Collagen Type II Enhances Chondrogenesis in Adipose Tissue–Derived Stem Cells by Affecting Cell Shape

Abstract

Ideally, biomaterials have inductive properties, favoring specific lineage differentiation. For chondrogenic induction, these properties have been attributed to collagen type II. However, the underlying mechanisms are largely unknown. This study aimed to investigate whether collagen type II favors chondrogenic induction by affecting cell shape through β1 integrins and Rho A/Rock signaling. For this purpose, adipose tissue–derived stem cells (ASCs) were encapsulated in collagen type I or II gels and cultured in plain and chondrogenic medium. It was demonstrated that (i) ASCs showed more efficient chondrogenic induction (higher collagen X, aggrecan, sox6, sox9, and collagen II gene expression) in both plain and chondrogenic media in collagen type II versus collagen type I gels; (ii) ASCs showed lower Rock 2 gene expression and a more rounded cell shape in collagen type II versus type I gels when grown in plain medium; (iii) Rock inhibitor (Y27632) more effectively enhanced chondrogenic gene expression of ASCs in collagen type I than in collagen type II gels, and diminished differences in chondrogenic gene expression and cell shape of ASCs between the two gel types; and (iv) β1 integrins blocking not only reduced the differences of chondrogenic gene expression but also eliminated the differences of Rock 1 and Rock 2 gene expressions and cell shape when comparing ASCs embedded in collagen type I and II gels. We conclude that collagen type II provides the inductive signaling for chondrogenic differentiation in ASCs by evoking a round cell shape through β1 integrin–mediated Rho A/Rock signaling pathway.

Introduction

Damaged cartilage tissues (e.g., articular cartilage and nucleus pulposus of the intervertebral disc) related to degenerative diseases and traumatic injury are a major health problem around the world,^1,2 because of their limited self-repair capability, low cell content, and low supply of progenitor cells.^3,4 Despite the recent advances in surgical and nonsurgical interventions, no treatments that can effectively restore or repair the damaged cartilage are yet available.⁵ Tissue engineering offers a unique therapeutic opportunity for the repair of damaged tissues.^4,5 There are three key elements in the concept of tissue engineering: cells, scaffolds (biomaterials), and bioactive factors. Their proper combination might create a surgically transplantable product that promotes tissue regeneration and restores the intrinsic biological function of the tissue.

Among the three key elements of tissue engineering concept, the choice of scaffolds (biomaterial) is critical for a successful tissue engineering approach. The ideal biomaterial should be able to provide suitable biomechanical and biocompatibility properties to the implanted tissue, and also a favorable environment for cell attachment. Moreover, in particular in the case of tissue engineering using stem cells, biomaterials should preferably have inductive properties that favor a specific lineage differentiation.^6,7 For instance, bone matrix extracts favor osteoblastic differentiation of human bone marrow–derived mesenchymal stem cells (MSCs)⁸; cartilaginous extracellular matrix (ECM) components, however, favor the chondrogenic differentiation of MSCs.^9,10 Therefore, it is imperative to know the underlying mechanisms by which the different biomaterial-based scaffolds exert their distinct lineage differentiation induction.

Cell shape has recently been identified as an important determinant for stem-cell differentiation,^11,12 and chondrogenesis benefits from round cell shape.^13–15 When cells are seeded in/on a three-dimensional scaffold, cell shape can be largely affected by cell membrane receptor–scaffold (ECM) interactions.^16,17 β1 integrins are one of the main cell surface receptors to interact with ECM molecules. The extracellular domain of β1 integrins recognizes and binds to ECM proteins, whereas the intracellular domain associates with cytoskeleton elements and subsequently regulates cell shape and the differentiation of MSCs.¹⁴ As an upstream regulator of cytoskeleton dynamics, Rho A/Rock signaling, which stimulates stressed fibril formation and promotes elongated cell shape, has been shown to inhibit the multiple stages of chondrogenic differentiation and instead promotes chondrocyte proliferation.^13,18 In our previous study, we proposed that β1 integrin–mediated Rho A/Rock signaling pathway modulates the shape of the cells and subsequently influences the chondrogenic differentiation.

Collagen type II, one dominant component in a pool of ECM molecules of cartilage tissue, has the preference for both maintaining and inducing chondrogenic phenotype.^9,19–21 In this study, we hypothesized that collagen type II may enhance chondrogenic differentiation by supporting round cell shape through β1 integrin–mediated Rho A/Rock signaling pathway. To test this hypothesis, first, chondrogenic induction properties of collagen type I and II were compared by embedding adipose tissue–derived stem cells (ASCs) in collagen type I and II gels, and Rock 1 and 2 gene expressions and cell shape (characterized by actin cytoskeleton visualization) of ASCs were also compared; second, the effects of Rock inhibitor (Y27632) on chondrogenic differentiation and cell shape between ASCs in collagen I or II gels were further analyzed and compared; finally, the roles of β1 integrins in ASCs in determining chondrogeneis, Rock signaling, and cell shape were investigated by blocking ASCs with neutralizing β1 integrin antibodies before collagen type I/II embedment.

Materials and Methods

ASC isolation and culture

Retrieval of all human specimens was approved by the human ethical committee of the VU University Medical Center, and informed consent was obtained.

