Effects of Hypergravity on Osteopontin Expression in Osteoblasts

Mechanical stimuli play crucial roles in bone remodeling and resorption. Osteopontin (OPN), a marker for osteoblasts, is important in cell communication and matrix mineralization, and is known to function during mechanotransduction. Hypergravity is a convenient approach to forge mechanical stimuli on cells. It has positive effects on certain markers of osteoblast maturation, making it a possible strategy for bone tissue engineering. We investigated the effects of hypergravity on OPN expression and cell signaling in osteoblasts. Hypergravity treatment at 20 g for 24 hours upregulated OPN expression in MC3T3-E1 cells at the protein as well as mRNA level. Hypergravity promoted OPN expression by facilitating focal adhesion assembly, strengthening actin bundles, and increasing Runx2 expression. In the hypergravity-triggered OPN expression pathway, focal adhesion assembly-associated FAK phosphorylation was upstream of actin bundle assembly.


Introduction
Mechanical stimuli play important roles in bone remodeling. Bone cells differentiate, proliferate, and mature under mechanical stress in vivo, and are extremely sensitive to mechanical cues [1][2][3][4][5]. Engineered bone tissue cultured in vitro always fails to provide adequate tissue structure and mechanical properties. This is largely attributed to the lack of mechanical stimuli during cell culture. Hypergravity, or centrifugation, is a convenient approach to place mechanical stress on cultured cells, and is known to have an impact on osteogenesis [6]. In previous studies, low-level (<10 ×g), mid-level (10-100 ×g), and high-level (>100 ×g) hypergravity conditions were all observed to have significant effects on the morphology, proliferation, gene transcription, extracellular matrix protein synthesis, and alkaline phosphatase activity of osteoblasts. These findings suggested that hypergravity is a possible approach for bone tissue engineering [7][8][9][10]. To evaluate the impact of hypergravity on osteogenesis, a detailed analysis of its effects on osteoblasts should be conducted and the mechanisms underlying osteoblast response should be explored.
During bone regeneration and development, OPN, a marker for osteoblasts, plays an important role in cell communication and matrix mineralization [11][12][13][14][15][16]. Recently, it was found that OPN is regulated by mechanical forces, and contributes to the mechanotransduction pathway. Shear stress, mechanical stretch, and water pressure are all known to affect OPN expression in a variety of cell types, such as osteoblasts, osteocytes, and bone marrow mesenchymal stem cells [17][18][19][20][21]. Gravity was also shown to have a significant impact on OPN expression in human periodontal ligament fibroblasts, MG-63 osteoblasts, endothelial cells, and ML-1 cells [22][23][24][25]. Ishijima et al. also showed that OPN was responsible for bone loss in the tail suspension mouse model that is often used to mimic microgravity effects on lower limbs in vivo on earth [26][27][28]. However, the mechanism underlying the change in OPN expression in cells due to altered gravity warrants further investigation.
In the present study, we aimed to understand the mechanism by which hypergravity affects OPN expression in osteoblasts. Using a custom-designed centrifuge, we cultured cells under 20-g hypergravity conditions, and found that OPN expression and secretion were upregulated in MC3T3-E1 osteoblast cells. Runx2, a major regulator of osteogenic differentiation, was required for hypergravity-induced OPN expression. A multi-index quantification showed that the cytoskeleton and focal adhesions of MC3T3-E1 cells were altered by hypergravity, and they also contributed to hypergravity-induced OPN expression. Taken together, our findings showed that hypergravity may affect OPN expression via changes in focal adhesions, the cytoskeleton, and Runx2.

