Constitutive expression of hypoxia-inducible factor-1 α in keratinocytes during the repair of skin wounds in horses

The granulation (t)issue: A narrative and scoping review of basic and clinical research of the equine distal limb exuberant wound healing disorder

Exuberant granulation tissue (EGT) is often observed during second intention wound healing in horses. Despite its impact on wound care, the basic mechanisms leading to EGT are still unclear and effective strategies to prevent and/or treat EGT are lacking. The development of EGT is a poorly understood, multifactorial process involving hyperproliferating fibroblasts and malfunctional differentiation of keratinocytes, suboptimal wound contraction, dysfunctional vascularisation, and chronic inflammation. To consolidate and describe basic and clinical research literature on EGT and to identify knowledge gaps and opportunities for future research, a search was systematically conducted using predefined search terms. Subsequently, a scoping review was conducted using specific criteria to select the peer-reviewed literature that described methods to treat and/or prevent EGT. Proposed mechanisms of effects as well as results and main conclusions were extracted and tabulated. The systematic search resulted in 1062 publications in PubMed and 767 in Web of Science. Twenty additional studies were later included. Of these, 327 studies were reviewed for the narrative review on basic research and 35 controlled clinical trials were eligible for the scoping review. All 35 studies were conducted in university hospitals, and all but one involved surgically induced non-infected wounds. The study population was predominantly horses (n = 230) with a small number of ponies (n = 18) and donkeys (n = 14). In conclusion, there remains a strong need for evidence-based recommendations on EGT treatment, preferably using multi-centre studies that represent the general population of horses, include higher numbers of animals, and are performed in naturally occurring wounds. This narrative and scoping review also emphasises the importance of incorporating basic research knowledge in the study design of clinical trials.

Histologic changes and gene expression patterns in biopsy specimens from bacteria-inoculated and noninoculated excisional body and limb wounds in horses healing by second intention

OBJECTIVE

To evaluate histologic changes and gene expression patterns in body and limb wounds in horses in response to bacterial inoculation.

SAMPLE

Wound biopsy specimens from 6 horses collected on days 7, 14, 21, and 27 after excisional wounds (20 wounds/horse) were created over the metacarpal and metatarsal region and lateral thoracic region (body) and then inoculated or not inoculated on day 4 with Staphylococcus aureus and Pseudomonas aeruginosa.

PROCEDURES

Specimens were histologically scored for the amount of inflammation, edema, angiogenesis, fibrosis organization, and epithelialization. Quantitative PCR assays were performed to quantify gene expression of 10 inflammatory, proteolytic, fibrotic, and hypoxia-related markers involved in wound healing.

RESULTS

Except for gene expression of interleukin-6 on day 27 and tumor necrosis factor-α on day 14, bacterial inoculation had no significant effect on histologic scores and gene expression. Gene expression of interleukin-1β and -6, serum amyloid A, and matrix metalloproteinase-9 was higher in limb wounds versus body wounds by day 27. Gene expression of cellular communication network factor 1 was higher in limb wounds versus body wounds throughout the observation period.

CONCLUSIONS AND CLINICAL RELEVANCE

The lack of clear markers of wound infection in this study reflected well-known difficulties in detecting wound infections in horses. Changes consistent with protracted inflammation were evident in limb wounds, and gene expression patterns of limb wounds shared similarities with those of chronic wounds in humans. Cellular communication network factor warrants further investigation and may be useful in elucidating the mechanisms underlying poor limb wound healing in horses.

Epithelial-to-mesenchymal transition and keratinocyte differentiation in equine experimental body and limb wounds healing by second intention

BACKGROUND

The re-epithelialization process in equine wound healing is incompletely described. For epithelial cells to migrate during embryogenesis they undergo epithelial-to-mesenchymal transition (EMT); this phenotypic transition occurs during wound healing in humans and rodents, but it has not been investigated in horses.

HYPOTHESIS/OBJECTIVES

To investigate keratinocyte differentiation and EMT in equine experimental excisional limb and body wounds healing by second intention.

ANIMALS

Six adult research horses.

