Cultured epidermal keratinocytes on a microspherical transport system are feasible to reconstitute the epidermis in full-thickness wounds

Research efforts to modify cultured autologous skin transplants for large full-thickness burn wounds and in chronic ulcers have shifted from multilayered differentiated grafts ("sheet" grafts) toward smaller units of basal undifferentiated single cell suspensions in a transport medium and subconfluently covered static carriers. It has been shown that wounds transplanted with single cell suspensions reconstitute the epidermis. However, this technique requires the detachment of the keratinocytes from the culture flasks by enzymatic digestion-digestion that might alter the anchoring proteins of the cells. A new approach might be to circumvent the enzymatic digestion to harvest the keratinocytes. This study reports a technique to culture epidermal cells on spherical microcarriers as a suspension culture and transport vehicle. The spherical microcarrier consists of a 100-microm-diameter collagen-coated dextran carrier (Cytodex 3 Pharmacia) and has been used previously for enzyme production commercially. With this new approach, we seeded the human keratinocytes in a spinner-like system onto microspheres and transplanted these micrografts onto full-thickness wounds on the back of nude mice. After 14 days, we showed a reconstituted epithelium that was multilayered and keratinized compared to control wounds. We believe that this is the first step of a new approach to increase the cell yield for seeding without altering the anchoring proteins by enzymatic steps, leading to a superior transplantation method for keratinocytes.     

Porous poly(lactic-co-glycolic acid) microsphere as cell culture substrate and cell transplantation vehicle for adipose tissue engineering

Tissue engineering often requires ex vivo cell expansion to obtain a large number of transplantable cells. However, the trypsinization process used to harvest ex vivo expanded cells for transplantation interrupts interactions between cultured cells and their extracellular matrices, facilitating apoptosis and consequently limiting the therapeutic efficacy of the transplanted cells. In the present study, open macroporous poly(lactic-co-glycolic acid) (PLGA) microspheres were used as a cell culture substrate to expand human adipose-derived stromal cells (ASCs) ex vivo and as a cell transplantation vehicle for adipose tissue engineering, thus avoiding the trypsinization necessary for transplantation of ex vivo expanded cells. Human ASCs cultured on macroporous PLGA microspheres in stirred suspension bioreactors expanded 3.8-fold over 7 days and differentiated into an adipogenic lineage. The apoptotic activity of ASCs cultured on microspheres was significantly lower than that of trypsinized ASCs. ASCs cultured on microspheres survived much better than trypsinized ASCs upon transplantation. The implantation of ASCs cultured on microspheres resulted in much more extensive adipose tissue formation than the implantation of ASCs cultured on plates, trypsinized, and subsequently mixed with microspheres. Ex vivo cell expansion and transplantation using this system would improve the therapeutic efficacy of cells over the current methods used for tissue engineering.     

Cultured human keratinocytes on type I collagen membranes to reconstitute the epidermis

The development of new techniques and modifications to overcome some of the disadvantages in cultured keratinocyte grafting has been motivated by several well-known drawbacks in the use of cultured epithelial autografts such as long culture periods, lack of adherence, difficulty in handling, lack of dermal substrates, and high costs. Two recent insights have influenced further research. On the one hand, it has been shown that the use of undifferentiated proliferative cells in fibrin glue suspensions is effective in epithelial reconstitution. On the other hand, the enzymatic release of cells from the culture surfaces is a critical step leading to at least temporary destruction of anchoring structures of the cultured cells. In this study, we tried to combine these two aspects in an attempt to modify common modalities of keratinocyte transplantation. To avoid dispase dissolving of the cultured cells, keratinocytes were seeded onto bovine collagen type I membranes without feeder layers and under serum-free culture conditions. Subconfluent monolayers of cultured human keratinocytes were transplanted as an upside-down graft on collagen membranes (keratinocyte collagen membrane grafts [KCMG], n = 12) after 3 days of culture or as membrane grafts alone (n = 12) onto standard nude mice full-thickness wounds. Fully differentiated epidermis was found at 21 days after grafting KCMG with persistence of human keratinocytes. This study demonstrates that upside-down grafts of undifferentiated monolayers of keratinocytes on non-cross-linked bovine type I collagen membranes do lead to an early reconstitution of multilayered squamous epithelium with enhanced wound healing compared to the control group. The upside down KCMG grafting technique is able to transfer actively proliferative keratinocytes and simplifies the application compared to conventional epithelial sheet grafting.     