Mechanical device
A custom-made low-speed centrifuge was used to apply hypergravity to the cells in this study. Cells in culture flasks (or dishes) were placed on the horizontal rotors in the centrifuge, and 20 ×g was applied for 24 h in an incubator at 37°C. The control cells were placed on the platform of the centrifuge, as shown in Fig 1. The temperature was measured on the platform, where the control cells were placed, and in the centrifuge. The temperature was measured by THG312 thermometer (Mettler Toledo, Switzerland), and the value was 36.7 ± 0.2°C at the platform, and 36.8 ± 0.3°C in the incubator.
Cell culture MC3T3-E1 osteoblastic cells (Cell Center; School of Basic Medicine of Peking Union Medical College) were purchased at the passage number of 20, and were maintained in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mmol/ L L-glutamine, and 1% penicillin-streptomycin at 37°C in 5% CO 2 . Cells were sub-cultured (1:3), every 3 days, using 0.25% trypsin with 1 mM EDTA. 15mM Hepes was added to adjust the pH in culture medium. Unless otherwise specified, 10 5 cells per flask were exposed to 20 ×g hypergravity for 24 h, after seeding and spreading for 2 h. The control cells were subjected to the same conditions as the hypergravity-exposed cells in terms of timing, incubation media, and other procedures. The MC3T3-E1 cells used in this study were from passages 25 to passages 30. We immediately collected n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples for western blotting analyses as well as n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples for real-time PCR, respectively.
In some experiments, groups of cells were placed in incubator for a period of time after 1 ×g and 20 ×g treatments; and samples were collected at 25 h, 26 h, and 27 h. We collected n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples from each class. In experiments with varies cell densities, 10 3 , 10 4 , 10 5 , or 10 6 cells were exposed to 20 ×g hypergravity for 24 h. In experiments with altered preparatory adherent period, cells were preparatory adhered over a range of 6 h, 12 h, and 24 h and then exposed to 20 ×g hypergravity for 24 h. We collected n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples from each class. In addition, cells were exposed to 20 ×g hypergravity for a longer period of 48 h or 72 h. We then collected n = 4 1 ×g controls and n = 4 20 ×g hypergravity treated samples from each condition for western blotting analyses.
In blebbistatin (Sigma, St. Louis, MO) treatment experiments, MC3T3-E1 cells were treated with 50 μM blebbistatin for 24 h during 20 ×g hypergravity stimulation (n = 6). In FAK inhibitor (PF573228; Sigma, St. Louis, MO) experiments, cells were treated with 10 mM PF573228 for 24 h during 20 ×g hypergravity stimulation. We collected n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples for western blotting analyses as well as n = 4 1 ×g controls and n = 4 20 ×g hypergravity treated samples for real-time PCR, respectively. To measure the secreted OPN in the culture medium, MC3T3-E1 cells were first seeded on culture flasks with complete media. After adhering and spreading, cells were cultured in serum-free media and then exposed to 20 ×g hypergravity for 24 h. The culture media were collected and concentrated using Millipore's Amicon Ultra 15 mL centrifugal filter device, according to the manufacturer's protocol. We collected 1 ×g controls and 20 ×g hypergravity treated samples without drug (n = 9), with blebbistatin (n = 6) or with PF573228 (n = 5) for western blotting analyses.
The primary osteoblasts were obtained from 3-day-old mouse calvarias following the published method [29,30]. In brief, the skulls were dissected and the periosteum and endosteum were stripped off. Then the bone was cut into pieces and digested with 2.5 mg/ml typsin for 30 min and 1.0mg/ml collagenase II twice for 1h. The cells were collected and cultured in α-MEM supplemented with 10% fetal bovine serum (FBS), 2 mmol/L L-glutamine, and 1% penicillinstreptomycin at 37°C in 5% CO 2 . The osteoblasts from passages 2 to passages 6 were used in this study. We collected n = 6 1 ×g controls and n = 6 20 ×g hypergravity treated samples for western blotting analyses.
A comparative CT method was used for quantification. The housekeeping gene GAPDH was used for internal normalization.

Immunofluorescence staining
Cells were washed with PBS, fixed with 4% paraformaldehyde for 10 min, and permeabilized in 0.1% Triton X-100 buffer for 5 min. After blocking in 5% BSA for 30 min at room temperature, the cells were incubated with primary antibodies against OPN (1:500, Abcam), Runx2 (1:200, Abcam), and paxillin (1:200, Santa Cruz) diluted in 5% BSA at 4°C overnight. Cells were then incubated with Anti-mouse IgG conjugated to TRITC (1:200, Abcam), anti-rabbit IgG conjugated to FITC (1:200, Abcam), and rhodamine phalloidin (1:100, Abcam) for 2 h at room temperature. Lastly, cell nuclei were visualized by DAPI (Sigma) and viewed under a Leica TCS SP5 confocal microscope system. The fluorescence images were selected randomly for quantification. The quantification of the fluorescence signal was done using the image processing software Image-Pro plus (version 6.0, Media Cybernetics, LP) and the ImageJ software. To measure the thickness of actin bundles, we drew a line across each cell, and the bundle width at the cross line section was quantified. To avoid background noise, only regions with a fluorescence intensity higher than 40 were considered as actin bundles and thus counted.

