The Preclinical Research Group of the Association of Dutch Burn Centres

The aim of the preclinical research group of the ADBC is to develop new treatment strategies for healing of (deep) burn wounds which improve the outcome of healing. Our attention is focused on fundamental research and efforts are made to detect the cellular and molecular mechanisms of wound healing and scarring. When the underlying processes which lead to scarring are understood, new treatments can be developed in which scarring is decreased or even prevented. In addition, the insight in processes which lead to a normal functioning skin are of interest in developing optimal skin substitutes, our second major focus of research.

Main Image: The Pre-clinical Group, Dutch Burn Centre.

• Top left: Magda Ulrich, Top right: Bouke Boekema,
• Centre left: Vincent van der Veer,
• Centre right: Michelle Verkerk,
• Bottom left: Neeltje Coolen, Bottom right: Marcel Vlig

Burn Wound Healing and Mechanisms of Scar Formation

The chances of survival of patients with severe burn wounds have improved over the last decennia. Because of the improvements in hygiene, nourishment and the policy regarding shock treatment the number of patients with extensive and severe burn wounds has increased considerably. Although the techniques for medical and surgery treatment have improved tremendously the outcome of healing of full thickness burn wounds is still poor. As a consequence, the number of patients with severe scarring has expanded as well.

Wound healing is a cascade of different processes that have to be closely orchestrated in order to regain normal functional restoration of the damaged tissue. Derailment of this system results in impaired wound healing and hypertrohic scar formation. A number of factors seem to be responsible for the generation of often invalidating scars, such as the depth and location of the wound and an excessive inflammatory reaction (possibly as a result of an infection). Although ample research has been performed on many aspects of wound healing, the precise mechanism of scar formation or scarless healing is still lacking.

Since several decades it is known that a lack of dermal tissue in the healing wound is contributing to scarring, along with other factors such as genetic properties, racial predisposition and complications during the healing process such as wound infection. Today the most used therapy and still gold standard to treat patients with deep burn wounds is transplantation with a (meshed) split skin autograft. However, patients with a high percentage of burned body surface are often confronted with a lack of intact skin which means that the possibility to transplant the patient with their own skin is restricted considerably.

This means that the demand for tissue engineered skin is large in burn wound care. Although there are several products on the market, these materials have not demonstrated satisfactory results.

Figure 1. Expression of MMP-1 and TIMP-1 in human dermal and scar tissue derived fibroblasts.

Dermal fibroblasts (upper left panel) express MMP-1 in most cells, only a few scar derived fibroblasts express MMP-1 (upper right panel). Dermal fibroblast do express TIMP-1 protein (lower left panel) while this protein is expressed abundantly by scar derived fibroblasts (lower right panel).

Key players in scar formation or normal skin regeneration are the fibroblasts. Fibroblasts, a heterogeneous group of cells, are responsible for extracellular matrix formation and remodelling. Depending on the tissues they reside in the cells exert a specific phenotype. The phenotype of a cell is largely defined by the microenvironment they live in. This, in case of fibroblasts, is ambiguous because these cells themselves are largely responsible for their environment. Cells bind to different macromolecules present in the ECM via specific receptors. Cell-matrix interactions not only control the morphology and orientation of cells but are also involved in regulation of various cellular functions like differentiation, migration, proliferation or gene expression profiles. Besides several other factors which influence the resident cells, are ‘trapped’ in the matrix, such as growth factors and cytokines.

The lack of a dermal template in deep burn wounds may therefore be a major cause of derailed tissue regeneration. Fibroblast like cells from other sources (circulation, adjacent tissues) will repopulate the wound, and because the proper signals are lacking, these cells acquire a defective phenotype.

In order to improve wound healing and reduce scar formation, treatment of wounds with artificial extracellular matrix scaffolds with the correct composition and architecture is very important. The dermal substitute has to function as a temporary scaffold to facilitate ingrowth of fibroblasts, and other cells such as endothelial cells, into the wound and to guide them into regeneration of the dermal tissue.

In wound healing a special sub-type of fibroblasts, the myofibroblast plays an important role. This sub-type is present in the wound for a certain period of time. A clear link between the appearance of myofibroblasts in the wound and scar formation has been demonstrated in foetal lamb wound healing.

Figure 2. The presence of the 'bone-like' cross-links in hypertrophic scars, normotrophic scars, and normal skin in humans.

The origin of myofibroblasts is still under debate. It is conceivable that various myofibroblast sub-populations are formed in the wound environment; however they could also be recruited from surrounding tissues. In normal wound healing, myofibroblasts either revert to the normal fibroblast phenotype or disappear from the wound by means of apoptosis. In deep (burn) wounds the myofibroblasts remain present in the wound environment and this will almost certainly result in severe (hypertrophic) scarring. The cause of myofibroblast persistence is still poorly understood, but could be related to the origin of the myofibroblast or a different subtype of fibroblast which creates a different microenvironment premeditated for maintaining its phenotype. Persistence of this sub-type in the healing wound prolongs processes such as wound contraction, increased ECM deposition and altered ECM remodelling, and results in scar formation instead of regeneration of skin. We have shown that collagen type I and III expression is increased in culture fibroblasts derived from scar tissue in comparison with dermal derived fibroblasts. Expression levels of collagen degrading enzymes MMP-1 and -3, are decreased in the scar derived fibroblasts and tissue whereas the inhibitors of these enzymes TIMP-1 and -3 are increased (figure 1).

