HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Longo, M. Articles by Teti, A. Search for Related Content PubMed PubMed Citation Articles by Longo, M. Articles by Teti, A.

Department of Experimental Medicine, University of L’Aquila, via Vetoio, Coppito, 2, 67100 L’Aquila, Italy
¹ European Molecular Biology Laboratories, Heidelberg, Germany
² Dompè S p A, via Campo di Pile, L’Aquila, Italy
³ Department of Medical Physiopathology, University ‘La Sapienza’ of Rome, Policlinico Umberto I, Rome, Italy

(Requests for offprints should be addressed to A Teti; Email: teti{at}univaq.it)

(S Migliaccio and A Teti contributed equally to this work)
(Gianfranco Caselli is now at Rotta Research Laboratorium S p A, Monza, Italy)

	Abstract

Top Abstract Introduction Materials and methods Results Discussion References

 
The human estrogen receptor (ER ) gene is driven by multiple promoters, of which the F promoter alone is found to be active in primary osteoblasts. The study was aimed at identifying new regulatory pathways affecting transcription of the receptor in this cell lineage. We generated human osteoblast-like cells, Saos-2, stably transfected with a luciferase-reporter gene downstream of the human ER F promoter (Saos F-Luc), and assayed the reporter response to differentiation-related signals. Over-confluence, shown to stimulate osteoblast differentiation, caused a time-dependent increase of F-promoter activity and correlated with an inactivation of protein kinase C (PKC ). PKC downregulation, obtained by long-term treatment with phorbol 12-myristate 13-acetate (PMA), resulted in promoter stimulation at similar levels in sub-confluent cells. The F promoter contains a putative PMA-responsive AP-1 site, but AP-1 activation was unremarkable in over-confluent cells. Treatment with PP1, a specific inhibitor of the non-receptor tyrosine-kinase c-Src, which is a negative regulator of osteoblast differentiation, showed that the activity of this kinase inhibits the F promoter. In PP1-treated cells, F-promoter activity was not further increased by PMA. Treatment with the generic kinase inhibitor 4-dimethylaminopyridine (DMAP) resulted in a dose-dependent induction of the promoter, which matched a parallel decrease of active c-Src. The effect was c-Src dependent, as DMAP caused no further promoter induction in PP1-treated cells. Overexpression of exogenous human ER resulted in modest promoter stimulation, which required the ligand-independent activator function 1 of the receptor. In murine primary osteoblasts, additional ER signal was observed upon induction of F promoter. In conclusion, we demonstrated a robust PKC/c-Src-dependent and estrogen-independent mechanism modulating transcription of ER in osteoblasts, probably affecting estrogen responsiveness during cell differentiation.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion References

 
Estrogens regulate a variety of tissues, including the bone (Couse & Korach 1999, Hall et al. 2001, Heshmati et al. 2002, Zaman et al. 2006). Estrogen deficiency leads to accelerated bone loss which is the primary cause of postmenopausal osteoporosis (Heshmati et al. 2002, Manolagas et al. 2002). The estrogen receptors (ERs) belong to the nuclear receptor superfamily, acting as ligand-inducible transcription factors (Couse & Korach 1999, Hall et al. 2001, Migliaccio & Marino 2003). Two genetically distinct forms of ERs (ER and ERß ) display non-identical expression profiles (Kuiper et al. 1996, Mosselman et al. 1996, Tremblay et al. 1997), with higher levels of ER relative to ERß in bone (Arts et al. 1997, Denger et al. 2001a,b).

Transcription of the human ER gene is driven by a rather complex region, comprising at least six core promoters (A–F), with some evidence for the existence of others (e.g. T promoter; Flouriot et al. 1998, Griffin et al. 1999, Brand et al. 2002, Lambertini et al. 2003). Alternative usage of these promoters results in tissue-specific mRNA heterogeneity (Flouriot et al. 1998, Griffin et al. 1999, Brand et al. 2002), predominantly residing in the 5'-untranslated regions (5'-UTRs), and arising from differential splicing of small untranslated exons bearing short open reading frames (ORFs). Importantly, differential splicing of the downstream, translated ORFs has also been observed and results in receptor heterogeneity at the protein level. Increasing evidence hints at physiological relevance of this variability in regulating the trans-acting properties of the receptor (Denger et al. 2001a,b). It is evident that the preferred observed receptor protein isoforms arise from preferential promoter usage, but this is likely to be the case, given the observation, in the human ER-positive breast cancer cell line MCF-7, of a naturally occurring ER splice variant originating exclusively from the promoters E and F, despite much higher activity of other ER promoters in these cells (Flouriot et al. 1998, Brand et al. 2002). There is also some evidence that the observed variability in the 5'-UTRs could be involved in physiological fine-tuning of the ER gene at the translation level in certain estrogen target tissues (Ko et al. 2002), through a mechanism directly involving the untranslated ORFs and their AUG (uAUG) ‘start’ codons. 5'-UTR-linked differential transcript folding, stability, or interaction with specific factors, are also good candidate mediators for upstream regulation (at a pretranslational level) of ER expression in the cell (Couse & Korach 1999, Migliaccio & Marino 2003).

The F (and, in some osteoblast-derived lineages, also the E) promoterhas been described as specifically active in bone cells (Flouriot et al. 1998, Griffin et al. 1999, Lambertini et al. 2003). No transcripts from other promoters have been detected in primary osteoblasts (Denger et al. 2001a,b, Lambertini et al. 2003), underscoring the absolute relevance of this promoter in bone physiology. One feature of the ER F promoter, namely, the presence of an AP-1 site closely upstream of a half estrogen-responsive element ( ERE) is shared by a variety of gene promoters (Denger et al. 2001a,b), well established as downstream targets of protein kinase C (PKC) pathways. In the present study, the involvement of PKCs in ER F-promoter modulation has been studied in proliferating and resting osteoblast-like cells, in the light of the well-known involvement of these kinases in osteoblast differentiation (Migliaccio et al. 1993, Davis et al. 1994) and their influence on AP-1-dependent transcription (Shaulian & Karin 2001). Furthermore, in view of our previous data pointing to PKC–c-Src interaction in osteoblast differentiation (Longo et al. 2004), c-Src relevance for the F promoter was also examined. Downregulation of c-Src activity is a well-established feature of differentiated osteoblasts (Marzìa et al. 2000), and both over-confluence-induced differentiation or c-Src inhibition are found to enhance F-promoter activity and increase ER protein in this work. In more detail, the increase of (a) putative shorter ER isoform(s) is observed in more differentiated osteoblasts at the mRNA and protein level (consistent with the results of Longo et al. 2004), and we hypothesized that lesser c-Src-dependent inhibition on the F promoter could trigger an increased production of (an) alternative ER isoform(s) during differentiation.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion References

 
Materials

Fetal bovine serum (FBS), glutamine, penicillin/streptomycin, and Dulbecco’s modified Eagle’s medium (DMEM) were purchased from HyClone. Trypsin 250, used for osteoblast isolation, was from Difco Laboratories. Hybond-C extra nitrocellulose membranes and the enhanced chemiluminescence (ECL) kit were from Amersham Bioscience. Mouse-monoclonal antibodies for PKC and PKC were purchased from BD Transduction Laboratories. The anti-Fyn and anti-p(Y416)Src antibodies were from Zymed Laboratories and Cell Signaling Technology respectively. PP1 and anti-v-Src antibody were from Calbiochem. The anti-ER antibodies (MC20 and H184) and the antibodies for Fos family, Jun family, and phosphorylated Jun-family isoforms, were purchased, along with anti-actin and all secondary antibodies, from Santa Cruz Biotechnology. RT-PCR reagents and lipofectamine were from Invitrogen. Brilliant SYBR Green QPCR master mix was purchased from Stratagene. Luciferase assay substrate was from Promega Corporation. All other reagents were of the purest analytical grade from Sigma Aldrich Company.

