- Email: [email protected]
Development of biodegradable antibacterial cellulose based hydrogel membranes for wound healing
International Journal of Biological Macromolecules 67 (2014) 22–27
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Development of biodegradable antibacterial cellulose based hydrogel membranes for wound healing Nelisa Türko˘glu Lac¸in ∗ Science and Technology Application and Research Center, Yıldız Technical University, 34349 Istanbul, Turkey
a r t i c l e
i n f o
Article history: Received 13 January 2014 Received in revised form 26 February 2014 Accepted 3 March 2014 Available online 12 March 2014 Keywords: Wound healing 2,3 Dialdehyde bacterial cellulose Antibacterial activity
a b s t r a c t Cellulose-based hydrogels have wide applications in tissue engineering and controlled delivery systems. In this study, chloramphenicol (CAP) loaded 2,3 dialdehyde cellulose (DABC) hydrogel membranes were prepared, characterized and their antibacterial efficacy was evaluated. Bacterial cellulose (BC) secreted by Acetobacter xylinum was modified to become DABC by oxidation via the sodium metaperiodate method. CAP–BC and CAP–DABC interactions were illustrated via ATR–FTIR analysis. Water retention capacity of BC and DABC membranes were determined as 65.6 ± 1.6% and 5.3 ± 0.3%, respectively. CAP release profiles were determined via HPLC analysis. The drug-loading capacities of BC and DABC membranes were 5 mg/cm2 and 0.1 mg/cm2 , respectively. Membranes released 99–99.5% of the contained CAP within 24 h and an initial burst release effect was not observed. In vitro antibacterial tests of BC and DABC, both CAPloaded, demonstrated their ability to inhibit bacterial growth for a prolonged duration. Antimicrobial effect against bacteria was still prevalent after 3 days of incubation period with disc diffusion tests. The MTT test results reveal that fibroblast adhesion and proliferation on CAP-loaded DABC membranes were noticeably higher than CAP-loaded BC membrane. This newly developed drug containing DABC membranes seem to be highly suitable for wound healing due to its unique properties of biodegradability, biocompatibility, and antimicrobial effectiveness. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Three-dimensional wound dressing substances must cover the wound, serve as a physical barrier against external infection, and provide cellular support. A successful tissue scaffold should possess the appropriate physical and mechanical characteristics, in addition to possessing an appropriate surface chemistry to facilitate cellular attachment, proliferation, and differentiation [1]. Cellulose, a linear polysaccharide, is one of the most abundant organic materials in nature with a variety of useful applications. A few bacterial species, taxonomically closely related to the genus Acetobacter xylinum (A. Xylinum), produce and extracellularly secrete cellulose in the form of fibre called BC [2]. It is a natural polymer with potential due to its unique structural and mechanical properties [3]. A. xylinum is a simple gram-negative bacterium that has an ability to synthesize high-quality cellulose composed of twisting ribbons of microfibrillar bundles [4]. The thick, gelatinous membrane formed in static culture conditions as a result of these processes is characterized by a 3D structure consisting of an ultrafine network of
∗ Tel.: +90 532 2257368; fax: +90 212 3832974. E-mail address: [email protected] http://dx.doi.org/10.1016/j.ijbiomac.2014.03.003 0141-8130/© 2014 Elsevier B.V. All rights reserved.
cellulose nanofibres (3–8 nm), which are highly uniaxially oriented [5]. BC is a natural hydrogel. Hydrogels are defined as threedimensional polymer networks swollen by large amounts of a solvent. BC is a hydrogel that can take water up to 99% its own weight, mainly due to its amorphous structure [6]. Owing to its unique nano-scaled three-dimensional network structure, BC has high water retention, high mechanical strength, and outstanding biocompatibility, which enable it to serve as a natural scaffolding material for the regeneration of a wide variety of tissues [7]. Dialdehyde cellulose (DABC) is a cellulose derivate produced by regioselective oxidation of cellulose with the use of periodate as an oxidation agent. It is biodegradable and biocompatible and has a large potential to be used in many applications [8,9]. Wei et al. have prepared a type of new functional dry BC film containing benzalkonium chloride as a potential antimicrobial wound dressing material and reported that it has strong antibacterial properties that especially resist both Staphylococcus aureus and Bacillus subtilis (gram-positive bacteria). Also, a stable and lasting release of the antimicrobial agent for at least 24 h was reported [10]. There are also commercial cellulose-based wound healing systems such as Biofill, Gengiflex, and XCell. Cellulose-based hydrogels are biocompatible, have low production costs, and are non-toxic. Therefore,
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
cellulose-based hydrogels have wide applications in tissue engineering and controllable delivery systems [11,12]. The high water absorption ability of the antimicrobial BC dry film is crucial for wound dressing to absorb blood and tissue fluid on acute traumas and would promote wound healing [10]. CAP was chosen as a model biologically active agent because it presents a wide antibacterial spectrum with a bactericide activity on gram-negative and grampositive bacteria [13]. The aim of this study was to develop CAP loaded membranes as antimicrobial wound healing materials, investigate the antimicrobial activities against model gram-positive and gram-negative bacteria (S. aureus, Streptococcus pneumoniae and Escherichia coli), and to observe the proliferation and attachment of fibroblasts on drug-loaded membranes. 2. Experimental procedure 2.1. Materials E. coli (ATCC 25293), S. aureus (ATCC 25925) and S. pneumoniae (ATCC6301) were purchased from Institute of Microbiology Chinese Academy of Science. 2.2. Production of BC and preparation of DABC membranes The production of bacterial cellulose membranes (BCMs) was achieved by growing A. xylinum (ATCC 10245) in Hestrin–Schramm medium, pH 5.1(adjusted by 1 M HCl). The medium containing 20 g/L glucose, 10 g/L bactopeptone, 10 g/L yeast extract, 4 mM KH2 PO4 , and 6 mM K2 HPO4 was used to produce cellulose pellicles in static culture. The inoculum was prepared by growing A. xylinum at 30 ◦ C using a rotary shaker for 3 days. The BC nanofibre formation was allowed to occur in a period of 7 days after inoculating the subculture in the proportion of 1:10 in petri dishes statically. The harvested BC membranes were washed with distilled water to remove s medium components and then incubated in 1 M NaOH solution at 80 ◦ C for 2 h to eliminate attached cells and other impurity. After that, the BC membranes were further purified to remove other residues by distilled water washing until the pH of the washing liquid was neutral. Finally, the bacterial cellulose membranes were cut into 1 cm2 size [14]. The morphology and microstructure of membranes were characterized by scanning electron microscopy (FE-SEM, Zeiss supra 55) and AFM (Park systems/XE-100E advance scanning probe microscope). Prior to the FE-SEM observation, all samples were sputter coated with a thin layer of platinum to avoid electrical charging. Periodate oxidized cellulose is often referred to as dialdehyde cellulose (DABC) [2]. Oxidation of cellulose using sodium metaperiodate has been extensively investigated in the literature and it is found to lead to selective cleavage at the C-2 and C-3 vicinal hydroxyl groups to yield a product with 2,3-dialdehyde units along the polymer chain [15,16]. The dialdehyde groups also serve as reactive chemical anchors for further reactions, conducive for the chemical derivatization of cellulose. The presence of dialdehyde groups in bacterial cellulose also improves biodegradability and they are discussed in relation to tissue-scaffold engineering [17,18]. Periodate oxidate BCMs were prepared by placing the membranes in DI water, adding 1.3 times the weight sodium metaperiodate and gently stirring at 55 ◦ C in dark for 4 h. After the excess periodate was decomposed with ethylene glycol, the DABC membranes were washed by DI. 2.2.1. Determination of dialdehyde content Determination of the aldehyde content of DABC was based on the oxime reaction between aldehyde group and NH2 OH·HCl. The periodate oxidized cellulose, which never underwent drying procedures, has been placed in a 250 mL beaker containing 1.39 g of
