ANTIMICROBIAL COMPOSITIONS

Abstract

The invention provides antimicrobial compositions comprising charged cellulose nanofibrils dispersed in an aqueous solution having a dissolved oxygen content of at least 20 mg/L, preferably from 20 to 100 mg/L. The cellulose nanofibrils may have an increased surface charge due to their carboxylic acid content which contributes to their antimicrobial properties. In particular, the carboxylic acid content may be at least about 1000 μmol/g cellulose, preferably at least about 1400 μmol/g cellulose. The compositions are suitable for use in the treatment of wounds, in particular chronic wounds.

Claims

1. An antimicrobial composition comprising charged cellulose nanofibrils dispersed in an aqueous solution, wherein said solution has a dissolved oxygen content of at least 20 mg/l.

2. A composition as claimed in claim 1, wherein said charged cellulose nanofibrils are present in an amount from 0.1 to 1.0 wt. %, preferably from 0.2 to 0.8 wt. %, e.g. from 0.3 to 0.5 wt. %, based on the total weight of the composition.

3. A composition as claimed in claim 1 or claim 2 which is provided in the form of a liquid or a viscous liquid, preferably having a viscosity in the range from 100 to 9,000 mPa.Math.s, e.g. from 100 to 600 mPa.Math.s when measured using a Brookfield viscometer at 10 rpm, 23° C.

4. A composition as claimed in claim 1 or claim 2 which is provided in the form of a hydrogel, preferably a hydrogel having a viscosity in the range from 10,000 to 20,000 mPa.Math.s when measured using a Brookfield viscometer at 10 rpm, 23° C.

5. A composition as claimed in any one of the preceding claims, wherein said charged cellulose nanofibrils are negatively charged.

6. A composition as claimed in any one of the preceding claims, wherein said charged cellulose nanofibrils are surface-oxidised, preferably TEMPO-oxidised.

7. A composition as claimed in claim 6, wherein the carboxylic acid content of the charged cellulose nanofibrils is in the range from 400 to 1750 μmol/g cellulose, preferably at least about 1000 μmol/g cellulose, e.g. at least about 1400 μmol/g cellulose.

8. A composition as claimed in claim 6 or claim 7, wherein the aldehyde content of the charged cellulose nanofibrils is in the range from 10 to 1700 μmol/g cellulose, preferably from 100 to 400 μmol/g cellulose.

9. A composition as claimed in any one of the preceding claims, wherein the charged cellulose nanofibrils are obtained from wood pulp, preferably softwood pulp, e.g. from Pinus radiata.

10. A composition as claimed in any one of the preceding claims, wherein the average diameter of the cellulose nanofibrils is in the range from 3 to 20 nm and/or wherein the average length of the cellulose nanofibrils is in the range from 5 to 10 μm.

11. A composition as claimed in any one of the preceding claims, wherein said solution has a dissolved oxygen content of from 20 to 100 mg/L oxygen, from 20 to 70 mg/L, from 20 to 60 mg/L, from 25 to 50 mg/L, or from 30 to 40 mg/L.

12. A composition as claimed in any one of the preceding claims which further comprises one or more active substances selected from the group consisting of: antibacterial agents, antifungal agents, antiviral agents, antibiotics, growth factors, cytokines, chemokines, nucleic acids, vitamins, minerals, anaesthetics, anti-inflammatory agents, moisturizers, extracellular matrix proteins, enzymes, stem cells from plants, extracts from eggs and eggshells, botanical extracts, fatty acids, and skin penetration enhancers.

13. A method for the preparation of a composition as claimed in any one of claims 1 to 12, said method comprising the step of combining an aqueous solution having a dissolved oxygen content of at least 20 mg/l with a preparation which contains charged cellulose nanofibrils, preferably wherein said preparation is an aerogel comprising charged cellulose nanofibrils.

14. A method for the preparation of a composition as claimed in any one of claims 1 to 12, said method comprising the following steps: (i) providing a dispersion of said charged cellulose nanofibrils in an aqueous solution; and (ii) oxygenating said dispersion.

15. A method as claimed in claim 14, wherein step (ii) comprises the following steps: introducing a liquid comprising said dispersion into a piping network to form a flow stream; injecting gaseous oxygen into the flow stream to produce a mixture of said liquid and oxygen bubbles; and passing the flowing mixture of liquid and gaseous oxygen bubbles through a venturi which is arranged to dissolve the gas into the liquid passing through the venturi.

16. A method as claimed in any one of claims 13 to 15 which further comprises the step of subjecting the resulting composition to cross-linking whereby to increase its viscosity.

17. A composition as claimed in any one of claims 1 to 12 for use as an antimicrobial agent, preferably for use in inhibiting the growth of at least one wound pathogen.

18. A composition as claimed in claim 17 for use in the treatment of a wound, preferably for use in the treatment of a chronic wound, more preferably a wound harbouring one or more bacteria selected from Bacteroides species, Clostridium species, Pseudomonas species, Enterococcus species, Enterobacteriacea species, Bacillus species, Streptococcus species, and Staphylococcus species, e.g. a wound harbouring Pseudomonas aeruginosa and/or Staphylococcus aureus.

19. A composition for use as claimed in any one of claims 1 to 12 in the prevention or treatment of a bacterial biofilm on a body surface, preferably on an external body surface, e.g. on the skin.

20. A wound covering (e.g. a bandage, gauze, patch or absorptive pad) having incorporated therein a composition as claimed in any one of claims 1 to 12.

21. A wound dressing in the form of a hydrogel comprising charged cellulose nanofibrils, wherein said hydrogel has a dissolved oxygen content of at least 20 mg/l.

22. A wound dressing as claimed in claim 21 which is formed by 3D printing.

23. A kit for use in treating a wound, the kit comprising: (a) a sterilised, sealed container or package containing an antimicrobial composition as claimed in any one of claims 1 to 12; (b) a wound covering, e.g. a wound dressing, bandage, gauze, patch or absorptive pad; and optionally (c) printed instructions for use of the components of the kit in the treatment of a wound.

24. A kit for use in treating a wound, the kit comprising: (a) a sterilised, sealed container or package containing an aerogel comprising charged cellulose nanofibrils; (b) an oxygenated aqueous liquid (e.g. oxygenated water or oxygenated saline) having a dissolved oxygen content of at least 20 mg/l; and optionally (c) printed instructions for mixing of the components whereby to form an oxygenated hydrogel and its use in the treatment of a wound.

Description

[0183] The invention will now be described further with reference to the following non-limiting Examples and the accompanying figures in which:

[0184] FIG. 1 shows the laser profilometry quantification of CNF film roughness in CNFs produced with increasing oxidation. The mean values for each lateral wavelength are given with the standard deviation of the mean (n=10).