Cell cultures

Calvarial mouse primary osteoblasts were obtained through a modification of the sequential collagenase/–trypsin digestion method. Briefly, calvaria obtained from 7 to 9-day-old CD1 mice were washed in sterile Hank’s balanced salt solution and sequentially digested for 10, 20, and 40 min at 37 ° C with gentle agitation in the same medium plus 1 mg/ml type IV collagenase and 0.25% trypsin. Cells released in the second and third digestion were collected by mild centrifugation, resuspended in DMEM plus 10% FBS, 4 mM L-glutamine, 5 IU/ml penicillin and 5 µ g/ml streptomycin, and plated. All animal experiments were conducted in accordance with the principles and procedures outlined in the NIH ‘Guidelines and Care and Use of Experimental Animals’ and in the Italian Legislative Decree 116/92.

Soas-2cells were cultured inthe same medium, and both were left to grow at 37 ° C in water-saturated, 5% CO₂-containing atmosphere. At confluence, cells were trypsinized and replated for characterization and experiments.

For serum starvations, FBS was replaced with 0.2% BSA. When needed, FBS was replaced with charcoal-stripped serum.

Plasmids, transfections, and cell screening

Proliferating Saos-2 cells were co-transfected, by means of a standard CaPO₄ technique, with two plasmids: the pcDNA3.1 vector expressing neomycin resistance and Promega basic pGL-2 vectors harboring ER promoter A, B, or F upstream of in-frame firefly luciferase sequence. After transfection, the cells were cultured in selection medium containing 400 µ g/ml neomycin, fed fresh medium plus the antibiotic (G418) every 3–4 days and kept within a confluence range of 20%. After a few weeks, neomycin-resistant transfectants were isolated and amplified in standard culture conditions.

Transient transfection experiments were performed by standard lipofectamine procedure. The expression vectors for wild-type Runx2, wild-type YAP, and truncated, ‘dominant-negative’ YAP were a kind gift of Dr Jane Lain (Department of Cell Biology and Cancer Center, University of Massachussets Medical School, Worcester, MA, USA), while the vector for wild-type c-Src was from Upstate Biotechnologies. The ‘pHEO’ plasmids for the overexpression of full-length human ER and exon 1-skipped ER were described previously (Denger et al. 2001b).

Total cell lysates

The cells from 60 mm dishes were rinsed twice with ice-cold PBS, scraped in 70 µ l (sub-confluent cells) or 210 µ l (over-confluent cells) of 1 x Promega Reporter Lysis Buffer plus phosphatase inhibitors (1 mM sodium orthovanadate and 2.5 mM sodium fluoride (pH 7)), and disrupted by snap-freezing in liquid nitrogen. Lysates were then incubated on ice for 15 min with magnetic stirring and centrifuged at 4 ° C for 10 min at 25 000 g. Supernatants were stored at – 70 ° C.

Sub-cellular fractionations

Cells from 150 mm dishes were washed twice in ice-cold PBS, gently scraped in 1 ml same buffer, and collected by mild centrifugation (2000 g) for 3 min at 4 ° C. Cell pellets were resuspended in hypotonic homogenization buffer (Hepes (pH 7.4), 20 mM; EDTA, 1 mM; EGTA, 1 mM) plus DTT (2 mM) and protease inhibitors (PMSF, 2 mM; aprotinin, 6 µ g/µ l; leupeptin, 12 µ g/µ l), thoroughly sonicated on ice (three 20 s pulses at mid-output), incubated on ice for 15 min with magnetic stirring, and ultra-centrifuged (250 000 g for 40 min at 4 ° C). Supernatants were collected as soluble fractions and stored at – 70 ° C. Pellets were resuspended in isovolumes of the same homogenization buffer plus 1% Triton X-100, thoroughly sonicated on ice (three 20 s pulses at mid-output), incubated on ice for 15 min with magnetic stirring, and ultra-centrifuged (250 000 g for 40 min at 4 ° C) to pellet Triton-insoluble fractions. Supernatants were collected as membrane fractions and stored at – 70 ° C. When needed, Triton-insoluble pellets were thoroughly resuspended in denaturing buffer (NaCl, 150 mM; Tris–Cl (pH 8), 50 mM; and SDS, 0.1%) and used as ‘insoluble’ fraction.

Western blotting

Twenty to forty micrograms of the lysates were diluted with the respective lysis buffers up to a volume of 32 µ l, mixed with 8 µ l reducing 5 x loading buffer (Tris–Cl (pH 6.8), 313 mM; 10% SDS, 50% glycerol, 0.05% bromophenol blue dye, and 25% ß-H₂S-ethanol), subjected to 10% SDS-PAGE, and electrotransferred onto nitrocellulose membranes by standard methods. Membranes were incubated 40 min at room temperature in PBS plus 5% non-fat dry milk and 0.05% Tween-20 to block non-specific protein binding, then overnight at 4 ° C with primary antibodies, 0.5–1.5 µ g/ml, in PBS plus 1% non-fat dry milk and 0.05% Tween-20, followed by three 10-min washings in PBS. Immunocomplexes were evidenced by incubation for 1.5 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibodies, followed by three 10-min washings in PBS, and subsequent detection with an ECL kit, according to the manufacturer’s instructions. When necessary, mild membrane stripping was performed by prolonged washing (5 x 30 min) with 0.1 M glycine (pH 2.3).

RT-PCR

Total RNA was extracted using standard Trizol procedure. One microgram of RNA was reverse transcribed and the equivalent of 0.1 µ g was used for each PCR. For real-time PCRs, the Brilliant SYBR Green QPCR master mix was used. Primer pairs and PCR conditions used for real-time PCRs were as follows. Murine GAPDH: forward 5'-CACCATGGAGAAGGCCGGGG-3' , reverse 5'-GACGGA-CACATTGGGGGTAG-3' ; murine alkaline phosphatase: forward 5'-CCAGCAGGTTTCTCTCTTGG-3' , reverse 5'-CTGGGAGTCTCATCCTGAGC-3' ; murine Runx2: forward 5'-AACCCACGGCCCTCCCTGAACTCT-3' , reverse 5'-CTGGCGGGGTGTAGGTAAAGGTG-3' ; human GAPDH: forward 5'-CTGCACCACCAACTGCTTAG-3' , reverse 5'-AGGTCCACCACTGACACGTT-3' ; human alkaline phosphatase: forward 5'-CCGTGGCAACTC-TATCTTTGG-3' , reverse 5'-GCCATACAGGATGGCA-GTGA-3' ; human Runx2: forward 5'-AACCCACGA-ATGCACTATCCA-3' , reverse 5'-CGGACATACCGAGG-GACATG-3' . PCR conditions were, 45 cycles: 95 ° C x 45 s, 60 ° C x 45 s, and 72 ° C x 45 s.

