SEARCH

SEARCH BY CITATION

Keywords:

  • mesenchymal stem cell;
  • matrix stiffness;
  • mechanotransduction;
  • osteogenic differentiation;
  • integrin

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Mesenchymal stem cells (MSCs) cultured on extracellular matrices with different stiffness have been shown to possess diverse lineage commitment owing to the extracellular mechanical stimuli sensed by the cells. The aim of this study was to further delineate how matrix stiffness affects intracellular signaling through the mechanotransducers Rho kinase (ROCK) and focal adhesion kinase (FAK) and subsequently regulates the osteogenic phenotype of MSCs. MSCs were cultured in osteogenic medium on tunable polyacrylamide hydrogels coated with type I collagen with elasticities corresponding to Young's modulus of 7.0 ± 1.2 and 42.1 ± 3.2 kPa. Osteogenic differentiation was increased on stiffer matrices, as evident by type I collagen, osteocalcin, and Runx2 gene expressions and alizarin red S staining for mineralization. Western blot analysis demonstrated an increase in kinase activities of ROCK, FAK, and ERK1/2 on stiffer matrices. Inhibition of FAK, an important mediator of osteogenic differentiation, and inhibition of ROCK, a known mechanotransducer of matrix stiffness during osteogenesis, resulted in decreased expression of osteogenic markers during osteogenic induction. In addition, FAK affects osteogenic differentiation through ERK1/2, whereas ROCK regulates both FAK and ERK1/2. Furthermore, α2-integrin was upregulated on stiffer matrices during osteogenic induction, and its knockdown by siRNA downregulated the osteogenic phenotype through ROCK, FAK, and ERK1/2. Taken together, our results provide evidence that the matrix rigidity affects the osteogenic outcome of MSCs through mechanotransduction events that are mediated by α2-integrin. © 2011 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

The physiology of a cell is affected by a variety of components of the extracellular environment. This comprises of not only the extracellular biochemical cues that are widely studied but also the biophysical effects that play a pivotal role in cellular physiology by relaying mechanotransduction signals into cells.1–4 Matrix stiffness is one such biophysical cue that cells respond to via initiation of mechanotransduction cascades to affect cellular differentiation. Mesenchymal stem cells (MSCs) are precursor cells capable of differentiating into mesenchymal lineage progenies such as osteoblasts, chondrocytes, adipocytes, and myocytes.5, 6

It has been shown that MSCs cultured on soft matrices mimicking the stiffness of brain will become neuronal-like, harder matrices mimicking muscle stiffness will cause cells to switch to a myogenic cell fate, whereas the hardest substrates mimicking collagenous bone lead to osteoblast formation.7 In addition, embryonic stem cells upregulated early mesendoderm markers, and osteogenic differentiation was enhanced when cultured on increased polydimethylsiloxane substrate stiffness.8

It has been reported that ERK1 binds and phosphorylates Ser301 and Ser319 within the proline/serine/threonine domain of Runx2 in vitro and in cell culture. Moreover, introduction of S301A and S319A mutations rendered Runx2 resistant to MAPK-dependent activation and reduced osteoblast-specific gene expression and differentiation after transfection into Runx2 null calvarial cells and mesenchymal cells.9 The extracellular matrix (ECM) compliance of collagen-functionalized poly(ethylene glycol) (PEG) hydrogels governs osteogenic differentiation of MC3T3-E1 preosteoblasts, and this effect was found to be mediated by the RhoA-ROCK-ERK pathway that promotes Runx2 osteoblast transcription factor activity.10 In addition, FAK is an important mediator in mechanotransduction pathways by responding to substrate rigidity in cancer cells.11 Knockdown of focal adhesion kinase (FAK) by siRNA downregulated osteogenic differentiation of MSCs through MAPK/ERK signaling.12 However, how Rho-kinase (ROCK) and FAK signaling are regulated to affect osteogenic differentiation on matrix stiffness remains unknown, and we hypothesize that there is an interplay of regulation between these two mechanotransducers, which are regulated by common signaling events. The purpose of this study was to delineate the relationship between two known mechanotransducers, ROCK and FAK, and surface membrane integrin on defined matrix stiffness that result in activation of signaling pathways regulating the osteogenic fate of MSCs.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Isolation, culture, and osteogenic differentiation of human bone marrow–derived MSCs