ASCs were isolated from subcutaneous adipose tissue obtained from patients undergoing elective surgical procedures as described previously.²² Briefly, harvested tissue was washed several times with phosphate-buffered saline (PBS) and then enzymatically dissociated with 0.5 U/mL Liberase Blendzyme 3 solution (Roche Diagnostics, Almere, The Netherlands). The dissociated tissue was then passed through a 100-μm mesh filter (Stokvis & Smith B.V., IJmuiden, The Netherlands) to remove debris, and the resulting single-cell suspension was centrifuged. Pelleted stromal cells were then washed several times with PBS. The stromal cells were plated in 25-cm² tissue culture flasks. Plating and expansion medium consisted of Dulbecco's modified Eagle's medium (DMEM; Gibco, Paisley, United Kingdom) supplemented with 500 μg/mL streptomycin sulfate (Sigma, St. Louis, MO), 600 μg/mL penicillin (Sigma), 50 μg/mL gentamycin (Gibco), 2.5 μg/mL fungizone (Gibco), and 10% fetal bovine serum (Hyclone, Logan, UT). Cultures were grown in a humidified incubator at 37°C in an atmosphere of 5% CO₂. ASCs used for the experiments were at passages 2–4.

Embedment of ASCs in collagen type I/II hydrogels and in vitro chondrogenesis

Collagen type I/II hydrogels were prepared by mixing five parts of collagen type I (rat tail; R&D Systems, Minneapolis, MN) or collagen type II (chicken sternal cartilage; Sigma), one part of 10 × DMEM, one part of reconstitution buffer (2.2 g NaHCO₃ in 100 mL of 0.05 N NaOH and 200 mM HEPES), one part of 20 mM acetic acid, and two parts of cell suspension in culture medium. The final concentration of collagen type I/II was 2.5 mg/mL, and the cell density was 1.5 × 10⁶ cells/mL gel. All the components above were mixed on ice, and 100 μL mixtures were seeded onto six-well plates for the study. The collagen gel lattices were formed by placing the plates in a humidified incubator at 37°C in an atmosphere of 5% CO₂ for 1 h. Thereafter, the culture medium, either plain medium (DMEM supplemented with ITS+™ Premix [final concentration in medium when diluted 1:100 insulin, 6.25 μg/mL; transferrin, 6.25 μg/mL; selenous acid, 6.25 ng/mL; bovine serum albumin, 1.25 mg/mL; linoleic acid, 5.35 μg/mL; R&D Systems], 500 μg/mL streptomycin sulfate [Sigma], 600 μg/mL penicillin [Sigma], and 2.5 μg/mL fungizone [Gibco]) or chondrogenic medium (plain medium supplemented with 7.5 ng/mL transforming growth factor beta 1 [TGF-β1; R&D Systems]) was carefully overlaid on each gel. The medium was refreshed every 3–4 days till day 14.

β1 integrins blocking and supplement of Rock inhibitor

In β1 integrin blocking studies, ASCs were preincubated for 45 min with plain medium supplemented with either mouse nonimmune IgG (50 μg/mL; Vector Labs, Burlingame, CA) as control or neutralizing β1 integrin-blocking antibody (25 μg/mL, CD29; R&D Systems) as experimental group before embedding in collagen type I/II gels. Thereafter, the pretreated cells were used for the preparation of collagen gels.

For Rock inhibiting experiments, the medium was supplemented with either dimethyl sulfoxide (DMSO) as vehicle control (1 μL in 1 mL) or 10 μM Y27632 (dissolved in DMSO) as experimental group. The inhibitors were added every 24 h, and the medium was changed every 3 days until harvesting.

Confocal microscopy

For observing F-actin distribution of ASCs, ASC-loaded gels were fixed in 4% formaldehyde in 0.1 M phosphate buffer (pH 7.2) for 20 min and subsequently treated with 2% glycine/1% bovine serum albumin in PBS for 30 min. Thereafter, cells were permeabilized with 0.5% Triton X-100 in PBS for 20 min. Actin fibers within the ASCs were stained with 2.5 U rhodamine phalloidin (Invitrogen, Breda, The Netherlands) in PBS in the dark for 1 h and subsequently observed by confocal fluorescence microscopy with a Leica TCS SP5 confocal microscope (Leica Microsystems, Wetzlar, Germany).

Real-time polymerase chain reaction analysis

Total RNA was extracted from cultured cells using Trizol (Invitrogen) according to the manufacturer's instructions. Seven hundred and fifty nanograms of total RNA was reverse transcribed in 20 μL of reverse transcriptase (RT)-polymerase chain reaction (PCR) mix (5 U transcriptor reverse transcriptase [Roche Diagnostics], 0.08 A260 units random primers [Roche Diagnostics], 1 mM of each of dNTP [Invitrogen], 10 U protector RNase inhibitor [Roche diagnostics], and 1× transcriptor RT buffer) at 55°C for 30 min and then at 85°C for 5 min to inactivate the transcriptase. Real-time PCR was performed to determine the gene expression of aggrecan, collagen type X, collagen II, sox5, sox6, sox9, Rock 1, and Rock 2 by lightcycler (Roche LC 480). All the target genes were normalized to housekeeping gene 18S to obtain the relative gene expression. Primers used in real-time PCR are listed in Table 1.

Table 1.