23
NH2 OH·HCl dissolved in 100 mL of 0.1 M acetate buffer (pH = 4.5). The beaker was covered with a thin rubber foil and the mixture was stirred for 48 h at room temperature with a magnetic stirrer. The product was filtrated and washed with 600 mL of deionized water after which it was dried in a freeze-dryer. For the determination of the aldehyde content of DABC, five membranes were periodate oxidized under the same conditions and the nitrogen content of the oxime derivate of DABCs was determined by using a Leco truSpec series elemental analyzer [19]. The results are expressed as the average values of five samples. 2.2.2. Water retention capacity of hydrogels In order to dry the membranes without harming the 3D nanonetwork structure, the porous membranes were frozen overnight in a freeze dryer (Christ Alpha 2-4 LD) at −80 ◦ C for 24 h. To determine the water retention capacity of membranes, the freeze dried 1 cm2 BC and DABC membranes were immersed in phosphate-buffered saline (PBS) at room temperature until equilibration. After that, the membranes were removed from the PBS and excess PBS at the surface of the membranes was blotted out with Kim wipes paper. The weights of the freeze dried membranes were measured, and then the membranes were swollen until no further weight change was observed and the procedure was repeated. The results are expressed as the average values of five samples. The water retention of membranes was calculated with the following formula: Wh : Weight of hydrate membrane, Wd : weight of dry membrane: WAC(%) = (Wh − Wd )/Wd × 100 [2]. 2.3. Preparation of antimicrobial membranes CAP was loaded into the hydrogels using the swelling-diffusion method. The interaction of CAP with the membranes was investigated by attenuated total reflectance—Fourier transform infrared (Perkin-Elmer). The 1 cm2 sized freeze-dried membranes were immersed in CAP stock solution (0.05 g/mL) (CAPstock sol. ) for 8 h. Membranes were removed from the CAP stock solution (CAPresidual ) and immersed in 1 mL distilled water for 10 s to remove the nonabsorbed CAP (CAPwashing sol. ). The amount of CAP loaded into membranes was calculated with the following formula:
CAPmembrane = (CAPstock sol. ) − CAPresidual + CAPwashing sol.
Afterwards, CAP-loaded membranes were placed in PBS (10 mL) and incubated at 37 ◦ C, 50 rpm for 24 h. CAP amount released from membranes was determined by high performance liquid chromatography (HPLC) to verify the calculation above. The same treatment was applied to each membrane and another freeze–drying step was administered. Finally, 30 min of UV sterilization was performed. 2.4. Determination of antibacterial activity The antimicrobial activities were investigated against E. coli, S. pneumoniae, and S. aureus using two different methods explained below. 2.4.1. Growth curve method Single colonies of E. coli, S. Aureus, and S. pneumoniae were grown on nutrient agar. Cell suspensions for inoculum were arranged according to standard no 0.5 as 1–1.5 × 108 CFU/mL. The positive control was the bacteria in nutrient broth, the negative control the culture with the pure membrane in nutrient broth, and the blank was the test tube only containing nutrient broth. Cultures were incubated at 37 ◦ C for 24 h. Samples drawn from the system every 2 h were analyzed spectrophotometrically by measuring the
24
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
absorbance at 600 nm. Triplicate experiments were carried out for each bacterium. 2.4.2. Disc diffusion method E. coli, S. aureus, and S. pneumoniae were spread on nutrient agar. Then, the CAP-loaded BCMs and DABC membranes were placed in the nutrient agar in petri dishes. BC and DABC membranes without the drug were the control group. The agar plates were then incubated at 37 ◦ C for 48 h. After incubation, the semi-diameters of the inhibition zone were measured. Each sample was tested in triplicate, using different plates. 2.5. The release of chloramphenicol from BC and DABC membranes In vitro CAP release profiles were observed by immersing CAPloaded hydrogel samples into 10 mL of PBS glass flask covered by aluminium foil. CAP-loaded membranes were incubated at 37 ◦ C, 50 rpm for 24 h. Samples removed from the release medium every 2 h were analyzed by HPLC [20]. HPLC experiments were conducted using an Agilent HPLC System. Separations of CAP were achieved using a 250 mm ACE 5 C18 column. Mobile phase compositions were [A] ACN, [B] H2 O ([A] 70% and [B] 30%, respectively). The flow rate was 1 mL/min, column temperature was 40 ◦ C and the analysis was performed by DAD detector at 278 nm [21]. Triplicate experiments were performed for each membrane. 2.6. Biocompatibility testing In order to evaluate the biocompatibility of the drug loaded membranes, fibroblast cell line L929 (ECACC, European Collection of Cell Cultures) was used. The cells were cultured in Dulbecco’s modified Eagle’s medium low glucose (Sigma-Aldrich) supplemented with 10% fetal bovine serum and 1% antibiotics/antimicotics (Sigma-Aldrich) until they reached the confluence. They were then trypsinized and seeded on UV-sterilized samples using the density of 2 × 104 cells/membrane. Cell proliferation was assessed by measuring the cellular enzyme activity using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. This assay consists of the bioreduction of the substrate by mitochondrial enzyme succinate dehydrogenase into a purple colour salt insoluble in water. According to the standard procedure, membranes were washed with PBS buffer to remove unattached cells and were placed into new well-plates after 1, 3, and 7 days of incubation. MTT reagent was added to each well and the plates were incubated for 4 h at 37 ◦ C in a humidified atmosphere containing 5% of CO2 . After the reaction occurred the intracellular formazan salt was solubilized using isopropanol. The absorbance of solubilized formazan was measured at 570 nm with a spectrophotometer (Biotek Instruments). The results are expressed as the average absorbance of triplicate samples. 3. Results and discussion 3.1. Characterization of BC, DABC and CAP-loaded BC/DABC membranes The 3D nano-network structure of BC was observed by AFM and SEM microphotographs (Figs. 1 and 2). Bacterial cellulose microfibril sizes range between 60 and 100 nm [22]. In this study, the width and pore size of cellulose nanofibres were identified to be between 80–100 nm and 60–90 nm, respectively. In other studies, the width of cellulose ribbons were observed to be between 10 nm and 100 nm [23,24]. Water retention capacity is the most important property directly involved in the biomedical applications of BCMs
Fig. 1. SEM microphotographs of 3D nano-network structure of BC.
