- Email: [email protected]
Template assisted synthesis of porous nanofibrous cellulose membranes for tissue engineering
Available online at www.sciencedirect.com
Materials Science and Engineering C 28 (2008) 549 – 554 www.elsevier.com/locate/msec
Template assisted synthesis of porous nanofibrous cellulose membranes for tissue engineering C.R. Rambo a,b,⁎, D.O.S. Recouvreux b , C.A. Carminatti b , A.K. Pitlovanciv b,c , R.V. Antônio b,c , L.M. Porto b a Group of Ceramic and Glass Materials – CERMAT, Brazil Genomic Engineering Group, Department of Chemical and Food Engineering, Brazil Department of Biochemistry, Federal University of Santa Catarina, P.O. Box 476, 88040-900 Florianópolis, SC, Brazil b
c
Received 29 November 2006; received in revised form 8 November 2007; accepted 30 November 2007 Available online 8 December 2007
Abstract Porous, nanofibrous bacterial cellulose (BC) membranes were produced by the bacterium Acetobacter xylinum. The bacterium was cultivated in an appropriate culture medium under static conditions. In situ pore formation was attained through the use of pin templates with diameters varying from 60 to 300 μm composed of polyestirene (ϕ = 300 μm) or optical fibers (ϕ = 60 μm), which were placed on culture medium with the pins immersed in the liquid. Cellulose biosynthesis occurred around the pins leaving tiny pores on the cellulose membrane. After removal of the template the biofilm was dried at 50 °C/24 h. Physico-chemical properties of BC membranes, like degree of crystallinity, swelling and tensile strength were not significantly altered after pore formation. Microstructure evaluation revealed that the film matrix is composed of long nanofibers isotropically distributed on its surface. Round-shaped pores with diameters varying between 60 and 300 μm, depending on the pin template used, were formed in the cellulose membranes. These pores exhibited no border failures that could start crack propagation along the film surface. Microporous membranes could be useful for applications in repairing tissues, which require high oxygenation rates or wound contracture delay. © 2007 Elsevier B.V. All rights reserved. Keywords: Tissue engineering; Porous membranes; Scaffolds; Bacterial cellulose
1. Introduction Tissue engineering is an interdisciplinary field that applies the principles of engineering and the life sciences toward the development of biological substitutes that restore, maintain, or improve tissue or a whole organ function. Typical application examples are implants of scaffolds composed of biodegradable polymers or inert materials coated with bioactive biomaterials that allow growth of new tissues of particular types of cells [1,2]. After the formation of the new tissue, polymeric scaffolds are gradually degraded into small molecular weight compounds, which can be absorbed by the body or excluded out of the body [3]. ⁎ Corresponding author. Genomic Engineering Group, Department of Chemical and Food Engineering, Brazil. E-mail address: [email protected] (C.R. Rambo). 0928-4931/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.msec.2007.11.011
In recent years the search for new classes of biopolymers, with specific properties to be used as scaffolds in tissue engineering, has attained great interest, like cellulose, polyhydroxyalkanoates, polylactates and blends of these materials [4–9]. In natural environments biofilms are associated with some microbial styles of life [10] and constitute a kind of selfprotection that allows microorganisms to survive in adverse conditions [11]. Bacterial cellulose (BC) is a kind of extra cellular polysaccharide present in the biofilm produced by several bacteria, notably by Acetobacter xylinum, as long nanofibers. This polymer is highly crystalline and its degree of crystallinity varies depending on the origin and mode of chemical treatment [12]. BC is composed by glucose molecules joined by β(1 → 4)-glycosidic bonds forming branchless linear chains [13,14]. Cellulose biosynthesis is a process involving several steps regulated by some individual
550
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
and well characterized complex enzymes and proteins. This process includes the synthesis of uridine diphosphate glucose (UDP-glucose), which is the precursor of cellulose biosynthesis, followed by glucose polymerization in β(1 → 4)-glucan chains by complex proteins [13,14]. Bacterial cellulose fibers exhibit a wide range of dimensions ranging from 1 to 25 nm in width, which corresponds to 10–250 polyglucan chains and from 1 to 9 μm in length (formed by 2000–18,000 glucose residues). Among six known polymorphs of cellulose, two common crystalline forms are of interest: cellulose I and II [15]. It is known that in the cellulose I, which is synthesized by plants and also by A. xylinum in static culture, parallel β(1 → 4)-glucan chains are uniaxially disposed, whereas β(1 → 4)-glucan chains of cellulose II are randomly arranged. Furthermore, crystalline cellulose I exhibits two main unit cell structures called cellulose Iα and Iβ, which consist of one chain triclinic and two chain monoclinic unit cells, respectively [16,17], which are difficult to distinguish by X-ray diffractometry. The cellulose synthesized by A. xylinum is known to be composed by cellulose Iα as majority phase and cellulose Iβ in smaller fraction [18–20]. The degree of crystallinity of cellulose influences some physico-chemical properties of cellulose like swelling and water binding [21]. Bacterial cellulose is used in a wide range of applications, from the food industry to electroacustic devices, as phone diaphragms. Bacterial cellulose is also potentially suitable for applications as scaffolds in tissue engineering due to its unique properties, including a high water retention capability (hydrophilic), a fine fibrous network, which allows cell growth and proliferation, high tensile strength, in situ moldability and low production cost [22–26]. In medicine, non-porous cellulose membranes are used as stent coatings, for dura-mater substitution in tumor or trauma cases or as skin protection in cases of burn and deep wounds. In odontology, cellulose films are applied for periodontal tissue recovering [27,28]. Despite the studies concerning bacterial cellulose membranes in medical applications as substitute for cartilaginous and skin tissues, production of porous bacterial cellulose films were reported only recently by Siqueira and Moreschi [29] and are not widely used in tissue engineering. Porous membranes
Fig. 2. Schematic flow chart for the production of porous bacterial cellulose membranes.
