2 activated by fluid shear stress promotes osteogenic differentiation of human bone marrow-derived mesenchymal stem cells through novel signaling pathways

The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

Contents lists available at ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

Extracellular signal-regulated kinase1/2 activated by fluid shear stress promotes osteogenic differentiation of human bone marrow-derived mesenchymal stem cells through novel signaling pathways Liyue Liu a , Lan Shao a , Bo Li b , Chen Zong a , Jianhu Li c , Qiang Zheng d , Xiangming Tong c , Changyou Gao b , Jinfu Wang a,∗ a

Institute of Cell Biology, College of Life Sciences, Zhejiang University, Zijingang Campus, Hangzhou 310058, PR China Institute of Medical Materials, College of Material and Chemistry, Zhejiang University, Hangzhou, Zhejiang 310028, PR China c Laboratory of Bone Marrow, The First Hospital, Zhejiang University, Hangzhou, Zhejiang 310006, PR China d Institute of Orthopedics, The Second Hospital, Zhejiang University, Hangzhou, Zhejiang 310009, PR China b

a r t i c l e

i n f o

Article history: Received 25 May 2011 Received in revised form 13 July 2011 Accepted 20 July 2011 Available online 26 July 2011 Keywords: Human mesenchymal stem cells Extracellular signal-regulated kinase1/2 Osteogenic differentiation Fluid shear stress Molecular signaling network

a b s t r a c t It is a classical signaling pathway that the activation of extracellular signal-regulated kinase1/2 (ERK1/2) results in the phosphorylation of runt-related transcription factor 2 (Runx2) and thereby initiates the transcription of osteogenic genes. Recently, it is found that the activation of ERK1/2 resulted from fluid shear stress (FSS) also increased the expression of Runx2 and ␤1 integrins, and finally enhanced osteogenic differentiation. However, it has been remained largely unknown how ERK1/2 regulates the expression of Runx2 and ␤1 integrins. We use the perfusion culture system to produce FSS exerting on human bone marrow-derived mesenchymal stem cells (hMSCs) and thus activate ERK1/2. Our study demonstrated that FSS-activated ERK1/2 mediated the expression of osteogenic genes via two novel signaling pathways except for the classical signaling pathway: feedback up-regulation of ␤1 integrins expression via activating nuclear factor kappa B (NF-␬B), and activation of bone morphogenesis proteins (BMPs)/mothers against decapentaplegic (Smad) pathway via activating NF-␬B and thereby regulating Runx2 expression. These signaling pathways combined with the classical signaling pathway, with ERK1/2 as a hub node molecule, form a molecular signaling cross-talking network to induce the osteogenic differentiation of hMSCs. The understanding on the mechanism of FSS inducing the osteogenic differentiation of hMSCs will not only be helpful to develop the bone tissue engineering but also provide new targets for drug discovery for treatment of osteoporosis and other related bone-wasting diseases. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Extracellular signal-regulated kinase1/2 (ERK1/2), a member of mitogen-activated protein kinase (MAPK) family, is an important mediator of multiple factors-induced proliferation and differentiation in various cell types (Lai et al., 2001; Kim et al., 2007). In osteogenic differentiation, the activation of ERK1/2 leads to the phosphorylation of runt-related transcription factor 2 (Runx2), an important osteogenic differentiation related transcription factor, and in turn promotes the transcription of osteogenic genes such as alkaline phosphatase (ALP), osteocalcin (OCN), collagen I (COL I) and osteopontin (Riddle et al., 2006; McAllister et al., 2000; Grellier et al., 2009; Arnsdorf et al., 2009; Sharp et al., 2009). Meanwhile, bone morphogenesis proteins (BMPs)/mothers

∗ Corresponding author. Tel.: +86 571 88206592; fax: +86 571 85128776. E-mail address: [email protected] (J. Wang). 1357-2725/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2011.07.008

against decapentaplegic (Smad) pathway is considered to be an important signal transduction pathway that stimulates the expression of Runx2 (Phimphilai et al., 2006; Fan et al., 2006). Some BMPs, such as BMP2 and BMP4 are potent osteogenic agents that stimulate mesenchymal stem cells (MSCs) to differentiate into osteoblasts (Lavery et al., 2008). BMPs can bind to BMP receptor I to activate receptor-regulated Smads (R-Smads), such as Smad1, 5 and 8 (Yamamoto et al., 1997; Nohe et al., 2004). Activated Smad1, 5, and 8 form complexes with the common partner, Smad4, and translocate into nucleus to regulate the transcription of Runx2. Human mesenchymal stem cells (hMSCs) have attracted great attention because of their multiple differentiations potent. These cells can differentiate into osteoblasts in response to multiple environmental factors such as mechanical stimulation (Manton et al., 2007; Stiehler et al., 2009). The molecular mechanism of osteogenic differentiation of hMSCs induced by environmental factors was complex and involved multiple signaling molecules and pathways

1592

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

including ion channels, nitric oxide (NO), integrins, prostaglandin E2 (PGE2) and ERK1/2 (Orciani et al., 2009; Kleiveland et al., 2008; Glossop and Cartmell, 2009; Kundu et al., 2009), and there is strong evidence that ERK1/2 activation is an important downstream event of these pathways (Liu et al., 2010). Recent research found that activated ERK1/2 not only resulted in the activation of Runx2 but also promoted the expression of Runx2 and ␤1 integrins in the osteogenic differentiation of hMSCs (Kundu et al., 2009; Kanno et al., 2007). ␤1 integrins, as important mechano-receptors in mechano-transduction (Katsumi et al., 2004; Chen et al., 1999), can detect mechanical stimulation and activate downstream kinases, such as focal adhesion kinase (FAK), which result in ERK1/2 activation (Lee et al., 2008). Although there is no evidence on the direct relationship between the expression of integrins and the activation of integrin signaling pathway, over-expression of ␤1 integrins can lead to their redistribution in cytomembrane and enhance their association with ECM. Therefore, the feedback up-regulation of ␤1 integrins expression by ERK1/2 reinforces the response of hMSCs to mechanical stimulation. Above studies implied that ERK1/2 should mediate multiple signaling pathways to promote the expression of osteogenic genes. However, it remains unknown how ERK1/2 modulates the expression of both ␤1 integrins and Runx2. Nuclear factor kappa B (NF-␬B) transcription factors family control the expression of over one hundred genes (Pahl, 1999). This family is composed of homo- or hetero-dimers formed by five subunits: RelA (p65), c-rel, RelB, p50 and p52. Furthermore, ERK1/2 can also modulate the activation of NF-␬B (Rangaswami et al., 2004). Therefore, dose ERK1/2 activated by fluid shear stress (FSS) modulate the expression of ␤1 integrins and Runx2 through activation of NF-␬B? To verify this speculation, the perfusion culture system was used to imitate FSS and thereby activate ERK1/2 in hMSCs seeded into the porous poly lactic co-glycol acid (PLGA) 3-D scaffolds. We measured the expression levels of osteogenic genes, Runx2 and ␤1 integrins as well as the activation level of ERK1/2, FAK and Runx2 in hMSCs under FSS application, examined the role of ␤1 integrins in the FSS-induced activation of FAK and ERK1/2 as well as the effect of ERK1/2 on the activation of Runx2 and the expression of osteogenic genes, Runx2 and ␤1 integrins, assayed whether BMP/Smad pathway was indispensable to FSS-increased expression of Runx2, and determined whether FSS-activated ERK1/2 could activated NF-␬B as well as whether NF-␬B acted as important transcription factors in FSS-induced up-regulation of ␤1 integrins and BMP expression. Finally, a molecular signaling network modulating the expression of osteogenic genes in hMSCs, with ERK1/2 as a hub node molecule, was analyzed.

every three days. hMSCs at passage 3 were used for following experiments.

