EUROPEAN  TISSUE  REPAIR  SOCIETY

RABBIT MODELS OF HEALING
Jeff Davidson, Professor of Pathology, Vanderbilt University School of Medicine, Nashville, USA

The rabbit ear excisional wound is a very useful experimental tool. By careful dissection of epidermis and dermis to the depth of the ear cartilage, one produces an ulcer-like lesion that has an avascular base and that is unable to close by contraction.
These properties differentiate this wound from most other small animal models and provide a greater resemblance to human wound closure. The ear wound is very amenable to recurrent treatment and dressing changes, in some cases with only modest restraint. Many parameters can be obtained from such wounds: morphometric estimates of wound closure and wound filling; non-invasive measurement of tissue perfusion or blood flow; histomor-phometric measurement of epithelization, granulation tissue and angiogenesis; biochemical measurement of wound constituents; quantitative and qualitative evaluation of gene expression. The wound healing progression also passes through a late stage of hypertrophy that may be useful for evaluation of anti-scarring agents.
Four wounds can readily be produced in each ear, providing sufficient numbers for within-animal replicates and dose-ranging. Healing can be further impaired by local ischemia, systemic factors or chemically induced diabetes. In addition, this model shows an age-dependent decline in healing capacity.

PIG MODELS OF HEALING
Esther Middelkoop, Dutch Burns Foundation, Beverwijk, The Netherlands

Wound healing studies in humans are often limited to non-invasive techniques. Although skin is readily accessible, quantitative non-invasive methods are sparse. Therefore, it is advantageous if one can study wound healing in an animal model. Some of the advantages are:
o one can compare different treatments in a standardised fashion in the same animal
o it is possible to take repeated tissue samples for histological, immunohistochemical, biochemical (e.g. enzymatic) or molecular biological analysis
o variation in wounds in a model is less than in a clinical setting
o timing of treatment can be planned and controlled.
Several aspects make the domestic pig a suitable animal to study wound healing processes. One of the most important aspects is the similarity between human and porcine skin. Especially the connection between the dermal layer and subcutaneous fat is important in this respect, and is essentially different in rodents e.g. It is due to this property, that wound contraction can be studied in a porcine wound model very effectively.
Another advantage of the pig over other animals such as mice or rat is that due to the relatively large size of the animal several wounds can be compared directly in one animal, which reduces inter-individual variation to intra-individual variation.
Of course one should not disregard the limitations of animal wound healing models: they lack the intrinsic complexity of a patient with a complicated wound healing problem. Nevertheless, wound healing studies in pig models can be of good assistance in learning to manage the different phases of wound healing.

KNOCK OUT MODELS OF HEALING
Sharon O'Kane, Renovo Ltd, Manchester, UK

Knockout and transgenic mice are valuable tools in elucidating the molecular and cellular biology of tissue repair. We have carried out wound healing experiments on many strains of mice that have genes either deleted or over-expressed, which have resulted in interesting findings. For example, Transforming Growth Factor beta 3 (TGFb3) is known to be an important molecule in preventing scarring in the adult, and in normal fetal wounds which don't scar, there are very high levels of TGFb3. We have now shown with very careful fetal wound healing studies in TGFb3 knockout mice that these fetal wounds scar, confirming the central role of TGFb3 in scarring. Another example is the gelsolin knockout mouse gelsolin is an actin binding protein involved in cytoskeletal turnover and cell movement. We have shown that wounds in the gelsolin knockout mouse heal much slower than normal wounds, resulting in a worse scar, due to altered cellular movement into the wound area. Another knockout mouse we have shown to also have delayed healing, but for different reasons, is the Collagenase resistant mouse, which has extremely impaired wound healing due to disruption in extracellular matrix turnover. We can therefore evaluate the contribution of various factors to the wound healing process using knockout and transgenic mice in this way.

TRANSGENIC MODELS OF HEALING
Sabine Werner, Professor of Cell Biology, Institute for Cell Biology, Zurich, Switzerland

Wound healing is a highly ordered and well co-ordinated process that involves inflammation, cell proliferation, matrix deposition and tissue remodeling. During the past few years, a series of candidate key players in the wound healing scenario have been identified. These include a variety of different growth factors and cytokines, but also molecules that are involved in cell-cell and cell-matrix interactions, and proteins responsible for cell stability and cell migration. In most cases, the suggested function of these molecules is based on descriptive expression studies and/or functional in vitro studies. By contrast, their in vivo function in wound repair has been poorly defined. The development of transgenic mouse technologies has already provided new insights into the function of many different genes during embryonic development. These technologies allow gain of function experiments (overexpression of ligands and receptors) as well as loss of function experiments (gene knock-outs by homologous recombination in embryonic stem cells or overexpression of dominant-negative acting molecules). A large number of viable genetically modified mice are now available which should not only be useful to determine the role of the targeted or overexpressed genes in normal physiology, but also for different types of repair processes. Indeed, the past five years have seen an exponential growth in the number of transgenic mice to have been used for wound healing experiments, and these studies have provided interesting, and in many cases unexpected, results concerning the in vivo function of growth factors, extracellular matrix molecules, proteinases and structural proteins in wound repair. In my presentation, I will describe advantages and problems of this strategy, and I will demonstrate an example from our own laboratory.

