Proc Natl Acad Sci U S A 106(44): 18587-18591

PMC: PMC2773973

PMID: 19843692

Priming integrin α5 promotes human mesenchymal stromal cell osteoblast differentiation and osteogenesis

Results

ITGA5 Is Up-Regulated During Osteoblast Differentiation of hMSCs.

As expected, dexamethasone increased ALP activity and other osteoblast markers at 1 and 3 days in primary hMSCs (Fig. S1 A–D). We analyzed the gene expression profile in these differentiating hMSCs using HG-U133 Plus 2.0 arrays consisting of 54,675 probe sets (21). Analysis of normalized data from three different donors revealed that ITGA5 expression was increased >2-fold in dexamethasone-treated hMSCs compared with that in untreated cells. Quantitative RT-PCR analysis (Fig. 1A) and Western blot analysis (Fig. 1B) confirmed that dexamethasone increased ITGA5 mRNA and protein levels in hMSCs. These results confirm that ITGA5 expression is rapidly increased during dexamethasone-induced differentiation of hMSCs.

An external file that holds a picture, illustration, etc.
Object name is zpq9990999000001.jpg

Fig. 1.

Integrin α5 (ITGA5) expression is up-regulated during osteoblast differentiation of human mesenchymal stromal cells (hMSCs). Adult hMSCs were treated with or without dexamethasone (Dex, 10^{M) to induce osteoblastic differentiation, and ITGA5 mRNA and protein levels were determined by quantitative PCR (}A) and Western blot analyses (B). Results are expressed as the mean ± SD of the ratios of Dex-treated to control cells after normalization to 18S rRNA expression. *, P < 0.05 (significant difference with control cells).

Forced Expression of ITGA5 Promotes Osteoblast Differentiation in hMSCs.

To determine the role of ITGA5 in MSC osteoblast differentiation, hMSCs were transduced with a lentiviral vector encoding ITGA5 and GFP. The LV-ITGA5-transduced cells showed increased GFP expression (Fig. 2A) and increased ITGA5 mRNA and protein levels (Fig. 2B and C). The ITGA5 transduction in hMSCs increased Runx2, ALP, and Col1A1 mRNA expression (Fig. 2D). Similar results were found in hMSCs obtained from three donors. Consistent with this effect, LV-ITGA5 transduction in hMSCs increased ALP activity and osteogenic capacity of hMSCs in vitro (Fig. 2E). These results demonstrate that forced expression of ITGA5 is sufficient to promote the expression of osteoblast markers and osteogenic capacity of primary hMSCs.

An external file that holds a picture, illustration, etc.
Object name is zpq9990999000002.jpg

Fig. 2.

Forced expression of integrin α5 (ITGA5) promotes osteoblast differentiation in human mesenchymal stromal cells (hMSCs). Adult hMSCs were transduced with LV-ITGA5 or empty vector, the efficacy of transduction was evaluated by GFP level under fluorescence microscopy (A), and ITGA5 mRNA and protein levels were determined by quantitative PCR (B) and Western blot analysis (C). Forced expression of ITGA5 in hMSCs increased Runx2, alkaline phosphatase (ALP), and type I collagen (Col1A1) mRNA expression (D) and enhanced ALP expression and matrix mineralization as determined by alizarin red staining (E) compared with control cells. Results are expressed as mean ± SD of the ratios of treated to control cells after normalization to 18S rRNA expression. *, P < 0.05 (significant difference with control cells).

Silencing of ITGA5 Mediated by shRNA Blunts Osteoblast Differentiation in hMSCs.

To further establish the role of ITGA5 in hMSC osteoblast differentiation, we determined whether RNAi-mediated silencing of ITGA5 expression interferes with osteoblast marker genes in hMSCs. Silencing of ITGA5 expression using two different shRNAs decreased ITGA5 mRNA by 40–60% and ITGA5 protein level by 70–80% at 24 h, whereas a nonrelevant control shRNA had no effect (Fig. 3A and B). In contrast to the control shRNA, the two ITGA5 shRNAs decreased Runx2, ALP, and Col1A1 mRNA levels (Fig. 3C). Silencing of integrin β1 (ITGB1) using a specific shRNA also decreased Runx2, ALP, and Col1A1 expression (Fig. 3D), supporting a role for the ITGA5/ITGB1 complex in osteoblast gene induction in hMSCs. Consistent with this effect, ITGA5 silencing abolished ALP activity and blocked the osteogenic capacity in primary hMSCs (Fig. 3E). These results indicate that ITGA5 is required for maintenance of the osteoblast phenotype in hMSCs.

An external file that holds a picture, illustration, etc.
Object name is zpq9990999000003.jpg

Fig. 3.

