lipoprotein lipase (LPL) at both enzyme and transcription ... LPL also are markers for adipocytes.(2,7 ..... NUMICO Research B.V. for financial support (to Z.C.D.).
JOURNAL OF BONE AND MINERAL RESEARCH Volume 17, Number 3, 2002 © 2002 American Society for Bone and Mineral Research
Exposure of KS483 Cells to Estrogen Enhances Osteogenesis and Inhibits Adipogenesis ¨ WIK Z.C. DANG, R.L. VAN BEZOOIJEN, M. KARPERIEN, S.E. PAPAPOULOS, and C.W.G.M. LO
ABSTRACT Osteoblasts and adipocytes arise from a common progenitor cell in bone marrow. Whether estrogen directly regulates the progenitor cells differentiating into osteoblasts or adipocytes remains unknown. Using a mouse clonal cell line KS483 cultured in charcoal-stripped fetal bovine serum (FBS), we showed that 17-estradiol (E2) stimulates the differentiation of progenitor cells into osteoblasts and concurrently inhibits adipocyte formation in an estrogen receptor (ER)– dependent way. E2 increased alkaline phosphate (ALP) activity and nodule formation and stimulated messenger RNA (mRNA) expression of core-binding factor ␣-1 (Cbfa1), parathyroid hormone/parathyroid hormone–related protein receptors (PTH/PTHrP-Rs), and osteocalcin. In contrast, E2 decreased adipocyte numbers and down-regulated mRNA expression of peroxisome proliferatoractivated receptor–␥ (PPAR␥)2, adipocyte protein 2 (aP2), and lipoprotein lipase (LPL). Furthermore, the reciprocal control of osteoblast and adipocyte differentiation by E2 was observed also in the presence of the adipogenic mixture of isobutylmethylxanthine, dexamethasone, and insulin. Immunohistochemical staining showed that ER␣ and ER were present in osteoblasts and adipocytes. A new mouse splice variant ER2 was identified, which differed in two amino acid residues from the rat isoform. E2 down-regulated mRNA expression of ER␣, ER1, and ER2. The effects of E2 are not restricted to the KS483 cell line because similar results were obtained in mouse bone marrow cell cultures. Our results indicate that estrogen, in addition to stimulation of osteogenesis, inhibits adipogenesis, which might explain the clinical observations that estrogendeficiency leads to an increase in adipocytes. (J Bone Miner Res 2002;17:394 – 405) Key words:
osteoblast, adipocyte, estrogen receptors, bone marrow, core-binding factor ␣-1, peroxisome proliferator-activated receptor-␥
INTRODUCTION is associated with a decrease in bone and a concurrent increase in bone marrow adipocytes.(1,2) The bone-forming osteoblasts arise from bone marrow progenitor cells, which also are the direct progenitor cells of adipocytes.(2– 4) Whether estrogen directly regulates the progenitor cells differentiating into osteoblasts or adipocytes remains unknown.(1,5) There is accumulating evidence that estrogen, in addition to osteogenesis, affects adipogenesis. Estrogen suppresses
E
STROGEN DEFICIENCY
The authors have no conflict of interest.
lipoprotein lipase (LPL) at both enzyme and transcription levels and regulates body fat distribution.(6 – 8) Furthermore, estrogen receptor ␣ (ER␣) as well as ER is present in adipose tissues.(9,10) Recently, Heine et al.(11) showed that ER␣ knockout mice increased their adipose tissue. In addition, Jones et al.(12) reported that mice with a deficiency of aromatase, an enzyme responsible for estrogen biosynthesis, enhanced their adiposity. An increase of adipocytes is found also in postmenopausal women as well as in ovariectomized animals, which can be prevented by estrogen replacement.(9,13) Several essential transcription factors for adipocyte and osteoblast differentiation have been identified.(14,15) The
Department of Endocrinology and Metabolic Diseases, Leiden University Medical Center, Leiden, The Netherlands.
394
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS
master transcription factor for adipocyte differentiation is peroxisome proliferator-activated receptor-␥ (PPAR␥), a member of the steroid receptor superfamily.(15,16) The expression of PPAR␥2 is considered as an early marker for adipocytes.(15–18) In addition, adipocyte protein 2 (aP2) and LPL also are markers for adipocytes.(2,7,8) For osteoblasts, core-binding factor ␣-1 (Cbfa1) is essential for osteoblast differentiation.(14) Other specific osteoblast markers include osteocalcin and parathyroid hormone/parathyroid hormone– related protein receptor (PTH/PTHrP-R).(19) Two types of ERs, namely, ER␣ and ER, mediate the biological effects of estrogen in target tissues including bone(1,20,21) and adipose tissue.(8,11,12) Both receptors can form heterodimers within the target cells and ER modulates ER␣ transcriptional activity.(22,23) Furthermore, ER2, a splice variant of ER, can function as a negative regulator of estrogen action.(24) Bone cells contain both ER␣ and ER.(1,20) In bone marrow and trabecular bone of adult rats, messenger RNA (mRNA) expression of ER␣ is high whereas the level of ER is low.(25) In contrast, in osteoblastic cells isolated from 1-day-old rat calvaria, the opposite has been reported.(26) During osteoblast differentiation, the high level of ER mRNA remained constant and that of ER␣ increased progressively. In human osteoblast-like cells, there is a progressive increase in mRNA of both ERs during differentiation.(27) The aim of this study is to examine the estrogen action on the progenitor cells. For this, we used KS483 cells, which is a clonal cell line from fetal mouse calvaria.(28,29) These cells are able to form both calcified nodules and adipocytes when they are cultured in charcoal-stripped fetal bovine serum (FBS).(30) In addition, we studied the effects of estrogen on osteogenesis and adipogenesis in mouse bone marrow cells. Our results showed that estrogen stimulates the differentiation of progenitor cells into osteoblasts and concurrently inhibits adipocyte formation in an ER-dependent way.