YOUNG INVESTIGATORS PRIZE SESSION

CHRONIC WOUND FIBROBLASTS DEMONSTRATE NO EVIDENCE OF SENESCENCE
Joanna Hilton, University of Wales, UK

Chronic leg wounds are characterised by defective extracellular matrix (ECM) remodelling, failure of re-epitheliali-sation and prolonged inflammation. The hypothesis that this defective EMC remodelling was associated with the accumulation of cells with a senescent phenotype within the wound was studied in chronic wound (CWF) and patient matched normal (NF) fibroblasts. Fibroblasts count cell divisions by utilising telomeres, special regions of DNA that cap and protect the ends of chromosomes. As they very end of the telomere is not fully duplicated during S phase, telomeres shorted with on-going cell division and at a given threshold the cells enter senescence. However, expression of telomerase (a reverse-transcriptase like molecule) results in the synthesis and addition of telomeric DNA de novo compensating for losses that naturally occur during DNA replication thereby prevent senescence.
CWF exhibited no differences in morphology, proliferation (p>0.1), population doubling levels (PDL: p>0.1) or senescence associated b Galactosidase staining (p>0.1) when compared to NF (n=3). This was in direct contrast to populations in in vitro senesced fibroblasts (PDL>64). Both CWF and NF did not express telomerase. Importantly, there were no differences in telomere length between CWF and NF (as assessed by telomere restriction fragment profiling) despite obvious differences caused by in vitro senescence.
These data suggest that the Impaired ability of CWF to reorganise ECM in vitro is not simply related to differences in population doubling levels or proliferation between chronic wound and patient-matched normal fibro-blasts. These findings further support the notion that the distinct phenotype of chronic wound fibroblasts is independent on senescence.

MICROVESSEL DENSITY PROFILE DURING HUMAN WOUND HEALING
Judith Gillard, University of Sheffield, UK

Wound healing requires an angiogenic response during early repair to supply the developing granulation tissue with oxygen and nutrients. Angiogenesis is tightly controlled by growth factors that alter their expression to favour either vessel growth or regression. Wound healing is thought to terminate leaving a relatively acellular and avascular scar but there is no human data investigating the fate of angiogenic vessels in maturing wounds. This study therefore aims to assess microvessel density (MVD) in human scar tissue at varying time-points after surgery.
Patients undergoing breast surgery or attending clinics were invited to participate. Punch biopsies were taken from the scar of patients from two to 104 weeks after surgery (n=65). Control tissue was taken during surgery (n=08). Biopsies were formalin fixed and wax embedded. Sections were stained with CD31 (anti-PECAM-1) antibody to visualise endothelial cells. A Chalkley count was performed in the five most vascular areas and the mean value calculated. Results were expressed as mean ± standard error of the mean. Statistical analysis was performed using a Mann Whitney U-test.
MVD was significantly increased (p<0.05) compared to control tissue (4.88 ± 0.24) at four weeks after surgery (7.4 ± 0.4), and remained significantly elevated until 24 weeks (6.1 ± 0.41). MVD then decreased but remained greater than controls at 36 (5.4 ± 0.31), 72 (5.6 ± 0.91), and 104 weeks (5.63 ± 0.23, p<0.05). The data indicates that angiogenesis is stimulated early in wound healing, but then undergoes a long remodelling process which is incomplete even at 104 weeks after injury. This suggests that human wound healing is considerably prolonged compared to previously reported animal studies.

IMPAIRED WOUND HEALING IN THE COLLAGENASE-RESISTANT MOUSE
Alice Beare, University of Manchester, UK

Collagen in the skin undergoes dramatic reorganisation during wound repair. Matrix metalloproteinases (MMPs) degrade and remodel the collagen in a tightly controlled process. The collagenase resistant mouse, Col1a1tm1Jae has been developed to produce collagen type I that is resistant to degradation by human MMP-1. The mice grow normally but show impaired tissue remodelling with increasing age. We investigated the effect of this mutant collagen on wound repair. Incisional wounds were made on Col1a1tm1Jae homozygous mutant (Col1a1r/r) and wild type (Col1a1+/+) mice and these wounds were harvested at one and six hours, 1, 2, 3, 7, 10, 14 and 70 days post wounding. Wound healing was severely delayed in Col1a1r/r wounds, with re-epithelialisation taking seven days longer than the wild type, and wound contraction and granulation tissue formation similarly compromised. The Col1a1r/r wounds had an early influx of neutrophils and a prolonged presence of macrophages. Immunostaining for MMPs revealed significant upregulation of MMP-13 in Col1a1r/r wounds, but minimal change in MMP-2, MMP-3, and MMP-9 compared to Col1a1+/+ wounds. There is no difference in scarring between Col1a1r/r and Col1a1+/+ wounds at 70 days.