as a dressing material. The proper moisture content of a dressing material accelerates the wound healing process and protects it against contamination [25]. Additionally, the high water absorption ability of the antimicrobial BCM is an important feature to promote wound healing for wound dressings [26]. Data shows good water-binding ability with a swelling ratio of 65.6% in PBS for BCM. The swelling ratio of DABC was determined to be 5.3%, which was significantly lower than wac of BC. BCM prepared in our work could hold water at least 65.6 times its own weight within 24 h. Kinetics of the swelling process is typical of hydrogels. The swelling degree limitations are specific to each hydrogel. Similarly, the time necessary to reach this threshold depend on the reaction conditions for the hydrogel. The greater maximal swelling degree (6000%) classifies these hydrogels as ‘superabsorbants’ [13]. The aldehyde content of DABC was calculated from nitrogen content of the membrane. One mol of aldehyde reacts with one mol of NH2 OH·HCl and the aldehyde content can be calculated directly from the nitrogen content of the product as described above [19]. In this study, according to an elemental analysis of DABC membranes, they have composition of 48% and 11% carbon and nitrogen, respectively. Sirvio et al. reported comparable results under similar conditions. Controlled and localized drug release offers many advantages over the current delivery methods of injection or ingestion. It avoids hepatic first pass metabolism and improves patient compliance. Controlled local release systems provide the desired constant drug concentrations at the delivery site, lower systemic drug level and a reduced potential for deleterious side effect [27]. Especially, delivery systems which act as biodegradable polymers do not require removal from the patient skin at the end of treatment period. As said above, BC is a promising wound healing material due to its chemical purity, mechanical, and physical properties, additionally when oxidized using sodium metaperiodate, it becomes a stronger alternative among others with its biodegradable property. Interaction between CAP and membranes was displayed by comparing the FTIR spectra of BC, DABC, CAP, CAP-loaded BC and CAP-loaded DABC (Fig. 3). The spectrum demonstrates the typical curves of BC, DABC, and CAP. It is well known that cellulose oxidized by periodate should exhibit a characteristic FTIR band around 1740 cm−1 . Intensity of the peak increases with increasing degree of oxidation [8,26]. A peak at 1725 cm−1 due to its carbonyl group occurred in FTIR spectrum of DABC. In CAP–DABC complex, the band appearing near 1682 cm−1 was due to the carbonyl group of the drug. The stretching bond observation of nitro
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
25
Fig. 2. AFM microphotographs of (a) fibre size and (b) pore size of 3D nano-network structure of BC.
group in free ligand appears at 1561 cm−1 and 1517 cm−1 . After the interaction, the peaks shifted and they appeared at 1563 cm−1 and 1513 cm−1 , respectively. Similarly, a stretching bond of nitro group at 1518 cm−1 was observed due to the complexation between CAP and BC [28].
Fig. 3. The ATR–FTIR spectra of BC, DABC, CAP, CAP–DABC and CAP–BC.
3.2. Antimicrobial activity CAP-loaded membranes were tested for their antimicrobial activity against S. aureus, S. Pneumonie, and E. coli. The difference of the semidiameter between inhibition zone of membranes containing CAP and pure membranes was measured (Fig. 4). Both of the CAP-containing BC and CAP-containing DABC membranes exhibited antimicrobial activities against the three model bacteria. Highest antimicrobial activity was displayed on S. pneumonie, followed by S. aureus and E. coli with the use of CAP-loaded BC with an inhibition zone of 13 mm, 11 mm, and 9 mm, respectively. No inhibition zones were observed with the pure membranes against all the three model bacteria. It could be concluded that the antimicrobial effect observed with disc diffusion tests is attributed to CAP adsorbed into bacterial cellulose membrane and could be said that after UV sterilization of CAP loaded membranes, any adverse effect on CAP is not determined. Wei et al. have prepared benzalkonium chloride containing antimicrobial BCM and achieved an inhibition zone of 8.5 mm against S. aureus [10]. In growth curve studies, the absorbances at OD600 for negative controls with S. pneumonie, S. aureus, and E. coli were observed 0.650, 0.460 and 0.300, respectively. The absorbances at OD600 for CAP-loaded BCMs with S. pneumonie, S. aureus, and E. coli were observed below 0.1 (Fig. 5a). Similar results were obtained with CAP-loaded DABC as seen in Fig. 5b. It could be concluded that the cell densities decreased significantly with the use of CAPloaded BC and DABC membranes, the antimicrobial activity was
26
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
Fig. 4. Growth inhibition effect of the CAP–BC against (A) S. pneumonia (B) S. aureus (C) E. coli and CAP–DABC against (D) E. coli (E) S. aureus (F) S. pneumonia by disc diffusion test.
Fig. 5. The bacterial growth inhibition effect of (A) CAP–BC and (B) CAP–DABC against S. pneumonia, S. aureus, E. coli.
only attributed to CAP absorbed into bacterial cellulose membrane and not due to bacterial cellulose itself. 3.3. Release of chloramphenicol from BC and DABC membranes
day 7 (Fig. 7). MTT results indicate that fibroblasts adhered more strongly to the DABC compared to the BC, most probably due to dialdehyde groups on DABC. Although the positive influence of these groups might be altered by initial drug release, the initial fibroblast attachment was not less than that on BC membranes
For an antimicrobial material, the permanence of the antimicrobial activity is important in practical applications, which need a stable and prolonged release of antimicrobial agents. In this study, the drug content of BC and DABC membranes were 5 mg/cm2 and 0.1 mg/cm2 , respectively. An initial burst release effect was not observed (Fig. 6). BC and DABC membranes released 99% and 99.5% of the loaded drug, respectively, within 24 h. Wei et al. have reported the release of 66% of benzalkonium chloride in 24 h and highlighted that the rate of drug release depends on the water content of the swollen hydrogel [10]. The highly porous structure easily permits the loading of drugs into the gel matrix and subsequent drug release at a rate depending on the diffusion coefficient of the small molecule or macromolecule through the gel network [29]. 3.4. Cell attachments on CAP–BC and CAP–DABC membranes Optical density (OD) values of day 1 confirm the presence of cells at all membranes, suggesting a successful loading of cells. OD values for CAP-loaded containing DABC increased at day 4 whereas OD values for CAP-loaded BC decreased. At day 7, the cell population on CAP–BC distinctly increased. However, cell population on CAP–DABC was still higher than the cell population on CAP–BC at
Fig. 6. Chloramphenicol releasing behaviour of CAP–BC and CAP–DABC membranes.