[0185] FIG. 2 shows AFM analysis of samples CNF_2.5, CNF_3.8 and CNF_6.0. The relatively thicker nanofibrils in the CNF_2.5 sample are indicated by arrows. The height plots were acquired at the middle of each image, indicated by a dotted line. Calibration and scale bars are given in nanometers. Height and width is measured on a single nanofibril (colored black) from the profile plot.

[0186] FIG. 3 shows the Brookfield viscosity measured at various speeds for CNFs produced with increasing oxidation.

[0187] FIG. 4 shows the antimicrobial effect of CNF gels on P. aeruginosa after 24 hours exposure, correlated to the negative control BHI100, which is set to 100%. The bars represent average and error bars represents SEM. N=5 in all groups.

[0188] FIG. 5 shows the quantification of light transmittance of 3D printed constructs (the mean value is given with the standard deviation, n=4), Target dimensions of the 3D printed constructs were 20 mm×40 mm×2 mm.

[0189] FIG. 6 shows an SEM assessment of freeze-dried constructs. The four columns provide four replicate SEM images for each series. The arrows indicate the printing direction. The right column yields the polar plots showing the main orientation of the surface structure.

[0190] FIG. 7 shows the Brookfield viscosity of 0.2 wt. % CNF and oxygenated CNF for CNF_2.5, CNF_3.8 and CNF_6.0 (Table 1). Data are expressed as average±SEM (n=10).

[0191] FIG. 8 shows an assessment of CNF dispersions with (A) a FiberTester (residual fibres and fines), and (B) a nanoparticle analyser (nano-sized fibres).

[0192] FIG. 9 shows the oxygenation of CNF and quantification of dissolved oxygen (DO). 0.2 wt. % CNF with different oxidation levels (CNF_2.5, CNF_3.8 and CNF_6.0, Table 1) was oxygenated and stored in sealed glass vials at room temperature (22° C.). The DO concentrations were measured at production date and 5 weeks later. Duplicate measurements for oxygenated CNF and singular measurements for CNF. Data are expressed as average±SEM.

[0193] FIG. 10 shows the antimicrobial effect of CNF and oxygenated CNF on P. aeruginosa. Bacterial survival of 0.2 wt. % CNF with different oxidation levels (CNF_2.5, CNF_3.8 and CNF_6.0, Table 1) on P. aeruginosa after 4 or 24 hours. Data are expressed as average±SEM. n=5 in all groups, except for CNF_6.0 4 hours and CNF 24 hours (n=4). BHI100 was used as negative control.

[0194] FIG. 11 shows the antimicrobial effect of CNF and oxygenated CNF on P. aeruginosa and S. aureus. Bacterial survival (Log10 CFU) of 0.2 wt. % CNF with different oxidation levels (CNF_2.5, CNF_3.8 and CNF_6.0, Table 1) on (A) P. aeruginosa and (B) S. aureus after 24 hours exposure. Data are expressed as Log10. N=5 in all groups, except CNF_3.8 in FIG. B (n=4). BHI100 and Prontosan were used as negative control and positive control, respectively.

[0195] FIG. 12 shows an SEM assessment of bacterial biofilms: (A) P. aeruginosa and CNF_6.0; (B) P. aeruginosa and CNF_6.0-Oxygenated; (C) S. aureus and CNF_6.0; and (D) S. aureus and CNF_6.0-Oxygenated.

[0196] FIG. 13 shows the effect of cross-linking oxygenated CNF with CaCl.sub.2. Upper figure: dissolved oxygen (DO) in 0.2 wt. % oxygenated CNFs with and without CaCl.sub.2 (50 mM or 100 mM), N=3. Lower figure: dissolved oxygen (DO) in 0.4 wt. % oxygenated CNFs with and without CaCl.sub.2 (50 mM or 100 mM), N=3 except for “Oxy 0.4% 100 mM CaCl.sub.2” (N=1).

[0197] FIG. 14 shows the Brookfield viscosities of CNFs at 0.2 wt. % and 0.4 wt. % (measured at 10 RPM).

[0198] FIG. 15 shows the dissolved oxygen (DO) content of CNFs injected through a 50 ml needle tip with 18G cannula. Upper figure: 0.2 wt. % CNF. Lower figure: 0.4 wt. % CNF. N=3.

[0199] FIG. 16 shows the antibacterial effect of CNF gels on P. aeruginosa after 24 hours exposure, correlated to the negative control BHI100, which was set to 100%. The bars represent average and the error bars represent standard error of the mean. n=15 in all groups. All samples were significantly different compared to the control (*, p<0.05).

[0200] FIG. 17 shows swimming levels of P. aeruginosa in agar gels containing CNFs (0.6 wt. %). The bars represent average and the error bars represent standard error of the mean, n=3 in all groups (*, p<0.05).

[0201] FIG. 18 shows the antimicrobial effect of CNF_3.8 and oxygenated CNF_3.8, assessed in vivo. Data are expressed as number of CFU. n=5 in all groups. (*) denotes significant difference (p<0.05).

EXAMPLES

Example 1—Preparation of Cellulose Nanofibrils (CNFs) and Characterisation

[0202] Preparation of CNFs:

[0203] Pinus radiata kraft pulp fibers were washed and autoclaved using NaOH as described by Nordli et al. (Carbohydrate Polymer 150, 65-73, 2016). This was performed to reduce the amount of endotoxins (Nordli et al., ACS Applied Bio Materials 2(3), 1107-1118, 2019). CNFs with varying surface chemistry were produced by TEMPO-mediated oxidation, applying three levels of oxidation, i.e. 2.5, 3.8 and 6.0 mmol hypochlorite (NaClO)/g cellulose and defined as CNF_2.5, CNF_3.8 and CNF_6.0, respectively (Saito et al., Biomacromolecules 5(5), 1983-1989, 2004). The CNFs were collected after passing the oxidized cellulose fibres three times through a homogenizer (Rannie 15 type 12.56X homogenizer, operated at 1000 bar pressure).

[0204] Characterisation of CA/Fs:

[0205] The content of carboxylic acid groups was quantified by conductometric titration according to Saito et al. (Biomacromolecules 5(5), 1983-1989, 2004). The content of aldehyde groups was determined based on a spectrophotometric method previously described by Jausovec et al. (Carbohydrate Polymer 116, 74-85, 2015).

