Abstract
This study evaluated the effect of TAK-778 [(2R, 4S)-(−)-N-(4-diethoxyphosphorylmethylphenyl)-1,2,4,5-tetrahydro-4-methyl-7,8-methylenedioxy-5-oxo-3-benzothiepin-2-carboxamide)] on in vitro osteogenic events and on gene expression of osteoblastic cells derived from human alveolar bone and the participation of estrogen receptors (ERs) on such effect. Osteoblastic cells were subcultured, with or without TAK-778 (10−5 M), to evaluate cell growth and viability, total protein content, and alkaline phosphatase (ALP) activity at 7, 14, and 21 days; bone-like formation at 21 days; and gene expression, using cDNA microarray, at 7 days. Also, osteoblastic cells were exposed to TAK-778 (10−5 M) combined to ICI182,780, a nonspecific ER antagonist (10−6 M), and gene expression was evaluated by real-time polymerase chain reaction (PCR) at 7 days. TAK-778 induced a reduction in culture growth and an increase in cell synthesis, ALP activity, and bone-like formation. The cDNA microarray showed genes associated with cell adhesion and differentiation, skeletal development, ossification, and transforming growth factor-β receptor signaling pathway, with a tendency to be higher expressed in cells exposed to TAK-778. The gene expression of ALP, osteocalcin, Msh homeobox 2, receptor activator of NF-kappa B ligand, and intercellular adhesion molecule 1 was increased by TAK-778 as demonstrated by real-time PCR, and this effect was antagonized by ICI182,780. The present results demonstrated that TAK-778 acts at a transcriptional level to enhance the in vitro osteogenic process and that its effect on gene expression of osteoblastic cells is mediated, at least partially, through ERs. Based on these findings, TAK-778 could be considered in the treatment of bone metabolic disorders.
Introduction
Bone tissue has a remarkable ability to regenerate, and remodeling and numerous regulatory mechanisms are involved in the maintenance of bone mass (1). In a homeostatic equilibrium, tissue formation and resorption are balanced and new bone continuously replaces old bone so that it adapts to mechanical load and strain (2). However, pathological situations may occur as a consequence of an inappropriate or uncoupled balance between bone resorption and formation (3, 4). Because of this, and based on the increased understanding of osteoblast and osteoclast biology and the physiology of bone diseases, some inhibitors of bone resorption have been developed (5, 6). Considering that most therapies involving bone tissue are focused on bone resorption, drugs that effectively stimulate bone formation could have an attractive application as new therapy to be used as adjuvants of bone resorption inhibitors (4).
In this context, a novel series of 3-benzothiepin-2-carboxyamide derivatives was synthesized by structure-activity relationship studies and, among these, TAK-778 [(2R, 4S)-(−)-N-(4-diethoxyphosphorylmethylphenyl)-1,2,4,5-tetrahydro-4-methyl-7,8-methylenedioxy-5-oxo-3-benzothiepin-2-carboxamide)] was selected for further investigation of its effects on osteoblastic cells (7). It has been demonstrated that TAK-778, at concentrations ranging from 10−7 M to 10−5 M, enhances osteoblastic differentiation and bone-like nodule formation in both human and rat bone marrow mesenchymal stromal cell cultures (8, 9). Based on these outcomes, we conducted further experiments to investigate the mechanism of action of TAK-778. In these studies, we observed that its effect on osteoblastic differentiation is mediated through estrogen receptors (ERs) (10, 11).
Although the stimulatory effect of TAK-778 has been observed in different cell culture systems, its effect on cells derived from human alveolar bone fragments remains to be determined. In addition, until the present time, changes in gene expression profile induced by TAK-778 in osteoblastic cells have not been investigated. Considering these factors, and because alveolar bone is one of the most active bones in the human body (12), we used this cell culture model to evaluate the effects of TAK-778 on key parameters of the in vitro osteogenic process and on gene expression by means of cDNA microarray and real-time polymerase chain reaction (PCR). Furthermore, we investigated whether TAK-778 acts on gene expression via an ER-dependent pathway.
Materials and Methods
Chemicals.
