Diffcrentiation (1990) 44:232-238

Differentiation Ontogeny and Neoplasia 0 Springer-Verlag 1990

Reconstituted skin in culture : a simple method with optimal differentiation Nicole Basset-SCguin *, Jean Francois Culard Ckcile Kerai ' , FredCric Bernard ', Annette Watrin Jacques Dernaille Jean Jacques Guilhou '

2,

2,

Laboratoire de Recherche Dermatologiquc, Scrvicc dc Dcrmatologie-Phltbologie,H6pital Saint-Charles. F-34059 Montpellier Cedex, France

' CNRS 8402-INSERM U249,

Accepted in revised form June 30, 1990

Abstract. Human skin is a unique organ, which can be reconstituted in vitro and represents an interesting system for studying cell proliferation and differentiation. A simple technique for producing reconstituted skin with optimal epidermal differentiation is described and characterized. A 4-mm punch biopsy of normal human skin is deposited on the epidermal side of mortified de-epidermized human dermis maintained at the air-liquid interface with a metallic support. The culture medium contains insulin, epidermal growth factor (EGF), cholera toxin, hydrocortisone, penicillin/streptoniycin and fungizone. A well-differentiated epidermis develops within 15 days. Morphological and ultrastructural studies show a neoepidermis resembling normal skin. Differentiation markers such as involucrin, filaggrin, and various cytokeratins detected with pancytokeratin antibody are present and confirm this resemblance. The keratin profile is comparable to that observed in other skin culture models. A basement-membrane-like structure is reconstituted with hemidesmosomes and anchoring-filament formation. Bullous pemphigoid (BP) antigen is observed at the dermo-epidermal junction after 21 days of culture. Moreover, both dermal substrates and punch biopsies can be kept frozen for long-term storage, with little or no loss of epidermal growth kinetics and morphology. This skin culture technique is rapid, simple, economical and reproducible. Characterization has here shown highquality epidermal differentiation. Scientists interested in epidermal in vitro studies should take interest in all these advantages.

Introduction The reconstruction of human skin in vitro has recently become an important goal of physicians, to provide treatment for extensive skin defects (e.g. burns, giant congenital cutaneous formations or leg ulcers). More-

* To whom

offprint requests should be sent

over, artificial skin is a useful tool for studies on cell proliferation and differentiation. There is a variety of techniques now available. The first was described by Green and involves the use of human keratinocytes seeded in lethally irradiated 3T3 fibroblasts [14, 271. According to this method, large areas of partially differentiated epidermis can be obtained rapidly, and the epidermis is able to complete its terminal differentiation after grafting [9]. Based on the supposed important role of the dermis in epidermal differentiation, others have used mortified dermis [24] or equivalent substrates [2, 31, on which isolated keratinocytes are seeded. Dubertret has described a variation of this technique which includes the insertion of a punch biopsy in a fibroblast-collagen matrix [6]. These methods provide an epidermal differentiation pattern which is close to that observed in vivo. Alternative, sometimes delicate, techniques providing diverse epidermal differentiation phenotypes include the growth of hair follicle keratinocytes on the eye lens [20, 321, and keratinocyte growth on polymeric or extracellular matrix components [30, 351. All of above-described methods are either technically difficult or require: (1) permanent maintenance of cells in culture (such as 3T3 fibroblasts, or human fibroblasts for collagen matrices) ; and/or (2) trypsin treatment, and thus keratinocyte selection, which could modify the properties of some cell components. In this paper, we describe here a simple technique for in vitro epidermal culture, combining the systems of Prunieras and Dubertret. This technique achieves optimal epidermal differentiation by taking all the parameters that have been shown to positively influence this differentiation into account: ( 1 ) the presence of a dermal substrate consisting of mortified de-epidermized dermis [16, 17, 211; (2) the persistence of some of the basement membrane components on the epidermal side of the dermal substrate [13,33]; (3) the presence of fibroblasts in the viable dermis of the punch biopsy [7, 391; (4) immersion of the culture, which is maintained at the air-liquid interface with a metallic support [25] ; ( 5 ) high calcium concentration [ 5 , 151. This technique is of major interest

233

because of its simplicity, the quality of the reconstituted epidermis and the possibility of cryopreservation of the culture substrates (i.e. dermal substrate and punch biopsy) thus reducing dependence on clinicians for obtaining skin specimens.

