TGF-β Enhances the Integrin α2β1-Mediated Attachment of Mesenchymal Stem Cells to Type I Collagen

Abstract

The heterodimeric integrins are important receptors for the attachment of cells to their extracellular matrix. Here, we studied the attachment of human mesenchymal stem cells (MSCs) to type I collagen (col-1), which is part of the extracellular matrix in bone, skin, and connective tissues. Furthermore, we examined how TGF-β influences the integrin expression and attachment of MSC. Using flow cytometry, immunoblot, and RT-PCR, we report that MSC express several integrin subunits, including the α₂β₁ integrin (VLA-2, CD49b/CD29). TGF-β increases the expression of integrin subunits α₂, α₆, and β₁ in MSC, thereby enhancing the attachment of MSC to col-1. The TGF-β-mediated up-regulation of the expression of the integrin subunits α₂ and α₆ is mainly mediated in MSC by Smad2.

Introduction

M esenchymal stem cells (MSCs) are multipotent cells [1 –4] that were first identified in bone marrow stroma. They seem to be closely related to adventitial reticular cells [5], and can also be found as bone lining cells in the periosteum [6] and endosteum. In vascularized tissues such as adipose tissue or dental pulp, MSC function as pericytes on the surface of blood vessels [7 –10]. The MSC are located within the bone marrow, but much less is known about the MSC in their microenvironment, the MSC niche, than about HSC in the well-characterized HSC niche [11]. In general, adult stem cell niches consist of surrounding niche cells, extracellular matrix (ECM) proteins, and growth factors. This environment prevents the differentiation and apoptosis of the stem cells and regulates their proliferation [12,13]. External stimuli activate stem cells to leave the niche, multiply, and undergo differentiation processes. The emerging progenitor cells contribute to tissue regeneration and wound repair. In this respect, it is important to characterize MSC niches with regard to their ECM composition and effects of soluble factors on MSC.

Because of their differentiation capacity, MSC are a highly promising tool in therapeutic approaches [14,15]. Once placed in the defective tissue in combination with an adequate carrier material, they are able to contribute to tissue regeneration. Important aspects in MSC research include finding optimal carrier materials for the tissue engineering application being developed and studying the effects of soluble factors that are deposited in the corresponding tissue during regenerative processes. Carrier materials should integrate easily into the target tissue, supporting the integration of MSC and providing factors necessary for their regenerative action.

In many tissues of the human body, collagens are prominent components of the ECM, so they would seem to be suitable candidates for tissue engineering carrier materials. Long-term bone marrow cultures and gene expression profiling revealed that bone marrow MSC themselves produce collagens as well as laminins 16 –19]. Furthermore, bone marrow contains cytokines and growth factors that are either soluble or bound to the ECM components [20]. Outside the bone marrow, MSC are situated near the basement membranes of blood vessels, which mainly consist of type IV collagen (col-4), laminin-411 (LM-411), and laminin-511 (LM-511) [21]. MSC express surface proteins for ECM attachment [22 –24] and growth factor receptors [25,26]. TGF-β is one of the factors to which the MSC respond [17,27].

In this study, we analyzed the effect of TGF-β on the attachment of MSC to type I collagen (col-1). Type I collagen (col-1) is the most abundant protein in the ECM. In bone 90% of the ECM protein mass is col-1, and it is found in other connective tissues as well. Receptors for the attachment to col-1 are integrins α₁β₁, α₂β₁, α₃β₁, α₁₀β₁, and α₁₁β₁ [28]. We also investigated the interaction of MSC with LM-111 and compared it to the attachment to col-1. LM-111 is the first laminin expressed during embryonic development [29]. In the adult, it is part of the basal membranes in epithelia. LM-111 could serve as a carrier material for bone tissue engineering or coating for prostheses, as it supports osteogenic differentiation of MSC [30,31]. Integrins binding to LM-111 are the α₁β₁, α₂β₁, α₆β₁, α₇β₁, and α₉β₁ receptors [29].

TGF-β is a growth factor involved in wound healing, in matrix turnover during fibrotic processes [32], as well as in osteogenesis [33], and it induces chondrogenic differentiation in MSC [1]. TGF-β regulates the attachment and integrin expression pattern in various cell types [34 –37], including synovial fibroblasts (SF) [38]. Here, we analyzed the attachment of MSC to col-1 and LM-111, and we examined the influence of TGF-β on the integrin expression pattern and the attachment of MSC.

Materials and Methods

Isolation and characterization of MSC

MSC were isolated by Ficoll-Paque^™ (Amersham Biosciences, Freiburg, Germany) gradient centrifugation from femoral bone marrow aspirates of patients undergoing endoprosthetic surgery. This study was approved by the local ethics committee. MSC were grown in Mesenchymal Stem Cell Growth Medium (Lonza, Basel, Switzerland).

Characterization of MSC was performed by flow cytometry and differentiation assays [1,39]. Flow cytometry analysis of MSC included staining with antibodies against the cell surface inclusion markers CD73 (BD Pharmingen, Heidelberg, Germany), CD90 (R&D Systems, Wiesbaden, Germany), CD105 (Serotec, Düsseldorf, Germany), and exclusion markers CD34 (Biolegend, Echingen, Germany), CD14, and CD45 (R&D Systems). Differentiation of MSC into adipogenic, osteogenic, and chondrogenic lineages was induced by specific differentiation media.

Adipogenic differentiation medium contained 0.1 mM indomethacin (Calbiochem, Darmstadt, Germany), 0.01 mg/mL insulin, 0.5 mM 3-isobutylmethylxanthine, and 1 μM dexamethasone (all from Sigma-Aldrich, Munich, Germany). Differentiation was detected by Oil Red staining. For this purpose, a 0.5% Oil Red stock solution in isoporopanol was diluted to 3:2 with water. After filtration, the cells were incubated in Oil Red staining solution for 30 min. Then cells were washed 3 times with PBS and recorded by microscopy (Leica DM-IRBE). Expression of LPL and PPARγ mRNA was measured by quantitative RT-PCR (qRT-PCR) using specific primers (Table 1).

Table 1.

