Kai P. Leung

Development of a novel, highly quantitative in vivo model for the study of biofilm-impaired cutaneous wound healing

A growing body of evidence suggests that in addition to hypoxia, ischemia‐reperfusion injury, and intrinsic host factors, bacterial biofilms represent a fourth major pillar in chronic wound pathogenesis. Given that most studies to date rely on in vitro or observational clinical data, our aim was to develop a novel, quantitative animal model enabling further investigation of the biofilm hypothesis in vivo. Dermal punch wounds were created in New Zealand rabbit ears, and used as uninfected controls, or inoculated with green fluorescent protein‐labeled Staphylococcus aureus to form wounds with bacteria predominantly in the planktonic or biofilm phase. Epifluorescence and scanning electron microscopy revealed that S. aureus rapidly forms mature biofilm in wounds within 24 hours of inoculation, with persistence of biofilm viability over time seen through serial bacterial count measurement and laser scanning confocal imaging at different time points postwounding and inoculation. Inflammatory markers confirmed that the biofilm phenotype creates a characteristic, sustained, low‐grade inflammatory response, and that over time biofilm impairs epithelial migration and granulation tissue in‐growth, as shown histologically. We have established and validated a highly quantitative, reproducible in vivo biofilm model, while providing evidence that the biofilm phenotype specifically contributes to profound cutaneous wound healing impairment. Our model highlights the importance of bacterial biofilms in chronic wound pathogenesis, providing an in vivo platform for further inquiry into the basic biology of bacterial biofilm–host interaction and high‐throughput testing of antibiofilm therapeutics.

Bacteriophage therapy for Staphylococcus aureus biofilm-infected wounds: a new approach to chronic wound care.

Background: Bacterial biofilms, which are critical mediators of chronic wounds, remain difficult to treat with traditional methods. Bacteriophage therapy against biofilm has not been rigorously studied in vivo. The authors evaluate the efficacy of a species-specific bacteriophage against Staphylococcus aureus biofilm–infected wounds using a validated, quantitative, rabbit ear model. Methods: Six-millimeter dermal punch wounds in New Zealand rabbit ears were inoculated with wild-type or mutant, biofilm-deficient S. aureus. In vivo biofilm was established and maintained using procedures from our previously published wound biofilm model. Wounds were left untreated, or treated every other day with topical S. aureus–specific bacteriophage, sharp débridement, or both. Histologic wound healing and viable bacterial count measurements, and scanning electron microscopy were performed following harvest. Results: Wild-type S. aureus biofilm wounds demonstrated no differences in healing or viable bacteria following bacteriophage application or sharp débridement alone. However, the combination of both treatments significantly improved all measured wound healing parameters (p < 0.05) and reduced bacteria counts (p = 0.03), which was confirmed by scanning electron microscopy. Bacteriophage treatment of biofilm-deficient S. aureus mutant wounds alone also resulted in similar trends for both endpoints (p < 0.05). Conclusions: Bacteriophages can be an effective topical therapy against S. aureus biofilm–infected wounds in the setting of a deficient (mutant) or disrupted (débridement) biofilm structure. Combination treatment aimed at disturbing the extracellular biofilm matrix, allowing for increased penetration of species-specific bacteriophages, represents a new and potentially effective approach to chronic wound care. These results establish principles for biofilm therapy that may be applied to several different clinical and surgical problems.

In vivo modeling of biofilm-infected wounds: a review.

Chronic wounds continue to represent a difficult and complex problem for both patients and healthcare providers. Bacterial biofilms represent a critical component of nonhealing wounds, utilizing several different mechanisms to inhibit innate inflammatory pathways and resist traditional therapeutics. Although in vitro biofilm systems have been well described and studied, understanding the intricacies of wound biofilm pathology requires appropriate in vivo models to understand the interactions between bacteria and host. In an effort to clarify the available literature, this review describes and critically evaluates all of the in vivo wound biofilm models currently published to-date, including model advantages and clinical applicability. We will also address the need for continued therapeutic development and testing using these currently available in vivo models.

Comparative Analysis of Single-Species and Polybacterial Wound Biofilms Using a Quantitative, In Vivo, Rabbit Ear Model