METHODS AND MATERIALS

Immunohistochemical analysis was used to detect expression of the differentiation markers cytokeratin (CK)10, CK14, loricrin and peroxisome proliferator-activated receptor alpha (PPAR-α), and of the EMT markers E-cadherin and N-cadherin in normal limb and body skin, and biopsies from limb and body wounds.

RESULTS

Loricrin and CK10 were expressed in normal skin and periwound skin but not in migrating epithelium of body and limb wounds. However, they reappeared at the migrating epithelial tip of body wounds only. CK14 and PPAR-α had uniform distribution throughout the migrating epithelium. N-cadherin was not expressed in normal unwounded skin but was detected in periwound skin adjacent to the wound margin. E-cadherin expression decreased at the wound margin.

CONCLUSIONS AND CLINICAL IMPORTANCE

Presence of N-cadherin suggests that cadherin switching occurred during wound healing, this may be an indication that EMT occurs in horses. To the best of the authors' knowledge, this has never been described in horses before and warrants further investigation to assess the clinical implications. The tip of the migrating epithelium in body wounds appeared more differentiated than limb wounds, which could be part of the explanation for the superior healing of body wounds.

Feasibility of a disposable canister-free negative-pressure wound therapy (NPWT) device for treating open wounds in horses

BACKGROUND

Wounds are among the most common medical conditions affecting horses and have a major economic impact on the horse industry. Wound healing in horses is distinct to that documented in other species, and often results in delayed healing and extensive scarring, with compromised functional and aesthetic outcomes. To date, there is no conventional method objectively proven to accelerate healing or to successfully prevent complications associated with second intention healing. Several effects of Negative Pressure Wound Therapy (NPWT) may be particularly useful to the management of wounds in horses. However, cumbersome designs of classic NPWT devices render them unsuitable for equine practice. A new lightweight, portable and disposable unit of NPWT (PICO®), should facilitate the use of this modality by equine practitioners. The aim of this study was to evaluate the feasibility of using this canister-free system to treat experimental open wounds in horses.

RESULTS

No difficulties were encountered with the application or maintenance of the PICO® system during the ex vivo experiment or during the preliminary in vivo experiment conducted on intact skin. All horses readily tolerated the PICO® but difficulties with adhesion and seal prevented the completion of the experimental wound study despite the use of many adjunctive adhesives.

CONCLUSION

The current PICO® dressing design is not suitable to be used as a dressing for open wounds in horses though the device is well tolerated by equine patients. A dressing with a wider adhesive edge, a superior adhesive and a more flexible pad would likely be better adapted to enable its future use in equine practice.

Treatment of limb wounds of horses with orf virus IL-10 and VEGF-E accelerates resolution of exuberant granulation tissue, but does not prevent its development

Bandaging of limb wounds in horses leads to formation of exuberant granulation tissue (EGT) that retards healing due to protracted inflammation, aberrant vascularisation and delayed epithelialisation. EGT is not observed if wounds are left undressed or when wounds are on the body. A previous study showed that short-term administration of proteins derived from orf virus dampened inflammation and promoted epithelialisation of open wounds in horses. Here, we investigated the impact of orf virus interleukin-10 and vascular endothelial growth factor-E on the development and resolution of EGT. Excisional wounds were created on the forelimb of four horses, and bandages were maintained until full healing to induce EGT formation. Matching body wounds were created to ensure EGT was limited to the limb, and to differentiate the effects of the viral proteins on normal healing and on EGT formation. Viral proteins or the hydrogel vehicle control were administered topically to site-matched wounds at day 1, with repeat administration at day 8. Wound healing and EGT formation were monitored macroscopically. Wound margin samples were harvested at 2, 7 and 14 days, and at full healing, with histology used to observe epithelialisation, immunofluorescence used to detect inflammatory cells, angiogenesis and cell death, and qPCR to measure expression of genes regulating inflammation and angiogenesis. Limb wounds developed EGT, and exhibited slower healing than body wounds. Viral protein treatment did not accelerate healing at either location nor limit EGT formation in limb wounds. Treatment of limb wounds did however increase epithelialisation and angiogenesis, without dampening inflammatory cell infiltration or gene expression. The healed wounds also had less occlusion and death of blood vessels and fewer epidermal rete ridges following viral protein treatment. These findings indicate that the viral protein treatment does not suppress wound inflammation or EGT formation, but does promote vascular and epidermal repair and EGT resolution.

