International Journal of Pharmaceutics 471 (2014) 45–55
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Active wound dressings based on bacterial nanocellulose as drug delivery system for octenidine Sebastian Moritz a , Cornelia Wiegand b , Falko Wesarg c , Nadine Hessler a , Frank A. Müller c,d, Dana Kralisch a,d , Uta-Christina Hipler b , Dagmar Fischer a,d, * a
Department of Pharmaceutical Technology, Friedrich-Schiller-University Jena, Otto-Schott-Str. 41, Jena 07745, Germany Department of Dermatology, University Medical Center Jena, Erfurter Str. 35, Jena 07743, Germany c Otto-Schott-Institute of Materials Research (OSIM), Friedrich-Schiller-University Jena, Loebdergraben 22, Jena 07743, Germany d Jena Center for Soft Matter (JCSM), Friedrich-Schiller-University Jena, Humboldtstraße 10, Jena 07743, Germany b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 February 2014 Received in revised form 25 April 2014 Accepted 29 April 2014 Available online 2 May 2014
Although bacterial nanocellulose (BNC) may serve as an ideal wound dressing, it exhibits no antibacterial properties by itself. Therefore, in the present study BNC was functionalized with the antiseptic drug octenidine. Drug loading and release, mechanical characteristics, biocompatibility, and antimicrobial efficacy were investigated. Octenidine release was based on diffusion and swelling according to the Ritger–Peppas equation and characterized by a time dependent biphasic release profile, with a rapid release in the first 8 h, followed by a slower release rate up to 96 h. The comparison between lab-scale and up-scale BNC identified thickness, water content, and the surface area to volume ratio as parameters which have an impact on the control of the release characteristics. Compression and tensile strength remained unchanged upon incorporation of octenidine in BNC. In biological assays, drug-loaded BNC demonstrated high biocompatibility in human keratinocytes and antimicrobial activity against Staphylococcus aureus. In a long-term storage test, the octenidine loaded in BNC was found to be stable, releasable, and biologically active over a period of 6 months without changes. In conclusion, octenidine loaded BNC presents a ready-to-use wound dressing for the treatment of infected wounds that can be stored over 6 months without losing its antibacterial activity. ã 2014 Elsevier B.V. All rights reserved.
PubChem classification: Octenidine dihydrochloride (PubChem CID: 51166) Keywords: Nanocellulose Wound dressing Octenidine Antiseptic Hydrogel
1. Introduction The increasing resistance of microorganisms against antibiotics in the treatment of acute and chronic colonized wounds has led to the renaissance of antiseptic drugs for local application in the prophylaxis and the treatment of wound infections. Octenidine dihydrochloride (octenidine) was introduced for skin, mucous membrane, and wound antisepsis more than two decades ago (for review, see Hübner et al., 2010). Due to its cationic
Abbreviations: BNC, Bacterial nanocellulose; MRSA, Methicillin-resistant Staphylococcus aureus; PHMB, Polyhexamethylene biguanide (polyhexanide); LS, Lab-scale; US, Up-scale; HoLiR, Horizontal lift reactor; SEM, Scanning electron microscopy; MLN, Microplate laser nephelometry; IC50, Half maximal inhibitory concentration; AUC, Area under the curve; LC50, Half maximal lethal concentration; SAV, Surface area to volume. * Corresponding author at: Department of Pharmaceutical Technology, FriedrichSchiller-University, Otto-Schott-Str. 41, 07745 Jena, Germany. Tel.: +49 3641 949941/949940; fax: +49 3641 949942. E-mail address: dagmar.fi[email protected] (D. Fischer). http://dx.doi.org/10.1016/j.ijpharm.2014.04.062 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
charge, it exhibits strong interactions with negatively charged components of microbial cell walls and membranes, leading to an inhibition of vital cell functions and a broad antimicrobial activity against Gram-positive and Gram-negative germs, plaque-forming bacteria, and fungi (Harke, 1989; Sedlock and Bailey, 1985; Slee and O‘Connor, 1983). Furthermore, octenidine is effective against Methicillin-resistant Staphylococcus aureus (MRSA) (Al-Doori et al., 2007) and exhibits a moderate virucidal effectivity against enveloped viruses (Hübner et al., 2010). Its activity was not found to be compromised by interfering substances like albumin or mucin (Pitten et al., 2003). A vast amount of knowledge on efficacy, tolerance, and safety has been collected from preclinical cell culture and animal studies as well as clinical trials (Hübner et al., 2010; Koburger et al., 2010; Vanscheidt et al., 2012). The following properties support the rationale for the selection of octenidine in this study: octenidine was found to be superior to polyhexanide (polyhexamethylene biguanide, PHMB), the agent of first choice for chronically poorly healing wounds and burns in terms of its therapeutic spectrum (Kramer et al., 2006). Consequently, it is nowadays an established
46
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
broad spectrum antiseptic in a wide range of applications and represents a valuable alternative to other antiseptics especially for sensitive patient groups due to its (i) excellent skin compatibility and (ii) 24 h long skin remanent effect as well as (iii) the absence of resistance induction or (iv) local resorption with subsequent systemic side effects (Harke, 1989; Hübner et al., 2010). Due to its superior safety and biocompatibility in comparison to older antiseptics it is hence thought that it will gradually replace conventional antiseptics like triclosan, PVP-iodine or chlorhexidine in the near future (Hübner et al., 2010). Octenidine is usually applied as gel or solution by wiping, spraying, pouring or compresses soaked with commercially available preparations immediately prior to application. Modern wound dressings are designed to fulfill many different requirements such as to promote a rapid and painless wound healing, maintain a moist environment, optimum pH and temperature, form an effective bacterial barrier, and provide protection from potentially irritating wound exudates (Wiegand and Hipler, 2010). Within the last years, a broad variety of modern wound dressings in different forms such as gels, foams or thin films and based on materials such as collagen, alginates, polyurethane, silicone or polyacrylates successfully entered the market. Against this, traditional wound dressings such as cotton wool, natural or synthetic bandages, lint or gauzes, designed to solely keep the wound dry, allow the evaporation of wound exudates and to prevent the entry of harmful bacteria into the wound significantly lost importance (Boateng et al., 2008). However, the search for the “ideal” wound dressing material is still ongoing, since most of the modern wound dressings also possess some drawbacks. Depending on the material used and its form, important criteria such as high moisture vapor transmission and fluid affinity, well balanced liquid uptake and retention without drying out the wound, mechanical stability (e.g., tensile strength) in combination with high softness following the limp contours, availability in different shapes as well as non-allergenic and sterile composition cannot be provided by all in one type of those dressings. Bacterial nanocellulose (BNC) fleeces produced by Gramnegative, aerobic strains of Komagataeibacter xylinus have been proven to be a perfectly suitable biomaterial that fulfills all performance requirements of ideal modern wound dressings (Bielecki et al., 2013). With fibers, 100-times thinner than for conventional cellulose dressings, BNC presents a fully biocompatible and mechanically stable hydropolymer that acts as a barrier to microbial contamination and traumas, but also provides gaseous exchange and maintains a moist wound environment while simultaneously absorbing exudates (Bielecki et al., 2013). While BNC itself has no antimicrobial activity, several approaches using silver (Berndt et al., 2013), silver sulfadiazine (Luan et al., 2012), benzalkonium chloride (Wei et al., 2011), and montmorillonite (UlIslam et al., 2013) were described to equip this biopolymer with antimicrobial properties. Up to now, only one BNC product containing polyhexanide (Suprasorb1 X + PHMB) has made it to the market as antiseptic wound dressing for critically colonized or infected, superficial or deep wounds with low to moderate exudation (Dissemond et al., 2010), but more products containing active drugs will surely follow in the future. In the present study, an active wound dressing based on bacterial nanocellulose loaded with the antiseptic octenidine was developed as drug delivery system for the treatment of acute and chronically infected wounds. The octenidine containing BNC fleeces were produced in 24-well plates under laboratory conditions [lab-scale (LS) BNC] and investigated regarding drug loading performance, controllable drug release, mechanical characteristics, biocompatibility, and antimicrobial efficacy. Additionally, the preservation of the drug release characteristics and the antimicrobial activity were determined in a six month
accelerated stability study. Furthermore, loading and release were compared to that of octenidine loaded BNC samples produced in a large scale process [up-scale (US) BNC] and cut into dimensions (100 mm 100 mm) typically used in clinical applications. 2. Materials and methods 2.1. Preparation and characterization of BNC fleeces For the biosynthesis of BNC, K. xylinus DSM 14666 (culture collection of the Friedrich-Schiller-University of Jena, deposited at the DSMZ, German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany) was cultivated at 28 C in Hestrin– Schramm medium (Kralisch et al., 2009). Under lab-scale static conditions, 24-well plates, 12-well plates (compression studies) (both from Sarstedt, Nuembrecht, Germany) or dumbbell-shaped culture containers (dimensions according to EN ISO 527-1, test specimen type 1A) consisting of polytetrafluoroethylene (RS Components, Moerfelden-Walldorf, Germany) (tensile tests) were used for cultivation of the bacteria as described before (Kralisch et al., 2009; Müller et al., 2013). In an up-scale process BNCs of two different thicknesses (4 mm: US4, 7 mm: US7) were produced as endless fleece under semi-continuous steady state conditions with continuous harvesting using the Horizontal Lift Reactor (HoLiR) process (Kralisch et al., 2009). After biosynthesis, the endless fleece was cut into pieces (100 mm 100 mm). All types of BNC were purified as described previously (Kralisch et al., 2009), sterilized by autoclaving (121 C, 20 min, 2 bar), and stored at 4 C until use. Compressed US BNC (USpress) was obtained by compression of the US7 fleeces between the plates of a hydraulic press (LaboPress P200S, Vogt Maschinenbau, Berlin, Germany) at 3 bar and 20 C for 3 min to form fleeces with a final thickness of about 3 mm. All BNC fleeces were characterized regarding their dimensions (edge length or diameter, height, and weight) as described before (Müller et al., 2013). For calculation of surface area and volume, the geometrical formulas of a circular cylinder or a cuboid were used (Müller et al., 2013). 2.2. Quantification of octenidine Octenidine was purchased from Schuelke & Mayr (Norderstedt, Germany) as a stock solution containing 0.5% octenidine. For quantification of the octenidine concentrations ultraviolet and visible (UV/vis) spectra of octenidine between 200–800 nm and calibration curves (3.4–11.1 mg/mL) were recorded in 0.1 M phosphate buffered solution (PBS) pH 7.4 (Carl Roth, Karlsruhe, Germany) using the Beckman DU 640 spectrophotometer (Beckman Instruments Inc., Fullerton, CA, USA). The time and storage temperature dependent stability of octenidine solutions was investigated by UV/vis measurements after storage at 20 C, 4 C, 20 C, and 32 C over 1, 2, 3, 4, 7, 9, 11, and 16 days and after autoclaving (121 C, 2 bar, 15 min). All experiments were run in triplicates and repeated once. 2.3. Loading and release experiments BNC fleeces produced in 24-well plates (LS samples) were incubated under submersed conditions in 10.0 mL 0.5% octenidine solution at 20 C on a temperature-controlled orbital shaker (KS 4000 ic control, IKA1-Werke, Staufen, Germany) at 70 rpm for 48 h. Fleeces loaded only with PBS were used as control. The drug loading is equal to the difference of the octenidine concentration of the loading solution before and after the loading process. To determine the octenidine release from BNC fleeces at 32 C, loaded BNC was removed from the loading solution and transferred into 20.0 mL PBS pH 7.4 as release medium. Samples were incubated
