Wound Repair and Regeneration

As antibiotic resistance continues to increase globally, there is an urgency for novel, non‐antibiotic approaches to control chronic drug‐resistant infections, particularly those associated with polymicrobial biofilm formation in chronic wounds. Also needed are clinically relevant polymicrobial biofilm models that can be utilized to assess the efficacy of innovative therapeutics against mature biofilms. We successfully developed a highly reproducible porcine ex vivo skin wound polymicrobial biofilm model using clinical isolates of multidrug‐resistant Pseudomonas aeruginosa, methicillin‐resistant Staphylococcus aureus, and Candida albicans. This ex vivo biofilm model was then used to assess the antimicrobial and antibiofilm properties of an easily fabricated chitosan hydrogel incorporating the natural antimicrobial peptide epsilon‐poly‐L‐lysine. Antimicrobial activity was evaluated against planktonic cultures in vitro and against mature biofilms ex vivo. The antibiofilm efficiency of the hydrogels was especially pronounced against Pseudomonas aeruginosa, whose counts were reduced by 99.98% after 2 hours in vitro and by 99.94% after treatment for 24 hours when applied to 24 hour ex vivo polymicrobial wound biofilms. The activity of the hydrogels was lower against Staphylococcus aureus and ineffective against Candida albicans. Gram, Hucker‐Twort staining of paraffin sections revealed balanced polymicrobial communities in mature 48 hour untreated biofilms. Treatment of 48 or 72 hour biofilms for 2 or 3 days with hydrogels that were applied within 5 hours after inoculation resulted in an impressive 96% and 97% reduction in biofilm thickness compared to untreated biofilms, respectively (P < .001). Likewise, topical gel treatment for 24 hours reduced biofilm thickness by 84% and 70%, respectively, when applied to mature biofilms at 24 and 48 hours after inoculation (P < .001). Thus, this ex vivo wound biofilm model provides a useful means to assess the efficacy of novel treatments to prevent and eradicate polymicrobial biofilms consisting of multidrug‐resistant Pseudomonas aeruginosa, methicillin‐resistant Staphylococcus aureus, and Candida albicans.

Here, we evaluate the efficacy of an emulgel dressing to control the release of an antifibrogenic factor, stratifin (SFN), along with an anti‐inflammatory drug, acetylsalicylic acid (ASA), to be used as a wound dressing with hypertrophic scar reducing features. Emulgel dressings were prepared by dispersing positively charged submicron vesicles in carboxymethyl cellulose gel. Release kinetics of SFN/ASA and toxicity for primary skin cells were assessed in vitro. Antifibrogenic efficacy of medicated emulgel dressings was tested on a rabbit ear fibrotic model. Following topical application on the wounds, emulgels formed an occlusive film and controlled the release of SFN and ASA for 7 and 24 hours, respectively. Wounds treated with SFN/ASA‐containing emulgel dressings showed an 80% reduction in scar elevation compared with untreated controls. Topical formulations were nontoxic for cultured human keratinocytes and fibroblasts. Inflammation was significantly controlled in treated wounds, as shown by a reduced number of infiltrated CD3⁺ T cells (p < 0.001) and macrophages. SFN/ASA‐treated wounds showed a significantly higher (p < 0.001) expression of matrix metalloproteinase‐1, resulting in reduced collagen deposition and less scarring. Film‐forming emulgel dressings that control the release of antifibrogenic and anti‐inflammatory factors provide an excellent treatment option for postburn hypertrophic scar management.

Chronic wounds expose the dermal matrix and underlying tissue to a diversity of microbes from the body and surrounding environment. We determined the microbial diversity of 19 chronic wounds using both molecular methods (sequence analysis of rRNA genes) and routine clinical culturing methods using swab samples. We identified 93 phylotypes in 2,653 rRNA clone sequences and found that compared with other environments, the microbial diversity of chronic wounds is relatively well characterized, i.e., 95% of sequences have ≥97% identity with known human commensals. In total, 75% of sequences belonged to four well‐known wound‐associated phylotypes: Staphylococcus (25%), Corynebacterium (20%), Clostridiales (18%), and Pseudomonas (12%). Approximately 0.5% of sequences (seven phylotypes) belonged to potentially new species. Individual wound samples contained four to 22 phylotypes, but in all wounds only a few (one to three) phylotypes were dominant. In more than half the wound specimens, polymerase chain reaction and culturing methods gave different diversity and dominance information about the microbes present. This exploratory study suggests that combining molecular and culturing methods provides a more complete characterization of the microbial diversity of chronic wounds, and can thereby expand our understanding of how microbiology impacts chronic wound pathology and healing.