A hypoxia response element in the Vegfa promoter is required for basal Vegfa expression in skin and for optimal granulation tissue formation during wound healing in mice

Hypoxia in skin wounds is thought to contribute to healing through the induction of hypoxia inducible factor-1 (HIF-1). Although HIF-1 can regulate the expression of vascular endothelial growth factor A (Vegfa), whether hypoxia and HIF-1 are required to induce Vegfa expression in the context of wound healing is unknown. To test this hypothesis, we evaluated Vegfa expression and wound healing in mutant mice that lack a functional HIF-1 binding site in the Vegfa promoter. Full-thickness excisional wounds were made using a biopsy punch, left to heal by second intention, and granulation tissue isolated on a time course during healing. mRNA levels of Vegfa and its target genes platelet-derived growth factors B (Pdgfb) and stromal cell-derived factor-1 (Sdf1) were measured by RT-qPCR, and HIF-1alpha and VEGFA protein levels measured by immunoblotting. Lower levels of Vegfa, Pdgf1 and Sdf1 mRNA were found in intact skin of mutant mice relative to wild-type controls (n = 6 mice/genotype), whereas levels in granulation tissue during wound healing were unaltered. VEGFA protein levels were also lower in intact skin of the mutant versus the wild-type mice. Decreased Vegfa mRNA levels in skin of mutant mice could not be attributed to decreased HIF-1alpha protein expression, and were therefore a consequence of the loss of HIF-1 responsiveness of the Vegfa promoter. Comparative histologic analyses of healing wounds in mutant and wild-type mice (n = 8 mice/genotype) revealed significant defects in granulation tissue in the mutant mice, both in terms of quantity and capillary density, although epithelialization and healing rates were unaltered. We conclude that HIF-1 is not a major regulator of Vegfa expression during wound healing; rather, it serves to maintain basal levels of expression of Vegfa and its target genes in intact skin, which are required for optimal granulation tissue formation in response to wounding.

Short-term treatment of equine wounds with orf virus IL-10 and VEGF-E dampens inflammation and promotes repair processes without accelerating closure

Healing is delayed in limb wounds relative to body wounds of horses, partly because of sustained inflammation and inefficient angiogenesis. In laboratory animals, proteins derived from orf virus modulate these processes and enhance healing. We aimed to compare immune cell trafficking and the inflammatory, vascular, and epidermal responses in body and limb wounds of horses and then to investigate the impact of orf virus interleukin-10 and vascular endothelial growth factor-E on these processes. Standardized excisional wounds were created on the body and forelimb of horses and their progression monitored macroscopically until healed. Tissue samples were harvested to measure the expression of genes regulating inflammation and repair (quantitative polymerase chain reaction) and to observe epithelialization (histology), innate immune cell infiltration, and angiogenesis (immunofluorescence). Delayed healing of limb wounds was characterized by intensified and extended pro-inflammatory signaling and exacerbated innate immune response, concomitant with the absence of anti-inflammatory eIL-10. Blood vessels were initially more permeable and then matured belatedly, concomitant with retarded production of angiogenic factors. Epithelial coverage was achieved belatedly in limb wounds. Viral proteins were administered to wounds of one body and one limb site/horse at days 1-3, while wounds at matching sites served as controls. Treatment dampened pro-inflammatory gene expression and the innate immune response in all wounds. It also improved angiogenic gene expression, but primarily in body wounds, where it altered blood vessel density and myofibroblast persistence. Moreover, the viral proteins increased epithelialization of all wounds. The short-term viral protein therapy did not, however, improve the healing rate of wounds in either location, likely due to suboptimal dosing. In conclusion, we have further detailed the processes contributing to protracted healing in limb wounds of horses and shown that short-term administration of viral proteins exerts several promising though transient effects that, if optimized, may positively influence healing.

