SEARCH

SEARCH BY CITATION

Keywords:

  • Laminin-5/laminin-332;
  • Mesenchymal stem cells;
  • Cell growth;
  • Chondrogenesis;
  • Osteogenesis

Abstract

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

Laminin-5 (laminin-332) is an important basement membrane protein that regulates cell attachment and motility. Recent studies have shown that laminin-5 is expressed in human mesenchymal stem cells (MSCs) in culture and that the laminin γ2 chain is transiently expressed in chondrocytes during development. These studies suggest that laminin-5 may be involved in the regulation of chondrogenic differentiation of MSCs. In this study, we examined a possible role of laminin-5 in the proliferation and differentiation of human MSCs. When MSCs were incubated in the presence of a coated or soluble form of laminin-5 in a growth medium, they proliferated more rapidly than nontreated cells, keeping their differentiation potential. On the other hand, laminin-5 potently suppressed the chondrogenic differentiation of MSCs. These activities were mediated mainly by integrin α3β1. However, laminin-5 had no effect on the osteogenic differentiation of MSCs. These results suggest that laminin-5 may contribute to the development of bone tissues by promoting the proliferation and by suppressing the chondrogenic differentiation of MSCs.


Introduction

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

Mesenchymal stem cells (MSCs) are pluripotent cells that can differentiate into osteoblasts and chondrocytes as well as adipocytes, hepatocytes, neural cells, muscle cells, and endothelial cells under appropriate conditions [1, 2]. Soluble factors such as ascorbic acid, dexamethasone, insulin-like growth factor-I, and transforming growth factor-β (TGF-β) are well-known as regulators of the MSC differentiation [1, [2]3]. A recent study has demonstrated that flat-shaped MSCs are induced to osteogenesis whereas round-shaped cells are induced to adipogenesis, indicating that the cell shape of MSCs is associated with their cell lineage commitment [4]. This suggests that extracellular matrix (ECM) proteins, which regulate actin cytoskeleton and hence cell structure through interaction with integrins, are important regulators of MSC differentiation. Indeed, some ECM proteins have been identified as the inducers of osteogenic differentiation [5].

Laminins are a family of ECM proteins that are localized mainly in basement membranes and regulate various cellular functions such as adhesion, motility, proliferation, apoptosis, and differentiation through interaction with specific cell surface receptors, integrins [6, 7]. The three subunits of laminins—α, β, and γ—form the laminin-specific cross-shaped structure linked together by disulfide bonds. Five α, three β, and three γ chains and at least 15 laminin isoforms consisting of different combinations of the three chains have been identified thus far [6]. Laminin-5, abbreviated here as Lm5, consists of α3, β3, γ2 chains and hence is named laminin-332 in a new nomenclature [8]. Lm5 has unique structural and biological properties. All of the three chains have N-terminally truncated structures as compared with laminin α1, β1, and γ1 chains, and the β3 and γ2 chains are found only in Lm5. Lm5 supports cell adhesion and migration more efficiently than other laminins [9, 10], and these activities are mediated mainly by integrin α3β1, α6β1, and α6β4 [11, 12]. Lm5 is an important component in the epidermal basement membrane of the skin and is also expressed in many other epithelial tissues (e.g., esophagus, lung, intestine, and kidney) [13]. In the skin, Lm5 plays essential roles in the stable epidermis and dermis connection [14, 15] and wound healing [16, 17].

We previously found that the γ2 chain of Lm5 is transiently expressed in embryonic cartilage [18] and that Lm5 has suppressive activity on chondrogenic differentiation of the mouse embryonal carcinoma-derived chondroprogenitor cell line ATDC5 [19]. In addition, another group reported that Lm5 is expressed in rat periosteum and human MSCs and induces osteogenic differentiation of MSCs [20]. These results suggest that Lm5 is a possible regulator for the differentiation of MSCs during bone development. In recent years, MSCs have attracted much attention as one of the most important stem cell types for regenerative medicine because of their pluripotency and capacity for self-renewal. In this study, we examined possible roles of Lm5 in the regulation of proliferation and chondrogenic and osteogenic differentiation of MSCs.

