Acta Biomaterialia 10 (2014) 2506–2517

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Enhanced cell adhesion on silk fibroin via lectin surface modification Andreas H. Teuschl a,b,⇑, Lukas Neutsch c, Xavier Monforte a,b, Dominik Rünzler a,b, Martijn van Griensven b,d,1, Franz Gabor c, Heinz Redl b,d a

University of Applied Sciences Technikum Wien, Department of Biochemical Engineering, Höchstädtplatz 5, 1200 Vienna, Austria The Austrian Cluster for Tissue Regeneration, Vienna, Austria c Department of Pharmaceutical Technology and Biopharmaceutics, University of Vienna, Vienna 1090, Althanstraße 14, Austria d Ludwig Boltzmann Institute for Experimental and Clinical Traumatology/AUVA Research Center, Vienna 1200, Donaueschingenstraße 13, Austria b

a r t i c l e

i n f o

Article history: Received 22 August 2013 Received in revised form 25 January 2014 Accepted 4 February 2014 Available online 12 February 2014 Keywords: Cell adhesion Lectin Silk fibroin Tissue engineering Biomaterial

a b s t r a c t Various tissue engineering (TE) approaches are based on silk fibroin (SF) as scaffold material because of its superior mechanical and biological properties compared to other materials. The translation of onestep TE approaches to clinical application has generally failed so far due to the requirement of a prolonged cell seeding step before implantation. Here, we propose that the plant lectin WGA (wheat germ agglutinin), covalently bound to SF, will mediate cell adhesion in a time frame acceptable to be part of a one-step surgical intervention. After the establishment of a modification protocol utilizing carbodiimide chemistry, we examined the attachment of cells, with a special focus on adipose-derived stromal cells (ASC), on WGA-SF compared to pure native SF. After a limited time frame of 20 min the attachment of ASCs to WGA-SF showed an increase of about 17-fold, as compared to pure native SF. The lectin-mediated cell adhesion further showed an enhanced resistance to trypsin (as a protease model) and to applied fluid shear stress (mechanical stability). Moreover, we could demonstrate that the adhesion of ASCs on the WGA-SF does not negatively influence proliferation or differentiation potential into the osteogenic lineage. To test for in vitro immune response, the proliferation of peripheral blood mononuclear cells in contact with the WGA-SF was determined, showing no alterations compared to plain SF. All these findings suggest that the WGA modification of SF offers important benefits for translation of SF scaffolds into clinical applications. Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Classical tissue engineering (TE) approaches such as matrixassisted autologous chondrocyte transplantation are based on a two-stage process: in the first step, cells from a harvested tissue are isolated and cultured until an adequate cell number is attained; and in the second step, the cell-seeded biomaterial is implanted [1,2]. As a consequence, these approaches involve two surgical interventions: tissue harvest and implantation. To avoid disadvantages such as high economical costs and low patient comfort (two operations with anesthesia), an alternative would be a so-called one-step surgical procedure for tissue regeneration [3]. Recently, the feasibility of this one-step procedure has been proven [4,5].

⇑ Corresponding author at: University of Applied Sciences Technikum Wien, Department of Biochemical Engineering, Höchstädtplatz 5, 1200 Vienna, Austria. Tel.: +43 1 333 40 77 964; fax: +43 1 333 40 77 99 964. E-mail address: [email protected] (A.H. Teuschl). 1 Current address: Department of Experimental Trauma Surgery, Klinikum rechts der Isar, Munich 81675, Ismaninger Straße 22, Germany. http://dx.doi.org/10.1016/j.actbio.2014.02.012 1742-7061/Ó 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

In a one-step procedure cells are harvested and added to a biomaterial in the same operation in which they are implanted. Essentially required biomaterial properties for such a one-step procedure include prompt and robust cell adhesion to ensure an adequate initial cell seeding efficiency. In general, the promotion of cell adhesion onto surfaces of biomaterials is of great interest for TE applications. Many attempts have been made to tailor material surfaces with bioactive molecules such as cell-binding peptide sequences including arginine– glycine–aspartic acid (RGD) or biomolecule-derived substances such as gelatin or fibronectin to improve cell adhesion [6,7]. The search for efficient strategies to improve cell–material adhesion is not restricted to the TE field but comprises a wide range of biopharmaceutical applications, e.g. the surface modification of polymeric drug delivery systems to promote the interaction with the intended treatment site [8]. In this context, the capability of carbohydrate-specific lectins to mediate site-directed targeting has been extensively investigated [9–11]. Lectins are generally defined as proteins or glycoproteins of non-immunological origin that bind carbohydrates [12], including

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

glycosylated membrane components (glycoproteins and glycolipids) exposed at the surface of mammalian cells. In previous studies [13,14], the dose-dependent improvement of cell adhesion on plastic culture dishes pre-treated with diluted solutions of various lectins was shown. Besides coating of material surfaces solely by physical interactions, lectins have also been chemically coupled to various materials, without interfering with the carbohydratebinding capability [11,15,16]. Here, we describe the functionalization of silk fibroin (SF) with wheat germ agglutinin (WGA), a plant lectin isolated from Triticum vulgare. This non-toxic, dietary glycoprotein shows binding affinity primarily for N-acetyl-D-glucosamine and sialic acid residues. Both sugar structures are ubiquitous in the glycocalyx of mammalian cells and therefore might provide versatile targets for adhesion improvements. In the last decade, SF has attracted attention as a biomaterial in various TE applications due to its mechanical properties, processability in various forms and its good biocompatibility [17–19]. Dependent on the preparation process, two types of SF are distinguished in TE applications: regenerated SF and native SF cocoon fibres. In both cases, the inherent inflammatory-response-eliciting sericin must be extracted. This outer gum-like cover of raw silk fibres is typically removed by a so-called degumming process, most often via boiling in alkaline solutions. SF is a protein of more than 5000 amino acids mainly composed of repetitive motifs of glycine and alanine. Nevertheless, SF does contain a sufficient fraction of reactive amino acids such as aspartic and glutamic acid, tyrosine, serine and threonine, all accessible for chemical modifications [20]. As the modification of SF with other biomolecules via carbodiimide chemistry has shown excellent results in previous studies, this form was chosen in this experiment. For example, carbodiimide binding chemistry has been used to couple SF with bone morphogenetic protein 2 (BMP-2) to induce bone formation [21,22], and to RGD peptide to promote cell attachment [23,24]. Moreover, the carbodiimide-mediated coupling of WGA was shown to result in stable immobilization of the dimeric protein with at least half of the binding domains being freely accessible for cell-associated carbohydrates [25]. In the current study we focused on working with adipose derived stromal cells (ASCs) as this cell type can be obtained by minimally invasive surgical procedures in acceptable cell numbers allowing one-step procedures. Besides ASCs we also worked with primary anterior cruciate ligament fibroblasts (ACLFs) as we want to use these primary cells in future ligament TE studies of our group. In upcoming examinations we envision to use the herein described WGA-modified SF as raw material for ligament scaffolds. We hypothesized that the coupling of WGA to SF would result in a robust and significantly higher cell adhesion compared to non-treated SF, facilitating the potential use of SF in one-step surgical procedures for tissue regeneration. 2. Materials and methods If not indicated otherwise, all reagents were purchased from Sigma (Vienna, Austria) and were of analytical grade. Fluorescein-labelled and pristine WGA was obtained from Vector Laboratories (Burlingham, CA, USA). HEPES (2-(4-(2-hydroxyethyl)-1 piperazinyl)-ethane-sulfonic acid) and urea were bought from Merck (Darmstadt, Germany).