Primers for the RT-PCR

Gene	Primer sequence	Product length	EMBL number
LPL	Forward 5′-GAG ATT TCT CTG TAT GGC ACC-3′	275 bp	NM_000237.1
	Reverse 5′-CTG CAA ATG AGA CAC TTT CTC-3′
PPARγ	Forward 5′-GCT GTT ATG GGT GAA ACT CTG-3′	330 bp	NM_015869.3
	Reverse 5′-ATA AGG TGG AGA TGC AGG CTC-3′
Alkaline phosphatase	Forward 5′-CCA CGT CTT CAC ATT TGG TG-3′	195 bp	NM_000478.2
	Reverse 5′-AGA CTG CGC CTG GTA GTT GT-3′
Runx-2	Forward 5′-TCTGGCCTTCCACTCTCAGT-3′	161 bp	NM_001024630.1
	Reverse 5′-GACTGGCGGGGTGTAAGTAA-3′

Osteogenic differentiation medium consisted of DMEM supplemented with 0.1 μM dexamethasone, 10 mM β-glycerophosphate (Merck, Darmstadt, Germany), and 50 μM l-ascorbate (Sigma-Aldrich). Differentiation was proven by von Kossa and Alizarin Red staining, and by quantification of mRNA encoding alkaline phosphatase, Runx-2 (Table 1), osteocalcin, and osteopontin (Search LC, Heidelberg, Germany). For the von Kossa staining, cells were fixed in methanol. After 3 washing steps in PBS, cells were incubated for 1 h with a 5% aqueous silver nitrate solution, followed by reduction with 5% sodium carbonate in 9% formalin and fixation in an aqueous solution of 5% sodium thiosulfate. For the Alizarin Red staining, cells were fixed in methanol, washed in PBS, and stained using an aqueous solution of 1.4% Alizarin Red (pH 4.0). Stained cells were photographed under a microscope (Leica DM IRBE).

Chondrogenic differentiation was initialized in pellet cultures using 10 ng/mL TGF-β3 (R&D Systems). After 4 weeks, proteoglycans in the pellets were stained with Alcian blue using a 1% Alcian blue solution in 3% acetic acid, and types I and II collagen were detected by immunohisto-chemical staining (monoclonal antibodies for types I and II from MP Biomedicals, Illkirch, France) and documented by microphotography [40].

Cell attachment assays

Attachment assays [41] were performed to investigate the binding of MSC to col-1 and LM-111 and to display the effect of TGF-β on the adhesion of MSC to these proteins. In brief, 0.01 fmol to 10 fmol of col-1 or LM-111 (both from BD Falcon Heidelberg, Germany) were dissolved in 2 μL PBS or 0.2 N acetic acid, respectively, spotted onto tissue culture dishes and air-dried, followed by a blocking step using 10 mg/mL BSA/PBS. Spots of 100 fmol and 1 pmol poly-l-lysine (PLL) served as controls.

To detect TGF-mediated effects, the MSC were preincubated with 10 ng/mL TGF-β1 (R&D Systems) for 24 h prior to the attachment assay or left untreated as controls. Then, cells were harvested by mild detachment, resuspended in serum-free medium with or without integrin-blocking antibodies (listed in Table 2)for 30 min, and seeded onto the protein spots in the cell culture dishes. After 15 min of incubation on the spots, the floating cells were removed and the adherent cells were recorded by microphotography.

Table 2.

Integrin-Blocking Antibodies Used in the Attachment Assay

Integrin	Clone	Source
α₂	P1E6	Chemicon
α₃	P1B5	Chemicon
α₆	G0H3	BD Pharmingen
α₉β₁	Y9A2	Abcam
β₁	4B4	Beckmann Coulter
Isotype control	W6/32.HL	[42]

Analysis of integrin expression by flow cytometry

The cells were either stimulated with 10 ng/mL TGF-β1 for 24 h or left untreated as controls. After harvesting and washing in PFEA buffer (PBS with 2 mM EDTA, 2 % FCS, and 0.01% sodium azide), primary antibodies specific for integrin β₁ (clone 4B4; Beckmann Coulter, Krefeld, Germany), α₁ (PE-labeled, clone SR84; BD Pharmingen), α₂ (PE-labeled, clone 12F1; BD Pharmingen), α₃ (clone P1B5; Chemicon, Schwalbach, Germany), α₆ (PE-labeled, clone 450–30A; Abcam, Cambridge, MA), α₉ (clone Y9A2; Abcam), and α₁₁ (Abnova, Taipei, Taiwan) were incubated for 20 min at 4°C. In the case of the unlabeled primary antibodies, the incubation with the PE-labeled secondary antibody (Jackson Immunoresearch, Newmarket, UK) was carried out for 20 min at 4°C. Cells were analyzed in a flow cytometer (BD LSR II) using the FACS^® DIVA software.

Quantitative RT-PCR analysis

For analysis of transcripts, RNA was isolated from MSC, which were either left untreated or stimulated with 10 ng/mL TGF-β1. To perform reverse transcription, we utilized 1 μg RNA (RT-for-PCR-Kit, Clontech, Saint-Germainen-Laye, France). Equal amounts of the cDNA were analyzed by qRT-PCR. Hot Start Taq polymerase and SybrGreen from the Light Cycler^® FastStart DNA Master kit (Roche, Penzberg, Germany) were employed for amplification and detection of the PCR products. Primers for the differentiation markers were designed according to published sequences [43] and are listed in Table 1. Integrin PCR primers are listed in Table 3. Primers for the differentiation markers are listed in Table 1. Primers for GAPDH and a recombinant GAPDH standard were obtained from Search LC (Heidelberg, Germany). The amplification was performed in the Light Cycler 2.0 (Roche), and relative quantification of the amplificates was carried out using the Fit Points method [44].

Table 3.

Primers Employed in qRT-PCR for the Quantification of Integrins

Gene	Primer sequence	Product length	EMBL number
Integrin α₁	Forward 5′-AACGAGGCACAATTCTGGAC-3′	247 bp	NM_181501
	Reverse 5′-CGGTAGCCCATCTTTGGATA-3′
Integrin α₂	Forward 5′-GGGACTTTCGCATCATCAAC-3′	227 bp	NM_002203
	Reverse 5′-CCGAGCTTCCATAAAATTGC-3′
Integrin α₃	Forward 5′-GAGACTGTGCTGACCTGTGC-3′	248 bp	NM_002204
	Reverse 5′-TCCGAGTCAATGTCCACAGA-3′
Integrin α₆	Forward 5′-CAG CCT TCA ACT TGG ACA CTC GG-3′	390 bp	NM_000210.2
	Reverse 5′-CAA AGA TGT CTC GGG ATT CCT GC-3′
Integrin α₇	Forward 5′-ATC AAG AGC TTC GGC TAC TCC-3′	358 bp	NM_002206.1
	Reverse 5′-GCT TGG GTT CTT CCA GGT TAC-3′
Integrin α₉	Forward 5′-AGT TAC AGT GTG CTG TTA GGC AAG-3′	383 bp	NM_002207.2
	Reverse 5′-GTC TCC AAC CTC CTT TGT ATA TGG-3′
Integrin α₁₀	Forward 5′-GTGGAAGCTTGGCTTCTTTG-3′	231 bp	NM_003637.3
	Reverse 5′-AGCAGCATCCCTAGGACTCA-3′
Integrin α₁₁	Forward 5′-CTGGATCATTGTAGGCAGCA-3′	211 bp	NM_001004439.1
	Reverse 5′-AGGCCTGGGTCTCAACACTA-3′
Integrin β₁	Forward 5′-GCGAAGGCATCCCTGAAAGT-3′	665 bp	NM_133376.1
	Reverse 5′-GGACACAGGATCAGGTTGGA-3′

