Materials Science and Engineering C 43 (2014) 290–299

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Embroidered polymer–collagen hybrid scaffold variants for ligament tissue engineering M. Hoyer a,b, N. Drechsel c, M. Meyer c, C. Meier a, C. Hinüber d,e, A. Breier d, J. Hahner d,e, G. Heinrich d,e, C. Rentsch f, L.-A. Garbe b, W. Ertel a, G. Schulze-Tanzil a,⁎,1, A. Lohan a,1 a

Department of Orthopaedic, Trauma and Reconstructive Surgery, Charité-Universitätsmedizin Berlin, Campus Benjamin Franklin, Garystrasse 5, 14195 Berlin, Germany Department of Bioanalytics, Technische Universität Berlin Seestr. 13, 13353 Berlin, Germany Research Institute of Leather and Plastic Sheeting-FILK, Meißner Ring 1–5, 09599 Freiberg, Germany d Department of Materials and Processing, Leibniz-Institut für Polymerforschung Dresden e.V., Hohe Strasse 6, 01069 Dresden, Germany e Institute of Material Science, Technische Universität Dresden Helmholtzstraße 10, 01062 Dresden, Germany f Department of Trauma and Reconstructive Surgery, University Hospital Carl Gustav Carus, Fetscherstrasse 74, 01307 Dresden, Germany b c

a r t i c l e

i n f o

Article history: Received 8 March 2014 Received in revised form 3 May 2014 Accepted 2 July 2014 Available online 9 July 2014 Keywords: Anterior cruciate ligament Ligament tissue engineering Polylactide-co-caprolactone Polydioxanone Embroidered scaffold Collagen

a b s t r a c t Embroidery techniques and patterns used for scaffold production allow the adaption of biomechanical scaffold properties. The integration of collagen into embroidered polylactide-co-caprolactone [P(LA-CL)] and polydioxanone (PDS) scaffolds could stimulate neo-tissue formation by anterior cruciate ligament (ACL) cells. Therefore, the aim of this study was to test embroidered P(LA-CL) and PDS scaffolds as hybrid scaffolds in combination with collagen hydrogel, sponge or foam for ligament tissue engineering. ACL cells were cultured on embroidered P(LA-CL) and PDS scaffolds without or with collagen supplementation. Cell adherence, vitality, morphology and ECM synthesis were analyzed. Irrespective of thread size, ACL cells seeded on P(LA-CL) scaffolds without collagen adhered and spread over the threads, whereas the cells formed clusters on PDS and larger areas remained cell-free. Using the collagen hydrogel, the scaffold colonization was limited by the gel instability. The collagen sponge layers integrated into the scaffolds were hardly penetrated by the cells. Collagen foams increased scaffold colonization in P(LA-CL) but did not facilitate direct cell-thread contacts in the PDS scaffolds. The results suggest embroidered P(LA-CL) scaffolds as a more promising basis for tissue engineering an ACL substitute than PDS due to superior cell attachment. Supplementation with a collagen foam presents a promising functionalization strategy. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The anterior cruciate ligament (ACL) rupture is the most frequent ligament injury of the knee joint [1], but the ACL is not able to selfregenerate [2,3]. The ruptured tissue is usually surgically reconstructed to restore joint stability [4]. Ligament reconstruction based on autologous tendons such as Musculus (M.) semitendinosus, M. gracilis or patellar tendons is associated with donor site morbidity [5] and restricted by

Abbreviations: ACL, anterior cruciate ligament; DAPI, 4',6-diamidino-2-phenylindole; DMEM, Dulbecco’s modified Eagle’s medium; DMMB, dimethyl methylene blue; ECM, extracellular matrix; EtBr, ethidium bromide; FCS, fetal calf serum; FDA, fluorescein diacetate; HE, hematoxylin and eosin staining; HMDS, hexamethyldisilazane; P(LA-CL), polylactide-co-caprolactone; PBS, phosphate buffered saline; PDS, polydioxanone; PFA, paraformaldehyde; PVA, polyvinyl alcohol; SEM, scanning electron microscopy; RT, room temperature; sGAG, sulfated glycosaminoglycan; TBS, TRIS buffered saline. ⁎ Corresponding author at: Dep. of Orthopaedic, Trauma and Reconstructive Surgery, Charité-Universitätsmedizin Berlin, Campus Benjamin Franklin, FEM, Garystrasse 5, 14195 Berlin, Germany. Tel.: +49 30 450 552385; fax: +49 30 450 552985. E-mail address: [email protected] (G. Schulze-Tanzil). 1 Joined senior authorship.

http://dx.doi.org/10.1016/j.msec.2014.07.010 0928-4931/© 2014 Elsevier B.V. All rights reserved.

autograft availability. Rathbone et al. showed that many surgeons would prefer a tissue engineering based ACL substitute compared with actual standard therapies for the case that the construct meets the biomechanical requirements of the ACL [6]. A biomaterial suitable for ACL reconstruction should possess properties such as high biocompatibility, suitable biomechanics and durability. Both synthetic materials used in this study, a copolymer from 70% polylactide and 30% polycaprolactone (polylactide-co-caprolactone: P [LA-CL]) and polydioxanone (PDS) are in clinical use as biocompatible, slow degrading suture materials [7–10] and have a suitable flexibility needed for the embroidery process. Both polymers are aliphatic polyesters that are degradable by hydrolysis [11] and are therefore suitable for ligament tissue engineering [7,12–14]. Two dimensional substrates of P(LA-CL) manufactured in different ratios were already tested with lapine ACL cells and revealed a biocompatibility comparable with commercially available cell culture plastic [12,15]. PDS has been successfully implanted in large tendon defects in rabbits [16] and could therefore also be promising for the ACL tissue engineering. Embroidery techniques allow the design of appropriate biomechanical properties of the scaffolds. Adaption of the embroidery patterns can