Materials and Methods

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

Laminins and Antibodies

Human recombinant Lm5 was purified as described previously [21]. This Lm5 preparation was composed mostly of the Lm5 with the proteolytically processed γ2 chain. In addition to the Lm5, an Lm5 variant with the unprocessed γ2 chain was prepared as described previously [19] and used in some experiments. Other laminins were used, and their sources were mouse laminin-1 (laminin-111) from Invitrogen (Carlsbad, CA, http://www.invitrogen.com), human laminin-2/4 (laminin-211/221) from Chemicon International (Temecula, CA, http://www.chemicon.com), and laminin-10/11 (laminin-511/521) from Sigma-Aldrich (St. Louis, http://www.sigmaaldrich.com). Function-blocking, anti-integrin antibodies used were the anti-α3-integrin antibody (P1B5), the anti-β1-integrin antibody (6S6) and the anti-β4-integrin antibody (3E1) from Chemicon International, and the anti-α6-integrin antibody (GoH3) from BD PharMingen (San Diego, http://www.bdbiosciences.com/pharmingen). Other antibodies used were the anti-osteopontin antibody from IBL (Gunma, Japan, http://www.ibl-japan.co.jp) and the anti-type II collagen antibody from Lab Vision Corporation (Fremont, CA, http://www.labvision.com).

Differentiation of MSCs

Human MSCs were purchased from Cambrex Bio Science Walkersville, Inc. (Walkersville, MD, http://www.cambrex.com) and maintained in MSC growth medium (maintenance medium; Cambrex Bio Science Walkersville, Inc.). Cultures were incubated at 37°C in a humidified atmosphere containing 95% air and 5% CO2.

To induce chondrogenesis, 2.5 × 105 cells were suspended in a chondrogenic medium containing ascorbate, dexamethasone, insulin, transferrin, sodium selenite, sodium pyruvate, proline, and TGF-β3 (Cambrex Bio Science Walkersville, Inc.) and centrifuged in a polypropylene tube at 150g for 5 minutes. The pellet was kept for 24 hours to produce a floating cell aggregate. This cell aggregate was kept floating with a medium change every second day. After 3 weeks, the aggregate was fixed in 10% formalin, embedded into paraffin, and cut into sections. The paraffin sections were stained for glycosaminoglycans with Alcian Blue (Sigma-Aldrich) or subjected to immunostaining as described below. For the Alcian Blue staining, the sections were soaked in 0.1% Alcian Blue in 0.1 N HCl for 2 hours at room temperature and washed three times with distilled water.

To induce osteogenesis, MSCs were inoculated at 3 × 103 cells per cm2 in an osteogenic medium containing ascorbate, dexamethasone, and β-glycerophosphate (Cambrex Bio Science Walkersville, Inc.) and incubated for 3 weeks with a medium change every third day. To assess the osteogenic differentiation, we measured alkaline phosphatase activity. For the assay, MSCs were washed with phosphate-buffered saline (PBS) and stained with 0.1 mg/ml naphthol ASBI (Sigma-Aldrich) plus 0.6 mg/ml Fast Red TR (Sigma-Aldrich) in 0.05 M Tris-HCl (pH 8.5) at 37°C for 30 minutes. After washing three times with PBS, cells were fixed in 10% formalin for 10 minutes. Osteopontin and osteocalcin in the cultures were detected as described below.

To examine the effects of Lm5 and other laminins in insoluble forms on the osteogenic differentiation of MSCs, 24-well plates (Sumibe Medical, Tokyo, http://www.sumibe.co.jp/sumilon) were coated with various concentrations of these substrate proteins at 4°C overnight and then treated with the PBS containing 1.2% (wt/vol) bovine serum albumin (BSA) at 37°C for 1.5 hours for blocking. MSCs were inoculated onto these coated plates and incubated in the differentiation medium. In control cultures, plates were coated with BSA alone. For the assay of Lm5 in a soluble form, MSCs were inoculated at the same density onto noncoated 24-well plates in the medium supplemented with purified Lm5 at indicated concentrations. In the case of chondrogenic differentiation, Lm5 was added into the differentiation medium after floating cell aggregates were made. In control cultures, the medium was supplemented with the vehicle alone. The vehicle, in which Lm5 was dissolved, was composed of 60 mM HEPES-NaOH (pH 7.3), 0.05% (vol/vol) trifluoroacetic acid, 0.005% Brij35, and 0.1% CHAPS.

Cell Adhesion Assay

Cell adhesion was assayed as described previously [22]. MSCs were harvested by trypsinization and washed twice with a serum-free medium, and 0.2-ml aliquots of MSC suspension (2 × 105 cells per ml) were inoculated per well of 96-well plates (Sumibe Medical) precoated with substrate proteins to be tested. After incubation for 5 minutes, unattached cells were removed and attached cells were stained with Hoechst 33342. Fluorescent intensity of each well was measured using a CytoFluor 2350 fluorometer (Millipore, Bedford, MA, http://www.millipore.com). For inhibition assay, MSCs were treated with function-blocking, anti-integrin antibodies for 5 minutes at room temperature before inoculation.