[0206] The CNF gels (concentration 0.6 wt. %) were printed on microscopy slides using a Regemat3D printing unit (version 1.0, Regemat3D, Granada, Spain). Solid areas of 10×20 mm were printed, 2 layers, using a nozzle of 0.58 mm and flow 3 mm/s. The gels were allowed to dry at room temperature (23° C.) and 40% relative humidity. A layer of gold was deposited on the printed structures and 10 laser profilometry images (1×1 mm) were acquired with a resolution of 1 μm/pixel. The laser profilometry images were bandpass-filtered and the surface roughness (root-mean-square) was quantified at various lateral wavelengths (Chinga-Carrasco et al., Micron 56, 80-84, 2014).

[0207] Atomic force microscopy (AFM) was performed on the three CNF samples. The samples were analyzed with a Veeco multimode V at room temperature. The AFM tips had a spring constant ˜0.4 N m.sup.−1 (Bruker AFM probes). The assessed local areas were 2×2 μm, with a resolution of 1.95 nm/pixel.

[0208] Viscosity of the CNFs was assessed with a Brookfield viscometer (Brookfield DV2TRV). The assessment was performed using spindle V-73 at a temperature of 23° C.±1° C. and at the following speeds: 0.6, 1, 2, 6 and 10 RPM.

[0209] Results and Discussion:

[0210] The carboxyl and aldehyde contents of the CNF gels and surface roughness of the CNF films are shown in Table 1:

TABLE-US-00001 TABLE 1 NaClO used during Carboxyl Aldehyde oxidation of content of content of CNF film fibers CNF CNF roughness (μmol/g) (μmol/g) (μmol/g) (μm) CNF_2.5 2500 1036 ± 41 351 ± 11.7 0.44 ± 0.02 CNF_3.8 3800 1364 ± 35 326 ± 3.1 0.31 ± 0.06 CNF_6.0 6000 1593 ± 10 223 ± 10 0.16 ± 0.01

[0211] The increase in the amount of NaClO led to an increase in the amount of carboxyl groups. Increasing the amount of carboxyl groups increases the repulsion forces between nanofibrils and this facilitates the production of individualized nanofibrils. This was confirmed by the laser profilometry data. The more oxidized the fibers, the higher the nanofibril yield and the smoother the surface of the CNF films. The relatively high roughness profile of CNF_2.5 is due to a major occurrence of residual micrometer-sized fibers. As the oxidation increases, the roughness decreases (see FIG. 1). This is due to the major fraction of individualized nanofibrils that are obtained (see FIG. 2).

[0212] AFM analysis revealed that the three samples contain nanofibrils (diameters less than 20 nm) (FIG. 2). The AFM analysis is valuable for providing a comparison between the 3 samples and suggests that sample CNF_2.5 contains relatively thicker nanofibrils (FIG. 2, arrows). This observation is also an indication of a structurally inhomogeneous sample, which confirms the roughness analysis (see Table 1).

[0213] The large fraction of individualized nanofibrils of sample CNF_6.0 causes an increase in viscosity of the corresponding gel (see FIG. 3). The three samples show a reduction of viscosity as the speed increases which can be explained from the shear thinning effect. Additionally, the viscosity data indicates that the sample CNF_6.0 has higher viscosity at a given speed, compared to the samples CNF_2.5 and CNF_3.8.

Example 2—Cytotoxicity and Skin Irritation Potential of CNFs

[0214] The cell viability and skin irritation potential of the three CNF samples produced in Example 1 was tested following standardized protocols for assessing medical devices. Six aerogels (20 g/m.sup.2) were prepared from each series. The gels were frozen at −20° C. and lyophilized during 24 h, using a Telstar LyoQuest −83 apparatus.

[0215] In Vitro EpiDerm Skin Irritation Test:

[0216] The skin irritating potential of the samples CNF_2.5, CNF_3.8 and CNF_6.0 was determined by irritation testing according to in vitro skin irritation for medical devices, using the in Vitro EpiDerm™ Skin Irritation Test kit (EPI-200-SIT; MatTek In Vitro Life Science Laboratories, Bratislava, Slovakia) and protocol “In vitro skin irritation test for medical device extracts” v.9.0 final. The test consists of topical exposure of extracts of the test item to the reconstructed human epidermis (RhE) model, followed by a cell viability assay using yellow water-soluble MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide which is metabolically reduced to a blue-violet insoluble formazan in viable cells. The number of viable cells correlates to the colour intensity determined by photometric measurements after dissolving the formazan in alcohol.

[0217] The RhE tissues were pre-incubated in 6-well plates in assay medium overnight (37±1° C., 5±1% CO.sub.2), after which 100 μL of test item extracts or control samples were added. The positive control was 1% sodium dodecyl sulfate solution (SDS, MatTek In Vitro Life Science Laboratories, Bratislava) in saline and sesame oil, and the negative control was Dulbecco's PBS without Ca.sup.2+ and Mg.sup.2+ (GE Healthcare Lifescience HyClone Laboratories, South Logan, Utah). The test item was extracted at 37±1° C. for 72±2 h. After 18 hours of exposure, the tissues were thoroughly rinsed with Dulbecco's PBS without Ca.sup.2+ and Mg.sup.2+ (GE Healthcare Lifescience HyClone Laboratories, South Logan, Utah) and incubated in 24-well plates with 1 mg/mL MTT (MatTek In Vitro Life Science Laboratories), for 3 hours (37±1° C. in 5±1% CO.sub.2), The MTT solution was removed, the tissues were immersed in 2-propanol (2 mL/tissue; MatTek), and the plates were shaken for two hours. The absorbance of the extracted formazan was thereafter measured at 570 nm using a spectrophotometer. Skin irritation potential of the test item is predicted if the remaining relative cell viability is below 50%.

[0218] Cytotoxicity:

[0219] The cytotoxic potential of the samples CNF_2.5, CNF_3.8 and CNF_6.0 was determined by cytotoxicity testing according to ISO 10993-5:2009 Annex C and RISE standard operating procedure SOP KM 11741. The test consists of exposure of extracts of the test item to a sub-confluent monolayer of L929 mouse, followed by a cell viability assay using yellow water-soluble MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromid which is metabolically reduced to a blue-violet insoluble formazan in viable cells. The number of viable cells correlates to the colour intensity determined by photometric measurements after dissolving the formazan in alcohol.