Western blot analysis

To investigate the expression of proteins, western blot analyses were performed as described [38]. MSC were lysed in RIPA buffer (c-c-pro, Oberdorla, Germany). The protein concentration in lysates was measured using the RCDC Kit (BioRad, Munich, Germany), and 100 μg of protein per lane were applied to the SDS-PAGE. After blotting, the nitrocellulose membrane was blocked with 5% milk/PBS/0.1% Tween-20 (blocking buffer). For detection of the integrin α₇ subunit, the blots were incubated overnight with antiintegrin α₇ antibody (1:1,000; Abcam) in blocking buffer. Secondary HRP-labeled goat anti-mouse antibody (Jackson Immunoresearch) was used in the ratio of 1:1,000 in blocking buffer. Binding of antibodies was visualized by chemiluminescence (ECL; GE Healthcare, Freiburg, FRG) and recorded by a luminescence-sensitive CCD camera system (Diana, Raytest Inc., Straubenhardt, FRG).

TGF-β signaling pathway

For investigation of the TGF-β-dependent signaling and gene regulation, qRT-PCR was employed. The cells were preincubated for 30 min with inhibitors of the TGF-β signaling pathway. The following inhibitors were used at the given concentrations: 100 nM of TGF-β RI kinase inhibitor, 20 μM of the MEK inhibitor PD98059, and 20 μM of the p38 inhibitor SB203580 (all from Calbiochem). After preincubation with the inhibitors, cells were exposed to 10 ng/mL of TGF-β1. Mock-treated MSC or MSC activated with 10 ng/mL TGF-β1 without inhibitors served as controls. After 24 h of incubation, RNA was isolated and qRT-PCR was performed to quantify integrin mRNA transcripts.

Statistical analyses

Statistical significance was assessed by 2-sided Student's t-tests. Probability values (P) ≤0.05 (*), 0.01 (**), or 0.001 (***) were considered to be statistically significant and marked in the figures accordingly.

Results

Characterization of MSC

To characterize the isolated and expanded MSC, we analyzed their differentiation potential and the expression of typical surface marker proteins [39]. The cells purified from bone marrow aspirates displayed strong staining for CD73, CD90, and CD105 and expressed low levels of CD14, CD34, and no CD45 (Fig. 1). Furthermore, upon appropriate stimulation, MSC differentiated into cells of the adipogenic, chondrogenic, and osteogenic lineages (Fig. 2). Figure 2A displays the spindle-like shape of undifferentiated MSC. Adipogenic differentiation of MSC was confirmed by Oil Red staining of the intracellular lipid vesicles (Fig. 2B) and by the enhanced expression of the adipogenic marker genes PPARγ (10.3 ± 1.7-fold, n = 4, P < 0.01) and LPL (7,425 ± 7,079-fold induction over undifferentiated control cells, n = 4, P > 0.05, Fig. 2C). Chondrogenically differentiated cell pellets contain proteoglycans, which were detected by Alcian blue staining (Fig. 2D). Immunostaining demonstrates that these pellets contain collagen type I (Fig. 2E), but more important for the chondrogenic differentiation, collagen type II (Fig. 2F), corroborating our recent in-depth study [40]. In osteogenically differentiated MSC, mineralization was proven by Alizarin Red (Fig. 2G) and von Kossa staining (Fig. 2H). Enhanced levels of mRNA encoding alkaline phosphatase (29 ± 39.8-fold induction over undifferentiated control cells, n = 4, P > 0.05), osteocalcin (3 ± 2.3-fold, n = 4, P > 0.05), osteopontin (2 ± 0.5-fold, n = 4, P < 0.05), and Runx-2 (6 ± 5.7-fold, n = 4, P > 0.05) were detected in osteogenically differentiated MSC in comparison to undifferentiated MSC (Fig. 2I).

FIG. 1.

Expression of cell surface antigens on mesenchymal stem cell (MSC). Human bone marrow-derived MSC were investigated for the expression of the MSC marker antigens CD73, CD90, and CD105 by flow cytometry. For MSC exclusion marker antigens are CD14, CD34, and CD45. Flow cytometry analysis revealed that cells isolated from bone marrow expressed the typical surface markers for MSC [39]. The dashed lines denote unstained cells, tinted areas the fluorescent-stained cells. The signal intensities are presented in the histogram mode on the x-axes and the number of cells staining on the y-axes.

FIG. 2.

Differentiation of mesenchymal stem cell (MSC) into adipogenic, osteogenic, and chondrogenic lineages. To confirm the differentiation capacity of the MSC after expansion in vitro, the cells were split, incubated for 28 days adipogenic, chondrogenic, or osteogenic differentiation media, and analyzed for a characteristic phenotype and gene expression. (A) Undifferentiated MSC. (B) Adipogenic differentiation was confirmed by Oil Red staining. (C) In adipogenically differentiated cells (black bars), mRNA expression of the adipogenic marker genes LPL and PPARγ was increased in comparison to undifferentiated control cells (blue bars). (D) Chondrogenically differentiated cell pellets were stained with Alcian blue to detect proteoglycans. (E) Detection of type I collagen (col-1) in chondrogenically differentiated cells. (F) Detection of type II collagen in chondrogenically differentiated cells. (G) After osteogenic differentiation, the mineralized extracellular matrix (ECM) was visualized using Alizarin Red, or (H) by von Kossa staining. (I) In osteogenically differentiated cells (black bars), expression of the osteogenic marker genes alkaline phosphatase (ALP), Runx-2, osteopontin (OPN), and osteocalcin (Ocal) was elevated compared to undifferentiated controls (blue bars). The qRT-PCR data represent the mean values ± SD from n ≥ 3 different experiments. Scale bars in the micrographs = 250 μm.

Attachment of MSC to extracellular matrix proteins

The attachment of MSC to immobilized spots of col-1 and LM-111 was investigated. MSC attach well to spots of 10 fmol and 1 fmol, to some extent to 0.1 fmol, and only weakly to 0.01 fmol of col-1. Furthermore, they attach well to 10 fmol or 1 fmol of LM-111, whereas 100 fmol of LM-111 per spot seemed to be above optimal ranges (Fig. 3). Attachment to 100 fmol of PLL served as control (Fig. 3).

FIG. 3.