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
27
developed CAP-loaded dialdehyde cellulose membrane may be superior to its counterparts in wound dressing due to its antimicrobial property, biodegradablity, and biocompatibility. Acknowledgements We thank to H. Bıyık and E. C¸oban from Adnan Menderes University for providing us A. xylinum (ATCC 10245). References
Fig. 7. Optical density values for fibroblast cell line L929 seeded on control BC (BC-Ctr), drug loaded BC (BC-DL), control DABC (DABC-Ctr), and drug loaded DABC (DABC-DL).
within first 24 h. After 4 and 7 days, proliferation was even much higher in DABC membranes. The MTT data describe the viability of fibroblasts growing on both of the drug loaded membranes. The data is comparable since the same number of cells was added to all samples. Bacterial cellulose (BC)-based biomaterials on medical device platforms have gained significant interest for tissue-engineered scaffolds or engraftment materials in regenerative medicine. In particular, BC has an ultrafine and highly pure nanofibril network structure and can be used as an efficient wound-healing platform since cell migration into a wound site is strongly mediated by the structural properties of the extracellular matrix [30]. Popadic et al. reported that CAP inhibits the in vitro proliferation of keratinocytes through mechanisms involving induction of apoptosis [31]. However, in this study we did not encounter such a problem with fibroblasts. 4. Conclusion In this research, extracellularly synthesized cellulose membranes by A. xylinum were functionalized by loading chloramphenicol, a type of broad spectrum antibiotic. The chloramphenicol loading capacities of bacterial cellulose and periodate oxidized bacterial cellulose (DABC) were quite different just like their water uptake capacities. Both BC and DABC membranes containing chloramphenicol resisted effectively against S. aureus, S. Pneumonie, and E. coli. However, the adhesion and proliferation of the fibroblast cell line L929 on CAP-loaded DABC membranes were noticeably higher than that on CAP-loaded BC membranes. The newly
[1] S. Zhong, Y. Zhang, C. Lim, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2 (2010) 510–525. [2] O.-a. Saibuatong, M. Phisalaphong, Carbohydr. Polym. 79 (2010) 455–460. [3] P. Ross, R. Mayer, M. Benziman, Microbiol. Rev. 55 (1991) 35–58. [4] R.M. Brown, J. Willison, C.L. Richardson, Proc. Nat. Acad. Sci. 73 (1976) 4565–4569. [5] W. Czaja, D. Romanovicz, R.M. Brown, Cellulose 11 (2004) 403–411. [6] S.T. Schrecker, P.A. Gostomski, Biotechnol. Lett. 27 (2005) 1435–1438. [7] F. Lina, Z. Yue, Z. Jinand, Y. Guang, Bacterial celullose for skin repair materials, in: Biomedical Engineering – Frontiers and Challenges, 2011, pp. 249–274. [8] U.J. Kim, S. Kuga, M. Wada, T. Okano, T. Kondo, Biomacromolecules 1 (2000) 488–492. [9] H.L. Ramirez, R. Cao, A. Fragoso, J.J. Torres-Labandeira, A. Dominguez, E.H. Schacht, M. Banos, R. Villalonga, Macromol. Biosci. 6 (2006) 555–561. [10] B. Wei, G.A. Yang, F. Hong, Carbohydr. Polym. 84 (2011) 533–538. [11] C.Y. Chang, J. Peng, L.N. Zhang, D.W. Pang, J. Mater. Chem. 19 (2009) 7771–7776. [12] C. Vinatier, O. Gauthier, A. Fatimi, C. Merceron, M. Masson, A. Moreau, F. Moreau, B. Fellah, P. Weiss, J. Guicheux, Biotechnol. Bioeng. 102 (2009) 1259–1267. [13] G. Buhus, M. Popa, J. Desbrieres, J. Bioact. Compat. Polym. 24 (2009) 525–545. [14] E. Tamahkar, C. Babac, T. Kutsal, E. Piskin, A. Denizli, Process Biochem. 45 (2010) 1713–1719. [15] R. Guthrie, M. Wolfram, Advances in Carbohydrate Chemistry, Academic, New York, 1961. [16] A.J. Varma, V.B. Chavan, Polym. Degrad. Stab. 49 (1995) 245–250. [17] Y. Liu, X. Cao, M. Luo, Z. Le, W. Xu, J. Colloid Interface Sci. 329 (2009) 244–252. [18] P. RoyChowdhury, S. Klemuk, I. Titze, V. Kumar, J. Biomed. Mater. Res. Part A 88A (2009) 680–688. [19] J. Sirvio, U. Hyvakko, H. Liimatainen, J. Niinimaki, O. Hormi, Carbohydr. Polym. 83 (2011) 1293–1297. [20] M.C.I.M. Amin, N. Ahmad, N. Halib, I. Ahmad, Carbohydr. Polym. 88 (2012) 465–473. [21] J.R. Koup, B. Brodsky, A. Lau, T.R. Beam, Antimicrob. Agents Chemother. 14 (1978) 439–443. [22] W.H. Tang, S.R. Jia, Y.Y. Jia, H.J. Yang, World J. Microbiol. Biotechnol. 26 (2010) 125–131. [23] H. Jia, Y.Y. Jia, J. Wang, Y. Hu, Y. Zhang, S.R. Jia, Proceedings of the 2009 Second International Conference on Biomedical Engineering and Informatics 1–4 (2009) 1138–1142. [24] D. Norouzian, A. Farhangi, S. Tolooei, Z. Saffari, M. Mehrabi, M. Chiani, S. Ghassemi, M. Farahnak, A. Akbarzadeh, Pak. J. Biol. Sci. 14 (2011) 780. [25] M. Ul-Islam, T. Khan, J.K. Park, Carbohydr. Polym. 88 (2012) 596–603. [26] Q.G. Fan, D.M. Lewis, K.N. Tapley, J. Appl. Polym. Sci. 82 (2001) 1195–1202. [27] A. Stoica-Guzun, M. Stroescu, F. Tache, T. Zaharescu, E. Grosu, Nucl. Instrum. Methods Phys. Res., Sect. B 265 (2007) 434–438. [28] W. Goh, A. Rosma, B. Kaur, A. Fazilah, A. Karim, R. Bhat, Int. Food Res. J. 19 (2012) 153–158. [29] T.R. Hoare, D.S. Kohane, Polymer 49 (2008) 1993–2007. [30] J. Kim, S.W. Kim, S. Park, K.T. Lim, H. Seonwoo, Y. Kim, B.H. Hong, Y.H. Choung, J.H. Chung, Adv. Healthcare Mater. 2 (11) (2013) 1525–1531. [31] S. Popadic, D. Popadic, Z. Ramic, M.M. Stojkovic, V. Trajkovic, M. Milinkovic, L. Medenica, Cell Biol. Toxicol. 22 (2006) 371–379.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Development of biodegradable antibacterial cellulose based hydrogel membranes for wound healing Nelisa Türko˘glu Lac¸in ∗ Science and Technology Application and Research Center, Yıldız Technical University, 34349 Istanbul, Turkey
a r t i c l e
i n f o
Article history: Received 13 January 2014 Received in revised form 26 February 2014 Accepted 3 March 2014 Available online 12 March 2014 Keywords: Wound healing 2,3 Dialdehyde bacterial cellulose Antibacterial activity