MATERIALS AND METHODS Cell cultures KS483 cells: KS483 cells were cultured in phenol red– free ␣-minimum essential medium (␣-MEM) supplemented with 10% non– charcoal-stripped FBS (Gibco BRL Life Technologies, Breda, the Netherlands) and penicillin/ streptomycin as maintenance medium at 37°C in a humidified atmosphere of 5% CO2 in air. Cells were passaged by 0.01% trypsin and 1 mmol/liter EDTA treatment at 3- to 4-day intervals. For the experiments, cells were plated at a density of 15,000 cells/cm2 in a 12-well plate and cultured in charcoal-stripped FBS with a supplement of ascorbic acid (50 g/ml) beginning on day 4 and with a supplement of -glycerophosphate (5 mM) beginning on day 11 after plating. Under this condition, alkaline phosphatase–positive cells (ALP⫹) and Oil Red O⫹ cells were concurrently found beginning on day 7 and the calcified nodules were observed beginning on day 11. This culture system was used further to examine the effects of 17-estradiol (E2) on osteogenesis and adipogenesis. Cells were exposed continuously to E2 or other substances beginning on day 1 on and the experiments
395
lasted for 18 days. The adipogenic insulin, dexamethasone, and 1-methyl-3-isobutylxanthine (IDI) mixture included 1.6 M of insulin, 0.25 M of dexamethasone, and 0.5 mM of 1-methyl-3-isobutylxanthine. Medium was changed every 3– 4 days. For time course experiments, cells were exposed continuously to E2 from the indicated day onward. The experiment was stopped at day 18. Mouse bone marrow cells: For each experiment, mouse bone marrow cells from tibias and femurs of 3–5 female Swiss albino mice aged 6 – 8 weeks were pooled. All cultures were carried out in phenol red–free ␣-MEM supplemented with 15% charcoal-stripped FBS, 50 g/ml of ascorbic acid, 10 mM of -glycerophosphate, and 10⫺8 M of dexamethasone as well as penicillin/streptomycin at 37°C in a humidified atmosphere of 5% CO2 in air. Cells were plated at a density of 6 ⫻ 105 cells/cm2 in a 12-well plate with continuous treatment from day 1 to day 21 when the experiment was ended. Medium was changed every 3– 4 days.
Assays for ALP activity, DNA, and calcium ALP activity and the content of DNA in the cells from the same well were measured. Cells were washed by phosphatebuffered saline (PBS), frozen immediately, and kept at ⫺20°C till use. A solution of 1⫻ SSC with 0.01% sodium dodecyl sulfate (SDS) was added to the cells and then sonicated. ALP activity was determined kinetically by colorimetry using p-nitrophenylphosphate as the substrate at pH 10.5 and reading the optical density at 405 nm. DNA was measured by the method of the enhancement of fluorescence using Hoechst 33258 (Sigma, St. Louis, MO, USA) binding to DNA.(31) The number of nodules and adipocytes was counted objectively under a light microscope by two different observers. Calcium was determined with a commercial endpoint assay based on the cresolphthalein complexone method (Sigma).
Stainings Oil Red O staining and ALP staining: Oil Red O staining is specific for the lipid droplets in the adipocytes. Oil Red O (4.2 g) was dissolved in 1200 ml of isopropanol and diluted with distilled water at a ratio of 4:3. Cell monolayers were rinsed with PBS and fixed with 4% formalin for 10 minutes. After washing with water, cells were stained for 1 h by immersion with Oil Red O solution. The staining was stopped by rinsing with water. Finally, histochemical staining for ALP activity in the cells was determined by using naphthol AS-MX phosphate (Sigma) as a substrate and fast blue BB salt as a coupler, as described previously.(32) Immunohistochemistry: The streptavidin-biotin immunohistochemistry of incorporated bromodeoxyuridine into DNA (BrdU; cat. no. 93–3943; Zymed, San Francisco, CA, USA) was used for staining the proliferating KS483 cells according to the manufacturer’s instructions. ER␣ and ER immunohistochemistry was carried out on cell monolayers. Endogenous peroxidase was blocked by 20 minutes incubation in 2% H2O2 methanol solution at room temperature. After washing twice in PBS, cells were incubated in 1% bovine serum albumin (BSA) in PBS for 2 h at room
396
DANG ET AL.
TABLE 1. PRIMERS USED
IN
RT-PCR ANALYSIS
Gene
Sense
Antisense
Reference
LPL aP2 PPAR␥2 Cbfa1 Osteocalcin PTH/PTHrP-R ER␣ ER 2-Microglobulin
5⬘-AGGTCATCTTCTGTGCTAGG 5⬘-TGTGATGCCTTTGTGGGAACC 5⬘-GGGTCAGCTCTTGTGAATGG 5⬘-CGGTGCAAACTTTCTCCAGG 5⬘-GCAGCTTGGTGCACACCTAG 5⬘-TGCTTGCCACTAAGCTTCG 5⬘-TGCAATGACTATGCCTCTGG 5⬘-GAACTGGTGCACATGATTGG 5⬘-TGACCGGCTTGTATGCTATC
5⬘-ATGCTGGAAGACCTGCTATG 5⬘-CGTCTGCGGTGATTTCATC 5⬘-CTGATGCACTGCCTATGAGC 5⬘-CAGGAAGTTGGGACTGTCGG 5⬘-GGAGCTGCTGTGACATCCAT 5⬘-TCCTAATCTCTGCCTGCACC 5⬘-TTCAACATTCTCCCTCCTCG 5⬘-TCTTCGAAATCACCCAGACC 5⬘-CAGTGTGAGCCAGGATATAG
Negishi et al.(35)
temperature. Polyclonal antisera of ER␣ or ER (MC-20; cat. no. sc-542 for ER␣; Y-19, cat. no. sc-6819 for ER; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) were applied overnight at a dilution of 1:200 and 1:50, respectively. Biotinylated anti-rabbit immunoglobulin G (IgG) or anti-goat IgG was used as second antibody for 1 h at room temperature and peroxidase-conjugated streptavidin was used for an additional 1 h. Between each step the cells were washed twice for 10 minutes in PBS. Thereafter, 3,3⬘-diaminobenzidine (DAB) in tris-buffered saline (TB) buffer (0.05 M of Tris-buffered saline, pH 7.6) with H2O2 (0.03%) was applied at room temperature. In controls, the first antiserum was omitted. Western blot was carried out for the specificity of ER␣ or ER antibody. Ten microliters of plasma membrane suspension (1–2 mg/ml of protein) was mixed with 10 l of sample buffer (2 mM dithiothreitol [DTT], 0.5% bromophenol blue, 30% glycerol, and 20 mM of Tris/HCl, pH 6.8) and boiled for 3 minutes. Samples were run on 10% polyacrylamide slab gels; kaleidoscope prestained markers (no. 161-0324; BioRad, Hercules, CA, USA) were used as reference. After electrophoresis, proteins were electroblotted to nitrocellulose membranes (pore size, 0.45 m; code 401196; Schleicher & Schuell, Dassel, Germany). After blocking the membranes with the mixture of 3% milk powder, 1% BSA, and 0.1% gelatin, the proteins were probed with the antibody overnight at 4°C. Anti-rabbit or anti-goat IgG-peroxidase conjugates were used to visualize the ER␣ and ER epitopes with DAB as the chromogen.