Attachment of mesenchymal stem cell (MSC) to extracellular matrix (ECM) proteins. To investigate the attachment of MSC to type I collagen (col-1) or LM-111, MSC were harvested and incubated in serum-free medium for 15 min on serial dilutions of col-1 (spots of 10 fmol and 0.01 fmol) and LM-111 (spots of 100 fmol and 1 fmol). Floating cells were removed and the MSC attached to the spots were recorded. Attached cells are visible as white dots on the protein spots, while areas coated by BSA only and those free of cells are visible as black zones. The MSC attached well to spots of 10 fmol and 1 fmol of col-1, to some extent to 0.1 fmol, and only weakly to 0.01 fmol of col-1 (top). MSC attached well to 10 fmol or 1 fmol of LM-111, to a lesser extent to 100 fmol of LM-111 (bottom). Attachment of MSC to poly-l-lysine (PLL, 100 fmol) served as a control.

The integrin β₁ subunit is part of the major extracellular matrix receptors for the binding of cells to collagens and laminins. Using the integrin β₁-specific antibody 4B4, we could completely block the attachment of MSC to col-1 and LM-111 (Fig. 4). Furthermore, the attachment of MSC to col-1 could be reduced by the integrin α₂-specific antibody P1E6, while the attachment to LM-111 was not affected by this antibody (Fig. 4). Using function-blocking antibodies against integrins α₃, α₆, and α₉ had no effect on the attachment of MSC to col-1. Anti-integrin α₉ antibody treatment had a minor effect on binding of MSC to LM-111. Cells on PLL spots served as controls. Incubation of the cells even with the β₁-specific integrin-blocking antibody had no effect on the attachment of MSC to PLL (Fig. 4).

FIG. 4.

Contribution of individual integrins to the attachment of mesenchymal stem cell (MSC) to type I collagen (col-1) and LM-111. The MSC were harvested and either preincubated with function-blocking antibodies (see Table 2)or left untreated, to investigate the binding on spots of col-1, LM-111, or poly-l-lysine (PLL). The MSC attached well to col-1, LM-111, and PLL (left). The attachment of MSC to col-1 and LM-111 but not PLL could be completely blocked using the integrin β₁ chain-specific antibody 4B4 (right). The integrin α₂-specific antibody P1E6 blocked the attachment of MSC to col-1 but not to LM-111 or PLL. Incubation of MSC with α₃-, α₆-, or α₉-specific antibodies did not influence the attachment of MSC to col-1 or PLL. The anti-integrin α₉ antibody reduced the attachment of MSC to LM-111 slightly. Preincubation of MSC with isotype control antibodies (isotype c) did not modulate the attachment of the cells to the spots. Cells attached to PLL served as additional controls, as the attachment to PLL is mediated by electrostatic interactions.

TGF-β influences the attachment of MSC to type I collagen

The effect of TGF-β1 on the attachment of MSC to col-1 and LM-111 was analyzed in a concentration-dependent cell attachment assay (Fig. 5). Cells were either left untreated or preincubated with TGF-β1 for 24 h prior to the attachment assay. TGF-β1 enhanced the attachment of MSC to col-1, as TGF-β-activated MSC displayed binding to 1 fmol, 0.1 fmol, and to some extent even to 0.01 fmol, whereas the mock-treated MSC failed to attach to 1 fmol of col-1. Spotting with concentrations as low as 10 fmol of col-1 revealed a difference in cell density of TGF-β1-activated cell attached to col-1 compared to the mock controls (Fig. 5). In contrast, the attachment of MSC to LM-111 was not affected by TGF-β1 (Fig. 5). The naïve MSC attached to spots of 10 fmol and 1 fmol of LM-111 with the same density as the TGF-β1-activated cells did. At 0.1 fmol or 0.01 fmol of LM-111 per spot specific binding of MSC fainted. Only some TGF-β1-activated MSC spread anywhere on the slides, but not specifically on the LM-111-coated spot (Fig. 5).

FIG. 5.

Influence of TGF-β1 on the attachment of mesenchymal stem cell (MSC) to type I collagen (col-1) and laminin-111. The activation of MSC by TGF-β1 (10 ng/mL, 24 h) enhances the attachment of the cells to col-1 (10, 1, 0.1 fmol) in comparison to untreated MSC. An enhanced attachment of TGF-β1-activated MSC to LM-111 (10, 1, 0.1, 0.01 fmol) was not observed.

Integrin expression pattern on MSC and SF

Attachment to col-1 and LM-111 is accomplished by integrin receptors, while adhesion to PLL occurs via electrostatic interactions between the negatively charged lipid membrane and the positively charged PLL. Our results suggested that TGF-β may change the integrin expression pattern in MSC, resulting in a stronger attachment to col-1. To confirm this hypothesis, integrin expression patterns were analyzed in more detail.

Performing RT-PCR analyses, we could detect the designated amplificates for integrin subunits α₁, α₂, α₃, α₆, α₇, α₉, α₁₀, α₁₁, and β₁ in MSC (Fig. 6A). Transcripts were then quantified by qRT-PCR (Fig. 6B). MSC show high transcript levels for integrins α₁ (5.9 × 10^−;3), α₃ (2 × 10^−;2), α₇ (1.9 × 10^−;2), α₁₁ (2.2 × 10^−;2), and β₁ (1.3 × 10^−;1) and low expression of α₂ (1.9 × 10^−;4), α₆ (6.8 × 10^−;4), α₉ (2.8 × 10^−;5), and α₁₀ (5.3 × 10^−;4) mRNA. The depicted transcript levels are normalized to the expression of the housekeeping gene GAPDH. Flow cytometry analyses revealed high expression of integrin subunits α₁, α₂, α₃, and β₁ on MSC, and a mid-level expression of integrin subunits α₆, α₉, and α₁₁ (Fig. 6C). A strong expression of integrin α₇ was detected on MSC by western blotting (Fig. 6D).

FIG. 6.

Integrin expression in mesenchymal stem cell (MSC). The expression of type I collagen (col-1)- and LM-111-binding integrins was analyzed by RT-PCR and qRT-PCR, flow cytometry, and immunoblot. (A) Detection of mRNA transcripts for the integrin chains α₁, α₂, α₃, α₆, α₇, α₉, α₁₀, α₁₁, and β₁ in MSC by PCR, followed by gel electrophoresis of the products. Detection of GAPDH transcripts served as a control. The length of the PCR products can be delineated from the 100 bp DNA size marker (right). (B) The expression levels of integrins were enumerated by qRT-PCR. The MSC expressed high amounts of integrin chains α₁, α₃, α₇, α₁₁, and β₁ and low amounts of integrin subunits α₂, α₆, α₉, and α₁₀ on the mRNA level. The data represent the mean transcript values ± SD from MSC of n ≥ 3 different donors normalized to the transcript amounts of GAPDH in each individual sample. (C) Detection of integrins on MSC by flow cytometry revealed that the cells presented high amounts of integrin subunits α₁, α₂, α₃, and β₁ and lower amounts of integrins α₆, α₉, and α₁₁ on their surfaces. Dashed lines denote unstained cells, dotted lines cells that reacted with secondary antibody only (if necessary), and solid lines those stained with integrin-specific antibody (and detection reagent as necessary). The signal intensities are displayed in the histogram mode on the x-axis, the number of stained cells on the y-axis. (D) As the anti-α₇ antibody at hand was not suitable for flow cytometry, expression of integrin α₇ in MSC at the protein level was confirmed by western blot. Extracts of Caco-2 cells served as positive control [38]. A protein size marker on the left indicated the molecular weight of the proteins detected.

