1 Introduction

Wound contamination happens when microscopic organisms enter a break in the skin [1]. The infection causes the impediment in the healing of the wounds and may lead to life-threatening complexities. Life-threatening wounds can pose both immediate and delayed risks of life. The former is due to underlying organ damage or severe exsanguinations, and the latter is attributable to infection and sepsis. Therefore, wound infection is the major concern in the wound care management. Treatment of infected injury relies on upon how serious the injury is, its area, and whether different territories are influenced [2]. One effective method for treating of wound contaminations is the application of a wound dressing that effectively hinders bacterial development inside and on the surface of dressing itself [3]. By blocking development of biofilm growth on the dressing, this would hinder microscopic organisms from reseeding the injury, which would permit the microscopic organisms to conquer the regular host protection components. To generate antimicrobial activity in wound dressing, the antimicrobial agents can be impregnated into the dressing [4, 5]. These active agents leach out into the wound bed gradually during their application and kill the invaded microbes may present in the wounded tissue [6]. Chemical bonding of antimicrobial agent to the dressing backbone and contact killing of pathogenic microorganisms is another approach for fulfilling this goal [7–10]. A combination of these two approaches is also possible [11, 12].

In all of the aforementioned methods, careful selection of antimicrobial agents possesses prime importance. Nanotechnology provides many potential opportunities to eliminate the microbial pathogen [13]. Distinctive antimicrobial metal and metal oxides nanoparticles, such as ZnO, Ag, Cu, Fe3O4, Al2O3, TiO3, and SiO3 are introduced in the scientific literature related to public health and some of these materials found their way into the commercial products including wound dressing membranes. Fortunately, these metals and metal oxides can effectively suppress the growth of various bacteria [14–17]. However, one of the main concerns in using such metallic particles is their toxicity to eukaryotic cells, since they have to be released from the dressing material in order to be effective. Graphenic materials (GMs) such as graphene (G), graphene oxide (GO), and reduced graphene oxide (rGO) are new nanomaterials that found a number of applications in the cutting-edge fields of biological and medical sciences [18–20]. Specifically, generous endeavours have been committed to investigating the more extensive therapeutic uses of G based materials to microbial infections [21, 22]. There are reports showing the dependency of antimicrobial activities of graphenic materials to their physicochemical or structural properties such as lateral size, layer number, morphology, and dispersibility [23, 24]. By proper control of these factors a superior inhibition ability towards bacteria growth can be attained, therefore, promising applications for this type of low-cost and highly effective carbon nanomaterials as the bactericidal material is expected. Application of GMs as the antimicrobial component of wound dressings has been considered very recently in few publications. In the article reported by Deepachitra and co-workers [25], GO was incorporated in the collagen-fibrin composite film and used as wound dressing material. Both in vitro and in vivo studies revealed faster wound healing in rats compared to those of collagen-fibrin and control. Khan [26] subjected GO to NIR radiation (λ = 1064 nm) to eliminate bacteria and fungi and control the wound infection. Madhavan [27] designed a type of electrospinning based on graphene–TiO2–ZnO composite nanofibres and claimed the potential use of these nanofibres in an antibacterial wound dressing. The electrospun nanofibres made from chitosan-poly (vinyl alcohol)-GO reported by Liu [28] displayed excellent antibacterial activity against E. coli, indicated their potential for wound healing. Chitosan-poly (vinyl alcohol) nanofibres containing G were also prepared for wound healing by Lu [29]. These electrospun fibres led to rapid and complete wound healing compared to neat fibres and controls. They argued that the existing free electron in G could inhibit the multiplication of prokaryotic cells without affecting the eukaryotic cells, thus making G suitable as wound dressing component. He [30] incorporated GO to sodium alginate fibres to enhance its strength. The hybrid film could swell into hydrogel suitable for fabrication of wound dressing. Fan [31] prepared a wound dressing by cross-linking the GO–Ag composites into the polymer hydrogel, which showed antibacterial activity and accelerated the wound-healing. Sun [32] prepared a kind of Band-Aid with antibacterial property suitable for wound disinfection based on G quantum dots and a low level of H2O2.

To expand the present knowledge about utilization of GMs for biomedical applications, a novel category of wound dressing composed of GO incorporated polyurethane/siloxane networks was introduced in the present study. The versatile chemistry followed in this work enabled us to tune the physicomechanical and biological properties of the final dressings in a way that the as-prepared membranes showed excellent wound exudates management, biocompatibility, antimicrobial activity and mechanical strength, which are necessary for proper wound care. The performance of the prepared materials as wound dressing membranes was evaluated based on data recorded under in vitro and in vivo conditions.