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

modulate the shape, pore sizes, stiffness and elasticity of the scaffolds [17]. Here mechanical requirements of the ACL are mimicked by the zig-zag embroidery pattern along the longitudinal axis so that the meshes can be extended during tension. In spite of the favorable properties of embroideries, to our knowledge this is the first study using embroidered hybrid scaffolds for ligament tissue engineering with primary ligament cells. Since the biomechanics of native ligaments depend on the parallel aligning of the collagen fibril bundles along the longitudinal axis of the tissue [12] and a wavelike extracellular matrix (ECM) texture in relaxed ligaments due to the presence of elastic fibers, a unidirectional zig-zag embroidery pattern of the scaffolds was selected for the present study. Also wrapping of the collagen fiber bundles by the ligament cells observed in ligaments and tendons could be mimicked by the ACL cells growing along the longitudinal threads [12]. With the aim to create a preformed tissue for the ACL reconstruction, the adherence of cells to the synthetic polymers P(LA-CL) and PDS has to be improved using functionalization techniques [18]. The embroidered structures were combined with collagen materials manufactured according to different preparation techniques. Collagen is the main component of connective tissue. The protein chains are twisted into triple helices, the latter in tissues being structured in microfibrils, fibers and fiber bundles [19]. Weak immunogenic reactions are observed, both humoral and the cell mediated, depending on the donor and the recipient species and reactions against antigens located in the triple helical as well as the telopetidal region [20]. For the clinical use triple helical collagen molecules may be assessed as immunocompatible [9,21] and the use of collagen may improve the biocompatibility of other implant materials. Furthermore, collagen is compatible with synthetic polymers [9] and the combination of both will result in hybridimplants [22], which are more stable than solely collagen-based scaffolds [12,23]. It is assumed that the combination of synthetic polymers in hybrid scaffolds with the biopolymer collagen, which is the main ECM component in ligaments, could be a suitable strategy to improve cell distribution and growth within the scaffold [22], without affecting polymer chemistry and thereby stability or biomechanics. Therefore, this study was executed to analyze whether P(LA-CL) or PDS is a suitable material to prepare embroidered scaffolds to be colonized by lapine ACL cells. Furthermore, strategies were developed to improve cell retention within the textile scaffolds using different collagen supplementations. The pore size of the embroidered structures shows a bottom limit, because with too low numbers of meshes the structure gets stiffer than desired to mimic ACL properties. To enhance biocompatibility of the synthetic material and to achieve a supporting biocompatible structure between the meshes the embroidered structures can be combined with collagen sponges, foams and hydrogel. In this study a commercially available chicken collagen gel was used as reference certified for establishing Threedimensional (3D) cultures for tissue engineering purposes, while bovine soluble collagen, bovine non soluble collagen (dispersion) and sponge from bovine collagen were prepared on our own according to medical device legislation. The bovine collagens were not treated with proteases. Therefore this collagen still contained telopeptides whereas the chicken collagen was described as atelomeric collagen by the supplier. In a first step thin films of P(LA-CL) and PDS, prepared by spin coating were further covered with fibrillated collagen to analyze the benefit of collagen in combination with P(LA-CL) and PDS. In a second step embroidered scaffolds from the synthetic polymers were combined with a collagen hydrogel for a homogeneous cell distribution, or the scaffolds were directly embroidered on sheets (plies) of a collagen sponge that should absorb the cell suspension and help to over span the embroidery pores. Finally, the scaffolds could also be impregnated using collagen foam that should increase the cell adherence and also initially scale down the scaffolds pore diameter.

291

2. Materials and methods 2.1. Preparation of P(LA-CL) and PDS embroidered scaffolds and thin films 2.1.1. Embroidered P(LA-CL) and PDS scaffolds Textile scaffolds were embroidered on a JCZ 0209-550 embroidery machine (ZSK, Germany) from surgical suture threads of P(LA-CL) (Gunze Ltd., Japan) and PDS (Samyang Biopharmaceuticals Corp., Korea) (thread diameters: 125 μm corresponds to the USP 6-0 [6-0] and 85 μm to the USP 7-0 [7-0], provided by the company Catgut, Germany). Threads were processed on plies of water soluble polyvinyl alcohol (PVA [Veline 581 white, Freudenberg Einlagestoffe, Germany]) as described elsewhere [24]. The base material was washed out after the embroidery process in several rinsing steps with aqua dest. All scaffolds were designed in a zig-zag stitch embroidery pattern with a stitch length of 1.5 mm and a stitch angle of 30° (Fig. 1). The height of the scaffolds was obtained by embroidering either three layers (6-0) or seven layers (7-0) one upon the other. All scaffolds were sterilized using ethylene oxide as proposed by Ray et al. [14]. Tensile testing of used monofilaments was executed following DIN EN 13895:2003 using a Zwick/Roell UPM 2.5 tensile testing machine (parameters: 15 mm clamping length, 50 mm/min testing velocity, rubberized mechanical clamps, 0.2 N minor load). Embroidered scaffolds were initially tested for proper biomechanics compared to native ACL references and were classified as suitable for ACL reconstruction in the rabbit (results not published yet).

2.1.2. Spin coating P[LA-CL] and PDS films were produced by spin coating (RC5; Suess Microtec, Germany). The suture materials P(LA-CL) and PDS were dissolved in chloroform (N99%, Merck, Germany) and 1,1,1,3,3,3hexafluoro-2-propanol (Sigma-Aldrich, Germany), respectively. Coverslips (d = 15 mm) were freshly cleaned with ethanol (J.T. Baker, Netherlands) for 30 min in ultrasonic bath and thereafter, to achieve adherent coating, the surface was modified with tridecafluoro-1,1,2,2tetrahyrdoctyl (trichlorosilane 97%, AB 111 444, ABCR, Germany) in gaseous phase, a procedure called silanization. Then, the surface was covered with 0.5% polymer solution. Spin coating was conducted at 3000 rpm for 30 s. The thickness of the polymer thin films was determined by ellipsometry (VASE 44 M; Woolam, USA); the contact angle was measured by means of dynamic contact angle measurement (OCA 30, dataphysics, Germany).

2.1.3. Bovine soluble collagen Soluble collagen was extracted from calf skin, which was obtained from a local abattoir. The extraction method was described in detail by [25]. In short, skin was shaved, cut and intensively washed and immersed into 0.1 M acetic acid (Carl Roth, Germany) under intermittently stirring for three days without using enzymes. This acidic solved collagen was then precipitated by adding solid sodium chloride (Carl Roth) up to a total concentration of 5%. The precipitate was centrifuged, dialyzed against deionized water and freezedried. This dried collagen was then re-dissolved in 0.01 M hydrochloric acid (Carl Roth) at 4 °C (2 mg/ml). For coating 50 μl TRISbuffer (50 mM Tris [Sigma-Aldrich], 150 mM NaCl [Carl Roth], pH 9) was placed at the surface of the thin film. 50 μl collagen solution was injected into the buffer; the final pH of this mixture was 8.9. Fibrillation started between some seconds and some minutes and a thin layer of collagen fibrils adsorbed at the thin film surface. The samples were allowed to incubate at 4 °C for 24 h – these conditions were found to be appropriate to generate fibrils – before being rinsed with deionized water. Finally, the thin films covered with this fibrillated collagen were dried at room temperature (RT).

292

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

A

embroidery pattern

B

C

P(LA-CL)

PDS

D

thin films glass

stitch length

stitch angle

loading axis

P(LA-CL)

P(LA-CL) + collagen

PDS

PDS + collagen

5000µm

E

PDS + collagen sponge

F

collagen sponge blank

G

P(LA-CL) + collagen foam

H

PDS + collagen foam

Fig. 1. Embroidered scaffolds and thin film polymer surfaces. Zig-zag embroidery pattern applied to the scaffolds (A) consisting either of violet colored P(LA-CL) (B) or PDS (C) suture material shown in 7-0 (B, C). Polymers P(LA-CL) and PDS spincoated on silanized cover slips with and without a bovine collagen coating (D). A scaffold modification is depicted in E, where PDS 7-0 scaffolds were combined with 5 plies of collagen sponge that were embedded into the embroidery. The collagen sponge as an 8 mm diameter blank is shown in F. G and H depict P(LA-CL) and PDS 6-0 scaffolds impregnated with collagen foam. Scale bars = 5000 μm.