2507

as described previously, by cooking in Na2CO3 solutions [26]. Bundles of silk fibres have been used to study the modification of silk scaffolds with WGA. To prepare silk films the degummed fibres were dissolved by boiling them in a ternary system consisting of calcium chloride, ethanol and ddH2O in a molar ratio of 1:2:8 for 6 h. The solution was then filtered (0.22 lm, RotilaboÒ, Roth (Karlsruhe, Germany) and dialysed against ddH2O using a Slide-a-LyzerÒ dialysis cassette form Pierce Biotechnology (Rockford, USA) with a molecular weight cutoff of 3.500 Da. The aqueous SF solution was lyophilized and the regenerated SF dissolved in hexafluoro-2-propanol to give a SF solution of 25 mg ml 1. This solution was used to prepare silk films in 96-well microtitre plates according to Sofia et al. [21]. In detail, 24 ll of the 25 mg ml 1 silk solution was used to coat one well of the 96-well plate, resulting in 600 lg dry fibroin per well. Prior to surface modification, SF films were fixed with MeOH/ddH2O (9+1) to induce the irreversible rearrangement of b-sheet structures, and air-dried under atmospheric pressure. 2.2. Modification of SF films/fibres with glycine, bovine serum albumin and WGA via carbodiimide chemistry Both films and fibres were hydrated in 20 mM HEPES/NaOH (pH 7.0) overnight (o/n) and thoroughly washed in the same buffer. Carboxyl groups were activated by immersing the films/fibres with a solution of 5.0 mg ml 1 1-ethyl-3-(dimethylaminopropyl) carbodiimide hydrochloride (EDC) and 7.0 mg ml 1 N-hydroxysuccinimide (NHS) in 20 mM HEPES/NaOH (pH 7.0), and incubated for 2 h. After activation of the films/fibres, the supernatant was discarded and the films/fibres were washed with 20 mM HEPES/NaOH (pH 7.0). SF films/fibres were then incubated o/n with a solution of either glycine (0.1 mg ml 1), BSA (0.015 mM) or WGA (0.015 mM) in HEPES/NaOH (pH 7.4) at 4 °C. In order to demonstrate the independency of enhanced cell adhesion due to the treatment of the SF films with the coupling reagents the glycine-modified SF was used as control group in the adhesion and proliferation experiments. We further included BSA-modified SF as control groups in the experiments in which possible cell behaviour changes (differentiation potential, in vitro immune response test) due to the modification process were investigated. After rinsing to remove excessive ligands, BSA- and WGAmodified films/fibres were immersed in a 0.1 mg ml 1 solution of glycine in HEPES/NaOH pH 7.4 for 30 min to saturate unreacted coupling sites. Finally, all films/fibres were thoroughly rinsed with 6 M urea and 20 mM HEPES/NaOH (pH 7.4), prior to storage in 20 mM HEPES/NaOH (pH 7.4) at 4 °C for up to 2 weeks. Control samples reacting with the same amount of fluorescein-labelled lectin (fWGA) instead of WGA were included in each batch as a readily traceable ligand to check for homogeneity and reproducibility of the coupling procedure. Images of silk fibres were taken using a Zeiss Epifluorescence Axio Observer.Z1 deconvolution microscopy system (Carl Zeiss, Oberkochen, Germany) equipped with LD Plan-Neofluar objectives and the LED illumination system ’’ColibriÒ’’. fWGA was monitored at excitation/emission 485/ 525 nm, and overlays with differential interference contrast images were acquired to facilitate spatial orientation. Samples incubated via the same protocol but without prior activation with EDC/NHS served as a control to assess unspecific adhesion of fWGA to the silk protein. For all samples, imaging conditions were kept constant in order to allow for direct comparison.

2.1. Silk fibres and films

2.3. Cells

White raw Bombyx mori silkworm fibres of 20/22 den, 250 T m 1, were purchased from Testex AG (Zürich, Switzerland). Prior to experiments the fibres were extracted from silk sericin,

2.3.1. NIH/3T3 The NIH/3T3 cell line was purchased from ECACC (European Collection of Cell Cultures, UK) and served as a control. NIH/3T3

2508

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

cells were cultured in Dulbecco’s modified Eagle medium (DMEM) containing 10% fetal calf serum (FCS) (Lonza Ltd., Basel, Switzerland) supplemented with 2 mM L-glutamine, 100 U ml 1 penicillin and 0.1 mg ml 1 streptomycin in plates coated with 0.2% gelatine solution. 2.3.2. ASC Subcutaneous adipose tissue was obtained during out-patient liposuction procedures under local tumescence anesthesia (IRB consent obtained). As previously described [27], adipose-derived stromal cells were isolated and cultured in medium containing DMEM–low glucose/HAM’s F-12 supplemented with 10% FCS (PAA, Pasching, Austria), 2 mM L-glutamine and 1 ng ml 1 recombinant human basic fibroblast growth factor (rhFGF, R&D Systems, Minneapolis, USA). ASCs up to passage 3 were used in the experiments for this study. 2.3.3. ACLFs The ethical review board of the AUVA (IRB) approved the procedure of collecting ACL tissue from patients undergoing total ACL reconstruction. Anterior cruciate ligament fibroblasts were obtained by a modified method based on a protocol of Nagineni et al. [28]. Briefly, ACL tissue was harvested and minced aseptically into small pieces of 1–3 mm3. Then these explants were transferred into 100 mm cell culture petri dishes and cultured in DMEM containing 10% FCS (PAA, Pasching, Austria), 2 mM L-glutamine, 100 U ml 1 penicillin and 0.1 mg ml 1 streptomycin. Cell outgrowth from the explants was monitored until near-confluency and then cells were trypsinized and used for further culture. In this study ACLFs from the cultures at the second to fourth passage were used for experiments. 2.3.4. Peripheral blood mononuclear cells (PBMCs) PBMCs were isolated from whole blood of healthy donors using density gradient centrifugation with Ficoll LSM 1077 (PAA, Pasching, Austria) in LeucoSep tubes (Greiner Bio-One GmbH, Kremsmünster, Austria) according to the manufacturer’s instructions. Then isolated PBMCs were resuspended in PBMC medium (RPMI 1640 (PAA, Pasching, Austria), 10% FCS, 2 mM L-glutamine (PAA, Pasching, Austria), penicillin/streptomycin solution (100 U ml 1 penicillinG and 0.1 mg ml 1 streptomycin (PAA, Pasching, Austria)) at 2  106 cells ml 1. 2.4. Determination of the WGA binding capacity of NIH/3T3, ACLF, ASC After harvesting the cells by trypsinization, 50 ll of a pre-cooled cell suspension (6  106 cells ml 1) were incubated with 50 ll of a fWGA solution (500 pmol ml 1 in phosphate-buffered saline (PBS + Ca/Mg, pH 7.4) for 30 min at 4 °C. To remove unbound fWGA, cells were washed three times with PBS and the relative cell-associated fluorescence intensity (RFI) was determined using an EPICS XL-MCL™ analytical flow cytometer (Beckman Coulter, Vienna, Austria) at 488/525 nm. A forward vs. side scatter gate was used to detect the single cell population and exclude debris, 2  103 cells were accumulated per analysis. The mean channel number of the logarithmic intensities of the individual peaks was used for further calculations; all assays were repeated in triplicate. To compensate for cell autofluorescence and exclude excessive aggregation phenomena induced by lectin binding, control samples with unlabelled cells were included. 2.5. Cell size determination The average size of NIH/3T3, ACLF and ASC cells was determined via differential interference contrast microscopy on a Zeiss Epifluorescence Axio Observer.Z1 deconvolution microscopy