The Primers Used for Real-Time Polymerase Chain Reaction

Gene		Oligonucleotide sequence	Product size (bp)
18S	Forward	5′-GTA ACC CGT TGA ACC CCA TT-3′	151
	Reverse	5′-CCA TCC AAT CGG TAG TAG CG-3′
Collagen type II	Forward	5′-GGA TGG GCA GAG GTA AAT G-3′	255
	Reverse	5′-GGT CCT TTG GGT CCT ACA A-3′
Collagen type X	Forward	5′ CAC TAC CCA ACA CCA AGA CA-3′	225
	Reverse	5′-CTG GTT TCC CTA CAG CTG AT-3′
Aggrecan	Forward	5′-CAA CTA CCC GGC CAT CC-3′	160
	Reverse	5′-GAT GGC TCT GTA ATG GAA CAC-3′
sox5	Forward	5′-CTC ACT TGA CAG GTT CAG-3′	208
	Reverse	5′-CGT CAC TCT CCT CTT CTT C-3′
sox6	Forward	5′-CTC CCA CCC ACA AAT TAA CC-3′	207
	Reverse	5′-GAG ACC GTC CCT GCT GTG TT-3′
sox9	Forward	5′-CCC AAC GCC ATC TTC AAG G-3′	242
	Reverse	5′-CTG CTC AGC TCG CCG ATG T-3′

Statistical analysis

Data were obtained from five independent donors. For statistical analysis, Levene's test was firstly performed to determine the homogeneity of variance for all the data, and then the data were analyzed by repeated measures in general linear model followed by Tukey post hoc analysis. When the data are not equally distributed, log transformation was used. SPSS 13.0 program (Chicago, Illinois) was employed for all statistic analysis, and differences were considered significant if p < 0.05.

Results

Collagen type II favors chondrogenic induction in ASCs compared with collagen type I

ASCs embedded in collagen type I or II gels were grown in either plain or chondrogenic medium for 4 and 14 days, and chondrogenic gene expression of ASCs (collagen X, aggrecan, sox5, sox6, sox9, and collagen type II) was analyzed. Our results showed that the responses of ASCs to chondrogenic induction largely depended on the collagen type in which they were encapsulated: for ASCs in collagen type II gels, TGF-β1 (chondrogenic condition) significantly upregulated the gene expression of collagen type X (Fig. 1A) but not of other chondrogenic markers (aggrecan, sox5, sox6, and sox9; Fig. 1B–E). TGF-β1 also advanced the induction time of collagen type II gene expression from day 14 in plain medium to day 4 in chondrogenic medium (Table 2). However, for ASCs in collagen type I gels, TGF-β1 (chondrogenic condition) significantly upregulated the gene expressions of collagen type X, sox6, and sox9 (Fig. 1A, D, E), and it also increased the induction frequency of collagen type II gene expression among the five donors (Table 2).

FIG. 1.

Comparisons of chondrogenic gene expression between adipose tissue–derived stem cells (ASCs) embedded in collagen type I and type II three-dimensional (3D) gels. After ASCs were embedded in either collagen type I or II gels and cultured for 4 and 14 days, ASCs showed different chondrogenic gene expression profiles in response to chondrogenic induction and the type of collagen used. (A) ASCs in both collagen type I and II gels displayed significant higher induction of collagen X gene expression under chondrogenic conditions. (B) There was no significant induction of aggrecan gene expression in ASCs in either collagen type I or II gels under chondrogenic conditions. (C) There was no significant induction of sox5 gene expression of ASCs in either collagen type I or II gels under chondrogenic conditions. (D) sox6 gene expression of ASCs was only significantly induced in collagen type I gels at day 14 under chondrogenic conditions. (E) sox9 gene expression of ASCs was only significantly induced in collagen type I gels at day 14 under chondrogenic conditions. (F) ASCs in collagen type II gels showed higher chondrogenic gene expressions than those in collagen type I gels in both plain (aggrecan, sox6, and sox9 at day 14) and chondrogenic (collagen X and sox9 at day 14) medium. Data were represented as mean ± standard error of mean (SEM), n = 5 [(A–E) relative gene; expression; (F) ratios of relative gene expression of ASCs in collagen type II gels vs. ASCs in collagen type I gels; dashed line, no difference]. Col, collagen. *p < 0.05.

Table 2.

Induction Frequency of Collagen Type II Gene Expression of Adipose Tissue–Derived Stem Cells Embedded in Different Collagen Gels and With Different Treatments

		Plain medium			Chondrogenic medium
Gel type	Treatments	Day 4	Day 14	Total	Day 4	Day 14	Total
Collagen type I	Control	0	2	2	2	3	4
	Y27632	4	4	4	5	3	5
	β1 integrin blocking	1	4	4	5	2	5
Collagen type II	Control	1	4	4	4	2	5
	Y27632	4	5	5	5	3	5
	β1 integrin blocking	4	5	5	5	3	5

Chondrogenic gene expression was also compared between ASCs in collagen type I and II gels. In both plain and chondrogenic medium, ASCs in collagen type II gels displayed more efficient chondrogenic induction than those in collagen type I gels: significant higher aggrecan, sox6, and sox9 gene expressions at day 14 in plain medium, and higher collagen type X and sox9 gene expressions at day 14 in chondrogenic medium (Fig. 1F). Under both plain and chondrogenic conditions, collagen type II gene expression was also induced more frequently and at an earlier time point when comparing type II and type I collagen gels (Table 2).