The two primer pairs for the PCRs amplifying F-promoter-derived murine ER mRNA have been described previously (Ko et al. 2000). Two rounds of PCR were performed, the second of which using the ‘nested’ primer pair and the entire product of the first PCR as template. Conditions were, 35 cycles: 95 ° C x 30 s, 62 ° C x 30 s, 72 ° C x 30 s for the first PCR round, and identical conditions, but 53 ° C instead of 62 ° C, for the ‘nested’ amplification.

Luciferase assays

Twenty microliters of total cell lysates, obtained as described previously, were assayed for luciferase activity with reconstituted luciferase assay substrate in a Netzschalter 090003 luminometer (GSG Nuclear, Milan, Italy), according to the manufacturer’s instructions. The obtained values of relative luciferase units (RLUs) were normalized for protein content of each sample, determined through the Bradford method. Lysates did not exceed a concentration range of 0.6–0.9 µ g/µ l.

Alkaline phosphatase assays

Equal cell numbers (Saos-2 or primary osteoblasts) were seeded on 60 mm dishes and allowed to grow up to the desired degree of confluence. Cells were then rinsed in ice-cold PBS and lysed in 100 µ l of 0.1% SDS. Fifty microliters of each lysate were then used for colorimetric evaluation of alkaline phosphate activity using the Sigma Diagnostics kit no. 104 according to the manufacturer’s instructions. The obtained relative intensity units were normalized against total protein content of replicate cultures lysed in homogenization buffer (see above).

Statistical analysis

Data were expressed as mean ± S.E.M. of at least three independent experiments and statistically evaluated by one-way ANOVA followed by the Student’s t-test. Differences were considered significant when a P < 0.05 was computed.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

 
F-promoter, but not A- and B-promoter-driven reporter genes are active in Saos-2 cells

Saos-2 cells are a human osteosarcoma-derived cell line expressing high levels of osteoblast markers, such as bone alkaline phosphatase and very low levels of ER protein compared with primary osteoblasts. In these cells, ER mRNA appears to arise exclusively from the F and also (at variance with primary, non-immortalized osteoblasts) E promoters of the gene (Denger et al. 2001b). In order to ascertain an intrinsic inactivity of other ER promoters in these osteoblastic cells, we transfected them with reporter genes in which different full-length promoters (~1 kb) of the human gene (Fig. 1A) drive transcription of the firefly luciferase. A one-tenth molar ratio of a neomycin-resistance-carrying vector was added in all transfections to allow subsequent isolation of stably transfected cell populations.

View larger version (17K): [in this window] [in a new window]   Figure 1 Basal activity of hER promoters in Saos-2 cells. (A) Organization of the promoter region of the human ER gene. Dotted arrow tips: location of different core promoters, each containing an alternative transcription start site. Additional core promoters have also been described, both upstream and downstream of the F promoter. Small boxes (Ex 1F-B): short untranslated exons, containing short ORFs. Usage of a given ‘X’ promoter leads to production of an mRNA in which Ex ‘1X’ is attached (after RNA splicing) to the first translated exon (i.e., Ex 1A in full-length ER ). F-derived mRNA is an exception, for it retains Ex 1E between Ex 1F and the first translated exon. Ex 1A contains the canonical translation start codon (ATG) and is skipped in some splice variants. (B) Equal cell numbers of parental Saos-2 cells (control) and of derived sub-populations containing luciferase reporter genes for the human A, B, or F promoter, as indicated, were seeded on 60 mm dishes and allowed to grow to 80% confluence prior to lysis. Cell lysates were assayed for their luciferase activity (RLU/µ g). The graph depicts the results of more than three independent experiments performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01.

F-promoter activity is upregulated in over-confluent, more differentiated Saos-2 cells

Cells of the osteoblast lineage become more differentiated at prolonged over-confluence (Migliaccio et al. 1998). To investigate whether confluence-induced differentiation correlated with changes of ER F-promoter activity, time-course experiments were performed in which Saos F-Luc cells were collected at increasing times from plating, from a state of active proliferation (sub-confluent cells) up to over-confluence and cell-cycle inhibition.

Two osteoblast markers, namely, the transcription factor Runx2 and the alkaline phosphatase, were evaluated at the mRNA level by real-time RT-PCR to ascertain increased differentiation of over-confluent Saos cell cultures. As a positive control, a parallel evaluation was performed also on primary osteoblasts cultured in the same conditions, and the alkaline phosphatase was also assayed biochemically as an internal control. Saos-2 cells display a much differentiated phenotype in all conditions, but, nonetheless, an increase of both markers was evidenced in over-confluent cultures (Table 1). Higher increases were recorded, as expected, in primary osteoblast counterparts (Table 1).

View larger version (10K): [in this window] [in a new window]   Figure 2 Confluence-induced stimulation of the F promoter. Equal cell numbers of Saos F-Luc cells were seeded on 60 mm dishes and allowed to grow to confluence (time 0) or for the indicated times thereafter. Cell lysates were assayed for their specific luciferase activity (RLU/µ g). The graph depicts the results of more than three independent experiments performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01 versus first column.

PKC activity is lower in over-confluent osteoblasts and inhibits the F promoter

Changes in specific PKC activities have been shown to correlate with osteoblast differentiation (Migliaccio et al. 1998, Fragale et al. 1999, Lemonnier et al. 2001, Longo et al. 2004). To investigate whether any PKC modulations occurred in over-confluent Saos-2 cells with increased F-promoter activity, western blot analysis of two PKC isoenzymes, PKC and PKC , was performed on sub-cellular soluble and membrane fractions from these cells. Translocation of these molecules to membrane fractions is indicative of enzyme activation, and we had already observed such translocation occured in a rat osteoblast-like cell system (Longo et al. 2004) during differentiation. The chosen isoforms are representative of the two diacyl-glycerol-dependent PKC sub-classes.

Higher amounts of PKC were detected in the membrane versus soluble fractions in sub-confluent cells (Fig. 3A, upper panel, lanes 1 and 2). Conversely, most PKC was in the soluble fraction at over-confluence (Fig. 3A, upper panel, lanes 5 and 6). Importantly, an analogous redistribution of the enzyme was observed also in primary osteoblasts (described below; see Fig. 10). PKC was very poorly detectable, but, consistent with PKC , maximal PKC signal was seen in the soluble fractions from over-confluent cells (Fig. 3A, lower panel, lane 5).

View larger version (33K): [in this window] [in a new window]   Figure 3 F-promoter induction upon PKC inactivation. (A) Western blot analysis for PKC and PKC in fractionated protein extracts from sub-confluent (Sub) or over-confluent (Over) Saos F-Luc cells. Prior to collecting, cells were serum-starved for 24 h and treated with PMA or vehicle, DMSO, as indicated, for a further 16 h. The figure is representative of three independent experiments with similar results. S, soluble fractions; M, membrane fractions (Triton-soluble). (B) Luciferase activity in the same cells. The graph depicts results of more than three experiments expressed as mean values ± S.E.M. *P < 0.01 versus first column.

View larger version (59K): [in this window] [in a new window]   Figure 10 ER upregulation in primary osteoblasts upon F-promoter stimulation. (A) Primary osteoblasts from newborn mice were seeded at passage 1 and allowed to grow to 80% confluence (Sub) or to full over-confluence (Over). Cells were then serum starved for 24 h and treated with PMA (10– 7 M), PP1 (2 x 10– 6 M), or vehicle, DMSO, as indicated, for further 24 h. Cell soluble (detergent free) lysates were then subjected to western blot analysis for ER with a C-terminus-mapping antibody (MC20). The figure is representative of more than three independent experiments with similar results. (B) Left panels: primary osteoblasts from newborn mice were seeded at passage 1 and allowed to grow to 80% confluence (Sub) or to full over-confluence (Over), and then subjected to western blot analysis for ER with an antibody mapping a broad region of the molecule (H184). The figure is representative of three independent experiments with similar results. Right panels: primary osteoblasts from newborn mice, treated as above, were subjected to RNA extraction and RT-PCR analysis for F-promoter-derived murine ER mRNA. *Primers.