Methods Antibodies. Mouse monoclonal antibodies to human filaggrin and rabbit polyclonal antibodies to human involucrin came from Euromedex (Strasbourg, France). Fluorescein-isothiocyanate (FITC)conjugated antibodies to mouse and rabbit IgG were purchased from Sigma, France. Monoclonal pancytokeratin antibodies were obtained from Boehringer Mannheim (Federal Republic of Germany). Rabbit antibodies to human laminin came from Calbiochem (La Jolla, Ca, USA). Rabbit antibodies to human C3d were obtained from Dako Corporation (Paris, France). Sera from patients with bullous pemphigoid (BP) or acquired bullous epidermolysis (EBA) were a generous gift from Dr. John Stanley (Dermatology Branch, National Cancer Institute, NIH) and were used as sources of antibody for these basement membrane antigens. Preparation of dermal substrates. Normal human skin was obtained from plastic surgery. For de-epidermisation, skin specimens were heated at 56" C under sterile conditions for 5-10 rnin and the epidermis peeled off with forceps. De-epidermized dermis (DED) was placed into sterile cryotubes and mortified by successive (n= 10) freezing and thawing in liquid nitrogen as described elsewhere [26]. DED was stored at -80" C until use. Skin cultures. To initiate the culture, a 4-mm punch biopsy of normal human skin was deposited, dermal side down, on top of the thawed DED in contact with the remnant basement membrane (BMZ) components. The DED was maintained at the air-liquid interphase with a metallic support (Fig. 1) in a 60-mm petri dish. Culture media (DMEM/Ham's F12 3:l) contained 10% fetal calf serum (FCS), 1O h penicillin/streptomycin, 1% fungizone, 0.4 pg/ml hydrocortisone, 5 mg/ml insulin, 0.1 nM cholera toxin, 10 ng/ml epidermal growth factor (EGF). Petri dishes were placed in a Jouan (France) incubator in 5% CO, atmosphere at 37" C and saturating humidity. Media were changed every 3 4 days. Serum delipidization was carried out by passage through a 0.1-pm millipore filter which does not allow passage of lipid miscelle. MCDB 153, a serum-free medium supplemented with bovine pituitary extract, was a gift from A. Taieb (Bordeaux, France). For cryopreservation, punch biopsies were placed in DMEM containing 20% FCS and 10% dimethyl sulfoxide (DMSO). Biop-

