Impact of shear stress and simulated microgravity on osteocytes using a new rotation cell culture device

Acta Astronautica 116 (2015) 286–298

Contents lists available at ScienceDirect

Acta Astronautica journal homepage: www.elsevier.com/locate/actaastro

Impact of shear stress and simulated microgravity on osteocytes using a new rotation cell culture device Xiao Yang a, Lian-Wen Sun a,n, Xin-Tong Wu a, Meng Liang a, Yu-Bo Fan a,b,n a Key Laboratory for Biomechanics and Mechanobiology of Ministry of Education, International Joint Research Center of Aerospace Biotechnology and Medical Engineering, Ministry of Science and Technology of China, School of Biological Science and Medical Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, 100191 Beijing, China b National Research Center for Rehabilitation Technical Aids, 100176 Beijing, China

a r t i c l e in f o

abstract

Article history: Received 24 March 2015 Received in revised form 10 July 2015 Accepted 16 July 2015 Available online 26 July 2015

Osteocytes sense the mechanical loading and weightlessness, and orchestrate bone remodeling by directing both osteoblast and osteoclast functions. Previous studies mostly focused on the impact of mechanical loading or microgravity on the osteocyte changes in themselves. However as the mechanosensors, it is more meaningful to make clear whether and how the mechanosensitivity of osteocytes affected by weightlessness, since the alteration of mechanosensitivity may further affect the mechanotransduction process in osteocytes, and finally altered the bone remodeling process. Whereas until now, this aspect had long been overlooked and was still not clear yet. To investigate how osteocytes respond to shear stress under weightlessness, a novel custom-made device was designed to simulate microgravity and load fluid flow on cells in vitro at the same time. The osteocyte-like cell line MLO-Y4 was loaded by this novel device at 15 dyn/cm2 shear stress with 15 rpm rotating speed. There were two loading durations for different determinations: 30 min of loading for the detection of nitric oxide and Prostaglandin E2, and 6 h of loading for the detection of three bone formation biomarkers (alkaline phosphatase, osteocalcin and procollagen type I N propeptide). In order to preliminarily explore the mechanism of this altered response to mechanical loading with different gravity in osteocytes, an observation of cytoskeleton and the determination of the elements in the Wnt/β-catenin signaling pathway were also performed after 6 h of loading. The results showed that (1) the mechanical response of both NO and PGE2 were increased higher during 15–30 min of shear stress under simulated microgravity than that under normal gravity; (2) the mechanical response of ALP activity was decreased, while that of OC and PINP content were increased by simulated microgravity. Moreover, the ALP activity and the OC content were related to the activity of Wnt signaling pathway, which plays a key role in regulating the bone formation; (3) the exploration for the mechanism of altered mechanical bio-response in osteocytes showed that F-actin filaments enhanced by shear stress under simulated microgravity were not so robust as that showed under normal gravity, and some short dendritic processes at cell periphery were only observed within the simulated microgravity groups; (4) further, the simulated microgravity significantly inhibited the mechanosensitivity of the Wnt/β-catenin signaling pathway on protein level in the osteocytes. These results suggested that the mechanosensitivity of osteocytes was altered by simulated microgravity, and this may be an adverse effect on the osteoblastic bone formation whereas be good for osteoclastogenesis. The findings in this study may provide

Keywords: Osteocytes Simulated microgravity Fluid shear stress Mechanosensitivity

n Correspondence to: No. 37 Xueyuan Road, School of Biological Science & Medical Engineering, Beihang University, Haidian District, 100191 Beijing, China. Tel.: þ8610 82339349; fax: þ86 10 82339349. E-mail addresses: [email protected] (X. Yang), [email protected] (L.-W. Sun), [email protected] (Y.-B. Fan).

http://dx.doi.org/10.1016/j.actaastro.2015.07.020 0094-5765/& 2015 IAA. Published by Elsevier Ltd. All rights reserved.

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

287

an important clue for searching an efficient approach to prevent osteocytes from changing the orchestration for bone remodeling under microgravity, and further this prevention may finally make the physical exercises capable against bone loss under microgravity. & 2015 IAA. Published by Elsevier Ltd. All rights reserved.

1. Introduction Physical activity (like exercise or physical training) would induce mechanical loading that is responsible for maintaining or improving bone mass and architecture [1] as well as for inhibiting bone loss with age [2]. However, weightlessness, or microgravity, is induced by spaceflight and leads to loss of bone mass invariably [3,4]. Even the resistance exercises in space usually cannot prevent this weightlessness-induced osteoporosis [5–8]. The main cause for this inability to recover bone mass is that the mechanism of microgravityinduced osteoporosis of bone is still unknown. Bone is constantly renewed by collaborations of boneforming osteoblasts and bone-resorbing osteoclasts, both of whose activities are determined mostly by locally produced signaling molecules [9]. The major producer of signaling molecules are osteocytes, which are the most abundant cells in mature bones [10]. Osteocytes sense the mechanical loading, especially shear stress caused by fluid flow [11]. Osteocytes also translate mechanical input into biochemical signals via the secretion of signaling molecules [12]. They then transmit these signaling molecules to effector cells [13–15] in order to orchestrate bone remodeling by directing both osteoblast and osteoclast function [16–19]. Conversely, the ablation of osteocytes rapidly leads to decreased bone strength, microfractures, and osteoporosis [20]. Therefore, osteocytes play a crucial role in bone remodeling as mechanosensors and modulators. Previous studies have proven that not only mechanical loading but microgravity also affects the activity of osteocytes [21]. For example, mechanical unloading increases the prevalence of osteocyte apoptosis in vivo [22,23]; in addition, the osteolytic activity in mature osteocytes is intensified by microgravity in space flight [23,24] and further causes bone loss in both cortical and trabecular bone [25]. According to these researches, osteocytes' behaviors appear to cause the disorder of bone remodeling under weightlessness and are relevant to microgravity-induced bone loss. These studies mentioned above showed us that the viabilities of osteocytes are affected by microgravity. However, the study on osteocytes should not be limited to studying the microgravity impact on osteocytes in themselves, what is more important is to explore whether the function of osteocytes as the mechanosensors were affected by microgravity. One possible scenario is that osteocytes mechanosensitivity could be changed with the gravity alteration; this effect may further directly affect the mechanotransduction process in osteocytes and the instructions on osteogenesis or osteoclastogenesis in bone tissue. Whereas until now, the study on osteocytes capabilities of sensing the mechanical loading (namely their mechanosensitivity) under microgravity has long been overlooked. In our