Silencing of integrin α5 (ITGA5) mediated by shRNA blunts osteoblast differentiation in human mesenchymal stromal cells (hMSCs). Adult hMSCs were transduced with two independent ITGA5 shRNAs, which efficiently reduced endogenous ITGA5 protein and mRNA levels as determined by Western blot (A) and quantitative PCR (B) analyses. A nonrelevant shRNA (sh-NR) was used as the control. The ITGA5 silencing reduced Runx2, alkaline phosphatase (ALP), and type I collagen (Col1A1) mRNA expression (C). A specific integrin β1 (ITGB1) shRNA also reduced Runx2, ALP, and Coll1A1 mRNA expression (D), indicating that the ITGA5/ITGB1 complex is required. Silencing ITGA5 abolished ALP activity and matrix mineralization as determined by alizarin red staining (E). Results are expressed as mean ± SD of the ratios of treated to control cells after normalization to 18S rRNA expression. *, P < 0.05 (significant difference with control cells).

Priming ITGA5 Promotes Osteoblast Differentiation in hMSCs.

We then determined whether activation of endogenous ITGA5 is sufficient to promote hMSC osteoblast differentiation. To this goal, we used a conformation-dependent anti-ITGA5 monoclonal antibody (SNAKA51) that primes and stimulates α5β1 integrin and promotes cell adhesion in fibroblasts (22). The addition of SNAKA51 at a dose that promotes ligand binding (10 μg/mL) (22) markedly increased Runx2, ALP, and Col1A1 expression in hMSCs, whereas a nonrelevant control antibody (IgG) has no effect (Fig. 4A). To determine whether activation of ITGA5 alone promoted osteoblast differentiation in hMSCs, we tested the effect of a synthetic peptide (CRRETAWAC) that binds and activates ITGA5 (22). As shown in Fig. 4B, CRRETAWAC (100 μg/mL) markedly increased the expression of osteoblast marker genes in hMSCs, whereas a nonrelevant control peptide (GRGESP) had no effect. As shown in Fig. 4C, the addition of SNAKA51 or CRRETAWAC greatly increased the osteogenic capacity of primary hMSCs in vitro. These results provide evidence that activation of ITGA5 using a specific monoclonal antibody or a peptide agonist that primes this integrin is sufficient to promote the expression of phenotypic osteoblast markers and osteogenic capacity in hMSCs.

An external file that holds a picture, illustration, etc.
Object name is zpq9990999000004.jpg

Fig. 4.

Priming integrin α5 (ITGA5) promotes osteoblast differentiation in human mesenchymal stromal cells (hMSCs). Treatment of adult hMSCs with the anti-α5 monoclonal antibody SNAKA51 (10 μg/mL) that primes ITGA5 (A) or the synthetic peptide CRRETAWAC (100 μg/mL) that specifically binds ITGA5 increased Runx2, alkaline phosphatase (ALP), and type I collagen (Col1A1) mRNA expression as determined by quantitative PCR analysis at 3 days (B) and enhanced matrix mineralization as determined by alizarin red staining at 14 days (C). Control IgG or control peptide GRGESP had no effect (A, B). Results are expressed as mean ± SD of the ratios of treated to control cells after normalization to 18S rRNA expression. *, P < 0.05 (significant difference with control cells).

ERK1/2 and PI3K Signaling Pathways Mediate ITGA5-Induced Osteoblast Differentiation in hMSCs.

We then determined the signaling mechanisms underlying the promoting effect of ITGA5 on osteoblast differentiation in hMSCs. The LV-ITGA5 transduction of hMSCs increased phosphorylation of focal adhesion kinase (FAK) (Fig. S2A). Biochemical ELISA analysis confirmed that LV-ITGA5 transduction of hMSCs increased FAK activity (Fig. S3A). Silencing of FAK using a specific shRNA that decreased FAK protein level by 60% (Fig. 3B) abolished the up-regulation of Runx2 induced by LV-ITGA5 or CRRETAWAC (Fig. S3C). These results support a critical role for FAK signaling in ITGA5-induced osteoblast differentiation of hMSCs. Consistent with FAK activation, LV-ITGA5 transduction of hMSCs increased ERK1/2 and PI3K but not p38 MAPK (Fig. S2A). Priming ITGA5 using the CRRETAWAC peptide also increased ERK1/2 and PI3K phosphorylation but not p38 MAPK (Fig. S2B). The ECM-induced ERK1/2 activation is known to induce phosphorylation of Runx2 (23). Both LV-ITGA5 and CRRETAWAC increased serine phosphorylation of Runx2 (Fig. S2C). These results identify ERK1/2, PI3K, and Runx2 phosphorylation as important signaling pathways that are induced by ITGA5 activation in hMSCs. We then sought to determine the relative importance of ERK1/2 and PI3K signaling pathways in osteoblast differentiation induced by ITGA5 in hMSCs. As shown in Fig. S4A, the mitogen-activated protein/extracellular signal-regulated kinase kinase (MEK) inhibitor U0126 (10 μM) abolished osteoblast marker gene expression induced by LV-ITGA5 in hMSCs. Consistently, transfection of hMSCs with dominant negative ERK1/2 (DN-ERK1/2) (24) fully abolished osteoblast marker gene expression induced by LV-ITGA5 (Fig. S4B). Additionally, {"type":"entrez-nucleotide","attrs":{"text":"LY294002","term_id":"1257998346","term_text":"LY294002"}}LY294002 (15 μM), a specific PI3K inhibitor, blunted osteoblast marker gene expression induced by LV-ITGA5 (Fig. S4C). Consistent with these findings, pharmacological inhibition of MEK or PI3K abolished osteoblast gene expression induced by the CRRETAWAC peptide (Fig. S4 D and E). These results indicate that FAK, ERK1/2, and PI3K signaling mediate the increased osteoblast marker gene expression induced by ITGA5 activation in hMSCs.

