Glucocorticoid Receptor and Sequential P53 Activation by Dexamethasone Mediates Apoptosis and Cell Cycle Arrest of Osteoblastic MC3T3-E1 Cells

Glucocorticoids play a pivotal role in the proliferation of osteoblasts, but the underlying mechanism has not been successfully elucidated. In this report, we have investigated the molecular mechanism which elucidates the inhibitory effects of dexamethasone on murine osteoblastic MC3T3-E1 cells. It was found that the inhibitory effects were largely attributed to apoptosis and G1 phase arrest. Both the cell cycle arrest and apoptosis were dependent on glucocorticoid receptor (GR), as they were abolished by GR blocker RU486 pre-treatment and GR interference. G1 phase arrest and apoptosis were accompanied with a p53-dependent up-regulation of p21 and pro-apoptotic genes NOXA and PUMA. We also proved that dexamethasone can’t induce apoptosis and cell cycle arrest when p53 was inhibited by p53 RNA interference. These data demonstrate that proliferation of MC3T3-E1 cell was significantly and directly inhibited by dexamethasone treatment via aberrant GR activation and subsequently P53 activation.


Introduction
Glucocorticoids (GCs) are the most frequently used antiinammatory and immunosuppressive drugs in clinic to treat a variety of diseases including inflammation, cancer, and autoimmune disorders [1]. However, it has been found that prolonged and/or overdose GCs treatment is the most common cause of osteonecrosis [2] and the third most common cause of osteoporosis [3]. It has been reported that GCs could induce apoptosis of osteoblasts and inhibit its proliferation, thus leading to osteoporosis and osteonecrosis [4]. However, the molecular mechanism of GCs involved in apoptosis and proliferation inhibition of osteoblast is still poorly understood.
The effects of GCs are primarily considered to be mediated by cytosolic glucocorticoid receptor (GR) activation [5], but the events leading from the activated GR to growth arrest are not yet elucidated completely. Previous studies have reported that GCs treatment induce osteoblast apoptosis by enhancing the expression of BH3-only protein Bim [6], down-regulation of TIMP-1 [7], and activation of glycogen synthase kinase 3 beta (GSK-3b) [8]. But to the best of our knowledge, there is no direct relationship between GR and these proteins, such as transcription-control or proteinprotein interaction. We reviewed previous studies in terms of the relationship between GR activation and apoptosis, and these studies has reported that p53 [9], granzyme A [10][11] or Glucocorticoid-induced leucine zipper (GILZ) [12][13] may be the downstream molecules of GR activation.
We postulate that GC activates GR, and then leads to activation of P53, granzyme A or GILZ, thereby inducing osteoblasts apoptosis and cell cycle arrest. The results of this study indicate that GR activation indeed up-regulates the expression of P53 and its downstream molecule, which results in growth inhibition.

Cell Culture
The murine osteoblastic cell line MC3T3-E1 was obtained from American Type Culture Collection (ATCC, Rockville, MD, USA).
Cells were cultured in a-MEM (Gibco BRL, Gaithersburg, MD, USA) supplemented with 10% FBS, 20 mM HEPES, 100 U/ml penicillin, 100 mg/ml streptomycin, and 50 mg/ml ascorbic acid Cells were incubated in a humid incubator at 37uC (95% O 2 and 5% CO 2 ) and maintained in a subconuent state unless otherwise indicated.

Cell Proliferation Assay
The MC3T3-E1 cells were inoculated at 2610 3 cells per well in 96-well plates. To assess the effects of dexamethasone on cell proliferation, the cells were incubated in growth medium or conditioned medium for 24 h at the concentration gradient from 0.001 mM to 10 mM. The sample cells were quantified using WST-8 assay, according to the instructions. Briefly, 10 ml of 2-(2methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (WST-8) solution reagent was added to 100 ml of culture medium in each well. After incubation for 2.5 h at 37uC, The absorbance of each well was read at a wavelength of 450 nm on a microplate reader. The measurements were represented by the means of at least three independent experiments, with each data point based upon three replicates.

