Received: 12 April 2016 | Accepted: 1 June 2016 DOI: 10.1111/cpr.12272 ORIGINAL ARTICLE Poly(ε-caprolactone)-based substrates bearing pendant small chemical groups as a platform for systemic investigation of chondrogenesis Min Chen1 | Lei Xu1 | Yan Zhou1 | Yan Zhang2 | Meidong Lang2 | Zhaoyang Ye1 | Wen-Song Tan1 State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai, China 1 Abstract Objectives: Physiochemical properties of biomaterials play critical roles in dictating Key Laboratory for Ultrafine Materials of Ministry of Education, School of Materials Science and Engineering, East China University of Science and Technology, Shanghai, China types of cell behaviour. In this study, a series of poly(ε-caprolactone) (PCL)-derived Correspondence Zhaoyang Ye and Wen-Song Tan, State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai, China. Emails: zhaoyangye@ecust.edu.cn and wstan@ecust.edu.cn Rabbit articular chondrocytes (rACs) and rabbit bone marrow–derived mesenchymal 2 polymers bearing different small chemical groups was employed as a platform to evaluate chondrogenesis of different cell types. Materials and methods: Thin films were prepared by spin-coating PCL derivatives. stem cells (rMSCs) were seeded on to the films, and cell adhesion, proliferation, extracellular matrix production and gene expression were evaluated. Results: The presence of hydrophilic groups (-NH2, -COOH, -OH and -C=O) promoted adhesion and proliferation of primary rACs and rMSCs. On these polymeric films, chondrogenesis of primary rACs depended on culture time. For passaged cells, redifferentiation was induced on these films by chondrogenic induction, but less for cells of passage 5 compared to passage 3. While films with hydrophilic groups favoured chondrocytic gene expression of both types of passaged cells, production of glycosaminoglycans (GAG) was similar for those of passage 3 on all films, and PCL-CH3 film better supported GAG production for cells of passage 5. Under chondrogenic conditions, rMSCs were more efficient at GAG production on PCL and PCL-NH2 films. Conclusions: This study demonstrates that different cells displayed distinct responses to substrate surface chemistry, implying that cell–biomaterial interactions can be developmental stage dependent. This provides a novel perspective for developing biomaterials for cartilage regeneration. 1 | INTRODUCTION vivo monolayer expansion of primary cells to obtain a sufficient quantity, which usually leads to dedifferentiation, namely downregulation Traumatic damage and osteoarthritic degeneration of articular carti- in typical cartilage gene expression and extracellular matrix (ECM) pro- lage are severe clinical issues due to its limited self-repair capacity. duction.3 On the other hand, tissue engineering aims at developing Current surgical practices are unable to achieve a long-term function- cartilage replacements in combination of biomaterials and cells such as al repair. Emerging technologies including cell therapy (i.e., autolo- chondrocytes and mesenchymal stem cells (MSCs).4 However, differ- gous chondrocyte implantation, ACI) and tissue engineering hold entiation of MSCs towards articular chondrocytes remains challenging great promise for cartilage regeneration.2 On one hand, ACI involves and current technologies often generate a hypertrophic, unstable car- implantation of isolated chondrocytes to defects and necessitates ex tilage phenotype.5,6 It is of note that engineered cartilage with MSCs 1 512 | © 2016 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/cpr Cell Proliferation 2016; 49: 512–522 | Chen et al. is still mechanically inferior to that with chondrocytes as well as native 7,8 513 on a substrate to solicit cell adhesion and favourable chondrogenesis.15,16 In a much simpler fashion, Ma et al. evaluated the effects cartilage. It has been recognized that in both ACI and tissue engineering, of hydroxyl, carboxyl and amide groups on PLLA films on chondro- biomaterials play very important roles. For instance, the second cyte growth.17 In a three-dimensional poly(ethylene glycol) (PEG) generation of ACI (so-called matrix-induced ACI, MACI) exploits hydrogel, the carboxylic group chemically tethered on PEG net- collagen as a cell delivery carrier to achieve better cell retention work was found to support the best chondrogenic differentiation and function.1 As a matter of fact, cells in vivo reside in a micro- of human MSCs among a series of small chemical groups.18 Most environment embodied with ECM, which plays important roles in recently, the design of surface chemistry to prevent cell adhesion regulating cellular fates. Hence, the design of biomaterial-based tis- and induce cell aggregation has been reported to promote chon- sue engineering scaffolds has been largely following the principle drogenesis of MSCs.19,20 For instance, by employing two polyelec- of mimicking native ECM, which are intended to provide a physi- trolytes with opposite charges, poly(l-glutamic acid) and chitosan, cochemical microenvironment for cell adhesion, growth, migration a hydrophilic surface in three-dimensional scaffolds was devel- and differentiation.9 It becomes very critical to gain understanding oped and aggregation of rabbit adipose-derived MSCs was encour- of