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
under shaking (200 rpm) and after certain time intervals (0, 0.5, 1, 2, 3, 4, 5, 6, 8, 24, 30, 48, 72, and 96 h), 0.5 mL of the release medium were withdrawn for octenidine quantification. The sampling volume withdrawn was not replaced. Calculations of the total amount of released antiseptic were based on the actual remaining volume of medium at each time point. Using US BNC fleeces from the HoLiR process, volumes of the loading solutions (250.0 mL) and release media (350.0 mL) as well as agitation rate (70 rpm) during release had to be adapted, whereas all other conditions were kept constant. UV/Vis spectrophotometry was used for quantification of octenidine as described above. Results were expressed as cumulative percentage release of the antiseptic against time. BNC samples were characterized regarding their physical dimensions before and after 48 h loading as described above. All experiments were performed in triplicates and repeated once. The semi-empirical Ritger–Peppas equation (Mt/M1 = ktn, power law) was used for modeling of the release data, where Mt and M1 are the cumulative amounts of released octenidine at time t as well as at infinite time. M1 corresponds to the amount of octenidine in BNC after 48 h loading. The logarithmic percentage cumulative release between 5–60% log (Mt/M1 100) was plotted against the logarithmic release time log (t). A linear regression line was fitted to the data, and the slope (n, diffusional exponent) was used as indicator for the release mechanism. Structural and geometrical parameters of the hydrogel were described by the kinetic constant k (Ritger and Peppas, 1987). 2.4. Mechanical characterization The Universal Testing Machine TIRAtest 2710 (Tira, Schalkau, Germany) with a load cell of 1 kN (compression tests) or 0.1 kN (tensile tests) was used for mechanical characterization at 20 C. For compression studies samples were fixed in a compression chamber (inner diameter: 21.5 mm). For each type of BNC at least seven samples were used. Loaded samples and negative controls (in purified water) were compressed (10 mm/min) in vertical direction to the fleece surface until 50% of the original sample thickness were reached. Compressive strength (s C) was calculated by F/S0, where F (N) is the instantaneous force and S0 (mm2) the original cross-sectional area of the BNC. For tensile tests, at least seven dumbell-shaped loaded fleeces and negative controls were fixed in screw clamps and stretched with a deformation speed of 50 mm/min until rupture of samples. Tensile strength (s TS) was calculated by F/A0, where F (N) is the instantaneous force and A0 (mm2) represents the original cross-sectional area. The corresponding strain at rupture (eT) was calculated by Dl/l0, where Dl (mm) represents the change of the length during stretching and l0 (mm) is the original length. 2.5. Scanning electron microscopy (SEM) For SEM analysis samples were lyophilized for 48 h at a pressure of 0.630 mbar and a condenser temperature of 85 C in an ALPHA 2–4 LSC freeze dryer (Martin Christ Gefriertrocknungsanlagen, Osterode, Germany). Cross-sections of freeze dried samples were sputter-coated with gold (Sputter Coater S150B, Edwards, Crawley, UK). The SEM investigations were performed at an acceleration voltage of 15 kV in vacuo (S440i, Leica Microsystems, Wetzlar, Germany). Images of the cross-sections were taken at a working distance of 10 mm. 2.6. Preparation of extracts for biological studies For extract preparation (according to DIN EN ISO 10,993-12), 1 g loaded BNC was used. All further preparation steps were carried
47
out as described for the preparation of a wound dressing extract (Wiegand et al., 2009). Dulbecco’s modified Eagle’s medium (DMEM, Promocell, Heidelberg, Germany) or Caso-bouillon (prepared from peptone and “Lab-Lemco” powder, Oxoid, Basingstoke, UK) were used as extraction media. The prepared extract is referred to as “original extract” (extraction ratio 1 g/50 mL). 2.7. Biocompatibility tests in human keratinocyte cultures Human HaCaT keratinocytes (kind gift from Prof. Dr. N.E. Fusenig, German Cancer Research Center, Heidelberg, Germany) were cultured in DMEM supplemented with 10% fetal calf serum (PromoCell, Germany) and 1% antibiotic/antimycotic solution consisting of 10,000 U/mL penicillin, 10,000 mg/mL streptomycin and 25 mg/mL amphotericin (Gibco Life Technologies, Darmstadt, Germany) for 5–7 days, harvested and seeded into 96-well plates as described previously (Wiegand et al., 2011). After 48 h, culture medium was replaced by the original extract, its dilutions (1:2, 1:4, 1:10, 1:100, 1:200, and 1:1000) or dilutions of the octenidine stock solution with concentrations from 0.5 to 500 mg/mL for further incubation (1, 24, 48 h). Untreated and Triton-X 100 treated cells (Merck Millipore, Schwalbach, Germany) served as negative and positive controls, respectively. Cell proliferation was determined using a luminometric adenosine triphosphate (ATP) assay (ATPliteTM-M assay, Packard Bioscience BV, Groningen, The Netherlands) as recommended by the manufacturer’s protocol. Luminescence was measured using a LUMIstar Galaxy (BMG Labtech, Ortenberg, Germany). A calibration curve was used for calculation of ATP concentration, and the number of the HaCaT cells was expressed as percentage of the cellular ATP content of the control cells (Wiegand et al., 2011). Measurements were performed eightfold and repeated once.
2.8. Microplate laser nephelometry (MLN) and direct contact test for determination of antimicrobial activity Staphylococcus aureus (S. aureus, ATCC 6538, DSMZ, Germany) was used as Gram-positive model germ for infected wounds. Inoculation of Caso-bouillon (20 mL) with S. aureus, incubation (37 C, 24 h) and preparation of a cell suspension (app. 5 103 colony forming units/mL) was carried out according to former reports (Wiegand et al., 2012). Octenidine stock solution (0.5%) and the original extracts were diluted using Caso-bouillon. Aliquots (100 mL) were pipetted in triplicate into the wells of a 96-well plate (Greiner Bio-One, Frickenhausen, Germany), and 100 mL cell suspension were added to each well. Pure Caso-bouillon was used as blank. Microplates were incubated at 37 C for 24 h in the microplate luminometer (NEPHELOstar Galaxy, BMG Labtech, Germany). Measurements were carried out automatically every hour (Wiegand et al., 2012). Experiments were run as quadruplicates and repeated once. For the direct contact test (according to JIS L 1902:2002), Casobouillon was inoculated with 1–2 colonies of S. aureus and cultivated at 37 C for 24 h under aerobic conditions. Loaded BNC was incubated in 400 mL bacterial suspension at 37 C for 24 h. A polyester material (kindly provided by the TITK, Thuringian Institute of Textile and Plastics Research, Rudolstadt, Germany) was used as growth control. Experiments were run in triplicates and repeated once. Extraction of incubated samples and germ number determination was performed as described before (Wiegand et al., 2011). Colony forming units per milliliter (cfu/ mL) were calculated, and growth reduction was compared to the starting value (log growth reduction = log cfu(negative control)(24 h) log cfu(sample)(24 h)) (Wiegand et al., 2011). A logarithmic reduction 1 and 3 represent a significant and values >3 a strong, antimicrobial activity. 2.9. Storage stability testing An accelerated stability study for drug products packaged in impermeable containers was conducted according to ICH Q1A (R2) (2003), using LS samples that were loaded under sterile conditions as described above. Loaded BNC fleeces were packed in an aluminum compound foil consisting of polyethylene terephthalate (PET), aluminum (Al) and polyethylene (PE) (Tesseraux, Buerstadt, Germany), sealed by a welding seam to prevent evaporation and stored at 40 C in an incubator (Heraeus kelvitron1 B6200, Kendro, Langenselbold, Germany). Loading and packaging were performed under aseptic conditions. After 0, 3 and 6 months, release characteristics, antimicrobial activity and cytotoxicity of loaded samples were investigated as described in the sections above. The experiments were run in four independent experiments.
Fig. 1. (A) Dimensions of the LS samples expressed as mean SD of n = 20 samples. Photographs of LS BNC samples cultivated under static conditions in 24-well plates before (B) and after (C) 48 h octenidine loading.
2.10. Data analysis and statistics The half maximal inhibitory concentration (IC50) of pure octenidine was calculated for the evaluation of antibacterial activity. For the respective octenidine concentration tested the AUC (area under the curve) was determined for each growth curve over 24 h and calculated as percentage of the untreated control. This was used to generate a dose-response-curve for octenidine vs. S. aureus from which the IC50 was calculated using a logistic fit function (y = A2 + (A1-A2)/(1 + (x/x0) ^p); A1: upper limit, A2: lower limit, x0: IC50, p: slope of the curve; Origin 7.5). For the assessment of cytotoxic effects, the LC50 value (half maximal lethal concentration for cells) for octenidine was acquired correspondingly. Dose-response curves were obtained from the cytotoxicity measurement results at 24 h. The LC50 was determined using the following logistic fit function (y = A2 + (A1-A2)/(1 + (x/x0) ^p); A1: upper limit, A2: lower limit, x0:LC50, p: slope of the curve; Origin 7.5). Statistically significant differences (p-values 0.05) were determined by ANOVA one way analysis of variance (Microsoft Excel1). 3. Results and discussion 3.1. Validation of the octenidine quantification The quantification of octenidine by UV/vis spectrophotometry (Papen-Botterhuis et al., 2009) was validated with regard to linearity, accuracy and stability. Octenidine showed an absorption maximum at 281 nm in purified water and in PBS. The detection limit was determined as 0.1 mg/mL octenidine with a corresponding UV absorbance of 0.007 0.001. UV/vis spectrophotometric measurements confirmed the stability of octenidine during loading, release and under storage conditions at 20, 4, 20, and 32 C without any loss of drug (data not shown). Octenidine stability during the autoclaving process at 121 C and 2 bar for 15 min was demonstrated by a recovery of 99.0 0.8%. This is in line with observations by Harke (1989) who demonstrated that octenidine could be sterilized in aqueous solution up to 130 C without losing its integrity or decreasing its efficacy. Due to its bispyridinamine structure (Bailey et al., 1984), octenidine differs from other antiseptics such as quaternary ammonium compounds or bisguanidines. Since the molecule does not contain amide or ester structures, octenidine was shown to be stable under different physical and chemical conditions and not to be prone to hydrolysis at pH 1.6–12.2 (Harke, 1989). Nevertheless, in certain solutions like
Ringers’solution stability issues of octenidine formulations may occur as described in the European patent EP2311456A1 (Baur et al., 2011).