Cell Death Detection
The MC3T3-E1Cells were seeded subconfluently into 12-well plates and treated with conditioned medium for 24 h at the concentration gradient from 0.001 mM to 10 mM. Cells were trypsinized and pelleted with cellular supernatant for 5 min at 400 g. After pellet was resuspended in 60 mL media, cell death rates were determined by counting cells using a hemocytometer after addition of Trypan blue, which stained the cytoplasm of dead cells but not live cells. Cell death rate (%) = number of dead cells/ number of total cells (6100%).

Cell Cycle Analysis
Cells were analyzed for their cell cycle distribution by flow cytometry. After being treated with dexamethasone for 48 h, the adherent cells were washed once with PBS, then trypsinized, and collected by centrifugation at 400 g for 5 min. The cells (106 cells per sample) were fixed in 4 ml of cold 70% ethanol at 220uC overnight. After centrifugation at 4006g for 5 min, cell pellets were incubated with 0.5 ml of PBS containing 100 g/mml RNase and 5 mg/ml propidium iodide (PI) at 37uC for 30 min. Cell cycle distribution was analyzed by measuring DNA content using a flow cytometer (Beckman Coulter, Inc.Fullerton, CA).

Assessment of Apoptosis
FITC Annexin V apoptosis determining kit was used to quantitatively determine the percentage of cells within a population that are actively undergoing early apoptosis per manufacture's instruction. Briefly, the cells were collected, washed twice with Annexin binding buffer and incubated with Annexin V-FITC and propidium iodide under the manufacturer's recommended conditions. Cells that stain positive for FITC Annexin V and negative for PI were considered undergoing early apoptosis.
TUNEL assay was performed using the in situ cell death detection kit according to the manufacturer's instructions. Briefly, MC3T3-E1 cells were plated onto coverslips and cultured overnight. After treated with various stress stimuli for 24 hours cells were fixed in 4% paraformaldehyde for 30 min at room temperature and permeabilized in 0.1% Triton X-100 for 2 min at 4uC. Cells were then stained with fluorescein isothiocyanate (FITC)-conjugated terminal deoxynucleotidyl TUNEL reaction mixture for 1 h at room temperature and incubated with DAB for 2 min. All dilutions were performed in PBS, and samples were analyzed and recorded by an OLYMPUS FLUOVIEW microscope.

Western Blot Analysis
Whole-cell extracts were prepared using Triton lysis buffer (50 mM Tris-HCl, pH 8.0 containing 150 mM NaCl, 1% Triton X-100, 0.02% sodium azide, 10 mM EDTA, 10 lg/ml aprotinin, and 1 lg/ml aminoethylbenzenesulfonyl fluoride). After the lysates were centrifuged for 12 min at 3,000 g to remove debris, protein concentrations were determined using the Bradford protein assay. 20-40 micrograms of protein from each cell layer homogenate was loaded onto a 10% or 12% polyacrylamide gel and transferred to a PVDF membrane. The membranes were blocked with 5% nonfat milk and incubated with primary antibodies. After extensive washing, the membranes were re-probed with peroxidase conjugated secondary antibody. Blots were processed using an ECL kit and exposed to film Controls for protein loading were made by bactins as the internal standard.

Statistical Analysis
All data were expressed as means 6 SD. The differences between groups was analyzed by one-way analysis of variance followed by Bonferroni's multiple comparisons test using SPSS 18.0 for windows. P-value (P) ,0.05 was considered statistically significant.

Dexamethasone Inhibit MC3T3-E1 Proliferation and Induce the Cell Death
The 24-hour exposure of murine osteoblast MC3T3-E1 to 0.001, 0.01, 0.1,1.10 mmol/L DEX respectively decreased cell proliferation in a concentration-dependent manner ( Figure 1A) Compared with ethanol control and low concentration dexamethasone, 1 and 10 mmol/L DEX reduced cell proliferation by 55.0164.01% and 60.0266.02%, respectively. To further confirm whether DEX toxicity inhibits cell proliferation, we used Typan blue incorporation to test dead cells. Typan blue assay suggested that treatment with 1 and 10 mmol/L dexamethasone remarkably increased dead cell population to 6.1060.00% and 7.7660.01%, respectively ( Figure 1B).