RNA isolation Total RNA was extracted using a guanidinium isothiocyanate-phenol-chloroform method.(33) The amount of RNA was determined by UV spectrophotometry at a wavelength of 260 nm and the quality was checked on 1% agarose gel, which contains a concentration of 0.5 g/ml of ethidium bromide.
Competitive and semiquantitative reverse-transcriptase polymerase chain reaction Competitive reverse-transcriptase polymerase chain reaction (RT-PCR) was performed according to the method described by Van Bezooijen et al.(34) In brief, complementary DNA (cDNA) was obtained by RT of total RNA.
Okazaki et al.(36) Deckers et al.(19) Deckers et al.(19)
Internal standard plasmid pMUS, which encodes for the housekeeping gene 2-microglobulin, was used at different dilutions and coamplified over 33 cycles with the cDNA at conditions of 94°C for 30 s, 56°C for 30 s, and 72°C for 1 minute. The equal amount of cDNA was defined if the ratio of cDNA and construct is one. Primers used in this study were described in Table 1. Some of the primers were designed by using software Primer III (http://www.genome. wi.mit.edu) and checked in Genbank to eliminate crossreactivity with the other sequences. Most of the primer sets used crossed intron/exon boundaries with such large introns that eventual contaminations with genomic DNA will not be amplified in the amplification process. Furthermore, a PCR using RNA samples without a proceeding RT reaction did not reveal any amplions, indicating the absence of DNA contamination. A PCR mixture contained 20 ng of cDNA in the presence and absence of internal standard PCR buffer (Eurogentec, Seraing, Belgium), 0.25 M of oligonucleotide primers (Eurogentec), 1.5 mM of MgCl2, 0.2 mM of deoxy-NTPs; Promega Corp., Leiden, The Netherlands), and 0.5 U/l of Goldstar DNA polymerase (Eurogentec). Semiquantitative PCRs were performed in the absence of an internal standard at 94°C for 3 minutes, 56°C for 30 s, and 72°C for 1 minute for 26 –35 cycles, depending on the abundance of the amplicons. The PCR for PPAR␥2 in KS483 cells was carried out for 38 cycles at 60°C instead of 56°C. Products of PCR were loaded in 1.5% agarose gel containing ethidium bromide, photographed under UV light, and quantified as previously described.(19,34)
Statistics Data are presented as means ⫾ SEM. Differences between groups were accepted at p ⬍ 0.05, which were assessed by one-way analysis of variance (ANOVA) and, subsequently, the Student–Newman–Keuls multiple comparisons test using software Instat (GraphPad, San Diego, CA, USA).
RESULTS Osteogenesis and adipogenesis of KS483 cells exposed to E2 KS483 cells were cultured in the medium supplemented with 10% of charcoal-stripped serum. Both osteoblasts and
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS
397
FIG. 2. Osteogenesis and adipogenesis of KS483 cells exposed to E2 at a concentration of 10⫺7 M at different time points. KS483 cells were first cultured in charcoal-stripped serum for different times as indicated in the figure and then exposed continuously to E2. After 18 days of cell culture, the number of nodules and adipocytes was quantified. Significant differences (*p ⬍ 0.05) were found between controls and E2 treatment at all the indicated time points except day 11. Each value is the mean ⫾ SEM of the results from three different wells and is representative of results from three different experiments.
FIG. 1. Osteogenesis and adipogenesis of KS483 cells exposed to E2. These cells were exposed continuously to E2 at a concentration of 10⫺15–10⫺5 M beginning on day 1. After 18 days of cell culture, (A) cellular ALP activity, (B) the number of nodules, and adipocytes were quantified. Significant differences (*p ⬍ 0.05) were found at 10⫺5– 10⫺13 M for ALP activity and adipocytes and 10⫺6–10⫺13 M for nodule formation. Each value is the mean ⫾ SEM of the results from three different wells and is representative of results from at least five different experiments.
adipocytes concurrently were present after 7 days of cell culture. Calcified nodule formation appeared after 11 days of culture. KS483 cells were treated with various concentrations of E2 ranging between 10⫺15 and 10⫺5 M beginning on day 1. There was a dose-dependent increase in ALP activity, with the maximal levels obtained at E2 concentrations between 10⫺12 and 10⫺7; higher E2 concentrations decreased the stimulatory effects on ALP activity (Fig. 1A). The increase in ALP activity was associated with a progressive increase in calcified nodule formation with maximal effect at 10⫺7 M and 10⫺8 M (Fig. 1B). In contrast, there was a dose-dependent decrease in the number of adipocytes,
with maximal inhibition starting from an E2 concentration of 10⫺9 M. We then examined the time course of the osteogenic and adipogenic responses to E2 at the concentration of 10⫺7 M. As shown in Fig. 2, a delayed addition of E2 to the culture was associated with a progressive decrease in calcified nodule formation and a progressive increase in adipocyte number. When E2 was added on the 11th day of culture, both calcified nodule formation and adipocyte numbers at day 18 were not different from controls (Fig. 2). Furthermore, when KS483 cells were exposed only to E2 (10⫺7 M) from day 1 to 4, the number of adipocytes per well at day 11 decreased from 313 ⫾ 23 in controls to 89 ⫾ 18 in E2 exposure. We further tested the effects of continuous treatment for 18 days with the well-known adipogenic IDI mixture on KS483 cells. As expected, IDI stimulated adipogenesis (Fig. 3A). E2 reduced the IDI-increased adipocyte numbers back to control levels. In addition, IDI stimulated ALP activity and nodule formation and E2 further potentiated this increase (Figs. 3B and 3C). Quantification of proliferation of the cells by BrdU staining showed that IDI and E2 increased proliferation of KS483 cells by 165 ⫾ 10% and 167 ⫾ 10.9%, respectively. Similarly, DNA content of cells continuously exposed to IDI or E2 for up to 18 days increased by 161 ⫾ 30% and 197 ⫾ 25%, respectively.