Cell Proliferation Assay

MSCs were inoculated on 24-well plates at a density of 3 × 103 cells per cm2 in the maintenance medium, and the cell number was determined every 4 days. To assess the effect of laminins, these proteins were individually precoated on the plates or directly added into the culture medium. In some experiments, a basal serum-free medium supplemented with 5% PANEXIN (PAN biotech GmbH, Aidenbach, Germany, http://www.pan-biotech.de) was used instead of the maintenance medium.

Immunohistochemistry

Paraffin sections of cell aggregates were treated with 0.4% pepsin in 0.01 N HCl for 15 minutes and then with 0.3% H2O2 in methanol for 30 minutes. The resultant sections were blocked with 5% (wt/vol) skim milk for 2 hours and incubated with 1 μg/ml of the anti-human-type II collagen monoclonal antibody for 24 hours at 4°C. Immunoreactivity was detected by incubating the sections with a biotinylated, goat anti-mouse-IgG antibody as the second antibody and then with streptavidin-conjugated horseradish peroxidase. Immunoreactive signals were developed in a reaction mixture containing DAB (3,3′-diaminobenzidine) and H2O2.

Analysis of Osteopontin by Immunoblotting

To detect osteopontin as an ostogenic differentiation marker by immunoblotting, cells on culture plates were lysed in 20 mM Tris-HCl (pH 7.5) buffer containing 150 mM NaCl, 1 mM EDTA, 1% Triton-X-100, and the protease inhibitor cocktail set III (EMD Biosciences, San Diego, http://www.emdbiosciences.com/html/CBC/home.html), and the cell lysates were discarded. After washing the plates three times with PBS, ECM proteins remaining on the plates were dissolved in the SDS sample buffer using a scraper and applied to SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions on a 10% gel. The separated proteins were transferred onto polyvinylidene difluoride membranes (Millipore), and osteopontin was detected with the anti-osteopontin antibody by the enhanced chemiluminescence detection method (GE Healthcare, Little Chalfont, Buckinghamshire, UK, http://www.gehealthcare.com).

Reverse Transcription-Polymerase Chain Reaction Analysis

Total RNAs were extracted with the TRIzol reagent (Invitrogen) from MSCs. cDNAs were prepared from 5 μg of total RNAs using a reverse transcription-polymerase chain reaction analysis (RT-PCR) kit (Toyobo, Osaka, Japan, http://www.toyobo.co.jp) according to the manufacturer's protocol, and the cDNAs were amplified using specific primers for osteocalcin and glyceraldehyde-3-phosphate dehydrogenase (Table 1). The PCR conditions employed were 1 minute at 95°C, 1 minute at 58°C, and 1 minute at 72°C for 30 cycles. Aliquots of the PCR products were electrophoresed on a 1% agarose gel in Tris-borate-EDTA buffer and stained with ethidium bromide.

Table Table 1.. Polymerase chain reaction primers used for detection of osteocalcin and GAPDH messages
Thumbnail image of

Results

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

Effect of Lm5 on Adhesion of MSCs

Lm5 is a large heterotrimeric protein consisting of laminin α3, β3, and γ2 chains. Recently, Klees et al. showed that MSCs express these chains of Lm5 [20]. To show the biological activity of Lm5 on MSCs, we first compared effects of Lm5 and three other laminins (laminin-1, laminin-2/4, and laminin-10/11) on the adhesion and spreading of MSCs. When each laminin was coated at a concentration of 1 μg/ml for Lm5 or 2 μg/ml for the other laminins on culture plates, Lm5 most efficiently promoted the attachment of MSCs to the plates even after a 5-minute incubation (Fig. 1A). The adhesion to the Lm5 substrate was reduced to a control level when the cells were pretreated with a function-blocking antibody to α3 or β1 integrin (Fig. 1B), suggesting that this adhesion was mediated mostly by integrin α3β1, which is consistent with the previous report [20]. After incubation for 10 minutes, MSCs became spread well on the Lm5 on laminin-10/11 substrates but not on the laminin-1 or laminin-2/4 substrate (Fig. 1C). These results indicate that Lm5 is the most effective substrate for the attachment and spreading of MSCs among the four laminins tested. We confirmed that there is no significant difference in the coating efficiency to plastic plates between Lm5 and laminin-10/11 (data not shown).

