Topical simvastatin promotes healing of Staphylococcus aureus‐contaminated cutaneous wounds

Abstract

Cutaneous wounds are prompt to be contaminated by bacteria, but the clinical benefits of applying antibiotics and antiseptics in wound management have not been proven. Statins are 3‐hydroxy‐3‐methylglutaryl‐coenzyme A (HMG‐CoA) reductase inhibitors commonly used to lower cholesterol levels. Studies indicated that statins, especially simvastatin, promote wound healing in experimental models. As Staphylococcus aureus is one of the most important microorganism responsible for wound infections, the aims of this study were to characterise the anti‐staphylococcal activity of simvastatin and to evaluate the application of simvastatin as a topical therapy for S. aureus‐contaminated wounds. In the present study, simvastatin was bacteriostatic against S. aureus at sub‐inhibitory concentrations up to 8 hours after exposure. Further increased concentrations of simvastatin above the minimal inhibitory concentration (MIC) did not enhance the growth inhibitory effect. By contrast, the ability of simvastatin to inhibit S. aureus biofilm formation was concentration dependent. Topical application of simvastatin at its MIC against S. aureus accelerated the healing and bacterial clearance of S. aureus‐contaminated wounds in an excisional mice wound model. This effective concentration is well below the safe concentration for topical use. Collectively, topical application of simvastatin has the potential as a novel modality for managing wound infections and promoting wound healing.

Background

Wounds, resulting from surgery, trauma events, burns, diabetes or chronic pressure, are prevalent clinical problems and are a burden to both the patients and the society 1. Wounds are vulnerable to bacterial insults, which disturb the healing process by inducing inflammation and tissue damage 2. The clinical benefit of using antimicrobials and antiseptics on open wounds is still controversial. Systemic or topical antibiotics have not been shown to promote wound healing, and the frequent applications of these agents can lead to the emergence of drug‐resistant microorganisms 1, 3. On the other hand, topical antiseptics, such as povidone‐iodine, chlorhexidine, alcohol, hydrogen peroxide and silver compounds, show toxicity to host cells, which potentially impedes the wound healing 3, 4, 5. Therefore, there is a need for better wound care modalities, which ideally comprise strategies both to decrease the bacterial contamination in the wound to prevent wound infections and to promote wound healing.

Statins are 3‐hydroxy‐3‐methylglutaryl‐coenzyme A (HMG‐CoA) reductase inhibitors, which are well known for their ability to lower serum cholesterol levels 6, 7, 8. The ability of statins to block the synthesis of many important non‐steroidal isoprenoids, while inhibiting mevalonate formation, has been suggested to account for the pleotropic effects of statins 9. Animal and observational studies have shown that statins are beneficial in promoting the rate of wound healing and wound strength 9, 10. A recent randomised, double‐blind, placebo‐controlled trial in patients with venous ulcers further demonstrated the beneficial effects of simvastatin on healing disorders in humans, in which patients given 40 mg simvastatin orally everyday had significant shorter wound healing time compared with the control patients 11. These data indicated that statin treatment has the potential to be a novel wound care modality, for healing disorders such as diabetic foot ulcers and venous ulcers. Statins have also been associated with reduced mortality in sepsis and infections in retrospective observational studies such as clinical cohort and case–control studies 12. This association, however, has not been confirmed in prospective randomised controlled trials. Experimental data on whether or not statin has therapeutic potential in managing infection are also limited.

In vitro minimal inhibitory concentration (MIC) values have provided experimental evidence on the direct antimicrobial properties of statins 13, 14. Masadeh et al. found that atorvastatin and simvastatin were potent against methicillin‐susceptible Staphylococcus aureus (MSSA), methicillin‐resistant S. aureus (MRSA) and other microorganisms commonly identified from wounds, such as Acinetobacter baumannii, Staphylococcus epidermidis and vancomycin‐susceptible and ‐resistant enterococci (VSE and VRE) 13. Fluvastatin and rosuvastatin, by contrast, did not exhibit comparable antimicrobial activity against MSSA and MRSA isolates as did simvastatin 13, 14. The mean MIC values of simvastatin against MSSA and MRSA reported in these studies were 29·2 and 75 µg/ml, respectively 14. Given the serum concentration of simvastatin of about 0·02 µg/ml, these MICs were inapplicable systemically in vivo 15. Rego et al. reported that topical simvastatin microemulsion (10 mg/ml) reduced bacterial loads, wound inflammation and neutrophil infiltration of polymicrobial‐contaminated skin wounds in a rat excisional model 16. To our knowledge, this study was the only direct assessment on the effects of statins on bacteria‐contaminated wounds. These studies provide important supporting evidence that the antimicrobial effect of statins, especially simvastatin, may offer additional clinical potential in managing infected skin wounds.