[0220] The test item was extracted at 37±1° C. for 24±2 h in Eagle's Minimum essential medium 1× with Earls balanced salts solution buffered with NaHCO.sub.3 (Gibco Life Technologies) supplemented with nonessential amino acids (Gibco Life Technologies), sodium pyruvate (GE Healthcare HyClone), 5% (v/v) Fetal Bovine Serum (Gibco Life Technologies), 4 mM Stable glutamine (Gibco Life Technologies), 100 IU/mL penicillin and 100 μg/mL streptomycin (GE Healthcare Hyclone) using a ratio of 0.1 g/mL. L929 mouse fibroblasts (ATCC NCTC clone 929: CCL-1) were seeded in a 96-well plate and cultured for at 37±1° C. and 5% CO.sub.2 24±2 h to form a subconfluent monolayer. 100 μL extracts from test item, positive control (Latex rubber, Gammex 91-325, AccuTech Ansell) and negative control (Thermanox Plastic Coverslips, Art no 174934, Thermo Scientific NUNC), as well as blanks (extraction vehicle to serve as a 100% measure of cell viability) were added to 6 replicate wells. The plate was incubated for 24 hours at 37° C. in 5% CO.sub.2. The extracts were removed and 50 μL of MTT solution was added to each well and the cells were incubated for 2 hours at 37° C. in 5% CO.sub.2. The MTT solution was removed and 100 μL of 2-propanol was added to each well. The plate was shaken rapidly until the formazan from the cells was extracted and formed a homogeneous solution. The absorbance was measured at 570 nm (reference wavelength 650 nm) and the viability of cells was calculated. The test item is considered cytotoxic if the cell viability is below 70%.

[0221] Results and Discussion:

[0222] Results are presented in Table 2:

TABLE-US-00002 TABLE 2 Cell viability of Viability of reconstructed human L929 mouse epidermis (RhE) (%) fibroblasts (%) Saline Sesame oil CNF_2.5 103.1 ± 3.6  92.2 ± 4.3  94.4 ± 13.1 CNF_3.8   110 ± 3.2  93.5 ± 0.7 103.9 ± 4.7 CNF_6.0   89 ± 1.7  98 ± 2.2 101.4 ± 9.2 Positive control  1.3 ± 0.2 2.6 ± 0  3.1 ± 0 Negative control   103 ± 2.6  100 ± 1.3   100 ± 1.3

[0223] The results confirm that the CNF samples do not have a cytotoxic potential (fibroblast cell viability was greater than 70%, see Table 2). According to criteria given in “In Vitro EpiDerm™ Skin Irritation Test (EPI-200-SIT)” and protocol “In vitro skin irritation test for medical devices” the materials are classified as non-irritant, i.e. the viability of RhE is above the limit to be considered with potential for skin irritation (De Jong et al., Toxicology in Vitro 50, 439-449, 2018). These findings confirm the development of a safe and biocompatible wound dressing material.

Example 3—Antimicrobial Properties of CNFs

[0224] The antimicrobial effect of the CNF gels produced in Example 1 (CNF_2.5, CNF_3.8 and CNF_6.0) on P. aeruginosa was assessed in vitro.

[0225] Method:

[0226] Overnight culture of P. aeruginosa (ATCC 15692) was set to the final bacterial concentration of 1×10.sup.7 colony forming units (CFU)/mL using optical density (OD) at 600 nm. 10 μL of the prepared bacterial suspension (1×10.sup.7 CFU/mL) were mixed with 500 μl CNF gel and incubated at 37° C. for 24 h. 230 μL of the mixture was suspended in 2 mL phosphate buffer (0.05% Triton X-100 in 0.0375 M phosphate) and diluted five times in ten-fold steps. 50 μL from each dilution was spread on horse blood agar plates and incubated overnight at 37° C. The number of CFUs on the blood agar plates was counted and the number of CFUs in the original tube with gel and bacteria mix was calculated. This was defined as bacterial survival after 24 hours treatment. For each gel, 5 replicates were performed, and as negative control 500 μL brain heard infusion medium diluted 100 times in H.sub.2O (BHI100) was used instead of CNF gel.

[0227] Results and Discussion:

[0228] The results in FIG. 4 confirm a dose-dependent antibacterial effect, i.e. increasing the concentration of CNF from 0.2 to 0.6 wt. % reduced the survival of P. aeruginosa. Additionally, it was found that the antimicrobial properties also depend on the surface charge of the CNF. The results show that increasing the surface charge from 1036 to 1593 μmol/g, reduced the bacterial survival. The reduction of bacterial survival may be attributed to the surface chemistry of the CNFs. Increasing the content of carboxyl groups leads to an increase in the nanofibrillation, i.e. a larger CNF yield is obtained during homogenization. The carboxyl content is expected to increase the repulsion forces between individual nanofibrils in the gel dispersion, thus potentially leading to a charge-dependent distribution of nanofibrils in the liquid medium. The higher the charge density, the more homogenously distributed the nanofibrils and the higher the viscosity (see Table 1 and FIG. 3), and potentially the larger the area the nanofibrils cover on the surface of the bacteria. The aldehyde content may also contribute to cross-link the proteins in the cell wall of the gram-negative bacteria, thus being unable to undertake essential functions. Although not wishing to be bound by theory, we postulate that these characteristics may contribute to limit the bacterial survival and growth.

Example 4—3D Printing of CNFs

[0229] The three CNF grades produced in Example 1 (concentration 0.6 wt. %) were tested for 3D printing.

[0230] Method:

[0231] 3D printing was performed with a Regemat3D printing unit. For each series (CNF_2.5, CNF_3.8 and CNF_6.0), four constructs (dimensions 20 mm×40 mm×2 mm) were printed using a 0.58 printing nozzle. The spaces between the printed tracks were 2 mm×2 mm. The height (2 mm) was composed of 4 printed layers. The flow speed during printing was 3 mm/s. As an additional test of print fidelity, the printing performance of the three CNF grades was assessed. Three replicates (20×40 mm) were printed. The structures were composed of only 1 layer for better assessment of printing performance. The distance between printed tracks was 2 mm. The flow speed was 3 mm/s. Images of the 3D printed structures were acquired immediately after printing with an Epson Perfection V750 PRO scanner, in transmission mode. The applied resolution was 2400 dots per inch. The transmission of light through the optical images was quantified with the ImageJ program (version 1.52h) and is reported as the fraction of light transmitted through the construct, relative to the background.

[0232] The 3D printed structures were frozen at −20° C. and lyophilized over 24 hours using a Telstar LyoQuest-83 apparatus. Scanning electron microscopy (SEM) assessment of the freeze-dried samples was performed with a Hitachi SU3500 Scanning Electron Microscope. Gold coating was performed with an Agar Auto Sputter Coater (Agar Scientific, Essex CM24 8GF United Kingdom). Images were acquired in secondary electron imaging (SEI) mode, using 5 kV and 6 mm acceleration voltage and working distance, respectively.

[0233] Grids were printed with diameter=20 mm, height=1 mm and composed of two layers. The printing nozzle was 0.58 mm. The flow speed during printing was 3 mm/s. The grids were immersed in CaCl.sub.2 (100 mmol) for at least 24 hours before mechanical assessment with a Tl950 Triboindenter from Bruker (former Hysitron). The nano-indentation parameters were: Conical tip; displacement controlled at peak indentation depth of 2000 nm; 0.125 s loading, 0.4 s holding, 0.125 s unloading (total testing time 0.65 s for one indent). At least 20 reproducible indents on random areas were undertaken, for each sample.