Confluence-induced stimulation of the F promoter does not rely on AP-1 activation

The human ER F promoter bears a putative AP-1 site, which is located upstream of, and very close to, a non-palindromic, half estrogen-responsive element (ERE; Dengeret al. 2001a,b), very proximally to the transcription start site(s). AP-1 complexes are well known for being strongly PKC/PMA responsive, which led us to ask whether the observed PKC/PMA-dependent increase of F-promoter activity could be mediated by this putative AP-1 site, through regulation of its cognate trans-acting factors of the Jun family and their Fos-family co-activators.

To explore this possibility, we performed western blot analyses for Jun- and Fos-family factors on total protein extracts from sub- and over-confluent serum-starved Saos F-Luc cells, either PMA-treated or not treated (Fig. 4). As a control for treatment efficacy, samples were also assayed for luciferase activity (Fig. 4, values on top panel). As is well known, Jun-family members require N-terminal phosphorylation on two serine residues, ser63 and ser73, for their transcriptional activity (Caelles et al. 1997, Behrens et al. 2000). Out of the two, phosphoser63-Jun proteins were virtually undetectable in untreated cells, either sub- or over-confluent (Fig. 4, first panel, first and third lanes). PMA treatment induced, as expected, a marked increase in this phosphorylation, both in sub- and over-confluent cells (Fig. 4, first panel, second and fourth lanes). Ser73 phosphorylation, on the other hand, was insensitive to PMA at sub-confluence, reduced in untreated over-confluent cells (Fig. 4, second panel; compare first and second lanes with third), and maximal in over-confluent PMA-treated counterparts (Fig. 4, second panel, fourth lane). Total Jun-family protein expression did not vary significantly (Fig. 4, third panel), while the expression of Fos-family co-activator factors was markedly reduced at over-confluence (Fig. 4, bottom panel). Therefore, the total AP-1 activation (Jun-family N-terminal phosphorylation plus Fos-family expression) appeared minimal at over-confluence, a condition whereby maximal reporter activity (2.3-fold relative to controls) was observed. This rules out the possibility of a main role for AP-1 in the F-promoter stimulation seen in over-confluent cells.

View larger version (67K): [in this window] [in a new window]   Figure 4 AP-1-independence of the F promoter. Sub-confluent (Sub) or over-confluent (Over) Saos F-Luc cells were serum-starved for 24 h and treated with PMA or vehicle, DMSO, as indicated, for further 16 h. Aliquots of cell lysates were then assayed for their luciferase activity or subjected to western blot analysis for the indicated AP-1 factors. Relative luciferase values (F Promoter %) on panel top are the means the of three experiments performed in triplicate, expressed as percentage of mean control (100 ± 2) value (P < 0.01). The western blot analysis shown displays results of one representative experiment.

The non-receptor tyrosine kinase c-Src is known to play a pivotal role in osteoblast differentiation, and c-Src-null primary osteoblasts appear to undergo ‘premature’ differentiation and show increased anabolism (Marzìa et al. 2000). Furthermore, our previous data (Longo et al. 2004) point to tight functional interactions between c-Src and PKCs in over-confluent, differentiating osteoblasts. In view of this, we explored the possibility of a potential role for c-Src on F-promoter regulation by means of the widely used, specific c-Src kinase inhibitor, PP1. Specificity and efficacy of PP1 in our cell systems was ascertained by analyzing its effect(s) on sub-cellular distribution of Fyn enzyme and c-Src activation respectively. Fyn is closest to c-Src within its enzyme family, and in parental Saos-2 cells, this enzyme appears to be restricted to triton-soluble and triton-insoluble/SDS-soluble fractions in basal conditions where many of its described targets are located. PP1 treatment was completely in influent on Fyn expression or distribution (Fig. 5A, left panels). In contrast, c-Src underwent a dramatic inactivation, as shown by a great decrease of its activating auto-phosphorylation at the tyrosine residue Y416, and also its total protein expression was altered, with slightly higher levels in treated cells (Fig. 5A, right panels).

View larger version (37K): [in this window] [in a new window]   Figure 5 c-Src inhibition causes F-promoter induction. (A) Specificity and effect of PP1. Western blot analysis for Fyn (left panels) in total and fractionated protein extracts from Saos-2 cells, and for total (Tot.) and activated c-Src (right panels), plus PKC (top right panel), in total extracts from control osteoblasts. Cells were serum starved for 24 h and treated with PP1 or vehicle (DMSO), as indicated, for a further 24 h. The figure is representative of three independent experiments with similar result. p(Y416) Src, phosphorylated c-Src form with full tyrosine-kinase activity; T, total (Triton based) extracts; S, soluble fractions; M, membrane fractions (Triton-soluble); I, Triton-insoluble/SDS-soluble fractions. (B) F-promoter induction by PP1. Equal cell numbers of Saos F-Luc cells were allowed to grow to 80% confluence, then were serum starved for 24 h and treated with PMA (10– 7 M), PP1 (2 x 10– 6 M) or vehicle, DMSO, as indicated, for further 24 h. Cell lysates were then assayed for their specific luciferase activity (RLU/µ g). The graph depicts the results of more than three independent experiments performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01 versus first column.

c-Src enzyme inactivation causes F-promoter induction

c-Src activation is regulated by phosphorylation of the enzyme at two C-terminal tyrosine residues, the Y527 and the above-mentioned Y416. Phosphorylation of the former tyrosine keeps the enzyme in a ‘closed’ conformation, while full enzyme activity is achieved after phosphorylation at Y416. The PP1 interferes only indirectly with this latter process. In fact, it inhibits c-Src catalytic site, which, in turn, affects c-Src auto-phosphorylation, an activity which can vary greatly depending on cell context/conditions. To further assess whether Y416 dephosphorylation resulted in enhanced F-promoter activity, Saos F-Luc cells were subjected to long-term (24 h) treatment with increasing doses of the generic tyrosine-kinase inhibitor, 4-DMAP, administered alone or in combination with PP1, and luciferase activity in cell lysates was measured. As shown in Fig. 6A (upper panel), DMAP caused a dose-dependent decrease of p(Y416)Src, which correlated with a parallel increase the F-promoter activity (Fig. 6B, first three columns). In PP1-treated cells, promoter activity was already high regardless of DMAP treatment. Only the highest DMAP dose resulted in some further induction (Fig. 6B). Notably, AP-1 activation, as indicated by p(ser63)Jun levels, was weak in all PP1-treated cells (Fig. 6A, middle panel, last three lanes), in sharp contrast with high promoter activity, thus further showing substantial independence of the F promoter from AP-1 complexes.