Given that S. aureus is one of the most prevalent causes of surgical wound infections 17, 18, 19, 20, 21, the objectives of the present study were to further characterise the anti‐staphylococcal activity of simvastatin and to examine the effect of topical application of simvastatin on cutaneous wounds contaminated with S. aureus.

Materials and methods

Chemicals and bacterial strains

S. aureus ATCC 29213 strain, a methicillin‐susceptible isolate originated from wound, was obtained from Bioresource Collection and Research Center, Taiwan (BCRC11863). Simvastatin, dimethyl sulfoxide (DMSO) and other chemicals used in this study were purchased from Sigma Aldrich (St. Louis, MO). Simvastatin stock was dissolved in 100% DMSO and stored in −20 °C as 25 mg/ml aliquots. For the bacterial experiments, the simvastatin stock was either added directly into the experimental groups or serial diluted with culture medium to obtain the designated concentrations. The vehicle control group contained the DMSO concentration as the highest simvastatin‐treatment group.

For the in vivo experiments, the stock solution was first diluted 1:200 with PBS to prepare 125 µg/ml working solution in 0·5% DMSO, and then diluted with 0·5% DMSO solution to 62·5 µg/ml. The solubility of simvastatin 125 µg/ml in 0·5% DMSO was verified by no visual precipitation.

MIC and bacterial growth inhibitory assays

The MIC of simvastatin against S. aureus was determined by broth microdilution method following the guidance from the Clinical and Laboratory Standards Institute (CLSI) 22. For bacterial growth inhibitory curves, bacterial macrodilution method was used. Briefly, one single colony of S. aureus was inoculated into Todd Hewitt broth and cultured at 37 °C with shaking at 200 rpm overnight. The bacterial pellet from the overnight culture was centrifuged with 16 000 g for 1 minute to pellet the bacteria. After removal of culture supernatant, the bacteria were then re‐suspended in fresh tryptic soy broth (TSB; Bacto, BD, Franklin Lakes, NJ) to 10⁶ CFU/ml. After the exposure of simvastatin, the bacteria were cultured at 37 °C with shaking at 200 rpm. The effects of simvastatin on bacterial growth were examined at times 0, 2, 8 and 24 hours by plating.

Antibiofilm assay

The biofilm protocol was adopted from Smeltzer and coworkers 23. Briefly, overnight culture of S. aureus was pelleted and reconstituted into biofilm‐conditioned broth (TSB with 0·5% glucose and 3% NaCl). The bacterial suspension was then diluted 1:200 with the biofilm‐conditioned broth and placed at a concentration of 100 µl/well into 96‐well tissue culture microtitre plates pre‐coated with 10% human plasma (Sigma, St. Louis, MO). To assess the effect of simvastatin on biofilm formation, the drug was added into the wells 1 hour after bacterial inoculation. After 24 hours of drug exposure, the biofilms were gently washed two times with 200 µl PBS to remove non‐adherent cells using a multi‐channel pipette. The biofilm was fixed with 100% ethanol for 10 minutes and then stained with 0·41% crystal violet in 12% ethanol for 2 minutes at room temperature. The stained wells were washed three times with PBS and air‐dried for 20 minutes before adding 100% ethanol to solubilise the crystal violet. The liquid was transferred to another microtitre plate and the absorbance at 595 nm was recorded by an ELISA plate reader.