2 Materials and methods

2.1 Materials

Poly(tetramethylene ether) glycol (PTMEG, Mn 2000 g mol−1) was supplied from Aldrich (Taufkirchen, Germany). It was freed from moisture by storing in a vacuum oven at 80 °C for 24 h. Poly(ethylene glycol) (PEG, Mn 2000 g mol−1) from Aldrich was dried through azeotropic distillation using dry toluene as solvent. Tetraethoxysilane (TEOS) and (3-aminopropyl)triethoxysilane (APS) were obtained from Aldrich and used as received. Isophorone diisocyanate (IPDI) from Merck (Darmstadt, Germany) was vacuum distilled. N-Methylpyrrolidone (NMP) and toluene were purified by distillation over CaH2 and sodium, respectively. xGnP® graphene nanoplatelets, grade C-750 (average surface areas of 750 m2 g−1, and atomic oxygen concentration about 8%) from XG sciences (Lansing, MI, US) was used without further treatments. Dibutyltin dilaurate (DBTDL) and tetrahydrofuran (THF) were purchased from Aldrich and used as received. Diethylene glycol from Merck was dried via storing in a vacuum oven at 80 °C overnight. S. aureus (ATCC 6538) and E. coli (ATCC 25922) bacteria and C. albicans (ATCC 10231) were purchased from Iranian Research Organization for Science and Technology (IROST, Tehran, Iran). Human skin fibroblast cells (Hu02) was purchased from Iranian Biological Resource Centre (IBRC, Tehran, Iran).

2.2 Synthesis of triethoxysilane-terminated polyurethane prepolymer (Si-PU)

Si-PU was prepared according to the procedure reported in our previous publication [33], briefly:

A polymerization kettle was charged with PEG (21.43 g), PTMEG (50.00 g) and NMP (40 ml). Under ambient temperature, IPDI (15.88 g) was introduced into the reactor during 30 min, then the temperature was increased to 85 °C. The reaction progress was monitored through measurement of free NCO content (ASTMD-2572). After completion of this step, a solution of diethylene glycol (1.89 g) dissolved in NMP (25 ml) was added and the reaction was continued at 85 °C for about 1 h. Then, the temperature was reduced to ambient condition. The reactor was charged with a solution of APS (7.90 g) dissolved in NMP (40 ml). The mixture was stirred for 30 min, and then the temperature increased to 50 °C for 3 h. The reaction product with predetermined solid content (50 w/w%) was transferred to a glass bottle, sealed and kept in the refrigerator for subsequent use.

2.3 Preparation of GO containing polyurethane-siloxane membranes (XSi-PU/GO)

The nanocomposite dressing membranes were prepared according to the formulations given in Table 1. At the first, the appropriate amount of GO powder was added to a 100 ml beaker containing NMP solvent (25 ml). The mixture was stirred at room temperature for 24 h using a magnetic stirrer. Desired amounts of Si-PU and TEOS were then added and the resulting mixture was subjected to ultrasound irradiation using ke 76 probe for 30 min. Finally, the catalytic amount of HCl solution (0.1 M) was dropped into the beaker and the resulting mixture was stirred vigorously for 1 min. After the removal of any trapped air by rapid evacuation, the mixture was cast slowly into a clean Teflon mold. The mold was kept at room temperature under ambient atmospheric condition overnight. It was then heated in an oven at 80 and 100 °C for 10 and 2 h respectively. The temperature was reduced again to 80 °C and high vacuum was applied to ensure complete removal of any residual solvent. The thickness of the membrane was adjusted by controlling the solid content of reaction mixture.

Table 1 Different formulations of wound dressing membranes
Full size table

For the preparation of a neat polyurethane/siloxane membranes, the mixture of Si-PU, TEOS and an appropriate amount of HCl solution was dissolved in NMP (Table 1) and the rest of procedure was repeated as described above.

For purification of the prepared membrane and removal of residual raw materials and solvents, all of the samples were immersed into ethanol (70% w/w) for 6 h, distilled water for another 6 h and fully dried under reduced pressure.

2.4 Characterization

2.4.1 Spectroscopic, tensile and dynamic-mechanical analysis

Fourier transform infrared spectra (FTIR) were recorded on a Bruker IFS 48 instrument (Bremen, Germany). All spectra were obtained in the air as a function of time with 16 scans at a resolution of 4 cm−1 and a spectral range of 400–4000 cm−1. The X-ray diffraction (XRD) pattern of the samples was recorded in a Siemens X-ray diffractometer (D5000 Germany) at crystal monochromated Cu Ka radiation in the angular range 2–15 (2θ) at 35 kV operating voltage and 25 mA current. Measurement of tensile strength, elongation-at-break and modulus of samples were carried out using a tensile tester (Instron 6025, MA, US) with a crosshead speed 50  mm min−1. Samples were cut into bars of 50 mm length and 5 mm width. The test was performed at room temperature. For each sample, five specimens were tested. Dynamic mechanical thermal analysis (DMA) was performed using a Triton instrument (model Tritec 2000, England) within a temperature range of −100–200 °C, a heating rate of 3 °C/ min−1 at a frequency of 1 Hz in tensile mode. The loading amplitude was set at 20 μm. Sample dimensions were 30 × 10 × 1 mm3.The values of storage modulus, loss modulus, and tan δ vs. temperature were recorded for each sample. The maximum temperature of tan δ curves was considered as glass transition temperature.