View larger version (44K): [in this window] [in a new window]   Figure 6 F-promoter activity matches c-Src inactivation, independent of AP-1. (A) Saos F-Luc cells were allowed to grow to 80% confluence, then serum starved for 24 h and treated with PMA (10– 7 M), or vehicle, DMSO, plus the generic tyrosine kinase inhibitor DMAP, 0.3–0.9 mM, as indicated, for a further 16 h, and then lysed and subjected to western blot analysis for p(Y416)Src and p(ser63)Jun. (B) Aliquots of the same lysates were assayed for their specific luciferase activity (RLU/µ g). The graph shows the results of three independent experiment performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01 versus first column; *P < 0.05 versus fourth column.

Over-confluence, DMAP, and PP1 all cause a decrease in c-Src activity, either arising from decreased levels of p(Y416)Src or from pure inhibition of kinase activity at the catalytic site of the enzyme. In all three cases, we observed an induction of the F promoter. As described previously, also PMA treatment stimulated the promoter, which prompted us to analyze its effect on c-Src. As shown in Fig. 7A, slightly lower levels of active c-Src, i.e., p(Y416)Src, were detected in soluble protein fractions from PMA-treated Saos-2 cells, with total c-Src unchanged. Therefore, importantly, lower c-Src activity is a common feature of all conditions in which the F promoter is stimulated.

View larger version (48K): [in this window] [in a new window]   Figure 7 c-Src/PKC crosstalk. (A) Saos-2 cells were allowed to grow to 80% confluence, then were serum starved for 24 h and treated with PMA or vehicle, DMSO (CTL) as indicated, for a further 16 h. Cell lysates were then subjected to western blot analysis for p(Y416)Src and total c-Src. (B) Saos-2 cells were allowed to grow to 80% confluence, then were serum starved for 24 h and treated with PMA, PP1, or vehicle, DMSO (CTL), as indicated, for a further 24 h. Cells were then lysed in hypotonic, detergent-free buffer and subjected to western blot analysis for sPKC .

Activator function 1 of ER modestly stimulates the F promoter, independent of c-Src

The human ER E and F promoters have been shown to be capable of driving transcription of RNAs whose subsequent splicing can originate exon 1-devoid product(s). Translated into protein, the resulting ER lacks its N-terminal hormone-independent trans-activation domain, denominated activator function 1 (AF1). In order to show potential auto-regulatory mechanisms, we transfected Saos F-Luc cells with expression construct for both the human full-length and the exon 1-devoid receptor isoforms, to investigate whether either affected F-promoter activity. Saos-2 cells contain very little endogenous ER protein(s), which makes these cells an ideal system for ER reintroduction studies. As shown in Fig. 8, the transfection of Saos F-Luc cells with increasing doses of full-length hER expression vector resulted in a significant, although modest, dose-dependent increase in promoter activity in steroid-starved cultures. The maximal upregulation observed ranged around 1.5-fold luciferase activity relative to controls transfected with green fluorescent protein (GFP) only. The observed induction was not affected by treatment with either estradiol (Fig. 8A, last column) or the pure anti-estrogen compound ICI (data not shown). In fact, both compounds invariably failed to affect reporter activity in a concentration range of 10^{– 7}–10^{– 9} M. In contrast with full-length ER , exogenous AF1-devoid ER exerted no effect on the reporter, either in the absence or in the presence of estradiol (Fig. 8B, last two columns).

View larger version (21K): [in this window] [in a new window]   Figure 8 AF1-dependent and hormone-independent induction of the F promoter by exogenous hER . (A) Equal numbers of Saos F-Luc cells were transfected with the indicated amounts of hER or GFP expression vectors and split into replicate cultures at the end of transfection. After 24 h cells were steroid deprived, treated with estradiol (E2) 5 x 10– 8 M or vehicle, DMSO, as indicated, and let to grow for further 24 h. Cell lysates were then assayed for their specific luciferase activity (RLU/µ g). Graph depicts the results of three independent experiment performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01 versus first column. (B) Equal numbers of Saos F-Luc cells were transfected with equal amounts of expression vectors for GFP (first and second column), or full-length (66 kDa) hER , or Exon 1-devoid (46 kDa) hER , and split into replicate cultures at the end of transfection. After 24 h, cells were steroid deprived, treated with estradiol (E2) 5 x 10– 8 M or vehicle, DMSO, as indicated, and let to grow for further 24 h. Cell lysates were then assayed for their specific luciferase activity (RLU/µ g). Graph depicts the results of three independent experiment performed in triplicate, expressed as mean values ± S.E.M. *P < 0.01 and +P < 0.05 versus first column.

The F promoter has three binding sites for the osteoblast-specific transcription factor, Runx2 (Lambertini et al. 2003). Zaidi et al.(2004) have shown that Runx2-dependent transcription of osteocalcin can be inhibited by c-Src through Yes-interActing Protein (YAP). By means of transient transfections, we tested whether an analogous mechanism underlay F-promoter inhibition by c-Src. Overexpression in Saos F-Luc cells (verified by western blot analysis) of exogenous wild-type Runx2 and wild-type or dominant-negative (Yagi et al. 1999) YAP failed to affect reporter activity (data not shown). Similar negative results were observed co-expressing wild-type or dominant-negative YAP along with wild-type c-src (data not shown). Therefore, the YAP/Runx2 pathway appears irrelevant for F-promoter activity in our cell system. Relevance of the cis-acting sequences harbored by the F promoter is summarized in Fig. 9.

View larger version (65K): [in this window] [in a new window]   Figure 9 Human ER F-promoter sequence. Shown is the region driving luciferase expression in the reporter gene used. Highlighted are consensus sequences for known transcription factors and two sequences binding unknown factors, plus their occupation/binding status.

Total ER is upregulated in over-confluent primary osteoblasts

As stated previously, Saos-2 cells are characterized by very low levels of ER protein(s). Therefore, we utilized primary osteoblasts to verify whether F-promoter-stimulating treatments led to increased levels of endogenous receptor. ER was evaluated both at the protein and at the mRNA level by western blot analysis and RT-PCR respectively.

For the former evaluation, various epitopes of the receptor were mapped, in view of ER polymorphism at the protein level. As shown in Fig. 10A, the appearance of additional ER immunosignals, corresponding to isoforms of lower ( < 66 kDa) molecular mass, was observed in cells subjected to F-promoter-inducing treatments (i.e. PMA, PP1, over-confluence). The most prominent additional signals were invariably found in over-confluent cells. In particular, antibodies recognizing broad regions of the receptor found a strong ER signal in over-confluent primary osteoblasts, corresponding to an isoform of ~46 kDa, absent in proliferating, sub-confluent controls (Fig. 10B, left panels). Of note, this finding is in agreement with our previous observations in rat osteoblasts (Longo et al. 2004, and unpublished observations).

To ascertain whether or not this additional ER signal originated from differential splicing of (upregulated) F-promoter-derived ER mRNA, we performed RT-PCR on RNAs from the same sub- and over-confluent cultures. The primer used (Ko et al. 2000) bind cognate sequences in the exon 1F (upstream primer) and exon 4 of the murine gene, so that they lead to specific amplification of F-promoter-derived mRNA, and to PCR products of different lengths according to differential splicing of the first exons. Despite low amplification efficiency (requiring two amplification rounds, described in Materials and methods), comparable levels of the same PCR product were obtained from sub- and over-confluent osteoblasts, and a shorter product only from the latter cells, which was absent in sub-confluent counterparts (Fig. 10B, right panels). Therefore, the additional ER signal(s) observed, at the protein level, in over-confluent osteoblasts originate(s), at least in part, from differential splicing of F-promoter-specific mRNA.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

 
Estrogen sensitivity represents a requirement for skeletal homeostasis. Osteoblasts express ER , and some of its activities could decline during differentiation (Migliaccio et al. 1993), suggesting modulated ER functions during the osteoblast lifespan. The mechanisms underlying the regulation of ER activity in bone cells are at present not well understood. In this study, we identified a pathway, acting at the transcriptional level, that could explain modulated sensitivity to estrogens in osteoblasts.