2.2. Fluid shear stress application Fluid shear stress application was performed using perfusion culture system. The system is composed of several culture cassettes, a multichannel cylinder pump (Cole Parmer MasterFlex 7524-55, Beijing, China) and a medium reservoir. hMSCs were seeded into PLGA 3-D scaffold as described previously (Yang et al., 2010). 3D PLGA scaffolds were friendly provided by Dr. Gao in College of Material and Chemistry, Zhejiang University. The scaffold was cylindrical (10 mm in diameter and 3 mm in height) and had a porosity of 95%. The mean pore diameter was 280–450 ␮m. Each scaffold was seeded with 1 × 106 hMSCs at passage 3, and then hMSC-scaffold complex was cultured with ordinary medium in the six-well plate overnight before hMSC-scaffold complex was placed into the culture cassette. The culture cassette and the medium reservoir were connected by silica gel tubes. Flow through each cassette was driven by the cylinder pump. The whole system was placed in an incubator that was maintained at 37 ◦ C with 5% CO2 in air. Intermittent FSS was loaded on cells in scaffold. Briefly, cellscaffold complex was exposed to FSS at a mean value of 4.2 dyn/cm2 (calculated according to literatures (Stephens et al., 2007; Bancroft et al., 2002)) for 1 h, and then was cultured under perfusion culture at flow rate of 0.3 ml/min (which may produce a low shear stress at mean value of 0.34 dyn/cm2 ) for 11 h as an interval. After the interval of 11 h, the cell-scaffold complex was again exposed to a FSS at mean value 4.2 dyn/cm2 for 1 h. In such way, the cell-scaffolds complex was exposed to FSS at 4.2 dyn/cm2 for 1 h twice a day. Ordinary medium was used and changed every three days. Cell-scaffold complexes were harvested for analysis after treated with FSS for 0 h, 1 h, 24 h, 4 d, 7 d, and 14 d. Two control cultures were set at the same time. In the positive control culture, cell-scaffold complex was cultured with osteogenic medium (D-MEM supplemented with 10% FBS, 50 ␮g/ml ascorbic acid, 10 mM sodium ␤-glycerophosphate, and 10−8 M dexamethasone; Sigma) under perfusion with a continuous flow rate of 0.3 ml/min. In the negative control culture, cell-scaffold complex was cultured with ordinary medium under perfusion with a continuous flow rate of 0.3 ml/min. Samples from both control cultures were collected at the corresponding times of intermittent FSS culture. Samples collected at time points of 0 h, 4 d, 7 d, 14 d were used for analysis of osteogenic differentiation, and those collected at time points of 1 h, 24 h, 4 d, 7 d, 14 d were employed for Western-bolt or immunoprecipitation. 2.3. Real-time PCR analysis

2. Materials and methods 2.1. Cell culture Human bone marrow was kindly provided by healthy male donors for annual health examination in the First People’s Hospital of Zhejiang, ranging in age from 19 to 45 years, given written consent to the use of bone marrow for research purpose according to procedures approved by the human experimentation committee at Zhejiang Public Health Bureau. hMSCs were isolated and cultured following a previously reported method with some modifications (Digirolamo et al., 1999). Briefly, hMSCs were cultured in the ordinary medium consisting of alpha minimum essential medium (␣-MEM, Gibco-BRL, Hangzhou, China), 10% fetal bovine serum (FBS, Gibco-BRL), 100 U/ml penicillin and 100 ␮g/ml streptomycin (Life Technologies, Beijing, China). Medium was changed

Total RNA was extracted from cell-scaffold using TRIzol (TaKaRa, Dalian, China) according to the manufacturer’s instructions. RNA sample was treated with DNase I (Fermentas, Shanghai, China) to remove residual genomic DNA. Total RNA was quantified at an absorbance of 260 nm by a spectrophotometer (Jenway Genova, Hangzhou, China). The RNA sample had an A260:A280 ratio of 2.0 to guarantee high purity. 2 ␮g of total RNA from each sample was subjected to first-strand cDNA synthesis using RevertAid First Strand cDNA synthesis Kit (Fermentas) with oligo d(T). mRNA of tested gene was quantified by the real-time PCR performed in Bio-rad Icycler 3.0 (Biorad, Shanghai, China) using QIAGEN Green Master Mix. Primer sequences were derived from genes sequences available through GenBank (Table 1). Quantification of genes mRNA was performed using the comparative threshold cycle method (Ct) with GAPDH as the internal reference, and relative gene expression was reported as 2−CT .

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

1593

Table 1 Primers for real-time PCR. GeneBank accession nos. ALP

NM 000478v

Runx2

NM 001024630

OCN

X53698

COL I␣

NM 000088

GAPDH

NM 002046

BMP2

NM 001200

BMP4

NM 130850

Annealing temperature (◦ C)

Primer sequence 



Forward: 5 -TGGACCTCGTTGACACCTGGAA-3 Reverse: 5 -CACCACCACCATCTCGGAGAGT-3 Forward: 5 -GCCACCTCTGACTTCTGCCTCT-3 Reverse: 5 -ATGGAGTGCTGCTGGTCTGGAA-3 Forward: 5 -AGCCACCGAGACACCATGAGAG-3 Reverse: 5 -GTGCCTGGAGAGGAGCAGAACT-3 Forward: 5 -ACCTGCCGTGACCTCAAGATGT-3 Reverse: 5 -ATGCTCTCGCCGAACCAGACA-3 Forward: 5 -TGACCACAGTCCATGCCATCAC-3 Reverse: 5 -CGCCTGCTTCACCACCTTCTT-3 Forward: 5 -GAGCCATTCCGTAGTGCCATCC-3 Reverse: 5 -CAGAAGTGTCGCCTCGAAGTCC-3 Forward: 5 -GTATCGCAGGCACTCAGGTCAG-3 Reverse: 5 -GCCACTTCCACCACGAATCCAT-3

