1α,25-Dehydroxyvitamin D3 Synergism toward Transforming Growth Factor-β1-induced AP-1 Transcriptional Activity in Mouse Osteoblastic Cells via Its Nuclear Receptor*

The present study demonstrates 1α,25-dehydroxyvitamin D3(1α-25-(OH)2D3) synergism toward transforming growth factor (TGF)-β1-induced activation protein-1 (AP-1) activity in mouse osteoblastic MC3T3-E1 cells via the nuclear receptor of the vitamin. 1α-25-(OH)2D3synergistically stimulated TGF-β1-induced expression of the c-jun gene in the cells but not that of the c-fos gene. We actually showed by a gel mobility shift assay 1α-25-(OH)2D3 synergism of TGF-β1-induced AP-1 binding to the 12-(O-tetradecanoylphorbol-13-acetate response element (TRE). 1α-25-(OH)2D3 markedly stimulated the transient activity of TGF-β1-induced AP-1 in the cells transfected with a TRE-chloramphenicol acetyltransferase (CAT) reporter gene. Also, a synergistic increase in TGF-β1-induced CAT activity was observed in the cells cotransfected with an expression vector encoding vitamin D3 receptor (VDR) and the reporter gene. However, the synergistic CAT activity was inhibited by pretreatment with VDR antisense oligonucleotides. In addition, in a Northern blot assay, we observed 1α-25-(OH)2D3 synergism of TGF-β1-induced expression of the c-jun gene in the cells transfected with the VDR expression vector and also found that the synergistic action was clearly blocked by VDR antisense oligonucleotide pretreatment. The present study strongly suggests a novel positive regulation by 1α-25-(OH)2D3 of TGF-β1-induced AP-1 activity in osteoblasts via “genomic action.”

TGF-␤1 1 is locally produced by osteoblasts and accumulates abundantly in bone matrix tissue (1)(2)(3). This local cytokine plays an important role as a "coupling factor" in bone remodeling (4 -7). We (8,9) previously demonstrated that TGF-␤1 is a potent activator of AP-1, a transcriptional factor that is a heterodimer of FOS and JUN proteins, in mouse osteoblastic MC3T3-E1 cells. AP-1 activates the transcription of target genes by binding to specific promoter elements called TRE. In fact, several studies (10 -16) have suggested that AP-1 is a regulatory factor in bone metabolism. Therefore, it is of interest to investigate the mechanism by which AP-1 regulates the metabolism of TGF-␤1-treated bone cells, which regulation may occur in an autocrine and/or paracrine fashion.
1␣-25-(OH) 2 D 3 also is an important systemic hormone in bone metabolism and regulates transcriptionally the expression of several genes involved in the differentiation of osteoblastic cells (17)(18)(19)(20)(21)(22)(23). Many studies (20 -26) have shown that the hormone binds to its receptor, VDR, in the cell nucleus and acts via binding of this complex to the VDR response element. This VDR-dependent action is referred to as the "genomic action" of the hormone. On the other hand, other recent studies (27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37) have suggested that the hormone is also able to induce several biological activities via protein kinase C, ceramide signaling pathways, and via an increase in the intracellular calcium concentration, which are called "nongenomic action." Several investigators (21)(22)(23)(24) have demonstrated negative regulation between AP-1 and VDR of the transcriptional activity of osteocalcin and collagen genes, both of which are involved in bone formation. However, positive regulation by 1␣-25-(OH) 2 D 3 of AP-1 transcriptional activity in osteoblastic cells has not been demonstrated in detail. Therefore, it is of interest to explore the possibility of 1␣-25-(OH) 2 D 3 positive regulation of AP-1 transcriptional activity via "genomic" or "nongenomic" action.
In this regard, we investigated in the present study the regulation by 1␣-25-(OH) 2 D 3 of TGF-␤1-induced AP-1 transcriptional activity in osteoblastic MC3T3-E1 cells. As a result, we demonstrated the presence of 1␣-25-(OH) 2 D 3 synergism toward TGF-␤1-induced AP-1 transcriptional activity via genomic action. This demonstration suggests the presence of a novel positive regulation by 1␣-25-(OH) 2 D 3 of AP-1 transcriptional activity in osteoblastic cells via the VDR-dependent pathway (genomic action).
