To assess the antibacterial efficacy of new composite materials developed from microparticles of 45S5 bioactive glass (BG) and agar–gelatin films.
To assess the antibacterial efficacy of new composite materials developed from microparticles of 45S5 bioactive glass (BG) and agar–gelatin films.
In vitro antibacterial activity was evaluated against Staphylococcus spp. because of the importance of this pathogen in damaged tissues and in failures associated with biomaterial implants. To our knowledge, this is the first paper reporting on the suitable combination of BG and agar–gelatin for bioactive and antibacterial films. Bacterial suspensions up or below 105 CFU ml−1 reflecting situations of wound infection and of noninfection, respectively, were prepared and then put in contact with the biomaterials at 37°C. After 24 and 48 h of incubation, the pH value was measured and the staphylococci strains viability was determined by counting in Mueller–Hinton agar plates. Moreover, the biomaterials were prepared for observation under scanning electron microscopy (SEM). Biocomposites (BCs) showed a strong antibacterial effect against all staphylococci strains tested. Some differences were found depending on the strain, the inoculum size and the contact time. This effect was correlated with an alkalinization of the media. By SEM analyses, no bacterial presence was observed on the surface of BCs in any of the cell concentrations tested at any time.
Overall, the coating of 45S5 BG on agar–gelatin films promoted BCs with strong antistaphylococcal activity. The effect was efficient under bacterial concentration up or below 105 CFU ml−1. Additionally, none of the strains were found on BCs surfaces.
45S5 bioglass/agar–gelatin biocomposite films are reported for the first time. The results suggest a potential application as wound dressing.
Staphylococcus aureus and Staphylococcus epidermidis are Gram-positive bacteria that are normal colonizers of human skin and mucous membranes, but can become pathogenic in the presence of wounds causing in some cases serious diseases (Lowy 1998; Mack et al. 2007). They are also associated with infection problems during orthopaedic failure of implants (Krimmer et al. 1999; Campoccia et al. 2008; Campoccia et al. 2009). Treatment of these infections is associated with high complication rates and implies in many cases prolonged hospital stay, increased morbidity and mortality, and serious economic costs (Gollwitzer et al. 2003). The infections may be difficult to treat with traditional antibiotics due to the emergence of some antibiotic resistance like methicillin-resistant S. aureus (MRSA) or because of biofilms formation (Raja et al. 2011).
Diverse strategies have been developed to avoid the staphylococci survival and adhesion on wound and implants including the development of antibacterial drug-delivery systems using different substrates (Itokazu et al. 1997; Schierholz and Beuth 2001; Gollwitzer et al. 2003; Von Eiff et al. 2005) or the alteration of the surface topography to prevent bacterial adhesion (Balazs et al. 2003; Chua et al. 2008). The disadvantages of these methods are the cytotoxicity of some antibacterial agents (Albers et al. 2013) or the persistent risk of an antibiotic resistance. Indeed many of them have a negative effect on the attachment of the host cells (Misra et al. 2010).
