Bacterial cellulose as a potential scaffold for tissue

Bacterial cellulose as a potential scaffold for tissue

(Parte 1 de 3)

Bacterial cellulose as a potential scaffold for tissue engineering of cartilage

aDepartment of Chemical and Biological Engineering, Tufts University, Boston, MA 02155, USA bDepartment of Material and Surface Chemistry/Biopolymer Technology, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden cDepartment of Orthopaedics, Cartilage Research Unit, G.oteborg University, Kungsbacka Hospital, SE-434 80 Kungsbacka, Sweden

Received 1 June 2003; accepted 19 February 2004


Tissue constructs for cartilage with native mechanical properties have not been described to date. To address this need the bacterial cellulose (BC) secreted by Gluconacetobacter xylinus (=Acetobacter xylinum) was explored as a novel scaffold material due to its unusual material properties and degradability. Native and chemically modified BC materials were evaluated using bovine chondrocytes. The results indicate that unmodified BC supports chondrocyte proliferation at levels of approximately 50% of the collagen type I substrate while providing significant advantages in terms of mechanical properties. Compared to tissue culture plastic and calcium alginate, unmodified BC showed significantly higher levels of chondrocyte growth. Chemical sulfation and phosphorylation of the BC, performed to mimic the glucosaminoglycans of native cartilage, did not enhance chondrocyte growth while the porosity of the material did affect chondrocyte viability. The BC did not induce significant activation of proinflammatory cytokine production during in vitro macrophage screening. Hence, unmodified BC was further explored using human chondrocytes. TEM analysis and RNA expression of the collagen I from human chondrocytes indicated that unmodified BC supports proliferation of chondrocytes. In addition, ingrowth of chondrocytes into the scaffold was verified by TEM. The results suggest the potential for this biomaterial as a scaffold for tissue engineering of cartilage. r 2004 Elsevier Ltd. All rights reserved.

Keywords: Cellulose; Cartilage tissue engineering; Chondrocytes; Scaffold

1. Introduction

The main purposes of articular cartilage, which contains a small number of cells (chondrocytes) in an extra cellular matrix (ECM) mainly composed of water, collagen type I and proteoglycans, are to cover the ends in bones in joints to provide frictionless movement and to distribute loads [1–3]. Ostheoarthritis is a disease of synovial joints resulting in pain and loss of function for the patient [1,4]. Damaged cartilage has limited regenerative capacity and therefore over 1 million patients in the United States require treatment for cartilage defects each year. However, presently these treatments result in limited pain relief and/or restorative tissue function

[5–10]. Thus, tissue engineering has potential to provide a supply of functional cartilage for the repair and regeneration of compromised native soft tissues [10,1].

The use of scaffoldsin the tissueengineeringof cartilage is essential in order to support cell proliferation and maintain their differentiated function in addition to definition of the shape of the new growing tissue [12]. To this end, a variety of scaffold materials have been evaluatedincludingnaturalpolymerslike collagen[13–17], alginate [13,16–18], hyaluronic acid [1,13,16,17],fi brin glue [13,16,17] and chitosan [13,16,17] and synthetic polymers including polyglycolic acid (PGA) [5,7,15–17], polylactic acid (PLA) [9,16,17], polyvinyl alcohol (PVA) [13,16,17,19], polyhydroxyethylmethacrylate (pHEMA)

[13,16,17] and polyN-isopropylacrylamide (pNIPAA)

[13,16,17]. However,tissue constructswith native mechanicalpropertieshave not yet been describedin the literature. It is also desirable to utilize a scaffold materialthat has the

*Corresponding author. Tel.: +46-317-723-407; fax: +46-317-723- 418. E-mail address: (P. Gatenholm).

0142-9612/$-see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.02.049

porosity necessary to support cell ingrowth and effective mass transport while also supplementing the mechanical properties of engineered tissue. This scaffold must also be biocompatible and support native ECM and biopolymer production, with degradation rates commensurate with the rate of the new tissue formation.

In the present study, we have explored a novel biomaterial, which meets these requirements, bacterial cellulose (BC). BC is secreted by Gluconacetobacter xylinus (=Acetobacter xylinum) and has unique properties including high water holding capacity, high crystallinity, a fine fiber network, and high tensile strength [20–24]. BC has recently been studied for use as artificial skin and blood vessels [25]. BC has potential to be used as a substrate for tissue engineering of cartilage due to its high strength in the wet state as well as its moldability in situ, biocompatibility and relatively simple, costefficient production.

We report the response of primary bovine chondrocytes on native and chemically modified BC. Phophorylation and sulfation of matrices were explored in order to add surface charges to mimic the glucosaminoglycans in cartilage tissue in vivo. The materials were analyzed for mechanical properties, microstructure and cell– material interactions in order to assess the potential for this matrix as a scaffold for tissue engineering of cartilage. Chondrocyte response to collagen type I, plant-derived cellulose, calcium alginate and tissue culture plastic were also compared to the BC matrices. The unmodified BC was further explored using human chondrocytes. The RNA-expression of ECM markers, including chondrocyte specific collagen I and fibroblast specific collagen I, was analyzed to study if BC supports proliferation of chondrocytes. In addition, the potential of BC to support cell ingrowth was investigated.