2.1.4. Preparation of scaffolds with collagen sponge Two batches of scaffolds, embroidered with P(LA-CL) and PDS in 7-0, were provided with a collagen interior by using the collagen sponge plies as base material instead of PVA as described in 2.1.1. The required height was obtained by embroidering three layers one upon the other, implying the insertion of five plies of the collagen sponge. These sponges were prepared by lyophilization of non-soluble collagen dispersions. For this, bovine hides were washed, split, limed, washed again, delimed, treated with hydrogen peroxide (1%, Carl Roth), chopped and acidified with hydrochloric acid to pH 3 [26,27]. The dry matter content was adjusted with aqua dest. to achieve a viscous liquid of 3% dry matter content which was then cast in Petri dishes and lyophilized by freezing at − 30 °C and drying in a freeze dryer (Epsilon 1-4, Christ, Germany) [26]. The dried sponges were then split to a final thickness of 1 mm with a splitting machine (Traco, CZ, modified) to achieve the plies. 2.1.5. Impregnation of scaffolds with collagen foam The scaffolds that were impregnated with collagen foam, were embroidered from P(LA-CL) and PDS (both 6-0), with zig-zag pattern and 1.8 mm stitch length and 15° stitch angle to create larger pores for better collagen foam penetration. To impregnate these scaffolds with collagen foam, they were soaked in 500 μl TRIS-buffer (pH 9) using reaction tubes, 500 μl of soluble collagen solution (bovine) was added, and the scaffolds were incubated for 1 h at 23 °C and rotated at 500 rpm. This technique led to interpenetrating fibrillated collagen gels as part of the embroidered scaffolds (refer to Supplemental Fig. 1). The wet scaffolds were fixed in polystyrene-boxes, frozen at −30 °C and lyophilized as described above. 2.2. Isolation of ligament cells from the lapine ACL Lapine ACL derived ligament cells were isolated from healthy ACLs of one female and 6 male New Zealand White Rabbits that weighed 2.8–

5 kg. ACLs were sliced into 2 mm2 pieces and placed into a culture flask with growth medium (Ham’s F-12/Dulbecco’s Modified Eagle’s Medium (DMEM) 1:1) containing 10% fetal calf serum (FCS), 50 IU/ml streptomycin, 50 IU/ml penicillin, 0.5 μg/ml partricin, essential amino acids (all from Biochrom, Germany) and 25 μg/ml ascorbic acid (Sigma-Aldrich). Ligament cells growing out of the tissue slices were harvested weekly. Cells were expanded until passage 4–6 before being used as described in 2.3. 2.3. Seeding of P(LA-CL) and PDS scaffolds and thin films without/with collagen 2.3.1. Dynamic seeding of ACL cells on P(LA-CL) and PDS embroidered scaffolds To analyze whether the ACL cells were able to adhere on P(LA-CL) and PDS-based 3D scaffolds, embroidered scaffolds (both 6-0) were seeded using a dynamic strategy (n = 5). The scaffolds were washed in sterile aqua dest. at 37 °C overnight to ensure the complete removal of PVA. Scaffolds were seeded under rotary shaking conditions with a suspension of 0.0037 × 106 ACL cells/mm3 scaffold in 7 ml growth medium in TubeSpin bioreactor tubes (TPP, Switzerland) using a rotator device (digital tube roller Stuart SRT9D, UK) with 12 rpm at 37 °C. Half medium changes were executed every 2 days, because of the high volume of culture medium in dynamic culture in contrast to statically seeded constructs where complete media changes were performed until the cultivation was stopped at day 7. 2.3.2. Seeding of ACL cells on thin films To examine whether the adherence and distribution of ACL cells on P(LA-CL) and PDS could be influenced by collagen, thin films of both biomaterials spin coated on glass slides either with or without an additional collagen layer were prepared and seeded with ACL cells. These spin coats as well as uncoated glass slides and silanized glass as controls were sterilized with 70% ethanol (J.T. Baker) for 10 s before seeding with

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

130 cells/mm2 in minimal growth medium (DMEM w/GLUTAMAX-I; Life technologies, Germany) containing 50 IU/ml streptomycin and 50 IU/ml penicillin (all from Biochrom) to enhance possible effects of the polymer coats on the cells morphology. To mimic in vivo conditions and long term culture of scaffolds, the minimal growth medium also contained 10% FCS (Biochrom). After 6 and 24 h the monolayers were rinsed with phosphate buffered saline ([PBS] Biochrom) and fixed with a ready to use 4% paraformaldehyde solution ([PFA] USB, USA) for 10 min and stored at 4 °C. 2.3.3. Static seeding of P(LA-CL) and PDS scaffolds with ACL cells and collagen hydrogel Embroidered P(LA-CL) and PDS scaffolds with thread sizes 6-0 (n = 3) and 7-0 (n = 6) were prepared as described in 2.3.1. The collagen hydrogel (chicken atelo-collagen, [Millipore Corporation, USA]) was generated as described in the manufacturer`s protocol. 0.0037 × 106 ACL cells/mm3 scaffold were suspended in the collagen solution before starting the re-fibrillation to form a hydrogel in the scaffolds. The seeded scaffolds were placed in agarose (1%, suitable for cell culture, SigmaAldrich) coated multiwell plates to prevent cell migration to the culture dish, overlaid with growth medium that was changed every two days and cultured until day 7. 2.3.4. Static seeding of PDS with collagen sponge and sponge blanks using ACL cells Dry PDS scaffolds (7-0, n = 3) with 5 plies of collagen sponge were placed in agarose coated multiwell plates to prevent cell migration to the culture dish. A cell suspension of 190 μl 0.0037 × 106 ACL cells/mm3 scaffold in growth medium was applied to the scaffold until the collagen sponge absorbed the whole volume. The seeded scaffolds were incubated for 1 h at 37 °C to enable the cells to adhere to the scaffold before they were overlaid with growth medium that was changed every two days and cultured until day 7. As a control, blanks punched-out of the collagen sponge (diameter: 8 mm, height: 1.14 mm) were also seeded with ACL cells and cultured as described for the scaffolds (n = 3).