TAK-778 was kindly supplied by Takeda Chemical Industries (Osaka, Japan) and ICI182,780, a nonspecific ER antagonist, was purchased from Tocris Cookson (Ellisville, MO). Both compounds were dissolved in a solution of ethanol-dimethylsulfoxide (1:1, v/v) and then diluted with culture medium to the final concentrations of 10−5 M to TAK-778 and 10−6 M to ICI182,780. These concentrations were based on previous studies carried out in our laboratory (8, 10). The 10−5 M TAK-778 concentration was most effective in stimulating in vitro osteogenic parameters and 10−6 M to ICI182,780 in inhibiting such effect.
Culture of Osteoblastic Cells.
Cells from human alveolar bone fragments were obtained from a total of 10 healthy donors, with cells from 4 donors used for evaluation of in vitro osteogenic events and real-time PCR and cells from 6 donors used for cDNA microarray analysis. This procedure followed the research protocols approved by the Committee of Ethics in Research from the University of Sao Paulo. Osteoblastic cells were released from these fragments by enzymatic digestion using type II collagenase (Gibco, Life Technologies, Grand Island, NY) as previously described (13). These cells were cultured in α-minimum essential medium (Gibco), supplemented with 10% fetal bovine serum (Gibco), 50 μg/ml gentamicin (Gibco), 0.3 μg/ml Fungizone (Gibco), 10−7 M dexamethasone (Sigma, St. Louis, MO), 5 μg/ml ascorbic acid (Gibco), and 7 mM β-glycerophosphate (Sigma). Subconfluent cells in primary culture were harvested after treatment with 1 mM EDTA (Gibco) and 0.25% trypsin (Gibco) and subcultured as described below. All cell cultures used in this study were maintained at 37°C in a humidified atmosphere of 5% CO2 and 95% air, and the medium was changed every 3 or 4 days.
In Vitro Osteogenic Events.
To evaluate the in vitro osteogenic parameters, cells were subcultured in 24-well culture plates (Falcon, Franklin Lakes, NJ) at a cell density of 2 × 104 cells/well in 2 distinct culture media containing (1) TAK-778 (10−5 M) and (2) the same volume of vehicle (control group) for periods of up to 21 days.
Culture Growth and Cell Viability.
To evaluate culture growth and cell viability, cells were subcultured for 7, 14, and 21 days. After these periods, cells were enzymatically released from the wells using 1 mM EDTA (Gibco), 1.3 mg/ml collagenase (Gibco), and 0.25% trypsin (Gibco). Viable and nonviable cells were detected by trypan blue (Sigma) and counted using a hemocytometer (Housser Scientific Company, Horsham, PA). Culture growth was expressed as number of cells × 104/well, and cell viability was expressed as a percentage of the viable cells.
Total Protein Content.
Total protein content was measured at 7, 14, and 21 days, according to the Lowry et al. (14) method. The culture medium was removed and the wells were filled with 2 ml of 0.1% sodium lauryl sulphate (Sigma). After 30 mins, 1 ml of the cell lysate from each well was mixed with 1 ml of Lowry solution (Sigma) and left at room temperature for 20 mins. Subsequently, it was added to 0.5 ml of the solution of phenol reagent of Folin and Ciocalteau (Sigma). This stood for 30 mins at room temperature to allow color development, and the absorbance was measured spectrophotometrically (CE3021–Cecil, Cambridge, UK) at 680 nm. The total protein content was calculated from a standard curve and expressed as μg/ml; data were normalized by the number of cells counted at 7, 14, and 21 days.
ALP Activity.
ALP activity was evaluated at 7, 14, and 21 days by the release of thymolphthalein from thymolphthalein monophosphate, using a commercial kit (Labtest Diagnostica SA, MG, Brazil). Briefly, 50 μl of thymolphthalein monophosphate was mixed with 0.5 ml of 0.3 mM diethamine buffer (pH 10.1) and left for 2 mins at 37°C. After this, 50 μl of the same lysates used for measuring total protein content were added. This was then left for 10 mins at 37°C, after which 2 ml of a solution of 0.09 mM Na2CO3 and 0.25 mM NaOH were added to allow color development. After 30 mins, absorbance was measured at 590 nm and ALP activity was calculated from a standard curve using thymolphthalein, giving a range from 0.012 to 0.40 μmol thymolphthalein/hrs/ml. Data were expressed as ALP activity normalized by the number of cells counted at 7, 14, and 21 days.