consisting of large pores are usually indicated for preventive and healing treatments of wounds, in particular those in which exudation and oxygenation are necessary [30]. Among the possibilities of using porous membranes, a particular case is of great interest in tissue engineering: wound healing mechanism. Porous membranes composed of collagen I and chondroitin-6sulfates are able to delay wound contracture and promote dermal regeneration. The optimal pore size lies between 20 and 120 μm, which promote the delay of fibroblast migration and avoids wound contraction and scar formation, allowing time for wound healing [31]. Fig. 1 shows the range of possible applications of porous cellulose membranes as a function of the pore diameter. This work reports the biosynthesis of nanofibrous bacterial cellulose films containing in situ produced micropores. An innovative design of features (porosity, pore size distribution and pore geometry) of the cellulose membrane was developed to modify/improve a determined property for a specific application, which in turn intrinsically depends on these features. The flow chart shown in Fig. 2 illustrates schematically the multidisciplinary approach for the manufacturing of the porous cellulose membranes. 2. Experimental 2.1. Synthesis
Fig. 1. Schematic diagram of the several applications of porous BC membranes in function of the pore diameter.
Bacteria A. xylinum ATCC 23769, acquired from “Collection of Tropical Culture (CCT)” (André Tosello Foundation) was used for the cellulose production. A. xylinum inoculum was prepared by its cultivation under static conditions, at 30 °C, in a 125 mL Erlenmeyer flasks, containing 25 mL of Hestrin & Schramm medium [32]. Cellulose production was carried out by the cultivation of the bacterium for 7 days, under static conditions at 30 °C, in 250 mL Erlenmeyer flasks containing 50 mL of modified Hestrin & Schramm medium, pH 6.6, containing (per liter): 5.0 g peptone, 5.0 g yeast extract, 1.15 g citric acid, 2.27 g Na2HPO4 and 20 g glucose. In order to produce porous
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
551
Microscopy (trinocolar, XDS-1B, 40 × objective) observations were carried out in order to evaluate the morphology and distribution of the micropores of dried PBC membranes. Tensile strength of BC and PBC at room temperature was determined on a set of 5 samples of each membrane with nominal dimensions of 70 × 25 mm using a universal testing device (Instron Corp., Instron 4202). The speed of the crosshead was set constant to 1 mm/min. 3. Results and discussion
Fig. 3. Photograph of the BC membrane in humid state (gel).
cellulose membranes, pin array structures (similar to a brush) with diameters varying from 60 to 300 μm composed of polystyrene (ϕ = 300 μm) or optical fibers (ϕ = 60 μm) were introduced into the media allowing in situ pore formation during cultivation and cellulosic membranes formation. The porous BC membranes were coded PBC300 and PBC60, which contained pore diameters of 300 μm and 60 μm, respectively. After the cultivation period, the formed gel at the liquid surface and in between the pin array template was removed, washed with deionized water and dried at 50 °C until constant weight. Subsequently, the cellulose membrane was treated with a 0.1 M NaOH solution at 90 °C for 30 min to remove bacterial impurities and eventual cell debris. The membrane was again washed in deionized water until neutral pH and then dried at 50 °C for 24 h.
In the humid state the biofilm formed on the surface of the culture medium of the bacterium A. xylinum is a homogenous transparent, moldable and handle-resistant gel (Fig. 3). After drying, the thickness of the membranes varied between 150 μm and 200 μm. A more detailed analysis by SEM revealed that the zooglea after drying is composed of long fibers. Fig. 4 shows the SEM micrograph of the dried membrane surface, highlighting details of the microstructure. The film consists of a dense, flawless and homogeneous cellulose matrix (Fig. 4a). The matrix is formed by an interlaced network of long cellulose nanofibers with high aspect ratio and mean diameters smaller than 100 nm (Fig. 4b). This high aspect ratio may affect the mechanical properties of the film such as elasticity and tensile strength. The reason for that fibrous morphology was subject of discussion in the last decade and is already established
2.2. Characterization Swelling behaviour of the BC membranes was determined by water and physiological solution (0.9%NaCl solution) absorption measurements in samples of 15 × 20 × 0.02 mm. The measurements consisted in immersing previously dried membranes into the solution for a determined period and measuring the mass, after removal of excess of solution on the surface. The experiment was carried on until no further absorption was observed. The percent weight gains of the samples were calculated and plotted against time. X-ray diffractometry (XRD, Philips, X'Pert) with CuKα radiation (λ = 1.45 nm) was used to identify the phases of the cellulose membrane. Dried BC and PBC60 membranes were placed on an aluminium plate and measured over a 2θ interval of 5°–40° with steps of 1 °/min. JCPDS-files were used for phase identification. The degree of crystallinity (XC) of the porous cellulose membrane was estimated from the diffractogram through the ratio between the areas under the (110) and (200) crystalline peaks and the total area, according to the method described by Watanabe and Tabuchi [24]. The microstructure of the cellulose membranes was characterized by Scanning Electron Microscopy (SEM, Philips, XL-30). For the observations, BC membranes with diameters of 5 mm were dried and placed over an aluminium support and sputtered with gold. Optical
Fig. 4. SEM micrographs of the surface of a bacterial cellulose membrane.