cell–biomaterial interactions in developing optimal biomaterials aged, showing improved chondrogenic differentiation.20 All these to drive cells towards specific tissue formation, and recent studies studies demonstrate that by designing surface chemistry, cellular have, in fact, documented that topography, mechanics and chemis- behaviours can be modulated to obtain improved chondrogene- try of biomaterials scaffolds are all significant in modulating cellular sis. However, so far, a comparable evaluation of chondrogenesis of behaviours. different cell types in a same biomaterials platform has not been 10 In particular, intensive efforts have been devoted to obtaining performed yet.21 biomaterials with appropriate chemistries that being both con- Previously, we had developed a novel series of poly(ε-caprolactone) ductive and inductive to chondrogenesis. Incorporation of natu- polymers bearing different pendent chemical groups (i.e. hydrox- rally derived molecules (e.g., collagen and chondroitin sulphate) yl, methyl, carboxyl, amino and carbonyl) and successfully applied either by chemical attachment or by physical entrapment in syn- in studying the behaviours of MSCs.22 The objective of this study thetic polymer substrates (e.g., poly[l-lactic acid], PLLA) has been was to perform a comparative investigation on different chondro- extensively studied to improve the chondrogenesis of both chon- genic cells by employing such a biomaterials platform. As illustrated drocytes and MSCs. Essentially, controlling surface chemical in Fig. 1, MSCs and chondrocytes (primary and passaged cells) were properties, such as electrostatic charge, wettability, and hydro- seeded onto this series of polymeric films and chondrogenesis was phobicity of a substrate that recapitulates the basic characteristics evaluated. This study would deliver a comprehensive understanding of native ECM, represents another efficient route to biomaterials of how biomaterials chemistry can influence chondrogenesis and offer development. For example, chargeable molecules such as polydo- a novel perspective to design of biomaterials scaffolds for cartilage pamine and poly(acrylic acid) have been utilized to confer charges regeneration. FIGURE 1 11–14 Experimental design 514 | Chen et al. 2 | MATERIALS AND METHODS sterile distilled water for 3 days. Then, after incubation in serum-free 2.1 | PCL polymers and fabrication of thin films in either 24-well plates (for films of 15 mm in diameter) or 6-well The series of PCL polymers bearing different pendant functional groups were prepared through the ring-opening co-polymerization between ε-caprolactone and distinct co-monomers (feeding molar ratio: 9/1) using benzyl alcohol as initiator and Sn(Oct)2 as catalyst and characterized as described in our previous work.22 PCL polymers bearing -NH2, -CH3, -COOH, –OH and -C=O were designated as PCL-NH2, PCL-CH3, PCL-COOH, PCL-OH and PCL-C=O in the following context, respectively. To prepare polymeric thin films, these polymers were dissolved in methylene chloride (10 mg/mL) and spin coated onto round glass coverslips (15 or 35 mm in diameter) for 15 seconds at a speed of 4000 rpm on a benchtop spin coater (Beijing JinSheng WeiNa, China). Residual solvent was removed in a vacuum oven at room temperature for 24 hours and stored in a desiccator until use. growth medium for 10 hours, polymeric films were seeded with cells plates (for films of 35 mm in diameter). For cell adhesion, rACs (passage 1, P1) and rMSCs (P5) were seeded at 1 × 104 cells/cm2 onto films in 24-well plates in 1 mL of GM1 and GM2, respectively, and cultured for 24 hours. For cell growth, rACs (P1) and rMSCs (P5) were plated at 2.5 × 103 cells/cm2 onto films in 24-well plates in 1 mL of GM1 and GM2 and maintained for 21 days and 16 days, respectively. Medium was refreshed twice a week. To induce chondrogenesis, passaged rACs (P3 and P5) and rMSCs (P5) were initially plated at 5×104 cells/cm2 onto films in 24-well plates in GM1 and GM2 for 24 hours, respectively. Then, growth media were replaced with chondrogenic induction medium (CIM) consisting DMEM supplemented with 6.25 μg/mL insulin, 6.25 μg/mL transferrin, 6.25 μg/mL selenious acid, 1.25 mg/mL bovine serum albumin, 5.35 ng/mL linoleic acid, 100 μg/mL sodium pyruvate, 50 μg/mL l-ascorbic acid-2-phosphate, 0.35 mmol L−1 l-proline, 0.1 μmol L−1 dexamethasone and 10 ng/mL TGF-β1. Culture was maintained for 14 days and medium 2.2 | Cell isolation was changed twice a week. As control, P3 and P5 rACs and P5 rMSCs Rabbit ACs (rACs) and rabbit MSCs (rMSCs) were isolated as tively. To ensure a sufficient cell number for gene analysis, films of described in our previous work.23–25 All animal experiments were performed at Shanghai Laboratory Animal Center (Shanghai) in accordance with the institutional guidelines of animal care and use were also cultured for a total of 15 days in GM1 and GM2, respec35 mm in diameter in six-well plates were for cell culture. committee. New Zealand white rabbits aged 2–3 months were killed, 2.4 | F-actin staining and cartilage tissue samples were harvested from knee joints carti- To observe cell morphology on polymeric films, samples were lage tissue and bone marrow from femoral and tibial bones. To isolate chondrocytes, cartilage