Statistical analysis
All experiments in this study were performed at least quadruplicate. The Wilcoxon matched pairs test and Kruskal-Wallis test were carried out to compare the differences between the experimental groups in realtime-PCR and western blotting. Unpaired t-test was used to compare the differences between the two experimental groups in immunofluorescence staining. A Pvalue < 0.05 was considered significant.

Hypergravity increases OPN and Runx2 expression
OPN and Runx2 are typical markers of osteoblasts maturation and differentiation. To investigate whether hypergravity affects OPN and Runx2 expression, we exposed MC3T3-E1 cells, a well-established murine osteoblastic cell model used in bone studies, to 20 ×g centrifugal force for 24 h. After exposure to hypergravity, cells were assayed for the mRNA and protein levels of OPN and Runx2 by real-time PCR, immunofluorescence staining, and western blotting, respectively (Fig 2A-2K). We observed that hypergravity promoted an increased expression of OPN at both the mRNA (2-fold of control) (n = 6, p<0.05) and protein level (1.6-1.8-fold of control) (n = 7, p<0.05) (Fig 2A, 2C, 2D, 2F and 2G). Secreted OPN in the culture medium was also upregulated (1.4-fold of control) after hypergravity treatment (n = 9, p<0.05) (Fig 2H  and 2I). The effect of hypergravity on OPN expression lasted for at least 3 hrs after the hypergravity exposure (n = 6, p<0.05) (Fig 2L and 2M). We also detected a 0.7-and 1.0-fold increase in Runx2 expression following hypergravity by immunofluorescence staining and western blotting (n = 6, p<0.05) (Fig 2C, 2E, 2J and 2K). Runx2 mRNA increased 1.2-fold after hypergravity treatment (n = 6, p<0.05) (Fig 2B). These results indicate that hypergravity at 20 ×g increases the expression and secretion of OPN, and the expression of Runx2.
For better understanding the effect of hypergravity on oesteoblasts, we assayed the effect of the time length for the gravity treatment, the preparatory adherent period before the hypergravity exposure, and cell density on hypergravity-induced increment of OPN and Runx2. To investigate the effect of cell density on hypergravity-increased OPN and Runx2 expression, the cells were seeded at the density of 10 3 , 10 4 , 10 5 , and 10 6 and exposed to hypergravity. By immunofluorescence staining, we observed that at the cell density of 10 3 , 10 4 , 10 5 , hypergravity up-regulated the expression levels of OPN and Runx2 (Fig 3A-3C). At the density of 10 6 , OPN and Runx2 expression were all suppressed in both control group and hypergravity-treated group; and no difference was observed between these two groups (Fig 3B and 3C). We also examined the effect of different preparatory adherent period over a range of 6 h, 12 h, and 24 h, and found that OPN and Runx2 expression were up-regulated after hypergravity treatment at 6 h (n = 6, p<0.05) (Fig 3D, 3G and 3H), and 12 h pre-adhesion groups (n = 6, p<0.05) (Fig 3E, 3G and 3H). However, when plating cells on the flask for 24 h before the hypergravity exposure, the difference of OPN and Runx2 expression between hypergravity groups and the control groups disappeared (n = 6, p>0.05) (Fig 3F, 3G and 3H). MC3T3-E1 cells exposed to hypergravity for 48 and 72 hours showed no difference in the level of OPN and Runx2 between the control and hypergravity-treated groups (n = 4, p>0.05) (Fig 3I-3P).
To test the effect of hypergravity on OPN and Runx2 expression in primary osteoblast, we exposed murine calvarial osteoblasts to 20 ×g centrifugal force for 24 h. By immunofluorescence staining, we detected a 0.88-and 0.4-fold increase in OPN and Runx2 expression following hypergravity, respectively (Fig 4A-4C). Moreover, hypergravity-increased expression of OPN and Runx2 in primary osteoblasts were confirmed by western blotting (n = 6, p<0.05) (Fig 4D-4G).
Since OPN plays crucial roles in bone development, regeneration, and mechanotransduction, we next adopt MC3T3 as a model to explore how hypergravity affects OPN expression.