552
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
[15,33]. Cellulose biosynthesis is characterized by unidirectional growth and crystallization, where glucose molecules are linear bonded by β(1 → 4)-glycosidic bonds. The union of glycosidic chains forms oriented microfibrils with intramolecular hydrogen bonds [34]. The preferential orientation of cellulose crystal growth during biosynthesis was already reported by Koama et al. [35]. The cellulose is crystallized outward the organisms, particularly in A. xylinum that synthesizes cellulose chains by introducing glucose units to the reducing ends of the polymer. The growth mechanism during bacterial activity determines the morphology of the final cellulose. Fig. 5 shows the swelling behaviour of the PBC60 membrane under water and physiological solution. The absorption behaviours under water and physiological solution were similar within the statistical deviation. After 30 min the weight gain reached a maximum (approximately 175%), slightly decreasing to 153% between 0.5 and 1 h. A stable state was achieved after 1 h. Swelling occurs by a combined effect of both fiber cell walls swelling and breaking of interfiber bonds that involves fiber displacement. Therefore, the displacement of the fibers causes an expansion of the volume and results in a free space for the swelling agent. The maximum expansion of the fibers depends on the absorption capacity of the fibers in the liquid, i.e. ability of the swelling liquid to form hydrogen bonds with the hydroxyl groups of cellulose [36]. Additionally cellulose water sorption may considerably affect the properties of cellulose-based materials, e.g. medical products. Fig. 6 shows the X-ray diffraction pattern of the surface of the bacterial cellulose membrane and of the porous membrane. Three main peaks can be identified in both spectra, which are assigned to the (110), (110) and (200) reflexions planes of cellulose I [18]. Despite the slight difference in peak intensities between BC (Fig. 6a) and PBC60 (Fig. 6b), no peak shift can be seen in the spectra, which indicates that no changes on the crystalline structure occurred that could be attributed to the influence of pin templates. The estimated degree of crystallinity of the BC membrane was 56 ± 5% and of the PBC60 membrane was 51 ± 5%. Although it was an estimative of XC based on the X-ray diffractograms, it is relevant to point out that the cellulose
Fig. 5. Swelling behaviour of the BC membranes in function of time for water and physiological solution.
Fig. 6. X-ray diffractogram of the BC (a) and PBC60 (b) membranes. Peaks indexed are assigned to cellulose I type.
membranes exhibit similar crystallinity within the accuracy of the method. Fig. 7 shows typical stress–strain curves of the BC and PBC300 membranes under tension. Both membranes exhibited relative low elongation under tension before breaking. Rupture occurred at a strain of 1.7% for the porous membrane (PBC300) and at 3.4% elongation for the BC membrane. Tensile strengths of the BC and PBC300 were 32 MPa and 28 MPa, respectively. Tensile strength of untreated BC membranes is reported to be higher than 65 MPa [37]. The lower strength of NaOH-treated membranes may be attributed to their swelling behaviour. During swelling of cellulose fibers, intermolecular bonds between the fibers break as a result of the internal stress produced by absorption [36]. Due to swelling, the ordering within the fibers will be reduced and this may contribute to the reduction in mechanical strength. It must be also noted that, although the PBC300 membrane is a porous material, it exhibited similar strength to the non-porous membrane, which is also an indication of the efficiency of the method of pore production, concerning the integrity of the PCB after processing.
Fig. 7. Typical room temperature stress–strain curves of BC and PBC300 membranes under tension.
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554 Table 1 Key properties of bacterial cellulose (Td = degradation temperature) Property
BC (other works)
Ref.
BC (this work)
PBC300
Tensile strength Td Crystallinity
10–70 MPa
[37,38]
32 ± 5 MPa
28 ± 5 MPa
N300 °C 40–70%
[12] [15,18,21,24]
– 56 ± 5%
341 °C 51 ± 5%
Table 1 summarizes the key properties of the BC and PBC membranes obtained in this work in comparison with other literature data. The physico-chemical properties of the porous membrane were not significantly affected by the presence of pores. Fig. 8 shows optical (a and b) and SEM images (c and d) of dried porous cellulose membranes with different pore diameters. Round-shaped pores with mean diameter of 300 μm distributed on the cellulose surface with a density of approximately 2 ppi (pores per linear inch) can be observed (Fig. 8a). The distance between the pores was set by arranging the polymeric pins in a distance of approximately a pin diameter from each other. The border of the pores exhibits no flaw and surrounds the pore uniformly, which provides isotropic strength distribution to the porous membrane (Fig. 8b). Large pores (N200 μm) are potentially useful in applications that allow high
553
oxygenation rates and wound exudation release. Concerning the last case, in treatments of deep wounds, where the exudation (with high or low viscosity) must be eliminated, porous scaffolds permit fast granulation and regeneration of the damaged tissue, with subsequent epithelial regeneration [29]. Fig. 8c shows the cellulose membrane synthesized with a template with pins of 60 μm of diameter. A pore round-shaped with approximately 60 μm of diameter is displayed. Some folds are also present around the pore, which resulted from the sample preparation for the SEM analysis. Details of the boarder of the pore can be seen in Fig. 8d. Analogous to the large pores, the small ones are round-shaped and exhibit no border failures that could start crack propagation along the film surface and therefore tearing the scaffold. The bacteria form a protective border that surrounds the pin surface leaving a rope-kind reinforcement structure, which could be the reason for the relative high tensile strength compared to the non-porous membrane. Pores with diameters below 100 μm are suitable for wound contracture by delaying fibroblast migration through the porous template. Beside the use of thin pore membranes with small pores for medical applications, pore sizes in the range of 10– 100 μm are interesting materials to promote selective cell migration through the pores for specific cell cultures, with different cell sizes.
Fig. 8. Optical (a,b) and SEM (c,d) micrographs of porous cellulose membranes. Figs. b and d are higher magnitude images from the same sample shown in Figs. a and c, respectively.