slices were digested with 0.1% collagenase II (200 U/mg; Invitrogen Carlsbad, California, USA). Harvested cells were then either cultured in chondrocyte growth medium (GM1) consisting of Dulbecco’s modified Eagle’s medium (Gibco Invitrogen, Carlsbad, California, USA) supplemented with 10% foetal bovine serum (FBS; Hyclone Logan, Utah, USA), 0.1 mmol L−1 nonessential amino acids, 0.4 mmol L−1 proline, 0.05 mg/mL vitamin C, 100 U/mL penicillin and 100 U/mL streptomycin at a cell density of 1.3 × 104 cells/cm2 for passaging or frozen down for future use. rMSCs were isolated from bone marrow using a density gradient method with Ficoll-Paque Plus solution (density: 1.077 g/mL; GE rinsed with phosphate buffered saline (PBS), fixed with 4% paraformaldehyde and permeated with 0.1% Triton-X 100 in PBS. After washing with PBS, samples were treated with Rhodaminephalloidin (70 nmol L−1; Invitrogen) in PBS containing 1% of BSA in dark for 20 minutes. Cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Fluorescence images were acquired on a fluorescence microscope (Eclipse Ti-S; Nikon ChiyodaKu, Tokyo, Japan). Cell spreading area per cell was determined using Image J software, and 4–5 images (200× magnification; 30–60 cells in total) were analysed for each group. Healthcare Pittsburgh, Pennsylvania, USA). Mononuclear cells col- 2.5 | Biochemical assays lected were initially plated at 1×105 cells/cm2 in growth medium Cell number on polymeric films was assessed using cell counting kit- (GM2), consisting of α-Minimum Essential Medium (α-MEM; Gibco) 8 (CCK-8, Dojindo, Japan). Cell-seeded polymeric films in 24-well supplemented with 10% FBS, 100 U/mL penicillin and 100 U/mL plates were rinsed with PBS; 0.5 mL of growth medium supplement- streptomycin in a humidified atmosphere of 5% CO2 at 37°C and ed with 50 μL of CCK-8 stock solution was added to each well and then subcultured at 5 × 103 cells/cm2. rMSCs were validated to be CD29+, CD44+, CD14− and CD45− and demonstrated the capability of differentiation into chondrocytes, osteoblasts and adipocytes (data not shown). incubated for 1 hour at 37°C. Culture wells without cells were added with CCK-8 solution and set as blank. The absorbance at 450 nm (optical density, OD450) was measured with a reference wavelength of 620 nm on a microplate reader (Bio-Tek Instrument Winooski, Vermont, USA). 2.3 | Cell culture on polymeric films Prior to cell seeding, polymeric films fixed on coverslips were exposed to UV light for 30 minutes for sterilization and then incubated in To quantify glycosaminoglycans (GAG), samples were treated with papain solution (125 μg/mL papain, 16–40 U/mg; Sigma St. Louis, Missouri, USA), 5 mmol L−1 −1 5 mmol L l-cysteine, 100 mmol L−1 Na2HPO4, EDTA and pH 6.2) at 60°C overnight. Contents of DNA | Chen et al. and GAG were measured by using Hoechst 33258 dye and dimethyl methylene blue (DMMB) method as described,23 respectively, and GAG content was normalized to DNA content as GAG/DNA (μg/μg). All biochemical assays were performed in triplicates. 2.6 | Histology 515 2.8. | Statistical analysis Values were expressed as mean ± standard deviation. Statistical analysis was performed with the SPSS 19.0 software. Comparisons among groups were performed using one-factor analysis of variance (ANOVA) and Tukey post hoc test. Statistical significance was set at P<.05. Deposition of GAG by cells on polymeric films was detected with both alcian blue and safranin O staining. Samples were rinsed with PBS, 3 | RESULTS fixed with 4% paraformaldehyde in PBS, stained with 1% alcian blue (Sigma) in 3% (v/v) acetic acid for 30 minutes or 0.5% safranin O in distilled water for 5 minutes at room temperature, and finally rinsed with distilled water. Images were acquired under a phase contrast microscope. 3.1 | Polymeric films Surface properties of the polymeric films were reported in our previous study.22 All films displayed flat, smooth surface under scanning electron microscope. However, the order of water contact angle of PCL 2.7 | Reverse transcription-polymerase chain reaction (RT-PCR) Cells on polymeric films (35 mm in diameter) were lysed with Trizol reagent (Invitrogen), and total RNA was extracted following the manufacturer’s protocol. cDNA was prepared with Superscript II Reverse Transcriptase (Invitrogen) using oligo(dT) and then amplified using PCR Master Mix (TaKaRa Kusatsu, Shiga, Japan) on a thermal cycler (TC-XP-A; Bioer, China Hangzhou). PCR reactions were performed as follows: 94°C for 10 minutes, 35 cycles of PCR (95°C for 15 seconds, 57°C for 45 seconds and 72°C for 45 seconds). PCR products were analysed by using 2% agarose gel electrophoresis and visualized under UV after ethidium bromide staining. Gapdh was used as a housekeeping gene and primer sequences were listed in Table S1. (87.6°) > PCL-CH3 (85.8°) > PCL-C=O (80.5°) > PCL-COOH (73.9°) ≈ PCL-OH (73.9°) > PCL-NH2 (72.8°), indicating different hydrophilicity. The total protein amount adsorbed on these films varied, with PCL, PCL-CH3 and PCL-C=O films having the highest protein contents and PCL-OH film the lowest. These together implied the distinct surface chemistries of these polymeric films. 