Bioactive glasses (BGs) are inorganic material with the ability to stimulate specific cellular responses at the molecular level as a result of the controlled release of ions from them (Hench and Polak 2002; Hoppe et al. 2011). The first system, 45S5 Bioglass®, constituted by (in wt%) 45% SiO2, 24·5% Na2O, 24·5% CaO and 6% P2O5 was developed by Hench and coworkers in 1971 (Hench et al. 1971). Subsequently, other compositions were obtained by replacing or incorporating different ions (Brink et al. 1997; Gorustovich et al. 2006, 2010; Vallet-Regí et al. 2006). 45S5 BG has been successfully employed to fill bone defects in clinic for its ability to promote osteogenesis. In addition, it was also demonstrated that 45S5 BG enhances the wound healing of soft tissues (Moosvi and Day 2009; Lin et al. 2012). On the other hand, numerous works reported on the antibacterial potential of 45S5 BG particles (Allan et al. 2001; Waltimo et al. 2007, 2009; Hu et al. 2009) including those of staphylococci (Hu et al. 2009; Misra et al. 2010). Biofilm viability was also prevented by particulate 45S5 BG (Allan et al. 2002). In recent years, 45S5 BGs have been incorporated into many natural or synthetic polymers in order to improve some characteristic of both materials like delivery or degradation (Blaker et al. 2005; Day et al. 2005; Hong et al. 2009; Gentile et al. 2010; Marelli et al. 2010). In that sense, some biocomposites (BCs) with antistaphylococcal activity were reported (Pratten et al. 2004; Misra et al. 2010). Nevertheless, to date, there have been no scientific reports of BCs based on 45S5 BG and agar–gelatin hybrids. Gelatin is obtained by thermal denaturation, physical or chemical degradation of collagen and has been used for medical applications such as wound dressings and adsorbent pads during surgery (Sakai et al. 2007). It has gained interest in biomedical engineering due to its low cost, wide commercial availability and characteristic of biocompatibility and biodegradability (Gentile et al. 2010). The disadvantage of gelatin is that it dissolves at the temperature of the human body as well as temperatures used for culturing cells (Van den Bosch and Gielens 2003; Sakai et al. 2007). Agar is a polysaccharide with higher thermostability than gelatin. Besides, agar gels behave like a sponge. Even though, agar has no significant moieties for the adhesion and proliferation of cells (Gruber et al. 1997).
Agar–gelatin scaffolds have been developed with promising application in tissue engineering (Sakai et al. 2007), evaluation of the toxicity of drugs and chemicals (Verma et al. 2009) and drug delivery (Saxena et al. 2011; Shome et al. 2011). The cytocompatibility of different weight ratio of these hybrids was investigated, and films and scaffolds containing agar and gelatin in 2 : 1 weight ratio exhibited the best growth kinetics of mouse fibroblast cell line NIH3T3 (Verma et al. 2007).
Wound infections are defined as a bacterial count over 105 micro-organisms g−1 tissue (Robson et al. 1999; Edwards and Harding 2004; O'Meara et al. 2006; Jacobsen et al. 2011). In the present work, micrometre 45S5 BG was employed to coat agar–gelatin films, and a series of experiments were carried out to evaluate the antibacterial properties of the BCs against four staphylococci strains. Two different inoculum sizes of Staphylococcus spp. strains were assessed to reflect a case of infection and one of noninfection.
Melt-derived 45S5 BG micrometre particles were in range size of 5–100 μm. The composition was (in%w/w): 45% SiO2, 24·5% Na2O, 24·5% CaO, 6% P2O5. Agar-agar was purchased from Britannia S.A. (Buenos Aires, Argentina). Edible gelatin Royal was obtained from Mondelez International (Buenos Aires, Argentina).
The agar–gelatin films were prepared in 2 : 1 weight ratio. Briefly, agar–gelatin 2 : 1 ratio was dissolved in distilled water making 1% homogenous solution. Twenty millilitres was poured in a Petri dish plate and kept in an incubator for 24 h at 30°C. Then, 0·54 g of 45S5 BG was suspended in 25 ml of isopropanol alcohol and finally incorporated to agar–gelatin gel. Once BG microparticles precipitated, the excess of isopropanol was removed with a pipette (this process accelerates the drying time). The new BCs were then incubated for 24 h to complete the drying. Subsequently, they were cut in discs shape of 5 mm of diameter. For biological assays, BCs and control films discs were UV sterilized for 20 min on each side. To determine the weight of each disc, coated and uncoated samples discs were weighed. The subtraction of both averages allowed determining the average weight in mg of 45S5 BG in each disc.
The materials obtained were morphologically characterized by scanning electron microscopy (SEM). For this, biomaterials were fixed with a 2·5% glutaraldehyde 0·1 mol l−1 PBS solution overnight at 4°C. The samples were then washed with distilled water and sequentially dehydrated through a graded series of ethanol solutions. After mounting on stubs and gold sputtering, the samples were examined with a scanning electron microscope (JSM 6480 LV, JEOL Ltd, Tokyo, Japan).