The ER gene generates a strikingly heterogeneous population of mRNAs. The various transcript variants appear to be distributed, to a certain degree, in a tissue-specific fashion (Flouriot et al. 1998, Griffin et al. 1999), and recent results shed light on their potential roles in the regulation of both ER expression (Ko et al. 2002) and, even more interestingly, ER function (Flouriot et al. 1998, 2000, Donaghue & Westley 1999, Bollig & Miksicek 2000, Brand et al. 2002, Zaman et al. 2006). To date, the least variability in ER mRNA has been observed in primary osteoblasts, where the only detectable transcripts appear to originate exclusively from the activity of the F promoter of the gene (Flouriot et al. 1998, Griffin et al. 1999), located far upstream (~12 kb; Inoue et al. 2002) of the translation start codon(s). In human osteosarcoma lines, also the E promoter appears to be active, indicating a tumor-related, non-physiological switching-on of this other promoter in some osteoblast-derived cells (Inoue et al. 2002). In yet other osteoblastic lines, such as the human MG-63 cells, the E promoter is not active, and all ER transcription relies, as in primary osteoblasts, on the F promoter alone. All osteoblasts express relatively low levels of ER protein and, consistently, appear to support limited transcription of the gene in comparison with estrogen target tissues, such as mammary gland or endometrium. Nevertheless, the activity of non-F promoters (namely, the B promoter) has been found also in a cell line, the HeLa cells, with extremely low ER transcription, strongly suggesting that the activity of the F promoter in osteoblasts should be regarded as specific rather than ‘residual’. In the present study, the widely active A and B promoters have been shown to be completely inactive, upstream of a reporter gene, in a human osteosarcoma line, the Saos-2 cells. In contrast, F-promoter-driven reporters proved fully active, which suggests an active choice of promoter in this osteoblast context rather than other mechanisms, such as lineage-dependent promoter silencing by methylation. In fact, such mechanism could not, presumably, affect the exogenous reporter genes.

In our transfected Saos-2 cells, we observed constant upregulation of the ER F promoter following over-confluence, a condition whereby cell proliferation is apparently almost abolished and which stimulates cell differentiation in osteoblast-derived lines as it does in primary osteoblasts (Migliaccio et al. 1998, Migliaccio & Marino 2003) and in our cell system (present study). At over-confluence, massive sub-cellular redistribution of a representative isoform of the PKC superfamily, PKC , was observed in our cells, in agreement with previous findings correlating osteoblast differentiation, PKCs, and ER function (Migliaccio et al. 1998, Fragale et al. 1999, Migliaccio & Marino 2003). Membrane tethering is a prerequisite for full enzyme activity of PKC, and membrane-bound PKC was greatly decreased in over-confluent cells, which suggested a possible inhibitory role exerted by PKCs on the F promoter. Consistent with this hypothesis, we could obtain reporter induction also in sub-confluent cells by means of the strong PKC downregulator PMA.

Incidentally, it has to be noted that we described some PKC activation, rather than inactivation, in other over-confluent osteosarcoma-derived cells, namely ROS cells (of rat origin; Longo et al. 2004). However, Saos-2 cells display a more differentiated phenotype than ROS, with high levels of alkaline phosphatase activity and low proliferation rates also at low cell densities, as well as the ability to form mineralized nodules (Lin et al. 2004).

The hER F-promoter harbors, very close to the transcription start site(s), a putative cis-acting AP-1 site located upstream of, and near to, a half, non-palindromic ERE sequence. Such configuration is shared by a number of known estrogen-responsive gene promoters (e.g. chicken ovalbumin; Griffin et al. 1999, Bollig & Miksicek 2000). AP-1 sites were previously known as TPA-responsive elements (TREs), with ‘TPA’ indicating the same compound as PMA, and sustained AP-1 activation is a well-known effect of PMA treatment. Since our promoter was stimulated by PMA and over-confluence at similar levels, we explored whether AP-1 activation could underlie both effects. PMA treatment caused a marked increase in Jun-family-activating phosphorylations, as expected, but these phosphorylations were minimal at over-confluence and, moreover, Fos-family factors were markedly reduced in this latter condition. Therefore, AP-1 activation cannot explain higher F-promoter activity at over-confluence. In view of such findings, it can be hypothesized that the effect of PMA in proliferating cells could not be AP-1 mediated. This would be consistent with our observation that PMA treatment of over-confluent cells results in no further F-promoter induction (data not shown). In conclusion, the proximal putative AP-1 could be altogether irrelevant for the F promoter. This would also be consistent with its estrogen-insensitivity, implying that the AP-1- ERE configuration harbored by the promoter is inactive in an osteoblast context. Moreover, a role for these sequences in osteoblastic MG-63 cells is excluded by Lambertini et al.(2003).

Beside the observed redistribution of PKC isoforms, decreased c-Src activity is a well-established feature of over-confluent, differentiated primary osteoblasts, as well as osteosarcoma cells (Longo et al. 2004). Many data points at close interaction between c-Src and members of the PKC superfamily (Longo et al. 2004, Takahashi et al. 2004), and much consistent evidence indicates an important role for c-Src in osteoblast differentiation (Marzìa et al. 2000). In view of these data, we explored the possibility of a c-Src role on F-promoter regulation. Specific enzyme inhibition (PP1) and/or inactivation (DMAP) caused remarkable reporter induction in sub-confluent cells, usually more pronounced than that obtained by PMA. Since both PP1 and PMA failed to further stimulate the reporter at over-confluence (data not shown), we can also conclude that F-promoter upregulation in this latter condition must rely on ‘already’ decreased c-Src/PKC activity. Which PKC isoforms (beside PKC ) exert an important upstream inhibition on the promoters remains to be elucidated.

The human ER F-promoter harbors multiple putative consensus sequences for Runx2, a transcription factor, whose role is pivotal for osteoblast differentiation (Runx2-null mice’s skeleton is entirely cartilaginous; Otto et al. 1997). Of these sequences, only one has been shown, in the above-mentioned extensive study by Lambertini et al.(2003), to be actually bound by Runx2 in the human preosteoblast cell-line MG-63. Runx2-dependent transcription can be inhibited by c-Src through YAP, a Src-family phosphorylation substrate, which acts as a Runx2-binding co-repressor in the nucleus (Zaidi et al. 2004). This mechanism acts, for instance, on one ‘canonical’ c-Src target gene, osteocalcin. By means of transient transfections, we demonstrated that no such mechanism affects the F promoter in our osteoblast-like cells. Few of the other known consensus sequences contained in the F promoter are found to be actually bound by MG-63 nuclear factors in the afore-mentioned study. Of these, the two most strongly bound sequences are not related to any known cognate transcription factor (shown in Fig. 9). PKC and c-Src-inhibitory effects on the F promoter could be exerted mainly on these cis-acting elements, through trans-acting factors yet to be unveiled. However, the role of cis-acting elements, such as SRY-factor-binding sequences remains to be elucidated.