2.4. Alkaline phosphatase activity assay Cellular ALP activity was measured using an ALP measurement kit (Jiancheng Biotechnology Institute, Nanjing, China) according to the manufacturer’s instructions. The activity of ALP was normalized against the phenol standards and expressed as ␮g p-nitrophenol produced per minute per million cells. The cell number of each scaffold had been assessed by quantifying DNA content using a Hoechst 33258 dye (Sigma, Hangzhou, China) assay followed the method as reported in previous study (Yang et al., 2010). The cell number of a scaffold is 1.0–2.4 × 106 . 2.5. Western-blot analysis Cell-scaffold was homogenized by sonication in RIPA lysis buffer (Beytime Biotech, Jiangsu, China) with 1 mM PMSF (Sigma, Shanghai, China) and 0.1% phosphatase inhibitors (Sigma). Cell homogenate was centrifuged at 12,000 × g for 5 min at 4 ◦ C and the supernatant was collected. The protein concentration of supernatant was determined by bicinchoninic acid (BCA) protein assay (Pierce Biotechnologies, Hangzhou, China). The 2× sample buffer was added to equal volume of each cell lysate sample and boiled for 5 min. Then, 15–25 ␮l lysate (50 ␮g protein every lane) was applied to SDS-PAGE in 8% gel for 2 h and a half at a voltage of 65 V. The proteins in the gel were transferred to PVDF membrane (Millipore). After blocked with 5% skimmed milk, the membrane was incubated at 4 ◦ C overnight in primary antibody against ERK1/2, p-Smad1/5/8 (1:1000, Cell Signaling Technology), p-ERK1 (pT202/Py204)/ERK2 (pT185/Py187), ␤1 integrin, FAK, p-FAK (Py397) (1:1000, Epitomics), ␤-tubulin (1:10000, Epitomics) and Runx2 (1:800, Abcam), followed by incubation at room temperature for 1 h in horseradish peroxidase (HRP)-conjugated goat anti-rabbit (1:3000, Huaan, Hangzhou, China) after washed with PBST (PBS supplemented 0.1% Tween-20) for three times. Immunoreactive bands were detected using the enhanced chemiluminescent reagent (Tiangen Biotech) and quantitatively analyzed in triplicate by normalizing band intensities to the controls on scanned films by Image J software.

60 60 60 60 59 59 59

of elution buffer was used to elute the labeled protein. Westernblot was performed to detect the phosphotyrosine level of Runx2 using phosphotyrosine antibody (1:1000, Millipore Corporation). Immunoreactive bands were quantitatively analyzed in triplicate by normalizing band intensities to total Runx2 on scanned films by Image J Software.

2.7. Immunofluorescence staining Cell-scaffold was paraffin-sectioned to a thickness of 6 ␮m. Sections were immunofluorescence-labeled according to the manufacturer’s instructions using a Cellular NF-␬B Translocation Kit (Beyotime Biotech, Suzhou, China) (Musa et al., 2006). Stained sections were analyzed with a Nikon fluorescence microscope (Nikon, Shanghai, China).

2.8. Selective inhibitors of signal transduction pathways To access the role of several signal transduction pathways in the FSS-induced osteogenic differentiation of hMSCs, we used selective inhibitors to inhibit the test signal transduction pathways: PD98059 (20 ␮M; Sigma) to ERK1/2 pathway, human recombinant noggin (10 ng/ml; Peprotech) to BMPs pathway, BAY11-7082 to NF-␬B (10 ␮M; Sigma) pathway, RGDS peptide (500 ␮g/ml; Tashi, Shanghai, China) to ␤1 integrins pathway. Briefly, cells seeded in scaffold were deprived of FBS overnight and then pretreated for 2 h with three inhibitors excluding RGDS before exposed to intermittent FSS. For RGDS treatment, cells were pre-incubated with RGDS peptide for 2 h before seeded into scaffold, and then cells were allowed to adhere to scaffold for 4 h before exposed to intermittent FSS. All above inhibitors were present during intermittent FSS loading.

2.6. Immunoprecipitation analysis 2.9. Statistical analysis Immunoprecipitation was performed to detect the phophoRunx2 level using immunoprecipitation kit (Genmed Scientifics Inc. Shanghai, China). Cell-scaffold was rinsed with cleaning buffer and homogenated in lysate buffer. 5 ␮g of Runx2 antibody (Abcam) was added into 250 ␮g of cell lysate following overnight incubation on a rocking-bed at 4 ◦ C. The supernatant was discarded after centrifuging at 3000 × g for 2 min and 50 ␮l of protein A resin was added to catch labeled protein incubated on a rocking-bed at 4 ◦ C for 1 h. 1 ml

All experiments were repeated for at least three times and the representative data were presented as means ± standard deviation where indicated. Statistical analysis was performed using factorial analysis of variance (two-way ANOVA), followed by Dunnett’s test for comparing treatments with controls. A probability value of less than 0.05 was statistically considered significant: * (P < 0.05), ** (P < 0.01), *** (P<0.001).

1594

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

Fig. 1. The expression level of ALP, Runx2, COL I␣, OCN, BMP2 and BMP4 in hMSCs from different culture: intermittent FSS culture (F), intermittent FSS plus osteogenic medium culture as a positive control culture (OM), perfusion culture at continuous 0.3 ml/min flow rate of ordinary medium as a negative control culture (F0.3 ), intermittent FSS plus PD98059 culture (PF), intermittent FSS plus Bay 11-7082 (BF) and intermittent FSS plus noggin (NF). (A) qRT-PCR analysis of ALP expression of hMSCs from different cultures. The mRNA level of ALP of samples from intermittent FSS culture was higher than those from negative control culture and intermittent FSS plus PD98059 culture. (B) Intermittent FSS as well as osteogenic medium can lead to high expression of Runx2 in hMSCs, and the intermittent FSS-induced up-regulation of Runx2 expression could be abrogated by PD98059, noggin and BAY 11-7082. (C and D) The expression level of OCN and COL I␣ in cells from intermittent FSS culture was significantly higher than those in cells from negative control culture and intermittent FSS plus PD98059 culture. (E and F) Intermittent FSS significantly increased the expression levels of both BMP2 and BMP4, but this up-regulation could be abolished by PD98059 and BAY. All results were representative of 3 experiments. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. *P < 0.05 was statistically considered significant, **P < 0.01 was statistically considered very significant, and ***P < 0.001 was considered extremely significantly.

3. Results 3.1. ERK1/2 mediates intermittent FSS-induced up-regulation of osteogenic gene expression by modulating the activation of Runx2 FSS at a mean value of 4.2 dyn/cm2 was applied intermittently to hMSCs in the scaffold and the mRNA level of ALP, OCN, COL I␣ and Runx2 was measured by quantitative real-time PCR (qPCR). The results showed that the expression level of ALP and Runx2 in hMSCs from intermittent FSS culture began to increase on day 4 and was significantly higher than those in hMSCs from negative control group (P < 0.05) (Fig. 1A and B). After 7 days of intermittent FSS culture, the expression level of OCN and COL I␣ also increased and was significantly higher than those in hMSCs from negative control culture (P < 0.05) (Fig. 1C and D). The expression level of ALP in hMSCs from intermittent FSS culture was similar to that in hMSCs from positive control culture (P > 0.05) (Fig. 1A). The expression levels of Runx2 and COL I␣ in hMSCs from intermittent FSS culture were also similar to those in hMSCs from positive control culture, especially at later stages (P > 0.05) (Fig. 1B and D). However, the expression level

of OCN in hMSCs from intermittent FSS culture was significantly lower than that in hMSCs from positive control culture on day 7 and day 14 (P < 0.05) (Fig. 1C). In addition, the activity of ALP in hMSCs from intermittent FSS culture was examined. The results showed that the ALP activity in hMSCs from intermittent FSS culture was not significantly different from that in hMSCs from positive control culture (P > 0.05), but markedly higher than that in hMSCs from negative control culture (P < 0.001) (Fig. 2). To investigate the mechanism that FSS led to up-regulation of osteogenic genes, we examined the phosphorylation of ERK1/2 and Runx2 by Western-blot and immunoprecipitation. A robust and sustained activation of ERK1/2 was observed in hMSCs from intermittent FSS culture while the phosphorylation of ERK1/2 in hMSCs from negative control culture was kept at a low level (Fig. 3A1, A2 and A5). Furthermore, the activation level of Runx2 in hMSCs from intermittent FSS culture also increased and was significantly higher than that in hMSCs from negative control culture (P < 0.01) (Fig. 3B1, B2 and B4). To examine whether the phosphorylation of ERK1/2 was essential to the intermittent FSS-induced activation of Runx2 and the

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

1595

intermittent FSS-induced expression of osteogenic genes, we used PD98059, a selective inhibitor of ERK1/2, to inhibit the activation of ERK1/2. The results showed that PD98059 inhibited the phosphorylation of ERK1/2 (Fig. 3A3 and A5). The intermittent FSS-induced phosphorylation of Runx2 was also abolished completely by PD98059 (Fig. 3B3 and B4). Moreover, PD98059 resulted in decrease of ALP, OCN, COL I␣ and Runx2 expression (Fig. 1A–D) and of ALP activity (Fig. 2). These results suggested that ERK1/2 was indispensable to the intermittent FSS-induced osteogenic differentiation of hMSCs and ERK1/2 mediated the expression of osteogenic genes by regulating the activation of Runx2.