system (Carl Zeiss, Oberkochen, Germany). Freshly trypsinized cells in single cell suspension (isotonic HEPES pH 7.4 buffer) were dispersed in LabTecÒ chamber slides™ (Electron Microscopy Sciences, Hatfield, USA) and the outer diameter was determined after settling via the integrated and pre-calibrated distance measurement tool. 50 cells were accumulated per measurement and all samples were analysed in triplicate. 2.6. Quantification of cell attachment A CyQUANT-NF cell proliferation assay kit (Invitrogen, Vienna Austria) was used to determine the number of cells bound to the surface of non-treated and modified SF films according to manufacturer’s protocol. Briefly, 100 ll of a Cyquant-NF solution was added to each well and incubated for 60 min in the dark. After incubation, the fluorescence of the wells was measured using a fluorometer (POLARstar Omega, LABTECH GmbH, Ortenberg, Germany) at 485/520 nm (ex/em), respectively. The number of cells attached to each SF film was determined using a standard calibration curve prepared with a cell suspension of known total number of ASCs. For each type of modification, at least four replicates were analysed. 2.7. Attachment of ASCs on the surface of SF films and SF fibres SF films with and without surface modification were incubated with ASCs by pipetting 40 ll of a 1  106 cells ml 1 ASC cell suspension on each silk film. The plates were incubated for 20 min at 37 °C in an atmosphere of 5% CO2 and 95% humidity. At the end of the incubation period the films were thoroughly rinsed with sterile PBS. To minimize variations in the mechanical stimulus caused by flushing with PBS, a MultipetteÒ Xstream plate washer (Eppendorf, Vienna, Austria) was used. Due to its automatic dispension and aspirating mechanisms, this electronic hand dispenser guarantees consistent rinsing of all wells. Then, 100 ll of ASC culture medium was added to each well. After 5 h of incubation at 37 °C (5% CO2 and 95% humidity) the number of cells attached to each silk film was quantified using a CyQuant-NF assay, as described above. In addition to silk films, the cell attachment to modified SF fibres was also investigated. Therefore, these SF fibre constructs were incubated for 5 min in a well of a 96-well plate containing 100 ll of an ASC cell suspension of 1  106 cells ml 1 under constant agitation. Then the fibres were rinsed thoroughly with PBS and immediately fixed in a 10% buffered formalin solution for propidium iodide (PI) nuclear staining. The fibres were incubated for 20 min with 0.25% Triton X-100 followed by incubation with PI solution for 10 min. Before observing the cells with an epifluorescence microscope Leica DMI6000B (Leica Microsystems GmbH, Wetzlar, Germany) samples were carefully washed with PBS. 2.8. Proliferation In a 24-well tissue culture plate ASCs were seeded at a density of 2  103 cells per well and then cultured for 4 days at 37 °C in a 5% CO2 incubator and on the fifth day proliferation was determined using a 5-bromo-2-deoxyuridine (BrdU) uptake assay (Cell Proliferation ELISA assay Kit, Roche Diagnostics, Vienna, Austria) according to the manufacturer’s instructions to demonstrate that the adhesion of ASCs on the WGA-SF does not influence proliferation behaviour. This assay is based on the measurement of BrdU incorporation (a thymidine analogue) during DNA synthesis in replicating cells and can therefore be used to determine inhibitory effects of various compounds/factors on cell proliferation. Furthermore, this assay has been widely used in measurements of the immunoreactivity of lymphocytes, stimulated by antigens

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

or mitogens. Briefly, cells were cultured with respective media containing 100 lM BrdU for 12 h at 37 °C. Then media was removed and the culture plates were air-dried for 15 min. FixDenatÒ solution was added for 30 min followed by the incubation with anti-BrdU POD (peroxidase) antibody for 60 min at room temperature (RT). The plate was washed three times with PBS and tetramethyl benzidine, as substrate, was added for 30 min. Then the reaction was stopped by adding 1 M H2SO4 and the absorption was measured at 450 nm against 690 nm as reference wavelength.

2509

as previously described [27] with osteogenic medium (OM) described by Pittenger et al. (DMEM (PAA, Pasching, Austria), 10% FCS (PAA, Pasching, Austria), 1% penicillin/streptomycin (PAA, Pasching, Austria), 1% L-glutamine (PAA, Pasching, Austria), 50 mM ascorbate-2-phosphate, 0.1 mM dexamethasone, 10 nM 1,25-dihydroxy-vitamin D3, 10 mM b-glycerophosphate (Stem Cell Technologies)) [30]. As a control, cells were cultured in control medium (CM, DMEM, 10% FCS, 1% penicillin/streptomycin, 1% L-glutamine). In all groups, medium was changed every 2–3 days for a total of 3 weeks.

2.9. Induction of resistance to trypsin and mechanical stimulus 2.11. Alkaline phosphatase (ALP) activity SF films with and without modifications were seeded with ASCs at a density of 2.5  105 cells per well for 24 h at 37 °C in a 5% CO2 environment. Before exposition of adhered cells to 0.05% trypsinethylenediaminetetraacetic acid (EDTA) (1; Life Technologies, Vienna, Austria), used as a protease model, cells were washed twice with PBS. After addition of trypsin, the wells were put on a mechanical shaker (Unimax 1010, Heidolph Instruments GmbH, Schwabach, Germany) for at least 10 s (in a range of 20 mm width at 30 rpm) between taking pictures at defined time points (30 s, 1 min, 2 min, 4 min, 8 min) using an epifluorescence microscope (Leica DMI6000B, Leica Microsystems GmbH, Wetzlar, Germany) equipped with a heating plate insert controlled by a temperature regulator (Pecon Tempcontrol 37, PeCon GmbH, Erbach, Germany). At the 4 min timepoint supernatants of the wells containing the released cells from the SF surface were collected. Their cell numbers were counted using a cell counter (TC20™ Automated cell counter, Bio-Rad Laboratories GesmbH, Vienna, Austria) and the percentage of cells remaining on the SF films to initial cells was calculated. Microscope glass slides were covered with a custom-made silicone sheet in which six channels had been cut out according to the geometry of the ibidi sticky-Slide VI0.4 flow chambers (Ibidi, Munich, Germany). This silicone sheet served as a mould to cover the glass surface with strips of SF and WGA-decorated SF film (SF-WGA), respectively. Then the self-adhesive underside of the bottomless blank slide of the sticky-slide chambers was adapted to the SF-coated glass slides. ASCs were harvested and incubated in the channels at 1  106 cells ml 1 for 20 min. Then cells were exposed to fluid flow in a ramped pattern by a steadily increasing delivery of 0–20 ml min 1 PBS using a syringe pump (PHD Ultra™, Harvard Apparatus) for a total time of 200 s (Fig. 5). This fluid flow induces shear stresses ranging from 0 to 28 dyn cm 2. During the exposition to fluid flow, the cells on the SF films were observed via an epifluorescence microscope Leica DMI6000B (Leica Microsystems GmbH, Wetzlar, Germany) equipped with a heating plate insert controlled by a temperature regulator (Pecon Tempcontrol 37, PeCon GmbH, Erbach, Germany). Pictures were taken of the remaining cells every 2 s and converted into a time-lapse video with a frame rate of 2 pictures s 1 (see Supplemental data) using NIH ImageJ (Version 1.47, http://rsb.info.nih.gov/ij/). To perform overlay pictures, pictures from before and after the applied fluid flow were taken. Using Matlab 7.1 (The MathWorks Inc., Natick, USA) the images were transformed to greyscale, inverted and layered into an RGB image where the blue and grey layers correspond to the final state and the red layer the initial state. A red boost filter was applied to highlight the released cells during the flow experiment.