Not surprisingly, a crosstalk between c-Src- and PKC-dependent signals was evident in our system. PP1 treatment, for instance, clearly affected PKC activity in Saos-2 cells. However, it is still noteworthy (although not surprising) that combined PMA/PP1 treatments did not stimulate the F promoter more than PP1 alone. Therefore, c-Src appears to act downstream of PKC, that is, PKC action could be partly due to modulation of c-Src. Consistent with this hypothesis, PKC downregulation by PMA resulted in less active (i.e. pY416-) c-Src. It has to be noted that this decrease in p(Y416)Src is not dramatic, which could explain the milder effect of PMA versus PP1 treatment on our reporter.

It is generally accepted that ER activation following estrogen binding eventually leads to downregulation of the receptor in the strongly ER-positive human breast cancer cell-line MCF-7 (Reid et al. 2002), in which all of the known promoters of the human ER gene seem to be, at various degrees, active. In contrast, our experiments of transient hER overexpression in Saos F-Luc cells resulted in a reproducible and dose dependent, although modest, upregulation of the reporter. Furthermore, such an effect was ligand independent, in that neither estradiol nor the pure anti-estrogen ICI, administered in lipid-deprived media, altered it. In view of such findings, we hypothesized that the observed stimulation could rely on the N-terminal, ligand-independent trans-activation domain of the receptor, denominated AF1 (activator function 1). Interestingly, there is evidence indicating that an AF1-devoid ER isoform naturally exists in the cell as a specific product (along with the full-length receptor) of F- and E-promoter-driven transcription (Flouriot et al. 1998, Brand et al. 2002). Such isoform can competitively inhibit AF1-dependent properties of the receptor (Flouriot et al. 2000), presumably through heterodimerization with the full-length form. Our reporter gene was altogether insensitive to AF1-devoid ER , consistent with our hypothesis. In other words, the F promoter appears, to some extent, AF1 dependent. Whether full-length ER trans-activates directly the F promoter remains unclear. In any case, it is interesting to note that massive overexpression of the receptor resulted in much lower reporter induction than did c-Src inhibition, which again underscores the importance of c-Src for F-promoter regulation. Therefore, the AF1 can be hypothesized to exert a fine-tuning role, rather than actual control, on the F promoter.

To gain insight into the biological significance of F-promoter induction by differentiation stimuli (i.e. confluence, c-Src inactivation, etc.), we analyzed endogenous ER expression in mouse primary osteoblasts cultured at over-confluence. In these cells, we invariably found increased ER immunosignal, which supports the idea that the same pathway(s) as in our osteosarcoma cells is active in all osteoblast lineages. Importantly, the higher ER signal observed was due to shorter proteins not observed in proliferating counterparts. The most prominent of these was of apparent molecular mass of 46 kDa, consistent with an exon 1-spliced, AF1-devoid protein. Moreover, RT-PCR showed the actual switch-on, in over-confluent osteoblasts, of an additional splicing eliminating a proximal (i.e. 5' ) exon on F-promoter-derived mRNA. Taken together, such findings strongly suggest that F-promoter stimulation might produce N-terminal-truncated protein(s) and receptor heterodimerization, with ‘dominant-negative’ effects. This would eventually lead to notable changes in ER-transcriptional activity in the cell, and perhaps to a tuning of ER gene itself at prolonged over-confluence (i.e. terminal differentiation), due to its AF1-dependence.

In summary, we have shown differentiation-dependent modulation, under strong c-Src/PKC control, of ER F-promoter activity in osteoblasts, which might result in modulated ratios among ER isoforms. We believe that further elucidation of the underlying mechanisms will contribute to a better understanding of estrogen physiology in bone tissue.

	Acknowledgements

 
This work was supported by the European Commission grant Genospora (contract no. QLK6-1999-02108, subcontractor) to A T. M L is recipient of a fellowship from the ‘Fondazione Italiana Ricerca sul Cancro’. We thank Ms Rita Di Massimo for her precious contribution towards the editing of this work. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.

	References

Top Abstract Introduction Materials and methods Results Discussion References

 
Arts J, Kuiper GG, Janssen JM, Gustaffson J-A, Lowik CW, Pols HAP & van Leeuwen JP 1997 Differential expression of estrogen receptor and ß mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 138 5067–5070.[Abstract/Free Full Text]

Behrens A, Jochun W, Sibilia M & Wagner EF 2000 Oncogenic transformation by ras and fos is mediated by c-jun N-terminal phosphorylation. Oncogene 19 2657–2663.[CrossRef][Web of Science][Medline]

Bollig A & Miksicek RJ 2000 An estrogen receptor alpha splicing variant mediates both positive and negative effects on gene transcription. Molecular Endocrinology 14 634–649.[Abstract/Free Full Text]

Brand H, Ko M, Denger S, Flouriot G, Gromoll J, Gannon F & Reid G 2002 A novel promoter is involved in the expression of estrogen receptor alpha in human testis and epididymis. Endocrinology 143 3397–3404.[Abstract/Free Full Text]

Caelles C, Gonzales-Sancho JM & Munoz A 1997 Nuclear hormone receptor antagonise with AP-1 by inhibition of the JNK pathway. Genes & Development 11 3351–3364.[Abstract/Free Full Text]

Couse JF & Korach KS 1999 Estrogen receptor null mice: what have we learned and were will they lead us? Endocrine Reviews 20 358–417.[Abstract/Free Full Text]

Davis VL, Couse J & Korach KS 1994 Correlation between low levels of estrogen receptors and estrogen responsiveness in two rat osteoblast-like cell lines. Journal of Bone and Mineral Research 9 983–991.[Web of Science][Medline]

Denger S, Reid G, Brand H, Ko M & Gannon F 2001a Tissue-specific expression of human ER and ERß in the male. Molecular and Cellular Endocrinology 178 155–160.[CrossRef][Web of Science][Medline]

Denger S, Reid G, Ko M, Flouriot G, Parsch D, Brand H, Korach KS, Sonntag-Buck V & Gannon F 2001b ER gene expression in human primary osteoblasts: evidence for the expression of two receptor proteins. Molecular Endocrinology 15 2064–2077.

Donaghue C & Westley BR 1999 Selective promoter usage of the human estrogen receptor gene and its regulation by estrogen. Endocrinology 13 1934–1950.

Flouriot G, Griffin C, Kenealy M, Sonntag-Buck V & Gannon F 1998 Differentially expressed messenger RNA isoforms of the human estrogen receptor gene are generated by alternative splicing and promoter usage. Molecular Endocrinology 12 1939–1954.[Abstract/Free Full Text]

Flouriot G, Brand H, Denger S, Metivier R, Ko M, Reid G, Sonntag-Buck V & Gannon F 2000 Identification of a new isoform of the human estrogen receptor-alpha (hER ) that is encoded by distinct transcripts and that is able to repress hER activation function 1. EMBO Journal 19 4688–4700.[CrossRef][Web of Science][Medline]

Fragale A, Tartaglia M, Bernardini S, Di Stasi AMM, Di Rocco C, Velardi F, Teti A, Battaglia PA & Migliaccio S 1999 Decreased proliferation and altered differentiation in osteoblasts from genetically and clinically distinct craniosynostotic disorders. American Journal of Pathology 154 1465–1477.[Abstract/Free Full Text]

Griffin C, Flouriot G, Sonntag-Buck V & Gannon F 1999 Two functionally different protein isoforms are produced from the chicken estrogen receptor gene. Molecular Endocrinology 13 1571–1587.[Abstract/Free Full Text]