Fig. 2. The ALP activity in hMSCs from different cultures: intermittent FSS culture (F), intermittent FSS plus osteogenic medium culture as a positive control culture (OM), perfusion culture at continuous 0.3 ml/min flow rate of ordinary medium as a negative control culture (F0.3 ), intermittent FSS plus PD98059 culture (PF). The ALP activity of hMSC-scaffold complexes from four cultures was analyzed quantitatively using ALP activity measurement kit. The activity of ALP was showed as ␮g p-nitrophenol produced per minute per million cells. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. *P < 0.05 was statistically considered significant, **P < 0.01 was statistically considered very significant, and ***P < 0.001 was considered extremely significantly.

3.2. ˇ1 integrins act as important mechano-receptors in the mechano-transduction of intermittent FSS Above results showed that intermittent FSS led to the activation of intracellular signal molecules such as ERK1/2 in hMSCs (Fig. 3). However, it remained elusive how the extracellular intermittent FSS signal was converted into the intracellular molecular signal in hMSCs. ␤1 integrins were considered as important mechanoreceptors in other types of cells and were proved to be very important in ECM-mediated osteogenic differentiation of hMSCs (Kundu et al., 2009; Lee et al., 2008). Therefore, we suspected that ␤1 integrins also acted as important mechano-sensors in the mechanism of osteogenic differentiation of hMSCs induced by intermittent FSS. To testify this possibility, we examined the tyrosine 397 phospho-

Fig. 3. Fluid shear stress increased the phosphorylation level of ERK1/2 and Runx2. (A) The phosphorylation level of ERK1/2 under four different conditions: (A1) the phosphorylation level of ERK1/2 in cells from negative control culture (F0.3 ); (A2) the phosphorylation level of ERK1/2 in cells from intermittent FSS culture (F); (A3) the phosphorylation level of ERK1/2 in cells from intermittent FSS plus PD98059 culture (PF); (A4) the phosphorylation level of ERK1/2 in cells from intermittent FSS plus RGDS peptide culture (RF); (A5) densitometric measure of band intensity for phospho-ERK1/2 was analyzed by Image J Software and normalized by corresponding total ERK1/2. (B) The phosphorylation level of Runx2 under four different conditions: (B1) the phosphorylation of Runx2 in cells from negative control culture; (B2) the phosphorylation of Runx2 in cells from intermittent FSS culture; (B3) the phosphorylation of Runx2 in cells from intermittent FSS plus PD98059 culture; (B4) densitometric measure of band intensity for p-Runx2 was analyzed by Image J Software and normalized by corresponding total Runx2. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. * (P < 0.05) was statistically significant, ** (P < 0.01) was statistically very significant and *** (P < 0.001) was extremely significant.

1596

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

Fig. 4. ␤1 integrins acting as important mechano-receptors in the mechano-transduction of intermittent FSS. Intermittent FSS can enhance the activation of FAK and the expression of ␤1 integrins. The selective inhibitors of ERK1/2, ␤1 integrins and NF-␬B can abrogate the enhancement of ␤1 integrins expression. The blockade of the connection between ␤1 integrins and ECM by RGDS peptide can abolish FSS-induced up-regulation of the phosphorylation of FAK. (A) The protein level of ␤1 integrins in cells under four different conditions: (A1) the protein level of ␤1 integrins in cells from negative control culture (F0.3 ); (A2) the protein level of ␤1 integrins in cells from intermittent FSS culture (F); (A3) the protein level of ␤1 integrins in cells from intermittent FSS plus PD98059 culture (PF); (A4) the protein level of ␤1 integrins in cells from intermittent FSS plus BAY 11-7082 culture (BF); (A5) densitometric measure of band intensity for ␤1 integrins was analyzed by Image J Software and normalized by corresponding housekeeping gene ␤-tubulin. (B) The phosphorylation level of FAK in cells under three different conditions: (B1) the phosphorylation level of FAK in cells from negative control culture; (B2) the phosphorylation level of FAK in cells from intermittent FSS culture; (B3) the phosphorylation level of FAK in cells from intermittent FSS plus RGDS culture; (B4) densitometric measure of band intensity for phospho-FAK was analyzed by Image J Software and normalized by corresponding total FAK. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. * (P < 0.05) was statistically significant, ** (P < 0.01) was statistically very significant and *** (P < 0.001) was extremely significant.

rylation level of FAK and then used RGDS peptide to block the connection between ␤1 integrins and ECM. The results showed that the phosphorylation of FAKtyr397 in hMSCs was significantly increased by intermittent FSS as compared with that in hMSCs from negative control culture (P < 0.001) (Fig. 4B1, B2 and B4). However, after cells were treated with RGDS peptide, intermittent FSS failed to activate FAK in hMSCs (Fig. 4B3 and B4), and the activity of ERK1/2 in hMSCs from intermittent FSS plus RGDS culture was also as low as that in hMSCs from negative control culture (P > 0.05) (Fig. 3A4 and A5). The results indicated that blocking the connection between ␤1 integrins and ECM could prevent intermittent FSS from activating ERK1/2, which evidently demonstrated that ␤1 integrins acted as important mechano-receptors of intermittent FSS in hMSCs. 3.3. ERK1/2 regulates the intermittent FSS-induced expression of ˇ1 integrins by activating NF-B The expression level of ␤1 integrins in hMSCs from intermittent FSS culture was examined to test whether intermittent FSS could up-regulate the expression of ␤1 integrins. It was found that the protein level of ␤1 integrins in hMSCs from intermittent FSS culture increased on day 4 and was significantly higher from day 4 to day 14 than that in hMSCs from negative control culture (P < 0.05) (Fig. 4A1, A2 and A5). Interestingly, we also found that the ␤1 integrins protein level in hMSCs from intermittent FSS plus PD98059