Cell Culture-Clonal osteoblastic MC3T3-E1 cells derived from C57BL/6 mouse calvaria were cultured in ␣-MEM supplemented with 10% FCS in plastic dishes at 37°C and 5% CO 2 in air and subcultured every 3 days as described previously (8,38). The cells (3 ϫ 10 5 cells) were cultured at 37°C under an atmosphere of 5% CO 2 in air in ␣-MEM supplemented with 10% FCS in 90-mm plastic dishes until nearly confluent. Then the cells were washed, incubated for 24 h in serum-free * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
␣-MEM, and subsequently treated for the desired periods in serum-free ␣-MEM with or without test samples at various concentrations.
Northern Blot Analysis-Total cellular RNA was extracted by the guanidine isothiocyanate procedure (39). As described previously (9,38), the RNA was subjected to 1% agarose electrophoresis and blotted onto a nylon membrane (MSI Magnagraph, Westboro, MA), and the membranes were subsequently baked, prehybridized, and then hybridized with mouse c-fos cDNA (Oncor, Gaithersburg, MD), mouse c-jun cDNA (ATCC, Rockville, MD), and human VDR cDNA (ATCC) probes labeled with 5Ј-[␣ 32 P]dCTP by use of the megaprime DNA labeling system. After hybridization, the membranes were washed, dried, and exposed to x-ray film (Eastman Kodak Co.) at Ϫ70°C. ␤-Actin was used as an internal standard for quantification of total RNA in each lane of the gel.
Nuclear Transcriptional (Run-on) Assay-This assay was performed according to the method of Groudine et al. (40) as described previously (8,41). Nuclei were prepared essentially as described by Diagnam et al. (42). In brief, cells (5 ϫ 10 7 cells) were cultured at 37°C under an atmosphere of 5% CO 2 in air in ␣-MEM supplemented with 10% FCS in 15-cm plastic dishes until nearly confluent. Then the cells were washed and incubated for 24 h in serum-free ␣-MEM. In addition, the cells were next treated or not for 24 h with 1␣-25-(OH) 2 D 3 and then for 40 min with TGF-␤1, scraped into phosphate-buffered saline, and centrifuged. Subsequently, the cell pellet was suspended in a lysis buffer (10 mM Tris (pH 7.4), 3 mM MgCl 2 , 10 mM NaCl, 0.5% Nonidet P-40) after which the nuclei were separated from the cytosol by centrifugation at 3,000 ϫ g for 15 min. Transcription initiated in intact cells was allowed to proceed for 30 min at 30°C in the presence of 5Ј-[␣-32 P]UTP, and the RNA was then isolated and hybridized to slot-blotted cDNA probes (5 g/slot). Blots were hybridized for 72 h and autoradiographed for 3 days. ␤-Actin gene was utilized as an internal standard.
Preparation of Nuclear Extracts-Confluent monolayers in 15-cm diameter dishes were treated with test samples as indicated in the figure legends, and then their nuclei were isolated as described above, and extracts of them were prepared as described previously (8,9). Protein concentration was measured by the method of Bradford (43).
Gel Mobility Shift Assay-This assay was carried out as described previously (8,9). Binding reactions were performed for 20 min on ice with 5 g of nuclear protein of 20 l of binding buffer (2 mM HEPES (pH 7.9), 8 mM NaCl, 0.2 mM EDTA, 12% (v/v) glycerol, 5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 g of poly(dI-dC)) containing 20,000 cpm of 32 P-labeled oligonucleotide in the absence or presence of nonlabeled oligonucleotide. Poly(dI-dC) and nuclear extract were first incubated at 4°C for 10 min before addition of the labeled oligonucleotide. 30-Mer double-stranded oligonucleotides containing the -TGACTCA-sequence (Oncogene Science, Inc., Manhasset, NY) of the AP-1 binding site were end-labeled by the oligonucleotide 5Ј-end-labeling [␥-32 P]ATP method. Reaction mixtures for the binding were incubated for 15 min at room temperature after addition of the labeled oligonucleotide. Unlabeled double-stranded oligonucleotide was used as the competitor. DNA-protein complexes were electrophoresed on native 6% polyacrylamide gels in 0.25ϫ TBE buffer (22 mM Tris, 22 mM boric acid, and 0.5 mM EDTA (pH 8.0)). Gels were vacuumed, dried, and exposed to Kodak x-ray film at Ϫ70°C.