opinion, yet studying and understanding this may explain to us why the bone remodeling regularly adapted to the physical countermeasures (mechanical loading at macrolevel) during spaceflight was unlike it showed on the earth. Therefore, in this study we focused for the first time on whether the osteocytes respond to shear stress (a prevalent model of mechanical loading on osteocytes) altered under simulated microgravity. We hypothesized that osteocytes would have a different mechano-response to fluid shear stress under simulated microgravity than that under normal gravity. In this case, it was necessary to make osteocytes to be stimulated by fluid flow shear stress under simulated microgravity environment, and a device that can make the shear stress and the simulated microgravity load on cells simultaneously was needed in our study. As we all know, plenty of ground-based experimental devices may realize the function of microgravity simulation in cell culture, such as the Rotating Wall Vessel bioreactor (RWV, NASA, USA), the Rotary Cell Culture System (RCCS, NASA, USA), and 2D/3D clinostats in studying cell behaviors directly [26,27]. We have also used one of these rotators to simulate microgravity with loading the fluid shear on osteocytes subsequently [28], trying to study the mechanoresponse of osteocytes after simulated microgravity. However, neither those simulation apparatuses nor the studies we did before can truly reflect the actual circumstance of mechanical stimulations on cells during spaceflight. In this case, a novel, custom-made ground-based experimental device was designed specifically for our study, which can achieve coupling shear stress with simulated microgravity. By using the device we made, two early mechanical secreted molecules, three bone formation biomarkers, the cytoskeletons, and seven primary components of the Wnt/β-catenin mechanical signaling pathway in osteocytes were investigated. By studying these, we could potentially improve the understanding on the mechanosensitivity of osteocytes under microgravity, which may further clarify the mechanism of osteoporosis induced by spaceflight, and can provide a clue for searching an efficient approach for the exercises more successfully countering the bone loss in space flight.

2. Materials and methods 2.1. Simulated microgravity with a novel custom-made device designed in our lab We designed a device to make the horizontal axis rotating, and medium flowing through flow chamber fixed on the axis at the same time (Fig. 1A). The device comprised

288

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

Fig. 1. The novel custom-made device. A. The model of the novel custom-made device. B. The schematic of the device. The computer controlled-motor drove the horizontal axis rotating, as well as controlled pump to regulate inflow rate. The flow chamber used in the device was a conventional parallel plate flow chamber with valid flow dimensions of 76.2 mm length (L)  0.3 mm height (H)  20 mm width (W). The valid region in flow chamber should meet the conditions of L»H, W»H and L/H410, so that the fluid could be the two dimension developed laminar flow. The residual acceleration was 1.7  10  3 g to reach the simulated microgravity environment when the device rotated with 15 rpm rotation speed in our experiment.

of flow chamber, fixed connectors, movable connectors, the pump and a computer-controlled motor (Fig. 1B). The computer-controlled motor drove the horizontal axis rotating with speed of 0–20 revolutions per minute (rpm), which produced an altered-vector of gravity relative to the fixed flow chamber and made the sum of the gravity force vector zero, in order to simulate microgravity like 2D clinostat which has been defined as a tool to obtain a “vectoraveraged” gravity and preventing cells from perceiving the gravitational acceleration vector [29,30]. In our experiment when the device rotated with 15 rpm rotation speed, the residual acceleration was 1.7  10  3 g to reach the simulated microgravity environment. The computer controlled-motor drove the horizontal axis rotating, as well as controlled pump to regulate inflow rate. Theoretically, wall shear stress in flow chamber can be estimated by the following formula [26]:

τw ¼

6μQ WH

2

F¼

6μ vF H

is the chamber width. F is the correction factor for a rectangular channel with a finite aspect ratio (when W/H 460, F is close to 1) [27], which can be derived from the standard solution of laminar flow. In the experiment, the flow chamber used in the device was a conventional parallel plate flow chamber with valid flow dimensions of 76.2 mm length (L)  0.3 mm height (H)  20 mm width (W), in which the fluid could be the two dimension developed laminar flow. In order to prove that the rotating flow chamber could be fit for loading shear stress to cells, a numerical simulation of τw in the flow chamber was calculated by FLUENT software in this study. The results showed the numerical τw was distributed uniformly at the bottom of the flow chamber, and thought to be nearly ideal conditions to study the effects of fluid shear on cells (result was not showed). 2.2. Cell culture

ð1Þ

Where m is the fluid viscosity, Q is the volumetric flow rate, ν is the inlet velocity, H is the chamber height, and W

MLO-Y4 cells, an osteocyte-like cell line, were kindly provided by Dr. Lynda F. Bonewald (Department of Oral Biology, University of Missouri at Kansas City School of

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

289

Dentistry, Kansas City, MO, USA). MLO-Y4 cells were cultured in Minimum Essential Medium (MEM) (Gibco, USA) supplemented with 2.5% (vol/vol) fetal bovine serum (FBS) (Hyclone, South America), 2.5% calf serum (CS) (Hyclone, New Zealand), 1% penicillin and streptomycin (1%P/S) (Invitrogen, USA) on culture flask coated with 0.15 mg/ml rat tail type I collagen (Millipore, USA).

procollagen type I N propeptide (PINP) were determined by ELISA according to the kit illustration (Rat OC ELISA Kit and Mouse PINP ELISA Kit; RapidBio Lab., California, USA). The absorbance was measured at 450 nm for both OC and PINP.

2.3. Loading tests on osteocytes

Immediately after 6 h of exposure to the simulated microgravity coupling with fluid shear stress (or only fluid shear stress, or only simulated microgravity, or control), all groups of cells on the glass slides were collected and fixed with 4% paraformaldehyde solution for 40 min, and permeabilized in 0.2% Triton  -100 (v/v) for 10 min in order to observe the cytoskeleton changes of osteocytes. A part of the samples were stained in rhodamine labeled phalloidin (Invitrogen, NY, USA) to visualize F-actin and 40-6Diamidino-2- phenylindole (DAPI) (Invitrogen, NY, USA) to visualize the nucleus. After that, cells were embedded overnight with antifade solution (ProLong Gold antifade reagent, Invitrogen, USA) and observed by using a Leica TCS SPE laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany). Rhodamine (red fluorescence) was excited at 544 nm and DAPI (blue fluorescence) at 372 nm.

MLO-Y4 cells were seeded onto collagen-coated glass slides at a concentration of 3  105 cells/slide 24 h prior to loading tests. Then the slides were placed into the flow chamber of the device. The horizontal axis of the device rotated at speed of 15 rpm, coupled with a unidirectional steady fluid flow shear stress of 15 dyn/cm2 applied on cells. The cells were loaded for 30 min to detect the early molecular mechanical response of osteocytes. The cells were also loaded for 6 h to detect the cytoskeleton, the components of the Wnt pathways and other bone formation biomarkers. The cells were divided into four groups: CON (normal gravity with no flow), CONS (normal gravity with fluid flow shear stress), SM (simulated microgravity with no flow) and SMOS (simulated microgravity with fluid flow shear stress).