[0234] Results and Discussion:

[0235] An adequate 3D printing process for the CNFs having a concentration of 0.6 wt. % was achieved, i.e. the deposited tracks did not collapse and 3D constructs could be printed.

[0236] Optical images of the 3D constructs (target dimensions 40 mm×20 mm×2 mm) were acquired and the light transmittance was quantified. The printed tracks of the samples CNF_2.5 and CNF_3.8 showed weaker definition compared to CNF_6.0. The CNF_6.0 sample demonstrated a 3D construct with well-defined tracks which is an indication of good print fidelity. Light transmittance through the constructs is shown in FIG. 5. When used as a wound dressing, the transparency facilitates the supervision of wound development.

[0237] For the 1 layer structures, the CNF_6.0 sample (having a relatively high viscosity and thus larger fraction of nanofibrils) was found to have particularly good printability, i.e. no major defects were observed on the printed structures.

[0238] The results of SEM analysis are presented in FIG. 6. The results indicate pore sizes in the micrometer scale, ranging from roughly 10 μm to 200 μm. A particular characteristic of CNF is the high aspect ratio of individualized nanofibrils, the length in the micrometer-scale, compared to the nanometric cross-sectional dimensions. Facilitated by these characteristics and the shear forces during extrusion, the nanofibrils align in the printing direction. The alignment of individual nanofibrils seems also to affect the self-assembly of the structure after lyophilization. Using computerized gradient analysis based on Sobel operators (Gadalamaria et al., Polymer Composite 14(2): 126-131, 1993) and Yoshigi et al., Cytom Part A 55a(2): 109-118, 2003), we were able to quantify the orientation of the aerogels texture. This is represented by polar plots of azimuthal facets, which indicate the main direction of orientation (Chinga et al., Journal of Microscopy-Oxford 227(3): 254-265, 2007). The more elongated the polar plot is the more pronounced is the orientation in a given direction. The polar plots of structures printed in a horizontal direction are obviously horizontally oriented, compared to the vertically oriented polar plots of structures printed vertically. Samples CNF_2.5 and CNF_3.8 have clear orientation patterns defined by the micrometer-sized surface pores. However, sample CNF_6.0 exhibits a more isotropic texture. The surface texture of CNF_6.0 is composed of flakes/walls of self-assembled nanofibrils. Controlling the orientation of the printed pattern is particularly interesting for scaffolds and tissue engineering to control the growth and proliferation of cells in a given direction.

[0239] Table 3 shows the stiffness and hardness (nano-mechanical properties) of the CNF hydrogels (0.6 wt. % concentration):

TABLE-US-00003 TABLE 3 Sample Elastic modulus (MPa) Hardness (MPa) CNF_2.5 2.10 ± 0.33 0.21 ± 0.07 CNF_3.8 3.17 ± 0.54 0.52 ± 0.16 CNF_6.0 2.44 ± 0.18 0.48 ± 0.09

[0240] The results yield the level of elastic modulus, i.e. ˜2-3 MPa and hardness (˜0.2-0.5 MPa) of the three set of gels. CNF_3.8 and CNF_6.0 have higher hardness values than CNF_2.5.

[0241] Stiffness, the resistance to deformation (in the elastic region) of a material upon an applied force, is important for the mechanotransduction response of cells. For example, cells respond to stiffness of biomaterials by reorganizing the cytoskeleton, affecting the cell spreading, proliferation and migration. Thus, the stiffness of the biomaterial affects the biological behavior of the cells and tissue, which may be important from a wound healing point of view.

[0242] Conclusions:

[0243] The CNFs are 3D printable and offer the capability to form wound dressings which may be adapted to specific requirements (shape and composition) in the x, y, and z directions. The CNF gels can be cross-linked with Ca.sup.2+ and easily managed to be applied in a wound situation. The wound dressing is in addition transparent which is expected to facilitate the wound healing management.

Example 5—Preparation of Oxygenated CNFs and Characterisation

[0244] Preparation of Oxygenated CNFs:

[0245] The CNFs produced in Example 1 (concentration 0.6 wt. % in water) with three different oxidation levels were denominated CNF_2.5, CNF_3.8 and CNF_6.0 (Table 1). The CNFs were diluted to 0.2 wt. % with purified water (Milli-Q water purifier, Millipore, Molsheim, France). The three grades of CNFs were sterilized in high-pressure steam for 20 minutes (121° C.) in an autoclave (TOMY, Autoclave SX-700E, Tokyo, Japan). The gels were kept at 4° C.

[0246] The three grades of CNFs were oxygenated by the OXY BIO System (Oxy Solutions, Oslo, Norway). A detailed description of the oxygenation device and production process is described in WO 2016/071691 (Oxy Solutions AS, Oslo, Norway). The OXY BIO System contains a piping system with venturi where oxygen gas (98%, Praxair, cat no. 500183, Oslo, Norway) and CNFs were mixed. During the production, the corresponding CNF was circulated through the oxygenation device continuously for a minimum of 10 minutes. To confirm if the desired oxygen concentration (>30 mg/l) was achieved under the production, the dissolved oxygen (DO) concentration was measured with Orion RDO Oxygen meter (Orion A323, Thermo Scientific, Massachusetts, USA). The production settings were 3.45 bar (liquid pressure) and 200 ml/min O.sub.2 (oxygen gas flow). The CNF was held cold during the whole production. After the production, oxygenated CNF was filled in glass vials (VWR, Pennsylvania, USA, cat. no. 216-3006) and sealed with aluminium center tear seals (VWR, Pennsylvania, USA, cat. no. 218-2117) and Bromobutyl stoppers (VWR, Pennsylvania, USA, cat. no. WHEAW224100-405).

[0247] Characterisation of Oxygenated CNFs:

[0248] Viscosity of the oxygenated CNFs was assessed with a Brookfield viscometer (Brookfield DV2TRV). The running parameters were: assessed volume: 200 mL. Temperature: 23° C.±1° C. Spindles: V-71.

[0249] Quantification of residual fibers was performed with a Fiber Tester (L&W Fiber Tester Plus, Code 912). The equipment quantifies the amount of residual fibers and fines that are larger than 7 μm. A volume of 40 ml of each CNF dispersion (0.2 wt. %) was prepared and quantified. The analysis was based on the acquisition of more than 7800 images. Two replicates were undertaken for each series. The

[0250] CNF dispersions were diluted to 0.1 wt. % and analyzed with a Particle size analyzer (N5 Submicron Particle Size Analyzer, Beckman Coulter), which can determine particle sizes in the range of 3 nm-3 μm.