554
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
4. Conclusions Porous nanofibrous bacterial cellulose membranes with controlled pore sizes and shapes were produced by appropriate cultivation of A. xylinum. The pin templates used as pore formers acted as bioinert surfaces, allowing the bacteria to synthesize cellulose fibers around the pins. The physical-chemical properties of BC membranes were not significantly affected after pore formation. Round-shaped pores with diameters in the range of 60 to 300 μm were achieved, depending on the pin diameter of the template. The pores were uniformly distributed on the cellulose membrane. After membrane drying the pore boarders exhibited no visible failures down to the submicrometer scale that could, under tension forces, start crack propagation along the film surface. The range of the pore diameter determines the most suitable application in tissue regeneration. The prepared membranes exhibited suitable macro- and microstructure that might be used for applications such as repairing tissue for cartilage and skin due to its nanofibrous network which could promote cell growth, and a porous network, which improves oxygenation of the repairing tissue, allow exudation flow and promote delay on wound contracture. Acknowledgements The authors thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil for the financial support. The authors also thank André Tosello Foundation for the gently donation of A. xylinum strain. References [1] E. Wintermantel, J. Mayer, J. Blum, K.-L. Eckert, P. Lüscher, M. Mathey, Biomaterials 17 (1996) 83. [2] C.R. Rambo, F.A. Mueller, L. Mueller, H. Sieber, I. Hofmann, P. Greil, Mater. Sci. Eng. C 26 (2006) 92. [3] R. Langer, J.P. Vacanti, Science 260 (1993) 920. [4] D. Chen, B. Sun, Mat. Sci. Eng. C 11 (2000) 57. [5] K. Zhao, Y. Deng, J.C. Chen, G.-Q. Chen, Biomaterials 24 (2003) 1041.
[6] S.V. Madihally, H.W.T. Matthew, Biomaterials 20 (1999) 1133. [7] Y.Z. Wan, Y. Huang, C.D. Yuan, S. Raman, Y. Zhu, H.J. Jiang, F. He, C. Gao, Mat. Sci. Eng. C 27 (2007) 855. [8] S. Ramakrishna, J. Mayer, E. Wintermantel, Kam W. Leong, Compos. Sci. Technol. 61 (2001) 1189. [9] F.A. Muller, L. Muller, I. Hofmann, P. Greil, M.M. Wenzel, R. Staudenmaier, Biomaterials 21 (2006) 3955. [10] P. Watnick, R. Kolter, J. Bacteriol. 182 (2000) 2675. [11] J.W. Costerton, P.S. Stewart, E.P. Greenberg, Science 284 (1999) 1318. [12] J. George, K.V. Ramana, S.N. Sabapathy, J.H. Jagannath, A.S. Bawa, Int. J. Biol. Macromol. 37 (2005) 189. [13] R.M. Brown Jr., J.H. Willison, C.L. Richardson, Proc. Natl. Acad. Sci., Cell Biology, vol. 73, 1976, p. 4565. [14] P. Ross, R. Mayer, M. Benziman, Microbiol. Rev. 55 (1991) 35. [15] A.C. O'Sullivan, Cellulose 4 (1997) 173. [16] D.L. VanderHart, R.H. Atalla, Macromolecules 17 (1984) 1465. [17] J. Sugiyama, J. Persson, H. Chanzy, Macromolecules 24 (1991) 2461. [18] C. Tokoh, K. Takabe, M. Fujita, H. Saiki, Cellulose 5 (1998) 249. [19] R.H. Atalla, D.L. VanderHart, Science 223 (1984) 283. [20] H. Yamamoto, F. Horii, Cellulose 1 (1994) 57. [21] K. Schenzel, S. Fischer, E. Brendler, Cellulose 12 (2005) 223. [22] A. Svensson, E. Nicklasson, T. Harrah, B. Panilaitis, D.L. Kaplan, M. Brittberg, P. Gatenholm, Biomaterials 26 (2005) 419. [23] H. Bäckdahl, G. Helenius, A. Bodin, U. Nannmark, B.R. Johansson, B. Risberg, P. Gatenholm, Biomaterials 27 (2006) 2141. [24] K. Watanabe, M. Tabuchi, Cellulose 5 (1998) 187. [25] J.D. Fontana, A.M. de Souza, C.K. Fontana, I.L. Torriani, J.C. Moreschi, B.J. Gallotti, S.J. de Souza, G.P. Narcisco, J.A. Bichara, L.F.X. Farah, Appl. Biochem. Biotech. 24 (1990) 253. [26] I. Pitanguy, F. Salgado, P.F.D. Maracaja, Rev. Bras. Cir. 78 (1988) 317. [27] R. Jonas, L.F. Farah, Polym. Degrad. Stab. 59 (1998) 101. [28] P.N. Galgut, Biomaterials 11 (1990) 561. [29] J.J.P. Siqueira, J.C. Moreschi, Cir. Vasc. Angiol. 16 (2000) 179. [30] J.J.P. Siqueira, Etica Psihiatr. 9 (2003) 10. [31] B.∅. Palsson, S.N. Bhatia, Tissue Engineering, Pearson Prentice Hall, Upper Saddle River, NJ, 2004. [32] M. Schramm, Z. Gromet, S. Hestrin, Biochem. J. 67 (1957) 669. [33] W. Helbert, J. Sugiyama, S. Kimura, T. Itoh, Protoplasma 203 (1998) 8490. [34] I.M. Saxena, K. Kudlicka, K. Okuda, R.M. Brown Jr., J. Bacteriol. 176 (1994) 5735. [35] M. Koyama, W. Helbert, T. Imai, J. Sugiyama, B. Henrissat, Proc. Natl. Acad. Sci., Biophysics, vol. 94, 1997, p. 9091. [36] G.I. Mantanis, R.A. Young, R.M. Rowell, Cellulose 2 (1995) 1. [37] J. George, K.V. Ramana, S.N. Sabapathy, A.S. Bawa, World J. Microb. Biotechnol. 21 (2005) 1323. [38] A.N. Nakagaito, S. Iwamoto, H. Yano, Appl. Phys., A 80 (2005) 93.