398
DANG ET AL.
FIG. 3. Osteogenesis and adipogenesis of KS483 cells exposed to E2 at a concentration of 10⫺7 M (E7) in the presence of adipogenic stimulator IDI. KS483 cells were exposed continuously to E2, IDI, or IDI plus E2. After 18 days of cell culture, the number of (A) adipocytes and (C) nodules as well as (B) ALP activity were quantified. Significant differences were found for all the treatments (a, compared with controls; b, compared with the treatment of IDI alone). Each value is the mean ⫾ SEM of the results from three different wells and is representative of results from three different experiments.
Consistent with the foregoing findings, E2 and IDI increased mRNA expressions of three specific osteoblast markers Cbfa1, PTH/PTHrP-R, and osteocalcin. The increases of these osteoblast markers that occurred on day 11 and day 14 correspond to the beginning of nodule formation (Fig. 4). The additive effects of E2 and IDI also were observed on mRNA expressions of these three osteoblastic parameters. We further assessed the effects of E2 and IDI alone or in combination on mRNA expressions of three specific adipocyte markers: PPAR␥2, aP2, and LPL. E2 suppressed mRNA expressions of these adipocyte markers. When mRNA expressions of these adipocyte markers were stimulated by IDI, E2 counteracted these increases and decreased the adipogenic mRNA expression. The suppression occurred beginning on day 7, the time when adipocytes became apparent (Fig. 5). Treatment of the cultures with the specific estrogen antagonist ICI182,780 blocked completely the stimulatory effects of E2 on ALP activity and calcified nodule formation. ICI182,780 given alone had no effect on these parameters and did not induce adipogenesis of KS483 (Table 2). However, this compound aborted the inhibitory effect of E2 on adipogenesis in KS483. Therefore, the effects of E2 on osteogenesis and adipogenesis of KS483 cells were ERmediated. The foregoing stimulatory effects of estrogen on osteogenesis and the inhibitory effects on adipogenesis were not limited to the concentration of 10⫺7 M presented. The similar results also were obtained at the other concentrations such as 10⫺8 M of estrogen (data not shown).
ER␣ and ER in KS483 cells ER␣ and ER were detectable by immunocytochemistry during 18 days of cell culture and they were found in both osteoblasts and adipocytes (Fig. 6). In contrast, in the negative controls (omitting first antibodies) of ER␣ and ER, staining in KS483 cells was not found (data not shown). The
specificity of ER␣ and ER antibodies was checked further by Western blot, showing an apparent molecular mass of 70 kDa and 55 kDa for ER␣ and 61 kDa for ER (data not shown). These are in line with earlier reports.(37) ER␣ and ER mRNA expression in KS483 was detected at all sampling points during 18 days of cell culture. One band was detected for ER␣ and two bands were detected for ER, namely, ER1 and ER2 (Fig. 7). Sequence analysis of ER1 and ER2 revealed that ER2 contained an additional in-frame 54 base pair (bp) insertion in the ligandbinding domain encoding 18 amino acid residues (assession number: GenBank AY054413). This differed in two amino acid residues from the rat isoform. In control cultures, there was an up-regulation of ER␣, ER1, and ER2 at 11 days of cell culture and a downregulation of ER␣ and ER1 after 14 days (Fig. 7). E2 down-regulated ER␣ beginning on day 11 and downregulated ER1 and ER2 beginning on day 4 at all sampling points.
Osteogenesis and adipogenesis of mouse bone marrow cells exposed to E2 To test whether the effects of E2 on osteogenesis and adipogenesis in KS483 cells are cell line dependent, we treated mouse bone marrow cells with E2 and IDI alone or in combination. E2 alone increased ALP activity and calcium deposition in these bone marrow cells. These responses were blocked by ICI182,780. IDI alone increased ALP activity and calcium deposition. When IDI and E2 were added to the cell culture, E2 further potentiated IDIstimulated ALP activity and calcium deposition (Figs. 8A and 8B). Furthermore, the increase in ALP activity and calcium deposition by E2 alone or in combination with IDI was abrogated by ICI182,780. E2 decreased adipocyte numbers, whereas IDI increased adipocytes. In the presence of IDI, E2 inhibited the stimulatory effects of IDI on adipo-
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS
399
FIG. 4. Osteogenesis of KS483 cells exposed to E2, IDI, or both assessed by mRNA expression of three specific osteoblast markers (A) Cbfa1, (B) PTH/PTHrP-R, and (C) osteocalcin at different time points. PCR pictures are representative for Cbfa1 and PTH/PTHrP-R on day 11and for osteocalcin on day 14. Each value is the mean ⫾ SEM of the results from three different experiments. The relative mRNA levels were compared with the highest values in each experiment. Significant differences were indicated as a, which were compared with controls, and b, which were compared with the treatment of IDI alone, in the same day.
genesis (Fig. 8C). The inhibitory effects of E2 alone or in combination with IDI on adipogenesis were blocked by ICI182,780, whereas ICI182,780 alone did not increase adipocyte numbers.