[0251] Results and Discussion:

[0252] Brookfield viscosity values of the oxygenated CNFs are shown in FIG. 7. There are two specific trends revealed by the viscosity data: (i) the viscosity decreases as the oxidation increases; and (ii) the oxygenation process decreases the viscosity of the corresponding samples, The reduction in viscosity with increasing oxidation at 0.2 wt. % concentration may be due to the residual fibres and fine materials. Residual fibres are relatively long objects that may contribute to increase the viscosity at low concentration of the dispersion.

[0253] In FIG. 8B, the analysis of the dispersion with a nanoparticle analyser shows that the mean object size decreases as the oxidation increases. Additionally, quantification with laser profilometry revealed that the fraction of residual fibres (micrometer-sized) decreases correspondingly. This is confirmed by quantifying a reduction of residual fibres and fines as a function of oxidation (FIG. 8A). Consequently, a higher fraction of relatively long objects may be the factor affecting the increase in viscosity of sample CNF_2.5, at diluted dispersions (0.2 wt. %). The reduction in viscosity with oxygenation may be attributed to mechanical stress of CNFs due to circulation through the OXY BIO System during the oxygenation process. An increased concentration of dissolved oxygen may also contribute to a reduction in viscosity, i.e. oxygen may act as a spacer between the nanofibrils.

Example 6—Shelf-Life Testing of Oxygenated CNFs

[0254] Oxygenated and non-oxygenated CNFs were stored in sealed glass vials at room temperature (22° C.) for 5 weeks. Dissolved oxygen (DO) concentrations were measured at production date (week 0) and 5 weeks later by Winkler titration as previously described (Moen et al., Health Sci. Rep. e57, 2018).

[0255] As shown in FIG. 9, no significant differences in DO levels were observed between the three CNF grades (31.2 mg/l for CNF_2.5, 29.6 mg/l for CNF_3.8 and 31.6 mg/l for CNF_6.0). This result demonstrates that CNFs with different surface chemistry and morphology can be oxygenated to approximately the same high levels of DO by the OXY BIO System. After 5 weeks storage, the levels of DO in 0.2 wt % CNF were reduced to 26.9%, 31.1% and 38.0%, respectively. Nevertheless, the DO levels were twice as high as control levels.

Example 7—Antimicrobial Testing of Oxygenated CNFs

[0256] Blinded samples of oxygenated CNF, non-oxygenated CNF and Prontosan wound gel as positive control (Braun Medical AG, Sempach, Switzerland, cat. no. 400515), were evaluated for their antimicrobial effect.

[0257] Method:

[0258] Overnight culture of P. aeruginosa (ATCC 15692) or S. aureus (ATCC 29213) was set to the final bacterial concentration of 1×10.sup.7 colony forming units (CFU)/ml using optical density (OD) at 600 nm. 10 μl of the prepared bacterial suspension (1×10.sup.7 CFU/ml) were mixed with 500 μl gel and incubated at 37° C. for 4/24 h. 230 μl was suspended in 2 ml phosphate buffer (0.05% Triton X-100 in 0.0375 M phosphate) and diluted five times in ten-fold steps. 50 μl from each dilution was spread on horse blood agar plates and incubated over night at 37° C. The number of CFUs on the blood agar plates was counted and the number of CFUs in the original tube with gel and bacteria mix was calculated. This was defined as bacterial survival after 4 and 24 hours treatment. For each gel, 5 replicates were performed, and as negative control 500 μl brain heard infusion medium diluted 100 times in H.sub.2O (BHI100) was used instead of gel.

[0259] In a first trial, the bacterial survival of the aerobic bacteria Pseudomonas aeruginosa (P. aeruginosa) was assessed. The quantification of bacterial survival was performed after 4 and 24 hours in order to verify a potential rapid antimicrobial effect after 4 hours. This rapid effect was shown after 4 hours (FIG. 10). Oxygenated CNF_2.5, CNF_3.8 and CNF_6.0 after 4 hours had significantly lower survival of P. aeruginosa (P<0.05, Independent 2-tailed t-test) compared to non-oxygenated CNF_2.5, CNF_3.8 and CNF_6.0, respectively. The results were confirmed after 24 hours. Increasing the charge of the CNFs caused a larger antimicrobial effect and this effect was potentiated by oxygenation.

[0260] In a second trial, the bacterial survival of the aerobic bacteria Pseudomonas aeruginosa (P. aeruginosa) and Staphylococcus aureus (S. aureus) after 24 hours were investigated (FIG. 9A-B). The trials started 1-3 weeks after the production of oxygenated CNFs. However, FIG. 9 confirms that the potential reduction of dissolved oxygen in the CNF gels is expected to be minor at the time of assessment. CNF (0.2 wt. %) with increasing oxidation levels (CNF_2.5, CNF_3.8 and CNF_6.0, Table 1) had a significant antimicrobial effect (P<0.05, Independent 2-tailed t-test) compared to BHI100 (negative control) in both trials (FIG. 11A-B).

[0261] These results confirm that carboxylated CNF gels have an antimicrobial effect. Oxygenated CNF_2.5 and CNF_6.0 had significantly lower survival of P. aeruginosa (P<0.05, Independent 2-tailed t-test) compared to non-oxygenated CNF_2.5 and CNF_6.0, respectively (FIG. 7A). The difference between CNF_3.8 and oxygenated CNF_3.8 was not significant. Lowest bacterial survival of P. aeruginosa was measured for oxygenated CNF_6.0 (FIG. 11A). These results indicate that the higher oxidation level (ONE_6.0), the better the antimicrobial effect. The effect of CNFs is further potentiated in the presence of high levels of dissolved oxygen. Similar results were observed with the bacteria strain S. aureus (FIG. 11B). The gels CNF_6.0 and CNF_6.0 oxygenated perform similar to the Prontosan gel which is a potent antimicrobial, used as control in this study. It is noted that the gels were diluted to 0.2 wt. % concentration for oxygenation by the OXY BIO system. Previously it has been demonstrated that increasing the concentration of carboxylated ONE increases the antimicrobial effect (Jack et al., Carbohydrate Polymers 157, 1955-1962, 2017). It can be expected that a highly oxygenated gel with a higher concentration of nanofibrils will be a potent antimicrobial agent.

Example 8—SEM Characterization of Biofilms

[0262] In order to shed more light on the mechanism of action of the ONE and oxygenated CNFs, biofilms of S. aureus and P. aeruginosa were grown on pig skin and agar and treated with the ONE gels. The samples were fixed, freeze-dried and assessed with SEM.