culture was as low as that in hMSCs from negative control culture (P > 0.05) (Fig. 4A1, A3 and A5). These results suggested that intermittent FSS could promote the expression of ␤1 integrins, and the intermittent FSS-induced up-regulation of ␤1 integrins expression was dependent on the activation of ERK1/2. The activation of MAPK/ERK1/2 could lead to the phosphorylation of IKB␣ and NF-␬B. Therefore, we suspected that NF-␬B should act as an important transcription factor in the intermittent FSS-induced up-regulation of ␤1 integrins and that ERK1/2 might modulate the expression of ␤1 integrins by crosstalking with NF-␬B. To confirm this hypothesis, we examined the nuclear translocation of NF-␬B, the phosphorylation level of NF-␬B p65 and IKB␣, and the effect of BAY 11-70820, a selective inhibitor of NF␬B, on the intermittent FSS-induced up-regulation of ␤1 integrins expression. As shown in Fig. 5, the phosphorylation levels of both p65 and IKB␣ in hMSCs increased significantly after intermittent FSS application for 1 h and were kept at a higher level during the period of intermittent FSS culture as compared with those in hMSCs from negative control culture (P < 0.05) (Fig. 5A1, A2, A5, B1, B2 and B5). However, the phosphorylation levels of NF-␬B p65 and IKB␣ in hMSCs from intermittent FSS plus PD98059 culture were kept at a very low level and were as low as those in hMSCs from negative control culture (Fig. 5A1, A3, A5, B1, B3 and B5). This indicated that intermittent FSS could up-regulate the activation of NF-␬B p65, and this up-regulation could be abrogated by the selective inhibitor of ERK1/2. The result was further confirmed by immunofluores-

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

1597

Fig. 5. The phosphorylation level of NF-␬B subunit p65 and IKB␣ induced by intermittent FSS can be decreased by PD98059. (A) The phosphorylation level of NF-␬B subunit p65 under four different conditions; (A1) the phosphorylation of p65 in cells from negative control culture (F0.3 ); (A2) the phosphorylation of p65 in cells from intermittent FSS culture (F); (A3) the phosphorylation of p65 in cells from intermittent FSS plus BAY 11-7082 culture (BF); (A4) the phosphorylation of p65 in cells from intermittent FSS plus PD98059 culture (PF); (A5) densitometric measure of band intensity for p-p65 was analyzed by Image J Software and normalized by corresponding total p65. (B) The phosphorylation level of IKB␣ subunit p65 under four different conditions: (B1) the phosphorylation of IKB␣ in cells from negative control culture; (B2) the phosphorylation of IKB␣ in cells from intermittent FSS culture; (B3) the phosphorylation of IKB␣ in cells from intermittent FSS plus BAY 11-7082 culture; (B4) the phosphorylation of IKB␣ in cells from intermittent FSS plus PD98059 culture; (B5) densitometric measure of band intensity for p-p65 and p-IKB␣ was analyzed by Image J Software and normalized by corresponding total p65 or ␤-tubulin. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. * (P < 0.05) was statistically significant, ** (P < 0.01) was statistically very significant and *** (P < 0.001) was extremely significant.

Fig. 6. The nuclear-translocation of NF-␬B in hMSCs from negative control culture (F0.3 ), intermittent FSS culture (F) and intermittent FSS plus PD980589 culture (PF). The nuclear translocation of NF-␬B was assayed by Immunofluorescence staining. Fluorescence microscopy (200×) showed the location of NF-␬B subunit, p65. The location of p65 (red fluorescence) was compared with nuclear (DAPI, blue). The nuclear translocation of p65 was increased by FSS but could be reduced by the inhibitor of ERK1/2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

1598

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

cence (Fig. 6). The nuclear translocation of p65 NF-␬B in hMSCs from intermittent FSS culture increased greatly. However, intermittent FSS failed to promote the nuclear translocation of p65 NF-␬B in the presence of PD98059 (Fig. 6). Then, we used BAY 11-7082, a selective inhibitor of NF-␬B, to treated hMSCs before and during intermittent FSS application and measured the protein level of ␤1 integrins. With the treatment of BAY 11-7082, intermittent FSS failed to increase the phosphorylation of IKB (Fig. 5B4 and B5). The protein level of ␤1 integrins in hMSCs from intermittent FSS plus BAY 11-7082 culture was similar to that in hMSCs from negative control culture from 1 h to day 14 (P > 0.05) (Fig. 4A1, A4 and A5), and was significantly lower than that in hMSCs from intermittent FSS culture on day 4 to day14 (P < 0.05) (Fig. 4A2, A4 and A5). The phosphorylation of p65 in hMSCs from intermittent FSS plus BAY 11-7082 culture was almost as high as that in hMSCs from intermittent FSS culture (P > 0.05) (Fig. 5A2, A4 and A5). The above results proved that the intermittent FSS-activated ERK1/2 promoted the expression of ␤1 integrins by modulating the activation and nuclear translocation of NF-␬B. 3.4. ERK1/2 regulates the expression of Runx2 by modulating the activation of NF-B and activating the BMPs/Smad pathway ERK1/2 had been proved to be critical to the intermittent FSS-induced up-regulation of Runx2 expression in above results. However, how does ERK1/2 regulate the expression of Runx2? BMPs/Smad pathway is believed to be a key signaling pathway responsible for the expression of Runx2. Therefore, we suspected that ERK1/2 modulated the expression of Runx2 by cross-talking with BMPs/Smad pathway. To make sure whether BMPs/Smad was involved in the mechano-transduction of intermittent FSS in hMSCs, we probed the mRNA level of BMP2 and BMP4 using qPCR. The results showed that the mRNA level of both BMP2 and BMP4 in hMSCs from intermittent FSS culture was markedly increased as compared with those in hMSCs from negative control culture and intermittent FSS plus PD98059 culture (P < 0.05) (Fig. 1E and F). It indicated that intermittent FSS could up-regulate the expression of BMPs and that this up-regulation was influenced by the activation of ERK1/2. Then, human recombinant noggin, a selective antagonist of BMPs, was used to block the BMPs/Smad pathway. The phospho-Smad1/5/8 level and the mRNA level of Runx2 were measured. The results showed that the phospho-Smad1/5/8 level in hMSCs from intermittent FSS culture was significantly higher than those in hMSCs from negative control culture, intermittent FSS plus noggin culture and intermittent FSS plus PD98059 culture (P < 0.01) (Fig. 7A–D and F). It indicated that intermittent FSS could lead to the activation of Smad1/5/8. The intermittent FSSinduced activation of Smad1/5/8 was dependent on the activation of ERK1/2, and the activation of ERK1/2 could result in the activation of Smad1/5/8 through up-regulation of BMP expression. Moreover, the intermittent FSS-induced up-regulation of Runx2 expression could be abrogated by noggin (Fig. 1B). Therefore, the activation of BMPs/Smad pathway was indispensable to the intermittent FSSinduced up-regulation of Runx2 expression. The further question was whether the intermittent FSSactivated ERK1/2 could up-regulate the expression of BMPs through activating NF-␬B in hMSCs? Hence, intermittent FSS plus BAY 11-7082 culture was used to treat hMSCs, and then the phosphorylation levels of Smad1 (Ser463/465)/Smad5 (Ser463/465)/Smad8 (Ser426/428) and the mRNA levels of BMP2, BMP4 and Runx2 were measured. Interestingly, the intermittent FSS-induced high expression of BMPs and Runx2 was abolished by BAY 11-7082 (Fig. 1B–F). In addition, the phospho-Smad1/5/8 level in hMSCs from intermittent FSS plus BAY 11-7082 culture was significantly lower than that in hMSCs from intermittent FSS culture (Fig. 7B–F). The above results showed that ERK1/2 influenced the intermittent

FSS-induced up-regulation of Runx2 by modulating the activation of NF-␬B to regulate the activation of BMPs/Smad pathway.