thumbnail image

Figure Figure 1.. Effects of four kinds of laminins on attachment and spreading of mesenchymal stem cells (MSCs). (A): Cell attachment. Each well of 96-well plates was coated with 2 μg/ml of laminin-1 (Lm1), laminin-2/4 (Lm2/4) or laminin-10/11 (Lm10/11), or 1 μg/ml of laminin-5 (Lm5) and then blocked with bovine serum albumin (BSA). Control wells (Cont.) were coated with BSA alone. MSCs were grown in the maintenance medium and harvested by trypsinization. The harvested MSCs were washed twice with, and then suspended in, a serum-free medium and then inoculated into each well of the plates. After incubation for 5 minutes, unattached cells were removed and attached cells were stained with Hoechst 33342. The fluorescent intensity of attached cells was measured. Each bar indicates the mean ± SD for triplicate assays. (B): Effects of function-blocking, anti-integrin antibodies on cell attachment to Lm5. MSCs were pretreated without (None) or with nonimmunized mouse immunoglobulin G (IgG), the anti-α3-integrin antibody (α3), the anti-α6-integrin antibody (α6), the anti-β1-integrin antibody (β1), or the anti-β4-integrin antibody (β4) and then incubated for 5 minutes on 96-well plates coated with 1 μg/ml Lm5 (Lm5) or with BSA alone (Cont.). Cell attachment was assayed as described above. (C): Cell-spreading activity. MSCs were incubated for 10 minutes on plastic plates precoated with the indicated laminins, and the cell morphology was examined under a phase-contrast microscope. Other experimental conditions are described in Materials and Methods.

Download figure to PowerPoint

Growth-Stimulating Activity of Lm5

Lm5 not only functions as an effective cell adhesion substrate but also stimulates proliferation of some cell lines [23, 24]. Therefore, we next examined the effect of Lm5 on the growth of MSCs. When Lm5 was coated at three different concentrations on plastic plates, it dose-dependently promoted the growth of MSCs in the maintenance medium (Fig. 2A, left panel). However, this growth-stimulating effect was scarcely observed on the cells in an osteogenic medium, suggesting that Lm5 promotes proliferation of only noncommitted MSCs (Fig. 2A, right panel). The growth-promoting activity of Lm5 in the maintenance medium was also obtained when Lm5 was directly added into the culture medium (Fig. 2B). These results suggest that Lm5 also functions as a soluble growth factor toward MSCs in the maintenance medium. Given that terminal deoxynucleotidyl transferase dUTP nick-end labeling assay revealed no difference between nontreated and Lm5-treated cells (data not shown), Lm5 seems to stimulate cell growth rather than inhibit apoptosis. To show whether the growth-promoting activity is specific to Lm5, we compared growth-promoting activities of four kinds of laminins toward MSCs, using the laminin-coated plates. Lm5 showed significantly higher growth activity than did laminin-10/11, laminin-1, and laminin-2/4 (Fig. 2C). Both laminin-10/11 and laminin-1, but not laminin-2/4, slightly promoted the MSCs growth.

thumbnail image

Figure Figure 2.. Effect of laminin-5 (Lm5) on proliferation of mesenchymal stem cells (MSCs). (A): Effect of three different concentrations of Lm5 in two types of culture media. Each well of 24-well plates was coated without (open circles) or with 1 (closed circles), 0.5 (closed triangles), or 0.25 (closed squares) μg/ml Lm5. MSCs were inoculated on the plates at a density of 5 × 103 cells per well in a maintenance medium (Mt. Med.; left panel) and an osteogenic medium (Ost. Med.; right panel). After incubation for the indicated periods, the number of cells was determined. Each bar indicates the mean ± SD for triplicate assays. (B): Growth stimulation by insoluble and soluble forms of Lm5. MSCs were inoculated on a 24-well plate precoated with 1.0 μg/ml Lm5 in a maintenance medium (closed circles) or on a noncoated plate in the medium with (closed triangles) or without (open circles) 0.5 μg/ml Lm5. The cell growth was determined as described above. (C): Growth-stimulating activity of four kinds of laminins. Each well of 24-well plates was coated with 2 μg/ml of laminin-1 (closed triangles), laminin-2/4 (closed squares), or laminin-10/11 (open triangles) or with 1 μg/ml Lm5 (closed circles) and then blocked with bovine serum albumin (BSA). Control wells (open circles) were coated with BSA alone. The cell growth was determined as described above. (D): Effects of function-blocking, anti-integrin antibodies on Lm5-stimulated growth of MSCs. MSCs were pretreated without (None) or with nonimmunized mouse immunoglobulin G (IgG), anti-α3-integrin antibody (α3), anti-α6-integrin antibody (α6), or both anti-α3-integrin and anti-α6-integrin antibodies (α3+α6). These cells were incubated on 24-well culture plates precoated with 1 μg/ml Lm5 for 6 days, and the number of cells was determined as described above. As control, nontreated MSCs were incubated on a noncoated plate (Control). (E): Growth stimulation by Lm5 in a serum-free medium. MSCs were incubated on 24-well culture plates precoated with or without 1 μg/ml Lm5 in a serum-free, 5% Panexin-containing medium with or without 1 ng/ml basic fibroblast growth factor (bFGF) or in the serum-containing, maintenance medium: Panexin alone/noncoated (open circles), Panexin+bFGF/noncoated (open triangles), serum/noncoated (open squares), Panexin alone/Lm5-coated (closed circles), Panexin+bFGF/Lm5-coated (closed triangles). The cell growth was determined as described above. Other experimental conditions are described in Figure 1 and Materials and Methods.