293

Cell monolayers on P(LA-CL) or PDS thin films either with or without collagen coating as well as on the glass and silanized glass slides as controls were rinsed in TRIS buffered saline (TBS: 0.05 M TRIS [Sigma-Aldrich], 0.015 M NaCl [Carl Roth], pH 7.6) before blocking with blocking buffer (5% donkey serum in TBS) for 30 min at RT. The samples were incubated with the primary mouse-anti-vinculin antibody (Sigma, USA) which was diluted in blocking buffer containing 0.1% Triton X-100 (Sigma-Aldrich) for 30 min in a humidified chamber at RT. After rinsing with TBS, the samples were incubated with the labeled secondary antibody (donkey anti-mouse Alexa488, Invitrogen, Germany) which was diluted in blocking buffer containing 0.1% Triton X-100, 0.1 μg/μl 4,6diamidino-2-phenylindole (DAPI, Roche, Germany) and 50 μg/ml phalloidin–CruzFluor555 (Santa Cruz Biotechnology, USA) for 1 h. After rinsing the labeled samples were overlaid with cover slides using fluoromount. The red (F-actin), green (vinculin) or blue (nuclei) fluorescence was monitored by confocal laser scanning microscopy (TCS SPE II, Leica Microsystems). For determination of average cell numbers, five areas (each 1.6 mm2) per donor (n = 3) and material were evaluated. The seeded collagen sponge blanks were embedded in paraffin wax, cross sectioned into 5 μm slices and then stained with 0.1 μg/μl DAPI to visualize cell migration into the sponge. 2.6. Scanning electron microscopy (SEM) Samples were fixed in 4% PFA solution overnight, washed in PBS and dehydrated in 10%, 25%, 50%, 70% and 90% ethanol followed by a 96% and 100% isopropanol step (15 min. each). Samples were infiltrated with hexamethyldisilazane (HMDS, Sigma-Aldrich) using a 25% and 50% isopropanol/HMDS solution for 15 min. Finally, samples were incubated in 100% HMDS and dried overnight under an extractor hood. The scaffolds were mounted on carriers and sputtered with a 50 nm gold layer (Bal-Tec Sputter Coater MED 010, Leica Microsystems). Samples were observed in a Philips FEG XL30 ESEM (FEI, USA) in Hi-Vak mode with acceleration voltages of 2 to 10 kV. 2.7. Sulfated glycosaminoglycan (sGAG) quantification

2.3.5. Dynamic seeding of P(LA-CL) and PDS scaffolds with collagen foam using ACL cells P(LA-CL) and PDS (both 6-0; n = 3) scaffolds with integrated collagen foam were mounted with a cell suspension (0.0037 × 106 ACL cells/mm3) in minimal growth medium and placed in TubeSpin bioreactor tubes on a rotary shaker with 12 rpm at 37 °C. The seeded scaffolds were overlaid with medium whose half volume was changed every two days and cultured until day 7. 2.4. ACL cell vitality in cultured scaffolds To estimate the cell vitality in seeded scaffolds, a fluorescein diacetate (FDA, Sigma-Aldrich) and ethidium bromide (EtBr, Carl Roth) staining was performed after 7 days of cultivation. A part of the scaffold was rinsed in PBS and incubated for 2 min in 9 μg/ml FDA and 10 μg/ml EtBr dissolved in PBS in the dark. The green (living cells, FDA) or red (dead cells, EtBr) fluorescence was monitored by fluorescence microscopy (Axioskop 40, Zeiss, Germany with camera XC30, Olympus, Germany) or by laser scanning microscopy (TCS SPE II, Leica Microsystems, Germany). 2.5. Histology and immunofluorescence staining After 7 days of culture, the scaffolds were fixed in 4% PFA solution and dehydrated in an ascending ethanol series before they were embedded in paraffin wax (Merck) and sectioned with a thickness of 7 μm. The general histological examination was performed by a hematoxylin (Sigma-Aldrich) and eosin Y (Sigma-Aldrich) staining (HE). All slices were analyzed by light microscopy (Axioskop 40, with camera XC30).

After 7 days of culture a similar sized piece of each scaffold was treated with proteinase K (Roche) solution (800 ng/ml proteinase K; 50 mM TRIS/HCl, 1 mM EDTA, 0.5% Tween 20 at pH 8.5, all Sigma-Aldrich) to digest total proteins overnight at 55 °C. Remaining threads were separated by centrifugation (10 min, 10,000 ×g). The total sGAG concentration in the supernatant was detected using dimethyl methylene blue (DMMB, Appli-Chem, Germany). After the dilution of the samples in a buffer containing 40 mM glycine (Sigma-Aldrich) and 40 mM NaCl (Carl Roth) at pH 3 DMMB (8.9 mM in ethanol) was added. Chondroitin sulfate (Sigma-Aldrich) was used to achieve a linear standard curve. The absorption shift from λ = 633 nm to λ = 552 nm was measured immediately after dye addition to prevent the formation of large aggregates. The results were normalized to the cell number according to the DNA content of the lysate as determined using the CyQuant NF Assay (Life Technologies). The assay was applied according to the manufacturer`s protocol using calf thymus DNA (Sigma-Aldrich) as standard. 2.8. Statistics The statistical analysis of average cell numbers on thin films of the tested materials was executed using Prism 5 for Windows (version 5.02, GraphPad Software Inc., USA). Using the Kolmogorov–Smirnov test with Dallal–Wilkinson–Lilliefor P value the normal distribution of the counting results (n = 5) per material and donor (n = 3) was proofed. After a one-way analysis of variance the mean results of all donors per material were post tested for significant differences using Bonferroni's multiple comparison test. The same program was used for the statistical analysis of sGAG concentrations (n = 3). In this case,

294

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

the Kruskal–Wallis test with Dunn's comparison of all columns post test was chosen for the evaluation of the results. The level of significance was P b 0.05 for all tests.

scaffold pores, spread within the whole scaffold and form a homogeneous tissue.

3.2. Compatibility of lapine ACL cells with P(LA-CL) and PDS 3. Results 3.1. Scaffolds and collagen materials Tensile testing of the threads resulted in a maximum tensile strength (Fmax [N]) of 6.47 and 3.50 for P(LA-CL) (6-0/7-0) and 5.56 and 4.40 for PDS (6-0/7-0) and in a maximum tensile strain (εmax [%]) of 83.4 and 60.7 for P(LA-CL) (6-0/7-0) and 49.8 and 33.9 for PDS (6-0/7-0). The P(LA-CL) and PDS embroidered scaffolds with thread sizes 6-0 or 7-0 and longitudinal zig-zag embroidery pattern (Fig. 1A–C) had an average dimension of 16.6 × 4.8 × 2 mm and were therefore based almost on in situ measured dimensions of native lapine ACLs (14 × 4 × 1 mm). The spin-coated polymer films covalently cross linked to glass coverslips via silanization exhibited a thickness of about 45–65 nm proven by ellipsometry. The hydrophilicity of the polymer films was determined by measuring the static advancing water contact angles and resulted in 79.5° for P(LA-CL) and 64.75° for PDS, which is therefore more hydrophilic than P(LA-CL). The contact angle measurement of the glass and silanized glass controls resulted in 62.7° for glass before and 108.1° for glass after silanization. The embroidered structures (Fig. 1B, C) were filled with a collagen hydrogel (not shown), which was already mixed with ACL cells. The integrated collagen sponge plies should soak the cell suspension in the PDS 7-0 scaffolds (Fig. 1E) and thereby retain the cells in place and possibly facilitate cell adhesion due to the presence of collagen. The scaffolds with integrated collagen sponge had comparable dimensions as scaffolds embroidered without collagen sponge (Fig. 1B, C, E). The collagen sponge was also tested without threads as an 8 mm blank that had a homogeneous surface with macroscopically no pores visible (Fig. 1F). Furthermore, the embroidered P(LA-CL) and PDS 6-0 scaffolds were impregnated with a collagen foam of re-fibrillated collagen (Fig. 1G, H). The collagen foam filled the scaffold pores with a finely collagen network (Fig. 1G, H), which should support the cells to fill the