Host responses to synthetic implants are analogous to healing, the process of repair that follows injury. Normally, the processes of wound healing follow well‐established patterns but conditions such as autoimmune diseases profoundly affect tissue repair. We have analyzed sponge‐induced wound healing responses in lupus‐prone New Zealand White and control (Balb/c) mouse strains by measuring inflammation, extracellular matrix deposition, angiogenesis, and cytokine production in polyether–polyurethane sponge implanted subcutaneously in male mice of these two strains. Although there was no difference in the gross appearance of the implants, further analysis of the wound healing responses, induced from 7 to 21 days post implantation, disclosed important differences between the New Zealand White and Balb/c strains. The intensity of inflammation (circulating tumor necrosis factor‐α and inflammatory leukocytes levels) was lower but implant fibrosis (collagen and transforming growth factor‐β1) was higher in New Zealand White, compared with Balb/c mice. Angiogenesis (hemoglobin, vascular endothelial growth factor, and vascularity) in New Zealand White implants peaked earlier than in Balb/c mice. In conclusion, we have shown that wound healing responses are clearly different in this strain of lupus‐prone mice and suggest that this pattern of repair was critically influenced by impaired inflammation and accelerated angiogenesis in the New Zealand White strain.

Although numerous factors are implicated in skin fibrosis, the exact pathophysiology of hypertrophic scarring remains unknown. We recently demonstrated that mechanical force initiates hypertrophic scar formation in a murine model, potentially enhancing cellular survival through Akt. Here, we specifically examined Akt‐mediated mechanotransduction in fibroblasts using both strain culture systems and our murine scar model. In vitro, static strain increased fibroblast motility, an effect blocked by wortmannin (a phosphoinositide‐3‐kinase/Akt inhibitor). We also demonstrated that high‐frequency cyclic strain was more effective at inducing Akt phosphorylation than low frequency or static strain. In vivo, Akt phosphorylation was induced by mechanical loading of dermal fibroblasts in both unwounded and wounded murine skin. Mechanically loaded scars also exhibited strong expression of α‐smooth muscle actin, a putative marker of pathologic scar formation. In vivo inhibition of Akt increased apoptosis but did not significantly abrogate hypertrophic scar development. These data suggest that although Akt signaling is activated in fibroblasts during mechanical loading of skin, this is not the critical pathway in hypertrophic scar formation. Future studies are needed to fully elucidate the critical mechanotransduction components and pathways which activate skin fibrosis.

Processes in the repair of a crevice in the knee joint meniscus were investigated in 10 dogs. Two 2‐mm cylindrical plugs from each medial meniscus were removed, rendered acellular by freezing and thawing, and then reinserted into the meniscus. Dogs were euthanized at intervals of 3–52 weeks after surgery. The crevice between the plug and meniscus at 3 weeks after surgery was filled with a tissue containing α‐smooth muscle actin‐positive cells. One year after surgery, the plug had remodeled and was populated with spindle‐shaped and fibrochondrocyte‐like cells. The plug had an appearance intermediate between that of hyaline and fibrocartilage at this time, with a seamless integration in sites between the remodeled plug and the surrounding meniscus. α‐smooth muscle actin‐positive cells were concentrated at the interface of the remodeled plug and adjacent meniscus and at the surface of the plug. Therefore, remodeling of both the plug and meniscal tissue and the participation of α‐smooth muscle actin‐positive cells appear essential for integration of the plug into the adjacent meniscal tissue. Cells in the superficial zone of the meniscus seem to be active in the repair process. A change in both the phenotype of the cells and the quality of the matrix toward a more hyaline state appears to be an integral part of the remodeling process in the meniscus.

Transforming growth factor‐β1 is a potent mediator of the differentiation of fibroblasts into myofibroblasts, which is characterized by the appearance of the cytoskeletal protein α‐smooth muscle actin. The aim of this study was to investigate the role of integrin extracellular matrix receptors in transforming growth factor‐β1–induced myofibroblast differentiation. We show that blockade of the αv and/or β1 integrins prevents the transforming growth factor‐β1–induced myofibroblast differentiation, seen by the increased expression of α‐smooth muscle actin and enhanced collagen gel contraction in three human fibroblast cell lines (from the mouth, skin, and kidney). Further, blockade of αv specific integrins αvβ5 and αvβ3 suppressed myofibroblast differentiation in fibroblasts from the mouth and skin; however, in the kidney cells, the prevention of differentiation was seen only with blockade of αvβ5 integrin but not αvβ3. A possible reason for this result may be different degrees of responsiveness to transforming growth factor‐β1 treatment seen from different anatomical origins of the cell lines. These data indicate a novel role for αv integrins in the differentiation of human fibroblasts from the mouth, skin, and kidney into myofibroblasts and suggest that there is a common differentiation pathway.