3.2. Octenidine loading and release of lab-scale BNC produced under static cultivation conditions LS BNC samples produced by K. xylinus under static cultivation conditions formed mechanically stable mat like hydrogels that were harvested at the interface between air and culture medium. As shown in Fig. 1(A), the round shaped LS samples (Fig. 1(B)) of about 16 mm diameter demonstrated highly reproducible sample dimensions with a height of about 7 mm and weights of about 1.5 g. Surface area and volume were calculated as 7.6 cm2 and 1.4 cm3 (Fig. 1(A)), respectively. The morphological characterization of the LS samples by SEM revealed the typical multilayered BNC structure with a dense top layer (Wesarg et al., 2012), a middle layer of nanofibrillar architecture with interconnected pores and tunnels, and residues of the prepolymer at the bottom (Klemm et al., 2001). Fig. 2(A) shows the cross-section of the middle layer of a LS BNC sample that is characterized by a nanodimensional network structure with a large surface area built up by single, randomly arranged BNC nanofibers. Active compounds could be loaded into the hydropolymer BNC either during polymer formation or afterwards in a post synthesis modification (Trovatti et al., 2011; Wesarg et al., 2012). For the loading of the BNC fleeces with octenidine, the post synthesis technique was selected, since it has the advantage of mild conditions, thus preventing drug damage during biosynthesis and purification of the BNC fleeces (Lin and Anseth, 2009). Additionally, octenidine as an antimicrobial active agent could impair the growth of the BNC producing bacteria (Yamanaka et al., 2000). After 48 h loading by immersion in an aqueous solution of octenidine at ambient temperature, the LS fleeces were found to be homogeneous without visible drug aggregates on the surface or any other macroscopically visual changes (Fig. 1(C)). The weight of the LS samples was found to be increased by 6.8 2.1% during the loading process (Fig. 3(A)) that corresponded to an amount of 4.2 0.1 mg octenidine/g loaded BNC (Fig. 3(B)). The incorporation of drugs into the BNC fleece was suggested to be accomplished by diffusion and capillary forces due to the large hydrophilic surface area (White and Brown, 1989). The increase of BNC weight during loading suggested that the water in the pores of the network was
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
49
Fig. 2. SEM images of BNC cross-sections of LS (A), US7 (B), and US4 (C) samples. The cross-section of the upper sample region of a USpress sample (SEM picture taken at 3.1 mm sample height) is shown in (D) while (E) represents the cross-section of the middle region (SEM picture taken at 2 mm sample height) (magnification 3000x, work distance 10 mm).
partially replaced by the antiseptic drug. In scanning electron microscopy, no relevant changes of the BNC network due to the interaction with octenidine (data not shown) were observed. The release of octenidine was investigated at skin temperature (32 C) over 96 h using an agitated release model (orbital shaker)
which is an established method for the testing of active wound dressings and has been described previously for investigating the release behavior of several drugs loaded into BNC hydropolymers (Wilhelms et al., 2007). Fig. 4 shows the cumulative release curve of octenidine loaded LS BNC plotted against time. The samples
Fig. 3. Loading of LS and US BNC fleeces (mean SD, n = 3). Increase of BNC weight (A), amount of loaded octenidine (mg)/g loaded BNC (B), and correlation between octenidine uptake and weight increase (C) after 48 h loading with 5 mg/mL octenidine under submersed conditions at 20 C.
50
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
Fig. 4. Cumulative octenidine release in PBS (pH 7.4) of BNC fleeces at 32 C over 96 h. The amount of released octenidine from loaded LS and US samples was quantified at specific time points by UV-spectrophotometry at 281 nm. Data are expressed as mean SD of 3 samples.
exhibited exponential release profiles with an initial burst (82.7 2.6%) within the first 8 h, reaching equilibrium conditions after 24 h and a total release of 91.8 2.0%. These release conditions favor the application as wound dressing since they provide a rapid increase of drug concentration in the wound to reach therapeutically relevant threshold amounts followed by a slower release to maintain the drug level (Wilhelms et al., 2007). 3.3. Comparison of lab-scale BNC fleeces with BNC produced under semi-continuous cultivation conditions 3.3.1. Macroscopic and microscopic morphology of the fleeces The transferability of the drug loading and release data of samples obtained from the static process to a large scaled, semicontinuous production process was tested next. In order to define the fleece parameters which might influence these results, LS samples were compared to US BNC. US samples of two different heights (7 mm and 4 mm) were square cut (US7: Fig. 5(A), US4: Fig. 5(B) with comparable edge lengths and surface areas of about 100 mm and 220 cm2 (Table 1), thus representing typical dimensions of commercial wound dressings (Hirsch et al., 2011). Additionally, the US7 fleeces were semi-dried by removal of water through a pressing step (USpress), thus including partially dried wound dressings in our studies which are used in the clinical practice for highly exudating wounds. By pressing, the initial mass
Fig.
of US7 samples was reduced from 82 g to 32 g, which corresponds to a removal of 50 g water. USpress samples (Fig. 5(C)) were characterized by a homogeneous and smooth surface, a macroscopical more compact fleece morphology, an edge length of 100 mm 100 mm, and a height of 3 mm (Table 1). Volume and weight of all US samples were found to be strongly depending on the thickness and the water content of the samples with a decrease from US7 > US4 > USpress (Table 1). The reduction of thickness had an only small impact on the calculated total fleece area due to a small impact of height compared to the edge length of the US fleeces (Table 1). SEM investigations of non-compressed BNC materials produced by the HoLiR process revealed that the network structure was neither changed by the up-scale process compared to LS fleeces obtained under static cultivation in small vessels (Kralisch et al., 2009) nor by the variation of fleece thickness. Cross-sections of US7 (Fig. 2(B)) and US4 (Fig. 2(C)) fleeces exhibited the same hierarchical network structure as LS fleeces, with randomly distributed BNC fibers and interconnected pores (Fig. 2(A)). Distinct changes could be observed for the USpress samples. After top-down compression, they were characterized in the upper and lower region (Fig. 2(D)) by dense layers of agglomerated fibers, partially connected by single BNC fibers which formed a more open network with slightly increased pore sizes and volumes. The network structure of the USpress middle region seemed to be less affected by compression (Fig. 2(E)), with a network found to be more horizontally oriented than those of US7 samples. While the entire surface of LS samples is exclusively defined by bacteria during biosynthesis, the cutting of US fleeces led to a structure with more open pores due to the cutting process (Wesarg et al., 2012). 3.3.2. Loading characteristics of US fleeces All parameters of the loading procedure were transferable from the lab-scale to the up-scale procedure with one exception: the volume of the loading solution had to be increased to ensure that the US fleeces were totally submersed in the loading solution. In preliminary studies, it could be shown that the ratio of medium volume to BNC weight had no effect on the amount of loaded drug as long as the loading solution is present in excess (data not shown). After 48 h of octenidine loading, none of the US fleeces showed any macroscopical changes (US7: Fig. 5(D), US4: Fig. 5(E), USpress: Fig. 5(F)). Measured immediately at ambient temperature and humidity after removal from the loading solution, US4 samples increased their weight by 8.2 0.7% whereas US7 demonstrated a significantly lower weight increase of 2.3 1.7% (Fig. 3(A)). In
5. Photographs of US7 (A, D), US4 (B, E) and USpress (C, F) BNC fleeces before (upper row, A–C) and after (lower row, D–F) 48 h loading with 5 mg/mL octenidine.
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
Table 1 Comparison of the physical dimensions edge length, height, weight, volume and surface area of the BNC fleeces produced in an up-scale process by the semicontinuous HoLiR technique. Ten samples of each BNC type were characterized, and results expressed as mean SD. Edge length (mm) Height (mm) Weight (g) Surface area (cm2) Volume (cm3)
US4 100.4 4.5 48.7 218.9 45.0
1.7 0.3 1.8 1.6 2.5
US7 98.6 7.6 80.8 224.2 73.8
2.1 0.4 1.8 3.4 3.7
USpress 101.4 1.9 3.1 0.1 31.4 2.4 216.7 3.6 31.0 0.5
contrast, the absorption of the compressed USpress fleeces was characterized by a higher weight gain of 57.5 8.0%. The calculated amount of octenidine after loading was increased in the order US7 < US4 < USpress (Fig. 3(B)). Also, the uptake of octenidine per gram unloaded BNC with US7 < US4 0.5% for all tested materials, independent of the storage time. This could be confirmed for storage times up to 6 months, where no loss of antimicrobial effectivity could be detected. The correlation of cytotoxic potential and antimicrobial efficiency revealed that antibacterial effects could be obtained at concentrations well below the values for cell damage effects. This is in accordance with Müller and Kramer (2008) who have shown that octenidine was more toxic to S. aureus than to murine fibroblasts under identical test conditions, resulting in a good biocompatibility with concomitant antibacterial activity. Notably, this also applies to octenidine loaded BNC where we could show that antibacterial activity is more pronounced than cytotoxicity against HaCaT cells. The antimicrobial efficacy of octenidine loaded BNC was further investigated according to JIS L 1902 (Fig. 9(C)) which describes a direct contact test. The results correlated well with the data of the nephelometric measurements. After 24 h of incubation, all
octenidine loaded BNC samples exhibited log reduction values 6 representing a strong antibacterial efficacy. This was also found after storage for 6 months. For comparison, the polyester growth control had no impact on bacterial growth. From the microbiological data it could be concluded that neither the interaction with BNC nor the storage over 6 months impairs the activity of octenidine. For comparison, Hirsch et al. (2011) reported earlier for combinations of octenidine with commercially available wound dressings that only 54% of all tested combinations were found to be therapeutically active. Because of its cationic character, octenidine seems to interact preferentially with anionic structures in these dressings resulting in a reduced antibacterial activity. 4. Conclusion In the present study, the antiseptic octenidine was incorporated in bacterial nanocellulose with the intension to develop a ready-touse system for wound treatment. The combination of both components leads to a wound dressing with advanced healing and superior material properties, combined with an efficient infection control and minimized unwanted side effects. Octenidine loaded BNC demonstrated release profiles that were comparable to already marketed products. They were stable for up to 6 months storage without losing their favorable physicochemical and biological characteristics. A study to scale up the fleeces to sizes typically used in the clinical praxis revealed parameters that can be
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