a b s t r a c t Cellulose-based hydrogels have wide applications in tissue engineering and controlled delivery systems. In this study, chloramphenicol (CAP) loaded 2,3 dialdehyde cellulose (DABC) hydrogel membranes were prepared, characterized and their antibacterial efficacy was evaluated. Bacterial cellulose (BC) secreted by Acetobacter xylinum was modified to become DABC by oxidation via the sodium metaperiodate method. CAP–BC and CAP–DABC interactions were illustrated via ATR–FTIR analysis. Water retention capacity of BC and DABC membranes were determined as 65.6 ± 1.6% and 5.3 ± 0.3%, respectively. CAP release profiles were determined via HPLC analysis. The drug-loading capacities of BC and DABC membranes were 5 mg/cm2 and 0.1 mg/cm2 , respectively. Membranes released 99–99.5% of the contained CAP within 24 h and an initial burst release effect was not observed. In vitro antibacterial tests of BC and DABC, both CAPloaded, demonstrated their ability to inhibit bacterial growth for a prolonged duration. Antimicrobial effect against bacteria was still prevalent after 3 days of incubation period with disc diffusion tests. The MTT test results reveal that fibroblast adhesion and proliferation on CAP-loaded DABC membranes were noticeably higher than CAP-loaded BC membrane. This newly developed drug containing DABC membranes seem to be highly suitable for wound healing due to its unique properties of biodegradability, biocompatibility, and antimicrobial effectiveness. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Three-dimensional wound dressing substances must cover the wound, serve as a physical barrier against external infection, and provide cellular support. A successful tissue scaffold should possess the appropriate physical and mechanical characteristics, in addition to possessing an appropriate surface chemistry to facilitate cellular attachment, proliferation, and differentiation [1]. Cellulose, a linear polysaccharide, is one of the most abundant organic materials in nature with a variety of useful applications. A few bacterial species, taxonomically closely related to the genus Acetobacter xylinum (A. Xylinum), produce and extracellularly secrete cellulose in the form of fibre called BC [2]. It is a natural polymer with potential due to its unique structural and mechanical properties [3]. A. xylinum is a simple gram-negative bacterium that has an ability to synthesize high-quality cellulose composed of twisting ribbons of microfibrillar bundles [4]. The thick, gelatinous membrane formed in static culture conditions as a result of these processes is characterized by a 3D structure consisting of an ultrafine network of
∗ Tel.: +90 532 2257368; fax: +90 212 3832974. E-mail address: [email protected] http://dx.doi.org/10.1016/j.ijbiomac.2014.03.003 0141-8130/© 2014 Elsevier B.V. All rights reserved.
cellulose nanofibres (3–8 nm), which are highly uniaxially oriented [5]. BC is a natural hydrogel. Hydrogels are defined as threedimensional polymer networks swollen by large amounts of a solvent. BC is a hydrogel that can take water up to 99% its own weight, mainly due to its amorphous structure [6]. Owing to its unique nano-scaled three-dimensional network structure, BC has high water retention, high mechanical strength, and outstanding biocompatibility, which enable it to serve as a natural scaffolding material for the regeneration of a wide variety of tissues [7]. Dialdehyde cellulose (DABC) is a cellulose derivate produced by regioselective oxidation of cellulose with the use of periodate as an oxidation agent. It is biodegradable and biocompatible and has a large potential to be used in many applications [8,9]. Wei et al. have prepared a type of new functional dry BC film containing benzalkonium chloride as a potential antimicrobial wound dressing material and reported that it has strong antibacterial properties that especially resist both Staphylococcus aureus and Bacillus subtilis (gram-positive bacteria). Also, a stable and lasting release of the antimicrobial agent for at least 24 h was reported [10]. There are also commercial cellulose-based wound healing systems such as Biofill, Gengiflex, and XCell. Cellulose-based hydrogels are biocompatible, have low production costs, and are non-toxic. Therefore,
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
cellulose-based hydrogels have wide applications in tissue engineering and controllable delivery systems [11,12]. The high water absorption ability of the antimicrobial BC dry film is crucial for wound dressing to absorb blood and tissue fluid on acute traumas and would promote wound healing [10]. CAP was chosen as a model biologically active agent because it presents a wide antibacterial spectrum with a bactericide activity on gram-negative and grampositive bacteria [13]. The aim of this study was to develop CAP loaded membranes as antimicrobial wound healing materials, investigate the antimicrobial activities against model gram-positive and gram-negative bacteria (S. aureus, Streptococcus pneumoniae and Escherichia coli), and to observe the proliferation and attachment of fibroblasts on drug-loaded membranes. 2. Experimental procedure 2.1. Materials E. coli (ATCC 25293), S. aureus (ATCC 25925) and S. pneumoniae (ATCC6301) were purchased from Institute of Microbiology Chinese Academy of Science. 2.2. Production of BC and preparation of DABC membranes The production of bacterial cellulose membranes (BCMs) was achieved by growing A. xylinum (ATCC 10245) in Hestrin–Schramm medium, pH 5.1(adjusted by 1 M HCl). The medium containing 20 g/L glucose, 10 g/L bactopeptone, 10 g/L yeast extract, 4 mM KH2 PO4 , and 6 mM K2 HPO4 was used to produce cellulose pellicles in static culture. The inoculum was prepared by growing A. xylinum at 30 ◦ C using a rotary shaker for 3 days. The BC nanofibre formation was allowed to occur in a period of 7 days after inoculating the subculture in the proportion of 1:10 in petri dishes statically. The harvested BC membranes were washed with distilled water to remove s medium components and then incubated in 1 M NaOH solution at 80 ◦ C for 2 h to eliminate attached cells and other impurity. After that, the BC membranes were further purified to remove other residues by distilled water washing until the pH of the washing liquid was neutral. Finally, the bacterial cellulose membranes were cut into 1 cm2 size [14]. The morphology and microstructure of membranes were characterized by scanning electron microscopy (FE-SEM, Zeiss supra 55) and AFM (Park systems/XE-100E advance scanning probe microscope). Prior to the FE-SEM observation, all samples were sputter coated with a thin layer of platinum to avoid electrical charging. Periodate oxidized cellulose is often referred to as dialdehyde cellulose (DABC) [2]. Oxidation of cellulose using sodium metaperiodate has been extensively investigated in the literature and it is found to lead to selective cleavage at the C-2 and C-3 vicinal hydroxyl groups to yield a product with 2,3-dialdehyde units along the polymer chain [15,16]. The dialdehyde groups also serve as reactive chemical anchors for further reactions, conducive for the chemical derivatization of cellulose. The presence of dialdehyde groups in bacterial cellulose also improves biodegradability and they are discussed in relation to tissue-scaffold engineering [17,18]. Periodate oxidate BCMs were prepared by placing the membranes in DI water, adding 1.3 times the weight sodium metaperiodate and gently stirring at 55 ◦ C in dark for 4 h. After the excess periodate was decomposed with ethylene glycol, the DABC membranes were washed by DI. 2.2.1. Determination of dialdehyde content Determination of the aldehyde content of DABC was based on the oxime reaction between aldehyde group and NH2 OH·HCl. The periodate oxidized cellulose, which never underwent drying procedures, has been placed in a 250 mL beaker containing 1.39 g of