DEX-induced MC3T3-E1apoptosis and G0/G1 Arrest are Abolished by RU486 Pre-treatment
To determine whether the inhibitory effect can be attributed to apoptosis and cell cycle arrest, and whether the effect results from GR activation, we treated MC3T3-E1 cells with PBS (control group), 1 mmol/L dexamethasone (DEX group), RU486 (RU486 group), and 1 mmol/L dexamethasone plus 10 mmol/L RU486 (DEX+RU group) respectively. The apoptotic cells to total cells ratio was significantly (P,0.05) higher in DEX group (12.160.04%) as compared to 3.0160.04% in the control group, 3.5460.05% in the RU486 group, and 5.3660.03% in the DEX+RU group. (Figure 2 A) The level of the cleaved-caspase 3 protein was significantly up-regulated in the DEX group, when compared to that in other groups. (P,0.05) (figure 2B). By adding RU486 2 hours prior to dexamethasone treatment, we blocked the GR and the apoptosis effect associated with GR activation.
To assess the change of cell cycle induced by dexamethasone in MC3T3-E1 cells, the cell nuclei were stained with propidium iodide and the cell cycle was analyzed through flow cytometry. The percentage of cells in the G0-G1 phase increased significantly (P,0.05) to 84.6761.61% in the DEX group, as compared to 64.6762.65% in the control group, 68.5.62.65% in the RU486 group, and 68.0060.76% in the DEX+RU group, respectively. The percentage of cells in the S phase was decreased significantly (P,0.05) to 8.9061.00% in the DEX group as compared to 27.6760.58% in the control group, 24.0061.00% in the RU group, and 23.8361.26% in the DEX+RU group, respectively. (Figure 2

C).
GRa Gene Silencing Inhibited the G0-G1 Arrest and Apoptosis Induced by DEX To investigate GRagene function, we tried to silence this gene in MC3T3-E1 cells using siRNA targeting GRamRNA. The elimination of the GRamRNA and protein in MC3T3-E1 cell cultures was achieved as determined by real-time PCR (Fig. 3A) and Western blotting (Fig. 3B), respectively. When compared to those in untreated and siRNA control (siC) group, the mRNA and protein expression level of GRa decreased significantly (P,0.05) in the siGR-1 and siGR-2 group. On the other hand, the b-actin did not vary significantly between the groups.
To further determine the role of GR in DEX-induced apoptosis and cell cycle, we treated MC3T3-E1 cells with PBS (control group), 1 mmol/L DEX (DEX group), siGR-1 (siGR-1 group), 1 mmol/L DEX plus siGR-1 (DEX+siGR-1 group) respectively. Cleaved Capase-3 was used to characterize apoptosis of these groups. Western blot show DEX can significantly activate caspase-3 and then up-regulate cleaved Capase-3 expression, this effect can be reversed by gene silencing of GRa ( Figure 3C). TUNEL staining was also performed which shows apoptotic cells increased in DEX group, this effect can abolisded by GR geen siliencing as demonstrated in the DEX+siGR-1 group (Fig. 3D). The apoptotic cells to total cells ratio was significantly (P,0.05) higher in DEX group (12.3460.65%) as compared to 4.1160.08% in the control group, 6.2360.27% in the siGR-1 group, and apoptotic cell ratio decreased significantly when GRawas silenced (6.9460.37%).
The percentage of cells in the G0-G1 phase increased significantly (P,0.05) to 81.1464.77% in the DEX group as compared to 67.2663.65% in the control group, 68.462.65% in the siGR-1 group, and 69.3460.76% in the DEX+siGR-1 group. The percentage of cells in the S phase was decreased significantly (P,0.05) to 11.0960.86% in the DEX group as compared to 24.7461.58% in the control group, 21.7562.00% in the siGR-1 group, and 20.8361.13% in the DEX+siGR-1 group. ( Figure 3E).