.+-

sies were sequentially frozen: first at -20" C for 48 h and then at - 180" C (liquid nitrogen) where they were kept for long-term storage. To initiate the culture, punch biopsies were rapidly thawed, washed, and used as described above. Time-course studies. For kinetics studies, the DED substrates (2 x 2 cm2) progressively covered by the newly formed epidermis, were cut at regular time intervals (53, 57, JIO, J15) and analysed by standard microscopy after hematoxylin and eosin staining. Electron-rnicroscopy studies. Small pieces of neo-epidermis (2 mm from the punch biopsy) were placed in Karnovsky-type fixative (2% paraformaldehyde, 2.5% glutaraldehyde, 0.2% tannic acid, 0.1 M cacodylate buffer, pH 7.2) for 1 h at 4" C. After fixation, skin specimens were washed in 0.1 M cacodylate containing 4.5% sucrose (pH 7.3) for 1 h at 4" C, post-fixed in 1 YOosmium tetroxide in 0.075 M sodium cacodylate for 2 h, and block-stained in a saturated aqueous uranyl acetate solution for 1 h at 4" C. Samples were then dehydrated and embedded in Epon. Ultrathin sections were stained with lead citrate. Immunofuorescence microscopy studies. lmmunofluorescence studies (IF) were performed on neoformed epidermis and DED using a modified version of our previously described technique [ 11. After excision, the samples were embedded in OCT (Miles Scientific, Paris), frozen and stored at -80" C. For IF, 6-pm sections were fixed in 3.7% formaldehyde for 10 min and permeabilized in methanol and then acetone at -20" C for 5 rnin each. The sections were then incubated at 37°C in a humidified chamber with the appropriate antibody or control reagent in 2% phosphate-buffered saline (PBS) bovine serum albumin (BSA). Slides were then washed three times in 2% PBS/BSA, Nonidet P40 (0.2%) for 5 rnin each time, mounted and studied with a Zeiss Axioscop microscope equipped with a camera. No formaldehyde fixation was performed for filaggrin studies. For BMZ-component studies, neither fixation nor permeabilization were carried out. Keratin extraction. We isolated keratins from both normal epidermis and neoformed epidermis of the same origin after 20 days of culture. The epidermis of normal human (NE) skin was peeled off only after separation by heating at 56" C for 5 min. The neoformed epidermis (NeoE) was easily detached from its DED substrate. NE and NeoE were pulverized in liquid nitrogen and homogenized at 4" C in 25 mM Tris-HCI (pH 7.4) containing protease inhibitors (ImM phenyl methyl sulphonyl fluoronate, PMSF; 10 mg/ml leupeptin; 10 mg/ml soybean trypsin inhibitor, 1 mM EDTA). Homogenization was performed with a Potter homogenizer using a Teflon pestle. After centrifugation for 5 rnin at 12000 g and 4"C, the pellet was washed with 25mM Tris-HCI, 1 mM EDTA and 2% NP40. The keratin containing the final pellet was extracted in 25 mM Tris-HCI, 2% sodium dodecyl sulfate (SDS) and 10 mM dithiothreitol, 8 M urea at 37" C for 15 rnin, and heated at 100" C for 2 min for one-dimensional electrophoresis. For two-dimensional electrophoresis, the final pellet was extracted for 15 min at 37" C in 9.5 mM urea, 50 m M dithiothreitol, 25 mM Tris (pH 7.4). Gel electrophoresis. One-dimensional SDS-polyacrylamide gel electrophoresis (10%) was performed according to the method of Laemmli [19]. Two-dimensional gel electrophoresis was done as described by O'Farrell [23]. Proteins were then visualized by Coomassie blue staining.

Results Epidermal outgrowth Fig. I. A 4-mm punch biopsy is depositcd on the basement membrane side of de-epidermized dermis, which is emersed on a metallic support

Growth of the new epidermis (NeoE) was easily detectable as a yellowish disk which progressively covered the

234

Fig. 2. A Kinetics studies show progressive organization of the neoepidermis as a pluristratified epithelia on days 3, 7, 10, and 15 (03, 0 7 , DIO and D f 5 , respectively). Complete morphological differentiation is obtained within 15 days. B Use of frozen biopsies for initiating the culture does not alter the morphology of the neo-epidermis (NeoE)

Fig. 3. Electron microscopy (EM) studies revealed the persistence of the lamina densa, an ultrastructural portion of the basement membrane zone (BMZ), at the top of the de-epidermized dermis (arrow). By immunofluorescence, the presence of laminin (L), KFI and epidermolysis bullosa acquisita (EBA) antigens and the presence of C3d defined the biochemical composition of the BMZ remnants. By contrast, the bullous pemphigoid (BP) antigen disappeared during the deepidermization

entire substrate. The NeoE stopped growing when it reached the external limits of the substrate. Time-course studies are shown in Fig. 2a: on day 3, a monolayer of keratinocytes was observed, at some distance from the punch biopsy, resulting from migration and growth of kerdtinocytes from the original biopsy. On day 7, migration from the punch biopsy was still observed and keratinocytes began to organize into two cell layers. After 10 days, a four-layer epidermis lacking stratum corneum was formed. Complete morphological differentiation was reached after 15-20 days. Cryopreservation of the punch biopsy had little or no effect on the morphology of the NeoE, but slightly slowed the growth rate (Fig. 2 b). In comparative studies of epidermal outgrowth from DED and punch biopsies of the same origin, samples in MCDBl53 showed an absence of growth whereas samples in complete medium covered the DED as expected within 15 days (data not shown). Similarly, when