Mouse wound infection model

The experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Kaohsiung Medical University. Male BALB/c mice were obtained from the National Laboratory Animal Center (NLAC, Taipei City, Taiwan). Mice were aged between 6 and 12 weeks at the time of the study. All mice were housed individually in P2 animal facility to prevent fighting and cross contamination after wound creation.

The protocol was adapted from Guo et al. with minor adjustments 24. Briefly, the mice were given intraperitoneal injection of 50 mg/kg Zoletil™ and subcutaneous injection of 5 mg/kg nalbuphine as anaesthesia and analgesia. The back of the mice were shaved and sterilised with 10% povidone‐iodine. Two 6‐mm full‐thickness excisional wounds were created on the back of each mouse, each side of the spine, using biopsy punches. The wounds were contaminated with 10⁶ CFU bacteria in 10 µl PBS. After 30 minutes of the inoculation, 20 µl of vehicle (0·5% DMSO in PBS) or simvastatin (62·5 µg/ml in 0·5% DMSO) was pipetted onto the wounds. Each mouse served as its own healing control, one of the wound received vehicle treatment and one of the wound received simvastatin treatment, to minimise the inter‐individual variability. The treatments were administered onto the wounds on days 0, 1, 3, 5, 7 and 9. The mice were euthanised on day 10. The data presented in this study were pooled from two independent experiments. Pictures of wounds were recorded by camera. Quantification of wound size and bacteria was performed by using ImageJ software version 1·46 (NIH, Bethesda, MD). Wound closure was presented as percentage of the wounded area on day 0. At the end of the study, the scars or the remaining wounds with surrounding healthy tissues were harvested for histology. Haematoxylin and eosin (HE) staining and Gram staining were performed to visualise tissue morphology and bacteria localisation, respectively.

To quantify the microvessels in the wounded area, pictures of three fields from each wound specimen were taken and examined under 200× microscopy (0·94 mm² per field) by a pathologist. Microvessel density was determined as the mean number of all vessels in the observation areas. Large vessels with thick muscular walls were excluded from the counts. To quantify the bacteria remaining in the wounded areas, pictures of at least four fields were taken from each specimens under oil‐immersion magnification of 1000×. The total pixels in Gram‐positive areas (dark green dots) were measured in all the fields and then normalised by the observed length of the wounds in the sections.

Statistical analysis

Statistical data were analysed and graphed by using GraphPad Prism Software Version 5.01 (GraphPad Software Inc., La Jolla, CA). One‐way ANOVA with Dunnett's multiple comparison test was used to compare the results in groups. For comparing the size of the wounds in the same mice, paired Student's t‐test was used. Data were represented as mean ± SEM. Statistical significant was set as P value <0·05.

Results

The anti‐staphylococcal effect of simvastatin

The MICs of simvastatin and vancomycin against S. aureus were 62·5 µg/ml and 1 µg/ml, respectively. At 2 hours, simvastatin significantly decreased the density of S. aureus at the concentration higher than 16 µg/ml. The anti‐staphylococcal effect of simvastatin was dose dependent at sub‐MIC concentrations; however, simvastatin concentration higher than 62·5 µg/ml did not exhibit better anti‐staphylococcal effect (Figure 1). The anti‐staphylococcal effect of simvastatin was further characterised up to 24 hours by bacterial growth inhibitory curve (Figure 2). Simvastatin at the concentration of up to 125 µg/ml inhibited S. aureus growth up to 8 hours. Simvastatin at 62·5 µg/ml significantly inhibited bacterial growth at 8 hours (6·6 ± 2·1 versus 9·3 ± 0·1 log CFU/ml in vehicle control group), where the growth inhibitory effect was better than that by vancomycin at its MIC. Together, simvastatin was bacteriostatic against S. aureus up to 8 hours, and the growth inhibitory effect was dose dependent up to the MIC. Concentrations much higher than the MIC did not improve its anti‐staphylococcal effect.

Figure 1.

Simvastatin exhibited anti‐staphylococcal effect 2 hours after exposure. The suspension was cocultured with simvastatin or vehicle dimethyl sulfoxide (DMSO). Bacterial densities were determined by plating. Asterisks represent statistically significant reduction in bacterial density from the vehicle control. Error bars represent SEM (N = 3).

Figure 2.