Influence of TGF-β on the integrin expression of MSC

To determine the effect of TGF-β1 on integrin expression, MSC were stimulated with TGF-β1 for 24 h, and the expression patterns of the integrin subunits were investigated by qRT-PCR and flow cytometry (Fig. 7). In MSC, TGF-β1 enhanced the mRNA encoding integrin subunits α₂ (15.8-fold induction over untreated controls, P ≤ 0.05), α₆ (4.7-fold induction, P ≤ 0.05), and β₁ (1.8-fold induction, P ≤ 0.01) statistically significant (Fig. 7A). The gain of the integrin β₁ transcripts was rather low, as this subunit was expressed in naïve MSC at high levels (Fig. 6B and 6C). The transcript levels of the other integrin chains under investigation were not regulated by TGF-β1 (Fig. 7A). The cell surface expression of the α₁, α₃, and α₉ integrin subunits was not regulated by TGF-β1, the signal observed for the α₁₁ integrin was reduced, but the expression of α₂, α₆, β₁ integrin subunits was slightly elevated by TGF-β1 (Fig. 7B). The TGF-β1-mediated gain of integrin α₂ signal intensity observed by FACS was low, and the peaks of the histograms were not clearly separated. In contrast, the mean of intensity for integrins α₆ and β₁ shifted to higher signal intensity, indicating an elevated assemblage of both subunits on the surface of the MSC (Fig. 7B).

FIG. 7.

Regulation of integrin expression in mesenchymal stem cell (MSC) by TGF-β1. MSC were activated by TGF-β1 (10 ng/mL, 24 h) or left untreated and regulation of integrin expression was investigated. (A) qRT-PCR analysis of integrin mRNA in MSC. Upon exposure to TGF-β1, levels of mRNA transcripts for integrin subunits α₂, α₆, and β₁ were significantly higher than the corresponding mRNA levels in unexposed control cells. The mRNA expression of all other analyzed integrin subunits was not significantly affected by TGF-β1. The data represent the mean values of mRNA induction ± SD from MSC of n ≥ 3 donors. The illustration displays the normalized induction index of intregrin transcripts in TGF-β1-activated MSC compared to the not-activated MSC. (B) Effect of TGF-β1 on integrin surface expression on MSC. Dashed lines: unstained cells, dotted lines: secondary antibody only (if necessary), black lines: integrin staining of cells without TGF-β exposure, tinted areas: integrin staining of TGF-β-treated cells. Upon exposure to TGF-β, MSC showed a slightly enhanced surface expression of integrin chain α₂, and a somewhat larger increase in the expression of integrin chains α₆ and β₁, while expression of the subunit α₁ was not affected by TGF-β1. The surface expression of integrin chains α₃ and α₁₁ was slightly down-regulated by TGF-β1. The signal intensities of the staining are displayed in the histogram mode on the x-axis, the number of stained cells on the y-axis.

TGF-β-mediated signaling pathway in MSC

MSC were incubated with TGF-β1 and in parallel exposed to inhibitors for different components of the TGF-β-signaling pathway, including SB203580, PD98059, and TGF-β RI kinase inhibitor. Twenty-four hours later, mRNA expression of TGF-β1-regulated integrin subunits α₂, α₆, and β₁ was measured by qRT-PCR.

The expression of the β₁ integrin was least modulated by interfering with the TGF-dependent intracellular signaling and p38 MAPK blocker SB203580, ERK blocker PD98059, TGF-β RI kinase inhibitor, which interferes with the SMAD2 signaling, had no significant effects (Fig. 8). The expression of the α₂ integrin was activated by TGF-β1 but the inhibition of the p38MAPK or ERK pathways by SB203580 or PD98059 had no effect. But blocking the SMAD2 signaling by addition of the TGF-β RI kinase inhibitor reduced the TGF-β1-induced expression of α₂ integrin (Fig. 8). TGF-β1 enhanced the expression of the α₆ integrin statistically significant (P < 0.05). Addition of SB203580 to the activated MSC did not reduce the expression of the α₆ integrin, but PD98059 reduced its expression to some extent. Inhibition of the SMAD2-signaling pathway by addition of the TGF-β RI kinase inhibitor reduced the TGF-β1-induced expression of α₆ integrin statistically significant (P < 0.05, Fig. 8). In MSC, TGF-β triggers the phosphorylation of Smad 2 at its C-terminus (Ser465/467), while the levels of phospho-ERK1/2 and phospho-p38 were not affected by TGF-β1 (data not shown). We conclude the phosphorylation of Smad 2 induces a signaling pathway that results in the enhancement of integrin α₂, α₆, and β₁ expression in MSC.

FIG. 8.

Modulation of the TGF-β1-induced expression of integrins. Mesenchymal stem cell (MSC) were incubated with TGF-β1 in the presence or absence of inhibitors for the TGF-β signaling pathway, and expression of mRNA encoding integrin subunits α₂, α₆, and β₁ was measured by qRTPCR. Untreated cells (white bars) and TGF-β1-activated cells (black bars) served as controls. The expression of the mRNA encoding the β₁, α₂, and α₆ integrin subunits was diminished by the TGF-β RI inhibitor, which precedes the phosphorylation of Smad2. The ERK inhibitor PD98059 reduced the induction of integrin α₆ chain, but not of α₂ and β₁ expression, and SB203580 (p38MAPkinase) did not influence the mRNA expression of the analyzed integrins. The figure displays the normalized transcript levels of the integrins in treated MSC as indicated compared to untreated MSC.

Discussion

In this study, we investigated the attachment of MSC to ECM proteins col-1 and LM-111 in the presence and absence of TGF-β1.

In general, attachment of cells to col-1 and LM-111 can be mediated by integrin receptors α₁β₁ and α₂β₁ (col-1 and LM-111), α₃β₁, α₁₀β₁, α₁₁β₁ (col-1), α₆β₁, α₇β₁, and α₉β₁ (LM-111). We report that MSC express high amounts of integrin subunits α₁, α₂, α₃, α₇, and β₁ and medium levels of integrin chains α₆, α₉, and α₁₁ on the protein level. The expression of integrin α₁₀ was observed by RT-PCR but not by flow cytometry.