DED were flipped, with the BMZ surface down, little or no epidermal growth was observed (data not shown). The dermo-epidermal junction

Immunohistochemistry studies showed the presence of classical epidermal basement membrane components such as laminin, KF1 and epidermolysis bullosa acquisita (EBA) antigens, and C3dg on the epidermal side of the DED. Electron-microscopy studies (EM) demonstrated persistence of the lamina densa (Fig. 3). However, the BP antigen had disappeared (Fig. 3). During NeoE formation, reconstitution of a normal BMZ-like structure could be seen between days 15 and 21. The presence of a space resembling a lamina lucida between the basal keratinocytes and the persisting lamina densa of the substrate could be seen by EM. As demonstrated in Fig. 4a, both hemidesmosomes and anchoring fila-

235

of keratinocytes, cuboidal in the basal and suprabasal layers (BL) and flattened in the upper spinous (SL) and granular layers (GL). The stratum corneum (SC)is clearly individualized. Interestingly, fewer nuclei contain condensed heterochromatin than in normal skin. At higher magnification, various structures (Fig. 5 b) which are classically recognized as markers of normal differentiation can be found. Desmosomes (D) and tight junctions (T) form connections between adjacent cells. In the spinous layers the typical interlacing network of tonofibrils (TNF) is seen. In the granular layers, keratohyalin granules (KHG) appear as amorphous electron-dense material, sometimes in contact with tonofibrils. Lamellar granules (LG), with typical periodicity, are also seen in the granular and adjacent layers. Finally, a cornified envelope (env) with modified desmosomes is seen at the junction between the stratum corneum and the granular layer and between corneocytes.

Differentiation markers Fig. 4. A The reconstruction of a BMZ-like Structure can be Seen within 21 days: a clear space ressembling a lamina lucida (LL) is formed, hemidesmosomes (H) and anchoring filament (AF) are observed. B Immunofluorescence studies revealed the reapparance of the BP antigen as a linear bright staining (urrowv) at the epidermal-dermaljunction

ments were observed at the neo-epidermal-dermal junction (Neo-BMZ). I.F. studies on day 21 revealed the presence of the BP antigen at the Neo-BMZ (Fig. 4b).

Ultrastructuralanalysis of the neo-epidermis An overview of the ultrastructure of the NeoE (20 days) is presented in Fig. 5a. The tissue is essentially composed

we Performed immunofluorescence studies using differ-

entiation markers such as pancytokeratin, inVOlUCrin and filaggrin and analysed extracted keratins to evaluate the degree of differentiation obtained in our culture system.

Immunojluorescence studies. Antibodies to human filaggrin and pancytokeratin gave staining patterns similar to those observed in normal skin (localized in the granular layer and suprabasal layers respectively). Antibodies to human involucrin gave specific cytoplasmic staining with membrane reinforcement, which was earlier than in normal skin, beginning in the suprabasal cells in the NeoE, whereas it was confined to granular cells in normal skin (Fig. 6).

Fig. 5. A Ultrastructural overview

of the neoepidermis at 20 days of culture. B Magnification of ultrastruct ural feature details: desmosomes (D).tight junction (7'). lonofibrils (TNF), lamellar granules (LG), keratohyalin granules (KHG), and cornified envelope ( e m )

236

OH-

B -. _50

445 KD

NS Keratin analysis. Keratins were extracted and analysed by one- and two-dimensional gel electrophoresis. Figure 7A, B shows that high-molecular-weight keratin (67 kDa) disappears in culture. Conversely, a faint band, corresponding to 67 kDa can be seen in Fig. 7 C (NS). This band was observed in several but not all one-dimensional electrophoresis experiments. Its variable intensity may be explained by quantitative differences in extraction efficiency, as well as differences in quantities loaded onto one- and two-dimensional gel electrophoresis. On the other hand, intermediate-size keratins were conserved (58-56 kDa) and low-molecular-weight keratinrelated products (48 kDa) were identified. After serum