Download figure to PowerPoint

As shown in Figure 1, Lm5 promotes the adhesion of MSCs through integrin α3β1. We also tested whether the growth activity of Lm5 toward MSCs is mediated by integrin α3β1, using the same function-blocking, anti-integrin antibodies (Fig. 2D). When the anti-integrin antibodies against integrin α3 or α6 were added separately into the MSC culture, each antibody partly inhibited the growth activity of Lm5. When both antibodies were added into the same culture, additional growth inhibition was obtained compared with the inhibition with either antibody. This suggests that Lm5 promotes the proliferation of MSCs through both integrins α3β1 and α6β1.

Basic fibroblast growth factor (bFGF) is known to stimulate the growth of MSCs [25]. Next, we examined the synergistic growth effect of Lm5 and bFGF in a serum-free medium. When MSCs were incubated on Lm5-coated plates in a serum-free defined medium supplemented with 10 μg/ml insulin, 10 μg/ml transferrin, 3 × 10−8 M sodium selenite, and 1 ng/ml bFGF, they could not proliferate appreciably (data not shown). Next, we used a commercial serum-free medium, PANEXIN (PAN biotech GmbH), which contains a relatively high concentration of unspecified proteins. Addition of bFGF into the PANEXIN medium significantly supported the growth of MSCs (Fig. 2E). When MSCs were incubated in the bFGF-PANEXIN medium on Lm5-coated plates, their growth was further improved and became comparable with that in the serum-containing maintenance medium, especially at an earlier growth phase. These results demonstrate that Lm5 and bFGF synergistically promote the proliferation of MSCs in a serum-free medium.

Effect of Lm5 on Differentiation Potential of MSCs

Because MSCs have pluripotency for differentiation, it seems important to show whether Lm5 affects their differentiation. First, we examined whether MSCs that had been cultured in the presence of Lm5 retain the capacity to differentiate to chondrocytes or osteoblasts. After cultivation on the Lm5 substrate in the maintenance medium for 8 days, MSCs were replated and induced for chondrogenic or osteogenic differentiation in the absence of Lm5. When MSCs that had been treated with Lm5 were cultured for 3 weeks in the chondrogenic medium, these cells expressed the typical chondrogenic markers, proteoglycans and type II collagen, at a level similar to nontreated cells (Fig. 3A, 3B). Similarly, when MSCs stimulated by Lm5 were induced for osteogenic differentiation, these cells, as well as nontreated cells, highly expressed the generally accepted osteogenic markers, alkaline phosphatase, osteopontin, and osteocalcin (Fig. 3C–3E). These results suggest that MSCs grown in the presence of Lm5 retain their pluripotency.

thumbnail image

Figure Figure 3.. Effect of Lm5 on differentiation potential of mesenchymal stem cells (MSCs). MSCs were cultured on a culture plate precoated without (Control) or with (Lm5) 1 μg/ml Lm5 in the maintenance medium for 8 days. The cells were harvested and then induced for chondrogenic differentiation (A, B) or osteogenic differentiation (C–E) for 3 weeks, as described in Materials and Methods. The degree of differentiation was determined by analyzing expression of the following differentiation markers. (A): After the incubation in the chondrogenic differentiation medium, the aggregates of MSCs were stained for glycosaminoglycans with Alcian blue. Magnifications: ×40 (upper panels), ×400 (lower panels). (B): Expression of type II collagen was detected by immunohistochemistry with the anti-type II collagen antibody. (C): After the incubation in the osteogenic differentiation medium, the alkaline phosphatase activity of the MSC cultures was detected by the method described in Materials and Methods. (D): Expression of OPN was detected by immunoblotting with the anti-OPN antibody. (E): Expression of osteocalcin (OC) mRNA was analyzed by reverse transcription-polymerase chain reaction. The GAPDH message was analyzed as a loading control. Other experimental conditions are described in Materials and Methods. Abbreviations: GAPDH, glyceraldehyde-3-phosphate dehydrogenase; LM5, laminin-5; CPN, osteopontin.