Isolation of human bone marrow MSCs was achieved as described previously.13, 14 Briefly, human bone marrow was aspirated from the iliac crest of healthy donors during fracture fixation surgeries with Institutional Review Board approval and informed consent. Mononuclear cells were obtained with a commercially available kit (RosetteSep, StemCell Technologies, Vancouver, British Columbia, Canada) according to the manufacturer's instructions. Nonadherent cells were washed away. Single-cell-derived, clonally expanded cells were obtained subsequently by limiting dilution and maintained in an expansion medium consisting of Iscove's modified Dulbecco's medium (IMDM; Invitrogen, Carlsbad, CA, USA) and 10% fetal bovine serum (Sigma-Aldrich, St Louis. MO. USA) supplemented with 10 ng/mL of basic fibroblast growth factor (Sigma-Aldrich) and 100 units of penicillin, 1,000 units of streptomycin, and 2 mmol/L L-glutamine (Sigma-Aldrich). These isolated bone marrow–derived MSCs were characterized as reported previously. Their surface phenotype was negative for CD7, CD14, CD16, CD19, CD33, CD34, CD38, CD45, CD127, CD133, CD135, and HLA-DR. On the other hand, they were positive for CD29, CD44, CD73, CD90, CD105, CD166, HLA-ABC, and MSC-specific antigen SH-2 (data not shown). The ability to differentiate into mesenchymal lineages, including osteoblasts, chondrocytes, and adipocytes, was confirmed before these cells were used for further experiments. To induce osteogenic differentiation, we treated cells with osteogenic medium, and the medium was changed twice per week. Osteogenic medium consists of IMDM supplemented with 0.1 mM dexamethasone (Sigma-Aldrich), 10 mM β-glycerol phosphate (Sigma-Aldrich), and 0.2 mM ascorbic acid (Sigma-Aldrich). FAK, ROCK, and ERK1/2 inhibition was performed during osteogenic induction by addition of FAK inhibitor (PF-573228; Tocris Biosciences, Bristol, UK), ROCK inhibitor (Y-27632; Sigma-Aldrich) and MEK/ERK inhibitor (U0126; Sigma-Aldrich) at concentrations of 100, 5, and 10 µM into the osteogenic medium, respectively.

Fabrication and characterization of polyacrylamide substrates

Substrates with tunable mechanical properties were fabricated according to reported methods in the literature.15 Briefly, solutions of 40% acrylamide (Bio-Rad, Hercules, CA, USA) and varying concentrations of bis-N,N'-methylene-bis-acrylamide (0.05% to 0.7%; Bio-Rad) were combined in 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, pH 8.5; Sigma-Aldrich). Then 10% ammonium persulfate and TEMED [1,2-di-(dimethylamino) ethane] were added, and the solution was filtered through a 0.22-µm filter. The sterile solutions were poured between 0.75-mm glass spacers and allowed to polymerize for 40 minutes. Then 0.25 mM N-sulfosuccinyimidyl-6-(4'-azido-2'-nitrophenylamino) hexanoate (sulfo-SANPAH; Pierce Biotechnology, Waltham, MA, USA) dissolved in 50 mM HEPES (pH 8.5) and 0.5% DMSO was covered over the polyacrylamide gels and exposed to 365-nm ultraviolet light for 70 minutes for photoactivation. The gels were washed with 50 mM HEPES three times to remove excess reagent. Type I collagen solution (25 µg/mL; Sigma-Aldrich) in phosphate-buffered saline (PBS) was added to the substrate to react overnight at 4 °C and washed three times with PBS to remove unbound proteins. Type I collagen bound to tissue culture polystyrene served as control. The Young's moduli of polyacrylamide gels were measured using MTS Synergie 100 (MTS Systems, Eden Prairie, MN, USA) with 10-N load/cell. Gels were formed into a bone shape and performed according to ASTM D638 standard (type V). Gels were attached and fixed to the tensile tester and subjected to a deformation rate of 1 mm/min.