Forced Expression of ITGA5 Promotes Osteogenesis in Vivo.

Our finding that ITGA5 plays a critical role in osteogenic differentiation of hMSCs in vitro prompted us to determine whether this can be translated to the induction of osteogenesis in vivo. To this end, we determined the bone-forming potential of hMSCs overexpressing ITGA5 in a standard ectopic osteogenic assay (25). Human MSCs were transduced with LV-ITGA5, associated with an effective osteoconductive carrier (7), and implanted s.c. into ectopic sites in the backs of immunocompromised mice (25). As shown in Fig. 5A, only a small fraction of the carrier implanted with control hMSCs showed de novo bone formation. In contrast, virtually 100% of the surface of the carrier was covered with new bone in hMSCs overexpressing ITGA5 (Fig. 5B). Remarkably, plump osteoblasts and osteocytes characterized the newly formed matrix (Fig. 5C and D). Quantification of the new bone formed showed that LV-ITGA5 transduction of hMSCs markedly increased the amount of bone matrix compared with that of control cells (Fig. 5E). Collectively, our data provide evidence that ITGA5 plays an essential role in the osteoblast differentiation program of hMSCs through activation of FAK, ERK1/2, and PI3K signaling and demonstrate that targeted induction of ITGA5 in hMSCs promotes the osteogenic capacity of hMSCs in vivo. This may serve as a basis to promote osteogenesis and tissue regeneration (Fig. 5F).

An external file that holds a picture, illustration, etc.
Object name is zpq9990999000005.jpg

Fig. 5.

Forced expression of integrin α5 (ITGA5) markedly promotes osteogenesis in vivo. (A–D) Adult human mesenchymal stromal cells (hMSCs) transduced with LV-ITGA5 or empty vector were implanted s.c. with coral/hydroxyapatite (Co) at ectopic sites in nude mice, and bone formation was examined after 8 weeks. Histological examination showed that ITGA5 overexpression in hMSCs increased new bone formation along the carrier (B) compared with that in control cells (A) with plump osteoblasts (Ob) and osteocytes (Ocy) (C, D). (E) Quantification of the newly formed bone confirmed the high bone-forming capacity of hMSCs overexpressing ITGA5. Results are expressed as mean ± SD. *, P < 0.05 (significant difference with control cells transduced with empty vector). (F) Mechanisms by which ITGA5 promotes hMSC osteogenic potential. The ITGA5 up-regulation induced by dexamethasone (Dex) or priming ITGA5 using agonists (specific antibody or peptide) activate focal adhesion kinase (FAK), ERK1/2, and PI3K signaling, resulting in increased osteoblast differentiation and osteogenesis in vitro and in vivo.

ITGA5 Is Up-Regulated During Osteoblast Differentiation of hMSCs.

Fig. 1.

Forced Expression of ITGA5 Promotes Osteoblast Differentiation in hMSCs.

Fig. 2.

Silencing of ITGA5 Mediated by shRNA Blunts Osteoblast Differentiation in hMSCs.

Fig. 3.

Priming ITGA5 Promotes Osteoblast Differentiation in hMSCs.

Fig. 4.

ERK1/2 and PI3K Signaling Pathways Mediate ITGA5-Induced Osteoblast Differentiation in hMSCs.

Forced Expression of ITGA5 Promotes Osteogenesis in Vivo.

Fig. 5.