DISCUSSION We showed that E2 directly regulates the differentiation of progenitor cells in an ER-dependent way in a clonal cell line KS483 and mouse bone marrow cells, resulting in a stimulation of osteoblast formation and an inhibition of adipocyte formation. These results are in agreement with observations in animal models of ovariectomy, ER␣ knockout and aromatase-deficient(8,9,11,12) and estrogen-deficient women.(38) KS483 is a fetal mouse calvaria-derived clonal cell line committed to the osteoblast lineage.(28,29) The finding that
these cells could differentiate into both osteoblasts and adipocytes shows that this cell line still has progenitor characteristics. In both KS483 and mouse mesenchymal bone marrow cultures, cells first proliferate and then differentiate into either osteoblasts that form calcified bone nodules or adipocytes. An increase in BrdU staining in the proliferation period and in DNA content measured at the end of the experiment showed that E2 enhanced proliferation of KS483 cells. These results are in line with the earlier reports in mouse and rats in which an increase of proliferation by E2 has been observed.(39,40) E2 could regulate progenitor cells reciprocally differentiating into osteoblasts and adipocytes. This conclusion was based on the following observations. There was an increase in ALP activity per well and an enhancement of ALP activity per cell in both KS483 and mouse bone marrow cells. These results fit with the notion that estrogen stimu-
400
DANG ET AL.
FIG. 5. Adipogenesis of KS483 cells exposed to E2, IDI, or both assessed by mRNA expression of three specific adipocyte markers (A) PPAR␥2, (B) aP2, and (C) LPL at different time points. PCR pictures are representative for PPAR␥2, aP2, and LPL expression on day 11. Each value is the mean ⫾ SEM of the results from three different experiments. The relative mRNA levels were compared with the highest values in each experiment. Significant differences were indicated as a, which were compared with controls, and b, which were compared with the treatment of IDI alone, in the same day.
TABLE 2. EFFECTS
OF
E2
ALP activity (mOD/minute per g DNA) Nodules (n/well) Adipocytes (n/well)
AND
ICI182,780
ON
OSTEOGENESIS
AND
ADIPOGENESIS
OF
KS483
C
E (10⫺9 M)
I (10⫺7 M)
E (10⫺9 M)/I (10⫺7 M)
153.7 ⫾ 8.6 15 ⫾ 2 186 ⫾ 7
354.7 ⫾ 8.6 84 ⫾ 4 40 ⫾ 2
139 ⫾ 15.1 13 ⫾ 2 191 ⫾ 8
135 ⫾ 4.6 13 ⫾ 2 182 ⫾ 11
lates osteoblast number and activity.(1,20,21) In contrast, compared with controls, less adipocytes were found in the cultures treated with E2 despite stimulation of proliferation. Moreover, despite the lesser expression of ERs in the cells in early stages than in late stages, the effects of E2 on the capacity of cells to differentiate into either phenotype was exerted early because the estrogen effects declined with the late addition of estrogen and the maximal estrogen effects appeared when estrogen was added after 1 day of plating. Our observations are in line with the observations that only multipotent bone marrow progenitor cells or cell lines other
than mature osteoblasts could differentiate into various phenotypes.(2,41,42) Interestingly, the well-known adipogenic mixture IDI stimulated, in addition, osteogenesis both in KS483 and in mouse bone marrow cells. This seems to contradict the notion of an apparent reciprocal relationship of these two phenotypes, that is, an increase of adipocytes occurring at the expense of osteoblasts.(2) However, IDI stimulated proliferation of KS483 cells. It is possible that these proliferating cells could differentiate into either osteoblasts or adipocytes, resulting in an increase in both osteoblast and
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS
401
FIG. 6. (A and B) Immunostaining of ER␣ and (C and D) ER in KS483 (A and C) before confluency and (B and D) after confluency. Adipocytes are indicated by arrows (magnification ⫻200).
adipocyte formation. Indeed, we observed that when KS483 cells were exposed to charcoal-stripped serum in the postproliferation period, that is, after day 4, these cells were still able to differentiate into both osteoblasts and adipocytes (Z. C. Dang, unpublished data, 2000). Further, it has been shown that dexamethasone stimulates the proliferation and differentiation of osteoprogenitor cells into osteoblasts and adipocytes.(43,44) As shown here, E2 regulates the key genes for osteogenic and adipogenic commitment, Cbfa1(14,45) and PPAR␥2,(15) respectively. PPAR␥2 suppresses the expression of Cbfa1 mRNA and plays a hierarchically dominant role.(46,47) The increase in osteoblasts observed in our KS483 cells and mouse bone marrow cells therefore could be caused by a duel effect of E2, namely, an enhancement of Cbfa1 and an inhibition of PPAR␥2 mRNA expression. Expressions of mRNA levels of PTH/PTHrP-R and osteocalcin in KS483 cells were increased during osteoblast differentiation,(19,29) indicating that E2 accelerates osteogenesis. Furthermore, we observed the dramatic effects of E2 on adipogenic markers and modest effects on osteoblastic markers. This might be because the majority of the cells in the culture were osteoblasts and adipocytes were only in a small amount. Therefore, the relative changes of osteoblast markers was smaller that those of adipocyte markers. E2 directly inhibited the differentiation of progenitor cells into adipocytes as evidenced by a decrease in adipocyte numbers and the down-regulation of mRNA expression of the adipogenic markers PPAR␥2, LPL, and aP2. However, compared with E2 effects on PPAR␥2 and aP2, E2 could not completely block the increase of LPL during differentiation. This may indicate that the regulation of LPL during differentiation of KS483 cells was different. The decrease of mRNA expression of LPL by E2 corroborates former reports that estrogen suppresses LPL mRNA expression in adipose tissue.(7,8,48) It has been shown that there is an estrogen-responsive suppressive element for LPL located at