The attachment of MSC to col-1 or LM-111 can be abolished by blocking the integrin β₁ chain with the antibody 4B4. Furthermore, the attachment of MSC to col-1 is at least partially blocked by the integrin α₂ antibody P1E6. The incubation with an antibody against integrin α₃ did not block attachment of MSC to col-1, although they express high amounts of this integrin subunit. This indicates that the integrin α₂β₁ (VLA-2, CD49b/CD29) is a key factor for the attachment of MSC to col-1. The role of other collagen-binding integrin subunits, including α₁, α₁₀, and α₁₁, must be further examined in this respect, when function-blocking antibodies become available.

The attachment of MSC to LM-111 could not be blocked by antibodies against integrins α₆ and α₉, although these integrin subunits are known to be laminin receptors [29]. The MSC strongly expressed integrins α₇ and α₁. These integrins chains may keep the MSC attached to LM-111, even when other integrins are blocked.

The integrins analyzed in this study are part of the interacting network in the bone marrow and the perivascular niche, anchoring MSC to the ECM of these tissues. Type I collagen is found in the bone marrow stroma, providing contact points for the MSC to attach. MSC themselves can express col-1 [17], thus contributing to the integrity of bone marrow stroma. Both the bone marrow and the basement membranes in blood vessels contain the laminin isoforms LM-411/421 and LM-511/521 [19,21]. Receptors responsible for the attachment to LM-411/421 and LM-511/521 include integrins α₂β₁, α₃β₁, α₆β₁, and α₇β₁. Showing strong expression of these integrins, MSC may attach well to laminins in the perivascular niche. Bone marrow-derived stromal cells, which contain to a large part MSC, expressed LM-411/421 and 511/521, and thus may contribute to the composition of their own niches [19].

TGF-β1 regulates the integrin expression pattern in MSC and enhances their attachment to col-1. The TGF-β1-induced signaling pathway mainly proceeds via TGF-βRI and subsequently phosphorylation of Smad2. In osteoblasts and vascular smooth muscle cells, an involvement of Smad proteins in the regulation of integrin expression was shown as well [45,46]. In chondrosarcoma cells, the LM-511 receptor integrin ανβ₃ is up-regulated by TGF-β via focal adhesion kinase (FAK), PI3 kinase, and NF-κB [47]. It is possible that these components are involved in the regulation of integrins in MSC as well. An influence of TGF-β on the integrin expression pattern was further reported in MG-63 human osteosarcoma cells [35] and alveolar epithelial cells [34]. Enhanced expression of integrin subunits α₁, α₂, α₃, α₅, and β₁ was reported in WI-38 lung fibroblasts after exposure to TGF-β [48]. In hepatocytes, TGF-β changes the expression of α₅β₁ and thus diminishes attachment to fibronectin [37]. By modulating the attachment of these cells to the ECM, TGF-β contributes to cell mobilization, organ development, wound healing, and fibrosis.

Recently, we reported that TGF-β1 enhanced the expression of matrix metalloproteinases and IL-16 in SF and also modulated the integrin expression pattern on SF [38,49]. TGF-β enhanced the expression of integrin β₁ in SF as well, but in contrast to the MSC, an up-regulation of the integrin α₂ or α₆ proteins could not be found in SF [38]. However, the TGF-β1-induced modification of the integrin expression pattern in SF is involved in the pathology of rheumatoid arthritis, where TGF-β1 enhances the attachment of the fibroblasts to LM-111. The co-signaling processes between TGF-β1 and integrin receptors induce the expression of matrix-degrading proteases in SF, leading to destruction of the rheumatic tissue [49]. In contrast, the TGF-β1-enhanced attachment of MSC to col-I, which is mediated by integrin α₂β₁, might contribute to the embedding of MSC into defective tissues during wound healing.

MSC are mobilized into peripheral blood after skeletal muscle injury [50] and acute large skin damage [51], but it is not known whether TGF-β1 is involved in this process. A recombinant TGF-β1 fusion protein with a collagen-binding domain was shown to induce the migration of MSC [52]. But the effect of TGF-β alone on the migration of MSC remains controversial [53,54]. It is possible that additional factors stimulate mobilization of MSC into the peripheral blood, while TGF-β1 is locally secreted and promotes their engraftment into the defective tissue by enhancing the expression of specific integrins.

Migration of MSC can be stimulated by growth factors such as BMP-2, BMP-4, bFGF, EGF, PDGF-ab, and PDGF-bb [53,54]. This process is often coupled to changes in the integrin expression pattern of the target cells. PDGF-ab enhances the expression of integrin β₁ in dermal fibroblasts, thus affecting attachment to col-1 and fibronectin [55]. bFGF, which enhances the expression of integrin α₁₀ in MSC [56], regulates migration of MSC in a concentration-dependent manner [53,57]. Hypoxia is another stimulus for the mobilization of MSC into the peripheral blood [58]. This could be interesting in the context of myocardial infarction. After ischemia, the expression of TGF-β is induced in cardiomyocytes. Analysis of the intramyocardial migration of MSC showed that integrin β₁ is responsible for the engraftment of MSC in the ischemic myocardium [59].

Thus, TGF-β1 might be an important factor in MSC homing into the damaged tissue by enhancing the expression of integrins on MSC and thereby improving their attachment to col-1. Furthermore, col-1 seems to be a suitable scaffold material for tissue engineering, integrating well into the tissue and keeping the MSC at the site of injury during wound healing under the influence of TGF-β.

Footnotes

Acknowledgments

This project was supported in part by a DFG grant (Ai 16/19). We thank Anis Larbi, Ph.D. for support with the flow cytometry analyses; Tino Felka, MS for optimization of chondrogenic differentiation protocols; Diane Blaurock, MA for her invaluable help in proofreading of the manuscript; and Bernd Rolauffs, MD for help in preparation of the artwork.

Author Disclosure Statement

No competing financial interests exist.

References

Pittenger

, Mackay

, Beck

, Jaiswal

, Douglas

, Mosca

, Moorman

, Simonetti

, Craig

, Marshak

. 1999. Multilineage potential of adult human mesenchymal stem cells. Science, 284:143–147.

Song

, Song

, Zhang

, Cuevas

, Sanchez-Ramos

. 2007. Comparison of neuron-like cells derived from bone marrow stem cells to those differentiated from adult brain neural stem cells. Stem Cells Dev, 16:747–756.

Makino

, Fukuda

, Miyoshi

, Konishi

, Kodama

, Pan

, Sano

, Takahashi

, Hori

, Abe

, Hata

, Umezawa

, Ogawa

. 1999. Cardiomyocytes can be generated from marrow stromal cells in vitro. J Clin Invest, 103:697–705.

Ferrari

, Cusella-De Angelis

, Coletta

, Paolucci

, Stornaiuolo

, Cossu

, Mavilio

. 1998. Muscle regeneration by bone marrow-derived myogenic progenitors. Science, 279:1528–1530.

Jones

, McGonagle

. 2008. Human bone marrow mesenchymal stem cells in vivo. Rheumatology, 47:126–131.