Materials Science and Engineering C 28 (2008) 549 – 554 www.elsevier.com/locate/msec
Template assisted synthesis of porous nanofibrous cellulose membranes for tissue engineering C.R. Rambo a,b,⁎, D.O.S. Recouvreux b , C.A. Carminatti b , A.K. Pitlovanciv b,c , R.V. Antônio b,c , L.M. Porto b a Group of Ceramic and Glass Materials – CERMAT, Brazil Genomic Engineering Group, Department of Chemical and Food Engineering, Brazil Department of Biochemistry, Federal University of Santa Catarina, P.O. Box 476, 88040-900 Florianópolis, SC, Brazil b
c
Received 29 November 2006; received in revised form 8 November 2007; accepted 30 November 2007 Available online 8 December 2007
Abstract Porous, nanofibrous bacterial cellulose (BC) membranes were produced by the bacterium Acetobacter xylinum. The bacterium was cultivated in an appropriate culture medium under static conditions. In situ pore formation was attained through the use of pin templates with diameters varying from 60 to 300 μm composed of polyestirene (ϕ = 300 μm) or optical fibers (ϕ = 60 μm), which were placed on culture medium with the pins immersed in the liquid. Cellulose biosynthesis occurred around the pins leaving tiny pores on the cellulose membrane. After removal of the template the biofilm was dried at 50 °C/24 h. Physico-chemical properties of BC membranes, like degree of crystallinity, swelling and tensile strength were not significantly altered after pore formation. Microstructure evaluation revealed that the film matrix is composed of long nanofibers isotropically distributed on its surface. Round-shaped pores with diameters varying between 60 and 300 μm, depending on the pin template used, were formed in the cellulose membranes. These pores exhibited no border failures that could start crack propagation along the film surface. Microporous membranes could be useful for applications in repairing tissues, which require high oxygenation rates or wound contracture delay. © 2007 Elsevier B.V. All rights reserved. Keywords: Tissue engineering; Porous membranes; Scaffolds; Bacterial cellulose
1. Introduction Tissue engineering is an interdisciplinary field that applies the principles of engineering and the life sciences toward the development of biological substitutes that restore, maintain, or improve tissue or a whole organ function. Typical application examples are implants of scaffolds composed of biodegradable polymers or inert materials coated with bioactive biomaterials that allow growth of new tissues of particular types of cells [1,2]. After the formation of the new tissue, polymeric scaffolds are gradually degraded into small molecular weight compounds, which can be absorbed by the body or excluded out of the body [3]. ⁎ Corresponding author. Genomic Engineering Group, Department of Chemical and Food Engineering, Brazil. E-mail address: [email protected] (C.R. Rambo). 0928-4931/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.msec.2007.11.011
In recent years the search for new classes of biopolymers, with specific properties to be used as scaffolds in tissue engineering, has attained great interest, like cellulose, polyhydroxyalkanoates, polylactates and blends of these materials [4–9]. In natural environments biofilms are associated with some microbial styles of life [10] and constitute a kind of selfprotection that allows microorganisms to survive in adverse conditions [11]. Bacterial cellulose (BC) is a kind of extra cellular polysaccharide present in the biofilm produced by several bacteria, notably by Acetobacter xylinum, as long nanofibers. This polymer is highly crystalline and its degree of crystallinity varies depending on the origin and mode of chemical treatment [12]. BC is composed by glucose molecules joined by β(1 → 4)-glycosidic bonds forming branchless linear chains [13,14]. Cellulose biosynthesis is a process involving several steps regulated by some individual
550
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
and well characterized complex enzymes and proteins. This process includes the synthesis of uridine diphosphate glucose (UDP-glucose), which is the precursor of cellulose biosynthesis, followed by glucose polymerization in β(1 → 4)-glucan chains by complex proteins [13,14]. Bacterial cellulose fibers exhibit a wide range of dimensions ranging from 1 to 25 nm in width, which corresponds to 10–250 polyglucan chains and from 1 to 9 μm in length (formed by 2000–18,000 glucose residues). Among six known polymorphs of cellulose, two common crystalline forms are of interest: cellulose I and II [15]. It is known that in the cellulose I, which is synthesized by plants and also by A. xylinum in static culture, parallel β(1 → 4)-glucan chains are uniaxially disposed, whereas β(1 → 4)-glucan chains of cellulose II are randomly arranged. Furthermore, crystalline cellulose I exhibits two main unit cell structures called cellulose Iα and Iβ, which consist of one chain triclinic and two chain monoclinic unit cells, respectively [16,17], which are difficult to distinguish by X-ray diffractometry. The cellulose synthesized by A. xylinum is known to be composed by cellulose Iα as majority phase and cellulose Iβ in smaller fraction [18–20]. The degree of crystallinity of cellulose influences some physico-chemical properties of cellulose like swelling and water binding [21]. Bacterial cellulose is used in a wide range of applications, from the food industry to electroacustic devices, as phone diaphragms. Bacterial cellulose is also potentially suitable for applications as scaffolds in tissue engineering due to its unique properties, including a high water retention capability (hydrophilic), a fine fibrous network, which allows cell growth and proliferation, high tensile strength, in situ moldability and low production cost [22–26]. In medicine, non-porous cellulose membranes are used as stent coatings, for dura-mater substitution in tumor or trauma cases or as skin protection in cases of burn and deep wounds. In odontology, cellulose films are applied for periodontal tissue recovering [27,28]. Despite the studies concerning bacterial cellulose membranes in medical applications as substitute for cartilaginous and skin tissues, production of porous bacterial cellulose films were reported only recently by Siqueira and Moreschi [29] and are not widely used in tissue engineering. Porous membranes
Fig. 2. Schematic flow chart for the production of porous bacterial cellulose membranes.