Bone-Like Formation.
The bone-like formation was evaluated at 21 days. The culture medium was removed and the attached cells were fixed with 10% formalin for 24 hrs at 4°C. After this period, the specimens were dehydrated through a graded series of alcohol and processed for staining with Alizarin Red S (Sigma), in order to detect calcium deposits. The specimens were evaluated using an image analyzer (Image Tool—University of Texas Health Science Center, San Antonio, TX) and the amount of bone-like formation was calculated as a percentage of total well area.
Total RNA Extraction.
At Day 7, total RNA was extracted from the cells to be used in both complementary DNA (cDNA) microarray and quantitative real-time PCR assays, using Trizol reagent (Gibco) according to the manufacturer’s instructions. Total RNA concentration was determined by optical density at a wavelength of 260 nm and its integrity was evaluated by denaturing agarose gel electrophoresis under standard conditions.
cDNA Microarray.
To investigate the effect of TAK-778 on gene expression of osteoblastic cells, we performed cDNA microarray analysis in cells subcultured in 75 cm2 culture flasks (Corning Inc., NY) at a cell density of 8 × 104 cells/flask in culture medium containing (1) TAK-778 (10−5 M) and (2) the same volume of vehicle (control group). At 7 days, the total RNA was extracted as described above.
A microarray containing a total of 687 cDNA sequences was used and spotted in duplicate in the form of PCR products on 2.5 × 7.5 cm Hybond N+ nylon membrane (GE Healthcare, UK). The arrays were prepared using a Generation III Array Spotter (Amersham Molecular Dynamics, Sunnyvale, CA) and the 3′ expressed sequence tag cDNA clones were obtained from the I.M.A.G.E. Consortium (Huntsville, AL) (http://image.llnl.gov/image/). The cDNA inserts were homogeneous in size (near 1 kb) and cloned in 3 vectors (pT7T3D, pBluescript, and Lafmid) and were amplified in 384- or 96-well plates using vector-PCR amplification with the following primers, which recognize the 3 vectors: LBP 1S GTGGAATTGTGAGCGGATACC forward and LBP 1AS GCAAGGCGATTAAGTTGG reverse. This set of cDNAs included sequences involved in cell adhesion, proliferation and differentiation, cell cycle regulation, apoptosis, skeletal development, and bone mineralization. The membranes were first hybridized with LBP 1AS (γ-33P) dCTP-labeled oligonucleotide (vector hybridization) and quantification of the signals allowed to estimate the amount of cDNA insert fixed in each spot. After stripping, the membranes were used for hybridization with cDNA complex probes (sample hybridizati on). The characterization of each cDNA sequence in terms of molecular and biological functions and clone ID was updated with web data banks GenBank (www.ncbi.nlm.nih.gov) and S.O.U.R.C.E. (http://genome-www5.stanford.edu).
cDNA Microarray Complex Probe Preparation and Hybridization.
In this study, we refer to the radioactive cDNA originated from the cell culture RNA samples as the “complex probe” and the PCR product originated from the clones and deposited on the nylon microarrays as “targets.” The 33P-labeled cDNA complex probes were prepared by reverse transcription of 10 μg of total RNA, using dT12–18 as a primer. An aliquot of 100 μl of 33P-cDNA complex probe containing about 30–50 million cpm was hybridized with targets spotted on nylon microarrays as previously described (15–17).
cDNA Microarray Imaging Acquisition, Data Quantification, Normalization, and Analysis.
Imaging plates and a phosphor imager (Phosphor Imager, model Cyclone, Packard Instruments, Meridien, CT) were used to capture the hybridization signals and BZScan software (http://tagc.univmrs.fr/ComputationalBiology/bzscan/index.php) (18) was used to quantify the signals with local background subtraction, whose spots matched a template grid. A mathematical correction (normalization) was applied to permit a direct comparison among independent hybridizations. Data were normalized by dividing the individual hybridization value of each spot (cDNA target on the microarray) by the median calculated from all of the hybridization values obtained for the whole array. The mean number of normalized hybridization values for both sets of cDNA clones was also calculated. For data analysis, we used a hierarchical clustering algorithm, comparing means of different genes whose standard deviations do not overlap; our objective was to compute a dendrogram that assembles all elements into a single tree. The software for this algorithm can be obtained from the authors Eisen et al. (19) at http://rana.stanford.edu/clustering. Before clustering, the gene expression values were median-centered and log transformed, using the Pearson correlation distance metrics and average linkage for clustering organization from 6 determinations of each situation studied.