To quantitatively assess the osteogenic differentiation and to test whether the SF-coupled WGA interferes with differentiation, ALP assays were performed at the end of the culture time. Before measurement the cells were frozen at 25 °C and then lysed with 100 ll per well (96-well plate) of an ALP-assay buffer (2 mM MgCl2, 0.5 M 2-amino-2-methyl-1-propanol, pH 10.5) containing 0.25% Triton X-100 for 1 h at RT. After clearing the supernatants by centrifugation (13.000g, 10 min), 50 ll of 20 mM p-nitrophenylphosphate in ALP assay buffer was provided as substrate. In the enzymatic reaction, p-nitrophenylphosphate is converted to pnitrophenol by ALP. After 20 min the reaction was stopped by the addition of 50 ll of 0.2 M NaOH. The absorbance at 450 against 690 nm was read immediately thereafter on an automatic microplate reader (Spectra Thermo, TECAN Austria GmbH, Vienna, Austria). 2.12. von Kossa staining for mineralization After 3 weeks of differentiation, von Kossa staining for the deposition of minerals in the culture was carried out. The cells were washed three times with 1  PBS without calcium and magnesium (PAA, Pasching, Austria). After each of the subsequent working steps, cells were washed three times with ddH2O. After fixation with 10% neutral buffered formalin the cells were stained with a 5% (w/w) AgNO3 solution for 30 min at RT. Then the samples were developed with 5% (w/w) Na2CO3 in a 25% aqueous solution of formaldehyde. After a final fixation in a 5% (w/w) Na2S2O3 solution in ddH2O for 2 min the stained samples were observed by phase-contrast microscopy (Leica DMI6000B, Leica Microsystems GmbH, Wetzlar, Germany). 2.13. In vitro immune response test To test immunomodulatory properties of non-modified SF as well as WGA-modified SF, plates were coated with the respective materials according to the method described above. In detail, 5  104 PBMCs were seeded in 96-well flat bottom plates (Greiner Bio-One GmbH, Kremsmünster, Austria) in a final volume of 100 ll PBMC medium per well, in the respective coated wells. On day 5, 10 mM BrdU was added to each well, and BrdU enzyme-linked immunosorbent assay (ELISA; Roche, Mannheim, Germany) was performed on day 6 according to the manufacturer’s instructions. As an additional immune response test, using activated PBMCs, a phytohemagglutinin (PHA) activation assay modified from Le Blanc et al. [29] was performed. PBMCs were cultured as described above but were activated with 5 mg ml 1 PHA starting on day 3.

2.10. Differentiation potential 2.14. Statistical analysis To test the differentiation capability of ASCs on the modified SF surfaces, 1  103 ASCs were seeded in 96-well flat bottom plates (Greiner Bio-One GmbH, Kremsmünster, Austria), coated with SFWGA or unmodified SF respectively, in a final volume of 100 ll medium per well. In ASCs, osteogenic differentiation was induced

All calculations were performed using GraphPad software (GraphPad software, Inc., San Diego, CA, USA). Normal distribution of data was tested with the Kolmogorov–Smirnov test. One-way analysis of variance followed by Tukey´s post hoc test was used

2510

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

to conduct statistical significance and P-values below 0.05 were considered statistically significant. All graphs in this study are shown as mean ± standard deviation (SD). 3. Results 3.1. WGA-binding capacity of ASC, ACLF and NIH/3T3 cells To evaluate the binding capacity of NIH/3T3, ACLF and ASC single cells for WGA, a constant number of cells (3  105) was incubated with serial dilutions of fluorescein-labelled WGA. As illustrated in Fig. 1, increasing amounts of WGA led to a rise in the mean cell-associated fluorescence intensity, independent of cell type. The average cell-bound fluorescence levels for ASC and NIH/3T3 cells showed no substantial difference, ranging from 39.5 ± 1.35 to 129.8 ± 5.05 and from 38.5 ± 2.51 to 127.6 ± 2.93, respectively. ACLF cells, in contrast, were characterized by a significantly higher lectin binding potential, amounting from 74.3 ± 2.52 to 219.0 ± 6.08 for the highest fWGA concentration. As expected for a receptor-mediated interaction process, binding curves gradually levelled off with increasing lectin concentration. Still, the fully saturated state was not reached under the given conditions. In all experiments, autofluorescence levels were in the range of 0.63 ± 0.63. To exclude the possibility that differences in cell size and thus cell surface area are responsible for variances in the amount of bound lectin, cell size measurements have been performed. These measurements indicated that there was no statistically significant difference in cell size among the investigated cell types. 3.2. Coupling of WGA to SF To analyse the influence of crosslinker EDC/NHS concentration on the binding density of WGA to the SF surfaces, a constant area (per well of a 96-well plate) was activated with increasing amounts of EDC, followed by addition of constant amounts of fWGA. In both conditioning cases, addition of 0.1 mg ml 1 or 0.5 mg ml 1 f-WGA, binding of WGA correlated in a dose-dependent manner to the crosslinker concentration. However, above the addition of 2.5 mg ml 1 EDC, a saturation effect (Fig. 2) could be seen. 3.3. ASC adhesion properties to WGA-modified SF films The capacity of WGA-decorated SF films to initiate cell adhesion of ASC was quantitatively examined. Therefore, SF-WGA as well as

Fig. 2. Influence of the crosslinker concentration on the fWGA density at the SF surface. All data represent means of at least three independent experiments ± SD, ⁄⁄⁄ significant from 0.1 mg ml 1 fWGA group (P < 0.001).

control groups consisting of SF and tissue culture polystyrene (TCPS) were incubated with a cell suspension of a distinct cell concentration (1  106 cells ml 1) for 20 min at 37 °C in 5% CO2 and 95% humidity. After incubation the surfaces were rinsed two times with PBS. Fig. 3A shows representative bright-field microscopy images of these experiments that confirmed the mediated cell adhesion as a result of WGA conjugation to SF films. The glycinedecorated silk compared to pure SF showed a tendency towards an elevated cell adhesion capacity but was not significant. However, on the WGA-modified SF, the adhesion of ASCs was remarkably enhanced compared to pure SF and glycine-decorated SF (Fig. 3A). In the WGA-SF groups ASCs showed an even distribution over the surface with a high number of cells compared to SF and glycine-coupled SF (SF-Gly) films. Precisely, a more than 17fold increase in initial cell density could be observed owing to the modification with WGA, when compared to pure SF. 3.4. Proliferation assay ASC on SF-modified films The representative histogram in Fig. 3B which shows the proliferation of ASC cells by incorporation of BrdU, demonstrating differences between cells cultured 4 days on tissue culture plastic substratum and on the SF films, treated (either chemically coupled with glycine, BSA or WGA) or non-treated. More importantly, no significant difference in proliferation could be detected between

Fig. 1. Concentration-dependent binding of fWGA to single cells (A): ASCs, ACLFs and fibroblastic cell line NIH/3T3. (B) Cell size measurements showing no significant differences in cell size among the investigated cell types. All data represent means ± SD of at least three independent experiments, ⁄⁄⁄significant from ACLF and NIH/3T3 groups (P < 0.001).