DS

Fig. 7. Two-dimensional electrophoresis of keratin extracts from: normal human skin ( A ) ; NeoE (20 days of culture) ( B ) ; Onedimensional-electrophoresis study of NeoE keratins ( 0 :N S , normal serum; DS, after 2 weeks of culture with delipidized serum. Keratins are labelled with their respective molecular weights

delipidization, an increase in the amount of 67-kDa keratin was observed consistently and is illustrated in Fig. 7 C (DS).

Discussion This study describes a simplified method for preparing newly formed epidermis for in vitro studies. The principle is based on the combination of two previously described systems with the use of a natural substrate mortified de-epidermized dermis as described by Prunitras [24], on which a punch biopsy of human skin is ~

237

deposited according to Coulomb’s technique [6]. A variant of this method using guinea-pig dermis was initially described by Freeman but did not give such satisfactory differentiation features [lo]. The main advantages of the technique described here are that cells do not need to be permanently maintained in culture, nor is it necessary to reconstruct a dermal substrate. The system is also very economical since only limited quantities of medium are used, and it is very practical since both dermal substrates and punch biopsies can be kept frozen until use. Finally, the epidermal differentiation obtained is close to that observed in vivo and equivalent to, or even better than, that observed in the other systems reported to date.

envelope [3, 17, 22, 28, 31, 341. Other markers such as involucrin, filaggrin and keratin patterns have additional differentiation-specific biochemical characteristics. In the literature all of these markers are variably present and analyzed in the different skin culture systems. The differentiation obtained in the culture system described here is excellent because all these markers are present and their biochemical analysis is close to that observed in vivo. Optimization of epidermal differentiation by emersion of the culture [25], and the use of delipidized serum [ll], is also observed in this study. Moreover, the influence of C a + + , and more recently, the precise range of Ca+ concentration which improves epidermal differentiation have been reported [ 15, 361. Although epidermis obtained in culture may be of good quality, no model yet allows the reconstruction of an epidermis and its appendages as observed in vivo, since Langerhans cells, Merkel cells and epidermal appendages are always missing. Melanocytes, on the other hand, can be maintained [4]. +

Epidermal outgrowth

The growth rate of the NeoE is comparable to that observed with other skin culture methods [2, 3, 6, 9, 14, 24, 27, 30, 32, 351. To be exact, in the system described here, the time required to obtain a well-differentiated NeoE covering 10- to 12-fold the original area of the punch biopsy is 15 3 days. Variations in growth rate were observed and expected [ 171, depending essentially on the age of the donors. Growth kinetics were easily followed morphologically by analysis of portions of the skin culture system at various time intervals. This could be useful for the study of epidermal wound-healing processes under various culture conditions. These conditions are in fact very important because direct comparative studies using either complete culture medium or MCDB 153 have resulted in strikingly different epidermal outgrowth. A fully differentiated epidermis already covered the entire DED with the complete culture medium after 15 days of culture, whereas with the MCDBl53 there was no significant epidermal outgrowth. When using frozen biopsies to initiate the culture, the growth rate was slowed slightly (approximately 20 days to cover the substrate), but NeoE morphology was very acceptable. NeoE growth seemed to be favored by the persistence of a partial BMZ of the DED surface. Interestingly, little or no growth was seen after the DED was flipped, with the BMZ down, whereas normal growth was observed when the biopsy was in contact with the remnant BMZ of the DED. However, it is not clear whether this is related to specific BMZ properties for epidermization as suggested in several studies [ 13, 331, or if it is due to the flatness of the surface, facilitating keratinocyte migration and proliferation.