2.4.2 Measurement of water vapor transmission rate (WVTR)

The moisture permeability of the synthesized materials was estimated by the measurement of the water vapor transmission rate (WVTR) according to the procedure reported in ASTM E96/E96M. The following equation was applied:

$${\rm{WVTR}} = \frac{{{W_{\rm{i}}} - {W_{\rm{t}}}}}{{A \times t}}$$

WVTR was expressed in g−1 m2 per day. A, W i and W t designated the area of cup mouth (m2), the weight of water containing cup before and after placing in the oven at 37 °C and 35% humidity and t showed the duration of measurements.

2.4.3 Gel content, surface, and bulk hydrophilicity

The progress of crosslinking reaction for the prepared samples was estimated by measuring their gel content. For this purpose, the membranes were dried under vacuum for 24 h at room temperature and weighed. Then, the samples were extracted by THF in a Soxhlet extractor for 24 h. The insoluble part was dried at 50 °C and weighed. The gel content was defined as follows:

$${\rm{Gel}}\,{\rm{content}}\,{\rm{\% }} = \frac{{{W_{\rm{d}}}}}{{{W_{\rm{i}}}}} \times 100,$$

where W d and W i designated the weight of dried membrane after extraction and the initial weight of the membrane, respectively.

The static contact angle of water droplets on as-prepared films surface was utilized as a means for measurement of surface hydrophilicity. For this purpose, an optical video contact angle system (OCA-15-plus, Dataphysics, Germany) was used. Bulk hydrophilicity of the samples was also assessed based on measurement of equilibrium water absorption (EWA %) in PBS buffer. Completely dried and weighed samples (20 × 10 mm) were placed in the beaker containing PBS. The swollen samples were wiped gently with a soft paper tissue before being weighed. EWA% was determined using the following equation:

$${\rm{EWA\% }} = \frac{{{W_{\rm{s}}} - {W_{\rm{d}}}}}{{{W_{\rm{d}}}}},$$

where W d and W s are the weights of dry and swelled films, respectively.

2.4.4 Conductivity analysis

The electrical conductivity (S cm−1) of the samples was measured at room temperature using a home-made four-point probe instrument, utilizing the following equation:

$$\sigma = \frac{{I\,{\rm{Ln}}\,{\rm{2}}}}{{V\pi {d_{\rm{n}}}}},$$

where V, I and d n represented potential in volt, current in ampere and thickness of the sample in cm, respectively.

2.4.5 Cytotoxicity of dressing membranes

The freshly prepared dressing membranes were immersed in the culture medium for 7 days at 37 °C. The possible existence of toxic material in this medium was examined by evaluating the metabolic activity of Hu02 fibroblast cells cultured in medium containing extracted leachates from different samples. The procedure reported in our previous work [10] was followed for this assessment. The percentage of relative cell viability was calculated according to following equation:

$${\rm{Cell}}\,{\rm{viability}}\,{\rm{\% }} = \frac{{{\rm{O}}{{\rm{D}}_{{\rm{sample}}}}}}{{{\rm{O}}{{\rm{D}}_{{\rm{negative}}\,\,{\rm{control}}}}}},$$

where OD designates the optical density. A growth medium, containing cells but no extracted leachates were considered as negative control. Each measurement was repeated at least three times.

2.4.6 Cell adhesion and proliferation on dressing membranes

Hu02 cells were cultured at 37 °C/5% CO2 in 10% bovine calf serum, high glucose-DMEM culture medium supplemented with 1% penicillin, streptomycin mixture. Cells at 90% confluency were detached from the flask by treating with a trypsin-EDTA solution (0.05%). The cells were counted using a Neobar lam. The dressing membranes soaked in PBS overnight and sterilized by UV irradiation (15 min each side). They were cut to a size fit the bottom of each well of a 96-well culture plate. Then, the culture medium containing 5 × 103 Hu02 fibroblast cells were placed on the membrane and incubated at 37 °C for 45 min to enable the cells to attach on the membrane surface. The membrane/cell constructs were supplied with more culture medium and the incubation period was extended to predetermined time points (1, 3 and 5 days). At the end of certain time points, the seeded membranes were washed with PBS to remove unattached cells and then the membranes were transferred to new culture plates. Cells were harvested after 3 min incubation with 0.25 w% trypsin and 1 mM EDTA at 37 °C, and the quantification of cells was done by the MTT assay using following equation:

$${\rm{Cell}}\,{\rm{viability}}\,{\rm{\% }} = \frac{{{\rm{O}}{{\rm{D}}_{{\rm{sample}}}} - {\rm{O}}{{\rm{D}}_{{\rm{positive}}\,\,{\rm{control}}}}}}{{{\rm{O}}{{\rm{D}}_{{\rm{negative}}\,\,{\rm{control}}}}}},$$

A growth medium, containing cells but no membrane was considered as the negative control and a medium containing membrane but no cells were considered as positive control. Each measurement was repeated at least five times.