2.6. Immunofluorescent observation of cell cytoskeleton

2.7. Wnt/β-catenin signaling pathway determination

2.4. Early mechanical response of osteocytes detection Nitric Oxide (NO) and Prostaglandin E2 (PGE2) are two early responding molecules to mechanical loading in osteocytes. The media of all groups after loading were collected at 0, 5, 10, 15, 20 and 30 min. Then NO concentration measured as nitrite (NO2) accumulation in conditioned medium was assayed using the Griess Reagent method according to kit illustration (Nitric Oxide Assay Kit; Beyotime; China), and PGE2 in conditioned medium was determined using Enzyme-linked Immunosorbent assays (ELISA) according to the kit illustration (Mouse PGE2 ELISA Kit; RapidBio Lab, California, USA). The absorbance was measured at 540 nm for NO and 450 nm for PGE2. 2.5. Bone formation biomarkers determination To identify the potential osteogenesis regulation in MLO-Y4 cells adapted to mechanical loading and unloading, alkaline phosphatase (ALP) activity, osteocalcin (OC) and procollagen type I N propeptide (PINP) were used as three biomarkers for bone mineralization and bone matrix synthesis after 6 h of loading. For ALP test, cells were conducted with 0.1% Triton  -100 (v/v) for 10 min and kept in 80 1C freezer overnight. Then it was freeze-dried on the lyophilizer. ALP activity was examined by p-nitro phenyl phosphate (PNPP) method (Alkaline Phosphatase Kit; Biosino Bio-tec., China), and the absorbance was measured at 405 nm. For OC and PINP test, cells were conducted with RIPA Lysis Buffer (Applygen Technologies Inc, China) on the ice for 20 min. Then cell supernatants were collected and kept in  20 1C. The proteins of osteocalcin (OC) and

After 6 h of loading, proteins of primary components in the Wnt signaling pathway were chosen to be detected western blotting. Cells were washed with PBS and lysed in RIPA buffer with protease inhibitor (Applygen, China). Samples were boiled for 5 min and were electrophoresed on a 12% SDS-PAGE gel. The separated proteins were transferred to a PVDF membrane which was blocked in 5% non-fat milk for 1.5 h. The membrane was incubated overnight with rabbit-polyclonal or mouse/rat-monoclonal primary antibody at a proper diluted concentration overnight at 4 1C (Table 1). Following the primary antibody incubation, the membrane was incubated with the appropriate goat-anti- mouse/rabbit or rabbit-anti-rat HRP secondary antibody (Boster; China) at a 1:10,000 (or 1:6000) dilution for 2 h. Then the chemiluminescence was detected using the ECL Smartchemi System (Beijing Sage Creation Science Co., Ltd., China). The expression level of each protein was then quantified by band density using Glyko BandScan 5.0 (Glyko, Hayward, CA, USA), and was normalized with respect to GAPDH expression. 2.8. Statistical analysis All data were analyzed using SPSS 19.0 software. Firstly, effects of gravity and shear stress on all variables were examined by a two-way ANOVA, and simple main effects analysis were performed to detect the differences within the same gravity groups and within the same static/shear stress groups. Secondly, the shear stress-to-static ratios of NO and PGE2 were performed by one-way ANOVA to detect the differences when the gravity changed. Finally, association between the elements of the Wnt signaling pathway and bone formation biomarkers was assayed by

290

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

Table 1 Primary antibodies and dilution. Proteins

Primary antibody

Producer

Wnt 3a LRP5 Β-catenin Sclerostin (sost) LEF1 CyclinD1 Connecxin43 GAPDH

Polyclonal rat Polyclonal rabbit Monoclonal mouse Polyclonal rabbit Polyclonal rabbit Monoclonal mouse Monoclonal mouse Polyclonal rabbit

R&D System, MN, U.S.A. Abnova, Taipei, Taiwan BD Bioscience, CA, U.S.A. Santa Cruz Biotechnology Inc., TX, U.S.A. Cell Signaling Technology, Inc., MA, U.S.A. BD Bioscience, CA, U.S.A. BD Bioscience, CA, U.S.A. Hangzhou Goodhere Biotechnology Co., Ltd., China

Table 2 Simple main effects analysis within groups of NO release.

CON vs. SM CON vs. CONS SM vs. SMOS CONS vs. SMOS

5 min

10 min

15 min

20 min

30 min

p ¼0.355 p o0.001nnn p o0.001nnn p ¼0.099

p ¼ 0.159 p o 0.001nnn p o 0.001nnn p o 0.001nnn

p ¼0.009nn p o0.001nnn p o0.001nnn p ¼0.997

p¼ 0.047n po 0.001nnn po 0.001nnn po 0.001nnn

p ¼ 0.001nnn p o 0.001nnn p o 0.001nnn p o 0.001nnn

* Statistically significant, **p o0.01, ***p o0.001.

partial correlation analysis. The data are presented as mean 7SD of at least three determinations. The significance level for statistical analyses was set as p o0.05.

3. Results 3.1. Effects of simulated microgravity on the early molecular response and on the shear stress-modulated early molecular response of osteocytes 3.1.1. Effects of simulated microgravity on the NO release and on the shear stress- modulated NO release The results showed that both shear stress and gravity had significant effects on the NO release (po0.001), and the cross effects between the shear stress and gravity on the NO release was also statistically significant (p ¼0.018). Simple main effects analysis (Table 2) indicated that firstly, NO released decreased under simulated microgravity compared to normal gravity during the whole experiment time, and had significant differences at 15 min (p ¼0.009) and 30 min (p ¼0.001); secondly, NO release increased significantly after exposure to fluid shear stress under both normal gravity (CON vs. CONS) and simulated microgravity (SM vs. SMOS) (Fig. 2A and B). The results also showed that although the amount of NO release under simulated microgravity was decreased and the peak time was delayed compared to that under normal gravity, the result of shear-to-static ratio showed NO releasing ratio was higher under simulated microgravity than that under normal gravity at 15 min and 30 min (Fig. 2C), indicated that the NO mechanical response was relatively promoted by simulated microgravity after the first 10 min of the mechanical stimulation.

