Caroline A. Harrison

The success of laboratory‐expanded autologous keratinocytes for the treatment of severe burn injuries is often compromised by their lack of dermal remnants and failure to establish a secure dermo‐epidermal junction on the wound bed. We have developed a tissue‐engineered skin substitute for in vivo use, based on a sterilized donor human dermis seeded with autologous keratinocytes and fibroblasts. However, culture rates are currently too slow for clinical use in acute burns. Our aim in this study was to increase the rate of production of tissue‐engineered skin. Two approaches were explored: one using a commercial low‐calcium media and the other supplementing well‐established media for keratinocyte culture with the calcium‐chelating agent ethylene glutamine tetra‐acetic acid (EGTA). Using commercial low‐calcium media for both the initial cell culture and subsequent culture of tissue‐engineered skin did not produce tissue suitable for clinical use. However, it was possible to enhance the initial proliferation of keratinocytes and to increase their horizontal migration in tissue‐engineered skin by supplementing established culture medium with 0.04 mM EGTA without sacrificing epidermal attachment and differentiation. Enhancement of keratinocyte migration with EGTA was also maximal in the absence of fibroblasts or basement membrane.

Background The transglutaminase (TG) family consists of eight distinct isoforms. TG types 1, 3 and 5 play a major role in normal skin development, with TG2 also being elevated during dermal wounding. TG1, 3 and 5 are responsible for the cross‐linking of keratin precursors and formation of the cornified envelope during keratinocyte differentiation. TG2 may play a role in keratinocyte basement membrane cross‐linking. Abnormal TG expression has been demonstrated in Darier disease, Netherton syndrome, psoriasis and lamellar ichthyosis. During a recent investigation of skin contraction in tissue‐engineered skin, transglutaminase inhibitors were found to produce hyperproliferation and parakeratosis. Objectives Accordingly, this study was designed to study the effect of pan‐transglutaminase inhibition on morphology of tissue‐engineered skin and expression of keratinocyte differentiation and proliferation‐associated antigens. Methods We used a tissue‐engineered model of human skin, based on de‐epidermized acellular human dermis, seeded with normal keratinocytes and dermal fibroblasts and cultured at an air–liquid interface. The pan‐transglutaminase inhibitors putrescine, NTU283 (1‐dimethyl,2‐[(oxopropyl)thio]imidazolium) and NTU285 (N‐benzyloxycarbonyl‐l‐glutaminyl‐6‐dimethylsulfonium‐5‐oxo‐l‐norleucine) were added to the culture medium. After 28 days, histology and immunohistochemistry for collagen IV, involucrin and cytokeratins 6, 10 and 16 were performed. Results Keratinocyte hyperproliferation and parakeratosis were seen in response to transglutaminase inhibition. Inhibition of transglutaminase also resulted in loss of basement membrane collagen IV. Involucrin and cytokeratins 6 and 16 were confined to the basal layers in control composites but expressed throughout the epidermis in response to transglutaminase inhibition. A distinct band of expression of cytokeratin 10 was seen in the upper stratum granulosum of control composites but only patchy expression was seen after transglutaminase expression. Conclusions Pan‐transglutaminase inhibition inhibits terminal differentiation of keratinocytes, leading to a hyperproliferative epidermis with parakeratosis and enhanced expression of involucrin and cytokeratins 6 and 16. Expression of the differentiation‐associated cytokeratin, cytokeratin 10, is reduced. Basement membrane integrity is also lost as a result of transglutaminase inhibition.

Skin graft contraction leading to loss of joint mobility and cosmetic deformity remains a major clinical problem. In this study we used a tissue-engineered model of human skin, based on sterilized human adult dermis seeded with keratinocytes and fibroblasts, which contracts by up to 60% over 28 days in vitro, as a model to investigate the mechanism of skin contraction. Pharmacologic agents modifying collagen synthesis, degradation, and cross-linking were examined for their effect on contraction. Collagen synthesis and degradation were determined using immunoassay techniques. The results show that skin contraction was not dependent on inhibition of collagen synthesis or stimulation of collagen degradation, but was related to collagen remodelling. Thus, reducing dermal pliability with glutaraldehyde inhibited the ability of cells to contract the dermis. So did inhibition of matrix metalloproteinases and inhibition of lysyl oxidase-mediated collagen cross-linking, but not transglutaminase-mediated cross-linking. In summary, this in vitro model of human skin has allowed us to identify specific cross-linking pathways as possible pharmacologic targets for prevention of graft contracture in vivo.

Background Hypertrophic scarring and skin graft contracture are major causes of morbidity after burn injuries. A prominent feature is excessive fibroplasia with accumulation of increased fibrillar collagen relative to normal scar tissue. The application of split‐thickness skin grafts or cultured epithelial autografts to burn wounds is known to reduce scarring and contraction. Objectives To investigate further how the keratinocyte influences underlying fibroblast behaviour by examining the influence of keratinocytes on fibroblast collagen synthesis, using a new assay for collagen synthesis never previously applied to skin cell biology. Methods We investigated the influence of the keratinocyte on fibroblast synthesis of type I collagen using an immunoassay for the aminoterminal propeptide of type I collagen (P1NP) in conditioned medium from monocultures and cocultures of keratinocytes and fibroblasts over 14 days. The importance of the physical presence of the keratinocyte was investigated by comparing cocultures of keratinocytes and fibroblasts against fibroblast monocultures with keratinocyte‐conditioned medium. Pharmacological agents known to promote fibroblast proliferation [basic fibroblast growth factor (bFGF)], keratinocyte proliferation [insulin‐like growth factor (IGF)‐1], modify scarring in vivo[tumour necrosis factor (TNF)‐α] or modify collagen biochemistry [putrescine, estrone, estradiol and β‐aminopropionitrile (β‐APN)] were then investigated for their effect on collagen synthesis in fibroblasts and in keratinocyte/fibroblast cocultures. Results Keratinocytes in coculture with fibroblasts, and keratinocyte‐conditioned medium, both reduced fibroblast P1NP synthesis. Of the pharmacological agents investigated, bFGF, IGF‐1, TNF‐α and β‐APN all increased collagen synthesis both in monocultures of fibroblasts and in cocultures of keratinocytes and fibroblasts. Conclusions Fibroblast collagen synthesis appears to be downregulated by keratinocyte‐derived cytokines. Fibroblast growth factors and proinflammatory cytokines appear to be able partially to overcome this downregulation and to increase collagen synthesis.