Plasmid Construction for Transient Expression Assay-The plasmid pTRE-TK-CAT was constructed by inserting a synthetic oligonucleotide containing the -TGACTCA-motif with HindIII-XbaI sites into the cor-responding sites of pTK-CAT, which contains the herpes simplex virusthymidine kinase promoter enhancer region located upstream of the CAT gene. The pGL-TK used also contains the herpes simplex virusthymidine kinase promoter enhancer region, but it is located upstream of the luciferase gene, and this plasmid was made from a pGL3-Enhancer Vector (Promega, Madison, WI). pSG-VDR is a rat VDR expression vector.
Transient Expression Assay-This assay was performed according to the method of Felgner et al. (44,45). The cells (1 ϫ 10 6 cells) were inoculated in 5-cm diameter dishes and incubated for 12 h in 10% FCS containing ␣-MEM. Then the cells were washed 3 times with Opti-MEM (Life Technologies, Inc.), transfected with a total of 7 g of DNA by use of 10 g of LipofectAMINE (Life Technologies, Inc.), and incubated for 6 h in serum-free Opti-MEM. The cells were transfected with 2 g of reporter plasmid, and the expression vector for the nuclear receptor (4 g of each expression vector) was used. The assay was performed in the presence of 1 g of pGL-TK, a luciferase expression plasmid, used as an internal control to normalize for variations in transfection efficiency. Bluescribe M13 ϩ (Stratagene, La Jolla, CA) was used as a carrier to adjust the total amount of DNA. The cells were washed three times after the transfection, and 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M or its analogs at 10 Ϫ8 M in serum-free ␣-MEM was then added. After a 24-h incubation, the cells were treated for 6 h with TGF-␤1. The cellular extracts were prepared by use of Reporter Lysis Buffer (Promega) and subjected to the CAT assay after normalizing luciferase activity. CAT activity was determined by autoradiography of thin layer chromatography (TLC) plates following completion of the CAT reaction using the appropriate concentration of D-threo-[dichloroacetyl-1-14 C]chloramphenicol (Amersham Pharmacia Biotech, Japan) as described previously (46,47).
Preparation of VDR Antisense or Sense Oligonucleotide-VDR antisense (5Ј-GCT GGC TGC CAT TGC CTC-3Ј) phosphorothioate oligodeoxynucleotide was synthesized and purified as described previously (8,9,41). These nucleotide sequences were complementary to the first 18 bases following the AUG sequence of mouse VDR mRNA. Also, the corresponding sense oligonucleotide was prepared and used as a control.

RESULTS
We (8,9) previously demonstrated by the gel mobility shift assay that TGF-␤1 induces the binding of AP-1 to TRE in osteoblastic MC3T3-E1 cells. Since AP-1 typically appeared when the cells were treated for 3 h with TGF-␤1 at 1 ng/ml, in this study we investigated the regulatory action of 1␣-25-(OH) 2 D 3 on TGF-␤1-induced AP-1 activity in the cells under these experimental conditions. Thereafter, total RNA in the cells was prepared at 1.5 h after the initiation of the cytokine treatment. Quantification of c-jun mRNA level was done by densitometry and is expressed as a percentage of maximum. B, cells were incubated for the indicated times in the presence or absence of 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M and then treated with TGF-␤1 (1 ng/ml). Thereafter, total RNA in the cells was prepared at 1.5 h after the initiation of the cytokine treatment. Northern blot analysis was performed with c-jun, c-fos, and ␤-actin cDNAs used as probes. Quantification of c-jun mRNA level was done by densitometry and is expressed as a percentage of maximum. An identical experiment independently performed gave similar results.
observed at a physiological concentration (10 Ϫ10 M) of the hormone. However, such action was not observed in the expression of the cytokine-induced c-fos gene in the cells. Also, the synergistic effect of 1␣-25-(OH) 2 D 3 on the expression of the c-jun gene in the cells was observed even when the hormone pretreatment time was only 1 h before the cells were incubated with TGF-␤1 and tended to be pretreatment time-dependent (Fig. 1B).