Immunofluorescent staining and quantification

Polyacrylamide gels were washed with PBS and blocked with 5% fetal bovine serum for 60 minutes and then incubated with primary antibodies against type I collagen (1:400; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) for 1 hour. After washing, cells were incubated with Alexa fluor 488–labeled immunoglobulin G secondary antibody (1:1000; Invitrogen) for 1 hour. Analysis of the extent of binding of type I collagen to the hydrogels was performed with ImageJ software (http://rsbweb.nih.gov/ij/).

Mineralization stain

For evaluation of mineralized matrix, cells were fixed with 3.7% formaldehyde and stained with 1% alizarin red S (Sigma-Aldrich) solution in water for 10 minutes.

Flow cytometry

Cells were stained with phycoerythrin-coupled integrin α2 (BD Pharmingen, Franklin Lakes, NJ, USA) and isotype antibodies and analyzed using a FACSCalibur (Becton Dickinson, Franklin Lakes, NJ, USA) flow cytometer.

RNA extraction, reverse transcription, and semiquantitative RT-PCR

Total RNA was extracted using the RNeasy Kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. The concentration of RNA samples was quantified using OD 260/280, and RNA samples were reverse-transcribed using reagents (Genemark Technology, Taipei, Taiwan) according to the manufacturer's instructions. Roche universal probes (Roche, Basel, Switzerland) were used to detect gene expression during amplification after initial denaturation at 95 °C for 10 minutes and then 95 °C for 10 seconds and 60 °C for 1 minute for every cycle for 40 cycles on Roche LightCycler 480 (Roche). The primer sequences are as follows: Osteocalcin (forward: TGAGAGCCCTCACACTCCTC; reverse: ACCTTTGCTGGACTCTGCAC), type I collagen (forward: CCCCTGGAAAGAATGGAGAT; reverse: AATCCTCGAGCACCCTGAG), Runx2 (forward: CACCATGTCAGCAAAACTTCTT; reverse: TCACGTCGCTCATTTTGC), and GAPDH (forward: AGCCACATCGCTCAGACAC; reverse: GCCCAATACGACCAAATCC).

Phosphorylated kinase assay

Activation of signaling pathways was assessed by a commercial human phosphose kinase array (R&D Systems, Minneapolis, MN, USA) that analyzes the phosphorylation profiles of 46 kinases and their protein substrates. All methods were performed according to manufacturer's instructions. Briefly, cells were harvested, and protein lysates were isolated by lysis buffer (Sigma-Aldrich) and protease inhibitor (1%). Nitrocellulose membranes with antibodies were blocked with buffer against nonspecific binding, and protein concentration was quantified. Then 100 µg of protein extract for all groups was incubated with each piece of nitrocellulose membrane for 1 hour at 25 °C. After three washes, detection antibodies were added and incubated overnight at 4 °C. After several washes, the membranes were incubated with streptavidin–horseradish peroxidase antibodies (Sigma-Aldrich) for 30 minutes and subsequently incubated in chemiluminescent reagents and exposed to X-ray films.

siRNA transfection of α2-integrin

For knockdown of α2-integrin, siRNA sequences targeting α2-integrin (Invitrogen) were transfected into MSCs according to the manufacturer's instructions. Briefly, 10 nM of siRNA for α2-integrin (sense: UUUACUACCAUAUUCUUCU GCAGGC; antisense: GCCUGCAGAAGAAUAUGGUAGUAAA) and control siRNA were transfected with RNAimax transfection reagent (Invitrogen) for 4.5 hours at 37 °C. Cells were washed with PBS and cultured in maintenance medium for 24 hours before the onset of osteogenic induction.

Western blots

Cells were lysed in lysis buffer containing protease and phosphatase inhibitors (Sigma-Aldrich). Cell lysate was quantified, and electrophoresis was carried out on 8% polyacrylamide gels and then transferred to polyvinylidene fluoride (PVDF) membranes. Membranes were blocked and hybridized with ERK1/2 (Cell Signaling, Danvers, MA, USA), phospho-ERK1/2 (Cell Signaling), β-actin (Sigma-Aldrich), phospho-FAK (Santa Cruz Biotechnology), FAK (Santa Cruz Biotechnology), phospho-MYPT (Millipore, Billerica, MA, USA), β-catenin (Cell Signaling), and GAPDH (Sigma-Aldrich) primary antibodies overnight at 4 °C. Secondary antibodies were hybridized the next day, and membranes were covered in chemiluminescent reagent for X-ray film exposure.