2. Methods 2.1. Production of BC

G. xylinus (ATCC 10245) purchased from the American Type Culture Collection was grown in 10g/l Bactopeptone (Difco), 10g/l yeast extract (Fisher),

4mm KH2PO4 (Sigma), 6mm K2HPO4 (Sigma) and 20g/l d-glucose dissolved in deionized water (DI). The pH of the medium was adjusted to 5.1–5.2. Five milliliters media were inoculated with culture (third cultivation from single colonies) over night at 0.6% v/v in 6-well tissue culture plates and incubated statically at 30 C for 7 days. BC pellicles were purified by treatment of 4% SDS (Fisher) in DI at 70 C for 3h and then 4% NaOH in DI at 70 C for 90min. Samples were rinsed with DI to pH=7 and stored in DI at room temperature prior to use.

2.2. Chemical modifications

2.2.1. Phosphorylation

Water was removed from samples by solvent exchange (70% ethanol-100% ethanol-DMF). Samples were phosphorylated at 120–130 C with reaction media of 159ml DMF, 83ml Toluene and 8ml 85%

H3PO4 for 30min (BC-P1) and 2h (BC-P2), respectively. Toluene was refluxed during the reaction and water was collected with a Dean–Stark apparatus. Samples were serially rinsed in DMF, ethanol and DI at room temperature. Samples treated by the identical procedure but without the addition of acid were prepared as controls (BC-P1c and BC-P2c). All samples were lyophilized overnight prior to analysis with ESCA.

2.2.2. Sulfation Water was removed as above. Twenty-five grams sulfamic acid (NH2SO3H) (Sigma) was dissolved in 103ml DMF. This solution was added to 147ml DMF at 80 C. Samples were added and the reaction was performed at 80 C for 90min. Samples were rinsed with DMF, 100% ethanol, 70% ethanol at room temperature and then with DI at 80 C. Samples treated by identical procedure without the addition of acid were prepared as controls (BC-Sc). All samples were lyophilized overnight prior to analysis with ESCA.

2.3. Control samples

2.3.1. Collagen

Two-hundred and fifty microliters of 0.1% Collagen type I (Davol, Cranston, RI) solution in 0.2% acetic acid was cast in 24-well plates, dried overnight at room temperature in a fume hood and ethanol sterilized (20min contact and 1h drying at room temperature).

2.3.2. Alginate

High viscosity alginate, sodium salt (Macrocystis pyrifera) was purchased from Sigma, thoroughly mixed as a 1% solution in DI and stored overnight at 4 C. The resulting gel was pipetted into a Delrin mold of appropriate dimension and allowed to crosslink for

48h in 1m CaCl2 with stirring. After gelation was complete, disks of appropriate thickness were sectioned, rinsed twice in an excess of DI and immediately autoclaved at 121 C for 20min.

2.3.3. Tissue culture plastic

Tissue culture plastic refers to the unmodified surface of Falcon (3524) 24-well tissue culture plates.

2.3.4. Plant derived cellulose

Whatman Qualitative filter paper (cellulose), rinsed in an excess of DI and autoclaved, was used as a source of plant derived cellulose (cotton linters) surfaces.

2.4. Characterization

Water content was determined by thermo gravimetric analysis (TGA) using a Perkin-Elmer TGA 7 (thermo gravimetric analyzer) with a TAC 7/DX (thermal analysis controller) under an oxygen free nitrogen atmosphere. The surface elemental composition of the samples was analyzed by electron spectroscopy for chemical analysis (ESCA) using a Surface Science SSX-100 ESCA Spectrometer from Surface Science. Scanning electron microscopy (SEM) Zeiss DSM 940A was used to study the morphology of the materials before and after chondrocyte growth. Phase-contrast microscopy (Axiovert S100 from Zeiss) was used to study cell growth during cell response tests. To investigate the morphology of human chondrocytes as well as their penetration of the BC scaffold a transmissions electron microscopy was used.

ConfocalmicroscopyBioRadMRC1024equippedwith fiber coupled ArKr laser was used to study the morphology and porosity of the materials in wet state. Filterswere chosen withregard to the emissionwavelength of the dye (lex ¼ 495nm and lem ¼ 516nm). The wet BC samples were fluorescently labeled using DTAF (5-((4,6- dichlorotriazin-2-yl)amino)fluorescein hydrochloride) (Sigma), which binds to exposed hydroxyl groups on the cellulose chains. About 250mg wet sample per condition wass tained for2 0h with 2ml0 .1m NaOH solution containing 5mg DTAF. After staining, the excess of DTAF was removed by placingthe samplesin a B.uchner funnel and rinsing with running deionized water for approximately5min. The samples were rinsed further in 200ml deionized water with mild stirring overnight. The water was changed twice.