Introduction The recent literature suggests that chronic wound biofilms often consist of multiple bacterial species. However, without appropriate in vivo, polybacterial biofilm models, our understanding of these complex infections remains limited. We evaluate and compare the effect of single- and mixed-species biofilm infections on host wound healing dynamics using a quantitative, in vivo, rabbit ear model. Methods Six-mm dermal punch wounds in New Zealand rabbit ears were inoculated with Staphylococcus aureus strain UAMS-1, Pseudomonas aeruginosa strain PAO1, or both, totaling 10∧6 colony-forming units/wound. Bacterial proliferation and maintenance in vivo were done using procedures from our previously published model. Wounds were harvested for histological measurement of wound healing, viable bacterial counts using selective media, or inflammatory cytokine (IL-1β, TNF-α) expression via quantitative reverse-transcription PCR. Biofilm structure was studied using scanning electron microscopy (SEM). For comparison, biofilm deficient mutant UAMS-929 replaced strain UAMS-1 in some mixed-species infections. Results Bacterial counts verified the presence of both strains UAMS-1 and PAO1 in polybacterial wounds. Over time, strain PAO1 became predominant (p<0.001). SEM showed colocalization of both species within an extracellular matrix at multiple time-points. Compared to each monospecies infection, polybacterial biofilms impaired all wound healing parameters (p<0.01), and increased expression of IL-1β and TNF-α (p<0.05). In contrast, mixed-species infections using biofilm-deficient mutant UAMS-929 instead of wild-type strain UAMS-1 showed less wound impairment (p<0.01) with decreased host cytokine expression (p<0.01), despite a bacterial burden and distribution comparable to that of mixed-wild-type wounds. Conclusions This study reveals that mixed-species biofilms have a greater impact on wound healing dynamics than their monospecies counterparts. The increased virulence of polybacterial biofilm appears dependent on the combined pathogenicity of each species, verified using a mutant strain. These data suggest that individual bacterial species can interact synergistically within a single biofilm structure.

Deficient cytokine expression and neutrophil oxidative burst contribute to impaired cutaneous wound healing in diabetic, biofilm-containing chronic wounds.

Diabetic patients exhibit dysregulated inflammatory and immune responses that predispose them to chronic wound infections and the threat of limb loss. The molecular underpinnings responsible for this have not been well elucidated, particularly in the setting of wound biofilms. This study evaluates host responses in biofilm‐impaired wounds using the TallyHo mouse, a clinically relevant polygenic model of type 2 diabetes. No differences in cytokine or Toll‐like receptor (TLR) expression were noted in unwounded skin or noninoculated wounds of diabetic and wild‐type mice. However, diabetic biofilm‐containing wounds had significantly less TLR 2, TLR 4, interleukin‐1β, and tumor necrosis factor‐α expression than wild‐type wounds with biofilm (all p < 0.001). Both groups had similar bacterial burden and neutrophil infiltration after development of biofilms at 3 days postwounding, but diabetic wounds had significantly less neutrophil oxidative burst activity. This translated into a log‐fold greater bacterial burden and significant delay of wound epithelization for biofilm‐impaired diabetic wounds at 10 days postwounding. These results suggest that impaired recognition of bacterial infection via the TLR pathway leading to inadequate cytokine stimulation of antimicrobial host responses may represent a potential mechanism underlying diabetic susceptibility to wound infection and ulceration.

Topically Delivered Adipose Derived Stem Cells Show an Activated-Fibroblast Phenotype and Enhance Granulation Tissue Formation in Skin Wounds

Multipotent mesenchymal stem cells (MSCs) are found in various tissues and can proliferate extensively in vitro. MSCs have been used in preclinical animal studies and clinical trials in many fields. Adipose derived stem cells (ASCs) have several advantages compared to other MSCs for use in cell-based treatments because they are easy to isolate with relative abundance. However, quantitative approaches for wound repair using ASCs have been limited because of lack of animal models which allow for quantification. Here, we addressed the effect of topically delivered ASCs in wound repair by quantitative analysis using the rabbit ear model. We characterized rabbit ASCs, and analyzed their multipotency in comparison to bone marrow derived-MSCs (BM-MSCs) and dermal fibroblasts (DFs) in vitro. Topically delivered ASCs increased granulation tissue formation in wounds when compared to saline controls, whereas BM-MSCs or DFs did not. These studies suggest that ASCs and BM-MSCs are not identical, though they have similar surface markers. We found that topically delivered ASCs are engrafted and proliferate in the wounds. We showed that transplanted ASCs exhibited activated fibroblast phenotype, increased endothelial cell recruitment, and enhanced macrophage recruitment in vivo.

Quantitative Comparison and Analysis of Species-Specific Wound Biofilm Virulence Using an In Vivo, Rabbit-Ear Model

BACKGROUND Although bacterial biofilm is recognized as an important contributor to chronic wound pathogenesis, differences in biofilm virulence between species have never been studied in vivo. STUDY DESIGN Dermal punch wounds in New Zealand white rabbit ears were inoculated with Klebsiella pneumoniae, Staphylococcus aureus, or Pseudomonas aeruginosa, or left uninfected as controls. In vivo biofilm was established and maintained using procedures from our previously published wound biofilm model. Virulence was assessed by measurement of histologic wound healing and host inflammatory mediators. Scanning electron microscopy (SEM) and bacterial counts verified biofilm viability. Extracellular polymeric substance (EPS)-deficient P aeruginosa was used for comparison. RESULTS SEM confirmed the presence of wound biofilm for each species. P aeruginosa biofilm-infected wounds showed significantly more healing impairment than uninfected, K pneumoniae, and S aureus (p < 0.05), while also triggering the largest host inflammatory response (p < 0.05). Extracellular polymeric substance-deficient P aeruginosa demonstrated a reduced impact on the same quantitative endpoints relative to its wild-type strain (p < 0.05). CONCLUSIONS Our novel analysis demonstrates that individual bacterial species possess distinct levels of biofilm virulence. Biofilm EPS may represent an integral part of their distinct pathogenicity. Rigorous examination of species-dependent differences in biofilm virulence is critical to developing specific therapeutics, while lending insight to the interactions within clinically relevant, polybacterial biofilms.