Discussion

The identification of mechanisms that direct adult MSC osteogenic differentiation is of prime interest for developing therapeutic strategies to promote bone formation and regeneration. In this study, we establish the essential role of ITGA5 in hMSC differentiation and show that specific activation of this integrin in hMSCs is sufficient to promote osteoblast differentiation. This provides a tool to promote hMSC osteogenic capacity, which may be exploited for bone tissue regeneration.

We first identified by transcriptome analysis that dexamethasone-induced osteoblast differentiation in hMSCs is associated with up-regulation of ITGA5. Dexamethasone was found previously to regulate α_Vβ3 and α_Vβ5 integrins in normal human osteoblastic cells (26). However, this is the first demonstration that ITGA5 is up-regulated during osteogenic differentiation in hMSCs. Prompted by this finding, we showed that forced expression of ITGA5 was sufficient to promote osteogenic differentiation in hMSCs. Notably, forced expression of ITGA5 increased the expression of Runx2, an essential osteoblast transcription factor (9, 10), and ALP and Col1A1, which are required for osteogenesis in vivo (27). Consistent with this finding, ITGA5 overexpression promoted the in vitro osteogenic capacity, whereas specific ITGA5 silencing reduced osteoblast marker gene expression and blocked osteogenic differentiation of hMSCs. Taken together, these data provide evidence that ITGA5 is essential for osteoblast differentiation of hMSCs and is an important activator of osteogenic differentiation by hMSCs in vitro.

We then investigated the potential mechanisms through which ITGA5 may promote hMSC osteogenic differentiation. Integrins are known to interact with several signaling molecules (15). We showed here that FAK, ERK1/2, and PI3K are necessary for ITGA5-induced osteoblast gene expression, thus identifying major signal transduction pathways that converge to mediate osteoblast differentiation induced by ITGA5 in hMSCs. Several mechanisms may act downstream of ERK1/2 and PI3K to activate hMSC osteoblast differentiation. Notably, ERK1/2 signaling activates SMAD and activating protein 1 transcription factors and induces Runx2 phosphorylation, which subsequently induces expression of osteoblast marker genes (10, 28). We show here that activation of ITGA5 induced Runx2 phosphorylation in hMSCs, indicating that activation of Runx2 is involved in subsequent induction of osteoblast gene expression induced by ITGA5 in hMSCs.

We then asked whether specific activation of ITGA5 may be sufficient to generate osteogenic differentiation of hMSCs. One possible approach to activate endogenous integrin expression is to promote ligand binding independently of ECM through conformational modulation (29 –31). Inactive, primed, and ligand-bound conformations regulate ligand binding affinity and intracellular signaling (32). However, most antibodies that modulate integrin activation alter the structure of the ligand binding pocket, resulting in stimulation or inhibition of ligand binding (22, 29). Although high-affinity ligands for integrins have been developed (16, 33, 34), a limited number of molecules act as agonists to selectively activate integrin-dependent molecular events. The SNAKA51 anti-ITGA5 monoclonal antibody changes the conformation and primes the integrin into a ligand-competent form, resulting in increased cell adhesion to fibronectin and matrix formation (22). We show here that priming ITGA5 integrin using the SNAKA51 antibody was sufficient to induce osteoblast gene expression and osteogenic differentiation in hMSCs. Additionally, we demonstrate that the CRRETAWAC peptide that binds ITGA5 (22) promoted hMSC osteoblast differentiation and in vitro osteogenesis. This demonstrates that SNAKA51 and the CRRETAWAC peptide act as ITGA5 agonists and functionally activate the osteogenic capacity of human hMSCs. An important issue for tissue regeneration is, however, to evaluate whether gene therapy targeting ITGA5 in hMSCs may promote osteogenesis in vivo. We showed that up-regulation of ITGA5 in hMSCs markedly promoted bone formation in a standard osteogenic assay, demonstrating that ITGA5 manipulation is sufficient to promote the osteogenic differentiation program in vivo. This finding validates the proposal that ITGA5 gene therapy in hMSCs may be an efficient strategy to promote osteogenesis in vivo.

In summary, this study demonstrates that ITGA5 is essential for the osteoblast differentiation program in adult hMSCs and promotes osteogenesis. Our results suggest an approach to enhance hMSC osteogenic capacity, which may have several therapeutic implications. The use of ITGA5 agonists may be used to promote hMSC osteoblast differentiation and bone formation in skeletal disorders where the recruitment of MSCs is compromised. This strategy also may prove to be useful to engineer biomaterials with appropriate ITGA5 agonists that then may be exploited in therapeutic protocols to efficiently promote bone regeneration.

Methods

Cells and Materials.