De Bari

, Dell'Accio

, Vanlauwe

, Eyckmans

, Khan

, Archer

, Jones

, McGonagle

, Mitsiadis

, Pitzalis

, Luyten

. 2006. Mesenchymal multipotency of adult human periosteal cells demonstrated by single-cell lineage analysis. Arthritis Rheum, 54:1209–1221.

da Silva Meirelles

, Chagastelles

, Nardi

. 2006. Mesenchymal stem cells reside in virtually all post-natal organs and tissues. J Cell Sci, 119,Pt 11:2204–2213.

da Silva Meirelles

, Sand

, Harman

, Lennon

, Caplan

. 2008. MSC frequency correlates with blood vessel density in equine adipose tissue. Tissue Eng Part A.

Schwab

, Gargett

. 2007. Co-expression of two perivascular cell markers isolates mesenchymal stem-like cells from human endometrium. Hum Reprod, 22:2903–2911.

10.

Shi

, Gronthos

. 2003. Perivascular niche of postnatal mesenchymal stem cells in human bone marrow and dental pulp. J Bone Miner Res, 18:696–704.

11.

Wilson

, Trumpp

. 2006. Bone-marrow haematopoietic-stem-cell niches. Nat Rev Immunol, 6:93–106.

12.

Moore

, Lemischka

. 2006. Stem cells and their niches. Science, 311:1880–1885.

13.

Chen

, Dusevich

, Feng

, Manolagas

, Jilka

. 2007. Extracellular matrix made by bone marrow cells facilitates expansion of marrow-derived mesenchymal progenitor cells and prevents their differentiation into osteoblasts. J Bone Miner Res, 22:1943–1956.

14.

Caplan

. 2005. Review: mesenchymal stem cells: cell-based reconstructive therapy in orthopedics. Tissue Eng, 11:1198–1211.

15.

Abdallah

, Kassem

. 2008. Human mesenchymal stem cells: from basic biology to clinical applications. Gene Ther, 15:109–116.

16.

Dexter

, Spooncer

, Simmons

, Allen

. 1984. Long-term marrow culture: an overview of techniques and experience. Kroc Found Ser, 18:57–96.

17.

Silva

Jr , Covas

, Panepucci

, Proto-Siqueira

, Siufi

, Zanette

, Santos

, Zago

. 2003. The profile of gene expression of human marrow mesenchymal stem cells. Stem Cells, 21:661–669.

18.

, Kortesmaa

, Tryggvason

, Persson

, Ekblom

, Jacobsen

, Ekblom

. 2003. Laminin isoform-specific promotion of adhesion and migration of human bone marrow progenitor cells. Blood, 101:877–885.

19.

Siler

, Seiffert

, Puch

, Richards

, Torok-Storb

, Müller

, Sorokin

, Klein

. 2000. Characterization and functional analysis of laminin isoforms in human bone marrow. Blood, 96:4194–4203.

20.

Paralkar

, Vukicevic

, Reddi

. 1991. Transforming growth factor beta type 1 binds to collagen IV of basement membrane matrix: implications for development. Dev Biol, 143:303–308.

21.

Hallmann

, Horn

, Selg

, Wendler

, Pausch

, Sorokin

. 2005. Expression and function of laminins in the embryonic and mature vasculature. Physiol Rev, 85:979–1000.

22.

Docheva

, Popov

, Mutschler

, Schieker

. 2007. Human mesenchymal stem cells in contact with their environment: surface characteristics and the integrin system. J Cell Mol Med, 11:21–38.

23.

Young

, Steele

, Bray

, Hudson

, Floyd

, Hawkins

, Thomas

, Austin

, Edwards

, Cuzzourt

, Duenzl

, Lucas

, Black

Jr . 2001. Human reserve pluripotent mesenchymal stem cells are present in the connective tissues of skeletal muscle and dermis derived from fetal, adult, and geriatric donors. Anat Rec, 264:51–62.

24.

Gronthos

, Simmons

, Graves

, Robey

. 2001. Integrin-mediated interactions between human bone marrow stromal precursor cells and the extracellular matrix. Bone, 28:174–181.

25.

Satomura

, Derubeis

, Fedarko

, Ibaraki-O'Connor

, Kuznetsov

, Rowe

, Young

, Gehron Robey

. 1998. Receptor tyrosine kinase expression in human bone marrow stromal cells. J Cell Physiol, 177:426–438.

26.

Kratchmarova

, Blagoev

, Haack-Sorensen

, Kassem

, Mann

. 2005. Mechanism of divergent growth factor effects in mesenchymal stem cell differentiation. Science, 308:1472–1477.

27.

Ramalho-Santos

, Yoon

, Matsuzaki

, Mulligan

, Melton

. 2002. “Stemness”.: transcriptional profiling of embryonic and adult stem cells. Science, 298:597–600.

28.

Tulla

, Pentikäinen

, Viitasalo

, Käpylä

, Impola

, Nykvist

, Nissinen

, Johnson

, Heino

. 2001. Selective binding of collagen subtypes by integrin alpha 1I, alpha 2I, and alpha 10I domains. J Biol Chem, 276:48206–48212.

29.

Schéele

, Nyström

, Durbeej

, Talts

, Ekblom

. 2007. Laminin isoforms in development and disease. J Mol Med, 85:825–836.

30.

Cool

, Nurcombe

. 2005. Substrate induction of osteogenesis from marrow-derived mesenchymal precursors. Stem Cells Dev, 14:632–642.

31.

Dennis

, Haynesworth

, Young

, Caplan

. 1992. Osteogenesis in marrow-derived mesenchymal cell porous ceramic composites transplanted subcutaneously: effect of fibronectin and laminin on cell retention and rate of osteogenic expression. Cell Transplant, 1:23–32.

32.

Broekelmann

, Limper

, Colby

, McDonald

. 1991. Transforming growth factor beta 1 is present at sites of extracellular matrix gene expression in human pulmonary fibrosis. Proc Natl Acad Sci USA, 88:6642–6646.

33.

Geiser

, Zeng

, Sato

, Helvering

, Hirano

, Turner

. 1998. Decreased bone mass and bone elasticity in mice lacking the transforming growth factor-beta1 gene. Bone, 23:87–93.

34.

Kumar

, Sigurdson

, Sheppard

, Lwebuga-Mukasa

. 1995. Differential modulation of integrin receptors and extracellular matrix laminin by transforming growth factor-beta 1 in rat alveolar epithelial cells. Exp Cell Res, 221:385–394.

35.

Heino

, Massagué

. 1989. Transforming growth factor-beta switches the pattern of integrins expressed in MG-63 human osteosarcoma cells and causes a selective loss of cell adhesion to laminin. J Biol Chem, 264:21806–21811.

36.