the LPL promoter.(8) E2 suppresses this functional LPL promoter activity and therefore repressed LPL gene expression at the transcriptional level. Furthermore, our results of E2 inhibition on adipogenesis are consistence with the findings that genistein, an estrogenic compound, inhibited adipogenesis in 3T3-L1 preadipocyte cell line.(49) The parallel changes in adipocyte numbers and mRNA expression of PPAR␥2 support the central role of PPAR␥2 in adipogenesis. Indeed, we observed that an up-regulation of PPAR␥2 is paralleled with an increase of adipocytes during the differentiation of KS483. E2 inhibited PPAR␥2 during the whole period of culture and therefore low numbers of adipocytes were found. In contrast, IDI up-regulated PPAR␥2 and more adipocytes appeared in the culture. Moreover, E2 also could inhibit IDI-induced adipogenesis. The activity of nuclear receptors like PPAR␥ and ERs is determined not only by ligand concentration but also by the expression of coactivators.(50) Tcherepanova et al.(51) recently showed that peroxisome proliferator-activated receptor-␥ coactivator-1 (PGC-1) serves as a convergence point between PPAR␥ and ER␣. The interaction of PPAR␥ and PGC-1 is different from that of ER␣ and PGC-1. E2 enhances the binding of PGC-1 to ER␣ but not to PPAR␥.(51) Interestingly, in addition to PPAR␥2 and ER␣, we also detected PGC-1 in this KS483 cell line (Z. C. Dang, unpublished results, 2001). Therefore, it is speculated that an inhibition of adipogenesis by E2 in KS483 cells could be caused by the competition between ER␣ and PPAR␥ for the limited amount of coactivators like PGC-1. Another possibility is that direct interaction between ERs and PPAR␥ results in down-regulation of PPAR␥ activity. Further investigations on the relations of these steroid receptors and their coactivators are being carried out in our laboratory. ER␣ and ER were detected during the whole period of cell culture. Our results of ER⫹ osteoblasts and adipocytes are consistent with early reports showing the expression of both ERs in osteoblasts(1,20,21) and in adipose tissues.(9,10)
402
DANG ET AL.
FIG. 7. mRNA expression of ER␣, ER1, and ER2 in KS483 cells exposed to E2. PCR picture represents expression of ERs on day 11. Bold letters of amino acids indicate the difference of the splice variant ER2 between rats and mice. Quantification of mRNA expression of ER␣, ER1, and ER2 in KS483 cells exposed to E2 at different time points is shown in the bar graph. Each value is the mean ⫾ SEM of the results from three different experiments (*p ⬍0.05).
Furthermore, we showed that E2 effects on osteogenesis and adipogenesis in KS483 and mouse bone marrow cells were ER dependent. This conclusion is based on the observations that the specific ER antagonist ICI182,780(52–54) blocked the effects of E2 on osteogenesis and adipogenesis in KS483 cells as well as in mouse bone marrow cells. These results are in line with previous reports in rat and mouse, which showed that stimulation of osteogenesis could be blocked by ER antagonists.(39) The decrease of ALP activity in mouse bone marrow cells treated with ICI182,780 may indicate the toxic effects also occurred at this concentration. Furthermore, we found that the 10⫺8 M of ICI182,780, which had no toxic effects on mouse bone marrow cells, still blocked the 10⫺10 M of E2 effects.
We identified a new mouse splice variant of ER, which differed in two amino acids from rat isoform. The functions of ER splice variants in the differentiation of progenitor cells into osteoblasts or adipocytes are not known. It has been shown that both ER␣ and ER are coexpressed in many cells and ER modulates ER␣ transcriptional activity as an inhibitor or activator. ER acts as an activator at a low concentration of E2 and as an inhibitor at a high concentration of E2.(22,23) Five different splice variants of ER have been identified and they could function differently.(24,55,56) The wide range of effective E2 concentrations and their similar effects on osteogenesis and adipogenesis observed in KS483 support the notion of ER as an activator or inhibitor of estrogen signaling. We showed that two splice
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS
403
variants of ER were up-regulated during the differentiation of KS483 cells. Similar up-regulation of ER has been reported in human osteoblast-like SV-HFO cells during differentiation.(27) The up-regulation of ER␣ and ER during the differentiation of KS483 cells was not only at mRNA expression levels but also at the protein levels because we observed immunostaining intensity of these two antibodies was increased after nodule formation (Z. C. Dang, unpublished data, 2000). In conclusion, we showed that E2 directly regulates progenitor cells differentiating into more osteoblasts and less adipocytes in an ER-dependent way. The results of this study might be relevant for the pathogenesis and the treatment of estrogen-deficient osteoporosis.(57,58)
ACKNOWLEDGMENTS We thank C. van der Bent for the technical support and NUMICO Research B.V. for financial support (to Z.C.D.).
REFERENCES
FIG. 8. Osteogenesis and adipogenesis of mouse bone marrow cells exposed to E2 at a concentration of 10⫺9 M in the presence of adipogenic stimulator IDI. These cells were exposed continuously to E2, IDI, or IDI plus E2 beginning on day 1. After 21 days of cell culture, (A) cellular ALP activity, (B) calcium deposition, and (C) the number of adipocytes were quantified. ER antagonist ICI182,780 at a concentration of 10⫺7 M(I7) blocked the effects of E2. Each value is the mean ⫾ SEM of the results from three different wells and is representative of results from three different experiments.