2.4.7 Antibacterial study

Antimicrobial activity of prepared membranes was evaluated against S. aureus (ATCC 6538) and E. coli (ATCC 25922) bacteria and C. albicans (ATCC 10231) using “colony forming count” method based on procedure reported in ASTM E 2180-07. Bacteria at inoculated concentration of 2 × 108 CFU ml−1 and membranes with the dimensions of 1 cm × 1 cm were used throughout the tests. Bacterial cultures grew for 18 h at a specified temperature in tryptic soy broth. 1.0 ml of inoculated agar slurry was pipetted onto the test and control samples and allowed to gel. 24 h after contact; surviving microorganisms are recovered via elution of the agar slurry inoculums from the test substrate into neutralizing broth and extracted via vortexing that provides complete removal of the inoculum from the test polymers. Serial dilutions were made and then spread plates were made of each dilution and incubated at 37 °C for 48 h. Colony numbers were counted and recorded for each dilution plate. The geometric mean of the number of organisms that recovered and percent reduction was determined by the following equations:

$${\rm{Geometric}}\,{\rm{mean}} = \frac{{{\rm{lo}}{{\rm{g}}_{10}}{\rm{X}}1 + {\rm{lo}}{{\rm{g}}_{10}}{\rm{X}}2 + {\rm{lo}}{{\rm{g}}_{10}}{\rm{X}}3}}{3}$$
$${\rm{\% }}\,{\rm{Reduction = }}\frac{{\left( {{\rm{a}} - {\rm{b}}} \right){\rm{ \times 100}}}}{a}$$

2.4.8 In vivo assay of dressing membranes

The wound healing characteristics of the prepared wound dressings were evaluated using a rat model. All experiments were performed with the approval of the Animal Use and Care Administrative Advisory Committee of Tarbiat Moddares University. The Wistar rats with an average body weight of 200–250 were employed through these experiments. The animals were anesthetized with an intraperitoneal injection of ketamine (85 mg kg−1) and xylazin (15 mg kg−1) and then the dorsal hair of rats was shaved and a full-thickness skin wound at about 1.5 cm2 area was prepared by excising the dorsum of the animals. Thirty-six rats were randomly divided into three groups, named as group 1, 2 and 3. The wound of group 1 rats (control group) were covered by sterilized cotton gauze without any other intervention. The wounds of group 2 and 3 rats were covered with neat Si-PU and Si-PU/GO5.0% dressing membrane, respectively. The dressings were fixed in the wound position by an elastic adhesive bandage.

The dissected wounds were not sterilized and the tested animals kept in separate cages under the environmental condition to avoid antibacterial interference by external disinfectant agents. Macroscopic photographs of the wounds were taken, and the wound area was measured after certain postoperative time. The percentages of wounds closure rate were calculated according to the following formula:

$${\rm{Wound}}\,{\rm{closure}}\,{\rm{rate}}\left( {\rm{\% }} \right) = \frac{{{A_0} - {A_{\rm{t}}}}}{{{A_0}}} \times 100,$$

where A 0 and A t were designated to initial wound area and wound area at designated time, respectively.

For the histological analysis, the dressings were removed from sacrificed rats and then, the healed wound skin tissues were dissected. The cut tissue was fixed in 10% v/v buffered formalin, dehydrated in a graded ethanol and embedded in paraffin. Tissue sections were processed for Hematoxylin & Eosin (H&E) and Malory Trichrome staining.

2.5 Statistical analysis

Statistical analyses were performed via PASW statistics program package, version 18 (SPSS Inc., Chicago, IL, USA). Comparison of the obtained data for different samples was performed with One-Way ANOVA using Tukey posthoc test. The significance level was set at p ≤ 0.05.

3 Results

3.1 Synthesis and spectroscopic characterization of dressings

The methodology followed for the preparation of dressing membranes was described in Supplementary Information (Schemes S1 and S2), and the formulations for these dressing were presented in Table 1.

FTIR spectrum of Si-PU (Supplementary Information Fig. S1a) displayed a broad band centred at 3401 cm−1 which was related to the stretching vibration of urethane and urea N-H groups. The urethane (NH–CO–O) and urea (–NH–CO–NH–) carbonyl groups were combined and detected as a strong broad peak centred at 1685 cm−1. The splitting of carbonyl peaks was not detected. Since the FTIR spectrum of Si-PU was taken as an admixture in NMP solvent, therefore, the amide bond of the solvent was mixed with peaks of carbonyl groups of Si-PU. The peak observed at 1543 cm−1 was related to C–N stretching, combined with N–H out-of-the-plane bending vibrations of the urethane and urea groups. The peak appeared at about 1100 cm−1 was attributed to ether (C–O–C) bonds arising from polyol segments. The presence of O–Si–O and urea carbonyl groups peaks at 778 and 1682 cm−1 and the absence of NCO peak at 2230 cm−1 indicated the complete reaction of isocyanate and NH2 groups.