Quantitative Real-Time PCR.
The participation of ERs on the effect of TAK-778 on gene expression of osteoblastic cells was investigated in cells subcultured in 25-cm2 culture flasks (Corning Inc.) at a cell density of 2.6 × 104 cells/flask in culture medium containing (1) TAK-778 (10−5 M), (2) TAK-778 (10−5 M) + ICI182,780 (10−6 M), (3) ICI182,780 (10−6 M), and (4) the same volume of vehicle (control group). At 7 days, the total RNA was extracted as described above.
Synthesis of cDNA was carried out using 2 μg of total RNA through a reverse transcription reaction (M-MLV reverse transcriptase, Promega, Madison, WI). Analyses of mRNA were performed in an ABI Prism 7000 Sequence Detection System, using the SybrGreen system (Applied Biosystems, Warrington, UK). SybrGreen PCR MasterMix (Applied Biosystems), specific primers, and 2.5 ng of cDNA were used in each reaction. The standard PCR conditions were 95°C (10 mins) and 40 cycles of 94°C (1 min), 56°C (1 min), and 72°C (2 mins), followed by the standard denaturation curve. The sequences of human primers were designed using the Primer Express software (Applied Biosystems). Primer sequences, predicted amplicon sizes, and the annealing (Ta) and melting (Tm) temperatures are depicted in Table 1. For mRNA analysis, the relative level of gene expression was calculated in reference to both β-actin expression in each sample and its respective control, using the cycle threshold method (20).
Statistical Analysis.
All experiments were done in quintuplicate, with exception of real-time PCR, which was conducted in triplicate. Data related to in vitro osteogenic parameters were submitted to the nonparametric Mann-Whitney U test for independent samples (level of significance: 5%). Real-time PCR data were submitted to the nonparametric Kruskal-Wallis test for independent samples (level of significance: 5%), followed by the Fisher’s least significant difference multiple comparisons procedure.
Results
In Vitro Osteogenic Events.
Cell number was not affected at 7 days (P > 0.05), but was reduced (P < 0.05) by TAK-778 at 14 and 21 days (Fig. 1A). Cell viability was not affected (P > 0.05) by TAK-778 in all evaluated periods (Fig. 1B). Total protein content and ALP activity were higher (P < 0.05) in cultures exposed to TAK-778 in all evaluated periods (Figs. 1C and D). At 21 days, bone-like formation was increased (P < 0.05) by TAK-778 (Fig. 1E).
cDNA Microarray.
The complete color heat-map is presented in Fig. 2A. The hierarchical cluster analysis for a total of 687 genes demonstrated a different gene expression profile by comparing TAK-778 and the control group. The dendrogram showed that this differential gene expression profile allowed the distinction of samples from both groups (Fig. 2B). In addition, it was possible to identify several clusters of genes with a tendency to be induced and repressed; 4 of these (Fig. 2B), which included genes related to osteoblastic differentiation process, were selected for discussion. Among the genes differentially expressed in the 4 clusters, 13 genes are associated with cell adhesion and differentiation, skeletal development, ossification, and transforming growth factor-β receptor (TGF-β) signaling pathway (Table 2, Refs. 21–32).
Quantitative Real-Time PCR.
All data related to quantitative real-time PCR analysis are presented in Figure 3. The results showed that TAK-778 induced a significant increase (P < 0.05) in the gene expression of ALP, osteocalcin (OC), Msh homeobox 2 (MSX-2), and receptor activator of NF-kappa B ligand (RANKL) (Fig. 3A–D); a nonsignificant increase (P > 0.05) in the gene expression of intercellular adhesion molecule 1 precursor (ICAM-1) (Fig. 3E); and no effect (P > 0.05) on runt-related transcription factor 2 (Runx2) (Fig. 3F) and osteoprotegerin (OPG) (Fig. 3G). This effect was abolished (P < 0.05) in cultures exposed to TAK-778 in the presence of ICI182,780, with gene expression of ALP, OC, and MSX-2 being lower (P < 0.05) than the control (Fig. 3A–D). In addition, ICI182,780 itself had no effect (P > 0.05) on gene expression.