consisting of large pores are usually indicated for preventive and healing treatments of wounds, in particular those in which exudation and oxygenation are necessary [30]. Among the possibilities of using porous membranes, a particular case is of great interest in tissue engineering: wound healing mechanism. Porous membranes composed of collagen I and chondroitin-6sulfates are able to delay wound contracture and promote dermal regeneration. The optimal pore size lies between 20 and 120 μm, which promote the delay of fibroblast migration and avoids wound contraction and scar formation, allowing time for wound healing [31]. Fig. 1 shows the range of possible applications of porous cellulose membranes as a function of the pore diameter. This work reports the biosynthesis of nanofibrous bacterial cellulose films containing in situ produced micropores. An innovative design of features (porosity, pore size distribution and pore geometry) of the cellulose membrane was developed to modify/improve a determined property for a specific application, which in turn intrinsically depends on these features. The flow chart shown in Fig. 2 illustrates schematically the multidisciplinary approach for the manufacturing of the porous cellulose membranes. 2. Experimental 2.1. Synthesis
Fig. 1. Schematic diagram of the several applications of porous BC membranes in function of the pore diameter.
Bacteria A. xylinum ATCC 23769, acquired from “Collection of Tropical Culture (CCT)” (André Tosello Foundation) was used for the cellulose production. A. xylinum inoculum was prepared by its cultivation under static conditions, at 30 °C, in a 125 mL Erlenmeyer flasks, containing 25 mL of Hestrin & Schramm medium [32]. Cellulose production was carried out by the cultivation of the bacterium for 7 days, under static conditions at 30 °C, in 250 mL Erlenmeyer flasks containing 50 mL of modified Hestrin & Schramm medium, pH 6.6, containing (per liter): 5.0 g peptone, 5.0 g yeast extract, 1.15 g citric acid, 2.27 g Na2HPO4 and 20 g glucose. In order to produce porous
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
551
Microscopy (trinocolar, XDS-1B, 40 × objective) observations were carried out in order to evaluate the morphology and distribution of the micropores of dried PBC membranes. Tensile strength of BC and PBC at room temperature was determined on a set of 5 samples of each membrane with nominal dimensions of 70 × 25 mm using a universal testing device (Instron Corp., Instron 4202). The speed of the crosshead was set constant to 1 mm/min. 3. Results and discussion
Fig. 3. Photograph of the BC membrane in humid state (gel).
cellulose membranes, pin array structures (similar to a brush) with diameters varying from 60 to 300 μm composed of polystyrene (ϕ = 300 μm) or optical fibers (ϕ = 60 μm) were introduced into the media allowing in situ pore formation during cultivation and cellulosic membranes formation. The porous BC membranes were coded PBC300 and PBC60, which contained pore diameters of 300 μm and 60 μm, respectively. After the cultivation period, the formed gel at the liquid surface and in between the pin array template was removed, washed with deionized water and dried at 50 °C until constant weight. Subsequently, the cellulose membrane was treated with a 0.1 M NaOH solution at 90 °C for 30 min to remove bacterial impurities and eventual cell debris. The membrane was again washed in deionized water until neutral pH and then dried at 50 °C for 24 h.
In the humid state the biofilm formed on the surface of the culture medium of the bacterium A. xylinum is a homogenous transparent, moldable and handle-resistant gel (Fig. 3). After drying, the thickness of the membranes varied between 150 μm and 200 μm. A more detailed analysis by SEM revealed that the zooglea after drying is composed of long fibers. Fig. 4 shows the SEM micrograph of the dried membrane surface, highlighting details of the microstructure. The film consists of a dense, flawless and homogeneous cellulose matrix (Fig. 4a). The matrix is formed by an interlaced network of long cellulose nanofibers with high aspect ratio and mean diameters smaller than 100 nm (Fig. 4b). This high aspect ratio may affect the mechanical properties of the film such as elasticity and tensile strength. The reason for that fibrous morphology was subject of discussion in the last decade and is already established
2.2. Characterization Swelling behaviour of the BC membranes was determined by water and physiological solution (0.9%NaCl solution) absorption measurements in samples of 15 × 20 × 0.02 mm. The measurements consisted in immersing previously dried membranes into the solution for a determined period and measuring the mass, after removal of excess of solution on the surface. The experiment was carried on until no further absorption was observed. The percent weight gains of the samples were calculated and plotted against time. X-ray diffractometry (XRD, Philips, X'Pert) with CuKα radiation (λ = 1.45 nm) was used to identify the phases of the cellulose membrane. Dried BC and PBC60 membranes were placed on an aluminium plate and measured over a 2θ interval of 5°–40° with steps of 1 °/min. JCPDS-files were used for phase identification. The degree of crystallinity (XC) of the porous cellulose membrane was estimated from the diffractogram through the ratio between the areas under the (110) and (200) crystalline peaks and the total area, according to the method described by Watanabe and Tabuchi [24]. The microstructure of the cellulose membranes was characterized by Scanning Electron Microscopy (SEM, Philips, XL-30). For the observations, BC membranes with diameters of 5 mm were dried and placed over an aluminium support and sputtered with gold. Optical
Fig. 4. SEM micrographs of the surface of a bacterial cellulose membrane.