4. Discussion In this study, FSS produced by flowing medium was exerted on hMSCs seeded in the PLGA porous scaffold. The mechanics environment of cells in this 3D perfusion culture is similar to that of cells in vivo (Scaglione et al., 2008; Bancroft et al., 2002). Therefore, our results should reflect the actual effect of FSS on hMSCs in bone in vivo. Our previous study had proved that a low flow rate of 0.3 ml/min had little effect on the osteogenic differentiation of hMSCs but can improve the delivery of oxygen and nutrient and the removal of metabolites (Yang et al., 2010). Therefore, we set a continuous perfusion culture at 0.3 ml/min flow rate of ordinary medium as negative control culture, and hMSCs in the interval period of intermittent FSS culture was also treated with perfusion culture at 0.3 ml/min flow rate. In our study, the mRNA level of Runx2, ALP, COL I␣ and OCN and the activity of ALP was examined to determine the effect of intermittent FSS on the osteogenic differentiation of hMSCs. The results showed that the mRNA levels of these osteogenic genes and the ALP activity increased with intermittent FSS application. It confirmed previous studies that intermittent FSS could induce the osteogenic differentiation of hMSCs (Glossop and Cartmell, 2009; Stiehler et al., 2009). We also examined the activation of ERK1/2 under intermittent FSS application, and used the selective inhibitor of ERK1/2, PD98059, to determine the effect of ERK1/2 on the intermittent FSS-induced osteogenic differentiation of hMSCs. Our result showed that intermittent FSS could lead to ERK1/2 activation and that ERK1/2 was critical to the intermittent FSS-induced osteogenic differentiation of hMSCs. However, it remained unclear how the mechanical signal of intermittent FSS was detected by hMSCs and transmitted into hMSCs to activate intracellular signal molecules such as ERK1/2. ␤1 integrins were considered as important mechanoreceptors and had been proved to play a critical role in responses of osteoblasts to FSS (Lee et al., 2008; Pavalko et al., 1998; Chen et al., 1999). In addition, FAK is an important kinase that can disseminate integrin signal and lead to the activation of ERK1/2 via Grb2-Sos-Ras pathway (Laser et al., 2000; Young et al., 2009). To determine whether the intermittent FSS-induced ERK1/2 activation in hMSCs was also dependent on the sensation of ␤1 integrins to the extracellular FSS signal, we examined the phosphorylation level of FAK in hMSCs with intermittent FSS application and the effect of blocking ␤1 integrins with RGDS peptide on the intermittent FSS-induced activation of ERK1/2 and FAK. It was found that intermittent FSS increased markedly the phosphorylation level of FAK in hMSCs. However, intermittent FSS failed to activate ERK1/2 and FAK after hMSCs were treated with RGDS peptide. This indicated that ␤1 integrins in hMSCs also acted as important mechanoreceptors to sense the stimulation of intermittent FSS and activate MAPK/ERK1/2 pathway by promoting the autophosphorylation of FAK. In addition, previous studies had demonstrated that the expression of ␤1 integrins was associated with the activation of integrin signaling pathway (Lee et al., 2000; Carvalho et al., 1995). Meanwhile, FSS could increase the expression of ␤1 integrins in other cell lines (Kapur et al., 2003; Soghomonians et al., 2002). It is possible that the expression of ␤1 integrins in hMSCs might also increase with FSS application. To access this possibility, we measured the protein level of ␤1 integrins in hMSCs treated with intermittent FSS. Our study showed that the protein level of ␤1 integrins in hMSCs increased significantly with intermittent FSS application. It indicated that intermittent FSS could up-regulate the expression of ␤1 integrins in hMSCs. Further, we used PD98059 to investigate the effect of ERK1/2 on the intermittent FSS-induced up-regulation of ␤1 inte-

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

1599

Fig. 7. Intermittent FSS-induced up-regulation of p-Smad1/5/8 can be decreased by PD98059, Noggin or BAY 11-7082. (A) The phosphorylation level of Smad1/5/8 in cells from negative control culture (F0.3 ). (B) The phosphorylation level of Smad1/5/8 in cells from intermittent FSS culture (F). (C) The phosphorylation level of Smad1/5/8 in cells from intermittent FSS plus PD98059 culture (PF). (D) The phosphorylation level of Smad1/5/8 in cells from intermittent FSS plus Noggin culture (NF). (E) The phosphorylation level of Smad1/5/8 in cells from intermittent FSS plus BAY 11-7082 culture (BF). (F) Densitometric measure of band intensity for p-Smad1/5/8 was analyzed by Image J Software and normalized by corresponding ␤-tubulin. The results were analyzed using two-way ANOVA followed by Dunnett’s test for comparing treatments with controls. The values are mean ± SD. * (P < 0.05) was statistically significant, ** (P < 0.01) was statistically very significant and *** (P < 0.001) was extremely significant.

grins expression. The treatment of PD98059 almost abolished the intermittent FSS-induced up-regulation of ␤1 integrins expression. In previous studies, it was controversial whether ERK1/2 had effect on the expression of ␤1 integrins (Kapur et al., 2003; Communal et al., 2003). Kapur et al. (2003) reported that ERK1/2 had no effect on the expression of ␤1 integrins. However, Communal et al. (2003) found that the inhibition of ERK1/2 could abolish ␣1-adrenergic receptors -stimulated increases of ␤1 integrins expression. Our study supports the opinion that the intermittent FSS-activated ERK1/2 is critical to the intermittent FSS-induced up-regulation of ␤1 integrins expression. Runx2 is an important osteogenic transcription factor which can bind to osteoblast-specific-acting element (OSE2) in the promoter region of osteogenic genes to initiate the expression of these genes (Ducy et al., 1997; Franceschi and Xiao, 2003). In osteogenic differentiation of hMSCs, the activation of ERK1/2 leads to the phosphorylation of Runx2 and in turn promotes the transcription of osteogenic genes. It is a classical signaling pathway (Riddle et al., 2006; McAllister et al., 2000; Grellier et al., 2009; Arnsdorf et al., 2009; Sharp et al., 2009). However, our further investigation with the PD98059 treatment demonstrated that intermittent FSS-activated ERK1/2 led not only to the phosphorylation of Runx2 but also to the increase of Runx2 expression and thereby enhanced the expression of osteogenic genes. How did ERK1/2 regulate the expression of Runx2? Although previous studies had also proved that the mechanical stimulation-induced up-regulation of Runx2 expression was dependent on ERK1/2 activation (Liu et al., 2009; Kanno et al., 2007), no reports elucidated this mechanism. Since it was widely accepted that the activation of BMPs/Smad pathway could lead to Runx2 expression, a question arose whether ERK1/2 influenced the intermittent FSS-induced up-regulation of Runx2 expression via cross-talking with BMPs/Smad pathway. Therefore, we measured the mRNA level of BMP2 and BMP4 in hMSCs exposed to intermittent FSS. The results showed that the mRNA level of both BMP2 and BMP4 in hMSCs with intermittent FSS application increased significantly. Companied with the high mRNA level