Effect of activated protein C in second intention healing of equine distal limb wounds: a preliminary study

OBJECTIVE

To investigate the effect of activated protein C (APC) on second intention healing of distal limb wounds in horses.

METHODS

In this experimental study of eight Standardbred geldings, six full-thickness skin wounds (2 × 1.5 cm) were created on one metacarpus (biopsy limb) and five similar wounds were created on the contralateral metacarpus (photographed limb). Three wounds on the biopsy limb were treated topically with 190 µg APC on days 1, 3, 6 and 9, while the remaining three wounds were untreated (control). One treated and one control wound were biopsied on days 4, 7 and 11 for histopathology. Wounds on the photographed limb were treated with either 66% Manuka honey gel, a commercial antibiotic ointment (bacitracin-neomycin-polymixin B ointment; BNP) or petrolatum daily throughout healing, treated on days 1,3,6 and 9 with 190 µg APC or left untreated. These wounds were digitally photographed and the wound area measured on day 1, then weekly until day 49. Overall time to healing was recorded.

RESULTS

There was no effect of APC on wound size, the rate of healing or the overall time to heal. However, compared with control wounds, histological scoring demonstrated enhanced epithelialisation (day 4) and angiogenesis (day 11). Wound healing variables for wounds treated with APC, Manuka honey gel and control wounds were not different and the variables for wounds treated with BNP and petrolatum demonstrated delayed healing.

CONCLUSION

The improvements in histological scores in APC-treated wounds suggest further study into the effect of APC on second intention wound healing in horses is warranted.

Immunohistochemical Expression of Collagens in the Skin of Horses Treated with Leukocyte-Poor Platelet-Rich Plasma

This study evaluated the immunohistochemical expression of type I (COL I) and III (COL III) collagens during the healing process of skin treated with leukocyte-poor platelet-rich plasma (LP-PRP). Seven healthy gelding crossbred horses aged 16 to 17 years were used. Two rectangle-shaped wounds were created surgically in the right and left gluteal regions. Twelve hours after wound induction, 0.5 mL of the LP-PRP was administered in each edge of the wounds of one of the gluteal regions. The contralateral region was used as control (CG). Three samples were obtained: after wound induction (T0), 14 days (T1) of healing process, and after complete closure of the skin (T2). The normal skin (T0) showed strong staining for type III and I collagen in papillary and reticular dermis, respectively. In the scar of the treated group, COL III showed important (p < 0.05) increase in immunoreaction in T2 compared with T1. The administration of a single dose of LP-PRP 12 h after induction of wound in horses does not influence formation of collagens I and III. However, the intense labeling for COL III suggests that the tissue was still weak during the macroscopic closure of the wound, demonstrating that healing was not completely finished.

mRNA expression of genes involved in inflammation and haemostasis in equine fibroblast-like synoviocytes following exposure to lipopolysaccharide, fibrinogen and thrombin

Background

Studies in humans have shown that haemostatic and inflammatory pathways both play important roles in the pathogenesis of joint disease. The aim of this study was to assess mRNA expression of haemostatic and inflammatory factors in cultured equine fibroblast-like synoviocytes exposed to lipopolysaccharide (LPS), fibrinogen and thrombin. Synovial membranes were collected from metacarpo-phalangeal joints of 6 skeletally mature horses euthanized for non-orthopaedic reasons. Passage 4 fibroblast-like synoviocytes were left non-treated or treated with either 0.1 μg/ml LPS, 5 mg/ml fibrinogen or 5 U/ml thrombin and harvested at time points 0, 6, 24 and 48 h. mRNA expression of serum amyloid A (SAA), interleukin-6 (IL-6), monocyte chemotactic protein 1 (MCP-1), tissue factor (TF), plasminogen activator inhibitor 1 (PAI-1), urokinase plasminogen activator (uPA), vascular endothelial growth factor (VEGF) and protease activator receptor 1 (PAR-1) was assessed using quantitative real time reverse transcriptase PCR.