used to selectively control the time-release curves of the BNC. Further investigations regarding a broader spectrum of microorganisms and wound studies in vivo will follow. Our findings suggest that the octenidine loaded BNC exhibit a tremendous potential as wound dressing with controlled drug delivery. Acknowledgements The authors would like to thank Ramona Brabetz and Elena Pfaff for their excellent technical assistance. All authors (NanocellCare) gratefully acknowledge the Thuringian Ministry of Education, Science and Culture and the European Fund for Regional Development (B714-10032). S.M. is grateful to the FAZIT Foundation, Gemeinnützige Verlagsgesellschaft mbH for financial support. References Al-Doori, Z., Goroncy-Bermes, P., Gemmell, C.G., Morrison, D., 2007. Low-level exposure of MRSA to octenidine dihydrochloride does not select for resistance. J. Antimicrob. Chemother. 59, 1280–1282. Amin, M.C.I.M., Ahmad, N., Halib, N., Ahmad, I., 2012. Synthesis and characterization of thermo- and pH-responsive bacterial cellulose/acrylic acid hydrogels for drug delivery. Carbohydr. Polym. 88, 465–473. Bailey, D.M., DeGrazia, C.G., Hoff, S.J., Schulenberg, P.L., O‘Connor, J.R., Paris, D.A., Slee, A.M., 1984. Bispyridinamines: a new class of topical antimicrobial agents as inhibitors of dental plaque. J. Med. Chem. 27, 1457–1464. Baur B., Behrends S., Siebert J., Dettmann A., Golombiewski M., 2011. EP2311456A1: Wound and mucous membrane antiseptic based on bispyridiniumalkanes. Schuelke & Mayr GmbH, 12. Berndt, S., Wesarg, F., Wiegand, C., Kralisch, D., Müller, F., 2013. Antimicrobial porous hybrids consisting of bacterial nanocellulose and silver nanoparticles. Cellulose 20, 771–783. Bielecki, S., Kalinowska, H., Krystynowicz, A., Kubiak, K., Kolodziejczyk, M., de Groeve, M., 2013. Wound dressings and cosmetic materials from bacterial nanocellulose. In: Gama, M., Gatenholm, P., Klemm, D. (Eds.), Bacterial NanoCellulose: A Sophisticated Multifunctional Material. Taylor and Francis (CRC Press), Boka Raton, pp. 157–174. Boateng, J., Matthews, K., Stevens, H., Eccleston, G., 2008. Wound healing dressings and drug delivery systems: a review. J. Pharm. Sci. 97, 2892–2923. Briese, V., Neumann, G., Waldschlaeger, J., May, T.W., Siebert, J., Gerber, B., 2011. Efficacy and tolerability of a local acting antiseptic agent in the treatment of vaginal dysbiosis during pregnancy. Arch. Gynecol. Obstet. 283, 585–590. Chiaoprakobkij, N., Sanchavanakit, N., Subbalekha, K., Pavasant, P., Phisalaphong, M., 2011. Characterization and biocompatibility of bacterial cellulose/alginate composite sponges with human keratinocytes and gingival fibroblasts. Carbohydr. Polym. 85, 548–553. Dissemond, J., Gerber, V., Kramer, A., Riepe, G., Strohal, R., Vasel-Biergans, A., Eberlein, T., 2010. A practice-oriented recommendation for treatment of critically colonised and locally infected wounds using polihexanide. J. Tissue Viabil. 19, 106–115. Gilbert, P., Moore, L.E., 2005. Cationic antiseptics: diversity of action under a common epithet. J. Appl. Microbiol. 99, 703–715. Goldstein, E.J., Citron, D.M., Nesbit, C.A., 1996. Diabetic foot infections: bacteriology and activity of 10 oral antimicrobial agents against bacteria isolated from consecutive cases. Diabetes Care 19, 638–641. Harke, H.P., 1989. Octenidine dihydrochloride, properties of a new antimicrobial agent. Zentralbl. Hyg. Umweltmed. 188, 188–193. Hirsch, T., Jacobsen, F., Rittig, A., Goertz, O., Niederbichler, A., Steinau, H.U., Seipp, H. M., Steinstraesser, L., 2009. A comparative in vitro study of cell toxicity of clinically used antiseptics. Hautarzt 60, 984–991. Hirsch, T., Limoochi-Deli, S., Lahmer, A., Jacobsen, F., Goertz, O., Steinau, H.-U., Seipp, H.-M., Steinstraesser, L., 2011. Antimicrobial activity of clinically used antiseptics and wound irrigating agents in combination with wound dressings. Plast. Reconstr. Surg. 127, 1539–1545. Huang, L., Chen, X., Nguyen, T.X., Tang, H., Zhang, L., Yang, G., 2013. Nano-cellulose 3D-networks as controlled-release drug carriers. J. Mater. Chem. B 1, 2976– 2984. Hübner, N.O., Siebert, J., Kramer, A., 2010. Octenidine dihydrochloride, a modern antiseptic for skin, mucous membranes and wounds. Skin Pharmacol. Physiol. 23, 244–258. ICHQ1A(R2), 2003. Stability testing of new drug substances and products. ICH Q8(R2), 2009. Pharmaceutical development. Klemm, D., Schumann, D., Udhardt, U., Marsch, S., 2001. Bacterial synthesized cellulose – artificial blood vessels for microsurgery. Prog. Polym. Sci. 26, 1561– 1603. Koburger, T., Hübner, N.O., Braun, M., Siebert, J., Kramer, A., 2010. Standardized comparison of antiseptic efficacy of triclosan, PVP-iodine, octenidine dihydrochloride, polyhexanide and chlorhexidine digluconate. J. Antimicrob. Chemother. 65, 1712–1719.
55
Kralisch, D., Hessler, N., Klemm, D., Erdmann, R., Schmidt, W., 2009. White biotechnology for cellulose manufacturing – the HoLiR concept. Biotechnol. Bioeng. 105, 740–747. Kramer, A., Müller, G., Assadian, O., 2006. Indications and agent selection for the antiseptic therapy of secondarily healing wounds. GMS Krankenhaushyg. Interdiszip. 1, 1–11. Lin, C.-C., Anseth, K.S., 2009. PEG hydrogels for the controlled release of biomolecules in regenerative medicine. Pharm. Res. 26, 631–643. Luan, J., Wu, J., Zheng, Y., Song, W., Wang, G., Guo, J., Ding, X., 2012. Impregnation of silver sulfadiazine into bacterial cellulose for antimicrobial and biocompatible wound dressing. Biomed. Mater. 7, 065006. Müller, A., Ni, Z., Hessler, N., Wesarg, F., Müller, F.A., Kralisch, D., Fischer, D., 2013. The biopolymer bacterial nanocellulose as drug delivery system: investigation of drug loading and release using the model protein albumin. J. Pharm. Sci. 102, 579–592. Müller, G., Kramer, A., 2008. Biocompatibility index of antiseptic agents by parallel assessment of antimicrobial activity and cellular cytotoxicity. J. Antimicrob. Chemother. 61, 1281–1287. Müller, G., Langer, J., Siebert, J., Kramer, A., 2014. Residual antimicrobial effect of chlorhexidine digluconate and octenidine dihydrochloride on reconstructed human epidermis. Skin Pharmacol. Physiol. 27, 1–8. Nakayama, A., Kakugo, A., Gong, J.P., Osada, Y., Takai, M., Erata, T., Kawano, S., 2004. High mechanical strength double-network hydrogel with bacterial cellulose. Adv. Funct. Mater. 14, 1124–1128. Papen-Botterhuis, N.E., ten Cate, A.T., Kolijn, J., Eversdijk, J., 2009. Enzyme-triggered release of antimicrobials from gelatin microcapsules. XVIIth International Conference on BioencapsulationGroningen, The Netherlands. Pitten, F.A., Werner, H.P., Kramer, A., 2003. A standardized test to assess the impact of different organic challenges on the antimicrobial activity of antiseptics. J. Hosp. Infect. 55, 108–115. Reynolds, T.D., Mitchell, S.A., Balwinski, K.M., 2002. Investigation of the effect of tablet surface area/volume on drug release from hydroxypropylmethylcellulose controlled-release matrix tablets. Drug Dev. Ind. Pharm. 28, 457–466. Ritger, P.L., Peppas, N.A., 1987. A simple equation for description of solute release II. Fickian and anomalous release from swellable devices. J. Control. Release 5, 37–42. Sanchavanakit, N., Sangrungraungroj, W., Kaomongkolgit, R., Banaprasert, T., Pavasant, P., Phisalaphong, M., 2006. Growth of human keratinocytes and fibroblasts on bacterial cellulose film. Biotechnol. Prog. 22, 1194–1199. Sedlock, D.M., Bailey, D.M., 1985. Microbicidal activity of octenidine hydrochloride, a new alkanediylbis[pyridine] germicidal agent. Antimicrob. Agents Chemother. 28, 786–790. Silva, N.H.C.S., Rodrigues, A.F., Almeida, I.F., Costa, P.C., Rosado, C., Neto, C.P., Silvestre, A.J.D., Freire, C.S.R., 2014. Bacterial cellulose membranes as transdermal delivery systems for diclofenac: in vitro dissolution and permeation studies. Carbohyd. Polym. 106, 264–269. Slee, A.M., O‘Connor, J.R., 1983. In vitro antiplaque activity of octenidine dihydrochloride (WIN 41464-2) against preformed plaques of selected oral plaque-forming microorganisms. Antimicrob. Agents Chemother. 23, 379–384. Torres, A.J., Zhu, C., Shuler, M.L., Pannullo, S., 2011. Paclitaxel delivery to brain tumors from hydrogels: a computational study. Biotechnol. Prog. 27, 1478–1487. Trovatti, E., Silva, N.H.C.S., Duarte, I.F., Rosado, C.F., Almeida, I.F., Costa, P., Freire, C.S. R., Silvestre, A.J.D., Pascoal, N.C., 2011. Biocellulose membranes as supports for dermal release of lidocaine. Biomacromolecules 12, 4162–4168. Ul-Islam, M., Khan, T., Khattak, W., Park, J., 2013. Bacterial cellulose-MMTs nanoreinforced composite films: novel wound dressing material with antibacterial properties. Cellulose 20, 589–596. Vanscheidt, W., Harding, K., Teot, L., Siebert, J., 2012. Effectiveness and tissue compatibility of a 12-week treatment of chronic venous leg ulcers with an octenidine based antiseptic – a randomized, double-blind controlled study. Int. Wound J. 9, 316–323. Wei, B., Yang, G., Hong, F., 2011. Preparation and evaluation of a kind of bacterial cellulose dry films with antibacterial properties. Carbohydr. Polym. 84, 533–538. Wesarg, F., Schlott, F., Grabow, J., Kurland, H.-D., Heßler, N., Kralisch, D., Müller, F.A., 2012. In situ synthesis of photocatalytically active hybrids consisting of bacterial nanocellulose and anatase nanoparticles. Langmuir 28, 13518–13525. White, D.G., Brown Jr., R.M., 1989. Prospects for the commercialization of the biosynthesis of microbial cellulose. In: Schürch, C. (Ed.), Cellulose and Wood: Chemistry and Technology. Wiley, New York, pp. 573–590. Wiegand, C., Abel, M., Ruth, P., Hipler, U.-C., 2009. HaCaT keratinocytes in co-culture with Staphylococcus aureus can be protected from bacterial damage by polihexanide. Wound Repair Regen. 17, 730–738. Wiegand, C., Abel, M., Ruth, P., Hipler, U.C., 2011. Superabsorbent polymercontaining wound dressings have a beneficial effect on wound healing by reducing PMN elastase concentration and inhibiting microbial growth. J. Mater. Sci.: Mater. Med. 22, 2583–2590. Wiegand, C., Abel, M., Ruth, P., Hipler, U.C., 2012. Analysis of the adaptation capacity of Staphylococcus aureus to commonly used antiseptics by microplate laser nephelometry. Skin Pharmacol. Physiol. 25, 288–297. Wiegand, C., Hipler, U.-C., 2010. Polymer-based biomaterials as dressings for chronic stagnating wounds. Macromol. Symp. 294, 1–13. Wilhelms, T., Schulze, D., Alupeil, C., I, Rohrer, C., Abel, M., Wiegand, C., Hipler, U.C., 2007. Release of polyhexamethylene biguanide hydrochloride (PHMB) from a hydroballanced cellulose wound dressing with PHMB. 17th Conference of the European Wound Management AssociationGlasgow. Yamanaka, S., Ishihara, M., Sugiyama, J., 2000. Structural modification of bacterial cellulose. Cellulose 7, 213–225.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Active wound dressings based on bacterial nanocellulose as drug delivery system for octenidine Sebastian Moritz a , Cornelia Wiegand b , Falko Wesarg c , Nadine Hessler a , Frank A. Müller c,d, Dana Kralisch a,d , Uta-Christina Hipler b , Dagmar Fischer a,d, * a
Department of Pharmaceutical Technology, Friedrich-Schiller-University Jena, Otto-Schott-Str. 41, Jena 07745, Germany Department of Dermatology, University Medical Center Jena, Erfurter Str. 35, Jena 07743, Germany c Otto-Schott-Institute of Materials Research (OSIM), Friedrich-Schiller-University Jena, Loebdergraben 22, Jena 07743, Germany d Jena Center for Soft Matter (JCSM), Friedrich-Schiller-University Jena, Humboldtstraße 10, Jena 07743, Germany b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 February 2014 Received in revised form 25 April 2014 Accepted 29 April 2014 Available online 2 May 2014
Although bacterial nanocellulose (BNC) may serve as an ideal wound dressing, it exhibits no antibacterial properties by itself. Therefore, in the present study BNC was functionalized with the antiseptic drug octenidine. Drug loading and release, mechanical characteristics, biocompatibility, and antimicrobial efficacy were investigated. Octenidine release was based on diffusion and swelling according to the Ritger–Peppas equation and characterized by a time dependent biphasic release profile, with a rapid release in the first 8 h, followed by a slower release rate up to 96 h. The comparison between lab-scale and up-scale BNC identified thickness, water content, and the surface area to volume ratio as parameters which have an impact on the control of the release characteristics. Compression and tensile strength remained unchanged upon incorporation of octenidine in BNC. In biological assays, drug-loaded BNC demonstrated high biocompatibility in human keratinocytes and antimicrobial activity against Staphylococcus aureus. In a long-term storage test, the octenidine loaded in BNC was found to be stable, releasable, and biologically active over a period of 6 months without changes. In conclusion, octenidine loaded BNC presents a ready-to-use wound dressing for the treatment of infected wounds that can be stored over 6 months without losing its antibacterial activity. ã 2014 Elsevier B.V. All rights reserved.