Exacerbated and prolonged inflammation impairs wound healing and increases scarring

Altered inflammation in the early stage has long been assumed to affect subsequent steps of the repair process that could influence proper wound healing and remodeling. However, the lack of explicit experimental data makes the connection between dysregulated wound inflammation and poor wound healing elusive. To bridge this gap, we used the established rabbit ear hypertrophic scar model for studying the causal effect of dysregulated inflammation. We induced an exacerbated and prolonged inflammatory state in these wounds with the combination of trauma‐related stimulators of pathogen‐associated molecular patterns from heat‐killed Pseudomonas aeruginosa and damage‐associated molecular patterns from a dermal homogenate. In stimulated wounds, a heightened and lengthened inflammation was observed based on quantitative measurements of IL‐6 expression, tissue polymorphonuclear leukocytes infiltration, and tissue myeloperoxidase activity. Along with the high level of inflammation, wound healing parameters (epithelial gap and others) at postoperative day 7 and 16 were significantly altered in stimulated wounds compared to unstimulated controls. By postoperative day 35, scar elevation of stimulated wounds was higher than that of control wounds (scar elevation index: 1.90 vs. 1.39, p < 0.01). Moreover, treatment of these inflamed wounds with Indomethacin (at concentrations of 0.01, 0.1, and 0.4%) reduced scar elevation but with adverse effects of delayed wound closure and increased cartilage hypertrophy. In summary, successful establishment of this inflamed wound model provides a platform to understand these detrimental aspects of unchecked inflammation and to further test agents that can modulate local inflammation to improve wound outcomes.

Impact of a novel, antimicrobial dressing on in vivo, Pseudomonas aeruginosa wound biofilm: Quantitative comparative analysis using a rabbit ear model

The importance of bacterial biofilms to chronic wound pathogenesis is well established. Different treatment modalities, including topical dressings, have yet to show consistent efficacy against wound biofilm. This study evaluates the impact of a novel, antimicrobial Test Dressing on Pseudomonas aeruginosa biofilm‐infected wounds. Six‐mm dermal punch wounds in rabbit ears were inoculated with 106 colony‐forming units of P. aeruginosa. Biofilm was established in vivo using our published model. Dressing changes were performed every other day with either Active Control or Test Dressings. Treated and untreated wounds were harvested for several quantitative endpoints. Confirmatory studies were performed to measure treatment impact on in vitro P. aeruginosa and in vivo polybacterial wounds containing P. aeruginosa and Staphylococcus aureus. The Test Dressing consistently decreased P. aeruginosa bacterial counts, and improved wound healing relative to Inactive Vehicle and Active Control wounds (p < 0.05). In vitro bacterial counts were also significantly reduced following Test Dressing therapy (p < 0.05). Similarly, improvements in bacterial burden and wound healing were also achieved in polybacterial wounds (p < 0.05). This study represents the first quantifiable and consistent in vivo evidence of a topical antimicrobial dressings impact against established wound biofilm. The development of clinically applicable therapies against biofilm such as this is critical to improving chronic wound care.

Understanding the host inflammatory response to wound infection : An in vivo study of Klebsiella pneumoniae in a rabbit ear wound model

Wound infection development is critically dependent on the complex interactions between bacteria and host. Klebsiella pneumoniae has become an increasingly common wound pathogen, but its natural history within wounds has never been studied. Using a validated, in vivo rabbit ear model, wounds were inoculated with K. pneumoniae at different concentrations (102–107 colony‐forming units) with measurement of viable and nonviable bacterial counts, histological wound‐healing parameters, and host inflammatory gene expression at multiple time points postinoculation (48, 96, and 240 hours). Bacteria and wound morphologies were evaluated with scanning electron microscopy. Comparable experiments were performed in ischemic ears to model immune response impairment. All wounds, despite different inoculants, equilibrated to similar bacterial concentrations by 96 hours. With a 106 colony‐forming units inoculant, wounds at 240 hours showed decreased bacterial counts (p < 0.01), with a corresponding improvement in healing (p < 0.01) and a decrease in inflammatory response (p < 0.05). In contrast, ischemic wounds revealed impaired inflammatory gene expression (p < 0.05) resulting in higher steady‐state bacterial concentrations (p < 0.01), impaired healing (p < 0.05), and biofilm formation on scanning electron microscopy. We conclude that a normal inflammatory response can effectively stabilize and overcome a K. pneumoniae wound infection. An impaired host cannot control this bacterial burden, preventing adequate healing while allowing bacteria to establish a chronic presence. Our novel study quantitatively validates the host immune response as integral to wound infection dynamics.