3.1.2. Effects of simulated microgravity on PGE2 release and on the shear stress-modulated PGE2 release The results showed that only shear stress had significant effects on PGE2 release (po0.001), while gravity had no significant effect on it (p¼0.183). There was a significant interaction between the effects of gravity and shear stress (p¼0.018). Simple main effects analysis (Table 3) showed that firstly, PGE2 release decreased under simulated microgravity compared to under normal gravity, while had no significant difference between these two groups until at 30 min (po0.001); secondly like NO release, PGE2 release also increased significantly after exposure to fluid shear stress under both normal gravity (CON vs. CONS) (Fig. 3A) and simulated microgravity (SM vs. SMOS) (Fig. 3B). The results also showed that although the peak time was delayed after shear stress under simulated microgravity compared to that under normal gravity, the amount of PGE2 release in SMOS group was higher than that in CONS group after 15 min. Further, the result of shear-to static ratio showed PGE2 releasing ratio was also higher under simulated microgravity than that under normal gravity after 15 min (Fig. 3C), implicated that the PGE2 mechanical response was also relatively promoted by simulated microgravity after the first 10 min of the mechanical stimulation. 3.2. Effects of simulated microgravity on bone formation biomarkers and on the shear stress-modulated bone formation biomarkers in osteocytes The results of main effects showed that the gravity had significantly effects on only ALP activity (po0.001), and the shear stress had significantly effects on both ALP activity (po0.001) and PINP content (p¼ 0.001). The cross effects of the gravity and the shear stress was statistically significant on ALP activity (p¼0.001) and OC content (p¼0.01).

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

291

Fig. 2. Comparison of shear stress-induced NO release within different gravity groups. (A) NO release under normal gravity. The NO release in CONS group sharply increased in 5 min and the peak time was at 10 min. (B) NO release under simulated microgravity. The NO release in SOMS group also increased significantly than that in SM group, while the peak time delayed to 15 min. The total amount of NO release in SMOS was lower than that in CONS group. (C) shear-to-static ratio of NO release. The ratios showed that the increase of NO release changed more under simulated microgravity than under normal gravity at 15 min and 30 min. npo 0.05, nnp o 0.01, nnnp o 0.001.

Table 3 Simple main effects analysis within groups of PGE2 release.

CON vs. SM CON vs. CONS SM vs. SMOS CONS vs. SMOS

5 min

10 min

15 min

20 min

30 min

p ¼ 0.556 p o 0.001nnn p o 0.001nnn p o 0.001nnn

p ¼0.669 p o0.001nnn p o0.001nnn p o0.001nnn

p¼ 0.318 po 0.001nnn po 0.001nnn po 0.001nnn

p ¼0.11 p o 0.001nnn p o 0.001nnn p o 0.001nnn

po 0.001nnn po 0.001nnn po 0.001nnn po 0.001nnn

n

Statistically significant.***p o 0.001

Simple main effects analysis exhibited that firstly only ALP activity (Fig. 4A) significantly increased under simulated microgravity compared to under normal gravity (p o0.001), while OC content (Fig. 4B) and PINP content (Fig. 4C) decreased under simulated microgravity with no significant difference compared to control group. Secondly, under normal gravity, only ALP activity was increased significantly by shear stress. On the other hand, under simulated microgravity, ALP activity was significantly decreased, while OC content and PINP content were significantly increased by shear stress. The results indicated that simulated microgravity changed the mechanical responses of bone formation biomarkers. 3.3. Effects of simulated microgravity on cytoskeleton and on the shear stress-modulated cytoskeleton in osteocytes After 6 h of fluid shear stress, robust, organized F-actin filaments were formed and alignment within the cells was

seen in CONS group (Fig. 5B) compared to CON (Fig. 5A). However, F-actin filaments reduced and disassembled in SM group, and many short dendritic processes were appeared at cell periphery (Fig. 5C). In SMOS group (Fig. 5D), F-actin filaments were re-polymerized to organized form, but not as robust as that in CONS group, and some dendritic processes also could be seen at the cell periphery.

3.4. Effects of simulated microgravity on the Wnt/β-catenin signaling pathway and on the shear stress-modulated Wnt/ β-catenin signaling pathway in osteocytes The results of main effects (Table 4) showed that the gravity had significant effects on β-catenin, LRP5/6 and CyclinD1, and the shear stress had significant effects on most of the variables except CyclinD1 (p ¼0.122) and LEF1 (p ¼0.068). The gravity and the shear stress had significant

292

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

Fig. 3. Comparison of shear stress-induced PGE2 release within different gravity groups. (A) PGE2 release under normal gravity. The PGE2 release in CONS group sharply increased in 5 min and the peak time was at 5 min. (B) PGE2 release under simulated microgravity. The PGE2 release in SOMS group also increased significantly than that in SM group, while the peak time delayed to 15 min. The total amount of PGE2 release in SMOS group were lower than that in CONS group at the first 10 min, and then was higher than that in CONS group after 15 min. (C) Shear-to-static ratio of PGE2 release. The ratios showed that the increase of PGE2 release changed more under simulated microgravity than under normal gravity after 15 min. np o 0.05, nnp o 0.01, nnnp o 0.001.

cross effects on most of the variables except Cx43 (p¼0.303) in the Wnt/ β-catenin signaling pathway. Simple main effects analysis exhibited that four proteins of all elements in the Wnt signaling pathway were significantly affected by simulated microgravity: β-catenin (p¼0.021, decreased), sost (p¼ 0.007, increased), CyclinD1 (po0.001, decreased) and LEF1 (p o0.001, increased), which indicated that simulated microgravity could affect the Wnt signaling pathway in osteocytes. But what is more important than that was the results showed the different mechano-responses of the same indicator under two different gravity levels. Under normal gravity, the proteins of all elements in the Wnt signaling pathway were affected by shear stress. However, under simulated microgravity, only four elements were affected by shear stress: β-catenin, sost, LEF1 and Cx43. In details of comparisons between gravity groups, under normal gravity, the protein expressions of Wnt3a (Fig. 6A), sost (Fig. 6D), and CyclinD1 (Fig. 6E) were decreased significantly, while β-catenin (Fig. 6B), LRP5/6 (Fig. 6C), LEF1 (Fig. 6F) and CX43 (Fig. 6G) were increased significantly by shear stress. However under simulated microgravity, firstly, Wnt3a, LRP5, and CyclinD1 had not significantly changed by shear stress; secondly, β-catenin and LEF1 had opposite changing trend by shear stress compared to that under normal gravity; finally, sost were decreased more, while CX43 were increased less by shear stress under simulated microgravity compared to that under normal gravity. The results indicated that mechanical responses of the

elements in the Wnt signaling pathway had been significantly affected by gravity alteration on protein level. Furthermore, partial correlation showed that some protein expressions in the Wnt signaling pathway had significant correlations with ALP activity and OC content (Table 5), which indicated that the mechanical responses of ALP activity and OC content could be affected by simulated microgravity via the changes of the Wnt signaling pathway. 4. Discussion and conclusion Osteocytes, as the most abundant cells in bone tissue, take responsibility of mechano-sensation and mechanotransduction [26,28]. By secreting the signaling molecules, osteocytes transmit the mechanical bio-signals to affect osteoblasts and osteoclast functions, and finally direct the bone remodeling [20]. However, the regular bone remodeling adapted to mechanical stress is disrupted by microgravity in space flight [31], whereas the resistance exercises in space cannot prevent microgravity-induced osteoporosis [8,20,31,32]. Since osteocytes are capable of orchestrating the cellular process of bone adaptation in response to mechanical loading [10,33], thus, there must be something wrong with the capabilities of osteocytes under the microgravity. In order to study whether the mechanosensitivity of osteocytes is altered under microgravity, we investigated the mechanical responses of early signaling molecules and of bone formation biomarkers;