PubChem classification: Octenidine dihydrochloride (PubChem CID: 51166) Keywords: Nanocellulose Wound dressing Octenidine Antiseptic Hydrogel
1. Introduction The increasing resistance of microorganisms against antibiotics in the treatment of acute and chronic colonized wounds has led to the renaissance of antiseptic drugs for local application in the prophylaxis and the treatment of wound infections. Octenidine dihydrochloride (octenidine) was introduced for skin, mucous membrane, and wound antisepsis more than two decades ago (for review, see Hübner et al., 2010). Due to its cationic
Abbreviations: BNC, Bacterial nanocellulose; MRSA, Methicillin-resistant Staphylococcus aureus; PHMB, Polyhexamethylene biguanide (polyhexanide); LS, Lab-scale; US, Up-scale; HoLiR, Horizontal lift reactor; SEM, Scanning electron microscopy; MLN, Microplate laser nephelometry; IC50, Half maximal inhibitory concentration; AUC, Area under the curve; LC50, Half maximal lethal concentration; SAV, Surface area to volume. * Corresponding author at: Department of Pharmaceutical Technology, FriedrichSchiller-University, Otto-Schott-Str. 41, 07745 Jena, Germany. Tel.: +49 3641 949941/949940; fax: +49 3641 949942. E-mail address: dagmar.fi[email protected] (D. Fischer). http://dx.doi.org/10.1016/j.ijpharm.2014.04.062 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
charge, it exhibits strong interactions with negatively charged components of microbial cell walls and membranes, leading to an inhibition of vital cell functions and a broad antimicrobial activity against Gram-positive and Gram-negative germs, plaque-forming bacteria, and fungi (Harke, 1989; Sedlock and Bailey, 1985; Slee and O‘Connor, 1983). Furthermore, octenidine is effective against Methicillin-resistant Staphylococcus aureus (MRSA) (Al-Doori et al., 2007) and exhibits a moderate virucidal effectivity against enveloped viruses (Hübner et al., 2010). Its activity was not found to be compromised by interfering substances like albumin or mucin (Pitten et al., 2003). A vast amount of knowledge on efficacy, tolerance, and safety has been collected from preclinical cell culture and animal studies as well as clinical trials (Hübner et al., 2010; Koburger et al., 2010; Vanscheidt et al., 2012). The following properties support the rationale for the selection of octenidine in this study: octenidine was found to be superior to polyhexanide (polyhexamethylene biguanide, PHMB), the agent of first choice for chronically poorly healing wounds and burns in terms of its therapeutic spectrum (Kramer et al., 2006). Consequently, it is nowadays an established
46
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
broad spectrum antiseptic in a wide range of applications and represents a valuable alternative to other antiseptics especially for sensitive patient groups due to its (i) excellent skin compatibility and (ii) 24 h long skin remanent effect as well as (iii) the absence of resistance induction or (iv) local resorption with subsequent systemic side effects (Harke, 1989; Hübner et al., 2010). Due to its superior safety and biocompatibility in comparison to older antiseptics it is hence thought that it will gradually replace conventional antiseptics like triclosan, PVP-iodine or chlorhexidine in the near future (Hübner et al., 2010). Octenidine is usually applied as gel or solution by wiping, spraying, pouring or compresses soaked with commercially available preparations immediately prior to application. Modern wound dressings are designed to fulfill many different requirements such as to promote a rapid and painless wound healing, maintain a moist environment, optimum pH and temperature, form an effective bacterial barrier, and provide protection from potentially irritating wound exudates (Wiegand and Hipler, 2010). Within the last years, a broad variety of modern wound dressings in different forms such as gels, foams or thin films and based on materials such as collagen, alginates, polyurethane, silicone or polyacrylates successfully entered the market. Against this, traditional wound dressings such as cotton wool, natural or synthetic bandages, lint or gauzes, designed to solely keep the wound dry, allow the evaporation of wound exudates and to prevent the entry of harmful bacteria into the wound significantly lost importance (Boateng et al., 2008). However, the search for the “ideal” wound dressing material is still ongoing, since most of the modern wound dressings also possess some drawbacks. Depending on the material used and its form, important criteria such as high moisture vapor transmission and fluid affinity, well balanced liquid uptake and retention without drying out the wound, mechanical stability (e.g., tensile strength) in combination with high softness following the limp contours, availability in different shapes as well as non-allergenic and sterile composition cannot be provided by all in one type of those dressings. Bacterial nanocellulose (BNC) fleeces produced by Gramnegative, aerobic strains of Komagataeibacter xylinus have been proven to be a perfectly suitable biomaterial that fulfills all performance requirements of ideal modern wound dressings (Bielecki et al., 2013). With fibers, 100-times thinner than for conventional cellulose dressings, BNC presents a fully biocompatible and mechanically stable hydropolymer that acts as a barrier to microbial contamination and traumas, but also provides gaseous exchange and maintains a moist wound environment while simultaneously absorbing exudates (Bielecki et al., 2013). While BNC itself has no antimicrobial activity, several approaches using silver (Berndt et al., 2013), silver sulfadiazine (Luan et al., 2012), benzalkonium chloride (Wei et al., 2011), and montmorillonite (UlIslam et al., 2013) were described to equip this biopolymer with antimicrobial properties. Up to now, only one BNC product containing polyhexanide (Suprasorb1 X + PHMB) has made it to the market as antiseptic wound dressing for critically colonized or infected, superficial or deep wounds with low to moderate exudation (Dissemond et al., 2010), but more products containing active drugs will surely follow in the future. In the present study, an active wound dressing based on bacterial nanocellulose loaded with the antiseptic octenidine was developed as drug delivery system for the treatment of acute and chronically infected wounds. The octenidine containing BNC fleeces were produced in 24-well plates under laboratory conditions [lab-scale (LS) BNC] and investigated regarding drug loading performance, controllable drug release, mechanical characteristics, biocompatibility, and antimicrobial efficacy. Additionally, the preservation of the drug release characteristics and the antimicrobial activity were determined in a six month
accelerated stability study. Furthermore, loading and release were compared to that of octenidine loaded BNC samples produced in a large scale process [up-scale (US) BNC] and cut into dimensions (100 mm 100 mm) typically used in clinical applications. 2. Materials and methods 2.1. Preparation and characterization of BNC fleeces For the biosynthesis of BNC, K. xylinus DSM 14666 (culture collection of the Friedrich-Schiller-University of Jena, deposited at the DSMZ, German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany) was cultivated at 28 C in Hestrin– Schramm medium (Kralisch et al., 2009). Under lab-scale static conditions, 24-well plates, 12-well plates (compression studies) (both from Sarstedt, Nuembrecht, Germany) or dumbbell-shaped culture containers (dimensions according to EN ISO 527-1, test specimen type 1A) consisting of polytetrafluoroethylene (RS Components, Moerfelden-Walldorf, Germany) (tensile tests) were used for cultivation of the bacteria as described before (Kralisch et al., 2009; Müller et al., 2013). In an up-scale process BNCs of two different thicknesses (4 mm: US4, 7 mm: US7) were produced as endless fleece under semi-continuous steady state conditions with continuous harvesting using the Horizontal Lift Reactor (HoLiR) process (Kralisch et al., 2009). After biosynthesis, the endless fleece was cut into pieces (100 mm 100 mm). All types of BNC were purified as described previously (Kralisch et al., 2009), sterilized by autoclaving (121 C, 20 min, 2 bar), and stored at 4 C until use. Compressed US BNC (USpress) was obtained by compression of the US7 fleeces between the plates of a hydraulic press (LaboPress P200S, Vogt Maschinenbau, Berlin, Germany) at 3 bar and 20 C for 3 min to form fleeces with a final thickness of about 3 mm. All BNC fleeces were characterized regarding their dimensions (edge length or diameter, height, and weight) as described before (Müller et al., 2013). For calculation of surface area and volume, the geometrical formulas of a circular cylinder or a cuboid were used (Müller et al., 2013). 2.2. Quantification of octenidine Octenidine was purchased from Schuelke & Mayr (Norderstedt, Germany) as a stock solution containing 0.5% octenidine. For quantification of the octenidine concentrations ultraviolet and visible (UV/vis) spectra of octenidine between 200–800 nm and calibration curves (3.4–11.1 mg/mL) were recorded in 0.1 M phosphate buffered solution (PBS) pH 7.4 (Carl Roth, Karlsruhe, Germany) using the Beckman DU 640 spectrophotometer (Beckman Instruments Inc., Fullerton, CA, USA). The time and storage temperature dependent stability of octenidine solutions was investigated by UV/vis measurements after storage at 20 C, 4 C, 20 C, and 32 C over 1, 2, 3, 4, 7, 9, 11, and 16 days and after autoclaving (121 C, 2 bar, 15 min). All experiments were run in triplicates and repeated once. 2.3. Loading and release experiments BNC fleeces produced in 24-well plates (LS samples) were incubated under submersed conditions in 10.0 mL 0.5% octenidine solution at 20 C on a temperature-controlled orbital shaker (KS 4000 ic control, IKA1-Werke, Staufen, Germany) at 70 rpm for 48 h. Fleeces loaded only with PBS were used as control. The drug loading is equal to the difference of the octenidine concentration of the loading solution before and after the loading process. To determine the octenidine release from BNC fleeces at 32 C, loaded BNC was removed from the loading solution and transferred into 20.0 mL PBS pH 7.4 as release medium. Samples were incubated
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