Simvastatin is bacteriostatic against Staphylococcus aureus. S. aureus was diluted to the density of 10⁶ CFU/ml in Todd‐Hewitt broth. The suspension was cocultured with simvastatin, vancomycin or vehicle control dimethyl sulfoxide (DMSO). Data from representative experiments are presented. Asterisks represent statistically significant reduction in bacterial density from the vehicle control. Error bars represent SEM (N = 6).

The effect of simvastatin on biofilm formation

The ability of simvastatin to inhibit S. aureus biofilm formation was evaluated by in vitro microplate assay as described above in the Materials and Methods section. At concentrations higher than 62·5 µg/ml, simvastatin significantly inhibited S. aureus biofilm formation (Figure 3). Vancomycin at the growth inhibitory concentrations also inhibited biofilm formation in a concentration‐dependent manner. Both simvastatin and vancomycin had no effect on existing biofilm (data not shown).

Figure 3.

Simvastatin inhibited biofilm formation in a concentration‐dependent manner. (A) The representative crystal violet‐stained biomass remaining on the microplate assay after simvastatin treatment. (B) Quantitative results of the microplate assay. The simvastatin data are from two independent experiments (n = 14 for simvastatin; n = 8 for vancomycin). Asterisks represent statistically significant reduction of biomass from the vehicle control. Error bars represent SEM.

The effect of topical simvastatin on the healing of S. aureus‐contaminated wounds in mice excisional wound model

A mice excisional wound model was used to assess the effects of topical simvastatin on S. aureus‐inoculated wounds. The wounds were inoculated with 10⁶ CFU S. aureus and then treated with vehicle control (0·5% DMSO) or simvastatin. The size of the wounds was monitored at days 1, 3, 7 and 10. By day 10, most of the wounds were completely healed. Three out of six (50%) control wounds still had scabs, whereas all simvastatin‐treated wounds were completely reepithelialised.

The wound area of the simvastatin‐treated wounds was about 15% smaller than that of the control wound on day 1. The average wound size of simvastatin‐treated wounds was consistently smaller than that of the vehicle‐treated group till the end of the experiment (Figure 4A). Significant smaller size of the wounded area in simvastatin‐treated group was observed on days 3 and 7. Wounds of the representative mice in each group are shown in Figure 4B.

Figure 4.

Topical simvastatin promoted the healing of Staphylococcus aureus‐contaminated wounds in vivo. (A) Wounded areas on days 1, 3, 7 and 10. Open circle denotes vehicle control group; close square denotes simvastatin 62·5 µg/ml‐treated group (n = 6). Data were pooled from two independent experiments. Error bars represent SEM. (B) Representative photos of the wounds. Scale bar = 5 mm. Asterisks represent statistically significant difference from the vehicle control group by paired t‐test.

Gram staining results showed that S. aureus can still be observed in the wounded area at day 10 (Figure 5). Most bacteria were localised in the corneal layers of the wounded area. Some hair follicles surrounding the wounded area were also contaminated with bacteria. There were visually more bacteria that remained in the corneal layers of the vehicle‐treated wounds compared with the simvastatin‐treated wounds under microscopy. Significantly higher pixels of dark green dots, which indicated Gram‐positive bacteria in the Gram‐stained sections, were measured in vehicle‐treated wounds compared with the simvastatin‐treated wounds. A great variation in the amounts of bacteria that remained in the vehicle‐treated wounds was noted.

Figure 5.

Simvastatin‐treated wounds showed less bacteria in histology at day 10. Representative images of Gram‐stained histological sections from S. aureus‐contaminated wounds at day 10. Panel (A) shows 1000× magnified vehicle [0·5% dimethyl sulfoxide (DMSO)]‐treated wound and panel (B) shows 1000×‐magnified simvastatin 62·5 µg/ml‐treated wound. Bar = 50 µm. The dark green circular dots are S. aureus (indicated by the arrows in panels A and B). Panel (C) shows the quantitative result of bacteria in the wounds. Bacteria number was estimated by measuring pixels of the green dots. The asterisk represents statistically significant difference in bacteria compared with the vehicle group.