The following strains were used in this study: S. aureus ATCC29213, S. aureus ATCC25923, S. aureus ATCC6538P and S. epidemidis ATCC12228. All strains were grown for 24 h in Muller–Hinton broth (Britannia S.A.) at 37°C. For the experiments, the bacterial cell suspensions were diluted to 4 and 6 log CFU ml−1, approximately.
The experiments were carried out in Hank′s balanced saline solution (HBSS) without Ca2+ and Mg+ (Life Technologies, Carlsbad, CA, USA). The biomaterials (3 discs) were incubated for 48 h at 37°C in 1 ml of cellular suspensions. Each staphylococci suspension in absence of biomaterial served as controls. Samples were collected after 24 and 48 h of incubation, and the viability of cells at 37°C were assessed by counting in Muller–Hinton agar plates. Also, at the end of each period, the pH value of the culture was determined. The results were expressed as log10 CFU ml−1 ± SD. On the other hand, the BCs and uncoated films were prepared for SEM observation. Previously, samples were rinsed in distilled water and vortexed for 1 min to wash away free bacteria.
Statistical analysis was performed using SPSS 15·0 statistical package software (IBM, Armonk, NY, USA) with appropriate statistical tests such as one-way analysis of (anova) with Dunnett's and Tukey's multiple comparison post-tests for intergroup analysis. Specifically, the data from cell suspension were used as a controland compared to the coated and uncoated films. The level of significance was set at a P-value of <0·05.
The mean weight of BCs discs was 2·93 ± 0·60 mg, of the uncoated films was 0·10 ± 0·01 mg. This means that the 45S5 BGs incorporation on each disc was 2·83 ± 0·59 mg.
The microstructure of agar–gelatin films and BCs are shown in Fig. 1. Figure 1(a) corresponds to agar–gelatin pure films. It can be seen that the surface of pure film is homogenous. Also, it is a continuous matrix without cracks with good structural integrity. Figure 1(b) shows the morphology of BCs. A homogeneous and continuous coating can be observed; pores were created as a consequence of the irregular size presented by 45S5 BG microparticles.
The antibacterial effects of BCs at different incubation periods are shown in Fig. 2. The initial cell concentration of the strains expressed as log CFU ml−1 was: S. epidermidis ATCC1228, 4·00 ± 0·43; S. aureus ATCC25923, 3·77 ± 0·34; S. aureus ATCC6538P, 3·65 ± 0·30; and S. aureusATCC29213, 3·82 ± 0·25. No statistical differences were found in the inoculum size between the strains, and pure films had no effects on the viability of the strains. On the other hand, the BCs strongly inhibited the cell growth of all strains. The inhibition increased with the time of exposition. After 24 h of incubation, the cell viability of the strains was reduced to 2·30 log CFU ml−1. While, after 48 h, S. epidermidis and S. aureus ATCC25923 were the most sensitive to BCs because their viability was below the detection limit of the technique (<2 CFU ml−1). In the case of S. aureus ATCC6538P, the bacterial count was reduced to 0·70 ± 0·38 log CFU ml−1. Even though S. aureus ATCC29213 was the least sensitive of the strains, at the end of the assays, the cell count for this strain was 1·25 ± 0·07 log CFU ml−1.
In all cases, it was observed a positive correlation between the growth inhibition of the cells and the alkalinization of the media culture containing BCs, indicating that the aqueous pH values increased with the increase of the incubation periods. Table 1 shows the values of these modifications.
|Conditions||Time of incubation (h)|
The results are shown in Fig. 3. The initial cell concentration expressed as log CFU ml−1 was: S. epidermidis ATCC1228, 6·40 ± 0·15; S. aureus ATCC25923, 6·16 ± 0·30; S. aureus ATCC6538P, 6·24 ± 0·09; and S. aureusATCC29213, 6·39 ± 0·13. No statistical differences were found in the inoculum size between the strains. Similar to the results shown above, the BCs strongly inhibited Staphylococcus spp. cell viability. The growth of S. epidermidis ATCC12228 and S. aureus ATCC25923 was the most strongly inhibited in presence of BCs after 24 h. After this period time, the cell viability was 3·80 ± 0·11 log CFU ml−1 for S. epidermidis and 2·90 ± 0·16 log CFU ml−1 for S. aureus ATCC25923. That means a reduction of 2·61 and 3·26 log CFU ml−1. After 48 h, no significant differences were found in cell viability for these strains.