23
NH2 OH·HCl dissolved in 100 mL of 0.1 M acetate buffer (pH = 4.5). The beaker was covered with a thin rubber foil and the mixture was stirred for 48 h at room temperature with a magnetic stirrer. The product was filtrated and washed with 600 mL of deionized water after which it was dried in a freeze-dryer. For the determination of the aldehyde content of DABC, five membranes were periodate oxidized under the same conditions and the nitrogen content of the oxime derivate of DABCs was determined by using a Leco truSpec series elemental analyzer [19]. The results are expressed as the average values of five samples. 2.2.2. Water retention capacity of hydrogels In order to dry the membranes without harming the 3D nanonetwork structure, the porous membranes were frozen overnight in a freeze dryer (Christ Alpha 2-4 LD) at −80 ◦ C for 24 h. To determine the water retention capacity of membranes, the freeze dried 1 cm2 BC and DABC membranes were immersed in phosphate-buffered saline (PBS) at room temperature until equilibration. After that, the membranes were removed from the PBS and excess PBS at the surface of the membranes was blotted out with Kim wipes paper. The weights of the freeze dried membranes were measured, and then the membranes were swollen until no further weight change was observed and the procedure was repeated. The results are expressed as the average values of five samples. The water retention of membranes was calculated with the following formula: Wh : Weight of hydrate membrane, Wd : weight of dry membrane: WAC(%) = (Wh − Wd )/Wd × 100 [2]. 2.3. Preparation of antimicrobial membranes CAP was loaded into the hydrogels using the swelling-diffusion method. The interaction of CAP with the membranes was investigated by attenuated total reflectance—Fourier transform infrared (Perkin-Elmer). The 1 cm2 sized freeze-dried membranes were immersed in CAP stock solution (0.05 g/mL) (CAPstock sol. ) for 8 h. Membranes were removed from the CAP stock solution (CAPresidual ) and immersed in 1 mL distilled water for 10 s to remove the nonabsorbed CAP (CAPwashing sol. ). The amount of CAP loaded into membranes was calculated with the following formula:
CAPmembrane = (CAPstock sol. ) − CAPresidual + CAPwashing sol.
Afterwards, CAP-loaded membranes were placed in PBS (10 mL) and incubated at 37 ◦ C, 50 rpm for 24 h. CAP amount released from membranes was determined by high performance liquid chromatography (HPLC) to verify the calculation above. The same treatment was applied to each membrane and another freeze–drying step was administered. Finally, 30 min of UV sterilization was performed. 2.4. Determination of antibacterial activity The antimicrobial activities were investigated against E. coli, S. pneumoniae, and S. aureus using two different methods explained below. 2.4.1. Growth curve method Single colonies of E. coli, S. Aureus, and S. pneumoniae were grown on nutrient agar. Cell suspensions for inoculum were arranged according to standard no 0.5 as 1–1.5 × 108 CFU/mL. The positive control was the bacteria in nutrient broth, the negative control the culture with the pure membrane in nutrient broth, and the blank was the test tube only containing nutrient broth. Cultures were incubated at 37 ◦ C for 24 h. Samples drawn from the system every 2 h were analyzed spectrophotometrically by measuring the
24
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
absorbance at 600 nm. Triplicate experiments were carried out for each bacterium. 2.4.2. Disc diffusion method E. coli, S. aureus, and S. pneumoniae were spread on nutrient agar. Then, the CAP-loaded BCMs and DABC membranes were placed in the nutrient agar in petri dishes. BC and DABC membranes without the drug were the control group. The agar plates were then incubated at 37 ◦ C for 48 h. After incubation, the semi-diameters of the inhibition zone were measured. Each sample was tested in triplicate, using different plates. 2.5. The release of chloramphenicol from BC and DABC membranes In vitro CAP release profiles were observed by immersing CAPloaded hydrogel samples into 10 mL of PBS glass flask covered by aluminium foil. CAP-loaded membranes were incubated at 37 ◦ C, 50 rpm for 24 h. Samples removed from the release medium every 2 h were analyzed by HPLC [20]. HPLC experiments were conducted using an Agilent HPLC System. Separations of CAP were achieved using a 250 mm ACE 5 C18 column. Mobile phase compositions were [A] ACN, [B] H2 O ([A] 70% and [B] 30%, respectively). The flow rate was 1 mL/min, column temperature was 40 ◦ C and the analysis was performed by DAD detector at 278 nm [21]. Triplicate experiments were performed for each membrane. 2.6. Biocompatibility testing In order to evaluate the biocompatibility of the drug loaded membranes, fibroblast cell line L929 (ECACC, European Collection of Cell Cultures) was used. The cells were cultured in Dulbecco’s modified Eagle’s medium low glucose (Sigma-Aldrich) supplemented with 10% fetal bovine serum and 1% antibiotics/antimicotics (Sigma-Aldrich) until they reached the confluence. They were then trypsinized and seeded on UV-sterilized samples using the density of 2 × 104 cells/membrane. Cell proliferation was assessed by measuring the cellular enzyme activity using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. This assay consists of the bioreduction of the substrate by mitochondrial enzyme succinate dehydrogenase into a purple colour salt insoluble in water. According to the standard procedure, membranes were washed with PBS buffer to remove unattached cells and were placed into new well-plates after 1, 3, and 7 days of incubation. MTT reagent was added to each well and the plates were incubated for 4 h at 37 ◦ C in a humidified atmosphere containing 5% of CO2 . After the reaction occurred the intracellular formazan salt was solubilized using isopropanol. The absorbance of solubilized formazan was measured at 570 nm with a spectrophotometer (Biotek Instruments). The results are expressed as the average absorbance of triplicate samples. 3. Results and discussion 3.1. Characterization of BC, DABC and CAP-loaded BC/DABC membranes The 3D nano-network structure of BC was observed by AFM and SEM microphotographs (Figs. 1 and 2). Bacterial cellulose microfibril sizes range between 60 and 100 nm [22]. In this study, the width and pore size of cellulose nanofibres were identified to be between 80–100 nm and 60–90 nm, respectively. In other studies, the width of cellulose ribbons were observed to be between 10 nm and 100 nm [23,24]. Water retention capacity is the most important property directly involved in the biomedical applications of BCMs
Fig. 1. SEM microphotographs of 3D nano-network structure of BC.