A systematic review and meta‐analysis were undertaken in order to explore the influence of matrix metalloproteinases and their diagnostic methods in chronic and acute wounds. Searches were conducted in the PubMed (Medline) and Embase (Elsevier) databases from inception to late November 2017. We included clinical trials enrolling patients with cutaneous chronic and acute wounds where a validated diagnostic method was employed for metalloproteinases. We excluded in vitro, animal or preclinical studies, nonoriginal articles, and studies without available data for analysis. In addition, references of narrative and systematic reviews were scrutinized for additional articles. Eight studies met the inclusion criteria. Results revealed that the most frequently determined matrix metalloproteinases were MMP‐2 and MMP‐9, and were found in 54.5% of wounds. MMP‐9 was present in more than 50% of the chronic wounds with a range from 37 to 78%. However, metalloproteinases were found in only 20% of acute wounds, and other types of metalloproteinases were also observed (MMP‐2 and MMP‐3). On the basis of the available evidence, high levels of metalloproteinases have been correlated with significantly delayed wound healing in wounds of a variety of etiologies.

Chronic wounds colonized with biofilm present a major burden to our healthcare system. While the current paradigm for wound healing is to maintain a moist environment, we sought to evaluate the effects of desiccation, and the ability of honey to desiccate wounds, on wound healing characteristics in Staphylococcus aureus biofilm wounds. In vivo biofilm wound healing after exposure to open‐air desiccation, honey, molasses, and saline was analyzed using a rabbit ear model of S. aureus biofilm wounds previously developed by our group. Wound morphology was examined using scanning electron microscopy and granulation tissue deposition was measured using light microscopy with hematoxylin and eosin staining. Viable bacterial counts in rabbit ear biofilm wounds and scabs were measured using a drop dilution method. In vitro S. aureus growth curves were established using tryptic soy broth containing honey and glycerol. Gene expression analysis of rabbit ear wounds was performed using reverse transcription quantitative PCR. Rabbit ear S. aureus biofilm wounds exposed to open‐air desiccation, honey, and molasses developed a dry scab, which displaced the majority of biofilm bacteria off of the wound bed. Wounds treated with open‐air desiccation, honey, and molasses expressed lower levels of the inflammatory markers tumor necrosis factor‐α and interleukin‐1β at postoperative day 12 compared with wounds treated with saline, and had increased levels of granulation tissue formation. In vitro growth of S. aureus in tryptic soy broth was inhibited by the presence of honey to a greater extent than by the presence of osmolality‐matched glycerol. Desiccation of chronic wounds colonized with biofilm via exposure to open air or honey leads to improved wound healing by decreasing bacterial burden and inflammation, and increasing granulation tissue formation. The ability of honey to help heal chronic wounds is at least in part due to its ability to desiccate bacterial biofilm, but other factors clearly contribute.

Honey has been used since ancient times for wound repair, but the subjacent mechanisms are almost unknown. We have tried to elucidate the modulatory role of honey in an in vitro model of HaCaT keratinocyte re‐epithelialization by using acacia, buckwheat, and manuka honeys. Scratch wound and migration assays showed similar increases of re‐epithelialization rates and chemoattractant effects in the presence of different types of honey (0.1%, v/v). However, the use of kinase and calcium inhibitors suggested the occurrence of different mechanisms. All honeys activated cyclin‐dependent kinase 2, focal adhesion kinase, and rasGAP SH3 binding protein 1. However, vasodilator‐stimulated phosphoprotein, integrin‐β3, cdc25C, and p42/44 mitogen activated protein kinase showed variable activation pattern. Re‐epithelialization recapitulates traits of epithelial‐mesenchymal transition (EMT) and the induction of this process was evaluated by a polymerase chain reaction array, revealing marked differences among honeys. Manuka induced few significant changes in the expression of EMT‐regulatory genes, while the other two honeys acted on a wider number of genes and partially showed a common profile of up‐ and down‐regulation. In conclusion, our findings have shown that honey‐driven wound repair goes through the activation of keratinocyte re‐epithelialization, but the ability of inducing EMT varies sensibly among honeys, according to their botanical origin.