Reconstruction of a BMZ-like structure As in our model some BMZ components are maintained

on the DED surface, after epidermization, a complete BMZ-like structure is reconstituted in vitro. Basal keratinocytes are organized so that a clear space resembling the lamina lucida remains between their basal poles and the lamina densa of the DED. Hemidesmosomes and anchoring filaments are formed and the BP antigen is present, as a linear staining at the demo-epidermal junction within 21 days of culture. Nevertheless, the NeoE was easily detachable from its substrate. The BMZ-like structure present in our model could be useful for in vitro analysis of interactions, between leucocytes and the BMZ, and for studying the deposition of complement-fixing autoantibodies at the BMZ [12], as observed in different pathological situations [ 181. In conclusion, a simplified method to obtain artificial skin in culture is reported here. Its advantages are multiple : simplicity, economy, rapidity, quality, reproducibility and long-term storage. We present it as a possible useful tool for scientists interested in in vitro epidermal analysis, particularly for various biochemical and/or pharmacological applications.

References Differentiation of the newly formed epidermis

1. Basset-Stguin N, Dersookian M, Cehrs K, Yancey K B (1988)

The various methods used for the reconstruction of living skin give different epidermal-cell phenotypes depending on the culture conditions [ 171. Morphological and ultrastructural markers of epidermal differentiation include: individualization of the epidermal layers such as the basal, spinous, and granular layers and stratum corneum ; the presence of desmosomes, keratohyalin granules, lamellar bodies; and identification of a corneal cell

C3dg is present in the normal epidermal basement membrane zone. J Immunol 141 :1273-1280 2. Bell E, Ivarson B, Merril C (1979) Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro. Proc Natl Acad Sci USA 176: 1274-1278 3. Bell E, Sher S, Hull B, Merril C, Rosen S, Chamson A, Asselineau D, Dubertret L, Coulomb B, Lapiere C, Nusgens B, Neveux Y (1983) The reconstruction of a living skin. J Invest Dermatol 81 : 2 ~ - 1 0 ~

238 4. Bertaux B, Morliere P, Moreno G, Courtalon A, Masse JM, Dubertret L (1988) Growth of melanocytes in a skin equivalent model in vitro. Br J Dermatol 119: 503-512 5. Boyce ST, Ham RG (1983) Calcium regulated differentiation of normal human epidermal keratinocytes in chemically defined clonal culture and serum-free serial culture. J Invest Dermatol 81:33s40s 6. Coulomb B, Saiag P, Bell E, Breitburg F, Lebreton C, Heslan M, Dubertret L (1986) A new method for studying epidermalization in vitro. Br J Dermatol 114:91-101 7. Coulomb B, Lebreton C, Dubertret L (1989) Influence of human dermal fibroblasts on epidermalization. J Invest Dermatol 92: 122-1 25 8. Ersham A, Larese A, Latarjet J, Blondet R, Faure M, Thivolet J (1987) Cryoconservation de peau de cadavre et des epidermes de culture. J Med Esth Chir Dermatol XIV:99-100 9. Faure M, Mauduit G, Schmitt D, Kanitakis J, Dedidem A, Thivolet J (1987) Growth and differentiation of human epidermal cultures used as auto and allografts in humans. Br J Dermato1 1161161-170 10. Freeman AE, lgel HJ, Herman BJ. Kleinfeld KL (1976) Growth and characterization of human skin epithelial cell cultures. In vitro 12:352-362 11. Fuchs E, Green H (1981) Regulation of terminal differentiation of cultured keratinocytes by vitamin A. Cell 25: 617-625 12. Gammon WR, Yancey KB, Mangum KL, Hendrix JD, Hammer CH (1989) Generation of C5-dependent bioactivity by tissue-bound anti-BMZ autoantibodies. J Invest Dermatol 93: 195-200 13. Gospodarowicz D (1983) The control of mammalian cell proliferation by growth factors, basement lamina and lipoproteins. J Invest Dermatol81:40~-50s 14. Green H, Kehinde 0, Thomas J (1979) Growth of cultured human epidermal cells into multiple epithelia suitable for grafting. Proc Natl Acad Sci USA 76: 5665-5668 15. Hennings H, Michael D, Cheng C , Steinert P, Holbrook K, Yuspa SH (1980) Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell 22: 629-632 16. Hirone T, Taniguchi S (1979) Basal lamina formation by epidermal cells in cell culture. In: Bernstein IA, Seiji M (eds) Biochemistry of normal epidermal differentiation. Current problems in dermatology. S Kerger AG, Basel Vol 10, pp 159-169 17. Holbrook KA, Hennings H (1983) Phenotypic expression of epidermal cells in vitro: A review. J Invest Dermatol 81 :11 s24s 18. Katz S (1984) The epidermal basement membrane zone-structure, ontogeny, and role in disease. J Am Acad Dermatol 11 : 1025-1037 19. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature 227:686 685 20. Limat A, Noser F (1986) Serial cultivation of single keratinocytes from the outer root sheath of human scalp hair follicles J Invest Dermatol 87:485488