Results

LPS caused a significant increase in mRNA expression of SAA, IL-6, MCP-1 and uPA, and a decrease in TF, PAI-1 and PAR-1 when compared to non-treated cells. Treatment with thrombin resulted in increased mRNA expression of SAA, IL-6, MCP-1 and PAI-1, and a decreased PAR-1 expression compared to non-treated cells. The fibrinogen-treated synoviocytes showed significantly increased mRNA expression of IL-6, MCP-1, TF and PAI-1, and decreased PAR-1 expression compared to non-treated cells.

Conclusion

LPS, fibrinogen and thrombin induced an increased gene expression of inflammatory markers in isolated equine fibroblast-like synoviocytes. LPS caused changes in gene expression promoting increased fibrinolysis, while fibrinogen and thrombin changed the gene expression resulting potentially in reduced fibrinolysis. Overall, it appeared that both inflammatory and haemostatic stimuli affected expression of genes involved in inflammatory and haemostatic pathways, supporting their importance in equine joint diseases.

Quantificação de fatores de crescimento na pele de equinos tratada com plasma rico em plaquetas

O plasma rico em plaquetas (PRP) é um produto derivado da centrifugação do sangue total, sendo rico em fatores bioativos, como os de crescimento. Apesar da ampla utilização em processos cicatriciais, há controvérsia sobre a eficácia da terapia na cicatrização cutânea. O objetivo desse estudo foi quantificar e comparar a concentração dos fatores TGF-β1 e PDGF-BB no PRP, plasma sanguíneo e pele, durante diferentes fases do processo de cicatrização da pele tratada ou não com PRP. Foram utilizados sete equinos machos castrados, mestiços, hígidos, com idade entre 16 e 17 (16,14±0,63) anos. Três lesões em formato quadrangular (6,25cm²) foram produzidas cirurgicamente nas regiões glúteas direita e esquerda de todos os animais. Doze horas após indução das feridas, 0,5mL do PRP foi administrado em cada uma das quatro extremidades das feridas de uma das regiões glúteas (Grupo tratado = GT), escolhida aleatoriamente. A região contralateral foi utilizada como controle (GC). As feridas foram submetidas à limpeza diária com água Milli Q, e amostras foram obtidas mediante biópsias realizadas com Punch de 6mm. Foram obtidas seis biópsias de pele, sendo a primeira realizada logo após a produção da ferida (T0), e as demais com 1 (T1) 2 (T2) 7 (T3) e 14 (T4) dias após a indução da lesão. A sexta biópsia (T5) foi obtida após completo fechamento da pele, que ocorreu aproximadamente aos 37 dias (36,85±7,45, GC; 38,85±6,46, GT). Também foram obtidas amostras de sangue com EDTA em todos os tempos mencionados. A quantificação dos fatores de crescimento TGF-β1 e PDGF-BB na pele, PRP e plasma sanguíneo foi realizada pela técnica ELISA. Os dados foram analisados estatisticamente pelo teste t, correlação de Pearson e regressão, utilizando nível de significância de 5%. Não houve diferença entre os grupos, nos valores dos dois fatores de crescimento mensurados na pele, nos diferentes tempos. Também não houve correlação entre a quantidade dos fatores de crescimento presentes na pele e no plasma. Por outro lado, correlação positiva foi observada entre PRP e pele no grupo tratado, para os fatores de crescimento TGF-β1 (r=0,31) e PDGF-BB (r=0,38), bem como entre ambos os fatores de crescimento presentes no PRP (r=0,81). Considerando as concentrações dos fatores de crescimento no T0, os maiores valores cutâneos (p<0,05) do TGF-β1, em ambos os grupos, ocorreram nos tempos T3 e T5. Valores mais elevados (p<0,05) do PDGF-BB ocorreram no T4 (GT) e T5 (GC). No plasma não houve alteração nas concentrações desses fatores em relação ao T0, o que sugere que o PRP não acarreta efeito sistêmico, quando os procedimentos adotados na presente pesquisa são utilizados. A administração local de PRP no volume estudado, 12 h após indução cirúrgica de ferida cutânea na região glútea de equinos não ocasiona maiores concentrações dos fatores de crescimento TGF-β1 e PDGF-BB no plasma sanguíneo e pele, durante o processo de cicatrização.