Collagen type II gel embedment of ASCs results in lower Rock gene expression and more round cells presented than collagen type I gel embedment

To address the question why collagen type II gel favors chondrogenic differentiation of ASCs, gene expression of Rock 1 and Rock 2 as well as the ASC cell shape in collagen type I and II gels were determined. When ASCs embedded in either collagen type I or II gels were cultured in plain or chondrogenic medium for 4 and 14 days, the following comparisons were made.

The effects of chondrogenic condition on Rock gene expression of ASCs in either collagen type I or II gels: it was found that only at the 14-day time point, Rock 1 and Rock 2 gene expressions of ASCs in collagen type I gels were significantly downregulated by chondrogenic condition but not in collagen type II gels (Fig. 2A, B).

The effects of collagen gel type embedded on Rock gene expression of ASCs under plain or chondrogenic conditions: under plain medium, a significant downregulation was observed for Rock 2 at day 4 and a similar trend for Rock 1 at the same time point in collagen type II gels versus collagen type I gels, and expression differences were lost after 14 days. However, chondrogenic conditions leveled out the differences observed under plain conditions (Fig. 2C).

Previously, we reported that ASCs seeded in collagen type II gels, but not in collagen type I gels, are prone to aggregate after several days of culturing in plain medium⁹ and that only small amounts of ASCs in collagen type I gels displayed rounded cell shapes, whereas most of them remained elongated with stressed actin cytoskeleton across the whole cell.²³ The latter observations were confirmed in this study (Fig. 3A). In contrast, most of the ASCs embedded in collagen type II were characterized with round cell shapes, and relaxed actin cytoskeleton smoothly distributed underneath of the cell membranes (Fig. 3D). Under chondrogenic conditions, cell shapes between ASCs in collagen type I and II gels did not show apparent differences; in other words, chondrogenic conditions induced round cell shapes of ASCs in collagen type I gels but did not further change the round cell shapes of ASCs in collagen type II gels except that they prevented ASC cell aggregation (Fig. 3G, J).

FIG. 2.

Comparisons of Rock 1 and 2 gene expression between ASCs embedded in collagen type I and collagen type II 3D gels. After ASCs were embedded in either collagen type I gels or II gels and cultured for 4 and 14 days, ASCs showed different Rock 1 and 2 gene expression profiles in response to chondrogenic induction and the type of collagen used. (A) Rock 1 gene expression of ASCs was only significantly downregulated in collagen type I gels at day 14 under chondrogenic conditions. (B) Rock 2 gene expression of ASCs was only significantly downregulated in collagen type I gels at day 14 under chondrogenic conditions. (C) In plain medium, Rock 2 gene expression of ASCs in collagen type II gels was significantly lower than in collagen type I gels. Data were represented as mean ± SEM, n = 5 [(A, B) relative gene expression; (C) ratios of relative gene expression of ASCs in collagen type II gels vs. ASCs in collagen type I gels; dashed line, no difference]. Col, collagen. *p < 0.05.

FIG. 3.

Comparisons of cell shape or actin cytoskeleton distribution between ASCs in collagen type I or II 3D gels before and after treatment with Rock inhibitor (Y27632) or β1 integrin blocking. (A–F) In plain medium, most of the ASCs in collagen type I gels remained elongated, containing stressed actin cytoskeleton across the whole cell (A); in contrast, most of the ASCs in collagen type II were characterized with round cell shapes and relaxed actin cytoskeleton smoothly distributed underneath the cell membranes (D); Y27632 induced some round ASCs in collagen type I gels with less stressed and/or granule-like actin cytoskeleton (B); however, Y27632 did not further change the round cell shape of ASCs but induced some granule-like actin cytoskeleton as well (E); β1 integrin blocking induced a round cell shape in ASCs in collagen type I gels (C), but it did not further change the distribution of the actin cytoskeleton and round cell shape of ASCs in collagen type II gels (F). (G–L) Under chondrogenic conditions, all the ASCs in collagen type I and II gels looked similar and displayed round cell shapes irrespective of whether they were embedded in collagen type I or II gels or treated with Y27632 or β1 integrin blocking. Col, collagen. Scale bars: 50 μm. Color images available online at www.liebertonline.com/ten.

Rock inhibitor differently affects chondrogenic gene expression and cell shape of ASCs in collagen type I or II gels

Next, it was examined whether collagen type II exerted its chondrogenic induction properties through the modulation of Rock signaling. ASCs embedded in either collagen type I or II gels were cultured in medium supplemented with Rock inhibitor (Y27632) or vehicle (DMSO) for 4 and 14 days. Differential responses to Y27632 were found for ASCs in collagen type I and II gels with respect to chondrogenic gene expression.

