Article pubs.acs.org/Biomac
Silk Fibroin/Hyaluronic Acid 3D Matrices for Cartilage Tissue Engineering Cristina Foss,†,‡,§ Enrico Merzari,†,§ Claudio Migliaresi,†,‡,§ and Antonella Motta*,†,‡,§ †
BIOtech, Department of Industrial Engineering, University of Trento, via delle Regole 101, 38123 Mattarello, Trento, Italy INSTM, Consorzio Interuniveristario Nazionale per la Scienza e Tecnologia dei Materiali, via G. Giusti 9, 50121 Firenze, Italy § European Institute of Excellence on Tissue Engineering and Regenerative Medicine, Trento, Italy ‡
ABSTRACT: In spite of commercially available products, the complete and sustained repair of damaged articular cartilage still presents various challenges. Among biomaterials proposed for cartilage repair, silk fibroin (SF) has been recently proposed as a material template for porous scaffolds cultured with chondrocytes and investigated in static and dynamic conditions. In addition to fibroin-based constructs, literature has reported that the combination of hyaluronic acid (HA) with other scaffold materials can protect the chondral phenotype and the cells in vitro response to the scaffold. In this study, the effect of the addition of HA on the physical properties of SF sponges, with and without cross-linking with genipin, was investigated. Salt-leached scaffolds were characterized in terms of morphology and structural and physical properties, as well as mechanical performance. Un-cross-linked sponges resulted in the physical separation of highly hydrophilic HA from the SF, while cross-linking prevented this phenomenon, resulting in a homogeneous blend. The presence of HA also influenced fibroin crystallinity and tended to decrease the cross-linking degree of the scaffolds when compared to the pure SF material.
■
INTRODUCTION In the articular joint, cartilage protects the subchondral bone from high stresses through an efficient load distribution, affording low-friction movement to the joint.1−3 Cartilage damage, due to traumas, injuries, or pathologies (e.g., osteoarthritis) is usually associated with severe pain, loss of articular motion, and joint instability1,4 and rarely self-repairs due to the avascular, aneural, and alymphatic nature of the tissue, resulting in a lack of the typical pathways of wound healing.2,5,6 Nowadays, several surgical methods are employed for the treatment of chondral defects, but none of these procedures is able to result in a complete, long-term repair of the joint.1,7 For example, mosaicplasty consists of the transplantation of osteochondral plugs (autografts) harvested from nonbearing sites of the patient to the site of injury; in autologous chondrocyte transplantation (ACT), lesions are filled with precultured chondrocytes isolated from the patient, expanded in vitro, and confined under a periosteal flap to the damaged site. Both techniques suffer from the scarce availability and morbidity of donor sites, low integration with the surrounding tissues, and difficulty matching the graft to the shape of the injury; moreover, the healing process usually proceeds through the formation of fibrocartilage, which is structurally different and mechanically inferior than native hyaline cartilage.3,6,8,9 To overcome these limitations, tissue engineering (TE) principles have been proposed to obtain an implantable graft with tailored morphological, chemical, mechanical, and biological properties necessary to obtain the complete and durable repair of damaged articular cartilage.3,5,7,9,10 Scaffolds for tissues, and in particular for cartilage regeneration, should allow for cell migration and organization, be able to induce assembly of new extracellular matrix, and possess suitable mechanical properties to sustain the necessary © 2012 American Chemical Society
physiological load. Hydrogels, sponges, and meshes have been proposed to meet these requirements, using both synthetic (e.g., poly(α-hydroxy esters), polyethylene glycol, and polyurethane) and natural (e.g., agarose, cellulose, collagen, chitosan, gelatin, silk fibroin, and hyaluronic acid) biomaterials.11−13 Silk fibroin (SF) is a protein produced by several insects and spiders. Silkworm fibroin that is usually isolated from the silk filaments produced by Bombyx mori silkworms can be processed to obtain films, nonwoven nets, sponges, and hydrogels able to sustain cell adhesion, proliferation, and ECM production both in vivo and in vitro4−21 while also possessing a low inflammatory potential.22 For cartilage TE applications, 3D fibroin sponges produced by salt leaching were seeded with human articular chondrocytes or mesenchymal stem cells (MSCs) and subsequently resulted in cartilage ECM deposition.23−25 SF scaffolds were also combined with growth factors, which induced a chondrogenic response in MSCs26 and were cultured in dynamic culture conditions to enhance chondrocytes proliferation, differentiation, and extracellular matrix (ECM) production.27 Hyaluronic acid (HA) is a natural polysaccharide present in the natural cartilage ECM. HA interacts with chondrocytes through several specific receptors, including CD44, and modulates cellular behavior preserving the chondral phenotype, which is a critical outcome for in vitro chondrocyte cultures.12,28 For cartilage tissue engineering applications, hyaluronic acid was combined with chitosan,29 hydrogels,30 poly(D,L-lactic-co-glycolic) acid,28 and collagen,31,32 with significant improvements in the production of cartilaginous ECM and chondrogenesis reported. Received: July 26, 2012 Revised: November 7, 2012 Published: November 7, 2012 38
dx.doi.org/10.1021/bm301174x | Biomacromolecules 2013, 14, 38−47
Biomacromolecules
Article