The vitality staining of dynamically seeded P(LA-CL) and PDS 6-0 scaffolds with ACL cells revealed a high cytocompatibility of both materials with a majority of viable cells and only few dead cells (Fig. 2A, B). The SEM showed a homogeneous distribution of ACL cells spreading over the whole surface of P(LA-CL) threads and aligning mostly according to the thread orientation (Fig. 2C). In contrast, the cells were not able to adhere in a similar manner on PDS threads, where they tended to aggregate more often at the crossing points of the threads and seemed to have looser contacts to the surface (Fig. 2D). The direct influence of the thread materials and an additional collagen coating on the ACL cells' ability to adhere and spread was therefore tested on 2 dimensional P(LA-CL) and PDS thin films with and without a collagen coating. ACL cells showed cluster formation and less vinculin expression in silanized compared to untreated glass (Fig. 3A–D, B2, D2) and tended to a lower number of cells attached after 6 h (not significant, Fig. 3N). After 24 h, the silanized glass surface showed a similar number of adherent cells compared with the untreated glass control, however most of the cells on silanized glass formed clusters (Fig. 3D2, N). The untreated glass control revealed more cells attached after 6 h than all other collagen free thin films (not significant, Fig. 4N), whereas this difference was not further detectable after 24 h with P(LA-CL) but not with PDS thin films (not significant, Fig. 3A–L, M). ACL cells on P(LA-CL) (Fig. 3E) showed a distinctly lower adherence and spreading (significant with P b 0.01, Fig. 3N) as well as a less developed F-actin cytoskeleton than in P(LA-CL) with collagen coating (Fig. 3G) after 6 h. The same result was recognized on PDS with and without collagen coating (Fig. 3I, K, significant with P b 0.01, Fig. 3N). However, after 24 h this difference between uncoated and collagen coated polymer surfaces was distinctly decreased on P(LA-CL) (Fig. 3F, H, M) while on PDS it was still detectable (Fig. 3J, L, M). The additional collagen coating of both polymers led to a distinctly more homogeneous cell distribution than on pure polymers (Fig. 3F2–L2). Significantly more

P(LA-CL) 6-0

PDS 6-0

A

B

100µm

C

D

100µm

Fig. 2. Adhesion and survival of ACL cells on P(LA-CL) and PDS scaffolds. P(LA-CL) 6-0 (A, C) and PDS 6-0 (B, D) scaffolds dynamically seeded with lapine ACL cells after 7 days. The vitality staining (A, B) resulted in a majority of viable (green) and few dead (red) cells as well as the beginning degradation of PDS threads indicated by a reddish color of scaffold threads (B) (n = 5); arrows indicate cells. C and D are SEM micrographs showing the cell morphology, orientation and distribution on the threads (n = 1). Cell orientation in C is indicated by the double head black arrow aligned in the direction of a representative group of cells bordered by a black line. Non seeded thread is blue colored (C, D). Scale bars = 100 μm.

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

295

Fig. 3. Cytoskeleton architecture and cell adhesion in response to polymers, collagen coating and cultivation time. Vinculin immunolabeling (green) and F-actin staining (red), combined with blue counterstaining of the cell nuclei. Lapine ACL cells were labeled after 6 h (A, C, E, G, I, K) and 24 h (B, D, F, H, J, L) (n = 3). Ligament cells adhered to untreated glass (A, B), silanized glass (C, D), P(LA-CL) (E, F), P(LA-CL) combined with collagen (G, H), PDS (I, J) and PDS combined with collagen (K, L). All glass cover slips with polymer coating had also a previous silanization (*). After 24 h the collagen coated surfaces were distinctly more homogeneous seeded than any other collagen free variant, where cell cluster formation was observed (B2, D2, F2, H2, J2, L2). Scale bars = 100 μm. Average number of adhered cells after 6 (M) and 24 h (N). M–N: means with standard deviation (n = 3), * significantly different with P b 0.05, ** significantly different with P b 0.01.

cells adhered and spread on P(LA-CL) (P b 0.05) than on PDS thin films despite the presence of several cell clusters on PDS thin films (Fig. 3J2, M).

3.3. Collagen hydrogel vs. collagen sponge as additional matrix in embroidered scaffolds Collagen could facilitate cell adhesion and could support the ACL cells to fill the scaffolds pores. Therefore, collagen hydrogel or collagen sponge was combined with the scaffolds. By embedding of the cells directly in the hydrogel it was possible to create a homogeneous cell distribution, independent of the scaffold materials and thread sizes (Fig. 4). ACL cells were able to re-elongate in the hydrogel (Fig. 4A–D). Also the cytocompatibility of the hydrogel with the ACL cells was high since most of the cells survived the hydrogel polymerization procedure and the subsequent cultivation embedded in the gel (Fig. 4E–H). These observations were also confirmed by the HE staining (Fig. 4I–L). Since ACL cells did not form direct thread contacts in most cases (Fig. 4I–L) the success of a permanent colonization of the scaffolds depended on the stability of the used collagen hydrogel that keeps the cells in the construct.

Therefore, another collagen based matrix was tested for the seeding of PDS scaffolds (Figs. 1E, 5A). Collagen sponge plies integrated into the embroidered scaffolds should provide an improved initial cell-scaffold contact. Most cells were vital (Fig. 5C) but the cells only adhered to the surface and neither penetrated the collagen sponge plies nor homogeneously colonized the scaffold (Fig. 5E). In the single sponge blank, the cell growth was also restricted to the surface and most of the cells were not able to migrate into the collagen sponge (Fig. 5B, F). The majority of cells survived (Fig. 5D). Due to these results P(LA-CL) threads with collagen sponge plies were not further tested. The sGAG concentration in both P(LA-CL) and PDS scaffolds with collagen hydrogel was distinctly lower (not significant) than that in the PDS scaffolds with integrated collagen sponge plies (Fig. 5G). The combined sponge–PDS scaffolds also met the sGAG expression range of the sponge blanks (Fig. 5G). 3.4. Collagen foam as additional matrix in embroidered scaffolds Dynamically seeded P(LA-CL) and PDS scaffolds impregnated with collagen foam (Fig. 1G, H), resulted in a homogeneous cell distribution with mostly elongated cells that survived during 7 days of rotary culture in a P(LA-CL) scaffold (Fig. 6A). The seeding success of the PDS scaffolds

296

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

P(LA-CL) 6-0

P(LA-CL) 7-0

A

PDS 6-0

PDS 7-0

B

C

D

F

G

H

* J

K

200µm

E

200µm

I

*

* *

*

* 200µm

L

*

*

*

*

Fig. 4. Statically seeded P(LA-CL) and PDS scaffolds with collagen hydrogel. Statically seeded P(LA-CL) 6-0 (A, E, I), P(LA-CL) 7-0 (B, F, J), PDS 6-0 (C, G, K) and PDS 7-0 (D, H, L) scaffolds after 7 days. ACL cells were embedded by the collagen hydrogel within in the scaffolds (A-D). E–H: Vitality staining indicates viable (green) and dead (red) cells. I–L: HE staining depicts pink to violet ligament cells in a light violet collagen hydrogel. Thread sizes 6-0 (n = 3) and 7-0 (n = 7); arrows indicate cells; asterisks indicate thread positions. Scale bars = 200 μm.