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

293

Fig. 4. Bone formation biomarkers expression in osteocytes. (A) ALP activity of osteocytes. The ALP activity under normal gravity was increased by shear stress, while that was decreased by shear stress under simulated microgravity. (B) OC content of osteocytes. OC content in osteocytes had no significant difference under normal gravity after exposed to shear stress, while it was increased significantly by shear stress under simulated microgravity. (C) PINP content of osteocytes. Like, OC content, under normal gravity, PINP content was not significantly changed by shear stress, while it was also increased significantly by shear stress under simulated microgravity. np o0.05, nnp o0.01, nnnpo 0.001.

further, in order to do some shallow studies on the mechanism of the mechanosensitivity alteration of osteocytes, we also investigated the mechanical response of the cytoskeleton and Wnt/β-catenin signaling pathway in osteocytes. Below are the details about the findings in the study. Previous studies have proved that NO and PGE2 release were considered as initial parameters for osteocytes activation during early time of the mechanical loading [21,34,35], and were reported to be promoted by shear stress [12,33,36], which were also demonstrated in our study. More than this, we also found that under the simulated microgravity, although the total amount of the molecular release was inhibited, the mechanical responses of NO and PGE2 were both relatively promoted compared to that under normal gravity. NO and PGE2 as two early signaling response molecules involve in the anabolic response of bones to mechanical loading by modulating recruitment, differentiation, and activity of osteoblasts and osteoclasts [37,38]. However, it is important to realize that depending on the different concentrations, NO and PGE2 have two opposite anabolic effects on bone tissue and on osteoblast or osteoclast activities. Previous studies proved that mechanical loading induced NO and PGE2 release could inhibit osteoblast apoptosis, and stimulate osteoblastic bone formation [39]; however, over-high amount of

NO and PGE2 release inhibited osteoblasts proliferation and the collagen synthesis in bone formation, while stimulated osteoclastic bone resorption [36,37]. In our study, the NO and PGE2 were increased much higher by shear stress after 15 min under simulated microgravity. According to previous studies, this increasing amplitude may still good for osteogenesis under normal gravity. On the contrary, since the gravity declined, the increasing amplitudes of these two signaling molecules may be too high that will have adverse effects on bone formation. Therefore, we surmised molecular over-response to shear stress under simulated microgravity would be a signal for inhibition of osteogenesis and the activation of osteoclastogenesis in osteocytes. For three biomarkers in osteocytes, ALP activity, osteocalcin and PINP, expressed in osteocytes, were reported to be stimulated by shear stress or microgravity [33,40], and were considered to indicate bone formation [5,7,41–44]. In our study, the mechano-response of these biomarkers under normal gravity was consistent with the previous studies [7,45,46], which showed the proper approaches for studying. However, we were more concerned about the effect of simulated microgravity in this study. It was showed that the mechanical response of ALP activity was negatively reversed, while mechanical response of osteocalcin and PINP were promoted significantly by simulated

294

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

Fig. 5. F-actin of osteocytes. A. CON group; B. CONS group; C. SM group; D. SMOS group. The white dashed boxes showed that the polymerized F-actin filaments with increased fluorescence in CONS and SMOS group compared to their corresponding static groups. The white arrows showed that some short but thick dendritic processes appeared in SM and SMOS groups.

microgravity, implicated that the mechanical responses of these bone formation biomarkers were also affected by the gravity alteration. It was believed that the ALP activity was positively related and OC expression was negatively related to osteoblastic bone mineralization [44,47], while PINP expression was positively related to the collagen synthesis. Considering these three biomarker expressed in different stages of bone remodeling, and since the bone mineralization is the next stage of collagen synthesis in bone formation, our study provided a possible scenario that the early stage of bone formation was promoted by simulated microgravity, while bone formation would be inhibited at the stage of bone mineralization, and could not form the new mature bone tissue finally. The behaviors of osteocytes on cellular level, including the molecules secretion and the expressions of functional biomarkers, showed us that the mechanosensitivity of osteocytes was indeed adversely affected by simulated microgravity. This mechano-behavior alteration in osteocytes by simulated microgravity may be due to the changes of mechanical stimuli induced-mechanotransduction. In this case, the shallow exploration for some factors involved in

mechanotransduction was also studied in this paper. These factors include the cytoskeleton and the activities of the Wnt signaling pathway, both of which directly involved in cellular mechanotransduction in osteocytes, and further affected bone remodeling adapted to mechanical loading [44]. Thus, studying the mechano-response changes of both cytoskeleton and Wnt signaling pathway may help us to learn more about the way of osteocytes responding to mechanical stimuli under microgravity. In detail, the cytoskeleton plays as the load-bearing architecture of the cell, is very sensitive to altered gravity and highly responsive to external physical stimuli [48–50]. The cytoskeleton plays a key role in the mechanical signal to a biochemical one [44], and is an essential factor in mechanosensing and mechanotransduction of osteocytes [36,47]. In this study, we found the F-actin filaments were disrupted and the fluorescent intensity of them was weakened by simulated microgravity. Although the filaments of F-actin were reorganized and the fluorescent intensity were enhanced by shear stress in simulated microgravity group, the filaments were still not so robust and the fluorescent intensity was still not so strong like

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

295

Table 4 Main effects of gravity and shear stress on protein of variables in Wnt pathways. Wnt3a Gravity effect Shear stress effect Gravity  shear stress

p ¼0.545 p o 0.001nnn p¼ 0.002nn

β-catenin nnn

p o0.001 p¼ 0.022n p o0.001nnn

LRP5/6 nnn

p o 0.001 p o 0.001nnn p o 0.001nnn

Sost p ¼ 0.5 p o0.001nnn p ¼ 0.002nn

CyclinD1 nnn

po 0.001 p ¼0.122 p ¼0.008nn

LEF1

CX43

p¼ 0.423 p ¼ 0.068 p o 0.001nnn

p¼ 0.204 po 0.001nnn p ¼ 0.303

* Statistically significant, **p o 0.01, ***p o 0.001.