Immunoblot semiquantitative analysis

Phosphorylated kinase array expressions on X-ray films were analyzed semiquantitatively by ImageQuant TL (GE Healthcare, Piscataway, NJ, USA).

Statistical analysis

Statistical analyses were performed with one-way ANOVA followed by Tukey's post hoc test or Student's paired t test. Different letters represent significance at p < .05 by ANOVA with Tukey's post hoc test and Student's t test.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Measurement of matrix elasticity and extent of binding of type I collagen on different matrices

A higher ratio of bis-acrylamide in 8% acrylamide gels results in higher stiffness of polyacrylamide substrates in which 0.05%, 0.2%, 0.5%, and 0.7% bis-acrylamide ratios result in an elasticity of 7 ± 1.2, 19.4 ± 3.9, 26.5 ± 2.7 and 42.1 ± 3.2 kPa (mean ± SD), respectively, as measured by a tensile tester (Fig. 1A). Groups with different letters are significantly different, whereas groups with the same letters are not. Photomicrographs showed the presence of type I collagen (green), as visualized with immunofluorescent stain for type I collagen bound to different polyacrylamide substrates (Fig. 1B). Semiquantification of fluorescence intensity from Fig. 1B showed no differences among 7-, 26.5-, and 42.1-kPa bis-acrylamide gels and tissue culture polystyrene (TCPS; Fig. 1C).

thumbnail image

Figure 1. Characteristics of polyacrylamide hydrogels. (A) Young's modulus of gels made with different ratios of bis-acrylamide. (B) Immunofluorescent stain of type I collagen (green) bound to hydrogels. (C) Relative intensity of semiquantitative analysis of type I collagen immunofluorescence from panel B. Data represent mean ± SD. ANOVA (Tukey post hoc test). Groups with different letters are significantly different, whereas groups with same letters are not; p < .05, n = 3. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

Regulation of matrix stiffness on osteogenic differentiation of MSCs

MSCs undergoing osteogenesis were cultured on different matrices for up to 3 weeks and analyzed for osteoblast markers by quantitative real-time PCR (qPCR). Results demonstrated that type I collagen expression was increased on the relatively stiff 42-kPa matrix compared with the relatively soft 7-kPa matrix after 3 weeks of induction, whereas osteocalcin and Runx2 expression was elevated after 2 weeks (Fig. 2A). Alizarin red S stain for mineralization also demonstrated a higher level of mineralization on stiffer matrices after 4 weeks (Fig. 2B). We also performed phosphorylated kinase array to compare 47 kinases of MSCs cultured on gel substrates (data not shown). Results from semiquantitative analysis of blot expressions demonstrated that MSCs induced for 1 week of osteogenic differentiation possessed significantly higher expression on stiffer matrices for phospho-FAK, phospho-STAT5a, phospho-Akt, phospho-ERK1/2, phospho-Jnk, and β-catenin (Fig. 2C). Groups with different letters are significantly different, whereas groups with the same letters are not. Western blot analyses also demonstrated an increase in phospho-ERK1/2, phospho-FAK, and β-catenin expressions on stiffer matrices after 1 week of osteogenic induction (Fig. 2D).

thumbnail image

Figure 2. Osteogenic differentiation of MSCs on different matrix stiffnesses. (A) Type I collagen, osteocalcin, and Runx2 gene expressions on different matrices after 1, 2, and 3 weeks of differentiation. Data represent mean ± SD. Statistical analysis (Student's t test, *p < .05) was compared between MSCs on 7- and 42-kPa gels. (B) Alizarin red S stain for mineralization after 4 weeks of differentiation. (C) Semi-quantitative expression of activated proteins pFAK, pSTAT5a, pAkt, pERK1/2, pJNK, and β-catenin on matrices from phosphorylated kinase array after 1 week of differentiation. Groups with different letters are significantly different, whereas groups with same letters are not, p < .05, n = 3. (D) Western blot of activated FAK, ERK1/2, and β-catenin on matrices. Data represent mean ± SD. ANOVA (Tukey post hoc test). Different alphabets represent different levels of significance, p < .05, n = 3. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