as a dressing material. The proper moisture content of a dressing material accelerates the wound healing process and protects it against contamination [25]. Additionally, the high water absorption ability of the antimicrobial BCM is an important feature to promote wound healing for wound dressings [26]. Data shows good water-binding ability with a swelling ratio of 65.6% in PBS for BCM. The swelling ratio of DABC was determined to be 5.3%, which was significantly lower than wac of BC. BCM prepared in our work could hold water at least 65.6 times its own weight within 24 h. Kinetics of the swelling process is typical of hydrogels. The swelling degree limitations are specific to each hydrogel. Similarly, the time necessary to reach this threshold depend on the reaction conditions for the hydrogel. The greater maximal swelling degree (6000%) classifies these hydrogels as ‘superabsorbants’ [13]. The aldehyde content of DABC was calculated from nitrogen content of the membrane. One mol of aldehyde reacts with one mol of NH2 OH·HCl and the aldehyde content can be calculated directly from the nitrogen content of the product as described above [19]. In this study, according to an elemental analysis of DABC membranes, they have composition of 48% and 11% carbon and nitrogen, respectively. Sirvio et al. reported comparable results under similar conditions. Controlled and localized drug release offers many advantages over the current delivery methods of injection or ingestion. It avoids hepatic first pass metabolism and improves patient compliance. Controlled local release systems provide the desired constant drug concentrations at the delivery site, lower systemic drug level and a reduced potential for deleterious side effect [27]. Especially, delivery systems which act as biodegradable polymers do not require removal from the patient skin at the end of treatment period. As said above, BC is a promising wound healing material due to its chemical purity, mechanical, and physical properties, additionally when oxidized using sodium metaperiodate, it becomes a stronger alternative among others with its biodegradable property. Interaction between CAP and membranes was displayed by comparing the FTIR spectra of BC, DABC, CAP, CAP-loaded BC and CAP-loaded DABC (Fig. 3). The spectrum demonstrates the typical curves of BC, DABC, and CAP. It is well known that cellulose oxidized by periodate should exhibit a characteristic FTIR band around 1740 cm−1 . Intensity of the peak increases with increasing degree of oxidation [8,26]. A peak at 1725 cm−1 due to its carbonyl group occurred in FTIR spectrum of DABC. In CAP–DABC complex, the band appearing near 1682 cm−1 was due to the carbonyl group of the drug. The stretching bond observation of nitro
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
25
Fig. 2. AFM microphotographs of (a) fibre size and (b) pore size of 3D nano-network structure of BC.
group in free ligand appears at 1561 cm−1 and 1517 cm−1 . After the interaction, the peaks shifted and they appeared at 1563 cm−1 and 1513 cm−1 , respectively. Similarly, a stretching bond of nitro group at 1518 cm−1 was observed due to the complexation between CAP and BC [28].
Fig. 3. The ATR–FTIR spectra of BC, DABC, CAP, CAP–DABC and CAP–BC.
3.2. Antimicrobial activity CAP-loaded membranes were tested for their antimicrobial activity against S. aureus, S. Pneumonie, and E. coli. The difference of the semidiameter between inhibition zone of membranes containing CAP and pure membranes was measured (Fig. 4). Both of the CAP-containing BC and CAP-containing DABC membranes exhibited antimicrobial activities against the three model bacteria. Highest antimicrobial activity was displayed on S. pneumonie, followed by S. aureus and E. coli with the use of CAP-loaded BC with an inhibition zone of 13 mm, 11 mm, and 9 mm, respectively. No inhibition zones were observed with the pure membranes against all the three model bacteria. It could be concluded that the antimicrobial effect observed with disc diffusion tests is attributed to CAP adsorbed into bacterial cellulose membrane and could be said that after UV sterilization of CAP loaded membranes, any adverse effect on CAP is not determined. Wei et al. have prepared benzalkonium chloride containing antimicrobial BCM and achieved an inhibition zone of 8.5 mm against S. aureus [10]. In growth curve studies, the absorbances at OD600 for negative controls with S. pneumonie, S. aureus, and E. coli were observed 0.650, 0.460 and 0.300, respectively. The absorbances at OD600 for CAP-loaded BCMs with S. pneumonie, S. aureus, and E. coli were observed below 0.1 (Fig. 5a). Similar results were obtained with CAP-loaded DABC as seen in Fig. 5b. It could be concluded that the cell densities decreased significantly with the use of CAPloaded BC and DABC membranes, the antimicrobial activity was
26
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
Fig. 4. Growth inhibition effect of the CAP–BC against (A) S. pneumonia (B) S. aureus (C) E. coli and CAP–DABC against (D) E. coli (E) S. aureus (F) S. pneumonia by disc diffusion test.
Fig. 5. The bacterial growth inhibition effect of (A) CAP–BC and (B) CAP–DABC against S. pneumonia, S. aureus, E. coli.
only attributed to CAP absorbed into bacterial cellulose membrane and not due to bacterial cellulose itself. 3.3. Release of chloramphenicol from BC and DABC membranes
day 7 (Fig. 7). MTT results indicate that fibroblasts adhered more strongly to the DABC compared to the BC, most probably due to dialdehyde groups on DABC. Although the positive influence of these groups might be altered by initial drug release, the initial fibroblast attachment was not less than that on BC membranes
For an antimicrobial material, the permanence of the antimicrobial activity is important in practical applications, which need a stable and prolonged release of antimicrobial agents. In this study, the drug content of BC and DABC membranes were 5 mg/cm2 and 0.1 mg/cm2 , respectively. An initial burst release effect was not observed (Fig. 6). BC and DABC membranes released 99% and 99.5% of the loaded drug, respectively, within 24 h. Wei et al. have reported the release of 66% of benzalkonium chloride in 24 h and highlighted that the rate of drug release depends on the water content of the swollen hydrogel [10]. The highly porous structure easily permits the loading of drugs into the gel matrix and subsequent drug release at a rate depending on the diffusion coefficient of the small molecule or macromolecule through the gel network [29]. 3.4. Cell attachments on CAP–BC and CAP–DABC membranes Optical density (OD) values of day 1 confirm the presence of cells at all membranes, suggesting a successful loading of cells. OD values for CAP-loaded containing DABC increased at day 4 whereas OD values for CAP-loaded BC decreased. At day 7, the cell population on CAP–BC distinctly increased. However, cell population on CAP–DABC was still higher than the cell population on CAP–BC at
Fig. 6. Chloramphenicol releasing behaviour of CAP–BC and CAP–DABC membranes.