The histological examination of the back skin tissues harvested at day 10 showed that the healed regions in general had a significantly thickened epidermal layer compared with the uninjured skin regions. At day 10, the simvastatin‐treated wounds demonstrated complete wound healing with the appearance of hair follicles, whereas the vehicle‐treated wounds showed thickened and prominent epidermal layers as well as granular layers (Figure 6). Simvastatin‐treated group showed no difference in microvessel density in the dermal and hypodermal regions of the wounded areas compared with the vehicle‐treated group at day 10.

Figure 6.

Simvastatin‐treated wounds showed better healing in histology at day 10. Representative images of haematoxylin and eosin‐stained histological sections from Staphylococcus aureus‐contaminated wounds at day 10. Panel (A) shows 200× magnified vehicle [0·5% dimethyl sulfoxide (DMSO)]‐treated wound and panel (B) shows 200×‐magnified simvastatin 62·5 µg/ml‐treated wound. Bar = 100 µm. Panel (C) depicts the quantitative result of microvessel density in the wounds (indicated by the arrows in panels A and B). Microvessel density was quantified as vessels/mm² in the 200×‐magnified fields.

Discussion

Cutaneous wound healing is a complex biological process orchestrated by the coordination of multiple cells. Bacterial stimulation, whether or not induced clinical symptoms, causes inflammation and induces cytotoxicity to the cells, thereby delaying wound healing 25, 26. In the present study, we used S. aureus, a clinically important microorganism responsible for wound infections, to characterise the antimicrobial activity of simvastatin and provide evidence that topical simvastatin has the potential as a novel wound care modality for S. aureus‐contaminated wounds.

Simvastatin has been indicated to possess antibacterial effects against various organisms in vitro 13, 14, 27. The MIC values from these studies indicated that the antimicrobial effect of simvastatin differs significantly depending on the organisms examined. Simvastatin was shown to be bacteriocidal against Streptococcus pneumonia and Moraxella catarrhalis; by contrast, the growth of Haemophilus influenza was not affected by simvastatin 27. Both the MICs of simvastatin against S. pneumonia and M. catarrhalis were 15 µg/ml. Simvastatin has been shown to possess antimicrobial activity against multiple representative pathogens with the MICs ranging from 26 (against MSSA) to 167 µg/ml (against MRSA) 13. The mode of action of simvastatin against S. aureus has not been reported. In the present study, we chose a clinical wound isolate with the MIC of simvastatin close to the reported mean MIC for MRSA 14. It was observed that simvastatin at sub‐MIC concentrations exhibited bacteriostatic effect against S. aureus, and the growth inhibitory effect lasted up to 8 hours. The growth inhibitory activity of simvastatin against S. aureus was surprisingly similar to that of vancomycin at its MIC concentration within 8 hours. Moreover, the ability of both simvastatin and vancomycin to inhibit biofilm formation was dependent on bacterial growth inhibition. The data showed that simvastatin is bacteriostatic against S. aureus and the anti‐staphylococcal activity of simvastatin is concentration dependent only at sub‐inhibitory concentrations. Further increased simvastatin concentrations did not appear to increase the anti‐staphylococcal effect. Because simvastatin is a prodrug, the simvastatin used in the present study was not activated in vitro, and not all clinically effective statins have comparable in vitro anti‐staphylococcal effect; the anti‐staphylococcal activity of simvastatin shown in this study is independent of its HMG‐CoA reductase inhibitory activity. Our Gram staining results of the wounds, which showed fewer bacteria remaining in simvastatin‐treated wounds compared with vehicle‐treated wounds, provided evidence that the experimental concentration of simvastatin had anti‐staphylococcal effect in vivo in the animal model.