After 24 h, BCs inhibited S. aureus ATCC6538P and ATCC29213 cells almost 1 order of magnitude. However, after 48 h, the cell viability for S. aureus ATCC6538P was 2·90 ± 0·45 and 4·25 ± 0·14 log CFU ml−1 for S. aureus ATCC29213. Once again, this last strain results the less sensitive to BCs at the end of the experiments. The pH of the culture also constantly increased during the incubations periods (see Table 1).
Representative morphological changes in samples with conditioning time are illustrated in Fig. 4. No cells of Staphylococcus spp. were found on BC-coated surfaces after 24 or 48 h post-incubation (a,b). No cells were found either on the uncoated faces of BCs (c). When bacterial cells were incubated with control samples, S. aureus ATCC 29213 cells were found on gelatin–agar surface, but only after 48 h of incubation period and when the inoculum size were the highest. It was the only case of bacterial presence found on agar–gelatin films. None of the rest of the strains was found on pure films.
In this work, the antibacterial effects of novel 45S5 BG/agar–gelatin films against four staphylococci strains at different incubation periods were investigated. The experiments were performed at bacterial concentrations up and below 105 CFU reflecting an infective and a subinfective level of bacteria, respectively (Robson et al. 1999; Edwards and Harding 2004; Jacobsen et al. 2011). It has been proposed that subinfective levels would accelerate wound healing and formation of granulation tissue and an increase of collagen formation (Edwards and Harding 2004). The results obtained in this work showed that BCs were highly effective in keeping or reducing the bacterial number to this level, even though some differences were obtained depending on the strain and the incubation periods.
The antistaphylococcal effects observed in this study are in good agreement with a previous investigation demonstrating that a biocomposite made of Poly(3-hydroxybutyrate) and nanoparticles of 45S5 BG (P(3HB)/n-BG 10% wt) inhibited the cell growth of S. aureus NCTC6571. The inhibitory effects in that work increased with the incubation periods (up to 48 h) and also were in correlation with a pH increment of the media culture (Misra et al. 2010).
Also, the results of the current study for ATCC29123 concur with those of Hu et al. who reported a strong antibacterial effects in vitro of 45S5 microparticles (<50 μm) against this strain and also against a strain of S. epidermidis (0·5–2 × 108 CFU ml−1) in a high alkaline environment (Hu et al. 2009).
Nevertheless, Bellantone et al. reported that 45S5 particles (90–710 μm) had no effect in vitro on S. aureus NCIMB11852 even at a 10 mg ml−1 of concentration (Bellantone et al. 2002). Also, Gorriti et al. (2009) reported that scaffolds made from 45S5 BGs did not exhibit antibacterial effect against S. aureus ATCC25923, ATCC29213 and ATCC6538P after 1 and 24 h of incubation period. Finally, Xie et al. (2008) showed that 300 mg of 45S5 BG (355–500 μm) failed to prevent in vivo S. aureus ATCC25923 infection of open tibial fractures in rabbits; these results were obtained for a single S. aureus strain and a single body site, a limitation that was recognized by the authors. Moreover, the common feature of all these works was the lack of pH increments in milieu in presence of 45S5 BG that could explain the discrepancies with the results of this work. In fact, the available evidence strongly suggests that the increase in aqueous pH value plays a critical role in 45S5 BG antibacterial effects (Allan et al. 2001; Hu et al. 2009) because it is well known that in general, a high alkaline environment is not well tolerated by the micro-organisms (Waltimo et al. 2007). The optimum pH value for the growth of staphylococci cells is between 7·0 and 7·5 (Misra et al. 2010). Thus, the increase in the pH value during the time of incubation found in this work explains, in part, the cell growth inhibition. Furthermore, there was a positive correlation between the antibacterial effects with the increase in pH during the incubation periods. Other reasons of the antibacterial properties may be related to an increment of the ionic strength, as the leaching of ions occurs from the BGs (Stoor et al. 1998), the high concentration of calcium and silica that could inhibit bacterial viability (Stoor et al. 1999; Zehnder et al. 2006), and also a physical damage to cell wall coming from BG debris could be related to its antibacterial effect (Hu et al. 2009).