Download figure to PowerPoint

Suppression of Chondrogenic Differentiation of MSCs by Lm5

Our previous study showed that Lm5 suppresses chondrogenic differentiation of the chondroprogenitor cell line ATDC5 [19]. This activity of Lm5 was examined for the chondrogenic differentiation of MSCs. When aggregates of MSCs were incubated in plastic tubes containing the differentiation medium, the size of the cell aggregates increased markedly (Fig. 4A and 4B, upper panel). Given that MSCs are known to scarcely proliferate in this chondrogenic medium, the expansion of the cell aggregates seemed to result from the accumulation of cartilage-specific ECM [26]. Indeed, the staining of paraffin sections of the aggregates with Alcian Blue revealed accumulation of proteoglycans in the ECM (Fig. 4A). When Lm5 was added into the chondrogenic differentiation medium, it strongly inhibited both the expansion of the MSC aggregates (Fig. 4A, upper panel) and the accumulation of proteoglycans in the ECM (Fig. 4A, lower panel).

thumbnail image

Figure Figure 4.. Suppression of chondrogenic differentiation of mesenchymal stem cells (MSCs) by laminin-5 (Lm5) and effect of function-blocking, anti-integrin antibodies on the Lm5 activity. The aggregates of MSCs were incubated for 3 weeks in the chondrogenic differentiation medium supplemented without (Control) or with 1 μg/ml Lm5 (Lm5), Lm5 plus anti-integrin-α3 antibody (Lm5 + anti-int-α3), or Lm5 plus anti-integrin-α6 antibody (Lm5 + anti-int-α6). As negative control, the aggregates were incubated in the differentiation medium without transforming growth factor-β3. (A): After incubation in the chondrogenic differentiation medium, the aggregates of MSCs were stained for glycosaminoglycans with Alcian blue. (B): The aggregates of MSCs were immunostained with the anti-type II collagen antibody. Other experimental conditions are described in Materials and Methods. Abbreviation: TGF, transforming growth factor.

Download figure to PowerPoint

We also examined which integrin is responsible for the inhibition of the chondrogenic differentiation by Lm5. When a function-blocking antibody to α3 or α6 integrin was added into the differentiation medium containing Lm5, the anti-α3-integrin antibody, but not the anti-α6-integrin antibody, partially recovered the proteoglycan accumulation suppressed by Lm5 (Fig. 4A).

Essentially the same results were obtained when type II collagen in the cell aggregates was analyzed by immunostaining (Fig. 4B). These results demonstrate that Lm5 suppresses the chondrogenic differentiation of MSCs through binding to integrin α3β1.

Effect of Lm5 on Osteogenic Differentiation

Recently, Klees et al. reported that Lm5 induces osteogenic differentiation of MSCs without soluble osteogenic supplements [20]. Therefore, we examined whether Lm5 actually induces osteogenesis in our experimental conditions. For this analysis, MSCs were incubated in a maintenance or differentiation medium for 3 weeks on culture plates that had been coated with 0.25–1.0 μg/ml of recombinant Lm5. These cultures were then analyzed for alkaline phosphatase activity as the osteogenic marker. Unexpectedly, MSCs cultured on Lm5 in the maintenance medium scarcely expressed alkaline phosphatase (Fig. 5A, upper panel). In the differentiation medium, MSCs highly expressed alkaline phosphatase activity regardless of the presence or absence of Lm5 (Fig. 5A, lower panel). When soluble Lm5 was added into the culture medium, almost the same results were obtained as in the case of insoluble Lm5 (data not shown).