Fig. 6. Protein expressions of the components in the Wnt/β-catenin signaling pathway. (A–G) Wnt3a, β-catenin, LRP5, sost, Lef1, CyclinD1 and Cx43. (H and I) band graphs. Under normal gravity, the protein expressions of Wnt3a (A), Sost (D) and CyclinD1 (E) were significantly decreased after shear stress, while β-catenin (B), LRP5/6 (C), LEF1 (F) and Cx43 (G) were significantly increased after shear stress. Under simulated microgravity, the protein expressions of sost (D) and LEF1 (F) were significantly decreased after shear stress while Cx43 (G) was significantly increased after shear stress. np o 0.05, nnp o 0.01, nnn p o0.001.

that in normal gravity group. These findings suggested the cytoskeleton response to shear stress was affected by simulated microgravity. In previous study it was reported

that disruption of actin cytoskeleton inhibited the shear stress induced-release of NO and PGE2 in osteocytes [33,36]. In our study, although the shear stress

296

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

Table 5 Partial correlation between bone formation biomarkers and the protein expressions of variables in Wnt pathways ALP activity R Wnt3a β-catenin LRP5/6 Sost CyclinD1 LEF1 CX43

 0.676 0.748 0.665 0.528  0.575 0.852 0.268

OC content p Value nn

0.002 o 0.001nnn 0.003nn 0.024n 0.013n o 0.001nnn 0.282

PINP content

R

p Value

R

p Value

0.208  0.498  0.761  0.524 0.165  0.683  0.325

0.408 0.036n o 0.001nnn 0.026n 0.513 0.002nn 0.188

 0.336  0.073  0.058  0.279  0.351  0.267  0.393

0.173 0.774 0.820 0.262 0.153 0.284 0.107

* Statistically significant, **p o0.01, ***p o0.001.

induced-actin organization was weakened and disrupted by simulated microgravity, the mechano-responses of NO and PGE2 release were promoted by simulated microgravity instead, which was not same as it reported in previous studies. It is worth noting that the disruptions of cytoskeleton in previous studies were the drug interventions under normal gravity, while the disruption of F-actin in our study was caused by simulated microgravity. The simulated microgravity participated in all changes inside of osteocytes but not only just the cytoskeleton, which may cause the different behaviors of molecular secretion adapted to shear stress. However, how cytoskeleton interacts with these molecular components under simulated microgravity is still unclear yet and need to be determined in future. In osteocytes, the Wnt/β-catenin canonical signaling pathway has been implicated as an important mechanoresponsive pathway in bone and is used for transmitting mechanical signals to other functional cells [36,51,52]. Also, the activation of the Wnt signaling cascade enhances the mechanosensitivity of bone cells, lead to Wnt production by osteocytes, thereby tuning their own sensitivity to mechanical loading in a feedback loop [33]. Therefore whether the mechanical response of elements in the Wnt/β-catenin signaling pathway changed with the gravity alteration would be a possible mechanism for the alteration of osteogenesis in osteocytes. In our study, the results suggested in following four respects: 1) the mechanical response of Wnt3a and its receptor LRP5/6, two upstream components, did not respond to shear stress anymore under simulated microgravity; 2) the mechanical response of the key component, β-catenin, was adversely affected by simulated microgravity; 3) the target gene CyclinD1 did not respond to shear stress, and the mechano-response of LEF1 was also adversely affected by simulated microgravity; 4) the mechanical response of antagonist protein sost was instead promoted by simulated microgravity. This demonstrated that gravity alteration adversely affected the mechanical response of almost all the elements in the Wnt signaling pathway, further the capability of the Wnt signaling pathway in mechanotransduction and the regulation of bone formation adapted to mechanical stimuli would also be inhibited by simulated microgravity. It is noteworthy that in the Wnt signaling pathway has a specific target gene: CX43. It could be regulated by β/catenin, while in turn inhibits the expression of β/catenin

[53,54], as well as inhibits the expression of sost [36]. Furthermore, shear stress-induced NO and PGE2 release can increase the protein expression of CX43 [55,56], and upregulation of CX43 in turn can induce gap junction open to release PGE2 in osteocytes [54,57]. In our study, under normal gravity, when CX43 were promoted by shear stress, β/catenin, NO and PGE2 were all promoted as well, which were consistent with the previous studies. However, under simulated microgravity, when CX43, NO and PGE2 were still enhanced by shear stress, the expression of β/catenin was adversely affected by simulated microgravity. This result may have two explanations: first, CX43 was not only regulated by β/catenin, but also by NO and PGE2; second, the high expression of CX43 under simulated microgravity was a negative feedback to β/catenin. In brief, the result of our study suggested that CX43 may act as a potential negative regulator to osteogenesis, and somehow promoted the osteoclastogenesis in osteocytes under simulated microgravity. The limitation in this study is how the CX43 affects β/catenin signaling to osteoblastic bone formation under microgravity was not clear yet, and needs to be determined in future. It is interesting that partial correlation showed β-catenin, LRP5/6 and LEF1 in the Wnt signaling pathway had negative correlation with osteocalcin, and positively related to ALP activity in our study. High osteocalcin was reported to inhibit bone mineralization, while LEF1 transcription could inhibit the expression osteocalcin [58]. In our study, we found the inhibition of Wnt/β-catenin signaling pathway, the enhancement of osteocalcin and the reversed trend of ALP activity under the simulated microgravity, which all proved the conclusion in previous studies. These findings confirmed that the key role of the Wnt signaling pathway on osteogenesis regulation, and also suggested the bone mineralization adapted to mechanical stimuli may adversely affected by simulated microgravity through the mechanosensitivity decline of the Wnt signaling pathway. However, PINP had no correlation with the Wnt signaling pathway, which suggested Wnt signaling pathway inhibition may not affect the collagen synthesis. Therefore, mechanical loading induced mechanosensitivity and osteogenesis in osteocytes was suppressed through the inhibition of Wnt/β-catenin signaling pathway by simulated microgravity. In conclusion, the results of this study indicated that the mechanosensitivity of osteocytes was adversely affected by simulated microgravity, performed in the

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

over-response of signaling molecules and the inhibitions of bone formation biomarkers. The possible mechanism of this is firstly, the cytoskeleton capability of organization adapted to mechanical stimuli was weakened by simulated microgravity, which may be a signal for the inhibition of the mechano-sensing and mechanotransduction in osteocytes. What is more, the activity of the Wnt/β-catenin signaling pathway responding to shear stress was also suppressed by simulated microgravity. This suppression may adversely affect the mechanosensitivity and osteogenesis of osteocytes, and further inhibit the bone formation on cellular level at the stage of bone mineralization. It is conceivable that targeted therapy for the maintenance of the mechanosensitivity of osteocytes under simulated microgravity in future may be a novel clue and a breakthrough for helping the physical exercise fully countering osteoporosis in spaceflight.