In collagen type I gels, Y27632 significantly upregulated sox6 gene expression in ASCs cultured in plain medium (Fig. 4A) and also exhibited increased induction frequency and advanced induction time (from day 14 to day 4) of collagen type II gene expression in ASCs embedded in both plain and chondrogenic medium (Table 2). For ASCs in collagen type II gels, however, Y27632 did not significantly affect chondrogenic marker expression in ASCs except that it advanced the induction time (from day 14 to day 4) of collagen type II gene expression of ASCs cultured in both plain and chondrogenic medium (Fig. 4B and Table 2). The significant differences of chondrogenic gene expression (collagen type X, aggrecan, sox6, and sox9) observed in ASCs embedded in collagen type II versus type I gels (Fig. 1F) were largely diminished after Y27632 was added in the culturing media (Figs. 1F and 4C; Table 2).

FIG. 4.

Effects of Rock inhibitor on chondrogenic differentiation in ASCs in collagen type I or II 3D gels. After ASCs were embedded in either collagen type I or II gels and cultured in medium supplemented with Rock inhibitor (Y27632) for 4 and 14 days, ASCs in collagen type I and II gels showed different responses to Y27632 with respect to chondrogenic gene expression. (A) Y27632 significantly upregulated sox6 gene expression of ASCs when they were embedded in collagen type I gels. (B) Y27632 did not significantly affect any chondrogenic marker in ASCs embedded in collagen type II gels. (C) Only collagen type X gene expression was significantly higher in ASCs in collagen type II gels compared with those in collagen type I gels at day 14. Data were represented as the ratios of mean ± SEM, n = 5 [(A, B) relative gene expression of experimental group treated with Y27632 vs. control group treated with dimethyl sulfoxide; (C) relative gene expression of ASCs in collagen type II gels vs. ASCs in collagen type I gels; dashed line, no difference]. Col, collagen. *p < 0.05.

In concert with the effects of Y27632 on chondrogenic gene expression of ASCs, Y27632 also had differential effects on the cell shapes of ASCs in collagen type I and II gels. In plain medium in collagen type I gels, Y27632 induced some irregular ASCs, and these cells were characterized with less stressed and/or granule-like actin cytoskeleton compared with control incubations (Fig. 3A, B). For ASCs in collagen type II gels, however, Y27632 did not further change the round cell shapes of ASCs except to induce some granule-like actin cytoskeleton (Fig. 3D, E). In chondrogenic medium, ASCs in collagen type I or II gels still kept round cell shapes, but some actin cytoskeleton protruded from the cellular membranes of ASCs in both collagen type I and II gels (Fig. 3H, K).

β1 integrins blocking in ASCs differently affects chondrogenic and Rock gene expression and ASC shape in collagen type I and II gels

Finally, we investigated whether β1 integrins participated in collagen type II–mediated chondrogenic induction by regulating Rock signaling and cell shape. Comparisons of chondrogenic and Rock 1/2 gene expression and cell shape between ASCs in collagen type I and II gels were performed after ASCs were preincubated with neutralizing β1 integrin antibodies before collagen gel embedment. The significant differences of chondrogenic (collagen type X, aggrecan, sox6, and sox9) and Rock gene expressions of ASCs resulted from collagen type I and II gel embedment were eliminated after β1 integrins in ASCs were blocked by neutralizing β1 integrin antibodies (data not shown), although collagen type II gene expression of ASCs in collagen type II gels was still more frequently induced than those in collagen type I gels (Table 2).

In plain medium, as previously reported, β1 integrin blocking in ASCs dramatically changed the shapes of ASCs and induced round cell shapes when ASCs were embedded in collagen type I gels (Fig. 3C) However, β1 integrin blocking did not further change the round cell shape or actin cytoskeleton distribution of ASCs in collagen type II gels (Fig. 3F). In chondrogenic medium, ASCs in collagen type I or II gels remained round cell shapes with actin cytoskeleton distributed underneath of cellular membranes after β1 integrins were blocked by β1 integrin neutralizing antibodies (Fig. 3I, L).

Discussion

The function of scaffolds has been extended to biofunctionality: not only the fundamental function of supporting cell attachment and cell proliferation but also the inductive function of driving stem cells into specific lineage differentiation. Many efforts have been made to reveal the mechanisms of scaffold-mediated inductive potential, for example, that the stiffness of substrate or the pore size of scaffold can dramatically affect the differentiation of stem cells attached.^11,24,25 In this study, we demonstrated that collagen type II possessed a higher chondrogenic inductive property compared with collagen type I, and that collagen type II gel embedment of ASCs downregulated Rock 2 gene expression and supported round cell shape. Further, we showed that Rock inhibitor (Y27632) largely diminished the differences of chondrogenic gene expression and cell shape of ASCs induced by collagen type I and type II gel embedment. Finally, our data showed that β1 integrin blocking in ASCs also largely eliminated the differences of chondrogenic and Rock gene expression and cell shape of ASCs resulted from collagen type I and II gel embedment.

The stronger inductive property of collagen type II over collagen type I for maintaining and inducing a chondrogenic phenotype was shown by our group in various different circumstances (chondrogenic induction by soluble factors secreted by nucleus pulposus cells or by TGF-β1) and also by other groups.^9,19,20 This study was performed in an attempt to reveal the mechanisms underlying this phenomenon. We previously reported that ASCs seeded in collagen type II gels are prone to aggregate at around day 7 after they were cultured in plain medium.⁹ In the current study, a similar phenomenon was observed. Also it was found that most of the ASCs in collagen type II were characterized with round cell shapes, and actin cytoskeleton smoothly distributed underneath the cell membranes. It is very interesting to compare these results with data recently published by Wu et al.,¹⁰ which showed that a chondroitin sulfate–based microenvironment also promotes cell aggregation and subsequent chondrogenic differentiation. Since both collagen type II and chondroitin sulfate are main components of the cartilaginous ECM, combining these datasets suggests that the ECM microenvironment in cartilage can provide a good “soil” for guiding the chondrogenic differentiation of stem cells and that the ECM does so by influencing the shape of the encapsulated cells.