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

2511

Fig. 3. (A) Cell adhesion of ASCs to SF-WGA, determined by measuring DNA content of cells and by microscopical images. The different surfaces; TCPS, pure SF film, SF-Gly and SF-WGA were incubated for 20 min with an ASC cell suspension of 1  106 cells ml 1 succeeded by two washing steps with PBS to remove loose cells. All data are means of at least seven independent experiments ± SD. White arrows exemplarily indicate single cells. Scale bars are 500 lm. ⁄⁄⁄ indicates significant difference of P < 0.001, ⁄ indicates significant difference of P < 0.05. (B) Proliferation of ASCs on SF-WGA, SF-BSA, SF-Gly, pure SF and TCPS. Identically, all the different surfaces were seeded with 2  103 cells per well (24-well plate) and after 4 days’ proliferation was measured using a BrdU-assay. Bars indicate the mean values ± SD, n = 6.

the modified and unmodified silk groups. Among the different modified silk films a basal proliferation could be observed. 3.5. Resistance to proteases and mechanical stress of lectin-mediated cell adhesion Plain and WGA-modified SF films were seeded with ASCs for 24 h, washed twice with PBS and then exposed to trypsin–EDTA. ASCs cultured on lectin-functionalized SF films showed enhanced trypsin resistance compared to cells seeded on non-modified SF films. In plain SF films ASCs almost instantly became round and started to disperse from the film surfaces after 30 s. After 4 min almost 79% of the initially seeded cells were detached from plain SF films whereas in contrast in the WGA-SF group only 20% of cells were dispersed (Fig. 4). The modification of silk films with WGA provides resistance to trypsin–EDTA, the cells still adhered to the film surface even 8 min after starting the protease treatment. Besides investigating the effects of lectin modification on resistance to proteolysis, an experimental set-up was established to have a closer look at the impact on mechanical stability of the lectin-mediated cell adhesion. Therefore the substrata of commercially available fluid flow chambers were coated with SF and SFWGA, respectively, and seeded with ASCs for 20 min. Then fluid flow in a ramped profile (Fig. 5B) was applied to the cells and the cells on the film surface investigated by taking microscopic images (Fig. 5). In both groups a part of cells immediately gets flushed

away from the field of vision as they were not able to adhere because they were not in contact with the surface (in suspension). But with increasing fluid flow also cells on the film surface (in the focus plane) start to detach, especially in the non-modified SF group. In Fig. 5C (bottom panel) the cells released from the surfaces are coloured in red, cells still adherent to the substrate but slightly changed their position are coloured in blue and cells kept their position from the initial incubation step are coloured in grey.

3.6. Osteogenic differentiation potential To show that binding of cells via WGA does not lead to an inhibition of differentiation capabilities, ASCs were seeded on SF-WGA films and cultured in osteogenic medium. The osteogenic differentiation was determined by the expression of ALP and by von Kossa mineralization staining. As shown in Fig. 6A, ALP levels were significantly elevated in the presence of osteogenic medium in the SFWGA as well as the plain SF and TCPS group compared to cells cultured with control medium on plain SF and SF-WGA. No significant difference in ALP induction between SF-WGA OM and TCPS OM group was observable. Besides the ALP assay, von Kossa staining was performed to show the positive induction of osteogenic differentiation. After 21 days of culturing ASCs on SF and WGA-modified SF films in osteogenic medium, a high deposition of minerals in the culture could be observed (Fig. 6B).

2512

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

Fig. 4. Release of ASCs from the pure SF and WGA-SF films due to the treatment with trypsin–EDTA. Both SF films were seeded with 2.5  105 cells per well (24-well plate) for 24 h, washed twice with PBS and then treated with trypsin–EDTA. After 30 s, 1 min, 2 min, 4 min and 8 min representative microscopic images were taken. Scale bars are 500 lm. For the 4 min timepoint the number of cells remaining on the substratum was determined by counting the released cells. The data are shown as the mean ± SD. n P 4, ⁄⁄⁄ indicates significant difference of P < 0.001.

3.7. Immune responses in vitro To address the concerns about the possible immune response elicitation, the stimulation effect of WGA-modified SF on lymphocyte proliferation was investigated. As expected, the proliferation response of PBMCs was significantly higher in co-cultures with SF material compared to tissue culture plastic substratum (Fig. 7A). In parallel, PBMCs have been prestimulated with PHA, a well-known mitogen for PBMCs and cultured on the various substrata. In contrast to the unstimulated co-cultures of PBMCs (Fig. 7A), the investigated SF materials, modified or non-modified (chemically coupled with glycine, BSA or WGA), decreased the proliferation response of the activated PBMCs to PHA compared to the tissue culture plastic group (Fig. 7 B). However, no differences in proliferation-inducing properties among the untreated SF and the chemically modified SF types could be detected. 3.8. Coupling of WGA to degummed SF fibres The ability to covalently immobilize WGA on silk fibres was demonstrated using fluorescein-labelled WGA (fWGA). The incubation of SF fibres with fWGA leads to a slight unspecific binding (Fig. 8 first row, second column), detectable as a sparse green stain-

ing under the fluorescence microscope. After an additional washing step with 6 M urea this background signal was eliminated (data not shown), indicating that this background signal can be reduced to unspecific binding of WGA to SF. In contrast to fWGA incubation without prior activation of accessible ester groups, EDC/NHS-mediated crosslinking results in a notable staining of the silk fibre with the fluorescent lectin. The stain persists after thorough washing with 6 M urea, which indicates stable and covalent conjugation of the fWGA protein to the fibroin molecules. 3.9. Propidium iodide staining of cells on SF-WGA fibres Propidium iodide staining was performed to visualize ASCs over the surface of the WGA-modified SF fibres. Because SF fibres are not transparent, the attached cells were fixed in formalin and stained with PI nuclear staining. The silk fibres were positioned on a slide (with the surface containing the cells) facing the light beam of the microscope. Fig. 9 confirmed the regular distribution of ASCs on the SF-WGA fibres after 5 min of incubation in a 1  106 cell suspension. PI staining showed a high number of cells observed on the WGA-SF fibres in comparison with unmodified SF fibres, where little to no cells could be observed. At a magnification of 10  the adherence of cells to the single silk fibres is obvious.