The suitability of reaction condition for the sol–gel process was confirmed by high values for gel content of the prepared membranes measured using THF as solvent (Table 1). The chemical identity of the dressing membranes was studied using FTIR spectroscopic method. Representative spectra of polyurethane/siloxane networks prepared in the presence or absence of GO (XSi-PU and XSi-PU/GO5.0%) are shown in Supplementary Information Fig. S1b, c. For both materials, the presence of the peak at about 1030 cm−1, as the characteristic stretching vibration of the Si−O−Si bond and almost fading of the peak at 771 cm−1 related to stretching vibration of O−Si−O linkages confirmed the successful formation of siloxane network through sol–gel reaction in the presence/absence of GO additive. In addition, after removal of NMP solvent, the carbonyl peak of urethane groups was detected at 1717 cm−1 for dressing without GO. After the introduction of GO into the formulation of dressing, this peak was shifted to a lower frequency (1706 cm−1), as a result of hydrogen bonding type interaction of carbonyl groups with hydroxyl and carboxylic groups present on the surface of GO.

3.2 Hydrophilicity of wound dressing membranes

EWA and WVTR were measured to find insight about the ability of the designed dressing membranes for wound exudates management (Table 2). The XSi-PU membrane showed the EWA at around 70%. After incorporation of GO, the EWA significantly decreased and this reduction was the highest for that sample with a maximum amount of GO. The WVTR data for the prepared dressing membranes and some of the commercially available dressings are also summarized in Table 2. The XSi-PU sample showed a WVTR at about 1950 g m−2 day. Upon introduction of GO nanoplatelets into the dressing membrane a considerable reduction in WVTR was observed.

Table 2 Hydrophilicity and WVTR of the prepared wound dressing membranes
Full size table

Surface hydrophilicity of dressings is usually expressed in terms of contact angle for a water drop on the dressing surface. The pristine membrane without GO (XSi-PU) showed the contact angle of about 65°, however, the water droplet contact angle was increased up to 82° for XSi-PUGO7.5% with the highest concentration of GO nanoplatelets.

3.3 Viscoelastic, tensile strength and phase structure of dressing membranes

The viscoelastic behaviour of dressing materials with and without GO nanoplatelets was studied using DMA technique. The experiments were carried out in the tensile mode as a function of temperature from −100 to 150 °C. Typically, variations in storage modulus (E′), loss modulus (E′′) and loss tangent (tan δ) as a function of temperature were recorded and the results are displayed in Supplementary Information Fig. S2. The prepared membranes showed a two-phase structure since two main transitions were detected in their DMA curves. The positions of these two peaks were shifted to higher temperature upon raising of frequency, therefore, these peaks were attributed to the glass transition (Tg) of different domains in the structure of the prepared samples. Upon introduction of GO nanoplatelets into the dressing formulations, no considerable changes in the phase structure of polymeric networks were detected and the position of Tg peaks remained almost intact. However, an increase in the elastic modulus at rubbery plateau region was observed for these samples.

The mechanical strength of dressing membranes was evaluated by examining their stress–strain curves (Fig. 1). Based on these curves, all of the prepared membranes showed the general behaviour of lightly crosslinked elastomers as they exhibited reversible deformation, low initial modulus, and increased modulus and tensile strength at higher deformation due to the alignment of amorphous chains in the direction of tension. The tensile strength, modulus, and elongation at break of GO containing membranes were significantly higher than the neat network and these values increased with raising the GO nanoplatelets to content up to a maximum of 5 wt.%. The mechanical property of the dressing materials was also evaluated immediately after removing them from PBS solution at 37 °C for 2 days. The strain-stress curves of the dressing membranes recorded at hydrated state were also plotted in Fig. 1. A reduction in the tensile strength of dressing membranes was observed under a wet condition with the same trend recorded for these samples under the dry state.

Fig. 1
figure 1

Stress–strain curves of dressing membrane under a dry and b hydrated states

Full size image

The prepared membranes were also subjected to XRD analysis. The representative X-ray diffractograms for XSi-PU and XSi-PU/GO 5% dressing membranes are shown in Supplementary Information Fig. S3. The broad XRD diffraction peak centred at about 20° (2θ) recorded for XSi-PU sample was related to the characteristic diffraction peak of the amorphous polyurethane-siloxane network. With the introduction of GO nanoplatelets into the polyurethane-siloxane network, the height of observed peak was intensified while its broadness was decreased. The recorded data implied the improvement in the crystallinity of polyurethane-siloxane network upon introduction of GO.

3.4 Cytocompatibility and antibacterial activity of dressing membranes

To have an insight about cytocompatibility of dressings, two set of experiments were performed. Firstly, the dressings were put into PBS for 1 week and then the human dermal fibroblast cells were cultured in the medium containing leachates extracted from dressing membranes. The viability of cells was evaluated by MTT assay and no detectable reduction in viability of cells was recorded for the cultivated cells. (Supplementary Information Fig. S4). In addition, there was no significant difference (p ≥ 0.05) between the viabilities of cells exposed to the leachates of samples without GO or those containing different amounts of GO nanoplatelets. Secondly, the human dermal fibroblast cells were directly seeded on the dressing membranes covered the bottom of tissue culture plate wells. The proliferation rate of cells grown on the surface of dressing membranes was investigated by MTT assay at certain time interval. The recorded results at three certain time points (1, 2 and 3 days) are collected in Supplementary Information Fig. S5. There was no significant difference (p ≥ 0.05) between viability of cells grown on samples with various amounts of GO. With extending the culturing time, the cells were continued their growth and proliferation for all samples.