Primary hMSCs were derived from the bone marrow stroma of normal individuals after informed consent (35). The ITGA5-activating monoclonal antibody (SNAKA51) and the synthetic agonist peptide (CRRETAWAC) were from M.J. Humphries (University of Manchester, Manchester, U.K.). The rabbit IgG fraction (negative control) was from DAKO. The synthetic control peptide (GRGESP) was from Peptide 2.0. Mouse monoclonal anti-pERK1/2 and anti-FAK, rabbit polyclonal anti-ITGA5, and anti-ITGB1 were from Santa Cruz Biotechnology. Rabbit polyclonal anti-p-p38 was from Abcam. Rabbit polyclonal anti-pPI3K p85, anti-p-FAK, and anti-p-(Ser/Thr)Phe antibodies were from Cell Signaling Technology. Mouse monoclonal anti-Runx2 antibody was from G. Karsenty (Columbia University, New York, NY). Rabbit polyclonal anti-β-actin antibody, U0126, and {"type":"entrez-nucleotide","attrs":{"text":"LY294002","term_id":"1257998346","term_text":"LY294002"}}LY294002 were from Sigma-Aldrich.

Transcriptome Analysis.

Subconfluent hMSCs were incubated with or without 10^{M dexamethasone for 1, 3, or 7 days, total RNA was isolated, and gene expression profiling with Affymetrix HG-U133 Plus 2.0 arrays was performed as described in ref.}21. Raw gene expression data were processed with the GCOS 1.2 software for signal calculation, and comparison analysis and cellular pathways were analyzed using statistical tools. These data have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc={"type":"entrez-geo","attrs":{"text":"GSE18043","term_id":"18043"}}GSE18043 (dataset {"type":"entrez-geo","attrs":{"text":"GSE18043","term_id":"18043","extlink":"1"}}GSE18043).

Quantitative RT-PCR Analysis.

Total RNAs were isolated and relative, mRNA levels were evaluated by quantitative PCR analysis using microfluidic cards (Applied Biosystems) or LightCycler (Roche Applied Science) and normalized to 18S rRNA as described in ref. 36.

Plasmids.

The amplified ITGA5 coding sequence was cloned and transferred into the pLentiGW vector by in vitro recombination. An ITGA5 shRNA encoding sequence was obtained by PCR elongation, cloned into the pGEM-T Easy vector (Promega), and subcloned into the pLenti-RNAi vector. Another ITGA5 shRNA was obtained from Open Biosystems. Control shRNA (scrambled sequence), ITGB1 shRNA, and human FAK shRNA lentiviral particles were from Santa Cruz Biotechnology.

Lentiviral Production and Transduction.

Viral production was performed as described in ref. 36. For transduction, subconfluent cells were incubated with lentivirus and 4 μg/mL polybrene (Sigma) for 48 h.

Western Blot, Immunoprecipitation, and ELISA Analyses.

Cell lysates and immunoprecipitation analysis were prepared as described in ref. 36. Densitometric analysis was performed using ImageQuant software (Agfa). Phosphorylated Tyr397 FAK level relative to cell number was evaluated using the FACE ELISA kit (Active Motif).

Alkaline Phosphatase and Osteogenic Assays.

The ALP activity, protein content, ALP cytochemical detection, and in vitro matrix mineralization assays were performed as described in ref. 36.

In Vivo Osteogenesis Assay.

For the in vivo osteogenic ectopic assay, 2 × 10^{control or LV-ITGA5-transduced hMSCs were mixed with 40 mg of ProOsteon (coral/hydroxyapatite particles of 0.5–1.0 mm in size), human fibrinogen (30 mg/mL in PBS), and thrombin (25 units/mL in 2% CaCl}₂) and s.c. transplanted in the backs of 8-week-old athymic nude mice (Harlan) (25). After 8 weeks, the transplanted sites were fixed, embedded in paraffin, serial 6-μm-thick sections were stained with hematoxylin and eosin, and quantification of de novo bone formation was performed using Image Pro computerized system (Media Cibernetics) in five sections obtained from eight different explants from four mice per group.

Statistical Analysis.

The results are expressed as mean ± SD of at least five samples. Comparisons between data were performed using the Student's t test with P < 0.05 considered as significant.

Cells and Materials.

Transcriptome Analysis.

Quantitative RT-PCR Analysis.

Plasmids.

Lentiviral Production and Transduction.

Viral production was performed as described in ref. 36. For transduction, subconfluent cells were incubated with lentivirus and 4 μg/mL polybrene (Sigma) for 48 h.

Western Blot, Immunoprecipitation, and ELISA Analyses.

Alkaline Phosphatase and Osteogenic Assays.

The ALP activity, protein content, ALP cytochemical detection, and in vitro matrix mineralization assays were performed as described in ref. 36.

In Vivo Osteogenesis Assay.

Statistical Analysis.

The results are expressed as mean ± SD of at least five samples. Comparisons between data were performed using the Student's t test with P < 0.05 considered as significant.