could not be increased by the collagen foam, since only few clusters of viable, round cells adhered on PDS with loose cell-thread contact (Fig. 6B). 4. Discussion Tissue engineering of a cell-based graft, mimicking the biomechanical features of the native ACL seems to be a desirable alternative to ACL

200µm

40

20

0

PD sp on S ge bl an k

F cross section

E

60

S

200µm

collagen sponge

PD

D

collagen hydrogel

L)

C

sGAG concentration 80

-C

200µm

G

P( LA

B cross section

A

collagen sponge blank

sGAG [µg/106 cells]

PDS 7-0 + collagen sponge

reconstruction, since the ACL is not able to self-regenerate [3] and artificial prostheses and autografts are associated with a couple of risks [4, 28]. Small biopsies from the ruptured ACL stump, whose cells are viable and produce collagenous ECM up to one year after rupture [29] are needed to isolate sufficient autologous ligament cells. Brune et al. could show that in vitro cultivated ACL cells derived from healthy and ruptured ACL tissue did not express different characteristics in a tissue engineering based ACL replacement approach [2]. Therefore, lapine

Fig. 5. Collagen sponge as additional cell matrix integrated into the embroidered scaffolds. PDS 7-0 scaffolds with collagen sponge plies (A, C, E) and collagen sponge blanks (B, D, F) were cultured statically with ACL cells for 7 days (n = 3). A bright field micrograph (A) and a blue nuclear staining (B) are depicted. Vitality staining revealed a majority of elongated, viable (green) and fewer dead (red) cells (C, D). HE staining with pink to violet ligament cells in a light violet collagen sponge (E, F). The sGAG content was determined after 7 days. It is depicted as mean with standard deviation (G) (n = 3). Arrows indicate cells. Scale bars = 200 μm.

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

P(LA-CL) 6-0

297

PDS 6-0

A

B

* *

*

* 100µm

Fig. 6. ACL cells on P(LA-CL) and PDS scaffolds with collagen foam. Vitality staining with viable (green) and dead (red) cells of ACL cells on P(LA-CL) 6-0 (A) and PDS 6-0 (B) scaffolds impregnated with collagen foam after 7 days of dynamical culture. Arrows indicate cells; asterisks indicate thread positions. Scale bars = 100 μm.

ligament cells isolated from non-ruptured ACLs were used in the present study. Scaffolds were prepared in the dimensions of a lapine ACL in perspective of the use in an animal model. P(LA-CL) and PDS embroidered scaffolds are cytocompatible in vitro, as shown in our study and for P(LA-CL) also in vivo [17]. The ACL cells adhered and survived on both scaffolds using a dynamical seeding procedure, but on PDS the cell distribution was inhomogeneous with cells growing in clusters. Also the study of Hakimi et al. showed difficulties of tenocytes to homogeneously colonize PDS sutures. They argued that the smooth surface of the suture threads are not favorable for cell adhesion and were able to show that a rougher surface of electrospun PDS material was more suitable for colonization [30]. Although the SEM micrographs showed a somewhat structured PDS thread, this adumbrated groove pattern was apparently to less to support cell adhesion. In contrast, ligament cells on P(LA-CL) had a homogeneous distribution with a cell alignment mostly according to the P(LA-CL) thread orientation. Since it is known that the tensile strength, elasticity and ECM density of ligaments depend on such aligned structures and cell orientations [12] the applied embroidery pattern along the longitudinal axis seems to be advantageous in P(LA-CL) scaffolds. Lapine ACL cell monolayers were cultured on spin coated glass slides with P(LA-CL) or PDS to examine the ACL cells' ability to adhere and spread on P(LA-CL) and PDS suture materials in view of cytoskeletal architecture. After 6 h, the cells on both suture materials showed vinculin expression, a membrane-associated cytoskeletal protein in focal adhesion sites that is involved in linkage of the cell-ECM adhesion molecule integrin to the actin cytoskeleton. After 24 h, the ligament cells showed further adherence and spreading as well as a well-developed F-actin cytoskeleton. Nevertheless, on P(LA-CL) higher cell numbers adhered more homogeneously, while fewer cells spread on PDS thin films and formed several cell clusters. Probably the higher hydrophilicity of PDS in comparison to P(LA-CL) is responsible for the inhomogeneous colonization of pure PDS in 2 and 3 dimensional applications. High hydrophilicity is potentially disadvantageous for cell adhesion, since more moderate wettable polymers lead to better serum adsorption that is respectively followed by a higher cell adhesion rate [31]. Our own results in culturing monolayers on thin films of P(LA-CL) and PDS confirm our assumption that collagen increased adherence and spreading of ACL cells since it presents binding domains for transmembrane integrin receptors as precondition for focal adhesion formation [32]. Therefore, it was analyzed, whether the additional collagen supplementation in hybrid scaffolds could improve the ACL cells attachment. Collagen was chosen to create optimal conditions for ligament cells regarding their adherence, growth and ECM formation. The ECM of the native ACL consists of fibrillar collagen, thereby the ACL is perceived as a macroscopic fiber bundle [33]. With 90% of the total

collagen content of the native ACL ECM, type I and type III collagens are the most abundant collagen types in ligaments [34–36]. To take advantage of the stimulatory effect of collagen on ACL cells three P(LA-CL)/PDS polymer–collagen hybrid scaffold variants were compared: cells were either embedded in a collagen hydrogel within the scaffolds, or plies of collagen sponge were embroidered into the scaffolds structure or the scaffolds pores were filled with collagen foam. Atelomeric collagens are thought to be more immunocompatible than the native collagen, since the immunogenicity is up to this region and therefore a xenogenic application is possible [37]. Also Magarian et al. showed that hydrogels from atelomeric collagen were suitable for a xenogenic supplementation of ACL sutures [38]. Therefore the first hybrid scaffold variant was based on the embedding of ACL cells into a collagen hydrogel (atelo-chicken collagen) that was introduced in the P(LA-CL) and PDS embroideries. Thereby, a 3 dimensional cell culture system was designed that closely simulates the collagen-rich ECM of native ligament tissue [39,40]. Garvin et al. could show that tendon cells embedded in a mechanically loaded collagen gel construct assumed a phenotype that was similar to that of cells in a native tendon in terms of morphology and expression profile [41]. According to our previous results with the collagen coated thin films, lapine ACL cells were homogeneously distributed in both P(LA-CL) and PDS scaffolds supplemented with collagen gel with a high proportion of reelongated cells. However, the ACL cells had only few direct thread contacts and the success of a permanent colonization of the scaffolds depends on the stability of the used collagen hydrogel. Therefore, in a second approach a more stable hybrid scaffold was used. Collagen sponge plies were integrated into embroidered PDS scaffolds to examine if this could improve the colonization. The collagen sponge should absorb the cell suspension and hold the cells in their compartment. Sponges from type 1 collagen are known to be suitable for ACL suture supplementation [38] and were tested as growth factor supplemented wrapping of the bone insertion part of autologous ACL replacements [42]. Cells were able to colonize the scaffold surface, but did not penetrate the sponge plies. Since the histology of the cultivated sponge-PDS hybrid scaffold showed a tightening of the plies compared to their porosity before the embroidery process, collagen sponge blanks without threads were tested. However, cross sections of the collagen sponge blanks showed in spite of a higher porosity also that the cell colonization was mainly restricted to the blank surface. This observation is in agreement with a study of Quteish et al. where the authors reported that cells colonized only the surface of a type 1 collagen sponge without penetration into the center [43]. Probably the exclusive seeding of the scaffolds and blanks surface led to the comparable sGAG concentrations independently from their size difference. The sGAG concentrations of