under shaking (200 rpm) and after certain time intervals (0, 0.5, 1, 2, 3, 4, 5, 6, 8, 24, 30, 48, 72, and 96 h), 0.5 mL of the release medium were withdrawn for octenidine quantification. The sampling volume withdrawn was not replaced. Calculations of the total amount of released antiseptic were based on the actual remaining volume of medium at each time point. Using US BNC fleeces from the HoLiR process, volumes of the loading solutions (250.0 mL) and release media (350.0 mL) as well as agitation rate (70 rpm) during release had to be adapted, whereas all other conditions were kept constant. UV/Vis spectrophotometry was used for quantification of octenidine as described above. Results were expressed as cumulative percentage release of the antiseptic against time. BNC samples were characterized regarding their physical dimensions before and after 48 h loading as described above. All experiments were performed in triplicates and repeated once. The semi-empirical Ritger–Peppas equation (Mt/M1 = ktn, power law) was used for modeling of the release data, where Mt and M1 are the cumulative amounts of released octenidine at time t as well as at infinite time. M1 corresponds to the amount of octenidine in BNC after 48 h loading. The logarithmic percentage cumulative release between 5–60% log (Mt/M1 100) was plotted against the logarithmic release time log (t). A linear regression line was fitted to the data, and the slope (n, diffusional exponent) was used as indicator for the release mechanism. Structural and geometrical parameters of the hydrogel were described by the kinetic constant k (Ritger and Peppas, 1987). 2.4. Mechanical characterization The Universal Testing Machine TIRAtest 2710 (Tira, Schalkau, Germany) with a load cell of 1 kN (compression tests) or 0.1 kN (tensile tests) was used for mechanical characterization at 20 C. For compression studies samples were fixed in a compression chamber (inner diameter: 21.5 mm). For each type of BNC at least seven samples were used. Loaded samples and negative controls (in purified water) were compressed (10 mm/min) in vertical direction to the fleece surface until 50% of the original sample thickness were reached. Compressive strength (s C) was calculated by F/S0, where F (N) is the instantaneous force and S0 (mm2) the original cross-sectional area of the BNC. For tensile tests, at least seven dumbell-shaped loaded fleeces and negative controls were fixed in screw clamps and stretched with a deformation speed of 50 mm/min until rupture of samples. Tensile strength (s TS) was calculated by F/A0, where F (N) is the instantaneous force and A0 (mm2) represents the original cross-sectional area. The corresponding strain at rupture (eT) was calculated by Dl/l0, where Dl (mm) represents the change of the length during stretching and l0 (mm) is the original length. 2.5. Scanning electron microscopy (SEM) For SEM analysis samples were lyophilized for 48 h at a pressure of 0.630 mbar and a condenser temperature of 85 C in an ALPHA 2–4 LSC freeze dryer (Martin Christ Gefriertrocknungsanlagen, Osterode, Germany). Cross-sections of freeze dried samples were sputter-coated with gold (Sputter Coater S150B, Edwards, Crawley, UK). The SEM investigations were performed at an acceleration voltage of 15 kV in vacuo (S440i, Leica Microsystems, Wetzlar, Germany). Images of the cross-sections were taken at a working distance of 10 mm. 2.6. Preparation of extracts for biological studies For extract preparation (according to DIN EN ISO 10,993-12), 1 g loaded BNC was used. All further preparation steps were carried
47
out as described for the preparation of a wound dressing extract (Wiegand et al., 2009). Dulbecco’s modified Eagle’s medium (DMEM, Promocell, Heidelberg, Germany) or Caso-bouillon (prepared from peptone and “Lab-Lemco” powder, Oxoid, Basingstoke, UK) were used as extraction media. The prepared extract is referred to as “original extract” (extraction ratio 1 g/50 mL). 2.7. Biocompatibility tests in human keratinocyte cultures Human HaCaT keratinocytes (kind gift from Prof. Dr. N.E. Fusenig, German Cancer Research Center, Heidelberg, Germany) were cultured in DMEM supplemented with 10% fetal calf serum (PromoCell, Germany) and 1% antibiotic/antimycotic solution consisting of 10,000 U/mL penicillin, 10,000 mg/mL streptomycin and 25 mg/mL amphotericin (Gibco Life Technologies, Darmstadt, Germany) for 5–7 days, harvested and seeded into 96-well plates as described previously (Wiegand et al., 2011). After 48 h, culture medium was replaced by the original extract, its dilutions (1:2, 1:4, 1:10, 1:100, 1:200, and 1:1000) or dilutions of the octenidine stock solution with concentrations from 0.5 to 500 mg/mL for further incubation (1, 24, 48 h). Untreated and Triton-X 100 treated cells (Merck Millipore, Schwalbach, Germany) served as negative and positive controls, respectively. Cell proliferation was determined using a luminometric adenosine triphosphate (ATP) assay (ATPliteTM-M assay, Packard Bioscience BV, Groningen, The Netherlands) as recommended by the manufacturer’s protocol. Luminescence was measured using a LUMIstar Galaxy (BMG Labtech, Ortenberg, Germany). A calibration curve was used for calculation of ATP concentration, and the number of the HaCaT cells was expressed as percentage of the cellular ATP content of the control cells (Wiegand et al., 2011). Measurements were performed eightfold and repeated once.
2.8. Microplate laser nephelometry (MLN) and direct contact test for determination of antimicrobial activity Staphylococcus aureus (S. aureus, ATCC 6538, DSMZ, Germany) was used as Gram-positive model germ for infected wounds. Inoculation of Caso-bouillon (20 mL) with S. aureus, incubation (37 C, 24 h) and preparation of a cell suspension (app. 5 103 colony forming units/mL) was carried out according to former reports (Wiegand et al., 2012). Octenidine stock solution (0.5%) and the original extracts were diluted using Caso-bouillon. Aliquots (100 mL) were pipetted in triplicate into the wells of a 96-well plate (Greiner Bio-One, Frickenhausen, Germany), and 100 mL cell suspension were added to each well. Pure Caso-bouillon was used as blank. Microplates were incubated at 37 C for 24 h in the microplate luminometer (NEPHELOstar Galaxy, BMG Labtech, Germany). Measurements were carried out automatically every hour (Wiegand et al., 2012). Experiments were run as quadruplicates and repeated once. For the direct contact test (according to JIS L 1902:2002), Casobouillon was inoculated with 1–2 colonies of S. aureus and cultivated at 37 C for 24 h under aerobic conditions. Loaded BNC was incubated in 400 mL bacterial suspension at 37 C for 24 h. A polyester material (kindly provided by the TITK, Thuringian Institute of Textile and Plastics Research, Rudolstadt, Germany) was used as growth control. Experiments were run in triplicates and repeated once. Extraction of incubated samples and germ number determination was performed as described before (Wiegand et al., 2011). Colony forming units per milliliter (cfu/ mL) were calculated, and growth reduction was compared to the starting value (log growth reduction = log cfu(negative control)(24 h) log cfu(sample)(24 h)) (Wiegand et al., 2011). A logarithmic reduction 1 and 3 represent a significant and values >3 a strong, antimicrobial activity. 2.9. Storage stability testing An accelerated stability study for drug products packaged in impermeable containers was conducted according to ICH Q1A (R2) (2003), using LS samples that were loaded under sterile conditions as described above. Loaded BNC fleeces were packed in an aluminum compound foil consisting of polyethylene terephthalate (PET), aluminum (Al) and polyethylene (PE) (Tesseraux, Buerstadt, Germany), sealed by a welding seam to prevent evaporation and stored at 40 C in an incubator (Heraeus kelvitron1 B6200, Kendro, Langenselbold, Germany). Loading and packaging were performed under aseptic conditions. After 0, 3 and 6 months, release characteristics, antimicrobial activity and cytotoxicity of loaded samples were investigated as described in the sections above. The experiments were run in four independent experiments.
Fig. 1. (A) Dimensions of the LS samples expressed as mean SD of n = 20 samples. Photographs of LS BNC samples cultivated under static conditions in 24-well plates before (B) and after (C) 48 h octenidine loading.
2.10. Data analysis and statistics The half maximal inhibitory concentration (IC50) of pure octenidine was calculated for the evaluation of antibacterial activity. For the respective octenidine concentration tested the AUC (area under the curve) was determined for each growth curve over 24 h and calculated as percentage of the untreated control. This was used to generate a dose-response-curve for octenidine vs. S. aureus from which the IC50 was calculated using a logistic fit function (y = A2 + (A1-A2)/(1 + (x/x0) ^p); A1: upper limit, A2: lower limit, x0: IC50, p: slope of the curve; Origin 7.5). For the assessment of cytotoxic effects, the LC50 value (half maximal lethal concentration for cells) for octenidine was acquired correspondingly. Dose-response curves were obtained from the cytotoxicity measurement results at 24 h. The LC50 was determined using the following logistic fit function (y = A2 + (A1-A2)/(1 + (x/x0) ^p); A1: upper limit, A2: lower limit, x0:LC50, p: slope of the curve; Origin 7.5). Statistically significant differences (p-values 0.05) were determined by ANOVA one way analysis of variance (Microsoft Excel1). 3. Results and discussion 3.1. Validation of the octenidine quantification The quantification of octenidine by UV/vis spectrophotometry (Papen-Botterhuis et al., 2009) was validated with regard to linearity, accuracy and stability. Octenidine showed an absorption maximum at 281 nm in purified water and in PBS. The detection limit was determined as 0.1 mg/mL octenidine with a corresponding UV absorbance of 0.007 0.001. UV/vis spectrophotometric measurements confirmed the stability of octenidine during loading, release and under storage conditions at 20, 4, 20, and 32 C without any loss of drug (data not shown). Octenidine stability during the autoclaving process at 121 C and 2 bar for 15 min was demonstrated by a recovery of 99.0 0.8%. This is in line with observations by Harke (1989) who demonstrated that octenidine could be sterilized in aqueous solution up to 130 C without losing its integrity or decreasing its efficacy. Due to its bispyridinamine structure (Bailey et al., 1984), octenidine differs from other antiseptics such as quaternary ammonium compounds or bisguanidines. Since the molecule does not contain amide or ester structures, octenidine was shown to be stable under different physical and chemical conditions and not to be prone to hydrolysis at pH 1.6–12.2 (Harke, 1989). Nevertheless, in certain solutions like
Ringers’solution stability issues of octenidine formulations may occur as described in the European patent EP2311456A1 (Baur et al., 2011).