3.2 | Cell adhesion on polymeric films Adhesion of both P1 rACs and P5 rMSCs on this series of polymeric films within 24 hours was assessed with F-actin staining and CCK8 assay. According to F-actin staining, P1 rACs displayed a small rounded or polygonal shape on all films (Fig. 2a) and cell spreading area showed no significant difference among different films (Fig. 2b). Based on CCK-8 assay, a significantly higher absorbance was on PCL tethered with small chemical groups than PCL film, indicating a higher (a) (b) (c) F I G U R E 2 Adhesion of P1 rACs on polymeric films. P1 rACs were seeded at 1 × 104 cells/cm2 onto the films in GM1 and cultured for 24 h. (a) F-actin staining, (b) cell-spreading area (NS: not significant; n>30) and (c) CCK-8 assay (*: P<.05; n=4) 516 | Chen et al. cell number, and PCL-C=O film further showed slightly higher absorb- slight different curves were noticed. Although both cells gradually ance than PCL-NH2, PCL-CH3 and PCL-OH films (Fig. 2c). proliferated over time on all the films following a sigmoid curve, rACs For P5 rMSCs, according to F-actin staining images after dif- displayed varied growth rates before reaching the plateau on day 18 ferent time intervals, cells gradually adhered onto all the films and rMSCs grew all the way towards the maximum on day 12. Since within 24 hours, developed mature stress fibres and transformed P1 rACs and P5 rMSCs were seeded at a same cell density, consider- from a small rounded shape into a large elongated spindle- like ing a higher cell amount for rACs on day 12, P1 rACs grew much fast one (Fig. 3a). Notably, rMSCs tended to be smaller on PCL and than P5 rMSCs on all polymeric films. Additionally, cell growth on PCL PCL-CH3 films than others, which was confirmed by quantifying film was lower than that on the films of PCL derivatives bearing small cell spreading area (Fig. 3b). The adhesion kinetics of rMSCs was chemical groups, especially for rMSCs. determined based on CCK-8 assay, displayed a gradually increasing trend and reached a plateau within 24 hours on all the films (Fig. 3c). It was also found that cell adhesion on PCL and PCL- CH3 films was consistently lower than that on other films at all tested time points. 3.4 | Long-term maintenance of P1 rACs on polymeric films P1 rACs were cultured on these polymeric films for 3 weeks in GM1. Cell morphology was evaluated by F-actin staining. While cells 3.3 | Cell growth on polymeric films remained rounded on all the films on day 7, the spindle-like cells could be apparently discerned on day 21 only on PCL and PCL-CH3 Growth of both P1 rACs and P5 rMSCs on PCL films was plotted films, based on the aligned organization of F-actin (Fig. 5a). Safranin against culture time as shown in Fig. 4. For these two types of cells, O and alcian blue staining were applied to detect the production of (a) (b) (c) F I G U R E 3 Adhesion of rMSCs on polymeric films. P5 rMSCs were seeded at 1 × 104 cells/cm2 onto the films and cultured for 24 h. (a) Factin staining at 3 h, 9 h and 24 h, (b) cell spread area at 24 h (*: P<.05; n>30) and (c) adhesion kinetics within 24 h | Chen et al. (a) 517 (b) F I G U R E 4 Cell growth kinetics on polymeric films. P1 rACs (a) and P5 rMSCs (b) were seeded at 2.5 × 103 cells/cm2 onto the films and cultured for 21 and 16 days in GM1 and GM2, respectively (*: P<.05; n=4). cartilaginous ECM component, GAG. As shown in Fig. 5a, the positive and were barely stained by the two dyes. Conversely, in CIM, P3 rACs staining of safranin O was observed on all the films, but with different became rounded and were positively stained with both safranin O and intensities, and the intensity on each film tended to vary over time. alcian blue, and cellular aggregates could be noticed. For P5 rACs in On day 21, less staining intensity was found on PCL and PCL-CH3 CIM, some cells were induced to be round, and both dyes were noted films than that on others. Additionally, alcian blue staining followed on all the films, however, with less intensities compared with those the same trend as the safranin O staining (Fig. 5a). for P3 rACs. Quantification of GAG/DNA was shown in Fig. 5b, and P1 rACs DNA content and GAG/DNA were quantified as shown in Fig. 6b. cultured on 24-well tissue culture plates (TCP) for 3 weeks were For P3 rACs, no significant difference in DNA content was found on included as a control. DNA contents on all the substrates were not all the films in both GM1 and CIM. For P5 rACs, compared with PCL significantly different on day 21, and slight reduction was noted after and PCL-CH3 films, slightly higher DNA content was obtained on oth- 2 weeks on both TCP and PCL film, which was in contrast to the gradu- er films in both media. In addition, P5 rACs in CIM had higher DNA al increasing trend on other films. GAG/DNA increased within 2 weeks contents than P5 rACs in GM1 and P3 rACs in both GM1 and CIM on on polymeric films and rose afterwards only on PCL and PCL-COOH all the films. For both P3 and P5 rACs, GAG/DNA was trivial in GM1 films, while remaining steady on others. However, GAG/DNA on PCL and significantly increased in CIM. In addition, GAG/DNA was much and PCL-CH3 films was relatively lower than on other films with the higher for P3 rACs in CIM than P5 rACs in CIM on all respective films. highest value achieved on PCL-COOH film. In contrast, GAG/DNA on While GAG/DNA was