Hall JM, Couse JF & Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. Journal of Biological Chemistry 276 36869–36872.[Free Full Text]

Heshmati HM, Khosla S, Robins SP, O’Fallon WM, Melton LJ III & Riggs BL 2002 Role of low levels of endogenous estrogen in regulation of bone resorption in late postmenopausal women. Journal of Bone and Mineral Research 17 172–178.[CrossRef][Web of Science][Medline]

Inoue A, Hayashi S, Aoyagi K, Nishigaki M, Sasaki H & Kiyama R 2002 A reporter gene assay for evaluation of tissue-specific responses to estrogens based on the differential use of promoters A to F of the human estrogen receptor alpha gene. Journal of Pharmacological and Toxicological Methods 47 129–135.[CrossRef][Web of Science][Medline]

Ko M, O’Brien S, Flouriot G & Gannon F 2000 Tissue-specific expression of multiple RNA variants of mouse estrogen receptor alpha gene. FEBS Letters 477 15–20.[CrossRef][Web of Science][Medline]

Ko M, Denger S, Reid G & Gannon F 2002 Upstream open reading frames regulate the translation of the multiple mRNA variants of the estrogen receptor alpha. Journal of Biological Chemistry 277 37131–37138.[Abstract/Free Full Text]

Kuiper GJM, Enmark E, Pelto-Huikko M, Nilsson S & Gustafsson JA 1996 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor and ß . PNAS 93 5925–5930.[Abstract/Free Full Text]

Lambertini E, Penolazzi L, Giordano S, Del Senno L & Piva R 2003 Expression of the human oestrogen receptor-alpha gene is regulated by promoter F in MG-63 osteoblastic cells. Biochemical Journal 372 831–839.[CrossRef][Web of Science][Medline]

Lemonnier J, Hay E, Delannoy P, Fromigue O, Lomri A, Modrowski D & Marie P 2001 Increased osteoblast apoptosis in Apert craniosynostosis: role of protein kinase C and interleukin-1. American Journal of Pathology 158 1833–1842.[Abstract/Free Full Text]

Lin Y, Liu LJ, Murray T, Sodek J & Rao LJ 2004 Effect of raloxifene and its interaction with human PTH on bone formation. Journal of Endocrinological Investigation 27 416–423.[Medline]

Longo M, Brama M, Marino M, Bernardini S, Korach KS, Wetsel WC, Scandurra R, Faraggiana T, Spera G, Baron R et al. 2004 Interaction of estrogen receptor alpha with protein kinase C alpha and c-Src in osteoblasts during differentiation. Bone 34 100–111.[Medline]

Manolagas SC, Kousteni S & Jilka RL 2002 Sex steroids and bone. Recent Progress in Hormone Research 57 385–409.[Abstract/Free Full Text]

Marzìa M, Sims NA, Voit S, Migliaccio S, Taranta A, Bernardini S, Faraggiana T, Yoneda T, Mundy GR, Boyce BF et al. 2000 Decreased c-Src expression enhances osteoblast differentiation and bone formation. Journal of Cell Biology 151 311–320.[Abstract/Free Full Text]

Migliaccio S & Marino M 2003 Estrogens and estrogen receptors: new actors in the plot of transcriptional regulation of genomic responses. Calcified Tissue International 72 181–182.[CrossRef][Web of Science][Medline]

Migliaccio S, Wetsel WC, Fox WM, Washburn TF & Korach KS 1993 Endogenous protein kinase C activation in osteoblast-like cells modulates estrogen responsiveness and estrogen receptor levels. Molecular Endocrinology 7 1133–1143.[Abstract/Free Full Text]

Migliaccio S, Bernardini S, Wetsel WC, Korach KS, Faraggiana T & Teti A 1998 Protein kinase C modulates estrogen receptor in differentiated osteoblastic cells in vitro. Steroids 63 342–354.

Mosselman S, Polman J & Dijkema R 1996 ERß : identification and characterization of a novel human estrogen receptor. FEBS Letters 392 49–53.[CrossRef][Web of Science][Medline]

Otto F, Thornell AP, Crompton T, Denzel A, Gilmour K-C, Rosewell IR, Stamp GW, Beddington RS, Mundlos S, Olsen BR et al. 1997 Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89 765–771.[CrossRef][Web of Science][Medline]

Reid G, Denger S, Ko M & Gannon F 2002 Human estrogen receptor alpha: regulation by synthesis, modification and degradation. Cellular and Molecular Life Sciences 59 821–831.[CrossRef][Web of Science][Medline]

Shaulian E & Karin M 2001 AP-1 in cell proliferation and survival. Oncogene 19 2390–2400.

Takahashi H, Suzuki K & Namiki H 2004 Pervanadate-induced reverse translocation and tyrosine phosphorylation of phorbol ester-stimulated protein kinase C beta II are mediated by Src-family tyrosine kinases in porcine neutrophils. Biochemical and Biophysical Research Communications 314 830–837.[CrossRef][Web of Science][Medline]

Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F & Giguere V 1997 Cloning, chromosomal localization and functional analysis of the murine estrogen receptor . Molecular Endocrinology 11 353–365.[Abstract/Free Full Text]

Yagi R, Chen LF, Shigesada K, Murakami Y & Ito Y 1999 A WW domain-containing yes-associated protein (YAP) is a novel transcriptional co-activator. EMBO Journal 18 1562–2551.

Zaidi SK, Sullivan AJ, Medina R, Ito Y, Van Wijnen AJ, Stein JL, Lian JB & Stein GS 2004 Tyrosine phosphorylation controls Runx2-mediated sub-nuclear targeting of YAP to repress transcription. EMBO Journal 23 790–799.[CrossRef][Web of Science][Medline]

Zaman G, Jessop HL, Muzylak M, De Souza RL, Pitsillides AA, Price JS & Lanyon LL 2006 Osteocytes use estrogen receptor alpha to respond to strain but their ERalpha content is regulated by estrogen. Journal of Bone and Mineral Research 21 1297–1306.[CrossRef][Web of Science][Medline]

Received 18 August 2006
Accepted 1 September 2006

This article has been cited by other articles:

				X.-Y. Li, Y. Lu, H.-Y. Sun, J.-Q. Wang, J. Yang, H.-J. Zhang, N.-G. Fan, J. Xu, J.-J. Jiang, R.-Y. Liu, et al. G protein-coupled receptor 48 upregulates estrogen receptor {alpha} expression via cAMP/PKA signaling in the male reproductive tract Development, January 1, 2010; 137(1): 151 - 157. [Abstract] [Full Text] [PDF]

				T. A. Roepke, C. Xue, M. A. Bosch, T. S. Scanlan, M. J. Kelly, and O. K. Ronnekleiv Genes Associated with Membrane-Initiated Signaling of Estrogen and Energy Homeostasis Endocrinology, December 1, 2008; 149(12): 6113 - 6124. [Abstract] [Full Text] [PDF]

				A. Spreafico, S. Schenone, T. Serchi, M. Orlandini, A. Angelucci, D. Magrini, G. Bernardini, G. Collodel, A. Di Stefano, C. Tintori, et al. Antiproliferative and proapoptotic activities of new pyrazolo[3,4-d]pyrimidine derivative Src kinase inhibitors in human osteosarcoma cells FASEB J, May 1, 2008; 22(5): 1560 - 1571. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Longo, M. Articles by Teti, A. Search for Related Content PubMed PubMed Citation Articles by Longo, M. Articles by Teti, A.