ERK1/2 activation regulates osteogenic differentiation on different matrix stiffnesses

Since Erk1/2 was elevated in the phosphorylated kinase array (Fig. 2C) and Western blot (Fig. 2D), we further inhibited ERK1/2 activation with a MEK/ERK inhibitor, U0126, at 10 µM to inhibit the activation of ERK1/2 during differentiation. Results showed that ERK1/2 activation was significantly downregulated during osteogenic induction (Fig. 3A). Type I collagen gene expression was significantly decreased after 1, 2, and 3 weeks in the ERK1/2-inhibited group compared with its uninhibited control on all matrices, whereas osteocalcin was affected after 2 and 3 weeks of induction (Fig. 3B). The extent of mineralization also was decreased with inhibition after 4 weeks of induction (Fig. 3C).

thumbnail image

Figure 3. ERK1/2 inhibition affects osteogenic gene and activated kinase expressions. (A) ERK1/2 is inhibited at the concentration of 10 µM with U0126. (B) Type I collagen and osteocalcin gene expression on matrices after 1, 2, and 3 weeks of differentiation with or without ERK1/2 inhibitor. Data represent mean ± SD. Statistical analysis was compared between groups with or without inhibition. Student's t test. *p < .05, **p < .005, n = 3. (C) Alizarin red S stain for mineralization after 4 weeks of differentiation. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

FAK activation regulates osteogenic differentiation on matrices with different stiffnesses

Since FAK activation was elevated on stiffer matrices in Fig. 2B, C, and because it has been known to be a crucial mechanoresponsive element.11 we therefore were interested in whether it plays an active role in regulation of osteogenic differentiation by modulation of mechanotransduction through ERK1/2. Thus we further inhibited FAK activation with a FAK-specific inhibitor, PF573228, at 100 nM to inhibit the activation of FAK during differentiation. Results showed that FAK activation was downregulated significantly after addition of 100 nM during osteogenic induction (Fig. 4A). Type I collagen and osteocalcin gene expressions were significantly decreased after 1, 2, and 3 weeks of osteogenic induction in the FAK-inhibited group compared with its uninhibited control on all matrices (Fig. 4B). The extent of mineralization also was decreased with inhibition after 4 weeks of induction (Fig. 4C). In addition, inhibition of FAK also downregulated ERK1/2 activation on matrices after 3 days of osteogenic induction (Fig. 4D).

thumbnail image

Figure 4. FAK inhibition affects osteogenic gene and activated kinase expressions. (A) FAK is inhibited at the concentration of 100 nM with PF573228. (B) Type I collagen and osteocalcin gene expression on matrices after 1, 2, and 3 weeks of differentiation with or without FAK inhibitor. Data represent mean ± SD. Statistical analysis was compared between groups with or without inhibition. Student's t test. *p < .05, **p < .005, n = 3. (C) Alizarin red S stain for mineralization after 4 weeks of differentiation. (D) Western blot analysis of pERK1/2 during FAK inhibition on matrices after 3 days of differentiation. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

ROCK activation regulates osteogenic differentiation on different matrix stiffnesses

An increase in the ROCK substrate phospho-MYPT demonstrated an increase in ROCK activation on stiffer matrices (Fig. 5A). Thus we further inhibited ROCK activation with a ROCK-specific inhibitor, Y27632, at 5 µM during differentiation to investigate its effects on matrix-mediated osteogenic differentiation. Results showed that ROCK activation was significantly downregulated during osteogenic induction (Fig. 5B). Type I collagen and osteocalcin gene expression was decreased after 2 and 3 weeks in the ROCK-inhibited group compared with its uninhibited control on matrices (Fig. 5C). In addition, the extent of mineralization also was decreased with inhibition after 4 weeks of induction (Fig. 5D). Furthermore, inhibition of ROCK downregulated both FAK and ERK1/2 activation on matrices after 3 days of osteogenic induction (Fig. 5E).