2.5. Mechanical analysis

The mechanical properties of BC in the wet state (never dried) were determined using a custom-made confined compression chamber with an inner diameter of 10mm and an Instron Testing Machine 8511 with a 50lb load cell. Samples with a diameter of 10mm and thickness of 0.3mm were compressed at a speed of 10mm/s and compressive modulus was determined from the linear region of the compression phase. In order to test the influence of the thickness of the samples on the compressive modulus, samples with thickness 3mm of unmodified BC and calcium alginate, respectively, were compressed as above using an Instron 5542 with a 500N load cell. In addition, ultimate tensile strength and Young’s modulus were determined on 10 wet samples of each condition (25 5 0.3mm) at a speed of 0.5mm/ min using an Instron Testing Machine 8511 with a 50lb load cell. Samples were gripped with sandpaper. The thickness of the samples was measured with a micrometer (72mm).

2.6. Cell study I

2.6.1. Cell responses

Modified and unmodified BC samples were punched from mats (F ¼ 15mm), put in DI, sterilized by autoclave at 121 C and 18psi for 20min and transferred aseptically to 24-well tissue culture plates in triplicate. One milliliter DMEM (Gibco) was added to each well to soak the samples before cell seeding. Primary bovine chondrocytes (passage number 6 and 95% viability), which were obtained by enzymatic digestion of fullthickness articular cartilage harvested from the femoropatellar grooves of 2–3-week-old bovine calves within 8h of sacrifice [26,27], were seeded at a concentration of 25,0 cells per well. One milliliter media was added to each well and plates were incubated for 8 days at 37 C,

5% CO2. Media was changed every 3 days and images were captured each day. The cell culture media consisted of Dulbecco’s modified Eagle’s medium (DMEM) (Gibco) containing 1% penicillin–streptomycin (P/S), 0.2% fungizone, 1% N-2-hydroxyethylpiperazine-N0-2- ethane sulfonic acid (HEPES) and 10% fetal bovine serum (FBS) (Gibco) plus 1% non-essential amino acids (NEAA), 0.1m Proline and 50mg/ml l-ascorbic acid (Sigma). After 8 days, 50ml MTT reagent (5mg/ml, Sigma) were added to each well and plates were incubated in darkness for 4h at 37 C, 5% CO2. All media was removed gently, 500ml MTT solubilization solution (acidified iso-propanol) was added to each well and plates protected from light were placed on a shaker for 1.5h. Following incubation, absorbance was determined of the media at 570nm.

Samples of modified and unmodified BC were punched and sterilized as described above. Bovine chondrocytes were seeded at a concentration of 1.6 105 cells per well. One milliliter of culture media was added to each well and plate was incubated overnight at 37 C, 5% CO2. Following washing three times with 1ml of 0.2m sodium cacodylate buffer, the cells were fixed by adding 1ml SEM fixative (Karnovsky fixative) to each well. The plate was stored overnight at 4 C and then the samples were dehydrated by using an ethanol gradient (10%, 30%, 50%, 80%, 90%, 2 95% and 2 100% EtOH for 15min per step). Samples were immersed in 1ml 1,1,2-trichlorotrifluoroethane 9.8% (ACS spectrophotometric grade). After evaporation, the samples were analyzed with SEM.

2.7. Immune response

2.7.1. Macrophage stimulation and assay

Macrophages were stimulated with the various cellulose preparations as well as with the alginate sample and tissue culture plastic. RAW 264.7 cells were

plated at 8 104 cells in 200ml per well in 96-well tissue culture plates (for chondrocyte supernatants) or 2 105 cells in 1ml per well in 24-well plates (for BC and control disks). The chondrocyte supernatants were tested in the 96-well format, where cells were plated for 24h prior to stimulation. RAW cells were plated on top of the cellulose disks in the 24-well format. All stimulations, with the exception of LPS positive controls, were carried out in the presence of 5mg/ml polymyxin B to inhibit any LPS contamination. Macrophage supernatants were collected at 24h (96- and 24-well format), and 48h (24-well format) assayed for tumor necrosis factor (TNF).

2.7.2. Determination of TNF release

TNF was measured by ELISA according to manufacturer’s instructions (R&D Systems, Minneapolis, MN). TheELISAhasa detectionlimitof approximately5pg/ml. Briefly, a standard curve was generated using recombinant murine TNF with two-fold dilutions from 31 to 2000pg/ml. Capture antibody (goat anti-murine TNF) was plated at 0.8mg/ml in 1m carbonate buffer, pH 9.5 in Nunc-Maxisorp 96-well plates overnight at 4 C. Plates were washed three times with PBS-Tween 20 (0.005%) and blocked for 2h with 1% BSA in PBS at 37 C. Plates were again washed three times, and 50mlo f macrophage supernatants and recombinant TNF standards were added to the ELISA plate and incubated overnight at 4 C. Plates were washed as described above and secondary antibody (biotinylated goat anti-murine TNF) was added at 0.3mg/ml in PBS-Tween 20 (0.005%). After 2h incubation at 37 C, plates were again washed and incubated 30min at room temperature with a 1:200 dilution of HRP-streptavidin in PBSTween 20 (0.005%). Plates were developed with 100mlo f TMB (3,30,5,50-tetramethylbenzidine) liquid substrate

(Parte 1 de 3)