Simvastatin has been indicated to have effects on innate immunity and neutrophil functions. Simvastatin at a concentration as low as 1 μM, which is ∼0·4 µg/ml, has been shown to reduce neutrophil recruitment and activation without affecting neutrophil phagocytosis and bacterial killing activities in an endotoxin‐induced lung injury model 28. Simvastatin at 4 µg/ml was shown to reduce phagocytosis and oxidative burst but increase extracellular traps of S. aureus 29. Wounds treated with a much higher dose of simvastatin (10 mg/ml) also showed that the simvastatin‐treated wounds had reduced inflammation, neutrophil infiltration and bacterial loads 16. Given that the concentration we examined in this study was in the middle of the reported range, we reasonably assumed that simvastatin used in our experiments reduced neutrophil infiltration and wound inflammation without a negative impact on bacterial clearance by innate immunity. Therefore, the significantly reduced bacterial amounts in the simvastatin‐treated wounds observed in this study could also be influenced by simvastatin's effect on innate immunity as well as the faster wound repair process.

There is also increasing evidence of statins, systemic or topical applications, on the healing of wounds. Topical application of simvastatin (10 mg petroleum gel containing 50 µg of simvastatin every 3 days) was shown to promote wound closure in excisional diabetic wound healing model 30. Intraperitoneal administration of simvastatin at 5 mg/kg in mice with incisional diabetic skin wound demonstrated that simvastatin‐treated wounds have more breaking resistance than the control groups 31. A recent double‐blind placebo‐controlled trial further indicated that simvastatin given orally at therapeutic dose promoted the healing rate of venous ulcers 11. Promoting macrophage infiltration, inducing vascular endothelial growth factor (VEGF) production, decreasing inflammation and promoting reepithelialisation have been indicated as possible mechanisms of statins to promote wound healing 9, 30, 31, 32, 33. These pleotropic effects have mainly been attributed to the statins' ability to reduce isoprenylation of downstream targets of the mevalonate pathway, although alternative mechanisms, such as statins' ability to bind to several nuclear hormone receptors, have been proposed 10. Detailed reviews on current evidence of statins in skin healing disorders have recently been performed by Stojadinovic et al. 10 and Farsaei et al. 34. In our study, we did not observe increased microvessel density in simvastatin‐treated wounds at day 10, despite the significantly improved healing time and better tissue repair in histology. Whether or not the simvastatin concentration promotes angiogenesis at any time points during the process will require further investigation.

Among the experimental conditions in the healing of skin wounds, the potential application of simvastatin on infected wounds has only been previously evaluated by Rego et al. in a polymicrobial‐contaminated wound model 16. The wounds treated with simvastatin 10 mg/ml topically showed reduced bacterial loads, inflammation and neutrophil infiltration in the wounds. In their study, rat faeces were used to contaminate the wounds, thereby the microorganisms contaminating the wounds could not be specified. By contrast, our study used S. aureus, a clinically relevant microorganism, to show the effectiveness of simvastatin at much lower concentration that can still promote bacterial clearance and the healing of bacteria‐contaminated wounds. The significant healing rate difference between the simvastatin‐ and vehicle‐treated wounds was strong evidence that the topical simvastatin treatment, at only one‐tenth concentration of previously reported effective concentration, is overall beneficial for the healing of S. aureus‐contaminated wounds. The relative contribution of direct anti‐staphylococcal effect of simvastatin and its effects on host cells in reducing bacterial loads in the S. aureus‐infected wounds are worthy of further investigation.

Statins are potent inhibitors of HMG‐CoA reductase, and the exposure to high concentrations of statins topically can decrease the production of epidermal cholesterol by keratinocytes and result in barrier disruption 35. Furthermore, biphasic effects of simvastatin on host cells have been noted in in vitro experiments where high doses of statins induced cell apoptosis and inversely inhibited angiogenesis 10, 36. A recent study performed by Adami et al. also showed that topical application of 1% simvastatin (about 10 mg/ml) ointment twice daily for five consecutive days would not cause skin barrier disruption on inflamed skin 33. The results of our study indicated that topical application of simvastatin at such a lower concentration, compared with concentrations higher than 1%, is effective and has minimal safety concerns.

To sum up, this study supports the topical application of simvastatin as a safe and promising modality for the prevention and management of S. aureus wound infection. Topical application of simvastatin promotes the healing of S. aureus‐contaminated wounds through multiple aspects, including direct antibacterial activity, inflammation modulation and promoting wound healing. Such wound care strategy could potentially minimise the application of antibiotics in wound management and could also be applied to diabetic foot care. Further clinical trials are warranted to validate the findings in humans.