Reduced contraction of skin equivalent engineered using cell sheets cultured in 3D matrices

In order to alleviate their extensive contraction, human fibroblast sheets were cultured in combination with three-dimensional matrices (knitted poly(lactic-co-glycolic acid) (PLGA) mesh and collagen-hyaluronic acid (CHA) sponge) to form contiguous dermal constructs for tissue engineering a bilayered skin equivalent. The resulting constructs were viable, and supported the development of bilayered skin equivalents which did not contract over the 4-week culture period. When implanted into full-thickness wounds in nude rats, cultured skin equivalents based on PLGA meshes registered a take rate of 100% and showed an extent of wound contraction that was statistically similar to autografts, while wounds grafted with PLGA meshes without cell sheets contracted more than autografts. On the other hand, skin equivalents based on CHA sponges were all sloughed off within 2 weeks of transplantation. In all cell sheet-incorporated specimens, cells from the constructs infiltrated and produced extracellular matrix within the neo-dermis, shown by positive human leukocyte antigen and collagen I expression. This technique offers an alternative approach for scaffold-based tissue engineering to produce mechanically stable grafts with matured neo-tissue.     

In vitro cytotoxicity testing of a nanocrystalline silver dressing (Acticoat) on cultured keratinocytes

Acticoat is a polyethylene mesh coated with nanocrystalline silver. It has been used widely as a dressing for chronic wounds, acute partial-thickness burn wounds and donor sites. In this study, the in vitro cytotoxicity of Acticoat on cultured keratinocytes is tested. Human keratinocytes are cultivated on a pliable hyaluronate-derived membrane (Laserskin) using dermal fibroblasts as the feeder layer. When the cultured Laserskin (CLS) is subconfluent it is covered by Acticoat, which is exposed to water (Group 1), phosphate-buffered saline (Group 2) or culture medium (Group 3). The control group is not exposed to the Acticoat. After 30 minutes incubation at 37 degrees C, the inhibitory effect of the nanocrystalline silver on keratinocyte growth is measured by an MTT assay. Compared with the control, the relative viability of the CLS dropped to 0%, 0% and 9.3%, respectively. Thus, Acticoat is cytotoxic to cultured keratinocytes and should not be applied as a topical dressing on cultured skin grafts.     

Wound contraction is significantly reduced by the use of microcarriers to deliver keratinocytes and fibroblasts in an in vivo pig model of wound repair and regeneration

In full-thickness injuries caused by extensive burns or penetrating traumatic injuries, the natural epidermal stem cell niche is destroyed, and wound healing occurs through migration of cells from the wound edges and wound contraction. This can lead to significant contracture formation, especially in large full-thickness injuries, causing lack of mobility and pain. Contraction is reduced when wounds are treated using split-thickness skin grafts (STSG) or dermal substitutes, particularly in combination with cultured autologous keratinocytes, delivered as confluent sheets or sprayed as a single cell suspension (SAK). Here, we show that the application of keratinocytes alone or keratinocytes with fibroblasts, delivered on microcarriers, in combination with STSG or a dermal substitute, significantly reduces contraction of wounds in vivo in a porcine model of wound repair and regeneration. A decrease in alpha-smooth muscle actin-positive myofibroblasts, the cell type responsible for wound contraction, accompanies the reduction in contraction. These findings demonstrate the potential for a significant clinical advantage in the treatment of full-thickness injuries.     

Sucralfate induces proliferation of dermal fibroblasts and keratinocytes in culture and granulation tissue formation in full-thickness skin wounds

Sucralfate is used to induce healing of gastrointestinal tract ulcers. We evaluated its potential utility in the healing of skin wounds. Initial experiments examined the effects of the sucralfate on proliferation of cultured dermal fibroblasts and keratinocytes. Sucralfate induced proliferation in quiescent cultures of both cell types. Additionally, sucralfate enhanced prostaglandin E₂ synthesis in basal keratinocytes and in interleukin-1-stimulated keratinocytes and dermal fibroblasts. Basal interleukin-1 and 6 release were not affected by sucralfate, but the agent enhanced interleukin-1-stimulated interleukin-6 release from fibroblasts. When applied daily to full-thickness wounds in rats, sucralfate increased the thickness of granulation tissue when assessed at day 12.     