1␣-25-(OH) 2 D 3 Synergistic Effect on TGF-␤1-induced Expression of the c-jun Gene in MC3T3-E1 Cells Occurs at the Tran-
scriptional Level-Next we examined, using the run-on assay, whether or not the 1␣-25-(OH) 2 D 3 synergistic action toward TGF-␤1-induced expression of the c-jun gene in the cells operates at the transcriptional level. The cells were pretreated or not for 24 h with 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M and then were treated or not for 1 h with TGF-␤1 at 1 ng/ml. Thereafter, the run-on assay was performed using nuclei isolated from the cells. Fig. 2, A and B, shows that 1␣-25-(OH) 2 D 3 clearly stimulated the transcriptional activity of the TGF-␤1-induced c-jun gene. These results indicate the synergistic action of 1␣-25-(OH) 2 D 3 at the transcriptional level for TGF-␤1-induced expression of c-jun gene in the cells.
Synergistic Effect of 1␣-25-(OH) 2 D 3 on AP-1 Binding to TRE in TGF-␤1-treated MC3T3-E1 Cells-AP-1 is a heterodimer of FOS and JUN proteins and binds to the TRE consensus sequence (48). Synergistic stimulation by 1␣-25-(OH) 2 D 3 of TGF-␤1-induced expression of the c-jun gene in the cells suggested to us that the hormone may remarkably increase AP-1 binding to TRE in the cytokine-treated cells. Therefore, we examined this point by the gel mobility shift assay. As we expected, and as shown in Fig. 3, 1␣-25-(OH) 2 D 3 caused a synergistic increase in AP-1 binding to TRE in the TGF-␤1-treated cells in a dosedependent manner. The stimulated binding was completely inhibited by the unlabeled oligonucleotide containing TRE (data not shown).

Effect of VDR Expression on 1␣-25-(OH) 2 D 3 Synergy in c-jun Gene Expression in TGF-␤1-treated MC3T3-E1 Cells-To en-
sure that 1␣-25-(OH) 2 D 3 synergy toward TGF-␤1-induced AP-1 transcriptional activity in the cells is mediated through VDR, it is significant to understand the precise role of VDR in the hormone synergy toward c-jun gene expression in the cells. Therefore, we explored using a VDR expression vector (pSG-VDR) the functional role of endogenous VDR in this synergy. The cells were transfected with pSG-VDR or control vector pSG-5 and then pretreated or not with 1␣-25-(OH) 2 D 3 . Subsequently, the cells were treated or not with TGF-␤1, and c-jun gene expression was analyzed by the Northern blot assay. As shown in Fig. 6, A and

Synergistic Stimulation by 1␣-25-(OH) 2 D 3 of TGF-␤1-induced AP-1 Transcriptional Activity in MC3T3-E1 Cells Is Mediated via VDR-1␣-25-(OH) 2 D 3 synergy in TGF-␤1-induced
expression of the c-jun gene in the cells transfected with the VDR expression vector suggested to us the possibility that 1␣-25-(OH) 2 D 3 stimulates synergistically TGF-␤1-induced AP-1-transcriptional activity in the cells via VDR. Therefore, we explored this possibility by using a cotransfection system with VDR expression vector (pSG-VDR) and pTRE-TK-CAT. Fig. 7, A and B, shows that TGF-␤1-induced AP-1 transcriptional activity in pSG-VDR-transfected cells was approximately 1.5 times that in the cells transfected with pSG-5. In a CAT assay with the TK-CAT reporter plasmid (pTK-CAT), no effect of pSG-VDR or pSG-5 was observed (data not shown). These results indicate strongly that 1␣-25-(OH) 2 D 3 stimulation of TGF-␤1-induced AP-1 transcriptional activity in the cells occurs via VDR.