3.2. Octenidine loading and release of lab-scale BNC produced under static cultivation conditions LS BNC samples produced by K. xylinus under static cultivation conditions formed mechanically stable mat like hydrogels that were harvested at the interface between air and culture medium. As shown in Fig. 1(A), the round shaped LS samples (Fig. 1(B)) of about 16 mm diameter demonstrated highly reproducible sample dimensions with a height of about 7 mm and weights of about 1.5 g. Surface area and volume were calculated as 7.6 cm2 and 1.4 cm3 (Fig. 1(A)), respectively. The morphological characterization of the LS samples by SEM revealed the typical multilayered BNC structure with a dense top layer (Wesarg et al., 2012), a middle layer of nanofibrillar architecture with interconnected pores and tunnels, and residues of the prepolymer at the bottom (Klemm et al., 2001). Fig. 2(A) shows the cross-section of the middle layer of a LS BNC sample that is characterized by a nanodimensional network structure with a large surface area built up by single, randomly arranged BNC nanofibers. Active compounds could be loaded into the hydropolymer BNC either during polymer formation or afterwards in a post synthesis modification (Trovatti et al., 2011; Wesarg et al., 2012). For the loading of the BNC fleeces with octenidine, the post synthesis technique was selected, since it has the advantage of mild conditions, thus preventing drug damage during biosynthesis and purification of the BNC fleeces (Lin and Anseth, 2009). Additionally, octenidine as an antimicrobial active agent could impair the growth of the BNC producing bacteria (Yamanaka et al., 2000). After 48 h loading by immersion in an aqueous solution of octenidine at ambient temperature, the LS fleeces were found to be homogeneous without visible drug aggregates on the surface or any other macroscopically visual changes (Fig. 1(C)). The weight of the LS samples was found to be increased by 6.8 2.1% during the loading process (Fig. 3(A)) that corresponded to an amount of 4.2 0.1 mg octenidine/g loaded BNC (Fig. 3(B)). The incorporation of drugs into the BNC fleece was suggested to be accomplished by diffusion and capillary forces due to the large hydrophilic surface area (White and Brown, 1989). The increase of BNC weight during loading suggested that the water in the pores of the network was
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
49
Fig. 2. SEM images of BNC cross-sections of LS (A), US7 (B), and US4 (C) samples. The cross-section of the upper sample region of a USpress sample (SEM picture taken at 3.1 mm sample height) is shown in (D) while (E) represents the cross-section of the middle region (SEM picture taken at 2 mm sample height) (magnification 3000x, work distance 10 mm).
partially replaced by the antiseptic drug. In scanning electron microscopy, no relevant changes of the BNC network due to the interaction with octenidine (data not shown) were observed. The release of octenidine was investigated at skin temperature (32 C) over 96 h using an agitated release model (orbital shaker)
which is an established method for the testing of active wound dressings and has been described previously for investigating the release behavior of several drugs loaded into BNC hydropolymers (Wilhelms et al., 2007). Fig. 4 shows the cumulative release curve of octenidine loaded LS BNC plotted against time. The samples
Fig. 3. Loading of LS and US BNC fleeces (mean SD, n = 3). Increase of BNC weight (A), amount of loaded octenidine (mg)/g loaded BNC (B), and correlation between octenidine uptake and weight increase (C) after 48 h loading with 5 mg/mL octenidine under submersed conditions at 20 C.
50
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
Fig. 4. Cumulative octenidine release in PBS (pH 7.4) of BNC fleeces at 32 C over 96 h. The amount of released octenidine from loaded LS and US samples was quantified at specific time points by UV-spectrophotometry at 281 nm. Data are expressed as mean SD of 3 samples.
exhibited exponential release profiles with an initial burst (82.7 2.6%) within the first 8 h, reaching equilibrium conditions after 24 h and a total release of 91.8 2.0%. These release conditions favor the application as wound dressing since they provide a rapid increase of drug concentration in the wound to reach therapeutically relevant threshold amounts followed by a slower release to maintain the drug level (Wilhelms et al., 2007). 3.3. Comparison of lab-scale BNC fleeces with BNC produced under semi-continuous cultivation conditions 3.3.1. Macroscopic and microscopic morphology of the fleeces The transferability of the drug loading and release data of samples obtained from the static process to a large scaled, semicontinuous production process was tested next. In order to define the fleece parameters which might influence these results, LS samples were compared to US BNC. US samples of two different heights (7 mm and 4 mm) were square cut (US7: Fig. 5(A), US4: Fig. 5(B) with comparable edge lengths and surface areas of about 100 mm and 220 cm2 (Table 1), thus representing typical dimensions of commercial wound dressings (Hirsch et al., 2011). Additionally, the US7 fleeces were semi-dried by removal of water through a pressing step (USpress), thus including partially dried wound dressings in our studies which are used in the clinical practice for highly exudating wounds. By pressing, the initial mass
Fig.
of US7 samples was reduced from 82 g to 32 g, which corresponds to a removal of 50 g water. USpress samples (Fig. 5(C)) were characterized by a homogeneous and smooth surface, a macroscopical more compact fleece morphology, an edge length of 100 mm 100 mm, and a height of 3 mm (Table 1). Volume and weight of all US samples were found to be strongly depending on the thickness and the water content of the samples with a decrease from US7 > US4 > USpress (Table 1). The reduction of thickness had an only small impact on the calculated total fleece area due to a small impact of height compared to the edge length of the US fleeces (Table 1). SEM investigations of non-compressed BNC materials produced by the HoLiR process revealed that the network structure was neither changed by the up-scale process compared to LS fleeces obtained under static cultivation in small vessels (Kralisch et al., 2009) nor by the variation of fleece thickness. Cross-sections of US7 (Fig. 2(B)) and US4 (Fig. 2(C)) fleeces exhibited the same hierarchical network structure as LS fleeces, with randomly distributed BNC fibers and interconnected pores (Fig. 2(A)). Distinct changes could be observed for the USpress samples. After top-down compression, they were characterized in the upper and lower region (Fig. 2(D)) by dense layers of agglomerated fibers, partially connected by single BNC fibers which formed a more open network with slightly increased pore sizes and volumes. The network structure of the USpress middle region seemed to be less affected by compression (Fig. 2(E)), with a network found to be more horizontally oriented than those of US7 samples. While the entire surface of LS samples is exclusively defined by bacteria during biosynthesis, the cutting of US fleeces led to a structure with more open pores due to the cutting process (Wesarg et al., 2012). 3.3.2. Loading characteristics of US fleeces All parameters of the loading procedure were transferable from the lab-scale to the up-scale procedure with one exception: the volume of the loading solution had to be increased to ensure that the US fleeces were totally submersed in the loading solution. In preliminary studies, it could be shown that the ratio of medium volume to BNC weight had no effect on the amount of loaded drug as long as the loading solution is present in excess (data not shown). After 48 h of octenidine loading, none of the US fleeces showed any macroscopical changes (US7: Fig. 5(D), US4: Fig. 5(E), USpress: Fig. 5(F)). Measured immediately at ambient temperature and humidity after removal from the loading solution, US4 samples increased their weight by 8.2 0.7% whereas US7 demonstrated a significantly lower weight increase of 2.3 1.7% (Fig. 3(A)). In
5. Photographs of US7 (A, D), US4 (B, E) and USpress (C, F) BNC fleeces before (upper row, A–C) and after (lower row, D–F) 48 h loading with 5 mg/mL octenidine.
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
Table 1 Comparison of the physical dimensions edge length, height, weight, volume and surface area of the BNC fleeces produced in an up-scale process by the semicontinuous HoLiR technique. Ten samples of each BNC type were characterized, and results expressed as mean SD. Edge length (mm) Height (mm) Weight (g) Surface area (cm2) Volume (cm3)
US4 100.4 4.5 48.7 218.9 45.0
1.7 0.3 1.8 1.6 2.5
US7 98.6 7.6 80.8 224.2 73.8
2.1 0.4 1.8 3.4 3.7
USpress 101.4 1.9 3.1 0.1 31.4 2.4 216.7 3.6 31.0 0.5
contrast, the absorption of the compressed USpress fleeces was characterized by a higher weight gain of 57.5 8.0%. The calculated amount of octenidine after loading was increased in the order US7 < US4 < USpress (Fig. 3(B)). Also, the uptake of octenidine per gram unloaded BNC with US7 < US4 0.5% for all tested materials, independent of the storage time. This could be confirmed for storage times up to 6 months, where no loss of antimicrobial effectivity could be detected. The correlation of cytotoxic potential and antimicrobial efficiency revealed that antibacterial effects could be obtained at concentrations well below the values for cell damage effects. This is in accordance with Müller and Kramer (2008) who have shown that octenidine was more toxic to S. aureus than to murine fibroblasts under identical test conditions, resulting in a good biocompatibility with concomitant antibacterial activity. Notably, this also applies to octenidine loaded BNC where we could show that antibacterial activity is more pronounced than cytotoxicity against HaCaT cells. The antimicrobial efficacy of octenidine loaded BNC was further investigated according to JIS L 1902 (Fig. 9(C)) which describes a direct contact test. The results correlated well with the data of the nephelometric measurements. After 24 h of incubation, all
octenidine loaded BNC samples exhibited log reduction values 6 representing a strong antibacterial efficacy. This was also found after storage for 6 months. For comparison, the polyester growth control had no impact on bacterial growth. From the microbiological data it could be concluded that neither the interaction with BNC nor the storage over 6 months impairs the activity of octenidine. For comparison, Hirsch et al. (2011) reported earlier for combinations of octenidine with commercially available wound dressings that only 54% of all tested combinations were found to be therapeutically active. Because of its cationic character, octenidine seems to interact preferentially with anionic structures in these dressings resulting in a reduced antibacterial activity. 4. Conclusion In the present study, the antiseptic octenidine was incorporated in bacterial nanocellulose with the intension to develop a ready-touse system for wound treatment. The combination of both components leads to a wound dressing with advanced healing and superior material properties, combined with an efficient infection control and minimized unwanted side effects. Octenidine loaded BNC demonstrated release profiles that were comparable to already marketed products. They were stable for up to 6 months storage without losing their favorable physicochemical and biological characteristics. A study to scale up the fleeces to sizes typically used in the clinical praxis revealed parameters that can be