The capability of prepared dressing for preventing wound infection was evaluated by the quantitative “colony counting method” against E. coli and S. aureus as gram-negative and gram-positive bacteria, as well as C. albicans as a fungal strain (Fig. 2 and Table 3). Fortunately, antimicrobial activity was detected for those dressing membranes containing the GO nanoplatelets. Also, it was found that this activity was concentration dependent and increased with a higher concentration of GO.

Fig. 2
figure 2

Antibacterial activity of the prepared dressing membranes against S. aureus, E. coli, and C.albicans

Full size image
Table 3 Bacterial reduction percent of the dressing membranes
Full size table

3.5 In vivo wound healing study

For evaluating the wound healing process in the presence of GO containing dressing membrane, the XSi-PU/GO5% sample was utilized for covering a full thickness wound created in the dorsal area of Wistar rat. XSi-PU and cotton gauze were also used as control samples. The surfaces of the treated wounds were photographed and closure rate was calculated by measuring the area of the wounds at certain time intervals. The images of the open excision wounds and the percentage of wound closure at various time periods were shown in Fig. 3a and b, respectively. Based on Fig. 3a and b findings, much better healing efficacy was noticed for the GO containing wound dressing membrane (XSi-PU/GO5%) than the dressing without GO (XSi-PU) at the same time intervals. On day 7, the wound covered with XSi-PU/GO5% membrane showed 55% wound closure, whereas the XSi-PU treated wounds and control groups showed wound closures of 48 and 34%, respectively. After 14 days of treatment, there was 89% reduction in the size of wound covered by XSi-PU/GO5% dressing, whereas wounds size reduction in XSi-PU treated wounds and control group were 81 and 70%, respectively. On the 20th day, the wounds covered with XSi-PU/GO5% showed 98% wound closure, whereas, 91 and 81% decrease in the size of the wound were recorded for wounds protected by XSi-PU membrane and cotton gauze, respectively.

Fig. 3
figure 3

Photographs (a) and closure rate (b) of wounds treated with gauze (control), XSi-PU and XSi-PU/GO5% during the wound healing process for 20 days. According to the analysis of variances, *P < 0.05, values are significantly different from the previously compared group

Full size image

The H&E and Malory Trichrome (MT) stained sections of the wounds covered with XSi-PU/GO5% and XSi-PU dressings, as well as control group covered with cotton gauze on days 14 and 20 are shown in Fig. 4. Based on Fig. 4 images, at day 14, the H&E stained section of the wound covered with XSi-PU/GO5% dressing showed uniform and thick epithelium layer. Also, re-epithelization was pronounced in the wounds covered with XSi-PU sample, while in the case of the control group, epithelization was still in early stage. Furthermore, neovascularization at newly formed tissue was more pronounced for the wound covered with XSi-PU/GO5% compared to those treated with either XSi-PU or cotton gauze. Close inspection of H&E stained area of the wound dressed with XSi-PU/GO5% membrane at days 20 revealed the complete re-epithelization and reconstruction of epidermis and dermis. Also, some of the secondary structures of the healed skin, like hair follicle, could be observed, which were in their initial stage of development.

Fig. 4
figure 4

Representative images of MT and H&E stained histological sections a day 14 and b day 20 after initial wounding. Scale bars equal to 200 μm. Arrows indicated the blood vessels

Full size image

MT staining of the wound tissue at 14th day confirmed collagen deposition in sections of all treated wounds including XSi-PU/GO5%, XSi-PU, and cotton gauze. There was a clear difference between collagen expressions in the wounds covered with XSi-PU/GO5% and XSi-PU in comparison to the cotton gauze groups. Although, higher collagen expression was observed in both groups, but the thickness of deposited collagen in the wounds covered with XSi-PU/GO5% membrane was higher than that of XSi-PU treated the wound. On day 20, the thickness of deposited collagen in wound treated with XSi-PU/GO5% group was considerably higher than that of both XSi-PU and cotton gauze treated wounds. Close inspection of MT staining section wounds of XSi-PU/GO5% treated groups, showed dense and parallel collagen deposition with higher alignment and maturation level in comparison to other two groups.

4 Discussion

4.1 Synthesis and characterization of wound dressing membranes

To construct wound dressing with the ability to maintain optimum healing milieu over a damaged area and also providing the necessary protection from external mechanical stresses (pressure, friction, and shear) during the healing process, utilization of polyurethane/siloxane based material containing GO nanoplatelets was considered in the present study [10, 11]. To tune the overall ability of dressings for managing wound exudates, a mixture of two hydrophobic (PTMEG) and hydrophilic (PEG) polyols at 70/30 wt.% was introduced into the backbone of dressing membranes. The partial chain extension of the intermediate compound with an inferior molar concentration of diethylene glycol led to the formation of higher molecular weight NCO-terminated polyurethane prepolymer, which subsequently end-functionalized with APS coupling agent. This step was considered to take advantages of the intermolecular attraction of extra urethane bonds formed during chain extension reaction. The overall mechanical strength of dressing membrane was improved via increasing the inorganic siloxane domain content by addition of TEOS to the membrane base formulations.