found similar on all the films for P3 rACs in both TCP decreased over time and remained much lower than that on poly- GM1 and CIM, P5 rACs on PCL-CH3 film had a significantly higher meric films. GAG/DNA value than those on PCL-NH2, PCL-COOH, PCL-OH and Gene expression was evaluated and shown in Fig. 5c. On day 0, PCL-C=O films. P1 rACs on TCP expressed hyaline cartilage-specific genes, including Gene expression was shown in Fig. 6c. While P3 rACs cultured in Sox9, Acan and Col2a1, and fibrocartilage-specific genes Col1a1. After GM1 only expressed a high level of collagen I on all the films, P3 rACs 1 week on polymeric films, cells expressed Sox9, Col2a1 and Col1a1 were induced to express Sox9, Col2a1, Acan and Col1a1 in CIM, with on all the films and aggrecan was only detected on PCL-NH2 and PCL- slightly lower levels on PCL and PCL-CH3 films. Expression of Sox9 and COOH films. After 2 weeks, only on PCL and PCL-NH2 films, Sox9, Acan could be detected for P5 rACs in CIM, also with slightly lower Col2a1 and Col1a1 were expressed and Acan became disappeared on levels on PCL and PCL-CH3 films. all the films. However, after 3 weeks, expression of Sox9, Col2a1 and Col1a1 was upregulated on the films, except PCL-NH2 and PCL-C=O films, and expression of Acan still remained undetectable on all the films. 3.6 | Chondrogenic differentiation of rMSCs on polymeric films P5 rMSCs were cultured in CIM for 2 weeks and were also cultured 3.5 | Re-differentiation of passaged rACs on polymeric films in GM2 for 15 days as a negative control. In Fig. 7a, safranin O and alcian blue staining demonstrated that cells formed aggregates in CIM, with GAG deposited on all the substrates including TCP, while cells Both P3 and P5 rACs were cultured on polymeric films in CIM for in GM2 remained in monolayer with only trace staining. As shown 2 weeks, and these cells cultured in GM1 were included as controls. in Fig. 7b, DNA contents were found much lower in CIM than those Safranin O and alcian blue staining were shown in Fig. 6a. In GM1, in GM2 on all substrates including TCP and followed a similar order both P3 and P5 rACs displayed a fibroblast-like shape on all the films among different substrates with the highest obtained on PCL-COOH 518 | Chen et al. (a) (b) (c) F I G U R E 5 Phenotypic characterization and gene analysis of P1 rACs after a long-term culture on polymeric films. P1 rACs were plated at 2.5×103 cells/cm2 onto the films and cultured in GM1 for 21 days. (a) F-actin, safranin O and alcian blue staining, (b) GAG/DNA and DNA content (n=4) and (c) gene expression | Chen et al. 519 (a) (b) (c) F I G U R E 6 Re-differentiation of passaged rACs on polymeric films. P3 and P5 rACs were seeded at 5 × 104 cells/cm2 onto the films in GM1 for 24 h and then cultured in CIM for 14 days. P3 and P5 rACs were also cultured in GM1 for 15 days as control. (a) Safranin O and alcian blue staining, (b) GAG, DNA and GAG/DNA (*: P<.05; n=4) and (C) gene expression | 520 Chen et al. and PCL-C=O films. In GM2, only trivial GAG/DNA was detected on 2 weeks, with only Sox9 and Col2a1 detected on PCL and PCL-NH2 all substrates. However, upon chondrogenic induction, significantly films. However, a slight upregulation of Sox9 and Col2a1 was pres- higher GAG/DNA values (15–30 μg/μg) were achieved on all the pol- ent after 3 weeks, especially on PCL, PCL-CH3 and PCL-COOH films. ymeric films, with the highest ones on PCL and PCL-NH2 films, which For passaged rACs, re-differentiation could be induced on these films were also comparable to those produced by P3 rACs in CIM (Fig. 6b), upon chondrogenic induction for both P3 and P5 cells, but to differ- suggesting an efficient chondrogenic differentiation. ent extents, with P3 cells having much higher GAG production and chondrocytic gene expression. P3 rACs had similar GAG/DNA on all 4 | DISCUSSION the films and also slightly higher gene expression on hydrophilic films, especially PCL-COOH and PCL-C=O films. For P5 rACs, polymeric films bearing hydrophilic groups, especially PCL-NH2, promoted growth and Comparison among different stem cells has been explored concerning upregulation of Sox9 and Acan in chondrogenic condition and PCL-CH3 chondrogenic potential previously.21 However, a systemic evaluation film supported better GAG production. For rMSCs, cell adhesion was of different cells on a biomaterials platform would confer an insight found least on both PCL and PCL-CH3 films, and all the films bearing perspective to biomaterials design for cartilage tissue engineering. In pendant chemical groups were superior in supporting growth com- our previous study, a series of PCL polymer derivatives bearing differ- pared with PCL film. In chondrogenic condition, rMSCs showed slightly ent pendant chemical groups (i.e. -NH2, -CH3, -COOH, -OH and -C=O) more efficient chondrogenic differentiation on both PCL and PCL-NH2 were successfully synthesized and could be conveniently processed films. Collectively, these results suggested that different cells displayed into thin, flat films for cell culture.22 Such a polymer series can serve as distinct behaviours on this series of polymeric substrates. a potential platform for studying cell–biomaterial interactions. To apply chondrocytes for therapeutic applications, in vitro