298

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299

P(LA-CL)- and PDS–collagen hydrogel hybrid scaffolds were lower than in collagen sponge applications, certainly due to the fast hydrogel degradation and the associated sGAG and cell loss into the medium, which were not controlled. Our third used hybrid scaffold variant is based on P(LA-CL) and PDS scaffolds with a collagen foam impregnation from freeze dried collagen. A related application was described by Pang et al. [44] that introduced collagen micro sponges into a PLGA/β-TCP scaffold for bone remodeling, but this method differs in the preparation of the collagen before freeze drying, since our collagen was first freezed at −30 °C to arrange a finely, porous collagen network different from sponges. So to our knowledge the introduction of collagen foam into embroidered scaffolds and seeding with primary ACL cells is a new field of investigation in ligament tissue engineering. The collagen foam–polymer hybrid scaffolds allowed the formation of intensive cell-thread contacts in contrast to the hydrogel. In agreement with our previous results with dynamically seeded scaffolds without collagen, only few cell clusters but large unseeded areas were found on PDS–collagen foam scaffolds compared to P(LA-CL)–collagen foam scaffolds, where the cells could spread over the threads. It is not clear, whether its hydrophilicity or the smooth surface led to a weak attachment of the collagen foam to PDS resulting in a slip away from the threads. Therefore, a stabilization of the collagen foam by cross-linking of the collagen to improve collagen adhesion to the fibers will be a possible strategy in the future as described for collagen sponges [21,37,45].

5. Conclusion To sum up, P(LA-CL) embroidered scaffolds were more suitable for the seeding with lapine ACL cells compared to PDS, irrespective of the use in collagen supplemented hybrid scaffolds or in single use. The tested thread sizes of both biomaterials revealed no major differences in regard to cell adhesion and vitality and should therefore be chosen dependent on the desired biomechanical suitability. Thus for designing a tissue engineered ACL, biomechanical analyses of the hybrid scaffolds seeded with ACL cells are necessary. So far unpublished results revealed that scaffolds using both P(LA-CL) and PDS threads in combination could exhibit a biomechanical profile very comparable with the lapine ACL. Collagen hydrogel-based hybrid scaffolds lead to a high vitality and homogeneous cell distribution in both materials, but the permanent colonization of cells in the scaffolds remains limited by the stability of the collagen hydrogel due to few direct cell–thread contacts. Embroidered scaffolds with integrated collagen sponge plies are not suitable for a homogeneous cell distribution through the whole scaffold. Collagen foams represent an interesting novel technique for scaffold functionalization and this study revealed their high cytocom patibility for ACL cells. The not cross-linked collagen foams could improve the cell distribution in embroidered P(LA-CL) scaffolds but they did not increase the adhesion of the cells on PDS. So the collagen foam cross-linking should be tested in future experiments to functionalize PDS. An interesting perspective approach to support a longitudinal cell alignment and stimulate ECM secretion would be the biomechanical stimulation of the constructs in vitro. Embroidered scaffolds supplemented with collagen foams could be a basis for a transient tissue-engineered cell-seeded ACL construct. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2014.07.010.

Acknowledgements The authors are grateful for technical assistance of Benjamin Kohl and Marion Lemke. This study was funded by the German Research Foundation (DFG-SCHU1979/9-1) and the Sonnenfeld Foundation.