Thus not only collagen deposition is increased but degradation of the deposited collagen fibres is decreased. It is known that TGF-b1 plays an important role in fibrosis. TGF-b1 induces collagen deposition and reduces collagen degradation by decreasing the expression of matrix metalloproteinases (MMPs) and inducing the expression of tissue inhibitors of metallotroteinases (TIMPs).

Figure 3. Expression pattern of LH2B, the enzyme responsible for the bone-like cross linking of collagen.

1A. mRNA expression level of LH2b relative to the housekeeping gene b-2 microglobulin, during wound healing in the full thickness wound model in the pig. Presence of the bone like crossliniking type in the scar (8 weeks old) and normal pig skin.

Besides the balance between the proteolytic enzymes and their inhibitors, the accessibility of a collagen molecule for enzymes is an important decisive factor for collagen degradation. Collagen structure is partly determined by the type of cross-linking. The fibril forming collagens type I and type III are the main elements in the dermal extracellular matrix. The precursor of the collagen molecule is synthesised as a pre/pro/alpha chain. Three of these polypeptides coil into each other forming the triple helix and these helixes are subsequently linked together in the final step of collagen fibril biosynthesis. Two pathways of cross-linking are responsible for the formation of collagen fibrils; the allysine pathway and the hydroxyallysine pathway. Depending on the tissue in which the collagen fibrils are formed one of these pathways is followed. In a variety of tissues such as bone and cartilage the hydroxyallysine pathway is the main cross-linking pathway. In skin the allysine pathway is the main cross-linking pathway. The difference between these two pathways depends on whether or not a lysine residue in the telopeptide is hydroxylated into a hydroxylysine residue. If the hydroxylation occurs the hydroxyallysine cross-link is formed, if not the allysine cross-link is formed. Two telopeptide hydroxylysine residues subsequently react with a lysine or hydroxyllysine residue from the helix, maturing into the trifunctional cross-links lysylpyridinoline (LP) and hydroxylysylpiridinoline (HP) respectively.

Figure 4. mRNA expression levels of TGFbs relative to the housekeeping gene b-2 microglobulin in human fibroblasts derived from normal dermis and scar.

In pathological processes of skin fibrosis the type of cross-linking switches from the allysine pathway to the hydroxyallysine pathway. The increased formation of hydroxyallysine cross-links is related to the accumulation of collagen in fibrotic lesions. It was also shown that collagen fibres cross-linked via the hydroxyallysine pathway is less accessible for MMP-1. Recently the telopeptide lysyl hydroxylase was identified as LH2b. The expression of this protein is increased in fibroblasts derived from hypertrophic scars.

The cross-linking product of LH2b has been studied in tissue derived from hypertrophic scars of burn patients, in comparison with normal dermis of the same patients the hydroxyallysine cross-links were significantly increased (figure 2).

In the porcine full thickness wound model we were able to assess the expression levels of different scar related proteins in time. Also in this animal model the LH2 b and the LP/HP crosslinks were increased (figure 3).

It is known that TGF-b1 plays an important role in fibrosis. TGF-b1 induces collagen deposition and reduces collagen degradation by decreasing the expression of matrix metalloproteinases (MMPs) and inducing the expression of tissue inhibitors of metallotroteinases (TIMPs).

TGF-b1 exerts its effect in fibrosis by being involved in the transition of fibroblasts into myofibroblasts.

Figure 5. mRNA expression levels of TGFbs relative to the housekeeping gene b-2 microglobulin during wound healing in the full thickness wound model in the pig.

TGF-b3 has been described as an anti-scarring agent. In fibroblast culture TGF-b3 causes the same effects as TGF-b1, but in vivo counteracting effects to TGF-b1 have been reported. During the first phase of wound healing these growth factors are mainly released by platelets and inflammatory cells. We have shown that fibroblasts derived from scar express significantly more TGF-b1 than normal dermal fibroblasts (figure 4), whereas the expression of TGF-b3 is decreased (not significant). In the pig model we saw a significant increase in TGF-b3 in time (figure 5). This might account for the fact that these animals do not develop hypertrophic scars as seen in man.

To accomplish normal healing instead of scar formation future efforts should be directed to down regulation of scar associated genes in the wound. For the future the great challenge will be to development an intelligent dermal scaffold which creates a microenvironment enabling dermal regeneration by directing scar or myofibroblasts into a dermal phenotype. When this could be accomplished one possibly could use mesenchymal stem cells to manufacture cell populated dermal substitutes.

Dr Magda Ulrich

Association of Dutch Burn Centres, Preclinical Research, Beverwijk and the Department of Plastic, Reconstructive and Hand Surgery,
Free University Medical Centre, Amsterdam, The Netherlands.