thumbnail image

Figure Figure 5.. Effect of Lm5 on osteogenic differentiation of mesenchymal stem cells (MSCs). MSCs were inoculated in either the maintenance medium (Mt. Med.) or the osteogenic differentiation medium (Ost. Med.) on 24-well plates coated with the indicated concentrations of Lm5 and incubated for 3 weeks. The resultant cultures were analyzed for the expression of the following osteogenic markers. (A): MSCs were incubated on the plates precoated with 0.25–1.0 μg/ml of Lm5, and alkaline phosphatase activity was stained. (B): MSCs were incubated on the plates precoated with (Lm5) or without (None) 1 μg/ml Lm5. Expression of osteopontin (OPN) was detected by immunoblotting with the anti-OPN antibody. (C): Expression of osteocalcin mRNA in the cultures used in (B) was analyzed by reverse transcription-polymerase chain reaction. (D): The activity of the Lm5 with the unprocessed γ2 chain (Up-Lm5) was compared with the Lm5 with the processed γ2 chain (Pr-Lm5). MSCs were incubated on 5 μg/ml each Lm5, and OPN was detected by immunoblotting as described in (B). Other experimental conditions are described in Figure 3 and Materials and Methods. Abbreviation: GAPDH, glyceraldehyde-3-phosphate dehydrogenase; LM5, laminin-5 OC, osteocalcin.

Download figure to PowerPoint

To confirm these results, expression of two other osteogenic markers, osteopontin and osteocalcin, was analyzed by immunoblotting and RT-PCR, respectively. When MSCs were cultured on the Lm5 substrate in the maintenance or differentiation medium, neither osteopontin (Fig. 5B) nor osteocalcin (Fig. 5C) was induced by Lm5, in agreement with the results of alkaline phosphatase staining. Similar results were reproduced even when 10 times the concentration of Lm5 (10 μg/ml) was used or when a different lot of MSCs was used (data not shown).

It has been reported that the activity of Lm5 is modulated by proteolytic processing of the γ2 chain [19, 27, 28]. The human recombinant Lm5 used in this study was mostly the processed form. Therefore, we also examined the effect of the Lm5 with the unprocessed γ2 chain on the osteogenic differentiation. However, the unprocessed Lm5, like the processed Lm5, did not induce the expression of alkaline phosphatase (data not shown) or osteopontin (Fig. 5D) at 0.5 μg/ml or even at 5 μg/ml. These results clearly showed that our Lm5 preparations, regardless of the processed or unprocessed form, had little effect on the osteogenic differentiation of MSCs.

Discussion

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

In developmental and adult tissues, various factors regulate stem cells to maintain their potentials of self-renewal and specific differentiation [29]. In vitro, some protein factors, such as bFGF [25] and heparin-binding epidermal growth factor-like growth factor (HB-EGF) [30], stimulate the growth of MSCs. Obviously, some ECM molecules also contribute to the regulation of MSC functions. A recent study showed that an ECM produced by murine parietal yolk sac cells increases the growth rate and proliferative life span of human MSCs, although the active component of the matrix was not identified [31]. It was also reported that Lm5 is expressed in human MSCs and the periosteum of rat rib [20]. It is well-known that cell adhesion molecules cooperate with growth factors to induce intracellular growth signaling by interacting with different cell surface receptors [32]. In the present study, Lm5 stimulated the growth of MSCs synergistically with bFGF or serum factors (Fig. 2). Although Lm5 strongly blocked the chondrogenic differentiation of MSCs, their cultivation in the growth medium on the Lm5 substrate did not abolish their potential to differentiate to chondrocytes and osteoblasts. MSCs that had been cultured on Lm5 were able to undergo chondrogenic or osteogenic differentiation after specific stimulation (Fig. 3). We previously found that Lm5 and laminin-5B (laminin-3B32) have growth-stimulating activity, depending on the type of target cells [22]. Indeed, Lm5 does not stimulate the growth of ATDC5 cells [19]. In the present study, Lm5 stimulated the growth of MSCs in the maintenance medium, but it did not in the osteogenic differentiation medium. Our present findings together with the past findings suggest that Lm5 may be a specific regulator of the undifferentiated MSC to maintain the stem cell function of MSCs in vivo. It seems likely that Lm5 supports the self-renewal of MSCs, keeping their differentiation potential but preventing their chondrogenic differentiation even if they are stimulated by chondrogenic factors. Lm5 may also contribute to the specific localization of MSCs by its strong cell adhesion activity.