Involvement of Endogenous VDR in 1␣-25-(OH) 2 D 3 Synergy toward TGF-␤1-induced Expression of the c-jun Gene and AP-1
Transcriptional Activity in MC3T3-E1 Cells-As described above, since our data suggested involvement of an endogenous VDR in TGF-␤1-induced expression of the c-jun gene and in AP-1 transcriptional activity in the cells, in addition, we used mouse VDR antisense oligonucleotide to investigate the functional role of VDR in this synergy. Fig. 8A shows that VDR antisense, but not sense, oligonucleotide treatment clearly inhibited the synergistic action of 1␣-25-(OH) 2 D 3 in the cytokine- Quantification of c-jun mRNA level was done by densitometry and is expressed as a percentage of maximum. B, cells were transfected with a reporter plasmid (pTRE-TK-CAT) or control plasmid (pTK-CAT) and were washed three times and then incubated in serum-free ␣-MEM supplemented with or without 1␣-25-(OH) 2 D 3 or its analogs at 10 Ϫ8 M. After a 24-h incubation, the cells were treated or not for 6 h with TGF-␤1 (1 ng/ml). Thereafter, the cellular extracts were prepared and subjected to the CAT assay. All assays were performed in the presence of 1 g of pGL-TK, a luciferase expression plasmid, used as an internal control to normalize for variations in transfection efficiency. Quantification of the CAT activity was done by densitometry and is expressed as a percentage of maximum. An identical experiment independently performed gave similar results.

FIG. 6. Transfection with a VDR expression vector demonstrates receptor role in 1␣-25-(OH) 2 D 3 synergy in c-jun gene expression in TGF-␤1-treated MC3T3-E1 cells.
A, cells were transfected or not with the VDR expression vector (pSG-VDR) or control vector (pSG-5) and then were washed three times. Thereafter, the cells were incubated in serum-free ␣-MEM supplemented or not with 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M. After a 24-h incubation, the cells were washed and subsequently were treated or not with TGF-␤1 (1 ng/ml). Total RNA was prepared at 1.5 h after the initiation of the cytokine treatment. Northern blot analysis was performed with c-jun and ␤-actin cDNAs used as probes. B, quantification of c-jun mRNA level in A was done by densitometry and is expressed as a percentage of maximum.
induced expression of the c-jun gene in the cells.
Also, we examined involvement of the endogenous VDR in the cytokine-induced AP-1 transcriptional activity in the cells. The cells were transfected or not with pTRE-TK-CAT and then pretreated or not for 12 h with VDR sense or antisense oligonucleotide. Thereafter, in addition, the cells were pretreated or not for 24 h with 1␣-25-(OH) 2 D 3 , and subsequently treated or not with TGF-␤1. As shown in Fig. 8, B and C, the VDR antisense oligonucleotide inhibited the synergistic action of 1␣-25-(OH) 2 D 3 toward TGF-␤1-induced AP-1 transcriptional activity in the cells. However, its sense oligonucleotide was ineffective. These results suggested to us a functional role for endogenous VDR in the 1␣-25-(OH) 2 D 3 synergistic action in TGF-␤1-induced AP-1 transcriptional activity in MC3T3-E1 cells.
1␣-25-(OH) 2 D 3 Stimulates at the Transcriptional Level the AP-1-dependent Expression of the Osteopontin Gene in TGF-␤1treated MC3T3-E1 Cells-It is of interest to determine whether the 1␣-25-(OH) 2 D 3 synergism toward TGF-␤1-induced AP-1 transcriptional activity actually operates physiologically in osteoblastic cells. It is well known that osteopontin, an important matrix protein, is a marker of osteoblastic cell differentiation (50 -53). Although the data are not shown, we observed that TGF-␤1-stimulated expression of the osteopontin gene in MC3T3-E1 cells is mediated via AP-1. And, in fact, it has been demonstrated that a TRE sequence is located on the promoter region of the murine osteopontin gene (52,53). Therefore, finally, we investigated 1␣-25-(OH) 2 D 3 synergism toward TGF-␤1-stimulated expression of the osteopontin gene in the cells. As we expected, 1␣-25-(OH) 2 D 3 pretreatment at 10 Ϫ8 M stimulated expression of the osteopontin gene in the cytokine-treated cells (Fig. 9A), and a run-on assay (Fig. 9B) showed that 1␣-25-(OH) 2 D 3 operates at the transcriptional level in the synergistic action. Also, such synergism was observed with OCT pretreatment (Fig. 9C). These observations allow us to propose that 1␣-25-(OH) 2 D 3 actually stimulates AP-1-mediated differntiation of TGF-␤1-treated osteoblastic cells. DISCUSSION TGF-␤1 and 1␣-25-(OH) 2 D 3 are important local and systemic regulatory factors in bone remodeling (4 -7, 54 -56). For osteoblastic cells, both are potent factors in growth and differentiation (7-9, 20 -25). We previously demonstrated that TGF-␤1 strongly induces AP-1 in mouse osteoblastic MC3T3-E1 cells via the activation of serine/threonine kinase (8,9). Several studies (21-23) have documented that AP-1 regulates the differentiation of osteoblastic cells by interacting with VDR, which functions as a transcriptional factor. In general, 1␣-25- FIG. 7. Synergistic stimulation by 1␣-25-(OH) 2 D 3 of TGF-␤1-induced AP-1 transcriptional activity in MC3T3-E1 cells is mediated via VDR. A, cells were cotransfected with the reporter plasmid (pTRE-TK-CAT) and VDR expression vector (pSG-VDR) or control vector (pSG-5) and then were washed three times. The transfected cells were incubated in serum-free ␣-MEM supplemented or not with 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M. After a 24-h incubation, the cells were treated or not for 6 h with TGF-␤1 (1 ng/ml). Thereafter, the cellular extracts were prepared and subsequently subjected to the CAT assay. All assays were performed in the presence of pGL-TK, a luciferase expression plasmid, used as an internal control to normalize for variations in transfection efficiency. B, quantification of CAT activity in A was done by densitometry and is expressed as a percentage of maximum. An identical experiment independently performed gave similar results.