of BMPs, the phosphorylation of Smad1/5/8 was enhanced. The treatment of PD98059 abrogated the intermittent FSS-induced up-regulation of BMPs expression and Smad1/5/8 activation. Furthermore, the antagonist of BMPS, noggin, almost completely inhibited the intermittent FSS-increased Runx2 expression and the Smad1/5/8 phosphorylation. Therefore, ERK1/2 influenced the expression of Runx2 by modulating the expression level of BMPs to activate BMPs/Smad pathway. The final question is how the intermittent FSS-activated ERK1/2 up-regulates the expression of BMPs and ␤1 integrins in hMSCs. NF␬B transcription factors family control the expression of over one hundred genes (Pahl, 1999). Our study showed that the inhibition of ERK1/2 could abolish the intermittent FSS-induced up-regulation of both ␤1 integrins and BMPs expression. Moreover, previous studies indicated that ERK1/2 could regulate the activation of NF-␬B (Hsu et al., 2010; Rangaswami et al., 2004). Therefore, it is possible that ERK1/2 modulated the intermittent FSS-induced up-regulation of ␤1 integrins and BMPs in hMSCs via regulating the activation of NF-␬B. To verify this possibility, we examined the activation and nuclear translocation of NF-␬B p65 in hMSCs with intermittent FSS application, and used inhibitors of ERK1/2 and NF-␬B to determine the effect of ERK1/2 on the activation of NF-␬B and the effect of NF␬B on the intermittent FSS-induced up-regulation of ␤1 integrins, BMPs and Runx2 expression. The results showed that the nuclear translocation of NF-␬B p65 was enhanced by intermittent FSS. Likewise, the phosphorylation of p65 and IKB␣ in hMSCs was also increased by intermittent FSS. PD98059 abrogated the intermittent FSS-induced activation and nuclear translocation of NF-␬B p65. It indicated that the intermittent FSS-induced NF-␬B activation was dependent on ERK1/2 activation. In addition, the selective inhibitor of NF-␬B, BAY 11-7082, almost abolished the intermittent FSSinduced up-regulation of ␤1 integrins, BMPs and Runx2 expression. The intermittent FSS-induced activation of Smad1/5/8 was also inhibited by BAY 11-7082. Therefore, these results supported our hypothesis that the intermittent FSS-activated ERK1/2 led to NF-␬B activation. The activated NF-␬B translocated into nuclear to initi-

1600

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601

Fig. 8. Molecular signaling network involved in the intermittent FSS-activated ERK1/2 mediating the osteogenic differentiation of hMSCs.

ate the transcription of ␤1 integrins and BMPs, and BMPs combined with its receptors and thereby activated Smad1/5/8. The phosphorylated Smad1/5/8 resulted in the expression of Runx2. In summary, our study has not only confirmed that intermittent FSS could induce the osteogenic differentiation of hMSCs but also demonstrated that two novel important signaling pathways involved in the mechanotransduction of intermittent FSS in hMSCs, combined with the classical signaling pathway, formed a molecular signaling network (with ERK1/2 as a hub node molecule) to enhance the osteogenic differentiation of hMSCs (Fig. 8). First, ␤1 integrins act as important mechanoreceptors to sense the stimulation of extracellular intermittent FSS and in turn activate ERK1/2 by activating FAK. The activated ERK1/2 leads to the phosphorylation of Runx2, and the phosphorylated Runx2 initiates the transcription of osteogenic genes to promote hMSCs to differentiate into osteoblasts. Second, the intermittent FSS-activated ERK1/2 increases the expression of BMPs via activating NF-␬B, the increased BMPs results in the activation of BMPs/Smad pathway and finally leads to the expression of Runx2. Third, the intermittent FSS-activated ERK1/2 influences the expression of ␤1 integrins by mediating the activation of NF-␬B. We have demonstrated novel signaling pathways involving in how ERK1/2 regulates the expression of ␤1 integrins and Runx2. Our study has provided the better understanding how intermittent FSS promotes the osteogenic differentiation of hMSCs through a molecular signaling network and the important information to elucidate the mechanotransduction of intermittent FSS in hMSCs. The understanding on the mechanism of intermittent FSS inducing the osteogenic differentiation of hMSCs will not only be helpful to develop the bone tissue engineering but also provide new targets for drug discovery for treatment of osteoporosis and other related bone-wasting diseases.

Conflict of interest There is no conflict of interest that could inappropriately influence the work reported in this manuscript. Acknowledgements The authors thank the health donors of the First People’s Hospital of Zhejiang for kindly providing bone marrow and Dr. Erik form USA to check the manuscript. This study was financially supported by Scientific Research from Scientific Fund of Zhejiang (2009C13020) and National Natural Science Fund of China (30971460). References Arnsdorf EJ, Tummala P, Kwon RY, Jacobs CR. Mechanically induced osteogenic differentiation – the role of RhoA, ROCKII and cytoskeletal dynamics. J Cell Sci 2009;122:546–53. Bancroft GN, Sikavitsas VI, van den Dolder J, Sheffield TL, Ambrose CG, Jansen JA. Fluid flow increases mineralized matrix deposition in 3D perfusion culture of marrow stromal osteoblasts in a dose-dependent manner. Proc Natl Acad Sci U S A 2002;99:12600–5. Carvalho RS, Scott JE, Yen EH. The effects of mechanical stimulation on the distribution of beta 1 integrin and expression of beta 1-integrin mRNA in TE-85 human osteosarcoma cells. Arch Oral Biol 1995;40:257–64. Chen KD, Li YS, Kim M, Li S, Yuan S, Chien S, et al. Mechanotransduction in response to shear stress. Roles of receptor tyrosine kinases, integrins, and Shc. J Biol Chem 1999;274:18393–400. Communal C, Singh M, Menon B, Xie Z, Colucci WS, Singh K. Beta1 integrins expression in adult rat ventricular myocytes and its role in the regulation of beta-adrenergic receptor-stimulated apoptosis. J Cell Biochem 2003;89:381–8. Digirolamo CM, Stokes D, Colter D, Phinney DG, Class R, Prockop DJ. Propagation and senescence of human marrow stromal cells in culture: a simple colonyforming assay identifies samples with the greatest potential to propagate and differentiate. Br J Haematol 1999;107:275–81.