The stronger chondrogenesis-inductive property of collagen type II over collagen type I prompted us to evaluate its underlying mechanisms. Cell shape–dependent control of lineage commitment is believed to be mediated by Rho A signaling pathway, specifically via its effects on Rock-mediated cytoskeletal tension.^12,26,27 The evidence of the role of Rho A/Rock-mediated cytoskeleton (especially for actin cytoskeleton) organization in chondrogenic differentiation has accumulated over the last 20 years.¹⁴ It is well known that chondrocytes plated in monolayer culture tend to change their morphology to flattened cells and cease the production of collagen II and glycosaminoglycans (GAGs).^28,29 The mere addition of inhibitors of actin polymerization (dihydrochalasin B or cytochalasin D) can stimulate rounding of dedifferentiated chondrocytes and reexpression of these same chondrogenic markers.^30,31 First, we showed that encapsulating ASCs in collagen type II gels, on one hand, decreased gene expression of Rock 1/2 and caused less stressed actin cytoskeleton (round cell shape), and on the other hand promoted chondrogenic commitment. Second, although we did not observe that Rock inhibitor (Y27632) induced round cell shape and cortical distribution of cytoskeleton in ASCs as reported in other studies^13–15 (might be because of less sensitivity of ASCs to the stimulation of Y37632), Y37632 did exert its chondrogenic induction function by upregulating sox6 gene expression of ASCs in collagen type I gels and also increased the induction frequency of collagen type II gene expression of ASCs in both collagen type I and II gels. In addition, Y27632 largely diminished the differences of chondrogenic gene expression found after collagen type I and II embedment. Based on these data, we postulate that collagen type II enhances chondrogenic differentiation and supports round cell shape largely through Rho A/Rock-mediated actin cytoskeleton signaling pathway. However, it is notable that Rho A/Rock signaling might not be the only mechanism underlying the preferential chondrogenic induction property of collagen type II, because ASCs in collagen type II gels still displayed slightly more efficient chondrogenic induction than those in collagen type I gels as shown by more frequent induction of collagen type II gene expression.

We previously proposed a model for chondrogenic differentiation of ASCs induced by soluble inductive factors, cell shape, and cell–matrix interactions²³: cell–matrix interaction is able to promote the chondrogenic commitment of stem cells by affecting cell shape but is not sufficient for mature chondrogenic differentiation, and this effect can be largely replaced by the normal chondrogenic induction of growth factors (TGF-β1) through the Rho/Rock signaling pathway or direct Rho/Rock signaling inhibition (β1 integrins blocking or Rock inhibitor). The results of the current study were in concert with this proposal: there were significant differences between ASCs in collagen type I and II gels when they were cultured in plain medium, but these differences were largely eliminated by the introduction of chondrogenic conditions, β1 integrins blocking, or the addition of Y27632.

The roles of β1 integrins in chondrogenesis have been extensively studied but so far remain elusive.^32–36 Our previous study showed that β1 integrins play a more important role in the later stages than in the earlier stages of chondrogenesis, and Rho A/Rock signaling appeared to participate in this process.²³ This study further provided the evidence that ECM can modulate chondrogenesis through β1 integrins-Rho/Rock signaling pathway and showed that β1 integrins blocking eliminated or largely eliminated the differences of chondrogenic and Rock gene expression resulting from collagen type I and II embedment. However, because of using a general β1 integrin neutralizing antibody in our previous and current study, we could not deduce which particular subset of the αnβ1 integrin family was involved in this induction process. Our current, comparative studies on the differential effects of both types of collagen (I and II), combined with the fact that α1β1, α2β1, and α3β1 integrins are the main receptors for collagen type I^37,38 and α1β1, α2β1, and α10β1 integrins are the main receptors for collagen type II^34,39,40 suggest a pivotal role for the α10β1 integrin. Unfortunately, since no neutralizing antibodies for the α10 subunit are available, direct evidence is currently lacking, although some interesting in vitro and in vivo studies are in support of this hypothesis: (i) the α10β1 integrin is known to be highly expressed in chondrocytes⁴¹; (ii) α10β1 integrin is increasingly expressed in the progress of chondrogenic differentiation^42,43; and (iii) α10β1 integrin knockdown results in the failed maturation of chondrocytes in vivo.^42,43 Additional studies using different approaches, such as blocking α10 expression blocking studies using siRNA technology, are needed to provide more insight into the potential roles of the α10β1 integrin receptor. Interestingly, the fact that the differences of chondrogenic gene expression between ASCs in collagen type I and II gels were not completely eliminated by β1 integrin blocking may also suggest that so far unidentified additional collagen receptors may be involved in chondrogenic induction processes.