Loeser

. 1997. Growth factor regulation of chondrocyte integrins. Differential effects of insulin-like growth factor 1 and transforming growth factor beta on alpha 1 beta 1 integrin expression and chondrocyte adhesion to type VI collagen. Arthritis Rheum, 40:270–276.

37.

Binamé

, Lassus

, Hibner

. 2008. Transforming growth factor beta controls the directional migration of hepatocyte cohorts by modulating their adhesion to fibronectin. Mol Biol Cell, 19:945–956.

38.

Warstat

, Hoberg

, Rudert

, Tsui

, Pap

, Angres

, Essl

, Smith

, Cruikshank

, Klein

, Gay

, Aicher

. 2009. TGF-{beta}1 and laminin-111 cooperate in the induction of IL-16 expression in synovial fibroblasts from rheumatoid arthritis patients. Ann Rheum Dis.

39.

Dominici

, Le Blanc

, Mueller

, Slaper-Cortenbach

, Marini

, Krause

, Deans

, Keating

, Prockop

, Horwitz

. 2006. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy, 8:315–317.

40.

Felka

, Schäfer

, Schewe

, Benz

, Aicher

. 2009. Hypoxia reduces the inhibitory effect of IL-1β on chondrogenic differentiation of FCS-free expanded MSC. Osteoarthritis Cartilage, 17:1368–1376.

41.

Drumea-Mirancea

, Wessels

, Müller

, Essl

, Eble

, Tolosa

, Koch

, Reinhardt

, Sixt

, Sorokin

, Stierhof

, Schwarz

, Klein

. 2006. Characterization of a conduit system containing laminin-5 in the human thymus: a potential transport system for small molecules. J Cell Sci, 119,Pt 7:1396–1405.

42.

Barnstable

, Bodmer

, Brown

, Galfre

, Milstein

, Williams

, Ziegler

. 1978. Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens-new tools for genetic analysis. Cell, 14:9–20.

43.

http://www.ncbi.nlm.nih.gov/sites/entrez.

44.

Roche. Light Cycler 2.0 Instrument Operator's Manual.

45.

Lai

, Feng

, Nishimura

, Teitelbaum

, Avioli

, Ross

, Cheng

. 2000. Transforming growth factor-beta up-regulates the beta 5 integrin subunit expression via Sp1 and Smad signaling. J Biol Chem, 275:36400–36406.

46.

Kintscher

, Lyon

, Wakino

, Bruemmer

, Feng

, Goetze

, Graf

, Moustakas

, Staels

, Fleck

, Hsueh

, Law

. 2002. PPARalpha inhibits TGF-beta-induced beta5 integrin transcription in vascular smooth muscle cells by interacting with Smad4. Circ Res, 91:e35–e44.

47.

Yeh

, Chiao

, Kuo

, Hsiao

, Chen

, Wei

, Lai

, Fong

, Tang

. 2008. TGF-beta1 increases motility and alphavbeta3 integrin up-regulation via PI3K, Akt and NF-kappaB-dependent pathway in human chondrosarcoma cells. Biochem Pharmacol, 75:1292–1301.

48.

Heino

, Ignotz

, Hemler

, Crouse

, Massagué

. 1989. Regulation of cell adhesion receptors by transforming growth factor-beta. Concomitant regulation of integrins that share a common beta 1 subunit. J Biol Chem, 264:380–388.

49.

Warstat

, Pap

, Klein

, Gay

, Aicher

. 2008. Co-activation of synovial fibroblasts by laminin-111 and transforming growth factor-beta induces expression of matrix metalloproteinases 3 and 10 independently of nuclear factor-kappaB. Ann Rheum Dis, 67:559–562.

50.

Ramírez

, Lucia

, Gómez-Gallego

, Esteve-Lanao

, Pérez-Martínez

, Foster

, Andreu

, Martin

, Madero

, Arenas

, García-Castro

. 2006. Mobilisation of mesenchymal cells into blood in response to skeletal muscle injury. Br J Sports Med, 40:719–722.

51.

Mansilla

, Marín

, Drago

, Sturla

, Salas

, Gardiner

, Bossi

, Lamonega

, Guzmán

, Nuñez

, Gil

, Piccinelli

, Ibar

, Soratti

. 2006. Bloodstream cells phenotypically identical to human mesenchymal bone marrow stem cells circulate in large amounts under the influence of acute large skin damage: new evidence for their use in regenerative medicine. Transplant Proc, 38:967–969.

52.

Andrades

, Han

, Becerra

, Sorgente

, Hall

, Nimni

. 1999. A recombinant human TGF-beta1 fusion protein with collagen-binding domain promotes migration, growth, and differentiation of bone marrow mesenchymal cells. Exp Cell Res, 250:485–498.

53.

Ozaki

, Nishimura

, Sekiya

, Suehiro

, Kanawa

, Nikawa

, Hamada

, Kato

. 2007. Comprehensive analysis of chemotactic factors for bone marrow mesenchymal stem cells. Stem Cells Dev, 16:119–129.

54.

Fiedler

, Röderer

, Günther

, Brenner

. 2002. BMP-2, BMP-4, and PDGF-bb stimulate chemotactic migration of primary human mesenchymal progenitor cells. J Cell Biochem, 87:305–312.

55.

Kirchberg

, Lange

, Klein

, Jungtäubl

, Heinen

, Meyer-Ingold

, Scharffetter-Kochanek

. 1995. Induction of beta 1 integrin synthesis by recombinant platelet-derived growth factor (PDGF-AB) correlates with an enhanced migratory response of human dermal fibroblasts to various extracellular matrix proteins. Exp Cell Res, 220:29–35.

56.

Varas

, Ohlsson

, Honeth

, Olsson

, Bengtsson

, Wiberg

, Bockermann

, Järnum

, Richter

, Pennington

, Johnstone

, Lundgren-Akerlund

, Kjellman

. 2007. Alpha10 integrin expression is up-regulated on fibroblast growth factor-2-treated mesenchymal stem cells with improved chondrogenic differentiation potential. Stem Cells Dev, 16:965–978.

57.

Schmidt

, Ladage

, Schinköthe

, Klausmann

, Ulrichs

, Klinz

, Brixius

, Arnhold

, Desai

, Mehlhorn

, Schwinger

, Staib

, Addicks

, Bloch

. 2006. Basic fibroblast growth factor controls migration in human mesenchymal stem cells. Stem Cells, 24:1750–1758.

58.

Rochefort

, Delorme

, Lopez

, Hérault

, Bonnet

, Charbord

, Eder

, Domenech

. 2006. Multipotential mesenchymal stem cells are mobilized into peripheral blood by hypoxia. Stem Cells, 24:2202–2208.

59.

, Wu

, Huang

, Zhang

, Pratt

, Dzau

. 2007. Mesenchymal stem cells use integrin beta1 not CXC chemokine receptor 4 for myocardial migration and engraftment. Mol Biol Cell, 18:2873–2882.