expan- In this study, an investigation was performed to uncover how sion to obtain a sufficient amount is necessary, which causes dediffer- substrate surface chemistry can influence chondrogenesis of differ- entiation inevitably, and thus, a variety of strategies, including growth ent cell types by exploiting such a series of polymeric films. For P1 factor supplementation, three-dimensional and/or dynamic culture rACs, while all the films supported both adhesion and proliferation, and biomaterials design, have been explored to promote the chondro- PCL films bearing hydrophilic groups (-NH2, -COOH, -OH and -C=O), cytic phenotype during in vitro culture.23,26–30 It has been generally especially those with charges (-NH2 and -COOH), favoured cartilag- considered that expansion of chondrocytes beyond 4 passages leads inous ECM production (i.e. GAG). However, in general, gene expres- to irreversible dedifferentiation.29 Hence, in this study, P1, P3 and P5 sion displayed a dynamic variation on polymeric films within 3 weeks. rACs were all evaluated on this series of polymeric films to investigate Within 1 week, all the films supported expression of Sox9 and Col2a1, both maintenance and induction of re-differentiation of chondrocytes. and Acan only expressed on charged films (PCL-NH2 and PCL-COOH). For P1 rACs in GM1, it might be possible that more hydrophilic films, Significant downregulation of chondrocytic genes was observed after especially charged ones, were beneficial to the accumulation of GAG, (a) (b) F I G U R E 7 Chondrogenic differentiation of rMSCs on polymeric films. P5 rMSCs were seeded at 5 × 104 cells/cm2 onto the films and TCP in GM2 for 24 h and then cultured in CIM for 14 days. P5 rMSCs were also cultured in GM2 for 15 days as a control. (a) Safranin O and alcian blue staining and (b) GAG, DNA and GAG/DNA (*: P<.05; n=8) | Chen et al. 521 which potentially favoured chondrocytic gene expression at the early films, which mimics the condensation event during early cartilage stage, and more hydrophobic films tended to favour gene expression development.38 In essence, the balance between cell–cell and cell– at the late stage in culture, suggesting the potential in maintaining substrate interactions becomes critical, with a strong adhesion on chondrocytic phenotype in long-term in vitro culture. Several stud- substratum being unfavourable in conducting chondrogenesis.30,39 ies have shown that an enhancement in wettability of a hydrophobic However, this balance can be sensitive to the differentiation status of substrate, such as PCL, poly(l-lactic acid) and polystyrene, is beneficial the cells, which further varies dynamically during culture process. For for growth and GAG production of early passages of chondrocytes, example, a dynamic control over cell adhesion in a hydrogel has been which is in consistent with our observation with P1 rACs.11,16,17,31,32 demonstrated to promote the chondrogenic differentiation of human However, while Ma et al.17 demonstrated that hydroxyl (-OH) and MSCs.40 This might explain why different cells displayed different amide (-CONH2) on poly(l-lactic acid) substrates were more favour- behaviours on this series of polymeric films. Specifically, a weak able to chondrocyte growth than -COOH, the best GAG production adhesion on substratum not only favours the formation of cellular was found on PCL-COOH film among the series in this study. aggregates but also maintains cells in a rounded morphology, which So far, few studies have evaluated the effects of substrate surface intrinsically stimulates chondrogenesis.41 This is consistent with the properties on re-differentiation of passaged chondrocytes. In this study, fact that PCL and PCL-CH3 films, which were less cell adhesive com- for P3 and P5 rACs in CIM, induction of re-differentiation was possible pared to more hydrophilic films, were better in inducing chondrocytic on all the films, but to a less extent for P5 cells. It appeared that more gene expression for P1 rACs at the late stage and ECM deposition hydrophilic films supported better chondrocytic gene expression for of both P5 rACs and rMSCs. However, besides from the balanced both cells. But, less impact was observed for surface chemistry on GAG hydrophobicity/hydrophilicity, surface charges might represent an production, which was possibly due to the short culture period (14 days). additional instructive cue for cells to undergo chondrogenesis, by In addition, presentation of -NH2 on the hydrophobic PCL substratum recapitulating the essential characteristics of native cartilage tissue seemed to support better expression of Sox9 than other hydrophilic matrix (i.e. GAG).18,33 This is evidenced by the fact that PCL-NH2 or groups for inducing re-differentiation of P5 rACs. However, this is in PCL-COOH promoted either ECM production of P1 rACs or chondro- contrast to the study by Yang et al.30 wherein neutral or low negatively cytic gene expression of P5 rACs. Collectively, this may suggest the charged surfaces induced more efficient spontaneous re-differentiation compounding mechanisms that mediate cell–substrate interactions of P7 human ACs on a hydrophilic gel than positively charged one. during chondrogenesis. Chondrogenesis of MSCs has also been recognized to be