Conclusions

In summary, this study shows that collagen type II scaffolds not only provide support for ASCs but also mediate inductive signaling for chondrogenic differentiation by provoking a round cell shape through β1 integrin–mediated Rho A/Rock signaling. Moreover, collagen type II enhances the action of chondroinductive factors such as TGF-β1. These data highlight the importance of the coordination among cells, scaffolds, and bioactive factors, the three main components in the concept of tissue engineering, and they also have an important implication for designing appropriate scaffolds for cartilage tissue engineering, which not only support the cells but also shape the round cells.

Footnotes

Acknowledgment

This study is financially supported by the Dutch Tissue Engineering Program (DPTE # BGT.6734).

Disclosure Statement

No competing financial interests exist.

References

Buckwalter

J.A.

Articular cartilage injuries. Clin Orthop Relat Res, 402:21. 2002.

Martin

J.A.

, Buckwalter

J.A.

Aging, articular cartilage chondrocyte senescence and osteoarthritis. Biogerontology, 3:257. 2002.

Guilak

, Awad

H.A.

, Fermor

, Leddy

H.A.

, Gimble

J.M.

Adipose-derived adult stem cells for cartilage tissue engineering. Biorheology, 41:389. 2004.

Hardingham

, Tew

, Murdoch

Tissue engineering: chondrocytes and cartilage. Arthritis Res, 4,Suppl 3:S63. 2002.

Frenkel

S.R.

, Di Cesare

P.E.

Scaffolds for articular cartilage repair. Ann Biomed Eng, 32:26. 2004.

Falconnet

, Csucs

, Grandin

H.M.

, Textor

Surface engineering approaches to micropattern surfaces for cell-based assays. Biomaterials, 27:3044. 2006.

Furth

M.E.

, Atala

, van Dyke

M.E.

Smart biomaterials design for tissue engineering and regenerative medicine. Biomaterials, 28:5068. 2007.

El Sabban

M.E.

, El Khoury

, Hamdan-Khalil

, Sindet-Pedersen

, Bazarbachi

Xenogenic bone matrix extracts induce osteoblastic differentiation of human bone marrow-derived mesenchymal stem cells. Regen Med, 2:383. 2007.

Z.F.

, Zandieh

D.B.

, Wuisman

P.I.

, Bank

R.A.

, Helder

M.N.

Influence of collagen type II and nucleus pulposus cells on aggregation and differentiation of adipose tissue-derived stem cells. J Cell Mol Med, 12:2812. 2008.

10.

Y.N.

, Yang

, Hui

J.H.

, Ouyang

H.W.

, Lee

E.H.

Cartilaginous ECM component-modification of the micro-bead culture system for chondrogenic differentiation of mesenchymal stem cells. Biomaterials, 28:4056. 2007.

11.

Engler

A.J.

, Sen

, Sweeney

H.L.

, Discher

D.E.

Matrix elasticity directs stem cell lineage specification. Cell, 126:677. 2006.

12.

McBeath

, Pirone

D.M.

, Nelson

C.M.

, Bhadriraju

, Chen

C.S.

Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev Cell, 6:483. 2004.

13.

Woods

, Wang

, Beier

RhoA/ROCK signaling regulates Sox9 expression and actin organization during chondrogenesis. J Biol Chem, 280:11626. 2005.

14.

Woods

, Wang

, Beier

Regulation of chondrocyte differentiation by the actin cytoskeleton and adhesive interactions. J Cell Physiol, 213:1. 2007.

15.

Zanetti

N.C.

, Solursh

Induction of chondrogenesis in limb mesenchymal cultures by disruption of the actin cytoskeleton. J Cell Biol, 99:115. 1984.

16.

, Wu

, Shi

, Chen

, Wu

Migfilin and Mig-2 link focal adhesions to filamin and the actin cytoskeleton and function in cell shape modulation. Cell, 113:37. 2003.

17.

Zhang

, Guo

, Chen

, Wu

A critical role of the PINCH-integrin-linked kinase interaction in the regulation of cell shape change and migration. J Biol Chem, 277:318. 2002.

18.

Riento

, Ridley

A.J.

Rocks: multifunctional kinases in cell behaviour. Nat Rev Mol Cell Biol, 4:446. 2003.

19.

Bosnakovski

, Mizuno

, Kim

, Takagi

, Okumura

, Fujinaga

Chondrogenic differentiation of bovine bone marrow mesenchymal stem cells (MSCs) in different hydrogels: influence of collagen type II extracellular matrix on MSC chondrogenesis. Biotechnol Bioeng, 93:1152. 2006.

20.

Chen

C.W.

, Tsai

Y.H.

, Deng

W.P.

, Shih

S.N.

, Fang

C.L.

, Burch

J.G.

, Chen

W.H.

, Lai

W.F.

Type I and II collagen regulation of chondrogenic differentiation by mesenchymal progenitor cells. J Orthop Res, 23:446. 2005.

21.

Scully

S.P.

, Lee

J.W.

, Ghert

P.M.A.

, Qi

The role of the extracellular matrix in articular chondrocyte regulation. Clin Orthop Relat Res, 391:S72. 2001.

22.

Z.F.

, Zandieh

D.B.

, Wuisman

P.I.

, Bank

R.A.

, Helder

M.N.

Differentiation of adipose stem cells by nucleus pulposus cells: configuration effect. Biochem Biophys Res Commun, 359:991. 2007.