552
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
[15,33]. Cellulose biosynthesis is characterized by unidirectional growth and crystallization, where glucose molecules are linear bonded by β(1 → 4)-glycosidic bonds. The union of glycosidic chains forms oriented microfibrils with intramolecular hydrogen bonds [34]. The preferential orientation of cellulose crystal growth during biosynthesis was already reported by Koama et al. [35]. The cellulose is crystallized outward the organisms, particularly in A. xylinum that synthesizes cellulose chains by introducing glucose units to the reducing ends of the polymer. The growth mechanism during bacterial activity determines the morphology of the final cellulose. Fig. 5 shows the swelling behaviour of the PBC60 membrane under water and physiological solution. The absorption behaviours under water and physiological solution were similar within the statistical deviation. After 30 min the weight gain reached a maximum (approximately 175%), slightly decreasing to 153% between 0.5 and 1 h. A stable state was achieved after 1 h. Swelling occurs by a combined effect of both fiber cell walls swelling and breaking of interfiber bonds that involves fiber displacement. Therefore, the displacement of the fibers causes an expansion of the volume and results in a free space for the swelling agent. The maximum expansion of the fibers depends on the absorption capacity of the fibers in the liquid, i.e. ability of the swelling liquid to form hydrogen bonds with the hydroxyl groups of cellulose [36]. Additionally cellulose water sorption may considerably affect the properties of cellulose-based materials, e.g. medical products. Fig. 6 shows the X-ray diffraction pattern of the surface of the bacterial cellulose membrane and of the porous membrane. Three main peaks can be identified in both spectra, which are assigned to the (110), (110) and (200) reflexions planes of cellulose I [18]. Despite the slight difference in peak intensities between BC (Fig. 6a) and PBC60 (Fig. 6b), no peak shift can be seen in the spectra, which indicates that no changes on the crystalline structure occurred that could be attributed to the influence of pin templates. The estimated degree of crystallinity of the BC membrane was 56 ± 5% and of the PBC60 membrane was 51 ± 5%. Although it was an estimative of XC based on the X-ray diffractograms, it is relevant to point out that the cellulose
Fig. 5. Swelling behaviour of the BC membranes in function of time for water and physiological solution.
Fig. 6. X-ray diffractogram of the BC (a) and PBC60 (b) membranes. Peaks indexed are assigned to cellulose I type.
membranes exhibit similar crystallinity within the accuracy of the method. Fig. 7 shows typical stress–strain curves of the BC and PBC300 membranes under tension. Both membranes exhibited relative low elongation under tension before breaking. Rupture occurred at a strain of 1.7% for the porous membrane (PBC300) and at 3.4% elongation for the BC membrane. Tensile strengths of the BC and PBC300 were 32 MPa and 28 MPa, respectively. Tensile strength of untreated BC membranes is reported to be higher than 65 MPa [37]. The lower strength of NaOH-treated membranes may be attributed to their swelling behaviour. During swelling of cellulose fibers, intermolecular bonds between the fibers break as a result of the internal stress produced by absorption [36]. Due to swelling, the ordering within the fibers will be reduced and this may contribute to the reduction in mechanical strength. It must be also noted that, although the PBC300 membrane is a porous material, it exhibited similar strength to the non-porous membrane, which is also an indication of the efficiency of the method of pore production, concerning the integrity of the PCB after processing.
Fig. 7. Typical room temperature stress–strain curves of BC and PBC300 membranes under tension.
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554 Table 1 Key properties of bacterial cellulose (Td = degradation temperature) Property
BC (other works)
Ref.
BC (this work)
PBC300
Tensile strength Td Crystallinity
10–70 MPa
[37,38]
32 ± 5 MPa
28 ± 5 MPa
N300 °C 40–70%
[12] [15,18,21,24]
– 56 ± 5%
341 °C 51 ± 5%
Table 1 summarizes the key properties of the BC and PBC membranes obtained in this work in comparison with other literature data. The physico-chemical properties of the porous membrane were not significantly affected by the presence of pores. Fig. 8 shows optical (a and b) and SEM images (c and d) of dried porous cellulose membranes with different pore diameters. Round-shaped pores with mean diameter of 300 μm distributed on the cellulose surface with a density of approximately 2 ppi (pores per linear inch) can be observed (Fig. 8a). The distance between the pores was set by arranging the polymeric pins in a distance of approximately a pin diameter from each other. The border of the pores exhibits no flaw and surrounds the pore uniformly, which provides isotropic strength distribution to the porous membrane (Fig. 8b). Large pores (N200 μm) are potentially useful in applications that allow high
553
oxygenation rates and wound exudation release. Concerning the last case, in treatments of deep wounds, where the exudation (with high or low viscosity) must be eliminated, porous scaffolds permit fast granulation and regeneration of the damaged tissue, with subsequent epithelial regeneration [29]. Fig. 8c shows the cellulose membrane synthesized with a template with pins of 60 μm of diameter. A pore round-shaped with approximately 60 μm of diameter is displayed. Some folds are also present around the pore, which resulted from the sample preparation for the SEM analysis. Details of the boarder of the pore can be seen in Fig. 8d. Analogous to the large pores, the small ones are round-shaped and exhibit no border failures that could start crack propagation along the film surface and therefore tearing the scaffold. The bacteria form a protective border that surrounds the pin surface leaving a rope-kind reinforcement structure, which could be the reason for the relative high tensile strength compared to the non-porous membrane. Pores with diameters below 100 μm are suitable for wound contracture by delaying fibroblast migration through the porous template. Beside the use of thin pore membranes with small pores for medical applications, pore sizes in the range of 10– 100 μm are interesting materials to promote selective cell migration through the pores for specific cell cultures, with different cell sizes.
Fig. 8. Optical (a,b) and SEM (c,d) micrographs of porous cellulose membranes. Figs. b and d are higher magnitude images from the same sample shown in Figs. a and c, respectively.