FIG. 8. Involvement of endogenous VDR in 1␣-25-(OH) 2 D 3 synergy toward TGF-␤1-induced AP-1 transcriptional activity in MC3T3-E1 cells.
A, cells were pretreated or not with for 12-h VDR antisense or sense oligonucleotide (10 M) and subsequently treated or not with 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M. After a 24-h incubation with the hormone, the cells were treated or not for 1.5 h with TGF-␤1 (1 ng/ml). Thereafter their total RNA was prepared. Northern blot analysis was performed with c-jun and ␤-actin cDNAs used as probes. An identical experiment independently performed gave similar results. Quantification of c-jun mRNA level was done by densitometry and is expressed as a percentage of maximum. B, cells were transfected with the reporter plasmid (pTRE-TK-CAT) or control plasmid (pTK-CAT) and were washed three times and subsequently pretreated or not for 12 h with VDR antisense or sense oligonucleotide (10 M). Then, the cells were incubated with or without 1␣-25-(OH) 2 D 3 at 10 Ϫ8 M. After a 24-h incubation with the hormone, the cells were treated for 6 h with TGF-␤1 (1 ng/ml). Afterward, the cellular extracts were prepared and subjected to the CAT assay. All assays were performed in the presence of pGL-TK. C, quantification of CAT activity in B was done by densitometry and is expressed as a percentage of maximum. An identical experiment independently performed gave similar results.
Several studies (10 -16, 21-23) have well documented that AP-1 in osteoblasts is an important transcriptional factor in bone formation and resorption. We observed here that TGF-␤1induced expression of the c-jun gene in MC3T3-E1 cells was synergistically stimulated by 1␣-25-(OH) 2 D 3 pretreatment, although such synergistic effect was not observed for c-fos gene expression. The synergistic action of 1␣-25-(OH) 2 D 3 toward c-jun gene expression was dose-and pretreatment-dependent. Our run-on assay indicated that the 1␣-25-(OH) 2 D 3 synergistic action in the c-jun gene expression in TGF-␤1-treated cells resulted from stimulation of its transcription. Since these observations suggested the synergistic increase by the hormone of AP-1 binding to TRE in the cells, we explored this point by the gel mobility shift assay. This assay clearly demonstrated that the cytokine-induced AP-1 binding to TRE in the cells was synergistically increased by 1␣-25-(OH) 2 D 3 pretreatment. Of more interest was whether the hormone would be actually able to stimulate synergistically AP-1 transcriptional activity. In fact, the TRE-TK-CAT assay showed that 1␣-25-(OH) 2 D 3 clearly stimulated the AP-1 transcriptional activity in TGF-␤1-treated cells. In this regard, Sassone-Corsi and co-workers (48,57) showed that AP-1 transcriptional activity in mouse embryonal carcinoma F9 cells, which express constitutively c-jun and c-fos genes at a low level, is induced by transfecting the cells with a c-jun autonomous expression vector. We (8,58) observed that curcumin, a potent inhibitor of c-jun gene expression, completely inhibited the synergistic effect of 1␣-25-(OH) 2 D 3 on TGF-␤1-induced AP-1 transcriptional activity in MC3T3-E1 cells (data not shown). These observations of vitamin D 3 support the notion that the stimulated expression of the c-jun gene provided an important clue in the mechanism of 1␣-25-(OH) 2 D 3 synergism toward TGF-␤1-induced AP-1 transcriptional activity in the cells.