GO nanoplatelets were added to the base formulation at different weight percent. The hydrolysis reaction of alkoxysilane groups present in Si-PU and TEOS and co-condensation of these materials with the hydroxyl groups already available on the surface of GO nanoplatelets led to the simultaneous formation of siloxane domain and integration of GO into the final polyurethane/siloxane networks. High gel content of these networks and their spectroscopic characteristic confirmed successful incorporation

4.2 Physical and mechanical properties of wound dressing membranes

Creation and kee** up of the optimal moist condition over injured tissue are generally accepted as the best state for wound healing process. In fact, cells can grow, proliferate and migrate at an increased rate to optimize the formation of new tissue under moist environment [34]. In addition, aqueous wound medium with several nutrients and vitamins is not only essential for cell metabolism and growth, but also provides a transport medium for a variety of bioactive molecules such as enzymes, growth factors, and hormones. The different cells in the wound area communicate with each other via these mediators, making sure that the healing processes proceed in a coordinated manner.

Application of proper wound dressing is the main tool for moist wound therapy. To fulfill this condition, the optimal wound dressing should establish a balanced ability between absorption of wound exudates and evaporative water loss from the wound site. The proper ability of XSi-PU membrane for the absorption of water was devoted to embedded PEG segments with high ability for creating hydrogen bonding type interaction with water molecules. As expected, after incorporation of GO with the hydrophobic nature into the dressing membrane formulation, the EWA significantly decreased.

Low WVTR causes the exudate build-up below a wound dressing which is harmful to healthy skin surrounding the wounded area, meanwhile, rapid water loss across dressing membrane due to its high WVTR can lead to surface dehydration of wounded area and scab formation. Therefore, optimal WVTR is required to keep moist environment over damaged tissue [39]. To find an insight regarding this important issue, the tensile property of dressings membranes was investigated (Fig. 1). Generally, the tensile strength, modulus and elongation at break of GO containing membranes were significantly higher than the neat network and these values increased with raising the GO nanoplatelets content up to a maximum of 5 wt.%. This behaviour could be related to the presence of well dispersed GO nanoplatelets in the nanocomposites membranes. The recorded result confirmed excellent interfacial adhesion and strong interaction between GO nanoplatelets and polyurethane/siloxane networks resulting from successful participation of GO in sol–gel reaction. The slight deterioration in the tensile property of the sample with the highest loading of GO nanoplatelets (XSi-PU/GO 7.5%) can be explained by irregular dispersion of GO and consequent formation of GO aggregates.

The applied dressing materials on the wounds site are exposed to the wound exudates. The absorbed liquid can lead to a serious drop in the overall mechanical strength of dressing due to deterioration of intermolecular attraction of chain segments of the polymeric network. Hence, the mechanical property of the dressing materials was also evaluated after immersing them into PBS at 37 °C (Fig. 1). As expected, due to plasticization effect of absorbed water molecules, the tensile strength of dressing membranes was reduced. The trend of changing in tensile property of samples at hydrated state was similar to those observed under dry condition. Both of elongation at break and tensile strength of the dressing membranes were increased up to a maximum of 5 wt.% of added GO. A slight decrease in tensile strength was detected at higher loading of GO. The reduction in tensile property of dressing membrane at hydrated state was lower for those containing GO and this difference in property of the samples at dry and hydrated states was also less important with increasing the concentration of GO. This phenomenon was attributed to the reinforcing effect of GO nanoplatelets and also to the lower water absorption of the samples containing the higher amount of GO. The measured tensile strength of the prepared dressing membranes (2.5 to 12 MPa at dry and 1 to 6 MPa at hydrated state) was comparable with the tensile strength of healthy skin (2.5–16 MPa). It is worth to notice that, the developed GO containing membranes showed superior tensile strength than some of the commercially available dressings like Resolut LT or Kaltostat (R), with tensile strength values of 1.7 and 0.9 MPa, respectively.

XRD analysis of the prepared dressings showed enhanced crystallinity of polyurethane-siloxane network containing GO nanoplatelets. This observation was attributed to the presence of exfoliated and well-dispersed GO nanoplatelets into the polyurethane-siloxane matrix. In fact, the GO sheets merged into the polyurethane-siloxane matrix during the sol–gel reaction and acted as nucleating agent. This recorded phenomenon was in accordance with the previous report regarding the significant increase in hard segment crystallinity of polyurethane film upon incorporation of GO [40]. The outcome of this happening was an enhancement of tensile strength for GO containing dressing. In fact, this modification increased the potential ability of dressing for providing better protection of wounded tissue against external forces. Indeed this single property is not enough for proper protection of wounds. Kee** moist and hygiene environment over wounded tissue have determining effects on appropriate healing of damaged skin. These extra properties were tuned by balancing the hydrophilicity-hydrophobicity ratio of dressing attained by chemical modification of dressing backbone and utilization of GO as an antimicrobial agent.