Runx2 is upstream of hypergravity-regulated OPN expression
Runx2, an indispensable transcription factor in the osteoblastic lineage, is required for OPN promoter activity, and it is associated with the differentiation of osteoblasts [32][33][34][35]. However, whether Runx2 plays a role in hypergravity-induced OPN expression is unclear. To examine the role of Runx2 in hypergravity-induced OPN expression, we transfected the cells with Runx2 siRNA and exposed them to 20 ×g hypergravity for 24 h. Runx2 knockdown blocked the differences in OPN expression and secretion between the 20 ×g hypergravity-treated and control groups, and downregulated OPN expression in both the groups (n = 6, p<0.05) (Fig 5). These results indicate that Runx2 is involved in the hypergravity-regulated OPN expression pathway. We next explored the upstream factors regulating OPN and Runx2 expression.
Hypergravity strengthens the cytoskeletal system and focal adhesions in MC3T3-E1 cells The cytoskeletal system and focal adhesions (FAs) are known to be the mechanosensing elements in several mechanotransduction pathways [36][37][38]. Previous reports have demonstrated that the fluorescence intensity of labeled actin bundles increased in certain types of hypergravity-treated cells [25,39]. However, not only the intensity, but also the distribution and thickness of actin bundles are crucial in characterizing the effects of hypergravity on cells. Using an integrative approach, we introduced a multi-index assay to characterize the impact of hypergravity on actin fibers and FAs. We observed that compared with unloaded cells, 20 ×g hypergravity-treated cells had higher actin staining intensity and thicker actin filaments. The average thickness of actin bundles was 1.39 μm in unloaded cells and 1.6 μm in the centrifuged cells. In addition, 36% of all visualized fibers in the loaded cells were thicker than 1.5 μm, compared with those in unloaded cells (29% of all visualized fibers) (Fig 6).  Upon visualization of FAs with a paxillin antibody, we observed the fluorescence intensity, size, and number of FAs under normal gravity and at 20 ×g hypergravity. The number of observable FAs in the hypergravity-treated cells and the fluorescence intensity of each FA were higher than those in the unloaded cells (Fig 7A and 7C), while only a small difference in FA size was observed between unloaded (2.1 μm 2 ) and post-hypergravity (2.3 μm 2 ) samples ( Fig  7D). When mapping the 2D diagram coordinates for average size and intensity of FAs, more FAs in the hypergravity-treated cells fell into the high-FA intensity and/or size region, when compared with those in the unloaded cells (Fig 7B). p-FAK and paxillin, two important proteins in focal adhesions, were assayed to evaluate FA assembly. Both p-FAK and paxillin   expression were upregulated after hypergravity treatment (n = 6, p<0.05) (Fig 7E-7H). These results show that hypergravity strengthens actin fibers and focal adhesions in osteoblasts.

Inhibition of actomyosin contractility diminishes the hypergravity effects on focal adhesions and OPN expression
Actin bundles, bearing contractions generated by actomyosin sliding, are tracks that propagate forces, and play crucial roles in mechanosensory signal transduction. FAs consist of several signaling molecules (including Src, Cas, Fak, Vinculin, and Integrin) that can undergo tension-dependent conformational changes to affect kinase activity, phosphorylation site availability, intracellular localization, and/or ligand affinity. Actin bundles, FAs, and their interactions play pivotal roles in several mechanosensory pathways [40][41][42]. Therefore, we proposed that actomyosin contraction of actin bundles and FAs may contribute to hypergravity-regulated Runx2 and OPN expression, and also determined the mechanosensing element that is the furthest upstream.
To explore the effects of actomyosin contraction on hypergravity-triggered signals, we used 50 μM blebbistatin to inhibit actomyosin contraction and assayed FA assembly, Runx2, and OPN expression. Immunofluorescence staining showed that non-cortical actin alignment and the differences in F-actin intensity between 20 ×g-and 1 ×g-treated cells disappeared after the blebbistatin treatment (Fig 8A and 8B). The number of actin fibers in each cell considerably decreased after blebbistatin treatment (Fig 8C). Western blotting showed that blebbistatin treatment failed to diminish hypergravity-induced FAK phosphorylation (Fig 8D and 8E).
Next, we examined Runx2 and OPN protein expression after blebbistatin treatment. Western blotting showed that blebbistatin treatment inhibited hypergravity-induced Runx2 and OPN expression, and OPN secretion (Fig 9). These results support our hypothesis that hypergravity affects OPN expression and secretion via regulation of actin bundles.