Discussion
In the present study, we investigated the effects of TAK-778 on key parameters of the in vitro osteogenic process and on gene expression of osteoblastic cells derived from human alveolar bone fragments, including the participation of ERs on such effects. The results showed that TAK-778 enhances the osteogenic potential and changes the expression of several genes related to cell adhesion and differentiation, skeletal development, ossification, and the TGF-β signaling pathway of these cells. In addition, we have demonstrated that the stimulatory effect of TAK-778 on the expression of some genes’ encoding markers of the osteoblastic phenotype is mediated through ERs.
Regarding the effect of TAK-778 on in vitro osteogenic events, our results are in agreement with those of previous studies, which have shown that TAK-778 induces a decrease in culture growth without affecting cell viability and an increase in protein synthesis, ALP activity, and bone-like formation in cultures of cells derived from human bone marrow and rat calvaria bone (9, 11). As a reciprocal relationship between the decrease in cell proliferation and the subsequent induction of cell differentiation has been proposed (33), the decrease in cultures with growth induced by TAK-778 that we observed in this study may be related to the progression of cell differentiation into mature osteoblasts (8). Based on these findings, it is possible to suggest that TAK-778 stimulates osteogenic events, regardless of the cell source and stage of the osteoblastic differentiation, since its effect has been demonstrated in cultures from different species (e.g., human and rat) and sites clearly exhibiting cells with different degrees of differentiation (e.g., bone marrow, alveolar, and calvaria bone).
The parameters evaluated here take part in a complex sequence of events that has been called osteogenesis, which ultimately results in formation of new bone tissue (34). The transcriptional program that controls osteogenesis is based on an integrated cascade of gene expression, which initially supports cell adhesion and proliferation and afterward supports synthesis, organization, and mineralization of the extracellular matrix (reviewed in Refs. 35–37). Considering this and the osteogenic potential of TAK-778, we investigated whether gene expression of osteoblastic cells would be altered by TAK-778 exposition. The cDNA microarray technique allowed simultaneous analysis of the expression of several genes related to cell adhesion, proliferation, and differentiation; cell cycle regulation; apoptosis; skeletal development; and bone mineralization. With this analysis, remarkable transcriptional changes were noticed in cells exposed to TAK-778. Based on these outcomes and on the biological functions of the genes, we selected 4 clusters of genes that presented a tendency to be higher expressed in cells exposed to TAK-778 compared with cells in control cultures, which are shown in Table 2. Higher expression of these genes as demonstrated here indicates that the osteogenic potential of TAK-778 is related to its action at a transcriptional level in osteoblastic cells.
The effect of TAK-778 on the development of osteoblastic phenotype at a transcriptional level was confirmed by real-time PCR that demonstrated a higher expression of the genes of ALP, OC, MSX-2, and RANKL, compared with control cultures. The expression of ICAM-1 was also upregulated, although in a nonsignificant way, and Runx2 and OPG were not affected by TAK-778. Our results corroborate those reported by Gotoh et al. (38) that observed the same effect of TAK-778 on gene expression of ALP, OC, MSX-2, and Runx2 in cultures of rat bone marrow cells. With consideration that Runx2 is a prerequisite transcription factor for osteoblastic differentiation and that TAK-778 did not change its gene expression, it is possible to suggest that TAK-778 acts only on cells already committed to osteoblastic phenotype. In agreement with this, Notoya et al. (9) demonstrated that the effect of TAK-778 on cells from bone marrow is dependent on dexamethasone, which plays an essential role in the development of osteoblastic phenotype in cultures of mesenchymal stem cells (39). Regarding ICAM-1, RANKL, and OPG, an interaction among them is essential to induce osteoclastogenesis (40) and, because of this, it is possible to speculate that beyond its effect on bone formation, TAK-778 could activate the osteoclast formation pathway. The effect of TAK-778 on OPG and RANKL gene expression differs from that of 17β-estradiol, which increases OPG and represses RANKL expression (41). That this effect of TAK-778 is due to its action on ERs is confirmed by the fact that the increase in RANKL gene expression was abolished by ICI182,780, a nonspecific ER antagonist. Thus, the discrepancy between TAK-778 and 17β-estradiol may be related to differences in either their action on ERs or on the cell culture systems used to evaluate their effects. In agreement with the latter, it has been demonstrated that ER agonists produced different gene expression fingerprints, depending on the cell line studied (42).