An in vitro examination of an extracellular matrix scaffold for use in wound healing

This paper describes evidence that an extracellular matrix (ECM) secreted by human umbilical vein endothelial cells (HUVECs) assembled on gelatin coated plates overlaid by a mixed matrix secreted by human dermal microvascular endothelial cells (HDMECs) and human dermal fibroblasts provides a viable acellular scaffold for use in wound healing. Trypsinized epidermal keratinocytes or colonies from Dispase-digested fresh and cadaver skin tissue adhered and proliferated on either HUVECs ECM/gelatin or mixed matrix overlaid on HUVECs ECM/gelatin. An epithelial-mesenchymal interaction, previously thought to be tissue-specific, was exposed as well as concomitant integrin versatility. Furthermore, heterologous HDMECs and dermal fibroblasts attached and proliferated on the mixed matrix as well as HUVECs ECM. The conditioned medium from HUVECs (HUVECs CM) was found to neutralize the lingering after effects of Dispase, and could be used for the tissue culture of epidermal keratinocytes, HDMECs and dermal fibroblasts, which share related extracellular secretions. Taken together, these results indicate that cultured epithelial autografts can be redesigned to include both epithelial and dermal elements, and advances the acellular 'sandwich' ECM scaffold as a possible structural replacement for the lamina densa and lamina lucida, damaged or completely missing in some wounds and burns.     

In vitro characterization of an artificial dermal scaffold

The treatment of extensive burn injuries has been enhanced by the development of artificial skin substitutes. Integra Artificial Skin, an acellular collagen-glycosaminoglycan (C-GAG) dermal equivalent requires a two-stage grafting procedure. However, preseeding the C-GAG dermal equivalent with cultured fibroblasts and keratinocytes, with the aim of performing a single-stage grafting procedure, may be beneficial in terms of replacing the requirement for traditional split-skin grafts. In this comparative in vitro study, the interactions of cultured human dermal fibroblasts and epidermal keratinocytes in Integra Artificial Skin in comparison to cadaver deepidermalized dermis (DED) was investigated. An increase in cell proliferation and migration in the C-GAG dermal equivalent was observed over time. Cocultures of fibroblasts and keratinocytes on both dermal equivalents showed positive expression of proliferation, differentiation, and extracellular matrix (ECM) protein markers. Organization of keratinocytes in the epidermal layers of DED composites were better compared to the C-GAG composites. Deposition of ECM proteins was enhanced in the presence of keratinocytes in both dermal equivalents. Results demonstrate that in vitro the C-GAG dermal equivalent is biocompatible for cell attachment, migration, proliferation, and differentiation. Preseeding Integra Artificial Skin with cultured autologous fibroblasts and keratinocytes for in vivo application, as a single-stage grafting procedure, warrants testing. A better clinical outcome may be achieved as shown by our in vitro results of the coculture composites.     

Influence of an epidermal cell extract on skin healing and scar formation

We have examined the possible regulatory role of epidermal cell extract(s) (ECE) on skin cells, namely fibroblasts and keratinocytes, both in vivo and in vitro with particular reference to modification of scar formation. In an experimental wound model in pigs, it was found that extracts of cultured human and pig keratinocytes stimulated replication of epidermal cells and their migration from wound edges and remnants of hair follicles and sebaceous glands, together with hair growth, but at the same time suppressed fibroblast proliferation in the dermis. Sections of healing skin wounds that had been treated with ECE showed the presence of a thick layer of epidermal cells lying on relatively sparse dermis. There was little or no contraction in treated wounds and scarring was minimal. Clinical studies of granulomatous lesions of horses and ulcerated wounds in dogs that had been treated with ECE supported these findings. In contrast to its apparently general stimulation of keratinocytes in vivo, ECE had a highly selective effect in vitro on epidermal cells plated at low density in the absence of a feeder layer, which suggests that its action in vivo may be confined to a specific sub-population of rapidly proliferating keratinocytes or alternatively mediated through a second messenger from another type of cell. The inhibitory effect of epidermal cell extract on fibroblasts in vitro was shown by its ability to prevent the contraction of collagen sponges by fibroblasts. These results suggest an important role for epidermal factors in the growth regulation of both epidermal and dermal cells during wound healing.     