It is noteworthy that, among four types of laminins tested here, only Lm5 had prominent effects on the growth and adhesion of human MSCs. Many laminin isoforms commonly recognize integrins α3β1 and α6β1 and, in some cases, integrin α6β4 [6]. However, their activity significantly differs from one laminin isoform to another. Such differences are thought to come from differences in their structures and in affinity to integrins [6, 8, 32]. Lm5 is especially unique in its structure and biological activity [33]. For example, all three chains of Lm5 have N-terminally truncated structures as compared with typical laminin chains (e.g., α1, β1, and γ1 chains) [8]. The α3 and γ2 chains of Lm5 undergo specific proteolytic processing after secretion, changing its biological activity [27, 28, 34, 35]. Furthermore, Lm5 induces cell migration in a soluble form, but other laminins do not [36]. Indeed, the soluble form of Lm5 stimulated the growth of MSCs and inhibited the chondrogenic differentiation in this study. The differential biological activity of Lm5 toward MSCs, relative to other laminins, is likely attributable to its unique structure and proteolytic processing.

A recent study by Klees et al. has shown that Lm5 induces osteogenic differentiation of MSCs in vitro [20]. In the present study, however, we could not reproduce the induction of osteogenic differentiation by Lm5 (Fig. 5). The discrepancy between the two studies may be derived from a difference in the Lm5 preparations used, because other experimental conditions are similar to each other. The biological activity of Lm5 is modulated by the proteolytic processing of the short arm of the γ2 chain [19, 27, 28]. Although we tested the effects of both the Lm5 with the processed γ2 chain and one with unprocessed γ2 chain, neither type induced the osteogenic differentiation (Fig. 5D). Klees et al. used their Lm5 at a very high concentration (20 μg/ml) [20]. The origin and specific activity of the Lm5 were not described in their report. Our human recombinant Lm5 was purified by affinity chromatography with a specific antibody and scarcely contained contaminating proteins [21]. It showed maximal cell adhesion activity and growth-stimulating activity toward MSCs at 1 μg/ml or lower (Figs. 1A, 2A), but it had no significant effect on the osteogenic differentiation even at 10 μg/ml. At present, we speculate that an additional factor may be required for the stimulation of osteogenic differentiation by Lm5.

Klees et al. have shown that Lm5 is expressed by MSCs in vitro and deposited in the periosteum in vivo [20]. However, they failed to detect the deposition of Lm5 on the culture plates of MSCs, suggesting that the secretion and deposition of Lm5 are very low in the MSC cultures. We were also unable to detect Lm5 on the culture plates of MSCs (data not shown). Therefore, the effect of the endogenous Lm5 seems to be very low or negligible, if any, at least in the MSC cultures. Some microenvironmental factors may influence the production and deposition of Lm5 by MSCs.

In the process of bone development, the anlage of bone is first made with the cartilage and gradually replaced by the bone matrix [37]. This means that chondrogenesis determines the normal patterning of bone structure. Therefore, the strict regulation of chondrogenic differentiation by various factors is indispensable for the normal bone development [38, 39]. In this process, the transient suppression of chondrogenic differentiation may be critical for MSCs or chondroprogenitor cells to proliferate enough and construct the proportioned bone structure. Alternatively, the suppression of chondrogenesis by Lm5 may favor commitment of MSCs to other cell lineages (e.g., osteoblasts, adipocytes, myoblasts, or tenocytes). In any case, Lm5 seems to contribute to the normal bone development by the growth stimulation and the suppression of chondrogenesis of MSCs. Detailed analysis of the new function of Lm5 seems to contribute to understanding for the mechanism of bone development.

MSCs can be easily obtained by bone marrow and grown in culture while keeping their potential of multilineage differentiation. Therefore, much attention has been focused on the use of MSCs for regenerative medicine (e.g., osteoporosis, arthritis, periodontal disease, cardiac disease, muscular dystrophies, and degenerative nerve diseases) [40, [41]42]. For clinical use, it is essential to grow patient-derived MSCs rapidly in culture. The activity of Lm5, which synergistically stimulates MSC growth with bFGF or serum factors, may be useful for the future use of MSCs in regenerative medicine.

Conclusion

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

In the present study, we found that the epithelial basement membrane protein Lm5 potently stimulated the growth of human MSCs and suppressed their chondrogenic differentiation in culture. Furthermore, Lm5 strongly promoted the attachment and spreading of MSCs to culture plates. These activities were mediated mainly by integrin α3β1, and additionally by integrin α6β1 in the case of the growth stimulation. These results suggest that Lm5 may contribute to the development of bone tissues by regulating the proliferation and differentiation of MSCs. The unique activities of Lm5 also suggest that it may be useful for efficient isolation and growth of human MSCs.

Acknowledgements

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

We thank Drs. H. Yasumitsu, S. Higashi, T. Ogawa, and K. Yamamoto for helpful suggestions and discussions. This work was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

References

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