Recently, we have shown that TAK-778 stimulates the expression of markers characteristic of the osteoblastic phenotype (i.e., ALP activity and bone-like formation) in human bone marrow mesenchymal cells, at least in part via an ER-dependent pathway (10, 11). In this study, further experiments were carried out and we demonstrated that the transcriptional changes induced by TAK-778, which ultimately increase new bone formation, are mediated through ERs, since ICI182,780 inhibited the stimulatory effect of TAK-778 on the expression of genes of RANKL and ICAM-1 and reduced the gene expression of ALP, OC, and MSX-2 below that of the control. In our previous study, we observed a statistically nonsignificant reduction of ALP activity in cell cultures treated with TAK-778 and ICI182,780 (10). Here, this finding was confirmed by the more-sensitive real-time PCR method, suggesting that the predominant effect of TAK-778 is ER mediated; however, when classical ER signaling pathways are blocked, TAK-778 may trigger mechanisms with opposite effects.
In conclusion, although the osteogenic potential of TAK-778 has been documented in different systems of cell culture, this is the first experimental evidence showing that TAK-778 acts at a transcriptional level to enhance the in vitro osteogenic process in cultures of osteoblastic cells derived from human alveolar bone fragments. In addition, we demonstrated that the effect of TAK-778 on gene expression of osteoblastic cells is mediated at least partially through ERs. In light of these observations, such compounds may be considered in the treatment of bone metabolic disorders.
Primer Sequences and Reaction Properties
Selected Genes with a Tendency To Be Higher Expressed in Osteoblastic Cells Derived from Human Alveolar Bone Fragments Exposed to TAK-778 (10−5 M) at Day 7
Effect of TAK-778 on in vitro osteogenic events. Culture growth (A), cell viability (B), total protein content (C), and ALP activity (D) evaluated at 7, 14, and 21 days, and bone-like formation (E) evaluated at 21 days in subcultures of osteoblastic cells derived from human alveolar bone fragments exposed to or not exposed to TAK-778 (10−5 M). Data are reported as mean ± SD (n = 5). The asterisks (*) indicate P < 0.05 for comparisons between TAK-778 and the control group at the same time point.
Effect of TAK-778 on gene expression of osteoblastic cells derived from human alveolar bone fragments evaluated by cDNA microarray at Day 7. Heatmap showing the gene expression profile of cells subcultured with or without TAK-778 (10−5 M) (A). Dendrogram showing the distinction of samples from TAK-778 and control groups and the 4 clusters of genes selected to be presented. (B). Red spots indicate upregulation, green spots indicate downregulation, black spots indicate absence of modulation, and gray spots indicate absence of values. A color version of this figure is available in the online journal.
Participation of ERs on the effect of TAK-778 on gene expression of osteoblastic cells derived from human alveolar bone fragments evaluated by quantitative real-time PCR at Day 7. Effect of TAK-778 (10−5 M), TAK-778 (10−5 M) + ICI182,780 (10−6 M), and ICI182,780 (10−6 M) on ALP (A), OC (B), MSX-2 (C), RANKL (D), ICAM-1 (E), Runx2 (F), and OPG (G) mRNA expression. Data were calculated as the expression of the target mRNAs normalized to β-actin and to its respective control and are reported as mean ± SD (n = 3). Asterisks (*) indicate P < 0.05 for comparisons between the respective groups and the control.
Footnotes
We thank the State of Sao Paulo Research Foundation (FAPESP) and the National Council for Scientific and Technological Development (CNPq) for their financial support.
Acknowledgements
We are grateful to Junia Ramos, Fabíola S. de Oliveira, and Roger R. Fernandes for their helpful assistance during the experiments.