554
C.R. Rambo et al. / Materials Science and Engineering C 28 (2008) 549–554
4. Conclusions Porous nanofibrous bacterial cellulose membranes with controlled pore sizes and shapes were produced by appropriate cultivation of A. xylinum. The pin templates used as pore formers acted as bioinert surfaces, allowing the bacteria to synthesize cellulose fibers around the pins. The physical-chemical properties of BC membranes were not significantly affected after pore formation. Round-shaped pores with diameters in the range of 60 to 300 μm were achieved, depending on the pin diameter of the template. The pores were uniformly distributed on the cellulose membrane. After membrane drying the pore boarders exhibited no visible failures down to the submicrometer scale that could, under tension forces, start crack propagation along the film surface. The range of the pore diameter determines the most suitable application in tissue regeneration. The prepared membranes exhibited suitable macro- and microstructure that might be used for applications such as repairing tissue for cartilage and skin due to its nanofibrous network which could promote cell growth, and a porous network, which improves oxygenation of the repairing tissue, allow exudation flow and promote delay on wound contracture. Acknowledgements The authors thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil for the financial support. The authors also thank André Tosello Foundation for the gently donation of A. xylinum strain. References [1] E. Wintermantel, J. Mayer, J. Blum, K.-L. Eckert, P. Lüscher, M. Mathey, Biomaterials 17 (1996) 83. [2] C.R. Rambo, F.A. Mueller, L. Mueller, H. Sieber, I. Hofmann, P. Greil, Mater. Sci. Eng. C 26 (2006) 92. [3] R. Langer, J.P. Vacanti, Science 260 (1993) 920. [4] D. Chen, B. Sun, Mat. Sci. Eng. C 11 (2000) 57. [5] K. Zhao, Y. Deng, J.C. Chen, G.-Q. Chen, Biomaterials 24 (2003) 1041.
[6] S.V. Madihally, H.W.T. Matthew, Biomaterials 20 (1999) 1133. [7] Y.Z. Wan, Y. Huang, C.D. Yuan, S. Raman, Y. Zhu, H.J. Jiang, F. He, C. Gao, Mat. Sci. Eng. C 27 (2007) 855. [8] S. Ramakrishna, J. Mayer, E. Wintermantel, Kam W. Leong, Compos. Sci. Technol. 61 (2001) 1189. [9] F.A. Muller, L. Muller, I. Hofmann, P. Greil, M.M. Wenzel, R. Staudenmaier, Biomaterials 21 (2006) 3955. [10] P. Watnick, R. Kolter, J. Bacteriol. 182 (2000) 2675. [11] J.W. Costerton, P.S. Stewart, E.P. Greenberg, Science 284 (1999) 1318. [12] J. George, K.V. Ramana, S.N. Sabapathy, J.H. Jagannath, A.S. Bawa, Int. J. Biol. Macromol. 37 (2005) 189. [13] R.M. Brown Jr., J.H. Willison, C.L. Richardson, Proc. Natl. Acad. Sci., Cell Biology, vol. 73, 1976, p. 4565. [14] P. Ross, R. Mayer, M. Benziman, Microbiol. Rev. 55 (1991) 35. [15] A.C. O'Sullivan, Cellulose 4 (1997) 173. [16] D.L. VanderHart, R.H. Atalla, Macromolecules 17 (1984) 1465. [17] J. Sugiyama, J. Persson, H. Chanzy, Macromolecules 24 (1991) 2461. [18] C. Tokoh, K. Takabe, M. Fujita, H. Saiki, Cellulose 5 (1998) 249. [19] R.H. Atalla, D.L. VanderHart, Science 223 (1984) 283. [20] H. Yamamoto, F. Horii, Cellulose 1 (1994) 57. [21] K. Schenzel, S. Fischer, E. Brendler, Cellulose 12 (2005) 223. [22] A. Svensson, E. Nicklasson, T. Harrah, B. Panilaitis, D.L. Kaplan, M. Brittberg, P. Gatenholm, Biomaterials 26 (2005) 419. [23] H. Bäckdahl, G. Helenius, A. Bodin, U. Nannmark, B.R. Johansson, B. Risberg, P. Gatenholm, Biomaterials 27 (2006) 2141. [24] K. Watanabe, M. Tabuchi, Cellulose 5 (1998) 187. [25] J.D. Fontana, A.M. de Souza, C.K. Fontana, I.L. Torriani, J.C. Moreschi, B.J. Gallotti, S.J. de Souza, G.P. Narcisco, J.A. Bichara, L.F.X. Farah, Appl. Biochem. Biotech. 24 (1990) 253. [26] I. Pitanguy, F. Salgado, P.F.D. Maracaja, Rev. Bras. Cir. 78 (1988) 317. [27] R. Jonas, L.F. Farah, Polym. Degrad. Stab. 59 (1998) 101. [28] P.N. Galgut, Biomaterials 11 (1990) 561. [29] J.J.P. Siqueira, J.C. Moreschi, Cir. Vasc. Angiol. 16 (2000) 179. [30] J.J.P. Siqueira, Etica Psihiatr. 9 (2003) 10. [31] B.∅. Palsson, S.N. Bhatia, Tissue Engineering, Pearson Prentice Hall, Upper Saddle River, NJ, 2004. [32] M. Schramm, Z. Gromet, S. Hestrin, Biochem. J. 67 (1957) 669. [33] W. Helbert, J. Sugiyama, S. Kimura, T. Itoh, Protoplasma 203 (1998) 8490. [34] I.M. Saxena, K. Kudlicka, K. Okuda, R.M. Brown Jr., J. Bacteriol. 176 (1994) 5735. [35] M. Koyama, W. Helbert, T. Imai, J. Sugiyama, B. Henrissat, Proc. Natl. Acad. Sci., Biophysics, vol. 94, 1997, p. 9091. [36] G.I. Mantanis, R.A. Young, R.M. Rowell, Cellulose 2 (1995) 1. [37] J. George, K.V. Ramana, S.N. Sabapathy, A.S. Bawa, World J. Microb. Biotechnol. 21 (2005) 1323. [38] A.N. Nakagaito, S. Iwamoto, H. Yano, Appl. Phys., A 80 (2005) 93.