The use of silk fibroin and hyaluronic acid blends has recently been explored for TE applications demonstrating great potential. SF/HA freeze-dried sponges were prepared and cross-linked with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) for in vitro culture of neural cells, which maintained their phenotype after 5 days of culture.33 MSCs cultured on SF/HA scaffolds produced by freeze-drying lead to a higher deposition of collagens I and III with respect to a pure fibroin scaffold.34,35 Other applications include SF/HA hydrogels produced by ultrasonication36 and cardiac patches prepared from microparticles of silk fibroin and hyaluronan and cross-linked by genipin, a natural cross-linking agent for biopolymers containing amine groups.37 Genipin is extracted from the Gardenia fruit and reacts with varying kinetics to both primary and secondary amines, present on both silk fibroin and hyaluronic acid, respectively. Moreover, genipin is considered to be biocompatible and possesses weak anti-inflammatory properties,38,39 further highlighting its potential utility to stabilize biomaterials. As a result of these attractive features, genipin has been successfully used to cross-link several natural polymers, including silk fibroin,40 chitosan−fibroin,41 and collagen type II−chitosan−hyaluronic acid.42 In the present study, silk fibroin/hyaluronic acid scaffolds with variable HA content were produced by using a salt leaching technique to obtain an interconnected and controlled porosity with respect to freeze-drying. The process has been conducted in an aqueous environment, and to avoid a significant loss of HA, silk fibroin and hyaluronic acid were cross-linked by genipin to stabilize the scaffolds. The novelty of this study lies in the combination of the salt leaching technique with SF/HA blends and in the systematic study of how scaffold properties can be modulated by HA content and subsequent cross-linking. The stability of the system and hyaluronic acid distribution in the sponges were also evaluated in a qualitative and a quantitative way. On this basis, SF/HA scaffolds were characterized in terms of morphology, silk fibroin conformation, water content, porosity, cross-linking degree, and mechanical properties. Wettability properties were also evaluated on films produced by casting the same aqueous solution used for sponge preparation. Films were used as a model system to better understand silk fibroin/hyaluronic acid interaction and how it is modified by cross-linking without the additional complexity of the salt leaching process. To evaluate HA content and distribution in the scaffolds with and without cross-linking, hyaluronic acid was conjugated with a fluoresceinamine and the scaffolds were subsequently observed under confocal laser microscopy.
■
of 35 mm diameter. The salt was partially dissolved subtracting water to the fibroin so inducing fibroin gelation. After three days at room temperature, gels were washed for 4 days in distilled water to remove salt and then dried. SF/HA sponges were prepared in the same way, starting from the 8% fibroin water solutions added with 1, 2, and 5% by weight of HA referred to the total proteins weight (sample codes: HA1, HA2, HA5). Cross-linked SF and SF/HA materials were prepared by adding 0.5% by weight of genipin (sample codes: SFgen and HA1gen, HA2gen, HA5gen) to the above solutions. From the same solutions without salt and with or without genipin, SF/HA films were also casted. In this case, 7 mL of each aqueous solution were poured in a 60 mm Petri dish kept under a hood at room temperature until solvent evaporation. The resulting films were stabilized in water vapor for 24 h and finally dried at room temperature. 2. Scaffold Characterization. 2.1. Environmental Scanning Electron Microscopy. The morphology of the unmodified and crosslinked SF/HA sponges was evaluated using an ESEM (Environmental Scanning Electron Microscope) XL 30 (Fei Company). Sponges were completely dried at room temperature and images were collected in low vacuum mode (from 0.6 to 0.8 Torr) with a voltage range between 10 and 14 kV. 2.2. Fourier Transform Infrared Spectroscopy. Silk fibroin structure in the SF/HA sponges and films was evaluated by FTIRATR (Fourier Transform InfraRed, Attenuate Total Reflectance, Spectrum One Perkin-Elmer, U.S.A.). Spectra were collected from 4000 to 600 cm−1 as the mean of 16 scans. 2.3. Porosity. The open porosity of the SF/HA sponges was quantitatively evaluated using the principle of liquid displacement. Hexane (n-Hexane, Sigma-Aldrich, U.S.A.) was used because it fills the pores of the submerged fibroin/hyaluronic acid scaffolds without swelling. After drying at room temperature, each scaffold was weighted (PS), put in a Gay-Lussac pycnometer, and submerged in hexane for 10 min to allow the liquid to penetrate the pores. Then, the pycnometer was completely filled (nominal volume: VN = 25 mL) and the net weight was registered (PT). The volume of hexane in the pycnometer was simply calculated as follows: V1 =
PT − PS ρhex
(1)
where ρhex is the hexane density at the measurement temperature. Subsequently, the sample was removed and the net weight of the residual hexane was measured (PR). The porosity was finally calculated from the following equation: ε(%) =
V1 − PR /ρhex VN − PR /ρhex
× 100 (2)
2.4. Cross-Linking Degree. The degree of cross-linking of the SF/ HA materials was evaluated using two different methods. The first method used the reaction with ninhydrin of the free amino groups (both primary and secondary) remained after cross-linking of the blend materials. That reaction occurs with the production of purple products whose amount detected by a spectrophotometer at 570 nm can be related to the amount of the reacted groups.43 Sponges of SF/HA or pure fibroin, with or without cross-linking with genipin, were frozen at −20 °C and lyophilized for 24 h. A total of 15 mg of each sample were incubated in 5 mL of 0.35% w/V ninhydrin (Sigma-Aldrich, U.S.A.) in ethanol (