It is well documented that multiple biological actions of 1␣-25-(OH) 2 D 3 in osteoblastic cells are mediated via interaction of its receptor complex with specific DNA sequences (20 -22, 24, 25). On the other hand, recent studies (27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37) have demonstrated that 1␣-25-(OH) 2 D 3 expresses several biological actions via ceramide and protein kinase C signaling pathways and also via intracellular calcium signals. Thus, it was of interest to demonstrate whether 1␣-25-(OH) 2 D 3 synergy in the stimulation of AP-1 transcriptional activity of TGF-␤1-treated MC3T3-E1 cells is genomic action-dependent or -independent. Therefore, we explored this point by using 1) 1␣-25-(OH) 2 D 3 analogs having high and low affinities to VDR, 2) transfection assay with VDR expression and TRE-TK-CAT vectors, and 3) VDR antisense oligonucleotide to eliminate production of endogenous VDR. We observed that OCT synergy in the expression of the c-jun gene in TGF-␤1-treated MC3T3-E1 cells was the same as that of 1␣-25-(OH) 2 D 3 . Since OCT is a ligand having high affinity for VDR (33,37), these observations suggested to us that 1␣-25-(OH) 2 D 3 synergy in the stimulation of AP-1 transcriptional activity of TGF-␤1-treated cells may be genomic in nature. We proved this suggestion by using a cotransfection assay with VDR expression and TRE-TK-CAT vec- (1 ng/ml). Total RNA was prepared at 3 h after the initiation of the cytokine treatment. Northern blot analysis was performed with osteopontin, and ␤-actin cDNAs were used as probes. Quantification of osteopontin mRNA level was done by densitometry and is expressed as a percentage of maximum. B, cells were treated under the conditions described in A. Then their nuclei were incubated for 30 min in the presence of 5Ј-[␣-32 P]UTP, after which the total RNA was isolated. Transcriptional activity assay (run-on assay) was performed with osteopontin and ␤-actin cDNAs. pBR322, the vector plasmid, was used as a negative control. Quantification of osteopontin run-on activity was done by densitometry and is expressed as a percentage of maximum. C, cells were incubated for 24 h in the presence or absence of 1␣-25-(OH) 2 D 3 or its analogues at 10 Ϫ8 M. Thereafter, the cells were washed and treated or not with TGF-␤1 (1 ng/ml). Then total RNA was prepared at 3 h after the initiation of the cytokine treatment. Northern blot analysis was performed with osteopontin, and ␤-actin cDNAs were used as probes. Quantification of osteopontin mRNA level was done by densitometry and is expressed as a percentage of maximum. An identical experiment independently performed gave similar results.
Several studies (59 -62) have shown that the combination of 1␣-25-(OH) 2 D 3 and TGF-␤1 functionally regulates bone formation. Our previous study (9) showed that TGF-␤1-induced expression of RAR-␣, RAR-␥, and retinoic X receptor-␣ is mediated via the cytokine-induced AP-1 signaling pathway in MC3T3-E1 cells. Most recently, we observed that the cytokineinduced expression of RAR-␣, -␥, and retinoic X receptor-␣ genes were remarkably stimulated by pretreatment with 1␣-25-(OH) 2 D 3 . 2 In addition, as we also showed in this study, the hormone stimulated synergistically TGF-␤1-induced expression of the osteopontin gene at the transcriptional level in the cells. Since osteopontin is one of the extracellular non-collagenous matrix proteins and also is a marker of differentiation of osteoblastic cells (20, 50 -53), these observations are significant with respect to bone metabolism.
In further experiments, our interest will be to define the mechanism of 1␣-25-(OH) 2 D 3 synergism operating in TGF-␤1induced AP-1 transcriptional activity in mouse osteoblastic cells. We are examining two possibilities as follows: one is that the hormone stimulates c-jun kinase activity, and the other one is stimulation by the hormone of TGF-␤1 receptor-associated Smads in a VDR-dependent manner.