It has been proved that the adhesion of bacteria on to mammalian tissue surfaces or biomaterials constitutes an important initial step in the pathogenesis of an infectious process (Von Eiff et al. 2005). The SEM observations showed that up to 48 h, none of the cells of the strains tested were present on BCs surfaces supporting the strongly antibacterial effects determined by plate count. These results are consistent with those found by Gorriti et al. on S. aureus ATCC 29213 (Gorriti et al. 2009). In other works, 45S5 BG just limited the bacterial attachment when present on Poly(3-hydroxybutyrate) (Misra et al. 2010) or sutures (Pratten et al. 2004). Conversely, it has been reported that 45S5 BG presented a dose-dependent and bacteria-dependent bacterial adhesion (Hu et al. 2009). On the other hand, when cells were incubated with uncoated films, a differential behaviour among S. aureus ATCC29213 and the rest of the staphylococci tested was detected. S. aureus ATCC29213 were found on pure films only in the case of infection model and after 48 h. It is well known that bacterial adhesion is a complex process affected by many factors such as bacterial properties (hydrophobicity, cell concentration), environmental conditions (ionic strength, pH, etc.) and the material surface characteristics (regularities and porosity) (Katsikogianni and Missirlis 2004; McWhirter et al. 2002; Dass et al. 2009; Ketonis et al. 2010; Crawford et al. 2012).
Taken an overview of the results, the BCs may have a potential application as wound dressing as they have some advantageous features. Gelatin possesses the RGD (arginine-glycine-aspartic acid) sequences of collagen responsible of the cell adhesion and proliferation (Rosellini et al. 2009). Furthermore, gelatin results more convenient than collagen because it is known to have no antigenicity and also results more economical than collagen (Ulubayram et al. 2001). Because agar gels behave like a sponge, it will be ideal for adsorbing wound exudates. Another plus for these BCs is that agar–gelatin 2 : 1 weight ratio was proved to have good cytocompatibility on eukaryotic cells (Verma et al. 2007, 2009). Finally, there have been no reports of bacterial resistance to BG, and the preparation costs are relatively inexpensive compared to other antibacterial agents such as silver (Ong et al. 2008; Madhumathi et al. 2010) antibiotics (Choi et al. 1999; Denkbas et al. 2004) or iodine (Ignatova et al. 2008; Misra et al. 2010). Moreover, angiogenic effects have been reported in vitro and in vivo for 45S5 system (Day et al. 2004, 2005; Gorustovich et al. 2010) that could enhance the wound healing process.
Finally, it may be discussed if the the harmful effect in bacteria cells could be also negative for eukaryotic cells because of the alkaline pH. In reference to this, Misra et al. 2010; reported that P(3HB)/Bioglass BCs exhibited bactericidal properties in alkaline environment. In vitro study demonstrated that the biomaterials were suitable for MG-63 osteoblast cell attachment and proliferation. Also, when the foams were implanted in rats as subcutaneous implants resulted in a nontoxic and foreign body response after 1 week of implantation. Other investigations have shown that in fact, some healing processes such as the take rate of skin-grafts requires an alkaline milieu (Schneider et al. 2007).
This work was supported by the Consejo Nacional de Investigaciones Científicas y Técnicas, CONICET (PIP0184 to A.A.G). The authors declare no conflict of interest related to this work.