Culturing of skin fibroblasts in a thin PLGA-collagen hybrid mesh

A thin biodegradable hybrid mesh of synthetic poly(DL-lactic-co-glycolic acid) (PLGA) and naturally derived collagen was used for three-dimensional culture of human skin fibroblasts. The hybrid mesh was constructed by forming web-like collagen microsponges in the openings of a PLGA knitted mesh. The behaviors of the fibroblasts on the hybrid mesh and PLGA knitted mesh were compared. The efficiency of cell seeding was much higher and the cells grew more quickly in the hybrid mesh than in the PLGA mesh. The fibroblasts in the PLGA mesh grew from the peripheral PLGA fibers toward the centers of the openings, while those in the hybrid mesh also grew from the collagen microsponges in the openings of the mesh resulting in a more homogenous growth. The proliferated cells and secreted extracellular matrices were more uniformly distributed in the hybrid mesh than in the PLGA mesh. Histological staining of in vitro cultured fibroblast/mesh implants indicated that the fibroblasts were distributed throughout the hybrid mesh and formed a uniform layer of dermal tissue having almost the same thickness as that of the hybrid mesh. However, the tissue formed in the PLGA mesh was thick adjacent to the PLGA fibers and thin in the center of the openings. Fibroblasts cultured in the hybrid mesh were implanted in the back of nude mouse. Dermal tissues were formed after 2 weeks and became epithelialized after 4 weeks. The results indicate that the web-like collagen microsponges formed in the openings of the PLGA knitted mesh increased the efficiency of cell seeding, improved cell distribution, and therefore facilitated rapid formation of dermal tissue having a uniform thickness. PLGA-collagen hybrid mesh may be useful for skin tissue engineering.     

Cultured mucosal cell sheet with a double layer of keratinocytes and fibroblasts on a collagen membrane

The aim of this study was to develop a novel cultured mucosal membrane that was facile to prepare and easy to handle, and that could be applied to mucosal defects in the oral cavity. Human oral keratinocytes and fibroblasts were prepared from the oral mucosa. We made the following two types of cultured mucosal cell sheets: a monolayer sheet of keratinocytes cultured on a collagen membrane (K-S) and a double-layered sheet of keratinocytes and fibroblasts on a collagen membrane (KF-S). A collagen membrane was used as a control. Each type of sheet was transplanted onto dorsal skin defects of nude mice. The wound area was measured for the assessment of wound contraction and a specimen was harvested for histologic evaluation 1 week and 4 weeks after grafting. Wound contraction was minimal with KF-S grafts. Although histologic examination showed normal differentiation of the epithelium in all graft types, the involucrin expression pattern of KFS was most similar to that of normal epithelium. These results indicate that a double-layered sheet of keratinocytes and fibroblasts cultured on a collagen membrane may facilitate epithelial healing and prevent wound contraction.     

Cytotoxicity tests for antimicrobial agents using cultured skin substitutes fixed at interface of air and culture medium

The present study is focussed on a new cytotoxicity test using cultured dermal and epidermal sheets, which are fixed at the air and medium interface as a wound surface model. The cultured dermal sheet is composed of human fibroblasts and a collagen matrix, and the cultured epidermal sheet is composed of human keratinocytes and a collagen matrix. Each cultured sheet was fixed at the air and medium interface, over which a piece of test specimen was placed. The in vitro system created, provides a mimetic wound surface since during wound repair, fibroblasts are embedded in an extracellular matrix, while keratinocytes migrate and proliferate on provisional granulation tissue. The results thus obtained in this cytotoxicity test are useful for determining the efficacious amount of antimicrobial agent used in clinical cases.     

Human recombinant EGF protein delivered by a biodegradable cell transplantation system

We have previously shown a new approach to expand cultured human keratinocytes and reconstitute the epidermis in full-thickness wounds using a new microsperical transport system. This was a new approach to increase the cell yield for seeding without altering the anchoring proteins by enzymatic steps. That time we used Cytodex 3 which failed to be degraded and induced an inflammatory reaction in a t-cell-deficient organism. Therefore, we have investigated another microcarrier consisting of PLGA, which is a well-known carrier material for cell culture and transplantation. After coating the PLGA carrier with gelatine the seeding time of viable cells reached 4 h and the cell gain after 7 days of spinner culture was 16-fold. At 14 days after transplantation, we could detect a new stratified epithelium in our full-thickness wound healing model. Because cytokines play a major role in wound healing, we loaded this carrier material with different concentrations of rhEGF, showing a dose dependent release of the protein in vitro and in vivo. This result might lead to a different approach in the treatment of wounds.     