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

2513

Fig. 5. Release of ASCs from the pure SF and WGA-SF films due to the exposition to fluid flow. (A) Experimental fluid flow chamber slide set-up used to test the mechanical resistance of lectin-bound cells by applying fluid shear stress: (a) tubing connected with syringe pump (fluid in); (b) flow chamber placed on the stage of a microscope; (c) tubing connected to waste beaker (fluid out), (d) syringe for cell suspension supply via septum; (e, f) condenser head of microscope. (B) Fluid flow was applied in a ramped pattern by steadily increasing delivery of 0–20 ml min 1 (inducing fluid shear stresses from 0 to 28 dyn cm 2) PBS using a syringe pump for a total time of 200 s. (C) Respective microscopic images of the adherent cells (20 min incubation time, ASC cell suspension of 1  106 cells min 1) before (first row) and after (second row) the fluid flow exposition on SF and SF-WGA (see Supplementary videos 1 and 2). The third row shows a computationally generated overlay of the pictures above in which cells released, strongly adherent (not moved from their initial place) and low-adherent cells (still adherent to the film surface but slightly moved) are stained in red, grey and blue, respectively. Scale bars are 500 lm.

4. Discussion Driven by the idea to use SF as scaffold material in one-step surgical procedures in TE approaches, we chemically immobilized the plant lectin WGA on SF. This immobilization leads to a remarkable improvement of cell adhesion to SF without alterations in proliferation and differentiation behaviour of ASCs. Furthermore, in vitro

immune response tests were performed, indicating no induction of the proliferation of PMBCs in vitro. In previous studies modification of biomaterials with lectins were described. Most of them were in the area of drug delivery systems. For instance, the preparation of WGA-grafted PLGA nanoparticles as suitable drug vehicle is described by Weissenböck et al. [11]. In cartilage TE, the potential application of WGA to modify

2514

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

Fig. 6. Differentiation capability determined by ALP assay (A) of ASCs cultured in OM/CM, on either plain SF, SF-WGA or TCPS films. Bars indicate the mean values ± SD. n P 4, (B) Respective von Kossa mineralization stainings of ASCs cultured in control medium on WGA-modified SF film (SF-WGA CM), ASCs cultured in osteogenic medium on plain silk fibroin film (SF OM) and ASCs cultured in osteogenic medium on WGA-modified SF film (SF-WGA OM). No statistical differences between SF and SF-WGA films could be observed. Scale bars are 200 lm.

Fig. 7. In vitro immune response tests: proliferation of PBMCs on pure SF, SF-WGA, SF-Gly, SF-BSA and TCPS either (A) unstimulated or (B) stimulated with phytohaemagglutinin. The data are shown as the mean ± SD. n = 8.

low adhesive materials has been proposed [31] due to WGA’s superior adhesion to chondrocytes cell lines compared with other lectins. To the best of our knowledge, the only study showing the improvement of biomaterials as a result of chemical WGA conjugation to its surface is from Wang et al. [15]. We established a protocol for decoration of silk surfaces with WGA using carbodiimide chemistry. The binding studies of fWGA to SF films showed successful coupling of WGA (Fig. 2). In parallel, the binding capacities of ASCs and ACLFs, intended to be used in future ligament TE approaches to WGA, have been determined (Fig. 1A). The fibroblastic cell line NIH/3T3 served as additional control. Apparently, isolated ACLFs yielded remarkable higher fluorescence intensities (=labelled WGA binding) in the saturation experiments compared to ASCs and the fibroblastic cell line NIH/ 3T3, indicating higher lectin binding. These findings suggest that ACLF cells present higher amounts of the carbohydrate binding motifs for WGA N-acetyl-D-glucosamine and sialic acid on their surface. By microscopic cell size measurements we could exclude the possibility that the higher binding intensities are simply due to differences in the cell surface areas (Fig. 1B). WGA-modified scaffolds may thus be characterized by a preferential adherence

of those cells that differentiate into the ligament/osteogenic linage, leading to better tissue ingrowth and a lower tendency for fibroblastic encapsulation or chronic immunogenic reactions. As a result of our study we propose that the modification of SF with WGA improves the cell-adhesion capacity of SF in a magnitude that would enable cell seeding in a one-step surgical procedure within an acceptable time frame. Indeed, we could observe a dramatic increase of adherent ASCs on WGA-SF compared to pure SF (Fig. 3A) after a cell seeding time of only 20 min. The adhesion of cells to non-modified SF is not attributed to a specific biorecognition site but may be mainly based on electrostatic interaction between cells and SF [32]. The slight improvement of SF-Gly compared to pure SF might result from the coupling procedure, which possibly induces a surface change in the topography or electrostatic characteristics. Moreover, we could demonstrate that this cell-adhesion improvement, as a result of WGA modification, is also feasible for SF scaffolds based on native and non-regenerated silk fibres (Figs. 8 and 9). Degummed natural SF fibres are recognized, especially in ligament TE strategies [33–35,23], as these fibres still have superior mechanical properties due to the natural fibre spinning process [36] compared to regenerated SF.

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

2515

Fig. 8. Representative fluorescence micrographs of covalent modification of SF fibres with fWGA. SF fibres were incubated with fWGA without (upper panel) or with (lower panel) prior activation of reactive coupling sites by EDC and NHS. Images show individual and merged channels of differential interference contrast microscopy, and fluorescence microscopy (ex/em 485/525 nm). Exposure time and imaging conditions were kept constant in order to allow for direct comparison. Scale bars are 500 lm.

In the already mentioned study of Wang et al. [15], both an enhancement of cell adhesion to WGA-grafted chitosan films and an increased proliferation rate of fibroblasts on the modified chitosan surface were demonstrated. In our study, the WGA modification did not lead to an enhanced cell proliferation in comparison to plain SF (Fig. 3B). A likely explanation is the difference between chitosan and SF in supporting cell spreading and migration. Chitosan is known to impair cell spreading and is therefore often blended with other biomaterials, especially with gelatin [37,38]. In contrast, SF has been proven to exhibit cell attachment and growth comparable to collagen [32]. We assume that WGA molecules on a chitosan surface contribute as additionally added cell interaction points that drastically improve cell attachment and proliferation properties. On SF, in contrast, the cell growth capacity may already exhibit a level that is not able to be further enhanced by WGA modification. Rather, the lectin-mediated interaction may lead to a sorting effect with regard to the cell type(s) most prone to surface attachment (like ACLF). In a study of Nishimura et al. [14] various plant lectins, including WGA, were used in their unbound form to increase cell adhesion to a variety of material surfaces. Moreover, in this study a higher resistance of the cell adhesion to proteases as well as to mechanical stimulations was described. As according to Nishimura et al. [14] we used trypsin as a protease model and we could also observe a phenotypic change from spindle-shaped to roundish, although the cells still remained on the SF-WGA. Similarly, we could observe a higher cell-binding force to WGA-modified SF surface under applied fluid shear stresses (up to 28 dyn cm 2). Hence, the modification not only reduces cell adhesion time but could also preserve adhered cells to withstand mechanical forces during the surgical procedure and the handling of a cell-loaded scaffold as well as in the defect site where, once implanted, it has to withstand forces of adjacent tissue or of body fluids. Many TE strategies are based on the combination of biomaterials and progenitor/stem cells. The basic idea behind the use of stem cells is that these progenitor cells have the potential to differentiate into the desired cell type and to produce new extracellular matrix at the defect site. A resulting prerequisite therefore is that the used biomaterial does not negatively influence the differentiation process. Here, as a proof of concept the osteogenic differentiation capacity of ASCs on the SF-WGA was verified by ALP activation