21. Lillie JH, MacCallum DK, Jepsen A (1980) Fine structure of subcultivated stratified squamous epithelium grown on collagen grafts. Exp Cell Res 125:153-165 22. Madison C, Swartzendruber DC, Wertz W, Downing DT (1988) Lamellar granule extrusion lamellae in murine keratinocyte cultures. J Invest Dermatol90: 1 1 6 116 23. O’Farrell P (1975) High-resolution two-dimensional electrophoresis. J Biol Chem 250:4007-4021 24. Prunitras M (1979) Epidermal cell cultures as models for living epidermis. J Invest Dermatol 73: 135-1 37 25. PruniCras M, Regnier M, Woodley D (1983) Methods for cultivation of keratinocytes with an air-liquid interface. J Invest Dermatol81: 28 s-33 s 26. Regnier M, Prunieras M, Woodley D (1981): Growth and differentiation of adult human epidermal cells on dermal substrates. Front Matrix Biol9:435 27. Rheinwald JG, Green H (1975) Serial cultivation of strains of human epidermal keratinocytes: The formation of keratinizing colonies from single cells. Cell 6: 331-344 28. Ryle CM, Breitkreutz D, Stark HJ, Leigh I, Steinert P, Roop D, Fusenig N (1989) Density-dependent modulation of synthesis of keratins 1 and 10 in human keratinocyte line HACAT and in ras-transfected tumorigenic clones. Differentiation 40 42-54 29. Saiag P, Coulomb B, Bell E, Lebreton C, Dubertret L (1985) Psoriatic fibroblasts induce hyperproliferation of normal keratinocytes in a skin equivalent model in vitro. Science 230:669672 30. Tinois E, Faure M, Chatelain P, Vallier P, Schmitt D (1987) Growth and differentiation of human keratinocytes on extra cellular matrix. Arch Dermatol Res 279: 241-246 31. Tseng SCG, Jarvinen MJ, Nelson WG, Huang JW, WoodcockMitchell J, Sun TT (1982) Correlation of specific keratins with different types of epithelial differentiation: Monoclonal antibody studies. Cell 30:361-372 32. Vermorken AJM, Verhagen H, Vermeesch-Markslag AMG, Wirtz P, Bernard BA, Asselineau D, Lenoir MC, Kimenai PM, Shroot B (1985) Differentiation of keratinocytes in vitro: a new culture vessel mimicking the in vivo situation. Mol Biol Rep 10:205-213 33. Vraco R (1974) Basal Lamina Scaffold. Anatomy and significance for maintenance of orderly tissue. Structure. Am J Path01 77:314-329 34. Weiss RA, Eichner R, Sun TT (1984) Monoclonal antibody analysis of keratin expression in epidermal diseases: a 48 and 56 kD keratin as molecular markers for hyperproliferative keratinocytes. J Cell Biol98 :1397-1406 35. Yannas IV, Burke JF, Orgill DP, Skrabut EM (1982) Wound tissue can utilize a polymeric template to synthesize a functional extension of skin. Science 215: 174176 36. Yuspa SH, Kilkenny AE, Steinert P. Roop DR (1989) Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentrations in vitro. J Cell Biol 109:1207-1217:1989