References [1] B.N. Tillmann, F. Eckstein, in: K.T. Zilles, B.N. (Eds.), Anatomie, Springer Medizin Verlag, Berlin Heidelberg New York, 2010, pp. 76–94. [2] T. Brune, A. Borel, T.W. Gilbert, J.P. Franceschi, S.F. Badylak, P. Sommer, In vitro comparison of human fibroblasts from intact and ruptured ACL for use in tissue engineering, Eur. Cell Mater. 14 (2007) 78–90. [3] J.W. Freeman, M.D. Woods, C.T. Laurencin, Tissue engineering of the anterior cruciate ligament using a braid-twist scaffold design, J. Biomech. 40 (2007) 2029–2036. [4] F.P. Tjoumakaris, A.L. Herz-Brown, A.L. Bowers, B.J. Sennett, J. Bernstein, Complications in brief: anterior cruciate ligament reconstruction, Clin. Orthop. Relat. Res. 470 (2012) 630–636. [5] M. Ekdahl, J.H. Wang, M. Ronga, F.H. Fu, Graft healing in anterior cruciate ligament reconstruction, Knee Surg. Sports Traumatol. Arthrosc. 16 (2008) 935–947. [6] S. Rathbone, N. Maffulli, S.H. Cartmell, Most british surgeons would consider using a tissue-engineered anterior cruciate ligament: a questionnaire study, Stem Cells Int. 2012 (2012) 303724. [7] K.E. Rehm, K.H. Schultheis, Ligament replacement with polydioxanone (PDS), Unfallchirurgie 11 (1985) 264–273. [8] F.A. Barber, J.N. Click, The effect of inflammatory synovial fluid on the breaking strength of new "long lasting" absorbable sutures, Arthroscopy 8 (1992) 437–441. [9] Z. Ge, F. Yang, J.C. Goh, S. Ramakrishna, E.H. Lee, Biomaterials and scaffolds for ligament tissue engineering, J. Biomed. Mater. Res. A 77 (2006) 639–652. [10] M. Labet, W. Thielemans, Synthesis of polycaprolactone: a review, Chem. Soc. Rev. 38 (2009) 3484–3504. [11] K. Tomihata, M. Suzuki, T. Oka, Y. Ikada, A new resorbable monofilament suture, Polym. Degrad. Stab. 59 (1998) 13–18. [12] C.K. Kuo, J.E. Marturano, R.S. Tuan, Novel strategies in tendon and ligament tissue engineering: advanced biomaterials and regeneration motifs, Sports Med. Arthrosc. Rehabil. Ther. Technol. 2 (2010) 20. [13] C.G. Pitt, M.M. Gratzl, G.L. Kimmel, J. Surles, A. Schindler, Aliphatic polyesters II. The degradation of poly (DL-lactide), poly (epsilon-caprolactone), and their copolymers in vivo, Biomaterials 2 (1981) 215–220. [14] J.A. Ray, N. Doddy, D. Regula, J.A. Williams, A. Melveger, Polydioxanone (PDS), a novel monofilament synthetic absorbable suture, J. Surg. Gynecol. Obstet. 153 (1981) 497–507. [15] H.W. Ouyang, J.C.H. Goh, X.M. Mo, S.H. Teoh, E.H. Lee, Characterization of anterior cruciate ligament cells and bone marrow stromal cells on various biodegradable polymeric films, Mater. Sci. Eng. C Biomim. Supramol. Syst. 20 (2002) 63–69. [16] A. Oryan, A. Moshiri, A.M. Parizi, N. Maffulli, Implantation of a novel biologic and hybridized tissue engineered bioimplant in large tendon defect: an in vivo investigation, Tissue Eng. A 20 (2014) 447–465. [17] B. Rentsch, A. Hofmann, A. Breier, C. Rentsch, D. Scharnweber, Embroidered and surface modified polycaprolactone-co-lactide scaffolds as bone substitute: in vitro characterization, Ann. Biomed. Eng. 37 (2009) 2118–2128. [18] B. Dhandayuthapani, Y. Yoshida, T. Maekawa, D.S. Kumar, Polymeric scaffolds in tissue engineering application: a review, Int. J. Polym. Sci. 2011 (2011) 19. [19] P. Fratzl, Collagen: Structure and Mechanics, Springer, New York, 2008. [20] A.K. Lynn, I.V. Yannas, W. Bonfield, Antigenicity and immunogenicity of collagen, J. Biomed. Mater. Res. B 71 (2004) 343–354. [21] E. Song, S. Yeon Kim, T. Chun, H.J. Byun, Y.M. Lee, Collagen scaffolds derived from a marine source and their biocompatibility, Biomaterials 27 (2006) 2951–2961. [22] E. Gentleman, G.A. Livesay, K.C. Dee, E.A. Nauman, Development of ligament-like structural organization and properties in cell-seeded collagen scaffolds in vitro, Ann. Biomed. Eng. 34 (2006) 726–736. [23] F.A. Petrigliano, D.R. McAllister, B.M. Wu, Tissue engineering for anterior cruciate ligament reconstruction: a review of current strategies, Arthroscopy 22 (2006) 441–451. [24] K. Gliesche, A. Breier, G. Schmack, Herstellung textiler Scaffolds mittels Sticktechnik, Biomaterialien 6 (2011) 155–159. [25] H. Orekhovic, Procollagen of Skin, Biochimia 13 (1948) 55–60. [26] S. Heinemann, C. Heinemann, H. Ehrlich, M. Meyer, H. Baltzer, H. Worch, T. Hanke, A novel biomimetic hybrid material made of silicified collagen: perspectives for bone replacement, Adv. Eng. Mater. 9 (2007) 1061–1068. [27] M. Meyer, H. Baltzer, K. Schwikal, Collagen fibres by thermoplastic and wet spinning, Mater. Sci. Eng. C Biomim. Supramol. Syst. 30 (2010) 1266–1271. [28] S. Bernardino, ACL prosthesis: any promise for the future? Knee Surg. Sports Traumatol. Arthrosc. 18 (2010) 797–804. [29] K.P. Spindler, S.W. Clark, L.B. Nanney, J.M. Davidson, Expression of collagen and matrix metalloproteinases in ruptured human anterior cruciate ligament: an in situ hybridization study, J. Orthop. Res. 14 (1996) 857–861. [30] O. Hakimi, R. Murphy, U. Stachewicz, S. Hislop, A.J. Carr, An electrospun polydioxanone patch for the localisation of biological therapies during tendon repair, Eur. Cell Mater. 24 (2012) 344–357. [31] P.B. van Wachem, T. Beugeling, J. Feijen, A. Bantjes, J.P. Detmers, W.G. van Aken, Interaction of cultured human endothelial cells with polymeric surfaces of different wettabilities, Biomaterials 6 (1985) 403–408. [32] B. Leitinger, E. Hohenester, Mammalian collagen receptors, Matrix Biol. 26 (2007) 146–155. [33] A.A. Amis, The functions of the fibre bundles of the anterior cruciate ligament in anterior drawer, rotational laxity and the pivot shift, Knee Surg. Sports Traumatol. Arthrosc. 20 (2012) 613–620. [34] I.K. Lo, L.L. Marchuk, D.A. Hart, C.B. Frank, Comparison of mRNA levels for matrix molecules in normal and disrupted human anterior cruciate ligaments using reverse transcription-polymerase chain reaction, J. Orthop. Res. 16 (1998) 421–428.

M. Hoyer et al. / Materials Science and Engineering C 43 (2014) 290–299 [35] H.J. Shao, Y.T. Lee, C.S. Chen, J.H. Wang, T.H. Young, Modulation of gene expression and collagen production of anterior cruciate ligament cells through cell shape changes on polycaprolactone/chitosan blends, Biomaterials 31 (2010) 4695–4705. [36] K. Riechert, K. Labs, K. Lindenhayn, P. Sinha, Semiquantitative analysis of types I and III collagen from tendons and ligaments in a rabbit model, J. Orthop. Sci. 6 (2001) 68–74. [37] J. Glowacki, S. Mizuno, Collagen scaffolds for tissue engineering, Biopolymers 89 (2008) 338–344. [38] E.M. Magarian, P. Vavken, S.A. Connolly, A.N. Mastrangelo, M.M. Murray, Safety of intra-articular use of atelocollagen for enhanced tissue repair, Open J. Orthop. 6 (2012) 231–238. [39] H. Koyama, E.W. Raines, K.E. Bornfeldt, J.M. Roberts, R. Ross, Fibrillar collagen inhibits arterial smooth muscle proliferation through regulation of Cdk2 inhibitors, Cell 87 (1996) 1069–1078. [40] S.Y. Cho, R.L. Klemke, Extracellular-regulated kinase activation and CAS/Crk coupling regulate cell migration and suppress apoptosis during invasion of the extracellular matrix, J. Cell Biol. 149 (2000) 223–236.

299

[41] J. Garvin, J. Qi, M. Maloney, A.J. Banes, Novel system for engineering bioartificial tendons and application of mechanical load, Tissue Eng. 9 (2003) 967–979. [42] K. Anderson, A.M. Seneviratne, K. Izawa, B.L. Atkinson, H.G. Potter, S.A. Rodeo, Augmentation of tendon healing in an intraarticular bone tunnel with use of a bone growth factor, Am. J. Sports Med. 29 (2001) 689–698. [43] D. Quteish, S. Singrao, A.E. Dolby, Light and electron microscopic evaluation of biocompatibility, resorption and penetration characteristics of human collagen graft material, J. Clin. Periodontol. 18 (1991) 305–311. [44] L. Pang, W. Hao, M. Jiang, J. Huang, Y. Yan, Y. Hu, Bony defect repair in rabbit using hybrid rapid prototyping polylactic-co-glycolic acid/beta-tricalciumphosphate collagen I/apatite scaffold and bone marrow mesenchymal stem cells, Ind. J. Orthop. 47 (2013) 388–394. [45] B. Hoyer, A. Bernhardt, A. Lode, S. Heinemann, J. Sewing, M. Klinger, H. Notbohm, M. Gelinsky, Jellyfish collagn scaffolds for cartilage tissue engineering, Acta Biomater. 10 (2014) 883–892.