affected by substrate surface chemistry.15,18–20,33–35 In this study, while rMSCs formed aggregates on all the films, both positively charged 5 | CONCLUSIONS (PCL-NH2) and hydrophobic (PCL) surfaces seemed to be favourable for chondrogenesis-based GAG quantification. This finding is partial- In this study, a PCL-based polymer series bearing distinct pendant ly in agreement with the study by Guo et al.15 wherein polystyrene chemical groups was employed to study the effects of surface chemis- surfaces modified with polyallylamine (containing -NH2) supported try on chondrogenesis of different cell types. It was found that surface better chondrogenic differentiation than that with poly(acrylic acid) chemistry is influential on chondrogenesis in vitro. Importantly, differ- (with -COOH), which instead promoted growth of MSCs. In contrast, ent cell types demonstrated distinct behaviours on such a biomateri- in several other studies, -COOH was claimed to be beneficial for als platform concerning chondrogenesis, implicating a developmental chondrogenesis rather than -NH2.18,33,34 It is possible that different dependency on surface chemistry. These findings confer an insight substrates applied (hydrophilic glass16,17 and poly(ethylene glycol)18 versus hydrophobic polystyrene 15 perspective to designing biomaterials for cartilage tissue regeneration. and PCL in this study might have caused such variation. Notwithstanding, Phillips et al.36 found no difference in chondrogenesis of human MSCs on self-assembled monolayers bearing different terminal chemical groups. AC KNOW LEDG EM ENTS This work was financially supported by the National Natural Science Based on the above analysis, one important finding is that sub- Foundation of China (31000424 and 31170951) and the Basic strate surface chemistry had differential effects on chondrogenesis Research Project of Shanghai Science and Technology Commission of P1, P3, P5 rACs and P5 rMSCs. Moreover, P1 rACs displayed the (12JC1403101 and 15JC1401402). stage-specific dependence on surface chemistry during the long-term culture. Since these cells represent distinct differentiation status, these observations imply a possibly developmental stage-dependent mechanism for cell–biomaterial interactions concerning chondrogenesis, which is significant in devising strategies for cartilage regeneration. There is a view that dedifferentiated chondrocytes upon in vitro serial expansion might resemble certain biological characteristics of MSCs.37 Hence, both passaged rACs and rMSCs tended to form cellular aggregates upon chondrogenic induction on all the polymeric REFERENCES 1. Redman SN, Oldfield SF, Archer CW. Current strategies for articular cartilage repair. Eur Cell Mater. 2005;9:23–32. 2. Madeira C, Santhagunam A, Salgueiro JB, Cabral JM. Advanced cell therapies for articular cartilage regeneration. Trends Biotechnol. 2015;33:35–42. 3. Huey DJ, Hu JC, Athanasiou KA. Unlike bone, cartilage regeneration remains elusive. Science. 2012;338:917–921. 522 | 4. Bornes TD, Adesida AB, Jomha NM. Mesenchymal stem cells in the treatment of traumatic articular cartilage defects: a comprehensive review. Arthritis Res Ther. 2014;16:432. 5. Hellingman CA, Koevoet W, van Osch GJ. Can one generate stable hyaline cartilage from adult mesenchymal stem cells? A developmental approach. J Tissue Eng Regen Med. 2012;6:e1–e11. 6. Pelttari K, Winter A, Steck E, et al. Premature induction of hypertrophy during in vitro chondrogenesis of human mesenchymal stem cells correlates with calcification and vascular invasion after ectopic transplantation in SCID mice. Arthritis Rheum. 2006;54:3254–3266. 7. Erickson IE, Huang AH, Chung C, Li RT, Burdick JA, Mauck RL. Differential maturation and structure-function relationships in mesenchymal stem cell- and chondrocyte-seeded hydrogels. Tissue Eng Part A. 2009;15:1041–1052. 8. Huang AH, Yeger-McKeever M, Stein A, Mauck RL. Tensile properties of engineered cartilage formed from chondrocyte- and MSC-laden hydrogels. Osteoarthr Cartil. 2008;16:1074–1082. 9. Lutolf MP, Blau HM. Artificial stem cell niches. Adv Mater. 2009;21:3255–3268. 10. Keung AJ, Kumar S, Schaffer DV. Presentation counts: microenvironmental regulation of stem cells by biophysical and material cues. Annu Rev Cell Dev Biol. 2010;26:533–556. 11. Cui YL, Qi AD, Liu WG, et al. Biomimetic surface modification of poly(L-lactic acid) with chitosan and its effects on articular chondrocytes in vitro. Biomaterials. 2003;24:3859–3868. 12. Hsu SH, Chang SH, Yen HJ, Whu SW, Tsai CL, Chen DC. Evaluation of biodegradable polyesters modified by type II collagen and Arg-GlyAsp as tissue engineering scaffolding materials for cartilage regeneration. Artif Organs. 2006;30:42–55. 13. Ma Z, Gao C, Gong Y, Shen J. Cartilage tissue engineering PLLA scaffold with surface immobilized collagen and basic fibroblast growth factor. Biomaterials. 2005;26:1253–1259. 14. Varghese S, Hwang NS, Canver AC, Theprungsirikul P, Lin DW, Elisseeff J. Chondroitin sulfate based niches for chondrogenic differentiation of mesenchymal stem cells. Matrix Biol. 2008;27:12–21. 15. Guo L, Kawazoe N, Fan Y, et al. Chondrogenic differentiation of human mesenchymal stem cells on photoreactive polymer-modified surfaces. Biomaterials. 2008;29:23–32. 