p53 but not Granzyme A or GILZ is Up-regulated by DEX Treatment
To identify the downstream signaling pathways of GR activation, we performed Western blot analysis of proteins affected by GR activation such as granzyme A, GILZ and p53,respectively. This was done when MC3T3-E1 cells was treated with 0, 0.001, 0.01, 0.1, 1.0 and 10.0 mmol/L DEX, respectively. The levels of granzyme A and GILZ did not increased, but the expression of P53 was dose-dependently up-regulated ( Figure 4A). These results suggest that P53 is a mediator of GR activation and growth inhibition.
To further confirm that and identify downstream signaling pathways of P53, we performed western blot analysis of apoptosis and cell cycle associated proteins NOXA, PUMA and p21. In the DEX group, the level of the p53 protein was up-regulated, when compared with that of the control group, siGR-1 group and siGR-1+DEX group, which was consistent with the previous results. The level of the NOXA, PUMA and P21 protein was also up-regulated in the DEX group which suggested that P53 is a key molecular between GR activation and NOXA, PUMA and P21 upregulation. For comparison, b-actin protein was used as an internal reference, which did not vary remarkably. (Figure 4B).

p53 Gene Silienc by siRNA can Reverse DEX Induced Apoptosis and Cell Cycle Arrest of MC3T3-E1 Cells
To investigate the p53 gene function, we tried to silence this gene in MC3T3-E1 cells with siRNA targeting p53 mRNA. The knockdown of the p53 mRNA in MC3T3-E1 cell cultures was achieved as verified by RT-PCR and western blot (Fig. 5A, B). The protein expression level of p53 decreased significantly  We then treated MC3T3-E1 cells with PBS(control group ), 1 mmol/L dexamethasone(DEX group ), MC3T3-E1 cell silencing the p53 Gene (sip53-1 group), 1 mmol/L dexamethasone plus siRNA (DEX+sip53-1 group) respectively. Cleaved Capase-3 was used to characterize apoptosis of these groups. Western blot show DEX can significantly up-regulate cleaved Capase-3 expression, this effect can be reversed by gene silencing of p53 ( Figure 5C). TUNEL staining was also performed which shows apoptotic cells increased in DEX group, but when p53 was silenced apoptotic cells decreased significantly (Fig. 5D).The apoptotic cells to total cells ratio was significantly (P,0.05) higher in DEX group (13.1260.75%) as compared to 5.1360.17% in the control group, 5.7460.56% in the sip53-1 group, and this effect can abolisded by p53 geen siliencing ( Figure 5E).
The percentage of cells in the G0-G1 phase increased significantly (P,0.05) to 81.4463.56% in the DEX group as compared to 66.3662.44% in the control group, 67.5163.54% in the siGR-1 group,and 70.6861.67% in the DEX+siGR-1 group. The percentage of cells in the S phase was also decreased significantly (P,0.05) to 9.5461.07% in the DEX group as compared to 23.7360.97% in the control group, 26.5761.56% in the siGR-1 group, and 22.3461.34% in the DEX+ siGR-1 group. ( Figure 5E).

