Reconstruction of a skin equivalent using an immortalized human keratinocyte line, HaCaT, was investigated in an attempt to generate an in vitro system representative for human skin. Three different substrates were used to establish air-exposed cultures of HaCaT cells: de-epidermized dermis, collagen gels, and filter inserts. Effects of variations in culture conditions on tissue morphology, on the expression of proliferation-specific and differentiation-specific protein markers, and on lipid profiles were investigated. When grown at the air-liquid interface HaCaT cells initially developed a multilayered epithelium, but during the course of culture marked alterations in tissue architecture were observed. Ultrastructurally, a disordered tissue organization was evident as judged from the presence of rounded cells with abnormally shaped nuclei. Keratins K1 and K10 were irregularly expressed in all cell layers, including stratum basale. Staining of K6/K16 was evident in all cell layers. Locally, basal and suprabasal cells were positive for K4 and additionally expressed K13 and K19. The cornified envelope precursors were expressed only in older cultures (>2 wk after air exposure), except for transglutaminase and small proline rich protein 1, which were irregularly expressed in both early and older cultures. In addition, HaCaT cells showed an impaired capacity to synthesize lipids that are necessary for a proper barrier formation as indicated by the absence of free fatty acids and a very low content and incomplete profile of ceramides. Our data demonstrate that the ultimate steps of terminal differentiation in HaCaT cells do not occur irrespective of the type of substrate or the culture conditions.

The purpose of this study was to find an optimal polymer matrix and to optimize the culture conditions for human keratinocytes and fibroblasts for the development of a human skin substitute. For this purpose porous, dense bilayers made of a block copolymer of poly(ethylene glycol terephthalate) (PEGT) and poly(butylene terephthalate) (PBT; Polyactivetrade mark) with a PEGT/PBT weight ratio of 55/45 and a PEG molecular weight (MW) of 300, 600, 1000, or 4000 Da were used. The best performance was achieved with PEGT/PBT copolymer with MW of PEG 300 D (300PEG55PBT45). When fibroblasts were seeded into the porous underlayer and cultured for 3 weeks in medium supplemented with 100 microg/mL ascorbic acid, all pores were filled with fibroblasts and with extracellular matrix, which was judged from the presence of collagen types I, III, and IV, and laminin. When seeded onto the dense top layer of the bilayered (cell free or fibroblast populated) copolymer matrix, human keratinocytes grew out into confluent sheets. After subsequent lifting to the air-liquid interface, a multilayered epithelium with a morphology corresponding to that of the native epidermis was formed. Some differences could still be observed: the expression and localization of some differentiation specific proteins was different and close to that seen in hyperproliferative epidermis; a basal lamina and anchoring zone were absent.

The aim of this study was to assess the performance of a biodegradable dermal substrate in deep dermal skin defects. The substrate consisted of a synthetic biodegradable matrix called Polyactive, which is an elastomeric poly (ether)/ poly (ester) block copolymer. This matrix was manufactured either as a porous substrate, with gradually changing pore size (BISKIN-M), or as a bilayer consisting of a porous underlayer with a fully dense surface layer (BISKIN). Cell-free matrices and matrices seeded with autologous or allogeneic porcine fibroblasts were applied to full-thickness skin wounds in Yucatan miniature pigs. Biopsies were taken at different time intervals up to 24-months post-transplantation. Although all BISKIN substrates showed little or no adherence to the wound bed, the adherence of the BISKIN-M substrates to the underlying wound was achieved within minutes after application. Therefore, only BISKIN-M Polyactive substrates were further evaluated. Wound contraction was inhibited by both cell-free and fibroblast-populated substrates. All substrates showed extensive neovascular and fibrous tissue ingrowth within 2-weeks post-transplantation. Furthermore, during this time period, matrix degradation was observed, starting with the fragmentation of the polymers into particles, which were phagocytized by macrophages. These processes occurred actively up to 3 months and ceased thereafter. Cell-free substrates degraded faster, and also, the collagen deposition was lower as compared with cell-seeded substrates. The tissue surrounding the remnants of the Polyactive substrates after 24-months post-transplantation consisted of a mature connective tissue. The newly formed collagen had the same distribution pattern as observed in normal native dermis. We conclude therefore that treatment of full-thickness skin defects with fibroblast-populated BISKIN-M Polyactive substrates leads to satisfactory dermal regeneration.

A fully differentiated epithelium mimicking the features of native epidermis was obtained in vitro by culturing human or porcine epidermal keratinocytes on polyester filter substrate at the air-liquid interface. In addition, after 2 weeks of culture, hemidesmosome-like structures were formed along the basal area of the plasma membrane of the basal cells at the cell-filter interface. When grafted onto full-thickness skin wounds in pigs, the take of cell sheets detached from the filter with dispase was significantly higher (about 70%) in comparison to mechanically detached keratinocytes (about 15%). With dispase-treated keratinocytes alone, basement membrane formation took place within 7 days postgrafting as judged from the presence of a lamina lucida and positive staining for type IV collagen. Also, numerous hemidesmosomes and anchoring fibrils were observed at the basal cell-"neodermis" interface. The fully differentiated epidermis, generated by culturing keratinocytes at the air-liquid interface and detached from the substrate by dispase-treatment, is less fragile and easier to handle than epidermal autografts obtained by conventional culturing methods. Detachment by a short dispase-treatment appeared in our hands the only method for successful and complete epithelial regeneration in full-thickness wounds.