Regulation of hypoxia-inducible factor-1α and related genes in equine digital lamellae and in cultured keratinocytes

REASONS FOR PERFORMING STUDY

Hypoxia-inducible factor-1α (HIF-1A) is an important protein in the regulation/induction of many genes in the cellular and tissue response to hypoxia and a central mediator in inflammatory signalling. As both hypoxia and inflammatory events are purported to occur in the lamellar epidermis in sepsis-related laminitis in the equid, HIF-1A may play a central role in this disease process.

OBJECTIVESS

To assess the regulation of HIF-1A and HIF-1A-related genes in the equine keratinocyte in vitro and in the lamellar tissue of horses with sepsis-related laminitis.

STUDY DESIGN

In vivo and in vitro experiments.

METHODS

Real-time quantitative PCR (RT-qPCR) and immunoblotting were performed to assess the mRNA and protein concentrations of HIF-1A and the mRNA concentrations of HIF-1A-related genes in cultured equine keratinocytes and in lamellar samples from black walnut extract (BWE)- and carbohydrate overload (CHO)-induced laminitis. Hypoxia-inducible factor-1α was further localised via indirect immunofluorescence in frozen lamellar tissue sections.

RESULTS

Hypoxia-inducible factor-1α appears to be regulated primarily at the post transcriptional level in the cultured equine keratinocyte, resulting in increased HIF-1A in response to hypoxia but not to lipopolysaccharide exposure. Hypoxia-inducible factor-1α is present at high concentrations in the normal equine lamina, and is increased in Obel grade 1 (OG1) stage laminitis in the CHO model of laminitis. Equine lamellar mRNA concentrations of cyclo-oxygenase-2 and inducible nitric oxide synthase, but not glucose transporter 1, are increased in the BWE and CHO models of laminitis.

CONCLUSIONS AND POTENTIAL RELEVANCE

These data indicate that the normal equine lamellae are profoundly hypoxic in comparison with other tissues. The increased mRNA concentrations of cyclo-oxygenase-2 and inducible nitric oxide synthase 2 in equine keratinocytes exposed to hypoxia and lipopolysaccharide, and in lamellar tissue from BWE and CHO models of sepsis-related laminitis, suggest that the marked lamellar inflammatory gene expression in sepsis-related laminitis may be due to an interaction of constitutively high lamellar keratinocyte HIF-1A signalling with inflammatory signalling, possibly induced by circulating inflammatory mediators.

Hypoxia regulates the expression of extracellular matrix associated proteins in equine dermal fibroblasts via HIF1

BACKGROUND

Exuberant granulation tissue (EGT), a fibrotic healing disorder resembling the human keloid, occurs almost exclusively in limb wounds of horses and may be caused in part by a relative state of hypoxia within the wound.

OBJECTIVE

The objectives of this study were therefore to (1) assess the effects of hypoxia on equine dermal fibroblast (EDF) proliferation and apoptosis, (2) study the effects of hypoxia on the expression of key extracellular matrix (ECM) associated proteins and determine if such effects are dependent on hypoxia-inducible factor (HIF), and (3) determine if EDFs from the body or limb respond differently to hypoxia.

METHODS

EDFs were isolated and cultured from skin from body or limb under normoxic or hypoxic conditions for up to 7days.

RESULTS

Hypoxia significantly stimulated EDF proliferation, but had no effect on cell survival. The hypoxia-mimetic agent CoCl(2) up-regulated COL1A1 expression and down-regulated MMP2 expression, suggesting an increase in ECM synthesis and a decrease in turnover. Both regulatory effects were inhibited by the addition of echinomycin, indicating that they are mediated by the transcriptional regulatory activity of HIF. No differences were observed between EDFs originating from body or limb for any effect of hypoxia or CoCl(2), suggesting that EGT development does not depend on intrinsic properties of limb fibroblasts.

CONCLUSIONS

We conclude that hypoxia regulates ECM remodeling via HIF1 in EDFs, and that this may be an important determinant in the pathogenesis of equine EGT.