Use of human fibroblasts in the development of a xenobiotic-free culture and delivery system for human keratinocytes

Previous work has shown that keratinocytes can be cultured serum-free on an acid-functionalized, plasma-polymerized surface (for subsequent delivery to patients' wound beds) by inclusion of a fibroblast feeder layer. This study seeks to extend this work by substituting human for murine feeder cells in serum-free culture and examining the performance of keratinocytes expanded in this way to transfer to an in vitro human dermal wound bed model. We compared murine and human fibroblasts (both short-term dermal fibroblasts and a fetal lung fibroblast cell line MRC-5, which has a long history in human vaccine production), alternative methods for growth-arresting fibroblasts, establishing culture of cells serum-free, and the impact of culture with fibroblasts on the differentiation of the keratinocytes. Irradiated human and murine fibroblasts were equally effective in supporting initial keratinocyte expansion, both in the presence and absence of serum. Keratinocytes were significantly less differentiated, as assessed by measuring involucrin expression relative to DNA when grown serum-free with fibroblasts than when grown with serum. Initial cultures of fibroblasts and keratinocytes could be initiated serum-free but were much slower to establish than if serum were used. Transfer of keratinocytes from keratinocyte/fibroblast co-cultures cultured on a plasma polymer surface to a human dermal wound bed model was as successful as from monocultures in both serum and serum-free cultures. In summary, we have revisited a well-accepted methodology for expanding human keratinocytes for clinical use and avoided the use of bovine serum and a mouse fibroblast feeder layer by introducing an irradiated human fibroblast feeder layer.     

Evaluation of functions and tissue compatibility of poly (D,L-lactic-co-glycolic acid) seeded with human dermal fibroblasts

In tissue engineering and wound-healing applications, dermal substitutes are used to provide fibroblasts with the mechanical support for their growth and then to facilitate the skin formation. In this study, three-dimensional porous poly(lactic-co-glycolic acid) (PLGA) 65/35 scaffolds were prepared and then the composites of the scaffolds and human fetal dermal fibroblasts were fabricated as a tissue-engineered dermal substitute. The function and tissue compatibility of the artificial dermal substitute were evaluated at the levels of gene expression (by RT-PCR) and protein expression (total collagen quantities), as well as by histological and immunohistochemical analysis. The PCR products indicated that the mRNA of type-I collagen, mainly secreted by the fibroblasts onto the PLGA scaffolds, was clearly expressed after 4 weeks. The amount of total collagen synthesized from the cells was shown to increase gradually during the initial culture period and slightly decreased afterwards. After 8 weeks of culture, the fibroblasts were well attached and migrated entirely throughout the pores of the PLGA scaffold with normal function. Furthermore, the positively stained type-I collagen was intensively detected throughout the pores. These results suggest that the function and tissue compatibility may be important criteria in evaluating an artificial tissue-engineered skin.     

Evaluation of functions and tissue compatibility of poly (D,L-lactic-co-glycolic acid) seeded with human dermal fibroblasts

In tissue engineering and wound-healing applications, dermal substitutes are used to provide fibroblasts with the mechanical support for their growth and then to facilitate the skin formation. In this study, three-dimensional porous poly(lactic-co-glycolic acid) (PLGA) 65/35 scaffolds were prepared and then the composites of the scaffolds and human fetal dermal fibroblasts were fabricated as a tissue-engineered dermal substitute. The function and tissue compatibility of the artificial dermal substitute were evaluated at the levels of gene expression (by RT-PCR) and protein expression (total collagen quantities), as well as by histological and immunohistochemical analysis. The PCR products indicated that the mRNA of type-I collagen, mainly secreted by the fibroblasts onto the PLGA scaffolds, was clearly expressed after 4 weeks. The amount of total collagen synthesized from the cells was shown to increase gradually during the initial culture period and slightly decreased afterwards. After 8 weeks of culture, the fibroblasts were well attached and migrated entirely throughout the pores of the PLGA scaffold with normal function. Furthermore, the positively stained type-I collagen was intensively detected throughout the pores. These results suggest that the function and tissue compatibility may be important criteria in evaluating an artificial tissue-engineered skin.     