(Fig. 6A) and by von Kossa mineralization staining (Fig. 6B). We concentrated on osteogenic differentiation due to the lack of ligament-specific differentiation markers. These results indicate that there is no negative interference with the mesenchymal differentiation potential on the WGA-modified SF, at least in the osteogenic lineage. These results are in accordance with a study of Nishimura et al., where other plant lectins (ConA and PHA) did not alter chondrogenic or osteogenic differentiation in mesenchymal stem cells of bone marrow origin [14]. Restricted to the effect of WGA on differentiation potential of mesenchymal stem cells, findings differ; just recently Talaei-Khozani et al. [39] showed that in contact with WGA, chondrogenesis and ossification decreased in limb bud stem cells, whereas in another study the production of bone-specific proteins [40] by bone marrow stem cells after binding WGA is described. All these studies have to be compared with caution with our results, as the majority of the studies expose the cells to unbound lectins in solution. In this free form, the interaction of lectins with cells is not limited to cell-membrane binding, but as demonstrated [10], can also result in lectin internalization, potentially leading to dynamic alteration/influence on intracellular trafficking and/or metabolic processes. We suggest preventing or at least minimizing these potential reactions by covalent binding of WGA to the SF surface. Due to the lectin immobilization via covalent bonds the immediate way of cell incorporation is restrained and would require a cleavage, for instance enzymatically, of the chemically generated bond between SF and WGA. To address the concerns of a potential immunogenic response upon WGA decoration, we evaluated the proliferation of PBMCs in the presence of SF-WGA compared to other SF modifications and plain SF. PBMCs represent a critical component of the immune system and have therefore often been used to test biocompatibility [41–44]. Seo et al. [45] concluded that in vitro PBMC cultures might be used to predict the in vivo biocompatibility of biomaterials. In their study SF showed higher cell compatibility in vitro and a lower inflammatory tissue reaction in vivo compared to polyglycolic acid. Our results do not indicate a higher proliferation rate of PBMCs in contact with WGA-modified SF as compared to nonmodified SF (Fig. 7A). Additionally, we observed the behaviour of activated PBMCs (after incubation with PHA) towards the modified and non-modified SF films (Fig. 7B). Comparable to the unstimulated experiment, we could not detect a significant difference in

2516

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

Fig. 9. Fluorescence micrographs of ASCs on pure SF or on SF-WGA. Pure SF fibres and SF-WGA were completely covered for 5 min with an ASC cell suspension of 1  106 cells ml–1 succeeded by two washing steps with PBS. Then cell nuclei were fixed and stained with propidium iodide. Images of the blue channel (blue) show autofluorescence of SF, cell nuclei were counterstained with propidium iodide (red channel; red). Blue/red indicates an overlay of the autofluorescence picture of silk (blue) and the propidium iodide stained cell nuclei (red). 5  and 10  indicate the applied magnification at the respective rows. Scale bars are 500 lm.

PBMC proliferation between WGA-modified and non-modified SF. Interestingly, the proliferation of PBMCs in all SF groups was significantly lower compared to TCPS control. The mechanisms behind this observation have not yet been fully elucidated, but we believe that the underlying effect results from the unspecific physical binding of PHA to SF. During the establishment of the WGA coupling to the SF we observed a strong unspecific binding of WGA, possibly mediated by electrostatic interactions. The same might account for PHA, and hinder it to bind to and crosslink cell surface glycoproteins (especially on T-cells), which impedes a stimulation of PBMC proliferation and thus an immune response. To summarize, in the in vitro immune tests we did not observe any activation effect on PBMCs. In future studies, we will test the WGA-SF biomaterial in vivo to verify its biocompatibility. Here, we will use intramuscular and subcutaneous implantation models. 5. Conclusion Our data demonstrate that the modification of SF with WGA improves the cell-adhesion capacity of SF in a magnitude that enables cell seeding in a one-step surgical procedure within an acceptable time frame and without causing adverse effects on proliferation and differentiation of ASCs or an induction of an in vitro immune

response in PBMCs. This tailoring feasibility and the resulting advances in cell adhesion might move SF scaffolds, intended to be used in various TE approaches, closer towards clinical application. Acknowledgments The authors would like to thank James Ferguson for native language review. The financial support by the City of Vienna (MA 27, Project 12-06 and MA 23, Project 14-06) and by the Austrian Research Agency FFG (Bridge 815471) is gratefully acknowledged. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figs. 5, 6, 8 and 9, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at http://dx.doi.org/10.1016/ j.actbio.2014.02.012. Appendix B. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2014.02. 012.

A.H. Teuschl et al. / Acta Biomaterialia 10 (2014) 2506–2517

References [1] Marlovits S, Zeller P, Singer P, Resinger C, Vécsei V. Cartilage repair: generations of autologous chondrocyte transplantation. Eur J Radiol 2006;57:24–31. [2] Batty L, Dance S, Bajaj S, Cole BJ. Autologous chondrocyte implantation: an overview of technique and outcomes. ANZ J Surg 2011;81:18–25. [3] Helder MN, Knippenberg M, Klein-Nulend J, Wuisman PIJM. Stem cells from adipose tissue allow challenging new concepts for regenerative medicine. Tissue Eng 2007;13:1799–808. [4] Jurgens WJ, Kroeze RJ, Bank R a, Ritt MJPF, Helder MN. Rapid attachment of adipose stromal cells on resorbable polymeric scaffolds facilitates the one-step surgical procedure for cartilage and bone tissue engineering purposes. J Orthop Res 2011;29:853–60. [5] Weber B, Scherman J, Emmert MY, Gruenenfelder J, Verbeek R, Bracher M, et al. Injectable living marrow stromal cell-based autologous tissue engineered heart valves: first experiences with a one-step intervention in primates. Eur Heart J 2011;32:2830–40. [6] Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as synthetic extracellular matrix materials. Biomaterials 1999;20:45–53. [7] Liu X, Won Y, Ma PX. Surface modification of interconnected porous scaffolds. J Biomed Mater Res A 2005;74:84–91. [8] Pisal DS, Kosloski MP, Balu-Iyer SV. Delivery of therapeutic proteins. J Pharm Sci 2010;99:2557–75. [9] Neutsch L, Plattner VE, Polster-Wildhofen S, Zidar A, Chott A, Borchard G, et al. Lectin mediated biorecognition as a novel strategy for targeted delivery to bladder cancer. J Urol 2011;186:1481–8. [10] Gabor F, Bogner E, Weissenboeck A, Wirth M. The lectin-cell interaction and its implications to intestinal lectin-mediated drug delivery. Adv Drug Deliv Rev 2004;56:459–80. [11] Weissenböck A, Wirth M, Gabor F. WGA-grafted PLGA-nanospheres: preparation and association with Caco-2 single cells. J Control Release 2004;99:383–92. [12] Barondes SH. Bifunctional properties of lectins: lectins redefined. Trends Biochem Sci 1988;13:480–2. [13] Rauvala H, Carter WG, Hakomori S. Studies on cell adhesion and recognition I. Extent and specificity of cell adhesion triggered by carbohyd rate-reactive proteins (glycosidases and lectins) and by fibronectin labeling of proteins. Cell 1981:88. [14] Nishimura H, Nishimura M, Oda R, Yamanaka K, Matsubara T, Ozaki Y, et al. Lectins induce resistance to proteases and/or mechanical stimulus in all examined cells–including bone marrow mesenchymal stem cells–on various scaffolds. Exp Cell Res 2004;295:119–27. [15] Wang Y-C, Kao S-H, Hsieh H-J. A chemical surface modification of chitosan by glycoconjugates to enhance the cell-biomaterial interaction. Biomacromolecules 2003;4:224–31. [16] Yin Y, Chen D, Qiao M, Wei X, Hu H. Lectin-conjugated PLGA nanoparticles loaded with thymopentin: ex vivo bioadhesion and in vivo biodistribution. J Control Release 2007;123:27–38. [17] Rockwood DN, Preda RC, Yücel T, Wang X, Lovett ML, Kaplan DL. Materials fabrication from Bombyx mori silk fibroin. Nat Protoc 2011;6:1612–31. [18] Kundu SC, Kundu B, Talukdar S, Bano S, Nayak S, Kundu J, et al. Non-mulberry silk biopolymers. Biopolymers 2012. [19] Ghanaati S, Orth C, Unger RE, Barbeck M, Webber MJ, Motta A, et al. Finetuning scaffolds for tissue regeneration: effects of formic acid processing on tissue reaction to silk fibroin. J Tissue Eng Regen Med 2010;4:464–72. [20] Murphy AR, Kaplan DL. Biomedical applications of chemically-modified silk fibroin. J Mater Chem 2009;19:6443–50. [21] Sofia S, McCarthy MB, Gronowicz G, Kaplan DL. Functionalized silk-based biomaterials for bone formation. J Biomed Mater Res 2001;54:139–48. [22] Karageorgiou V, Meinel L, Hofmann S, Malhotra A, Volloch V, Kaplan D. Bone morphogenetic protein-2 decorated silk fibroin films induce osteogenic differentiation of human bone marrow stromal cells. J Biomed Mater Res A 2004;71:528–37. [23] Chen J, Altman GH, Karageorgiou V, Horan R, Collette A, Volloch V, et al. Human bone marrow stromal cell and ligament fibroblast responses on RGDmodified silk fibers. J Biomed Mater Res A 2003;67:559–70. [24] Kardestuncer T, McCarthy MB, Karageorgiou V, Kaplan D, Gronowicz G. RGDtethered silk substrate stimulates the differentiation of human tendon cells. Clin Orthop Relat Res 2006;448:234–9.