23.

Z.F.

, Zandieh

D.B.

, Huang

C.L.

, Bank

R.A.

, Helder

M.N.

Beta1 integrins regulate chondrogenesis and rock signaling in adipose stem cells. Biochem Biophys Res Commun, 372:547. 2008.

24.

Hirata

, Nomura

, Ito

, Shimazu

, Okazaki

Acceleration of bone formation with BMP2 in frame-reinforced carbonate apatite-collagen sponge scaffolds. J Artif Organs, 10:212. 2007.

25.

Hutmacher

D.W.

, Schantz

J.T.

, Lam

C.X.

, Tan

K.C.

, Lim

T.C.

State of the art and future directions of scaffold-based bone engineering from a biomaterials perspective. J Tissue Eng Regen Med, 1:245. 2007.

26.

Harvey

K.A.

, Paranavitana

C.N.

, Zaloga

G.P.

, Siddiqui

R.A.

Diverse signaling pathways regulate fibroblast differentiation and transformation through Rho kinase activation. J Cell Physiol, 211:353. 2007.

27.

Varon

, Basoni

, Reuzeau

, Moreau

, Kramer

I.J.

, Genot

TGFbeta1-induced aortic endothelial morphogenesis requires signaling by small GTPases Rac1 and RhoA. Exp Cell Res, 312:3604. 2006.

28.

Grundmann

, Zimmermann

, Barrach

H.J.

, Merker

H.J.

Behaviour of epiphyseal mouse chondrocyte populations in monolayer culture. Morphological and immunohistochemical studies. Virchows Arch A Pathol Anat Histol, 389:167. 1980.

29.

von der

M.K.

, Gauss

, von der

M.H.

, Muller

Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture. Nature, 267:531. 1977.

30.

Benya

P.D.

, Brown

P.D.

, Padilla

S.R.

Microfilament modification by dihydrocytochalasin B causes retinoic acid-modulated chondrocytes to reexpress the differentiated collagen phenotype without a change in shape. J Cell Biol, 106:161. 1988.

31.

Benya

P.D.

, Padilla

S.R.

Dihydrocytochalasin B enhances transforming growth factor-beta-induced reexpression of the differentiated chondrocyte phenotype without stimulation of collagen synthesis. Exp Cell Res, 204:268. 1993.

32.

Aszodi

, Hunziker

E.B.

, Brakebusch

, Fassler

Beta1 integrins regulate chondrocyte rotation, G1 progression, and cytokinesis. Genes Dev, 17:2465. 2003.

33.

Fassler

, Meyer

Consequences of lack of beta 1 integrin gene expression in mice. Genes Dev, 9:1896. 1995.

34.

Loeser

R.F.

, Sadiev

, Tan

, Goldring

M.B.

Integrin expression by primary and immortalized human chondrocytes: evidence of a differential role for alpha1beta1 and alpha2beta1 integrins in mediating chondrocyte adhesion to types II and VI collagen. Osteoarthritis Cartilage, 8:96. 2000.

35.

Shakibaei

Inhibition of chondrogenesis by integrin antibody in vitro. Exp Cell Res, 240:95. 1998.

36.

Takahashi

, Onodera

, Sasano

, Mizoguchi

, Bae

J.W.

, Mitani

, Kagayama

, Mitani

Effect of stretching on gene expression of beta1 integrin and focal adhesion kinase and on chondrogenesis through cell-extracellular matrix interactions. Eur J Cell Biol, 82:182. 2003.

37.

Jokinen

, Dadu

, Nykvist

, Kapyla

, White

D.J.

, Ivaska

, Vehvilainen

, Reunanen

, Larjava

, Hakkinen

, Heino

Integrin-mediated cell adhesion to type I collagen fibrils. J Biol Chem, 279:31956. 2004.

38.

Lundstrom

, Holmbom

, Lindqvist

, Nordstrom

The role of alpha2 beta1 and alpha3 beta1 integrin receptors in the initial anchoring of MDA-MB-231 human breast cancer cells to cortical bone matrix. Biochem Biophys Res Commun, 250:735. 1998.

39.

Camper

, Hellman

, Lundgren-Akerlund

Isolation, cloning, and sequence analysis of the integrin subunit alpha10, a beta1-associated collagen binding integrin expressed on chondrocytes. J Biol Chem, 273:20383. 1998.

40.

Knudson

, Loeser

R.F.

CD44 and integrin matrix receptors participate in cartilage homeostasis. Cell Mol Life Sci, 59:36. 2002.

41.

Camper

, Holmvall

, Wangnerud

, Aszodi

, Lundgren-Akerlund

Distribution of the collagen-binding integrin alpha10beta1 during mouse development. Cell Tissue Res, 306:107. 2001.

42.

Bengtsson

, Aszodi

, Nicolae

, Hunziker

E.B.

, Lundgren-Akerlund

, Fassler

Loss of alpha10beta1 integrin expression leads to moderate dysfunction of growth plate chondrocytes. J Cell Sci, 118:929 3 1 2005.

43.

Varas

, Ohlsson

L.B.

, Honeth

, Olsson

, Bengtsson

, Wiberg

, Bockermann

, Jarnum

, Richter

, Pennington

, Johnstone

, Lundgren-Akerlund

, Kjellman

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. 2007.