thumbnail image

Figure 5. ROCK inhibition affects osteogenic gene and activated kinase expressions. (A) ROCK is inhibited at the concentration of 5 µM with Y27632. (B) Type I collagen and osteocalcin gene expression on matrices after 1, 2, and 3 weeks of differentiation with or without ROCK inhibitor. Data represent mean ± SD. Statistical analysis was compared between groups with or without inhibition. Student's t test. *p < .05, **p < .005, n = 3. (C) Alizarin red S stain for mineralization after 4 weeks of differentiation. (D) Western blot analysis of pERK1/2 and pFAK during ROCK inhibition on matrices after 3 days of differentiation. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

α2-Integrin expression during osteogenic differentiation

To further investigate which cell surface molecule is differentially expressed on different matrix stiffness, we stained for α2-integrin on MSCs cultured on matrices with different stiffnesses and analyzed its expression by flow cytometry after 1 day of osteogenic induction. Interestingly, the mean fluorescence intensity of α2-integrin expression per cell was statistically higher on stiffer matrices (groups with different letters are significantly different, whereas groups with same letters are not; Fig. 6A), whereas the percentage of cells in a population that express α2-integrin was not affected as a result of matrix stiffness (Fig. 6B). This demonstrates that although cells on different matrix stiffnesses express α2-integrin at the same ratio (approximately 90%), the amount of expression per cell is higher on stiffer matrices.

thumbnail image

Figure 6. α2-Integrin expression on matrices during osteogenic differentiation. (A) Relative quantitative expression after 1 day of osteogenic induction on 7-, 26.5-, and 42.1-kPa and TCPS matrices. Data represent mean ± SD. ANOVA (Tukey post hoc test). Groups with different letters are significantly different, whereas groups with same letters are not, p < .05, n = 3. (B) Percent cell expression of α2-integrin on matrices after 1 day of osteogenic induction. TCPS = tissue culture polystyrene.

Download figure to PowerPoint

α2-Integrin regulates osteogenic differentiation on different matrix stiffnesses

Since matrix stiffness mediated the differential expression of α2-integrin during osteogenic differentiation (Fig. 6A), we performed knockdown of α2-integrin on MSCs to further investigate its role in mechanotransduction during osteogenic differentiation. Flow cytometric analysis exhibited a 50% reduction in α2-integrin mean fluorescence intensity after 2 days of knockdown (Fig. 7A). This resulted in a significant decrease in type I collagen gene expression after 1, 2, and 3 weeks of osteogenic induction, whereas osteocalcin gene expression was decreased after 2 and 3 weeks in the knockdown groups on matrices (Fig. 7B). In addition, the extent of mineralization also was decreased with inhibition after 4 weeks of induction (Fig. 7C). Furthermore, knockdown of α2-integrin downregulated phospho-MYPT, phospho-FAK, and phospho-ERK1/2 expression on matrices after 3 days of osteogenic induction (Fig. 7D).

thumbnail image

Figure 7. α2-Integrin knockdown affects osteogenic gene and activated kinase expressions. (A) α2-Integrin is knocked down by siRNA and cultured in maintenance medium for 24 hours before induction of differentiation. Flow cytometric analysis revealed a 50% reduction in mean fluorescence after 1 day of differentiation. (B) Type I collagen and osteocalcin gene expression on matrices after 1, 2, and 3 weeks of differentiation with or without α2-integrin knockdown. Data represent mean ± SD. Statistical analysis was compared between groups with or without inhibition. Student's t test. *p < .05, **p < .005, n = 3. (C) Alizarin red S stain for mineralization after 4 weeks of differentiation. (D) Western blot analysis of pMYPT, pFAK, and pERK1/2 during α2-integrin knockdown on matrices after 3 days of differentiation. TCPS =  tissue culture polystyrene.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Since substrate stiffness affects many processes, such as cell growth, adhesion, migration, and differentiation, as a consequence, it is difficult to define a suitable scaffold stiffness that optimally stimulates a particular type of tissue regeneration.16 However, it is likely that a scaffold for skeletal regeneration should display the stiffness of a developing skeletal tissue, which might be lower than the stiffness of a mature tissue.17 From this perspective, it is reasonable to consider differentiation of cells on substrates with a stiffness that mimics the stiffness of the developing native tissue.