L. Liu et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1591–1601 Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 1997;89:747–54. Fan J, Shen H, Sun Y, Li P, Burczynski F, Namaka M, et al. Bone morphogenetic protein 4 mediates bile duct ligation induced liver fibrosis through activation of Smad1 and ERK1/2 in rat hepatic stellate cells. J Cell Physiol 2006;207:499–505. Franceschi RT, Xiao G. Regulation of the osteoblast-specific transcription factor, Runx2: responsiveness to multiple signal transduction pathways. J Cell Biochem 2003;88:446–54. Glossop JR, Cartmell SH. Effect of fluid flow-induced shear stress on human mesenchymal stem cells: differential gene expression of IL1B and MAP3K8 in MAPK signaling. Gene Expr Patterns 2009;9:381–8. Grellier M, Bareille R, Bourget C, Amedee J. Responsiveness of human bone marrow stromal cells to shear stress. J Tissue Eng Regen Med 2009;3:302–9. Hsu CJ, Lin TY, Kuo CC, Tsai CH, Lin MZ, Hsu HC, et al. Involvement of integrin up-regulation in RANKL/RANK pathway of chondrosarcomas migration. J Cell Biochem 2010;111:138–47. Kanno T, Takahashi T, Tsujisawa T, Ariyoshi W, Nishihara T. Mechanical stressmediated Runx2 activation is dependent on Ras/ERK1/2 MAPK signaling in osteoblasts. J Cell Biochem 2007;101:1266–77. Kapur S, Baylink DJ, Lau KH. Fluid flow shear stress stimulates human osteoblast proliferation and differentiation through multiple interacting and competing signal transduction pathways. Bone 2003;32:241–51. Katsumi A, Orr AW, Tzima E, Schwartz MA. Integrins in mechanotransduction. J Biol Chem 2004;279:12001–4. Kim SH, Choi YR, Park MS, Shin JW, Park KD, Kim SJ, et al. ERK 1/2 activation in enhanced osteogenesis of human mesenchymal stem cells in poly (lactic-glycolic acid) by cyclic hydrostatic pressure. J Biomed Mater Res A 2007;80:826–36. Kleiveland CR, Kassem M, Lea T. Human mesenchymal stem cell proliferation is regulated by PGE2 through differential activation of cAMP-dependent protein kinase isoforms. Exp Cell Res 2008;314:1831–8. Kundu AK, Khatiwala CB, Putnam AJ. Extracellular matrix remodeling, integrin expression, and downstream signaling pathways influence the osteogenic differentiation of mesenchymal stem cells on poly (lactide-co-glycolide) substrates. Tissue Eng Part A 2009;15:273–83. Lai CF, Chaudhary L, Fausto A, Halstead LR, Ory DS, Avioli LV, et al. Erk is essential for growth, differentiation, integrin expression, and cell function in human osteoblastic cells. J Biol Chem 2001;276:14443–50. Laser M, Willey CD, Jiang W, Cooper Gt, Menick DR, Zile MR, et al. Integrin activation and focal complex formation in cardiac hypertrophy. J Biol Chem 2000;275:35624–30. Lavery K, Swain P, Falb D, Alaoui-Ismaili MH. BMP-2/4 and BMP-6/7 differentially utilize cell surface receptors to induce osteoblastic differentiation of human bone marrow-derived mesenchymal stem cells. J Biol Chem 2008;283: 20948–58. Lee DY, Yeh CR, Chan F, Lee PL, Chien S, Cheng CK, et al. Integrin-mediated expression of bone formation-related genes in osteoblast-like cells in response to fluid shear stress: roles of extracellular matrix, Shc, and mitogen-activated protein kinase. J Bone Miner Res 2008;23:1140–9. Lee HS, Millward-Sadler SJ, Wright MO, Nuki G, Salter DM. Integrin and mechanosensitive ion channel-dependent tyrosine phosphorylation of focal adhesion proteins and beta-catenin in human articular chondrocytes after mechanical stimulation. J Bone Miner Res 2000;15:1501–9. Liu J, Zhao Z, Li J, Zou L, Shuler C, Zou Y, et al. Hydrostatic pressures promote initial osteodifferentiation with ERK1/2 not p38 MAPK signaling involved. J Cell Biochem 2009;107:224–32. Liu L, Yuan W, Wang J. Mechanisms for osteogenic differentiation of human mesenchymal stem cells induced by fluid shear stress. Biomech Model Mechanobiol 2010;9:659–70.

1601

Manton KJ, Leong DF, Cool SM, Nurcombe V. Disruption of heparan and chondroitin sulfate signaling enhances mesenchymal stem cell-derived osteogenic differentiation via bone morphogenetic protein signaling pathways. Stem Cells 2007;25:2845–54. McAllister TN, Du T, Frangos JA. Fluid shear stress stimulates prostaglandin and nitric oxide release in bone marrow-derived preosteoclast-like cells. Biochem Biophys Res Commun 2000;270:643–8. Musa H, Meek S, Gautel M, Peddie D, Smith AJ, Peckham M. Targeted homozygous deletion of M-band titin in cardiomyocytes prevents sarcomere formation. J Cell Sci 2006;119:4322–31. Nohe A, Keating E, Knaus P, Petersen NO. Signal transduction of bone morphogenetic protein receptors. Cell Signal 2004;16:291–9. Orciani M, Trubiani O, Vignini A, Mattioli-Belmonte M, Di Primio R, Salvolini E. Nitric oxide production during the osteogenic differentiation of human periodontal ligament mesenchymal stem cells. Acta Histochem 2009;111:15–24. Pahl HL. Activators and target genes of Rel/NF-kappaB transcription factors. Oncogene 1999;18:6853–66. Pavalko FM, Chen NX, Turner CH, Burr DB, Atkinson S, Hsieh YF, et al. Fluid shearinduced mechanical signaling in MC3T3-E1 osteoblasts requires cytoskeletonintegrin interactions. Am J Physiol 1998;275:C1591–601. Phimphilai M, Zhao Z, Boules H, Roca H, Franceschi RT. BMP signaling is required for RUNX2-dependent induction of the osteoblast phenotype. J Bone Miner Res 2006;21:637–46. Rangaswami H, Bulbule A, Kundu GC. Nuclear factor-inducing kinase plays a crucial role in osteopontin-induced MAPK/IkappaBalpha kinase-dependent nuclear factor kappaB-mediated promatrix metalloproteinase-9 activation. J Biol Chem 2004;279:38921–35. Riddle RC, Taylor AF, Genetos DC, Donahue HJ. MAP kinase and calcium signaling mediate fluid flow-induced human mesenchymal stem cell proliferation. Am J Physiol Cell Physiol 2006;290:C776–84. Scaglione S, Wendt D, Miggino S, Papadimitropoulos A, Fato M, Quarto R, et al. Effects of fluid flow and calcium phosphate coating on human bone marrow stromal cells cultured in a defined 2D model system. J Biomed Mater Res A 2008;86:411–9. Sharp LA, Lee YW, Goldstein AS. Effect of low-frequency pulsatile flow on expression of osteoblastic genes by bone marrow stromal cells. Ann Biomed Eng 2009;37:445–53. Soghomonians A, Barakat AI, Thirkill TL, BlankenshipTN, Douglas GC. Effect of shear stress on migration and integrin expression in macaque trophoblast cells. Biochim Biophys Acta 2002;1589:233–46. Stephens JS, Cooper JA, Phelan Jr FR, Dunkers JP. Perfusion flow bioreactor for 3D in situ imaging: investigating cell/biomaterials interactions. Biotechnol Bioeng 2007;97:952–61. Stiehler M, Bunger C, Baatrup A, Lind M, Kassem M, Mygind T. Effect of dynamic 3-D culture on proliferation, distribution, and osteogenic differentiation of human mesenchymal stem cells. J Biomed Mater Res A 2009;89:96–107. Yamamoto N, Akiyama S, Katagiri T, Namiki M, Kurokawa T, Suda T. Smad1 and smad5 act downstream of intracellular signalings of BMP-2 that inhibits myogenic differentiation and induces osteoblast differentiation in C2C12 myoblasts. Biochem Biophys Res Commun 1997;238:574–80. Yang J, Cao C, Wang W, Tong X, Shi D, Wu F, et al. Proliferation and osteogenesis of immortalized bone marrow-derived mesenchymal stem cells in porous polylactic glycolic acid scaffolds under perfusion culture. J Biomed Mater Res A 2010;92:817–29. Young SR, Gerard-O’Riley R, Kim JB, Pavalko FM. Focal adhesion kinase is important for fluid shear stress-induced mechanotransduction in osteoblasts. J Bone Miner Res 2009;24:411–24.