Fibroblast sheets enable epithelialization of sounds that do not support keratinocyte migration

Keratinocyte migration over the wound bed is the single most important parameter for wound epithelialization. Therefore, improvement of the wound bed matrix holds considerable promise for the shortening of hospitalization time in patients with ulcers, burns, and chronic wounds. We investigated wound epithelialization in athymic mice in the presence or absence of a sheet of cultured human fibroblasts. The physiology of keratinocyte growth on fibroblast sheets was investigated in tissue culture using histology, immunofluorescence, and electron microscopy. Keratinocytes from human skin explants were unable to attach or migrate on full-thickness dorsal wounds of athymic mice. Placement of a fibroblast-seeded polyglactin mesh on the wounds resulted in dramatically increased keratinocyte outgrowth. Similarly, human keratinocytes showed good outgrowth on fibroblast sheets at the air/liquid interface in tissue culture. Outgrowth was correlated inversely with fibroblast viability, indicating that the observed effect was due to the complex extracellular matrix secreted by the fibroblasts and matrix-bound growth factors rather than ongoing growth factor release. Collagen IV, a promoter of keratinocyte migration, was found to be abundant in the fibroblast-derived matrix. This study demonstrates that wounds which are unable to support keratinocyte migration can undergo epithelialization if a conducive substrate, supplying appropriate extracellular matrix and/or matrix-bound growth factors, is applied.     

UV radiation and prostaglandin E2 up-regulate vascular endothelial growth factor (VEGF) in cultured human fibroblasts

OBJECTIVE AND DESIGN: Exposure to UV radiation is responsible for skin erythema and inflammation. PGE2 is an important inflammatory mediator involved in this process and vascular endothelial growth factor (VEGF) is a potent vascular permeability factor mainly produced by epidermal keratinocytes. This study was aimed at determining whether UVB/A1 radiation and prostaglandin E2 (PGE2) could modulate the production of VEGF by cultured dermal human fibroblasts (HF) in comparison to keratinocytes (HK). MATERIALS AND METHODS: The skin cells derived from foreskin, were cultured in defined medium before treatment by either UVB/A1 radiation, or stimulation by addition of PGE2 (10(-8) to 10(-5) M). The expression of VEGF in cultured fibroblasts and keratinocytes was evaluated at the mRNA (RT-PCR) and protein levels (ELISA). RESULTS: The basal level of VEGF was lower in HF than in HK. Both UVB and UVA1 radiation strongly up-regulated VEGF mRNA and protein in HF whereas UVB but not UVA1 radiation induced a VEGF increase in HK only at the protein level. UVA1, when associated with UVB radiation, showed an additive effect on VEGF secretion in HF but not in HK. PGE2 increased in a dose-dependent manner the expression of VEGF in HF but not in HK. Indomethacin as well as the antioxidant alpha-tocopherol did not reduce UV-induced enhanced secretion of VEGF by both fibroblasts and keratinocytes whereas pyrolidine dithiocarbamate exerted an inhibition of this overexpression. CONCLUSIONS: These results indicate different signaling pathways in the PGE2 and UV-induced regulation of VEGF in dermal fibroblasts and epidermal keratinocytes. They also suggest a role for VEGF from both fibroblasts and keratinocytes in the UV-induced erythema, independent of PGE2. A dermal overexpression of VEGF by fibroblasts from UV-irradiated skin may contribute to dilated microvasculature, a feature of skin photoaging and more generally, to a more permissive stroma to tumor formation than unexposed skin.     

Thermo-responsive culture dishes allow the intact harvest of multilayered keratinocyte sheets without dispase by reducing temperature

To develop new technology for harvesting transplantable cultured epithelium without dispase treatment, human keratinocytes were plated on culture dishes grafted with a thermo-responsive polymer, poly(N-isopropylacrylamide). The grafted dish surfaces are slightly hydrophobic above 32 degrees C, but reversibly change to hydrophilic below this temperature. According to the method of Rheinwald and Green, keratinocytes proliferated and made a multilayer on the grafted surfaces at 37 degrees C, as on the nongrafted culture dishes. The multilayered keratinocyte sheets were detached from the grafted surfaces only by reducing temperature to 20 degrees C without need for dispase. No cell remnants were observed on the dishes. Such cell sheet detachment was not observed on nongrafted dishes. Immunoblotting of harvested keratinocyte sheets revealed that dispase treatment disrupted E-cadherin and laminin 5, while these molecules remained intact in the keratinocyte sheets harvested by only reducing temperature from the grafted dishes. Transmission electron microscopy revealed that desmosomes were destroyed in dispase treatment but retained in low-temperature treatment. Use of thermo-responsive dishes was examined as a new tool for tissue engineering to achieve the preparation of artificial epithelium for cell transplantation as well as for the investigation of intact multilayered keratinocyte sheets.