2517

[25] Ratzinger G, Länger U, Neutsch L, Pittner F, Wirth M, Gabor F. Surface modification of PLGA particles: the interplay between stabilizer, ligand size, and hydrophobic interactions. Langmuir 2010;26:1855–9. [26] Bessa PC, Balmayor ER, Azevedo HS, Nürnberger S, Casal M, van Griensven M, et al. Silk fibroin microparticles as carriers for delivery of human recombinant BMPs. Physical characterization and drug release. J Tissue Eng Regen Med 2010;4:349–55. [27] Wolbank S, Peterbauer A, Fahrner M, Hennerbichler S, van Griensven M, Stadler G, et al. Dose-dependent immunomodulatory effect of human stem cells from amniotic membrane: a comparison with human mesenchymal stem cells from adipose tissue. Tissue Eng 2007;13:1173–83. [28] Nagineni CN, Amiel D, Green MH, Berchuck M, Akeson WH. Characterization of the intrinsic properties of the anterior cruciate and medial collateral ligament cells: an in vitro cell culture study. J Orthop Res 1992;10:465–75. [29] Le Blanc K, Tammik L, Sundberg B, Haynesworth SE, Ringdén O. Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex. Scand J Immunol 2003;57:11–20. [30] Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–7. [31] Toegel S, Harrer N, Plattner VE, Unger FM, Viernstein H, Goldring MB, et al. Lectin binding studies on C-28/I2 and T/C-28a2 chondrocytes provide a basis for new tissue engineering and drug delivery perspectives in cartilage research. J Control Release 2007;117:121–9. [32] Minoura N, Aiba S, Gotoh Y, Tsukada M, Imai Y. Attachment and growth of cultured fibroblast cells on silk protein matrices. J Biomed Mater Res 1995;29:1215–21. [33] Altman GH, Horan RL, Lu HH, Moreau J, Martin I, Richmond JC, et al. Silk matrix for tissue engineered anterior cruciate ligaments. Biomaterials 2002;23:4131–41. [34] Liu H, Fan H, Wang Y, Toh SL, Goh JCH. The interaction between a combined knitted silk scaffold and microporous silk sponge with human mesenchymal stem cells for ligament tissue engineering. Biomaterials 2008;29:662–74. [35] Fan H, Liu H, Toh SL, Goh JCH. Anterior cruciate ligament regeneration using mesenchymal stem cells and silk scaffold in large animal model. Biomaterials 2009;30:4967–77. [36] Kinahan ME, Filippidi E, Sarah K, Hu X, Evans HM, Pfohl T, et al. Tunable silk: using microfluidics to fabricate silk fibers with controllable properties. Biomacromolecules 2011:1504–11. [37] Huang Y, Onyeri S, Siewe M, Moshfeghian A. Ã SVM. In vitro characterization of chitosan – gelatin scaffolds for tissue engineering. Biomaterials 2005;26:7616–27. [38] Xia W, Liu W, Cui L, Liu Y, Zhong W, Liu D, et al. Tissue engineering of cartilage with the use of chitosan-gelatin complex scaffolds. J Biomed Mater Res B Appl Biomater 2004;71:373–80. [39] Talaei-Khozani T, Monsefi M, Ghasemi M. Lectins influence chondrogenesis and osteogenesis in limb bud mesenchymal cells. Glycoconj J 2011;28: 89–98. [40] Van Vlasselaer P, Falla N, Snoeck H, Mathieu E. Characterization and purification of osteogenic cells from murine bone marrow by two-color cell sorting using anti-Sca-1 monoclonal antibody and wheat germ agglutinin. Blood 1994;84:753–63. [41] Bayrak A, Tyralla M, Ladhoff J, Schleicher M, Stock U a, Volk H-D. Human immune responses to porcine xenogeneic matrices and their extracellular matrix constituents in vitro. Biomaterials 2010;31:3793–803. [42] Wanschitz F, Nell A, Patruta S, Wagner A, Ewers R. Influence of three currently used bone replacing materials on the in vitro proliferation of human peripheral blood mononuclear cells. Clin Oral Implants Res 2005; 16:570–4. [43] Prymak O, Bogdanski D, Köller M, Esenwein S a, Muhr G, Beckmann F, et al. Morphological characterization and in vitro biocompatibility of a porous nickel-titanium alloy. Biomaterials 2005;26:5801–7. [44] Treves C, Martinesi M, Stio M, Gutiérrez A, Jiménez JA, López MF. In vitro biocompatibility evaluation of surface-modified titanium alloys. J Biomed Mater Res A 2010;92:1623–34. [45] Seo YK, Yoon HH, Park YS, Song KY, Lee WS, Park JK. Correlation between scaffold in vivo biocompatibility and in vitro cell compatibility using mesenchymal and mononuclear cell cultures. Cell Biol Toxicol 2009;25:513–22.