Acknowledgment The authors thank Professor Xiaodu Wang (University of Texas at San Antonio, TX, U.S.A) for the advices and modifications of the manuscript. This work was funded by grants from the National Basic Research Program of China (No. 2011CB710901), the National Natural Science Foundation of China (11472033 and 11421202), the 111 Project (B13003) and supported by the Innovation Foundation of BUAA for PhD Graduates. References [1] A.P. Lirani-Galvao, M. Lazaretti-Castro, Physical approach for prevention and treatment of osteoporosis, Arq. Bras. Endocrinol. Metabol. 54 (2) (2010) 171–178. [2] T. CH, F. MR, O. MW, Mechanotransduction in bone: do bone cells act as sensors of fluid flow? FASEB J. 8 (11) (1994) 875–878. [3] S. Judex, C.T. Rubin, Is bone formation induced by high-frequency mechanical signals modulated by muscle activity? J. Musculoskelet. Neuronal Interact. 10 (1) (2010) 3–11. [4] K. Ksiezopolska-Orlowska, Changes in bone mechanical strength in response to physical therapy, Pol. Arch. Med. Wewn. 120 (9) (2010) 368–373. [5] A.G. Robling, A.B. Castillo, C.H. Turner, Biomechanical and molecular regulation of bone remodeling, Annu. Rev. Biomed. Eng. 8 (2006) 455–498. [6] J.I. Aguirre, L. Plotkin, S.A. Stewart, et al., Osteocyte apoptosis is induced by weightlessness in mice and precedes osteoclast recruitment and bone loss, J. Bone Miner. Res. 21 (4) (2006) 605–615. [7] V.I. Sikavitsas, J.S. Temeno!, A.G. Mikos, Biomaterials and bone mechanotransduction, Biomaterials (2001) 2581–259322 (2001) 2581–2593. [8] T.F. Lang, A.D. Leblanc, H.J. Evans, et al., Adaptation of the proximal femur to skeletal reloading after long-duration space flight, J. Bone Miner. Res. 21 (8) (2006) 1224–1230. [9] J. Klein-Nulend, R.G. Bacabac, J.P. Veldhuijzen, et al., Microgravity and bone cell mechanosensitivity, Adv. Space Res. 32 (8) (2003) 1551–1559. [10] E.H. Burger, J. Klein-Nulend, Microgravity and bone cell mechanosensitivity, Bone 22 (5) (1998) 127S–130S. [11] L.F. Bonewald, Mechanosensation and transduction in osteocytes, Bonekey Osteovision 3 (10) (2006) 7–15. [12] C. Androjna, N.P. McCabe, P.R. Cavanagh, et al., Effects of spaceflight and skeletal unloading on bone fracture healing, Clin. Rev. Bone Miner. Metab. 10 (2012) 61–70. [13] K. Watanabe, K. Ikeda, Osteocytes in normal physiology and osteoporosis, Clin. Rev. Bone Miner. Metab. 8 (4) (2010) 224–232. [14] A.D. Bakker, J. Klein-Nulend, Mechanisms of osteocyte mechanotransduction, Clin. Rev. Bone Miner. Metab. 8 (2010) 163–169.

297

[15] D.P. Nicolella, J.Q. Feng, D.E. Moravits, et al., Effects of nanomechanical bone tissue properties on bone tissue strain: implications for osteocyte mechanotransduction, J. Musculoskelet. Neuronal Interact. 8 (4) (2008) 330–331. [16] S.D. Tan, T.J. deVries, A.M. Kuijpers-Jagtman, et al., Osteocytes subjected to fluid flow inhibit osteoclast formation and bone resorption, Bone 41 (5) (2007) 745–751. [17] T. Moriishi, R. Fukuyama, M. Ito, et al., Osteocyte network; a negative regulatory system for bone mass augmented by the induction of Rankl in osteoblasts and Sost in osteocytes at unloading, PLoS One 7 (6) (2012) e40143. [18] S.A.J. Lloyd, H.J. Donahue, Gap junctions and biophysical regulation of bone cells, Clin. Rev. Bone Miner. Metab. 8 (2010) 189–200. [19] A. Neve, A. Corrado, F.P. Cantatore, Osteocytes: central conductors of bone biology in normal and pathological conditions, Acta Physiol. 204 (3) (2012) 317–330. [20] J. Klein-Nulend, A.D. Bakker, R.G. Bacabac, et al., Mechanosensation and transduction in osteocytes, Bone 54 (2) (2013) 184–190. [21] A. Santos, A.D. Bakker, J. Klein-Nulend, The role of osteocytes in bone mechanotransduction, Osteoporos Int. 20 (2009) 1027–1031. [22] J. Klein-Nulend, A.D. Bakker, Osteocytes: mechanosensors of bone and orchestrators of mechanical adaptation, Clin. Rev. Bone Miner. Metab. 5 (2007) 195–209. [23] S.C. Manolagas, A.M. Parfitt, For whom the bell tolls: distress signals from long-lived osteocytes and the pathogenesis of metabolic bone diseases, Bone 54 (2) (2013) 272–278. [24] R.G. Bacabac, J.J.W.A.V. Loon, J.M.A.d Blieck-Hogervorst, et al., Microgravity and bone cell mechanosensitivity: flow experiment during the DELTA mission, Microgravity Sci. Technol. 19 (5/6) (2007) 133–137. [25] N. Basso, J.N.M. Heersche, Effects of hindlimb unloading and reloading on nitric oxide synthase expression and apoptosis of osteocytes and chondrocytes, Bone 39 (4) (2006) 807–814. [26] J. Cao, S. Usami, C. Dong, Development of a side-view chamber for studying cell-surface adhesion under flow conditions, Ann. Biomed. Eng. 25 (3) (1997) 573–580. [27] J.T. Patton, D.G. Menter, D.M. Benson, et al., Computerized analysis of tumor cells flowing in a parallel plate chamber to determine their adhesion stabilization lag time, Cell Motil. Cytoskelet. 26 (1) (1993) 88–98. [28] X. Yang, L.-w. Sun, X.-t. Wu, et al., Effect of simulated microgravity on osteocytes responding to fluid shear stress, Acta Astronaut. 84 (2013) (2013) 237–243. [29] R. Herranz, R. Anken, J. Boonstra, et al., Ground-based facilities for simulation of microgravity: organism-specific recommendations for their use, and recommended terminology, Astrobiology 13 (1) (2013) 1–17. [30] C.A. Nickerson, C.M. Ott, S.J. Mister, et al., Microgravity as a novel environmental signal affecting Salmonella enterica serovar Typhimurium virulence, Infect. Immun. 68 (6) (2000) 3147–3152. [31] J.J.W.A.v Loon, Mechanomics and physicomics in gravisensing, Microgravity Sci. Technol. 21 (2009) 159–167. [32] P.D. Pajevic, J.M. Spatz, J. Garr, et al., Osteocyte biology and space flight, Curr. Biotechnol. 2 (3) (2013) 179–183. [33] J. Klein-Nulend, R.F.M. van Oers, A.D. Bakker, et al., Nitric oxide signaling in mechanical adaptation of bone, Osteoporos. Int. 25 (5) (2014) 1427–1437. [34] S. Temiyasathit, C.R. Jacobs, Osteocyte primary cilium and its role in bone mechanotransduction, Ann N Y Acad Sci 1192 (2010) 422–428. (Skeletal Biology and Medicine). [35] J. Klein-Nulend, C.M. Semeins, N.E. Ajubi, et al., Pulsating fluid flow increases nitric oxide (NO) synthesis by osteocytes but not periosteal fibroblasts – correlation with prostaglandin upregulation, Biochem. Biophys. Res. Commun. 217 (2) (1995) 640–648. [36] J. Klein-Nulend, R.F. van Oers, A.D. Bakker, et al., Bone cell mechanosensitivity, estrogen deficiency, and osteoporosis, J. Biomech. 48 (2015) 855–865. [37] S.D. Tan, A.D. Bakker, C.M. Semeins, et al., Inhibition of osteocyte apoptosis by fluid flow is mediated by nitric oxide, Biochem. Biophys. Res. Commun. 369 (4) (2008) 1150–1154. [38] A. Santos, A.D. Bakker, B. Zandieh-Doulabi, et al., Early activation of the β-catenin pathway in osteocytes is mediated by nitric oxide, phosphatidyl inositol-3 kinase/Akt, and focal adhesion kinase, Biochem. Biophys. Res. Commun. 391 (1) (2010) 364–369. [39] J.A. Robinson, M. Chatterjee-Kishore, P.J. Yaworsky, et al., Wnt/betacatenin signaling is a normal physiological response to mechanical loading in bone, J. Biol. Chem. 281 (42) (2006) 31720–31728.