N.T. Lac¸in / International Journal of Biological Macromolecules 67 (2014) 22–27
27
developed CAP-loaded dialdehyde cellulose membrane may be superior to its counterparts in wound dressing due to its antimicrobial property, biodegradablity, and biocompatibility. Acknowledgements We thank to H. Bıyık and E. C¸oban from Adnan Menderes University for providing us A. xylinum (ATCC 10245). References
Fig. 7. Optical density values for fibroblast cell line L929 seeded on control BC (BC-Ctr), drug loaded BC (BC-DL), control DABC (DABC-Ctr), and drug loaded DABC (DABC-DL).
within first 24 h. After 4 and 7 days, proliferation was even much higher in DABC membranes. The MTT data describe the viability of fibroblasts growing on both of the drug loaded membranes. The data is comparable since the same number of cells was added to all samples. Bacterial cellulose (BC)-based biomaterials on medical device platforms have gained significant interest for tissue-engineered scaffolds or engraftment materials in regenerative medicine. In particular, BC has an ultrafine and highly pure nanofibril network structure and can be used as an efficient wound-healing platform since cell migration into a wound site is strongly mediated by the structural properties of the extracellular matrix [30]. Popadic et al. reported that CAP inhibits the in vitro proliferation of keratinocytes through mechanisms involving induction of apoptosis [31]. However, in this study we did not encounter such a problem with fibroblasts. 4. Conclusion In this research, extracellularly synthesized cellulose membranes by A. xylinum were functionalized by loading chloramphenicol, a type of broad spectrum antibiotic. The chloramphenicol loading capacities of bacterial cellulose and periodate oxidized bacterial cellulose (DABC) were quite different just like their water uptake capacities. Both BC and DABC membranes containing chloramphenicol resisted effectively against S. aureus, S. Pneumonie, and E. coli. However, the adhesion and proliferation of the fibroblast cell line L929 on CAP-loaded DABC membranes were noticeably higher than that on CAP-loaded BC membranes. The newly
[1] S. Zhong, Y. Zhang, C. Lim, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2 (2010) 510–525. [2] O.-a. Saibuatong, M. Phisalaphong, Carbohydr. Polym. 79 (2010) 455–460. [3] P. Ross, R. Mayer, M. Benziman, Microbiol. Rev. 55 (1991) 35–58. [4] R.M. Brown, J. Willison, C.L. Richardson, Proc. Nat. Acad. Sci. 73 (1976) 4565–4569. [5] W. Czaja, D. Romanovicz, R.M. Brown, Cellulose 11 (2004) 403–411. [6] S.T. Schrecker, P.A. Gostomski, Biotechnol. Lett. 27 (2005) 1435–1438. [7] F. Lina, Z. Yue, Z. Jinand, Y. Guang, Bacterial celullose for skin repair materials, in: Biomedical Engineering – Frontiers and Challenges, 2011, pp. 249–274. [8] U.J. Kim, S. Kuga, M. Wada, T. Okano, T. Kondo, Biomacromolecules 1 (2000) 488–492. [9] H.L. Ramirez, R. Cao, A. Fragoso, J.J. Torres-Labandeira, A. Dominguez, E.H. Schacht, M. Banos, R. Villalonga, Macromol. Biosci. 6 (2006) 555–561. [10] B. Wei, G.A. Yang, F. Hong, Carbohydr. Polym. 84 (2011) 533–538. [11] C.Y. Chang, J. Peng, L.N. Zhang, D.W. Pang, J. Mater. Chem. 19 (2009) 7771–7776. [12] C. Vinatier, O. Gauthier, A. Fatimi, C. Merceron, M. Masson, A. Moreau, F. Moreau, B. Fellah, P. Weiss, J. Guicheux, Biotechnol. Bioeng. 102 (2009) 1259–1267. [13] G. Buhus, M. Popa, J. Desbrieres, J. Bioact. Compat. Polym. 24 (2009) 525–545. [14] E. Tamahkar, C. Babac, T. Kutsal, E. Piskin, A. Denizli, Process Biochem. 45 (2010) 1713–1719. [15] R. Guthrie, M. Wolfram, Advances in Carbohydrate Chemistry, Academic, New York, 1961. [16] A.J. Varma, V.B. Chavan, Polym. Degrad. Stab. 49 (1995) 245–250. [17] Y. Liu, X. Cao, M. Luo, Z. Le, W. Xu, J. Colloid Interface Sci. 329 (2009) 244–252. [18] P. RoyChowdhury, S. Klemuk, I. Titze, V. Kumar, J. Biomed. Mater. Res. Part A 88A (2009) 680–688. [19] J. Sirvio, U. Hyvakko, H. Liimatainen, J. Niinimaki, O. Hormi, Carbohydr. Polym. 83 (2011) 1293–1297. [20] M.C.I.M. Amin, N. Ahmad, N. Halib, I. Ahmad, Carbohydr. Polym. 88 (2012) 465–473. [21] J.R. Koup, B. Brodsky, A. Lau, T.R. Beam, Antimicrob. Agents Chemother. 14 (1978) 439–443. [22] W.H. Tang, S.R. Jia, Y.Y. Jia, H.J. Yang, World J. Microbiol. Biotechnol. 26 (2010) 125–131. [23] H. Jia, Y.Y. Jia, J. Wang, Y. Hu, Y. Zhang, S.R. Jia, Proceedings of the 2009 Second International Conference on Biomedical Engineering and Informatics 1–4 (2009) 1138–1142. [24] D. Norouzian, A. Farhangi, S. Tolooei, Z. Saffari, M. Mehrabi, M. Chiani, S. Ghassemi, M. Farahnak, A. Akbarzadeh, Pak. J. Biol. Sci. 14 (2011) 780. [25] M. Ul-Islam, T. Khan, J.K. Park, Carbohydr. Polym. 88 (2012) 596–603. [26] Q.G. Fan, D.M. Lewis, K.N. Tapley, J. Appl. Polym. Sci. 82 (2001) 1195–1202. [27] A. Stoica-Guzun, M. Stroescu, F. Tache, T. Zaharescu, E. Grosu, Nucl. Instrum. Methods Phys. Res., Sect. B 265 (2007) 434–438. [28] W. Goh, A. Rosma, B. Kaur, A. Fazilah, A. Karim, R. Bhat, Int. Food Res. J. 19 (2012) 153–158. [29] T.R. Hoare, D.S. Kohane, Polymer 49 (2008) 1993–2007. [30] J. Kim, S.W. Kim, S. Park, K.T. Lim, H. Seonwoo, Y. Kim, B.H. Hong, Y.H. Choung, J.H. Chung, Adv. Healthcare Mater. 2 (11) (2013) 1525–1531. [31] S. Popadic, D. Popadic, Z. Ramic, M.M. Stojkovic, V. Trajkovic, M. Milinkovic, L. Medenica, Cell Biol. Toxicol. 22 (2006) 371–379.