Supplementary Material

Supporting Information:

Click here to view.

^{Institut National de la Santé et de la Recherche Médicale U606, Hôpital Lariboisière, Paris, France;}

^{University Denis Diderot, Paris, France;}

^{Charite University, Berlin, Germany;}

^{Institut National de la Santé et de la Recherche Médicale U966, University of Tours, France; and}

^{Anatomy and Cell Biology Department, Faculty of Medicine, Rappaport Institute, Haifa, Israel}

^{To whom correspondence should be addressed. E-mail:}rf.mresni@eiram.erreip

Edited by Richard O. Hynes, Massachusetts Institute of Technology, Cambridge, MA, and approved August 18, 2009

Author contributions: O.F., S.S., E.L., and P.J.M. designed research; Z.H., O.F., J.R., T.H., S.S., and E.L. performed research; P.V. and J.-C.P. contributed new reagents/analytic tools; Z.H., O.F., J.R., T.H., and P.J.M. analyzed data; and O.F., S.S., E.L., and P.J.M. wrote the paper.

^{Z.H. and O.F. contributed equally to this work.}

^{Present address: Unité Mixte de Recherche 6175, Institut National de la Recherche Agronomique, Nouzilly, France.}

Edited by Richard O. Hynes, Massachusetts Institute of Technology, Cambridge, MA, and approved August 18, 2009

Received 2008 Dec 5

Abstract

Adult human mesenchymal stromal cells (hMSCs) have the potential to differentiate into chondrogenic, adipogenic, or osteogenic lineages, providing a potential source for tissue regeneration. An important issue for efficient bone regeneration is to identify factors that can be targeted to promote the osteogenic potential of hMSCs. Using transcriptome analysis, we found that integrin α5 (ITGA5) expression is up-regulated during dexamethasone-induced osteoblast differentiation of hMSCs. Gain-of-function studies showed that ITGA5 promotes the expression of osteoblast phenotypic markers and in vitro osteogenesis of hMSCs. Down-regulation of endogenous ITGA5 using specific shRNAs blunted osteoblast marker gene expression and osteogenic differentiation. Molecular analyses showed that the enhanced osteoblast differentiation induced by ITGA5 was mediated by activation of focal adhesion kinase/ERK1/2-MAPKs and PI3K signaling pathways. Remarkably, activation of endogenous ITGA5 using agonists such as a specific antibody that primes the integrin or a peptide that specifically activates ITGA5 was sufficient to enhance ERK1/2-MAPKs and PI3K signaling and to promote osteoblast differentiation and osteogenic capacity of hMSCs. Importantly, we demonstrated that hMSCs engineered to overexpress ITGA5 exhibited a marked increase in their osteogenic potential in vivo. Taken together, these findings not only reveal that ITGA5 is required for osteoblast differentiation of adult hMSCs but also provide a targeted strategy using ITGA5 agonists to promote the osteogenic capacity of hMSCs. This may be used for tissue regeneration in bone disorders where the recruitment or capacity of hMSCs is compromised.

Keywords: mesenchymal stem cells, bone formation, agonist

Abstract

Mesenchymal stromal cells (MSCs) derived from the bone marrow stroma are capable of differentiating into chondroblasts, adipocytes, or osteoblasts (1, 2) under appropriate environmental conditions (3, 4). Adult human MSCs (hMSCs) are considered as a valuable source for bone tissue regeneration in human diseases (5, 6). However, the capacity of autologous hMSCs to differentiate along functional bone-forming osteoblasts remains relatively limited for bone regeneration in vivo (7). An important issue for efficient bone regeneration is therefore to target hMSCs to promote their osteogenic potential for in vivo bone regeneration.

The osteogenic differentiation process of MSCs is characterized by the expression of the main osteoblast transcription factor Runx2 and osteoblast markers such as alkaline phosphatase (ALP) and type I collagen (Col1A1) and is typified by ECM mineralization (8 –10). The ECM–osteoblast interactions generate important signaling mechanisms that converge to promote early osteoblast-specific gene expression and differentiation (11 –13). Cell–matrix interactions involve integrins, a family of transmembrane proteins that induce intracellular signals (14, 15). The α5β1 integrin is a cell surface receptor for fibronectin that has been implicated in cell spreading, proliferation, differentiation, migration, and survival in different cell types (16 –18). Osteoblasts express α5β1 integrin, which is involved in cell adhesion (11) and apoptosis in vitro and in vivo (19, 20). However, its role in the osteogenic differentiation program in hMSCs is not established. Using genome wide and functional analyses, we show in this study that the integrin α5 subunit (ITGA5) mediates osteoblast differentiation induced by the synthetic glucocorticoid dexamethasone in adult hMSCs. Consistent with this finding, silencing ITGA5 expression abolished osteoblast gene expression in hMSCs. We demonstrate that forced expression of ITGA5 or priming ITGA5 integrin using specific agonists was sufficient to induce osteoblast gene expression and in vitro osteogenic differentiation of hMSCs. Importantly, forced expression of ITGA5 in hMSCs results in a marked increase in de novo osteogenesis in vivo. These results identify a critical role for ITGA5 in osteogenic differentiation of adult human mesenchymal stromal cells and suggest a therapeutic strategy using ITGA5 agonists to promote hMSC osteoblast differentiation and osteogenesis that may be used for bone regeneration in conditions where the recruitment or capacity of MSCs is compromised.