1. Rickard DJ, Subramaniam M, Spelsberg TC 1999 Molecular and cellular mechanisms of estrogen action on the skeleton. J Cell Biochem Suppl 32/33:123–132. 2. Nuttall ME, Gimble JM 2000 Is there a therapeutic opportunity to either prevent or treat osteopenic disorders by inhibiting marrow adipogenesis? Bone 27:177–184. 3. Park SR, Oreffo ROC, Triffitt JT 1999 Interconversion potential of cloned human marrow adipocytes in vitro. Bone 24: 549 –554. 4. Bianco P, Robey PG 2000 Marrow stromal stem cells. J Clin Invest 105:1663–1668. 5. Jilka RL, Takahashi K, Munshi M, Williams DC, Roberson PK, Manolagas SC 1998 Loss of estrogen upregulates osteoblastogenesis in the murine bone marrow: Evidence for autonomy from factors released during bone resorption. J Clin Invest 101:1942–1950. 6. Bjorntorp P 1997 Hormonal control of regional fat distribution. Hum Reprod 12(Suppl 1):21–25. 7. Dieudonne MN, Pecquery R, Leneveu MC, Giudicelli Y 2000 Opposite effects of androgens and estrogens on adipogenesis in rat preadipocytes: Evidence for sex and site-related specificities and possible involvement of insulin-like growth factor 1 receptor and peroxisome proliferator-activated receptor ␥2. Endocrinology 141:649 – 656. 8. Homma H, Kurachi H, Nishio Y, Takeda T, Yamamoto T, Adachi K, Morishige K, Ohmichi M, Matsuzawa Y, Murata Y 2000 Estrogen suppresses transcription of lipoprotein lipase gene. J Biol Chem 275:11404 –11411. 9. Pedersen SB, Borglum JD, Eriksen EF, Richelsen B 1991 Nuclear estradiol binding in rat adipocytes regional variations and regulatory influences of hormones. Biochim Biophys Acta 1093:80 – 86. 10. Crandall DL, Busler DE, Novak TJ, Weber RV, Kral JG 1998 Identification of estrogen receptor beta RNA in human breast and abdominal subcutaneous adipose tissue. Biochem Biophys Res Commun 248:523–526. 11. Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS 2000 Increased adipose tissue in male and female estrogen receptor-␣ knockout mice. Proc Natl Acad Sci USA 97: 12729 –12734. 12. Jones MEE, Thorburn AW, Britt K, Hewitt KN, Wreford NG, Proietto J, Oz OK, Leury BJ, Robertson KM, Yao S, Simpson
404
13.
14. 15. 16. 17. 18. 19.
20. 21. 22.
23. 24.
25. 26. 27.
28.
29.
30.
DANG ET AL. ER 2000 Aromatase-deficient (ArKO) mice have a phenotype of increased adiposity. Proc Natl Acad Sci USA 97:12735– 12740. Gambacciani M, Ciaponi M, Cappagli B, Piaggesi L, De Simone L, Orlandi R, Genazzani AR 1997 Body weight, body fat distribution, and hormonal replacement therapy in early postmenopausal women. J Clin Endocrinol Metab 82:414 – 417. Yamaguchi A, Komori T, Suda T 2000 Regulation of osteoblast differentiation mediated by bone morphogenetic proteins, hedgehogs, and Cbfa1. Endocr Rev 21:393– 411. Rosen ED, Walkey CJ, Puigserver P, Spiegelman BM 2000 Transcriptional regulation of adipogenesis. Genes Dev 14: 1293–1307. Desvergne B, Wahli W 1999 Peroxisome proliferatoractivated receptors: Nuclear control of metabolism. Endocr Rev 20:649 – 688. Chawla A, Lazar MA 1994 Peroxisome proliferator and retinoid signaling pathways co-regulate preadipocyte phenotype and survival. Proc Natl Acad Sci USA 91:1786 –1790. Hozumi Y, Kawano M, Saitom T, Miyata M 1998 Effect of tamoxifen on serum lipid metabolism. J Clin Endocrinol Metab 83:1633–1635. Deckers MML, Karperien M, Van der Bent C, Yamashita T, Papapoulos S, Lo¨ wik CWGM 2000 Expression of vascular endothelial growth factors and their receptors during osteoblast differentiation. Endocrinology 141:1667–1674. Spelsberg TC, Subramaniam M, Riggs LB, Khosla S 1999 The actions and interactions of sex steroids and growth factors/ cytokines on the skeleton. Mol Endocrinol 13:819 – 828. Fiorelli G, Brandi ML 1999 Skeletal effects of estrogens. J Endocrinol Invest 22:589 –593. Hall JM, McDonnell DP 1999 The estrogen receptor -isoform (ER) of the human estrogen receptor modulates ER␣ transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 140:5566 –5578. Pettersson K, Delaunay F, Gustafsson J 2000 Estrogen receptor  acts as a dominant regulator of estrogen signaling. Oncogene 19:4970 – 4978. Maruyama K, Endoh H, Sasaki-Iwaoka H, Kanou H, Shimaya E, Hashimoto S, Kato S, Kawashima H 1998 A novel isoform of rat estrogen receptor beta with 18 amino acid insertion in the ligand binding domain as a putative dominant negative regulator of estrogen action. Biochem Biophys Res Commun 246:142–147. Lim SK, Won YJ, Lee HC, Huh KB, Park YS 1999 A PCR analysis of ERalpha and ERbeta mRNA abundance in rats and the effect of ovariectomy. J Bone Miner Res 14:1189 –1196. Onoe Y, Miyaura C, Ohta H, Nozawa S, Suda T 1997 Expression of estrogen receptor beta in rat bone. Endocrinology 138:4509 – 4512. Arts J, Kuiper GGJM, Janssen JMMF, Gustafsson J, Lowik CWGM, Pols HAP, Van Leeuwen PTM 1997 Differential expression of estrogen receptors ␣ and  mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 138:5067–5070. Yamashita T, Asano K, Takahashi N, Akatsu T, Udagwa N, Sasaki T, Martin J, Suda T 1990 Cloning of an osteoblastic cell line involved in the formation of osteoclast-like cells. J Cell Physiol 145:587–595. Yamashita T, Ishii H, Shimoda K, Sampath TK, Katagiri T, Wada M, Osawa T, Suda T 1996 Subcloning of three osteoblastic cell lines with distinct differentiation phenotypes from the mouse osteoblastic cell line KS-4. Bone 19:429 – 436. Dang ZC, van Bezooijen RL, Karperien M, Papapoulos S, Lo¨ wik C 2000 Genistein and 17-estradiol stimulate osteoblastogenesis but inhibit adipogenesis. J Bone Miner Res 15: S1;S496 (abstract).