298

X. Yang et al. / Acta Astronautica 116 (2015) 286–298

[40] M. Mullender, A.J. El Haj, Y. Yang, et al., Mechanotransduction of bone cells in vitro: mechanobiology of bone tissue, Med. Biol. Eng. Comput. 42 (1) (2004) 14–21. [41] C. Miyaura, Role of EP4 receptor in bone resorption induced by PGE, Nihon Yakurigaku Zasshi 117 (4) (2001) 293–297. [42] K. Miyajima, S. Suzuki, T. Iwata, et al., Mechanical stress as a stimulant to the production of osteocalcin in osteoblast-like cells, Aichi Gakuin Dent. Sci. 4 (1991) 1–5. [43] D. Pavlin, S. Dove, R. Zadro, et al., Mechanical loading stimulates differentiation of periodontal osteoblasts in a mouse osteoinduction model: effect on type I collagen and alkaline phosphatase genes, Calcif. Tissue Int. 67 (2) (2000) 163–172. [44] S.C. Cowin., Bone Mechanics Handbook, CRC Press, Boca Raton. London, New York, Washington, D.C, 2001. [45] D. Kaspar, W. Seidl, C. Neidlinger-Wilke, et al., Dynamic cell stretching increases human osteoblast proliferation and CICP synthesis but decreases osteocalcin synthesis and alkaline phosphatase activity, J. Biomech. 33 (2000) 45–51. [46] G.L. Galea, A. Sunters, L.B. Meakin, et al., Sost down-regulation by mechanical strain in human osteoblastic cells involves PGE2 signaling via EP4, FEBS Lett. 585 (2011) 2450–2454. [47] R.L.V. Bezooijen, B.A.J. Roelen, A. Visser, et al., Sclerostin is an osteocyte-expressed negative regulator of bone formation, but not a classical BMP antagonist, J. Exp. Med. 199 (6) (2004) 805–814. [48] S.M. Ponik, J.W. Triplett, F.M. Pavalko, Osteoblasts and osteocytes respond differently to oscillatory and unidirectional fluid flow profiles, J. Cell. Biochem. 100 (3) (2007) 794–807.

[49] A.R. Qian, L. Wang, X. Gao, et al., Diamagnetic Levitation causes changes in the morphology, cytoskeleton and focal adhesion proteins in osteocytes, IEEE Trans. Biomed. Eng. 59 (1) (2012) 68–77. [50] S.M. Di, A.R. Qian, L.N. Qu, et al., Graviresponses of osteocytes under altered gravity, Adv. Space Res. 48 (6) (2011) 1161–1166. [51] J.G. McGarry, J. Klein-Nulend, P.J. Prendergast, The effect of cytoskeletal disruption on pulsatile fluid flow-induced nitric oxide and prostaglandin E2 release in osteocytes and osteoblasts, Biochem. Biophys. Res. Commun. 330 (1) (2005) 341–348. [52] S.H. McBride, M.J. Silva, Adaptive and injury response of bone to mechanical loading, Bonekey Osteovis. 1 (2012) 1–8. [53] L.F. Bonewald, M.L. Johnson, Osteocytes, mechanosensing and Wnt signaling, Bone 42 (4) (2008) 606–615. [54] X. Li, C. Liu, P. Li, et al., Connexin 43 is a potential regulator in fluid shear stress-induced signal transduction in osteocytes, J. Orthop. Res. 31 (12) (2013) 1959–1965. [55] X. Xia, N. Batra, Q. Shi, et al., Prostaglandin promotion of osteocyte gap junction function through transcriptional regulation of connexin 43 by glycogen synthase kinase 3/beta-catenin signaling, Mol. Cell. Biol. 30 (1) (2010) 206–219. [56] N. Bivi, R. Pacheco-Costa, L.R. Brun, et al., Absence of Cx43 selectively from osteocytes enhances responsiveness to mechanical force in mice, J. Orthop. Res. 31 (7) (2013) 1075–1081. [57] E. Lau, S. Al-Dujaili, A. Guenther, et al., Effect of low-magnitude, high-frequency vibration on osteocytes in the regulation of osteoclasts, Bone 46 (6) (2010) 1508–1515. [58] A. Santos, A.D. Bakker, B. Zandieh-Doulabi, et al., Pulsating fluid flow modulates gene expression of proteins involved in Wnt signaling pathways in osteocytes, J. Orthop. Res. 27 (10) (2009) 1280–1287.