We used an in vitro hydrogel system to mimic the native tissue elasticity of developing bone (42.1 kPa) and compared osteogenic differentiation to a softer matrix (7 kPa). Similar to previously reported studies, we find that MSC commitment to the osteoblastic lineage is enhanced on ECM stiffness that mimics developing bone with a Young's modulus of 42.1 kPa and results in higher levels of osteogenic phenotype compared with softer gels.

Type I collagen was selected for binding on hydrogel in this study because it is the major ECM molecule in native skeletal tissues. Previous studies demonstrate the major integrin that interacts with collagen is α2β1 integrin.18 A wide range of integrins, including α2β1 integrin, are found on the surface membranes of MSCs that are required for survival and differentiation by functioning as cell adhesion molecules to the ECM.19 Disruption of α2-integrin–ECM interactions with a blocking antibody or DGEA peptide containing the cell-binding domain of type I collagen blocked activation of the mouse osteocalcin gene 2 promoter by ascorbic acid as well as induction of endogenous osteocalcin gene expression and mineralization. Furthermore, anti-α2-integrin-blocking antibody or peptide reduced ascorbic acid–dependent binding of Runx2 to OSE2.20 Integrins connect the ECM to intracellular cytoskeleton through focal adhesions with a wide range of focal adhesion proteins, including talin, paxillin, and vinculin.21 Formation of focal adhesions in MC3T3-E1 preosteoblasts was enhanced when ECM stiffness was increased from 20 to 110 kPa,22 as visualized by immunofluorescent staining for vinculin. It is intriguing to observe that on induction of osteogenic differentiation, α2-integrin expression was higher on stiffer matrices. Since it plays a pivotal role in cell attachment to the ECM, it is likely that α2-integrin mediates mechanotransduction from the cell membrane and serves as a mechanosensor and early mechanotransducer of matrix elasticity in osteogenic cells.

With the use of a tunable hydrogel system in our study, we have provided evidence of a mechanism that MSCs use in response to the mechanistic properties of matrix rigidity during osteogenic differentiation. Knockdown of α2-integrin reduced activation of phospho-MYPT, a substrate downstream of ROCK. Interestingly, ROCK inhibition also downregulated FAK activation, whereas FAK inactivation downregulated ERK1/2. Thus, during osteogenic induction, MSCs sense the mechanistic alteration of ECM and respond to this biophysical cue by adjusting the expression of surface α2-integrins, which subsequently dictate the differential activation of the osteogenic program, mediated via the mechanotransducers ROCK and FAK to subsequent activation of ERK1/2.

The control of differentiation is complex and can be multifaceted. The mechanism of MSC osteogenic commitment involves various intracellular signaling pathways such as MAP kinases p38 and JNK.23 Although we did not investigate the role of upregulated kinases from the kinase array further in this study, it is observed that matrix rigidity affects activation of multiple proteins such as phospho-Akt, β-catenin, phospho-Jnk, and phospho-STAT5a. Akt activation and increased expression of β-catenin are both implicated during induction of the skeletal phenotype,24, 25 whereas STAT5a has been shown to participate in 1,25-dihydroxyvitamin D3 upregulation of osteoblast markers.26, 27

Taken together, these results have provided evidence on how MSCs accommodate to ECM rigidity through an α2-integrin-ROCK-FAK-ERK1/2 axis during osteogenic differentiation. The results from this study will shed light on the future study of the central and functional role of biophysical forces during osteogenic lineage commitment.

Disclosures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

All the authors state that they have no conflicts of interest.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

We acknowledge financial support from the Taipei Veterans General Hospital (VGH99E1-014, VGH99S4-001), the National Science Council, Taiwan (NSC98-2627-B-010-004-MY3, NSC98-2314-B-010-001-MY3, NSC98-3111-B-010-003, NSC97-3111-B-010-003), and a National Yang-Ming University/Cheng Hsin General Hospital Grant (98F117CY10). This study also was supported by a grant from the Ministry of Education, Aim for the Top University Plan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References