Click here to view.

Acknowledgments.

We thank Biopredic (Rennes, France) for providing hMSCs, Dr. S. Kuwada (University of Utah, Salt Lake City, UT) for the ITGA5 plasmid, Dr. M.J. Humphries (University of Manchester, Manchester, U.K.) for the SNAKA51 antibody and CRRETAWAC peptide, and Dr. G. Pagès (Unité Mixte de Recherche Centre National de la Recherche Scientifique 6543, University of Nice-Sophia Antipolis, France) for the DN-ERK plasmid. This work was supported by the Etablissement Français du Sang (Scientific Council No. 2004–08 and 2005.11), the Integrated Project of the European Commission FP6 research funding program (LSHB-CT-2003–503161 GENOSTEM), and the Association Rhumatisme et Travail (Hôpital Lariboisière, Paris, France). The use of ITGA5 agonists for applications in tissue regeneration is covered by European Patent No. 08 290 752.8.

Acknowledgments.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Data deposition: The data reported in this paper have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. {"type":"entrez-geo","attrs":{"text":"GSE18043","term_id":"18043","extlink":"1"}}GSE18043).

This article contains supporting information online at www.pnas.org/cgi/content/full/0812334106/DCSupplemental.

Footnotes

References

1. Pittenger MF, et al Multilineage potential of adult human mesenchymal stem cells. Science. 1999;284:143–147.[PubMed][Google Scholar]
2. Bianco P, Riminucci M, Gronthos S, Robey PGBone marrow stromal stem cells: Nature, biology, and potential applications. Stem Cells. 2001;19:180–192.[PubMed][Google Scholar]
3. Kassem MMesenchymal stem cells: Biological characteristics and potential clinical applications. Cloning Stem Cells. 2004;6:369–374.[PubMed][Google Scholar]
4. Marie PJ, Fromigué OOsteogenic differentiation of human marrow-derived mesenchymal stem cells. Regen Med. 2006;1:539–548.[PubMed][Google Scholar]
5. Bianco P, Robey PGStem cells in tissue engineering. Nature. 2001;414:118–121.[PubMed][Google Scholar]
6. Prockop DJ, Gregory CA, Spees JLOne strategy for cell and gene therapy: Harnessing the power of adult stem cells to repair tissues. Proc Natl Acad Sci USA. 2003;100:11917–11923.[Google Scholar]
7. Petite H, et al Tissue-engineered bone regeneration. Nat Biotechnol. 2000;18:959–963.[PubMed][Google Scholar]
8. Aubin JEBone stem cells. J Cell Biochem. 1998;30–31(Suppl):73–82.[PubMed][Google Scholar]
9. Karsenty G, Wagner EFReaching a genetic and molecular understanding of skeletal development. Dev Cell. 2002;2:389–406.[PubMed][Google Scholar]
10. Lian JB, et al Regulatory controls for osteoblast growth and differentiation: Role of Runx/Cbfa/AML factors. Crit Rev Eukaryot Gene Expr. 2004;14:1–41.[PubMed][Google Scholar]
11. Moursi AM, Globus RK, Damsky CHInteractions between integrin receptors and fibronectin are required for calvarial osteoblast differentiation in vitro. J Cell Sci. 1997;110:2187–2196.[PubMed][Google Scholar]
12. Franceschi RTThe developmental control of osteoblast-specific gene expression: Role of specific transcription factors and the extracellular matrix environment. Crit Rev Oral Biol Med. 1999;10:40–57.[PubMed][Google Scholar]
13. Zimmerman D, Jin F, Leboy P, Hardy S, Damsky CImpaired bone formation in transgenic mice resulting from altered integrin function in osteoblasts. Dev Biol. 2000;220:2–15.[PubMed][Google Scholar]
14. Damsky CH, Ilić DIntegrin signaling: It's where the action is. Curr Opin Cell Biol. 2002;14:594–602.[PubMed][Google Scholar]
15. Hynes ROIntegrins: Bidirectional, allosteric signaling machines. Cell. 2002;110:673–687.[PubMed][Google Scholar]
16. Humphries JD, Byron A, Humphries MJIntegrin ligands at a glance. J Cell Sci. 2006;119:3901–3903.[Google Scholar]