[0263] Method:

[0264] Biofilms of P. aeruginosa (ATCC 15692) or S. aureus (ATCC 29213) were grown on pig skin and agar. After the incubation samples were fixed by 2.5% glutaraldehyde overnight, washed by buffer under agitation twice for 30 min, then fixed in 1% osmium tetroxide overnight, washed by ultrapure water under agitation twice for 30 min, plunge-frozen in liquid propane, and freeze-dried overnight. After, the samples were mounted on microscopy pins and coated by 15 nm of Au/Pt. The imaging was done by Zeiss Supra 40VP SEM, in secondary electrode image mode. The acceleration voltage and working distance were 3 kV and 12 mm, respectively.

[0265] Results and Discussion:

[0266] The results are presented in FIG. 12 and evidence the mode of action of the CNF. FIG. 12A-B are images of CNF entrapping P. aeruginosa. FIG. 120-D are images of CNF entrapping S. aureus. The nanofibrils appear to form a network which surround and entrap the bacteria. The spatial distribution of carboxylated nanofibrils seems to depend on the oxidation degree (carboxylic groups) and this may facilitate the interaction of the CNF with the bacteria. Additionally, the aldehyde groups encountered on the CNF surface (Table 1) may contribute to anchoring the individual nanofibrils to the proteins in the bacteria cell wall, thus entrapping the microorganisms and limiting their mobility and growth. Individual nanofibrils are playing a specific role on entrapping bacteria and potentially limiting their further mobility and growth (FIG. 12).

Example 9—Preparation of Oxygenated CNFs Hydrogels

[0267] Oxygenated hydrogels containing surface-charged nanofibrils were produced from corresponding oxygenated “CNF liquids” having a low concentration (0.2 wt. % or 0.4 wt. %) of nanofibrils by cross-linking (through the —COO.sup.− groups) with Ca.sup.2+ cations.

[0268] Method:

[0269] The dissolved oxygen (DO) content in 0.2 wt. % and 0.4 wt. % oxygenated nanocellulose, with or without CaCl.sub.2, was determined to test whether the addition of CaCl.sub.2 changes DO and viscosity. CaCl.sub.2 (50 mM or 100 mM) was added after each nanocellulose was oxygenated. The level of DO was then measured by Winkler titration (triplicate measurements) on production day and 1 month later. Changes in viscosity were visually observed.

[0270] Results and Discussion:

[0271] Results are presented in FIG. 13. The addition of CaCl.sub.2 to 0.2 wt. % oxygenated nanocellulose resulted in a small reduction in DO−6.9 mg/IlDO and 3.8 mg/l DO for 50 mM CaCl.sub.2 (production day and 1 month later, respectively) and 7.7 mg/l DO and 4.5 mg/l DO for 100 mM CaCl.sub.2 (production day and 1 month later, respectively). The addition of CaCl.sub.2 to 0.4 wt. % oxygenated nanocellulose resulted in a small reduction in DO−2.2 mg/l DO for 50 mM CaCl.sub.2 (1 month later) and 0.7 mg/l DO and 3.6 mg/l DO for 100 mM CaCl.sub.2 (production day and 1 month later, respectively). Cross-linking of the “CNF liquids” with Ca.sup.2+ was found to increase the viscosity due to cross-linking, but without unduly affecting the oxygenation level.

Example 10—Preparation of Oxygenated CNFs—Hydrogels

[0272] Oxygenated hydrogels containing surface-charged nanofibrils were produced from corresponding oxygenated “CNF liquids” having a low concentration of nanofibrils (0.2 wt. %) by mixing with non-oxygenated CNF gels having a higher CNF content (0.6 wt. %).

[0273] Method:

[0274] Oxygenated CNF (0.2 wt. %) was mixed with non-oxygenated CNF (0.6 wt. %) to obtain an oxygenated CNF with higher CNF concentration (0.4 wt. %). Details of the materials are set out in Table 4:

TABLE-US-00004 TABLE 4 Carboxyl CNF Material content of CNF concentration Code (mmol/g) (wt. %) Oxygenation 22_01 2.5 0.2 22_02 3.8 0.2 22_03 6.0 0.2 22_04 2.5 0.2 X 22_05 3.8 0.2 X 22_06 6.0 0.2 X 22_07 2.5 0.4 22_08 3.8 0.4 22_09 6.0 0.4 22_10 2.5 0.4 X 22_11 3.8 0.4 X 22_12 6.0 0.4 X

[0275] Brookfield viscosity of the materials was measured as set out in Example 1.

[0276] Results:

[0277] The viscosities of the materials are given in FIG. 14. The viscosities of samples 22_01 to 22_06 correspond to the results given in FIG. 5. A significant increase in viscosity of the 0.4 wt. % “gels” was observed.

Example 11—3D Printing of Ooxygenated CNFs

[0278] Oxygenated CNFs (CNF_6.0) having concentrations of 0.2 wt. % and 0.4 wt. % were extruded (i.e. injected) through a 50 ml needle tip with 18G cannula syringe (Braun, Einmal lnjektions-Kanule, 1.20×40 mm BC/SB 18Gx1 ½) to test the potential impact of 3D printing on their oxygen content.

[0279] The results in FIG. 15 show that the extrusion process did not lead to a significant loss of oxygen. The reductions in dissolved oxygen (DO) were small and not significant (p value=0.277 for 0.2 wt. % nanocellulose and p value=0.393 for 0.4 wt. % nanocellulose).

[0280] Similar experiments were carried out with CNF_2.5 and CNF_3.8 materials having concentrations of 0.2 wt. %. Injection through a 50 ml needle tip with 22G cannula syringe (Braun Sterican®, Einmal lnjektions-Kanule, 0.45×12 mm BULB 26Gx½) resulted in a small reduction in DO which was not significant.

Example 12—Antibacterial Properties of CNFs

[0281] The antibacterial effect of the CNF gels produced in Example 1 (CNF_2.5, CNF_3.8 and CNF_6.0) on P. aeruginosa (ATCC 15692, American Type Culture Collection, Manassas, Va.) and S. aureus (ATCC 29213, American Type Culture Collection, Manassas, Va.) was assessed in vitro.