4.3 Biological properties of wound dressing membranes under in vitro condition

Application of antibacterial agents on wound dressing materials may deteriorate their cytocompatibility and consequently have an adverse effect on growth and proliferation of cells at the damaged tissue. Therefore, evaluation and controlling of this issue are crucial for the new compounds developed in the present study. Based on two set of experiments performed for evaluation of Fibroblast cells viability in contact with either dressings or leachates extracted from dressing, no toxicity was detected. For all samples, the Fibroblast cells continued their growth and proliferation without any problem. The recorded results implied that the based polyurethane/siloxane network selected for the preparation of dressing membranes was not toxic and the added GO at the studied range of concentration was not influenced on excellent cytocompatibility of theses dressing membranes.

It is well accepted that the bacterial or fungal infections can delay the wound healing process and even cause serious complications [41]. Covering of wounds with antimicrobial dressings is an effective methodology for preventing wound infection and subsequently accelerating the healing process [42]. Concentration-dependent antimicrobial activity was detected for the dressing membranes. The XSi-PU/GO2.5% sample showed moderate antibacterial activity, with efficiency in the range of 50–60 % against all of the studied strains. By increasing the GO nanoplatelets concentration, both of XSi-PU/GO5% and XSi-PU/GO7.5%samples showed the almost complete killing activity for the evaluated microorganisms. The recorded results were in accordance with recent findings regarding antibacterial [21] and antifungal [43] activities of GMs. A three-stage mechanism was proposed for antimicrobial action of GMs, which includes initial cell deposition on GMs, membrane stress brought on by direct contact with sharp nanosheets, and the consequent superoxide anion-independent oxidation [22]. It is more probable that for GO containing dressings, the incorporated GO nanosheets migrated to the surface of dressing membranes served as “cutters” to disrupt and damage bacteria membranes. This phenomenon led to the loss of membrane integrity and releasing of intracellular contents, and eventually microbial cells death. It seems, due to the low electrical conductivity of the prepared GO containing wound dressings (2 × 10−7 to 1 × 10−5 S cm−1) the second possible mechanism for antimicrobial activity of GO containing membrane i.e. oxidative stress mediated by GO through reactive oxygen species-independent oxidative stress, was less probable in our case.

Based on data recorded by cytocompatibility and antimicrobial assays, the sample XSi-PU/GO5% with optimum physicochemical, mechanical and biological properties was selected for the in vivo assay.

4.4 Evaluation of wound healing under in vivo condition

In vivo evaluation of the prepared dressing was performed on Wistar rat as an animal model (Fig. 3). Accelerated wound closure and better healing of wound dressed with XSi-PU/GO5% membrane were attributed to the favourable moist and warm environment over the wounded tissue, as well as maintaining a hygiene environment over wounded tissue due to the antimicrobial activity of this sample [44].

The results extracted from Fig. 4 showed promoted wound healing and higher quality of generated new skin for wound covered with GO containing wound dressing. Previous studies demonstrated that the re-construction of the epithelium and dermis, including derma papillae and hair follicles, are associated with the formation of blood vessels [44–46]. Therefore, enhanced epithelization of wounded tissue treated with XSi-PU/GO5% membrane can be attributed to the presence of additional blood vessels, however, secondary structures of skin like derma papillae and hair follicles are yet to be developed at this time period.

Higher vascularization and complete re-epithelialization of wounds covered with XSi-PU/GO5% membrane were attributed to enhanced migration of epithelial cells resulting from the maintenance of the moist environment at the wounds site [47, 48]. Collagen is an integral component of skin and plays a crucial role in the reconstruction of the dermis at the wound site and plays a pivotal role in maintaining skin structure [49–51]. Parallel alignment and higher maturation level of collagen were detected for MT stained section of wounds covered with XSi-PU/GO5%.

The in vivo test results revealed that the XSi-PU/GO5% dressing could encourage healing process in terms of increased rate of wound closure, collagen deposition, and re-epithelization. In addition to providing optimum moist environment over the damaged tissue, the enhanced healing efficacy of XSi-PU/GO5% dressing could be related to the promising antimicrobial activity of this sample. This result is in accordance with the previous report about the application of antimicrobial dressings. In fact, the reduction in the number of pathogenic microorganisms and consequently, reduction of the inflammatory response at wound site can promote the healing process.

5 Conclusion

People may die due to severe infection, improper protection of damaged tissue from external mechanical stress and most likely due to dehydration of the wounded skin. GO nanoplatelets as an antimicrobial and reinforcing agent were introduced into the backbone of novel dressing membranes composed of a polyurethane/siloxane network. DMA and XRD data revealed the proper incorporation and distribution of GO nanoplatelets throughout the matrix. Very good tensile strength and flexibility of the dressing after incorporation of a proper amount of GO at both dried and hydrated states confirmed the ability of dressings to protect damaged tissue from external forces. The balanced EWA and WVTR of dressings enabled preservation of suitable moist environment over wounds with the low to moderate exudates formation. The GO nanoplatelets incorporated dressings could successfully kill bacterial and fungal strains without showing adverse effects on growth and proliferation of dermal fibroblast cells. In vivo assay on rat model confirmed that the designed GO containing dressing could effectively stimulate wound healing process and remodeling of newly generated skin tissue.