16. Tsai WB, Chen WT, Chien HW, Kuo WH, Wang MJ. Poly(dopamine) coating of scaffolds for articular cartilage tissue engineering. Acta Biomater. 2011;7:4187–4194. 17. Ma Z, Gao C, Gong Y, Shen J. Chondrocyte behaviors on poly-Llactic acid (PLLA) membranes containing hydroxyl, amide or carboxyl groups. Biomaterials. 2003;24:3725–3730. 18. Benoit DS, Schwartz MP, Durney AR, Anseth KS. Small functional groups for controlled differentiation of hydrogel-encapsulated human mesenchymal stem cells. Nat Mater. 2008;7:816–823. 19. Goldshmid R, Cohen S, Shachaf Y, et al. Steric interference of adhesion supports in-vitro chondrogenesis of mesenchymal stem cells on hydrogels for cartilage repair. Sci Rep. 2015;5:12607. 20. Zhang K, Yan S, Li G, Cui L, Yin J. In-situ birth of MSCs multicellular spheroids in poly(L-glutamic acid)/chitosan scaffold for hyaline-like cartilage regeneration. Biomaterials. 2015;71:24–34. 21. Seda Tigli R, Ghosh S, Laha MM, et al. Comparative chondrogenesis of human cell sources in 3D scaffolds. J Tissue Eng Regen Med. 2009;3:348–360. 22. Chen M, Zhang Y, Zhou Y, et al. Pendant small functional groups on poly(-caprolactone) substrate modulate adhesion, proliferation and differentiation of human mesenchymal stem cells. Colloids Surf B Biointerfaces. 2015;134:322–331. 23. Xu F, Xu L, Wang Q, Ye Z, Zhou Y, Tan WS. 3D dynamic culture of rabbit articular chondrocytes encapsulated in alginate gel beads using spinner flasks for cartilage tissue regeneration. Biomed Res Int. 2014;2014:539789. Chen et al. 24. Xu FY, Xu L, Wang Q, Zhou Y, Ye ZY, Tan WS. A three-dimensional dynamic coculture system enabling facile cell separation for chondrogenesis of mesenchymal stem cells. Biochem Eng J. 2015;103:68–76. 25. Xu L, Wang Q, Xu F, Ye Z, Zhou Y, Tan WS. Mesenchymal stem cells downregulate articular chondrocyte differentiation in noncontact coculture systems: implications in cartilage tissue regeneration. Stem Cells Dev. 2013;22:1657–1669. 26. Jakob M, Demarteau O, Schafer D, et al. Specific growth factors during the expansion and redifferentiation of adult human articular chondrocytes enhance chondrogenesis and cartilaginous tissue formation in vitro. J Cell Biochem. 2001;81:368–377. 27. Lee TJ, Bhang SH, La WG, et al. Spinner-flask culture induces redifferentiation of de-differentiated chondrocytes. Biotechnol Lett. 2011;33:829–836. 28. Schrobback K, Klein TJ, Schuetz M, Upton Z, Leavesley DI, Malda J. Adult human articular chondrocytes in a microcarrier-based culture system: expansion and redifferentiation. J Orthop Res. 2011;29:539–546. 29. Schulze-Tanzil G, de Souza P, Villegas Castrejon H, et al. Redifferentiation of dedifferentiated human chondrocytes in high-density cultures. Cell Tissue Res. 2002;308:371–379. 30. Yang JJ, Chen YM, Liu JF, Kurokawa T, Gong JP. Spontaneous redifferentiation of dedifferentiated human articular chondrocytes on hydrogel surfaces. Tissue Eng Part A. 2010;16:2529–2540. 31. Ma Z, Gao C, Shen J. Surface modification of poly-L-lactic acid (PLLA) membrane by grafting acrylamide: an effective way to improve cytocompatibility for chondrocytes. J Biomater Sci Polym Ed. 2003;14:13–25. 32. Tsai WB, Wei TC, Lin MC, Wang JY, Chen CH. The effect of radiofrequency glow discharge treatment of polystyrene on the behavior of porcine chondrocytes in vitro. J Biomater Sci Polym Ed. 2005;16:699–714. 33. Curran JM, Chen R, Hunt JA. Controlling the phenotype and function of mesenchymal stem cells in vitro by adhesion to silane-modified clean glass surfaces. Biomaterials. 2005;26:7057–7067. 34. Curran JM, Chen R, Hunt JA. The guidance of human mesenchymal stem cell differentiation in vitro by controlled modifications to the cell substrate. Biomaterials. 2006;27:4783–4793. 35. Gross-Aviv T, Vago R. The role of aragonite matrix surface chemistry on the chondrogenic differentiation of mesenchymal stem cells. Biomaterials. 2009;30:770–779. 36. Phillips JE, Petrie TA, Creighton FP, Garcia AJ. Human mesenchymal stem cell differentiation on self-assembled monolayers presenting different surface chemistries. Acta Biomater. 2010;6:12–20. 37. Tallheden T, Dennis JE, Lennon DP, Sjogren-Jansson E, Caplan AI, Lindahl A. Phenotypic plasticity of human articular chondrocytes. J Bone Joint Surg Am. 2003;85-A(suppl 2):93–100. 38. Bobick BE, Chen FH, Le AM, Tuan RS. Regulation of the chondrogenic phenotype in culture. Birth Defects Res C Embryo Today. 2009;87:351–371. 39. Woodfield TB, Miot S, Martin I, van Blitterswijk CA, Riesle J. The regulation of expanded human nasal chondrocyte re-differentiation capacity by substrate composition and gas plasma surface modification. Biomaterials. 2006;27:1043–1053. 40. Salinas CN, Anseth KS. The enhancement of chondrogenic differentiation of human mesenchymal stem cells by enzymatically regulated RGD functionalities. Biomaterials. 2008;29:2370–2377. 41. Rottmar M, Mhanna R, Guimond-Lischer S, Vogel V, Zenobi-Wong M, Maniura-Weber K. Interference with the contractile machinery of the fibroblastic chondrocyte cytoskeleton induces re-expression of the cartilage phenotype through involvement of PI3K, PKC and MAPKs. Exp Cell Res. 2014;320:175–187. SUP P O RT I NG I NFO R M AT I O N Additional Supporting Information may be found online in the supporting information tab for this article.