S. Moritz et al. / International Journal of Pharmaceutics 471 (2014) 45–55
used to selectively control the time-release curves of the BNC. Further investigations regarding a broader spectrum of microorganisms and wound studies in vivo will follow. Our findings suggest that the octenidine loaded BNC exhibit a tremendous potential as wound dressing with controlled drug delivery. Acknowledgements The authors would like to thank Ramona Brabetz and Elena Pfaff for their excellent technical assistance. All authors (NanocellCare) gratefully acknowledge the Thuringian Ministry of Education, Science and Culture and the European Fund for Regional Development (B714-10032). S.M. is grateful to the FAZIT Foundation, Gemeinnützige Verlagsgesellschaft mbH for financial support. References Al-Doori, Z., Goroncy-Bermes, P., Gemmell, C.G., Morrison, D., 2007. Low-level exposure of MRSA to octenidine dihydrochloride does not select for resistance. J. Antimicrob. Chemother. 59, 1280–1282. Amin, M.C.I.M., Ahmad, N., Halib, N., Ahmad, I., 2012. Synthesis and characterization of thermo- and pH-responsive bacterial cellulose/acrylic acid hydrogels for drug delivery. Carbohydr. Polym. 88, 465–473. Bailey, D.M., DeGrazia, C.G., Hoff, S.J., Schulenberg, P.L., O‘Connor, J.R., Paris, D.A., Slee, A.M., 1984. Bispyridinamines: a new class of topical antimicrobial agents as inhibitors of dental plaque. J. Med. Chem. 27, 1457–1464. Baur B., Behrends S., Siebert J., Dettmann A., Golombiewski M., 2011. EP2311456A1: Wound and mucous membrane antiseptic based on bispyridiniumalkanes. Schuelke & Mayr GmbH, 12. Berndt, S., Wesarg, F., Wiegand, C., Kralisch, D., Müller, F., 2013. Antimicrobial porous hybrids consisting of bacterial nanocellulose and silver nanoparticles. Cellulose 20, 771–783. Bielecki, S., Kalinowska, H., Krystynowicz, A., Kubiak, K., Kolodziejczyk, M., de Groeve, M., 2013. Wound dressings and cosmetic materials from bacterial nanocellulose. In: Gama, M., Gatenholm, P., Klemm, D. (Eds.), Bacterial NanoCellulose: A Sophisticated Multifunctional Material. Taylor and Francis (CRC Press), Boka Raton, pp. 157–174. Boateng, J., Matthews, K., Stevens, H., Eccleston, G., 2008. Wound healing dressings and drug delivery systems: a review. J. Pharm. Sci. 97, 2892–2923. Briese, V., Neumann, G., Waldschlaeger, J., May, T.W., Siebert, J., Gerber, B., 2011. Efficacy and tolerability of a local acting antiseptic agent in the treatment of vaginal dysbiosis during pregnancy. Arch. Gynecol. Obstet. 283, 585–590. Chiaoprakobkij, N., Sanchavanakit, N., Subbalekha, K., Pavasant, P., Phisalaphong, M., 2011. Characterization and biocompatibility of bacterial cellulose/alginate composite sponges with human keratinocytes and gingival fibroblasts. Carbohydr. Polym. 85, 548–553. Dissemond, J., Gerber, V., Kramer, A., Riepe, G., Strohal, R., Vasel-Biergans, A., Eberlein, T., 2010. A practice-oriented recommendation for treatment of critically colonised and locally infected wounds using polihexanide. J. Tissue Viabil. 19, 106–115. Gilbert, P., Moore, L.E., 2005. Cationic antiseptics: diversity of action under a common epithet. J. Appl. Microbiol. 99, 703–715. Goldstein, E.J., Citron, D.M., Nesbit, C.A., 1996. Diabetic foot infections: bacteriology and activity of 10 oral antimicrobial agents against bacteria isolated from consecutive cases. Diabetes Care 19, 638–641. Harke, H.P., 1989. Octenidine dihydrochloride, properties of a new antimicrobial agent. Zentralbl. Hyg. Umweltmed. 188, 188–193. Hirsch, T., Jacobsen, F., Rittig, A., Goertz, O., Niederbichler, A., Steinau, H.U., Seipp, H. M., Steinstraesser, L., 2009. A comparative in vitro study of cell toxicity of clinically used antiseptics. Hautarzt 60, 984–991. Hirsch, T., Limoochi-Deli, S., Lahmer, A., Jacobsen, F., Goertz, O., Steinau, H.-U., Seipp, H.-M., Steinstraesser, L., 2011. Antimicrobial activity of clinically used antiseptics and wound irrigating agents in combination with wound dressings. Plast. Reconstr. Surg. 127, 1539–1545. Huang, L., Chen, X., Nguyen, T.X., Tang, H., Zhang, L., Yang, G., 2013. Nano-cellulose 3D-networks as controlled-release drug carriers. J. Mater. Chem. B 1, 2976– 2984. Hübner, N.O., Siebert, J., Kramer, A., 2010. Octenidine dihydrochloride, a modern antiseptic for skin, mucous membranes and wounds. Skin Pharmacol. Physiol. 23, 244–258. ICHQ1A(R2), 2003. Stability testing of new drug substances and products. ICH Q8(R2), 2009. Pharmaceutical development. Klemm, D., Schumann, D., Udhardt, U., Marsch, S., 2001. Bacterial synthesized cellulose – artificial blood vessels for microsurgery. Prog. Polym. Sci. 26, 1561– 1603. Koburger, T., Hübner, N.O., Braun, M., Siebert, J., Kramer, A., 2010. Standardized comparison of antiseptic efficacy of triclosan, PVP-iodine, octenidine dihydrochloride, polyhexanide and chlorhexidine digluconate. J. Antimicrob. Chemother. 65, 1712–1719.
55
Kralisch, D., Hessler, N., Klemm, D., Erdmann, R., Schmidt, W., 2009. White biotechnology for cellulose manufacturing – the HoLiR concept. Biotechnol. Bioeng. 105, 740–747. Kramer, A., Müller, G., Assadian, O., 2006. Indications and agent selection for the antiseptic therapy of secondarily healing wounds. GMS Krankenhaushyg. Interdiszip. 1, 1–11. Lin, C.-C., Anseth, K.S., 2009. PEG hydrogels for the controlled release of biomolecules in regenerative medicine. Pharm. Res. 26, 631–643. Luan, J., Wu, J., Zheng, Y., Song, W., Wang, G., Guo, J., Ding, X., 2012. Impregnation of silver sulfadiazine into bacterial cellulose for antimicrobial and biocompatible wound dressing. Biomed. Mater. 7, 065006. Müller, A., Ni, Z., Hessler, N., Wesarg, F., Müller, F.A., Kralisch, D., Fischer, D., 2013. The biopolymer bacterial nanocellulose as drug delivery system: investigation of drug loading and release using the model protein albumin. J. Pharm. Sci. 102, 579–592. Müller, G., Kramer, A., 2008. Biocompatibility index of antiseptic agents by parallel assessment of antimicrobial activity and cellular cytotoxicity. J. Antimicrob. Chemother. 61, 1281–1287. Müller, G., Langer, J., Siebert, J., Kramer, A., 2014. Residual antimicrobial effect of chlorhexidine digluconate and octenidine dihydrochloride on reconstructed human epidermis. Skin Pharmacol. Physiol. 27, 1–8. Nakayama, A., Kakugo, A., Gong, J.P., Osada, Y., Takai, M., Erata, T., Kawano, S., 2004. High mechanical strength double-network hydrogel with bacterial cellulose. Adv. Funct. Mater. 14, 1124–1128. Papen-Botterhuis, N.E., ten Cate, A.T., Kolijn, J., Eversdijk, J., 2009. Enzyme-triggered release of antimicrobials from gelatin microcapsules. XVIIth International Conference on BioencapsulationGroningen, The Netherlands. Pitten, F.A., Werner, H.P., Kramer, A., 2003. A standardized test to assess the impact of different organic challenges on the antimicrobial activity of antiseptics. J. Hosp. Infect. 55, 108–115. Reynolds, T.D., Mitchell, S.A., Balwinski, K.M., 2002. Investigation of the effect of tablet surface area/volume on drug release from hydroxypropylmethylcellulose controlled-release matrix tablets. Drug Dev. Ind. Pharm. 28, 457–466. Ritger, P.L., Peppas, N.A., 1987. A simple equation for description of solute release II. Fickian and anomalous release from swellable devices. J. Control. Release 5, 37–42. Sanchavanakit, N., Sangrungraungroj, W., Kaomongkolgit, R., Banaprasert, T., Pavasant, P., Phisalaphong, M., 2006. Growth of human keratinocytes and fibroblasts on bacterial cellulose film. Biotechnol. Prog. 22, 1194–1199. Sedlock, D.M., Bailey, D.M., 1985. Microbicidal activity of octenidine hydrochloride, a new alkanediylbis[pyridine] germicidal agent. Antimicrob. Agents Chemother. 28, 786–790. Silva, N.H.C.S., Rodrigues, A.F., Almeida, I.F., Costa, P.C., Rosado, C., Neto, C.P., Silvestre, A.J.D., Freire, C.S.R., 2014. Bacterial cellulose membranes as transdermal delivery systems for diclofenac: in vitro dissolution and permeation studies. Carbohyd. Polym. 106, 264–269. Slee, A.M., O‘Connor, J.R., 1983. In vitro antiplaque activity of octenidine dihydrochloride (WIN 41464-2) against preformed plaques of selected oral plaque-forming microorganisms. Antimicrob. Agents Chemother. 23, 379–384. Torres, A.J., Zhu, C., Shuler, M.L., Pannullo, S., 2011. Paclitaxel delivery to brain tumors from hydrogels: a computational study. Biotechnol. Prog. 27, 1478–1487. Trovatti, E., Silva, N.H.C.S., Duarte, I.F., Rosado, C.F., Almeida, I.F., Costa, P., Freire, C.S. R., Silvestre, A.J.D., Pascoal, N.C., 2011. Biocellulose membranes as supports for dermal release of lidocaine. Biomacromolecules 12, 4162–4168. Ul-Islam, M., Khan, T., Khattak, W., Park, J., 2013. Bacterial cellulose-MMTs nanoreinforced composite films: novel wound dressing material with antibacterial properties. Cellulose 20, 589–596. Vanscheidt, W., Harding, K., Teot, L., Siebert, J., 2012. Effectiveness and tissue compatibility of a 12-week treatment of chronic venous leg ulcers with an octenidine based antiseptic – a randomized, double-blind controlled study. Int. Wound J. 9, 316–323. Wei, B., Yang, G., Hong, F., 2011. Preparation and evaluation of a kind of bacterial cellulose dry films with antibacterial properties. Carbohydr. Polym. 84, 533–538. Wesarg, F., Schlott, F., Grabow, J., Kurland, H.-D., Heßler, N., Kralisch, D., Müller, F.A., 2012. In situ synthesis of photocatalytically active hybrids consisting of bacterial nanocellulose and anatase nanoparticles. Langmuir 28, 13518–13525. White, D.G., Brown Jr., R.M., 1989. Prospects for the commercialization of the biosynthesis of microbial cellulose. In: Schürch, C. (Ed.), Cellulose and Wood: Chemistry and Technology. Wiley, New York, pp. 573–590. Wiegand, C., Abel, M., Ruth, P., Hipler, U.-C., 2009. HaCaT keratinocytes in co-culture with Staphylococcus aureus can be protected from bacterial damage by polihexanide. Wound Repair Regen. 17, 730–738. Wiegand, C., Abel, M., Ruth, P., Hipler, U.C., 2011. Superabsorbent polymercontaining wound dressings have a beneficial effect on wound healing by reducing PMN elastase concentration and inhibiting microbial growth. J. Mater. Sci.: Mater. Med. 22, 2583–2590. Wiegand, C., Abel, M., Ruth, P., Hipler, U.C., 2012. Analysis of the adaptation capacity of Staphylococcus aureus to commonly used antiseptics by microplate laser nephelometry. Skin Pharmacol. Physiol. 25, 288–297. Wiegand, C., Hipler, U.-C., 2010. Polymer-based biomaterials as dressings for chronic stagnating wounds. Macromol. Symp. 294, 1–13. Wilhelms, T., Schulze, D., Alupeil, C., I, Rohrer, C., Abel, M., Wiegand, C., Hipler, U.C., 2007. Release of polyhexamethylene biguanide hydrochloride (PHMB) from a hydroballanced cellulose wound dressing with PHMB. 17th Conference of the European Wound Management AssociationGlasgow. Yamanaka, S., Ishihara, M., Sugiyama, J., 2000. Structural modification of bacterial cellulose. Cellulose 7, 213–225.