FAs are involved in promoting hypergravity-induced OPN expression
To investigate the role of FAs in hypergravity-induced OPN up-regulation, we used 1 μM PF573228 to inhibit FAK phosphorylation and then analyzed the cytoskeleton, and the expression of Runx2 and OPN. We observed that PF573228 inhibited FAK phosphorylation and hypergravity-induced p-FAK expression (Fig 10A and 10B). Fluorescence staining demonstrated that PF573228 treatment diminished the intensity differences in actin bundles between hypergravity-treated and unloaded cells (Fig 10C and 10D). Both the mRNA and protein levels of Runx2 and OPN were inhibited by PF573228 and no differences were observed in Runx2 and OPN expression between hypergravity-treated and unloaded cells (Fig 11). Since blebbistatin treatments failed to ablate hypergravity-promoted p-FAK expression, it is likely that FAK phosphorylation and actin bundles are involved in the hypergravity-induced increase in Runx2 and OPN expression, and also that FAK phosphorylation is upstream of actin bundle assembly.

Discussion
In this study, we exposed osteoblasts to 20 ×g hypergravity by using a low-speed centrifuge for 24 h, and observed that hypergravity promoted OPN expression by facilitating FAs, actin bundle assembly, and Runx2 expression.
OPN expression is reported to be sensitive to altered gravity in space flight [43]. Our result, which showed that hypergravity can increase OPN expression in osteoblasts, supported the results of previous studies using other cell types, where short periods (60 min) of hypergravity stimulation increased OPN expression [44]. Here, we found that a medium-magnitude (20 ×g), 24 h continuous hypergravity stimulation upregulated OPN expression, and the hypergravityincreased OPN expression can last for at least 3 h after hypergravity deprivation. After 48 h hypergravity treatment, the expression of OPN and Runx2 returned to the same levels of the control group. Independent group has reported that OPN plays a role in the early the initial early molecular events during the course of bone and tendon remodeling [45,46]. The present work indicates that OPN is involved in the early response of osteoblast cells to hypergravity. The mechanism underlying the dynamically regulated expression of OPN is unknown and warrants further investigation.
Some transcription factors such as AP-1 and ERK1 are known to bind the OPN promoter and affect OPN expression in certain cell types under hypergravity conditions [44]. In this study, we observed that Runx2 was a necessary element in facilitating the hypergravity- regulated OPN expression pathway. Runx2 can regulate OPN expression by binding to the OPN promoter [15]; therefore, hypergravity may affect OPN via Runx2.
FAs and the cytoskeletal system, including actin bundles, microtubules, and intermediate filaments, have long been known to respond to hypergravity and microgravity [39,47,48]. Since proteins in FAs and actin bundles are well-documented mechanosensing/transducing elements, one may easily deduce that actin bundles and FAs may play important roles in the mechanotransduction pathway in cells under hypergravity. However, there are interactions and feedback pathways between FAs and actin bundle assembly, whose roles in the hypergravity-sensing pathway have not been fully elucidated. In the present study, using a multi-index analysis including the intensity, width, and distribution of actin bundles, we found that actomyosin contractility was important for hypergravity-induced upregulation of OPN and Runx2 expression, and that it was not responsible for FAK phosphorylation induced by hypergravity.   The inhibition of p-FAK diminished actin bundle assembly, as well as OPN and Runx2 upregulation under hypergravity conditions. These results indicated that both FAK phosphorylation and actomyosin contraction play roles in hypergravity-induced OPN expression, and suggested that FAK is upstream of actin assembly.u Since actin bundles are crucial to the cell's reaction to hypergravity, the signaling pathways interacting with actin bundles, such as Rho/ROCK, ERK/ERK1, and ERK-c-Fos, are worth investigating [23,44]. In addition, the effects of other cytoskeletal components, microtubules, and intermediate filaments on cells subjected to hypergravity warrant further exploration. Notably, the response of the cytoskeletal system to hypergravity may depend on the specific cell type and the dose of hypergravity. Rat bone marrow mesenchymal stem cells cultured on a smooth surface have been reported to have a nonsignificant cytoskeletal response to 10 ×g hypergravity treatments [49]. Here, we also demonstrated that the cell density, the time length for the hypergravity treatment and the preparatory adherent period before hypergravity treatment have effect on the expression of OPN and Runx2. Thus, the effects of hypergravity on cell function and the underlying mechanisms should be carefully evaluated in seeding cells used for tissue engineering.
In conclusion, our findings show that a 24-h medium hypergravity treatment can upregulate OPN expression and secretion in osteoblasts. Furthermore, FAK phosphorylation, actin bundle contractility, and Runx2 are closely involved in the hypergravity sensing and signal transduction process. These results suggest that hypergravity conditions induced by centrifugation may be a convenient approach in bone tissue engineering.