17. Ruoslahti E, Reed JCAnchorage dependence, integrins, and apoptosis. Cell. 1994;77:477–478.[PubMed][Google Scholar]
18. Zhang Z, Vuori K, Reed JC, Ruoslahti EThe α5β1 integrin supports survival of cells on fibronectin and up-regulates Bcl-2 expression. Proc Natl Acad Sci USA. 1995;92:6161–6165.[Google Scholar]
19. Kaabeche K, et al Cbl-mediated ubiquitination of α5 integrin subunit mediates fibronectin-dependent osteoblast detachment and apoptosis induced by FGFR2 activation. J Cell Sci. 2005;118:1223–1232.[PubMed][Google Scholar]
20. Dufour C, et al FGFR2-Cbl interaction in lipid rafts triggers attenuation of PI3K/Akt signaling and osteoblast survival. Bone. 2008;42:1032–1039.[PubMed][Google Scholar]
21. Haupl T, et al Gene expression profiling of rheumatoid arthritis synovial cells treated with antirheumatic drugs. J Biomol Screen. 2007;12:328–340.[PubMed][Google Scholar]
22. Clark K, et al A specific α₅β₁-integrin conformation promotes directional integrin translocation and fibronectin matrix formation. J Cell Sci. 2005;118:291–300.[Google Scholar]
23. Xiao G, et al MAPK pathways activate and phosphorylate the osteoblast-specific transcription factor, Cbfa1. J Biol Chem. 2000;275:4453–4459.[PubMed][Google Scholar]
24. Pagès G, et al Mitogen-activated protein kinases p42^{and p44}^{are required for fibroblast proliferation.}Proc Natl Acad Sci USA. 1993;90:8319–8323.[Google Scholar]
25. Bianco P, Kuznetsov SA, Riminucci M, Gehron Robey PPostnatal skeletal stem cells. Methods Enzymol. 2006;419:117–148.[PubMed][Google Scholar]
26. Cheng SL, Yang JW, Rifas L, Zhang SF, Avioli LVDifferentiation of human bone marrow osteogenic stromal cells in vitro: Induction of the osteoblast phenotype by dexamethasone. Endocrinology. 1994;134:277–286.[PubMed][Google Scholar]
27. Murshed M, Harmey D, Millan JL, McKee MD, Karsenty GUnique coexpression in osteoblasts of broadly expressed genes accounts for the spatial restriction of ECM mineralization to bone. Genes Dev. 2005;19:1093–1104.[Google Scholar]
28. Franceschi RT, Xiao GRegulation of the osteoblast-specific transcription factor, Runx2: Responsiveness to multiple signal transduction pathways. J Cell Biochem. 2003;88:446–454.[PubMed][Google Scholar]
29. Humphries MJ, Symonds EJ, Mould APMapping functional residues onto integrin crystal structures. Curr Opin Struct Biol. 2003;13:236–243.[PubMed][Google Scholar]
30. Mould AP, Askari JA, Humphries MJMolecular basis of ligand recognition by integrin α₅β₁. I. Specificity of ligand binding is determined by amino acid sequences in the second and third NH₂-terminal repeats of the α subunit. J Biol Chem. 2000;275:20324–20336.[PubMed][Google Scholar]
31. Humphries JD, et al Molecular basis of ligand recognition by integrin α₅β₁. II. Specificity of Arg-Gly-Asp binding is determined by Trp^{of the α subunit.}J Biol Chem. 2000;275:20337–20345.[PubMed][Google Scholar]
32. Mould AP, Humphries MJRegulation of integrin function through conformational complexity: Not simply a knee-jerk reaction? Curr Opin Cell Biol. 2004;16:544–551.[PubMed][Google Scholar]
33. Heckmann D, Kessler HDesign and chemical synthesis of integrin ligands. Methods Enzymol. 2007;426:463–503.[PubMed][Google Scholar]
34. Koivunen E, Gay DA, Ruoslahti ESelection of peptides binding to the α₅β₁ integrin from phage display library. J Biol Chem. 1993;268:20205–20210.[PubMed][Google Scholar]
35. Delorme B, Charbord PCulture and characterization of human bone marrow mesenchymal stem cells. Methods Mol Med. 2007;140:67–81.[PubMed][Google Scholar]
36. Hamidouche Z, et al FHL2 mediates dexamethasone-induced mesenchymal cell differentiation into osteoblasts by activating Wnt/β-catenin signaling-dependent Runx2 expression. FASEB J. 2008;22:3813–3822.[PubMed][Google Scholar]