Discussion
In this study, we have illustrated for the first time that GR activation by dexamethasone up-regulates P53 in MC3T3-E1 osteoblast cell line, thereby enhancing the transcriptional activatity of P53. And it ultimately induces MC3T3-E1 cells apoptosis and cell cycle arrest.
It has been reported that GCs have potent inhibitory effects on osteoblast proliferation [4]. Earlier research studies have suggested that GR activation play a pivotal role in GCs induced apoptosis [11]. In this study, we found that dexamethasone treatment not only induces MC3T3-E1 apoptosis but also arrest G1 cell cycle, which was also dependent on GR activation. We demonstrated that dexamethasone treatment can increase apoptosis more than two fold, which was consistent with previous studies [11]. However, Smith disclosed that dexamethasone arrest MC3T3-E1 at G2/M phase [14][15], which did not comply with our results. This may be because the culture condition in our experiments differs from that in the previous study.
The effect of GR activation by glucocorticoids has long been a research focus. It is well accepted that the effect and molecular mechanism of GR activation is cell context specific. For example, it has been reported that GR activation by dexamethasone promotes survival and inhibits apoptosis of mammary epithelial cells through transcriptional induction of serum and GC-inducible protein kinase-1 (SGK-1) and mitogen-activated protein kinase phosphatase-1 (MKP-1) [16]. On the contrary, other studies have reported that GR activation in thymocytes and lymphocytes triggers apoptosis through cathepsin B [17]. In osteoblasts, there is a consensus that glucocorticoid treatmentinduces osteoblast death both in vivo and in vitro [11,18]. It has been reported that several important protein participate in the apoptosis or cell cycle arrest of osteoblast [6][7][8]14] but to the best of our knowledge, there is no transcription-control or protein-protein interaction between GR and these proteins. So we need to further investigate the downstream executive molecules.
GZMA, one of the apoptotic effectors and direct transcriptional targets of glucocorticoid receptor [19], has been reported to mediate glucocorticoid-induced apoptosis of human leukemia cells [20]. Moreover, it is also reported to be involved in glucocorticoidinduced apoptosis of human osteoblast-like cells [21]. Besides GZMA, GILZ, an ubiquitous protein, whose expression is directly up-regulated by glucocorticoids, has various interaction with apoptosis and cell cycle related protein such as SGK1 and FOXO [21][22]. Based on the above analysis, we assumed these two protein are involved in DEX induced osteoblast apoptosis and cell cycle arrest. However, our results illustrated that DEX does not have an impact on their expression, but can significantly upregulate the expression level of P53 and enhance its transcriptional activity as evident by elevated expression of PUMA, NOXA and p21.
The crosstalk between GR and P53 has long been a dubious topic for quite some time. Many studies verified that GR activation strengthens the transcriptional activity of p53. In a neural cell line HT-22, GR activation by DEX enhances the transcriptional activity of p53, and thus induces apoptosis and cell cycle arrest [9]. In human lung carcinoma cells, suppression of p53 blocks DEXinduced p21 (WAF1/Cip1) over-expression and G1 growth arrest [23]. In consistent with these studies, Murphy and colleagues demonstrated that the loss of p53 impairs transcription of GR target genes, and subsequently impaired GCs rescue of death in a mouse model of LPS shock [24]. These reports suggested that p53 is necessary to GR transactivation and the downstream signal transduction.
On the contrary, numerous evidence argue that there is negative cross-talk between GR and p53. Sengupta reported that GR activation resulted in p53 cytoplasmic sequestration, thus inhibiting p53's transcriptional activity [25]. Similarly, Dex has been shown to induce ubiquitylation of GR and p53, inhibit transcriptional activity of both proteins in stressful HUVEC and normal hepatoma cells [26]. Furthermore, It has been shown that p53-mediated transcriptional activity is inhibited by GR co-GRasignificantly decreased in the siGR -1, and siGR -2 groups compared to that in the siC groups and FBS groups. SiGR-1 and siGR -2 group: Cells treated with siRNA molecules siGR -1 and siGR -2, respectively. siC group: Cells treated with a siRNA which had a randomized nucleotide sequence that had no significant homology to any part of the human genome. FBS group: Cells without any treatment. (C) The level of the cleaved-caspase 3 protein was significantly up-regulated in the DEX group, when compared to that in the control