[0282] Method—Determination of Bacterial Survival:

[0283] Colonies of P. aeruginosa or S. aureus were cultured on horse blood agar plates (Columbia agar, Oxoid, Basingstoke, UK) supplemented with 5% defibrinated horse blood (Swedish National Veterinary Institute, Uppsala, Sweden), then transferred into 10 ml 3.7% brain heart infusion (BHI) broth (Difco, BD Diagnostics, Franklin Lakes, N.J.) and incubated at +37° C., 250 rpm overnight. The bacterial suspension was centrifuged for 10 minutes at 2000×g. The supernatant was discarded and the pellet was re-suspended in 1 ml 0.037% BHI (BHI medium diluted 100 times in water, BHI100). This suspension was further diluted in BHI100 to reach the final bacterial concentration of 1×10.sup.8 colony forming units (CFU)/ml, as estimated by measuring optical density at 600 nm. 10 μL of the prepared bacterial suspension (1×10.sup.8 CFU/ml) were mixed with 500 μl CNF gel and incubated at 37° C. for 24 h. 230 μL of the mixture was suspended in 2 ml phosphate buffer (0.05% Triton X-100 in 0.0375 M phosphate) and diluted five times in ten-fold steps. 50 μL from each dilution was spread on horse blood agar plates and incubated overnight at 37° C. The number of CFUs on the plates were counted and the number of CFUs in the original tube with gel and bacteria mix was calculated and defined as bacterial survival after 24 hours treatment. Each gel was tested three times in three separate blinded trials on P. aeruginosa and in one trial on S. aureus. For each trial 5 replicates were performed, and as a negative control 500 μL BHI100 was used instead of CNF gel.

[0284] Method—Swimming Assay:

[0285] Luria-Bertani broth supplemented with 0.5% glucose and 0.3% agar was melted in boiling water and then cooled to 45° C. before adding the CNF (6 wt. %) to the melted agar in a 5% v/v mixture as described by Silva et al. (J. Mater. Sci. 54(18), 12159-12170, 2019). The mixture was poured into 55 mm petri dishes (7.5 ml per dish) and was cured for 3 hours with the lid tilted. One sample where the CNF gels were replaced with water was used as control. 5 μL of S. aureus (ATCC 29213) (non-flagellated bacteria) or P. aeruginosa (ATCC 15692) (flagellated bacteria) suspension (1×10.sup.9 CFU/ ml) was inoculated in the centre of each plate by dipping the pipette tip slightly into the agar. The CNFs and controls were tested in triplicates. The plates were incubated in upright position in aerobic conditions at 37° C. for 9 hours. Digital images were acquired of each agar plate and assessed with the ImageJ program. The images were automatically filtered with a median filter to remove noise and automatically thresholded into binary images to segment the bacteria halo. The Feret's diameter of the bacteria halo was quantified and reported as the degree of swimming of each tested sample.

[0286] Results:

[0287] The results in FIG. 16 confirm an antibacterial effect of the CNF gels. All samples were significantly different compared to the control. The results in FIG. 17 show the swimming levels of P. aeruginosa in the agar gels containing the CNFs.

Example 13In Vivo Surgical Site Infection (SSI model)—CNF and Oxygenated CNF

[0288] The antimicrobial effect of CNF_3.8 and oxygenated CNF_3.8 as prepared in accordance with Examples 1 and 5, respectively, was determined in vivo and compared to Protonsan® wound gel (obtained from B. Braun, Germany).

[0289] Method:

[0290] Bacterial preparation: Colonies of S. aureus (ATCC 29213) were cultured on horse blood agar plates (Columbia agar, Oxoid, Basingstoke, UK) supplemented with 5% defibrinated horse blood (Swedish National Veterinary Institute, Uppsala, Sweden), then transferred to 10 ml 3.7% brain heart infusion (BHI) broth (Difco, BD Diagnostics, Franklin Lakes, N.J.) and incubated at +37° C., 250 rpm overnight. The bacterial suspension was centrifuged for 10 minutes at 2000×g. The pellet was re-suspended in 1 ml BHI100 and the suspension was further diluted in BHI100 to reach 2×10.sup.9 CFU/ml, as estimated by optical density at 600 nm. 8 ml of bacterial suspension were transferred into a 15 ml tube and 3-0 silk sutures (684G, Ethicon, Sollentuna, Sweden) were soaked for 30 minutes in the suspension. The sutures were dried on filter paper at +4° C. and kept at +4° C. until use (a maximum of 4 hours). Approximately 5×10.sup.3 cells were adsorbed per cm suture as previously described (Hakansson et al., Antimicrob. Agents Chemother. 58(5), 2982-4, 2014). Animal model: The model used in this study has been published previously (Gisby et al., Antimicrob. Agents Chemother. 44(2), 255-60, 2000; Hakansson et al., Antimicrob. Agents Chemother. 58(5), 2982-4, 2014; McRipley Antimicrob. Agents Chemother. 10, 38-44, 1976; Rittenhouse et al., Antimicrob. Agents Chemother. 50, 3886-3888, 2006) and was modified as described below. All animal experiments were performed after prior approval from the local Ethics Committee for Animal Studies at the Administrative Court of Appeals in Gothenburg, Sweden. The animals were kept in a 12-hours light-dark cycle with free access to water and pellets (Lab For, Lantmännen, Malmö, Sweden), and were cared for in accordance with regulations for the protection of laboratory animals. Female CD1 mice (25-30 g, Charles River, Sulzfeldt, Germany) were anaesthetized with isoflurane (Isobavet, Shering-Plough Animal Health, Farum, Denmark). The back of the mouse was shaved with a clipper, washed with 70% ethanol and a 1 cm full-thickness incision wound was placed centrally on the back of the mouse at the neck region with a scalpel. Approximately 1 cm of the infected suture was placed into the wound and a single nylon suture 5-0 Ethilon*II (EH7800H, Ethicon, Sollentuna, Sweden) was used to close the wound to avoid the mouse from scratching. Buprenorfin (48 μg/kg, Temgesic, Shering-Plough, Brussels, Belgium) was given pre-operatively by intraperitoneal injection for post-surgical pain relief. 24 hours post-infection, 30 μl of placebo or active treatment was applied to the wound with a micropipette. 3 hours later, a second 30-μl treatment was applied to the wound. The placebo and treatment stayed in place in the wound. 2 hours after the last treatment, the mice were euthanized by cervical dislocation and an area of 2×1 cm around the wound (including the whole wound area and surrounding tissue) was excised and homogenized with a rotor stator homogenizer (T10 basic ULTRA-TURRAX, IKA-WerkeGmbH & Co. KG, Staufen, Germany) in 2 ml ice cold BIH100. The homogenate was diluted in six 10-fold steps by transferring 22.2 μl to 200 μl phosphate buffer (0.05% Triton X-100 in 0.0375 M phosphate) in a 96 well plate. 50 μl of each dilution was transferred to horse blood agar plates and incubated at +37° C. overnight. The colonies on the plates containing 30-300 CFU were counted and the number of CFUs/wound was determined.

[0291] Results:

[0292] Results are shown in FIG. 18.