31. Labarca C, Paigen K 1980 A simple, rapid, and sensitive DNA assay procedure. Anal Biochem 102:344 –352. 32. Van Beek E, Van der Wee-Pals L, van de Ruit M, Nijweide P, Papapoulos S, Lo¨ wik C 1993 Leukemia inhibitory factor inhibits osteoclastic resorption, growth, mineralization, and alkaline phosphatase activity in fetal mouse metacarpal bones in culture. J Bone Miner Res 8:191–198. 33. Chomzcynski P, Sacchi N 1987 Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156 –159. 34. Van Bezooijen RL, Farih-Sips HC, Papapoulos SE, Lo¨ wik CW 1998 IL-1alpha, IL-1beta, IL-6, and TNF-alpha steady-state mRNA levels analyzed by reverse transcription-competitive PCR in bone marrow of gonadectomized mice. J Bone Miner Res 13:185–194. 35. Negishi Y, Kudo A, Obinata A, Kawashima K, Hirano H, Yanai N, Obinata M, Endo H 2000 Multipotency of a bone marrow stromal cell line, TBR31–2, established from ts-SV40 T antigen gene transgenic mice. Biochem Biophys Res Commun 268:450 – 455. 36. Okazaki R, Toriumi M, Fukumoto S, Miyamoto M, Fujita T, Tanaka K, Takeuchi Y 1999 Thiazolidinediones inhibit osteoclast-like cell formation and bone resorption in vitro. Endocrinology 140:5060 –5065. 37. Dechering K, Boersma C, Mosselman S 2000 Estrogen receptors ␣ and : Two receptors of a kind? Curr Med Chem 7:561–576. 38. Rosen CJ, Bilezikian JP 2001 Anabolic therapy for osteoporosis. J Clin Endocrinol Metab 86:957–964. 39. Qu Q, Perala-Heape M, Kapanen A, Dahllund J, Salo J, Vaananen K, Harkonen P 1998 Estrogen enhances differentiation of osteoblasts in mouse bone marrow culture. Bone 22:201–209. 40. Damien E, Price JS, Lanyon LE 2000 Mechanical strain stimulates osteoblast proliferation through the estrogen receptor in males as well as females. J Bone Miner Res 15:2169 –2177. 41. Dennis JE, Merriam A, Awadallah A, Yoo JU, Johostone B, Caplan AI 1999 A quadripotential mesenchymal progenitor cell isolated from the marrow of an adult mouse. J Bone Miner Res 14:700 –709. 42. Diascro DD, Vogel RL, Johnson TE, Witherup KM, Pitzenberger SM, Rutledge SJ, Prescott DJ, Rodan GA, Schmidt A 1998 High fatty acid content in rabbit serum is responsible for the differentiation of osteoblasts into adipocyte-like cells. J Bone Miner Res 13:96 –106. 43. Ishida Y, Tertinegg I, Heersche JNM 1996 Progesterone and dexamethasone stimulate proliferation and differentiation of osteoprogenitors and progenitors for adipocytes and macrophages in cell populations derived from adult rat vertebrae. J Bone Miner Res 11:921–930. 44. Ishida Y, Heersche JNM 1996 Progesterone stimulates proliferation and differentiation of osteoprogenitor cells in bone cell populations derived from adult female but not from adult male rats. Bone 20:17–25. 45. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G 1997 OSF2/Cbfa1: A transcriptional activator of osteoblast differentiation. Cell 89:747–754. 46. Lecka-Czernik B, Gubrij I, Moerman E, Kajkenova O, Lipschitz D, Manolagas S, Jilka RL 1999 Inhibition of Osf2/Cbfa1 expression and terminal osteoblast differentiation by PPAR␥2. J Cell Biochem 74:357–371. 47. Manolagas SC 2000 Birth and death of bone cells: Basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr Rev 21:115–137. 48. Hozumi Y, Kawano M, Hakamata Y, Miyata M, Jordan VC 2000 Tamoxifen inhibits lipoprotein activity: In vivo and in vitro studies. Horm Res 53:36 –39.
ESTROGEN INFLUENCES OSTEOGENESIS AND ADIPOGENESIS 49. Harmon AW, Harp JB 2001 Differential effects of flavonoids on 3T3–L1 adipogenesis and lipolysis. Am J Physiol Cell Physiol 280:C807–C813. 50. Jenster G 1998 Coactivators and corepressors as mediators of nuclear receptor function: An update. Mol Cell Endocrinol 143:1–7. 51. Tcherepanova I, Puigserver P, Norris JD, Spiegelman BM, McDonnell DP 2000 Modulation of estrogen receptor-␣ transcriptional activity by the coactivator PGC-1. J Biol Chem 275:16302–16308. 52. Gunther DF, Calikoglu AS, Underwood LE 1999 The effects of the estrogen receptor blocker, Faslodex (ICI182,780), on estrogen-accelerated bone maturation in mice. Pediatr Res 46:269 –273. 53. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138:863– 870. 54. Kuiper GGJM, Lemmen JG, Carisson B, Christopher J, Safe SH, Van der Saag PT, Van der Burg B, Gustafsson J 1998 Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor . Endocrinology 139:4252– 4263. 55. Price RH Jr, Lorenzon N, Handa RJ 2000 Differential expression of estrogen receptor beta splice variants in rat brain: Identification and characterization of a novel variant missing exon 4. Mol Brain Res 80:260 –268.
405
56. Pettersson K, Grandien K, Kuiper GG, Gustafsson JA 1997 Mouse estrogen receptor beta forms estrogen response element-binding heterodimers with estrogen receptor alpha. Mol Endocrinol 11:1486 –1496. 57. Khastgir G, Studd J, Holland N, Alaghband-Zadeh J, Fox S, Chow J 2001 Anabolic effect of estrogen replacement on bone in postmenopausal women with osteoporosis: Histomorphometric evidence in a longitudinal study. J Clin Endocrinol Metab 86:289 –295. 58. Compston JE 2001 Sex steroids and bone. Phys Rev 81:419 – 447.
Address reprint requests to: C. Lo¨wik, Ph.D. Department of Endocrinology and Metabolic Diseases (C4-R) Leiden University Medical Center Albinusdreef 2 2300 RC Leiden, The Netherlands
Received in original form May 21, 2001; in revised form September 10, 2001; accepted October 23, 2001.