group, the siGR-1 group,DEX+ siGR-1 group. (D) Visualization of apoptotic cells by the TUNEL (DAB/haemotox) assay in MC3T3-E1 cells. The cell nuclei was stained with DAPI (blue) (49,6-diamidino-2phenylindole).And the apoptotic cells were visualized by TUNEL staining(brown). The apoptotic cells in DEX group is significantly increased compared to the control group, the siGR-1 group,DEX+siGR-1 group. (E) Assessment of apoptosis in MC3T3-E1 cell using flow cytometry with Annexin V-FITC/PI staining. Annexin V-FITC(+)PI(2) cells were considered as early apoptotic cells. The apoptotic cells to total cells ratio was significantly (P,0.05) higher in DEX group compared to the control group, the siGR-1 group,DEX+siGR-1 group. (F) Cell cycle analysis using flow cytometry with PI staining, showing representive histograms of MC3T3-E1 cells in control group, DEX group, the siGR-1 group,DEX+ siGR-1 group. The proportion of cells in the S phase decreased markedly in the DEX group in comparison to the control group, the siGR-1 group,DEX+ siGR-1 group. The proportion of cells in the G0-G1 significantly increased in the DEX group. (P,0.05) Control group: cells treated with PBS. DEX group: cells cells treated with 1 mmol/L dexamethasone. siGR-1 group: cells treated with siRNA molecules siGR -1. DEX+ siGR-1 group: cells cells treated with 1mmol/L dexamethasone plus siRNA molecules siGR -1. doi:10.1371/journal.pone.0037030.g003   sip53-1, sip53-2) targeting p53mRNA; the mRNA expression level of p53 in the sip53-1and sip53-2 groups decreased significantly (P,0.05) compared to that in the FBS group and the siC group. (B) Examination of the protein expression level of the p53agene by Western blotting following treatment with the indicated siRNA molecules sip53-1 and sip53-2. The protein expression level of P53 significantly decreased in the sip53 -1, and sip53 -2 groups compared to that in the siC groups and FBS groups. Sip53-1 and sip53 -2 group: Cells treated with siRNA molecules sip53 -1 and sip53 -2, respectively. siC group: Cells treated with a siRNA which had a randomized nucleotide sequence that had no significant homology to any part of the human genome. FBS group: Cells without any treatment. (C) The level of the cleaved-expression, and that wild-type p53 efficiently inhibited GR transcriptional activity [27]. These studies confirmed that mutual inhibition between GR and p53 existed in some cell types in some specific condition.
And some other studies have also propounded that there is no relationship between p53 and GR. Dex efficiently suppressed TNF alpha-induced apoptosis through abrogated Bcl-2 reduction, which is independent of p53's status [28]. In a similar manner, irrespective of p53 status, mammalian target of rapamycin inhibitors sensitize multiple myeloma cells to dexamethasoneinduced apoptosis [29].
In this study, we have illustrated that GR activation by dexamethasone up-regulates the expression level of P53, enhances its transcriptional activity, thus resulting in p21, PUMA and NOXA up-regulation. And this subsequently induces osteoblast cell cycle arrest and apoptosis. Furthermore, we verified that disruption of p53 activation by p53 siRNA leads to the decrease in the DEX-induced cytotoxic and apoptotic activities in osteoblast cells.
The tumor suppressor p53 provides exquisite decision between cell growth arrest and apoptosis in response to various cellular stress. Sustained stress or irreparable damage trigger p539s killer functions to initiate transcription of pro-apoptotic genes such as Puma, Noxa, Bax, and Bid. To prevent the unnecessary loss of cells which could cause premature aging, the killer functions of p53 are tightly regulated and balanced against protector functions that arrest cell cycle and support survival in response to low stress or mild damage [30][31]. Our results showed GR activation can upregulate P53 expression and transcriptional activity, which paralleled with cell cycle arrest and apoptosis. While MC3T3-E1 cells (have functional GR and p53) were readily arrested in G1 phase and undergone apoptosis in response to DEX treatment, the GR-silent and p53-silent MC3T3-E1 cells preceded into S phase and survived in the presence of DEX. These results demonstrated that both GR and P53 is necessary for GC induced osteoblast apoptosis and cell cycle arrest. Furthermore, western blot results have verified transcriptional activity of p53 is dependent on GR activation.
In conclusion, our findings suggested that GR activation by DEX can up-regulate P53, which then enhances its transcriptional activatity. This mechanism contribute to apoptosis and cell cycle arrest of MC3T3-E1 cells. P53 could be used as a new target for the treatment of GC-induced osteoporosis and osteonecrosis.