Cytotechnology 9: 163-171, 1992. 9 1992 Khtwer Academic Publishers. Printed in the Netherlands.
The organotypic culture of human skin keratinocytes and fibroblasts to achieve form and function Nancy L. Parenteau, Patrick Bilbo, Cynthia J.M. Nolte, Valerie S. Mason and Mireille Rosenberg
Organogenesis Inc., Cambridge, MA, 02142, USA Received 3 April 1992; accepted in revised form 27 August 1992
Key words." organotypic, skin, skin culture, tissue engineering, morphology, immunocytochemistry
Abstract We describe an organotypic model of human skin comprised of a stratified layer of human epidermal keratinocytes and dermal fibroblasts within a contracted collagen lattice. Feasible and reproducible production of the skin construct has required the use of traditional as well as specialized culture techniques. The configuration of the construct has been engineered to maintain polarity and permit extended culture at the air-liquid interface. Morphological, biochemical and kinetic parameters were assessed and functional assays were performed to determine the degree of similarity to human skin. Light and ultrastructural morphology of the epidermis closely resembled human skin. The immunocytochemical localization of a number of differentiation markers and extracellular matrix proteins was also similar to human skin. Kinetic data showed a transition of the epidermal layer to a more in vivo-like growth rate during the development of the construct at the air-liquid interface. The barrier properties of the construct also increased with time reaching a permeability to water of less than 2%.h after approximately 2 weeks at the air-liquid interface which is still on average 30-fold more water-permeable than normal human skin. The construct is currently used for in vitro research and testing and is also being tested in clinical applications.
Abbreviations: ALI: Air-liquid interface; CK: Cultured keratinocytes; HS: human skin; HSPG: Heparan sulfate proteoglycan; LSE: Living skin equivalent; SC: skin construct
Introduction The objective of organotypic culture is to develop cellular models which exhibit as many properties of the organ from which they were derived as possible. In the case of skin, this entails the formation of a dermal component consisting of connective tissue cells embedded in extracellular matrix overlaid with a stratified, comified epidermis. In simplest terms, an organotypic culture
of skin can be "engineered" using its three major components, collagen, dermal fibroblasts and epidermal keratinocytes. Combination of these components to achieve organotypic form and function requires proper structural as well as biological support together with a permissive culture environment to maximize phenotypic expression. In the model presented, the structural support is provided by the collagen lattice which surrounds and is contracted
164 by the dermal fibroblasts. The collagen lattice provides a three dimensional framework for the dermal fibroblast and also provides biological support both for the dermal fibroblasts and the overlying keratinocytes. Fibroblasts within a collagen lattice are biosynthetic (Nakagawa et al., 1989) and responsive to cytokines (Dubertret, 1992). While they are proliferative (Parenteau et al., 1991), the fibroblasts are not hyperproliferative as they would be on a culture dish or other nonbiological substrate (Triglia et al., (1991), but exhibit controlled proliferation with the collagen providing biological feedback. Keratinocytes attach readily to the collagen lattice. Basal keratinocytes are able to maintain a cuboidal to columnar morphology and exhibit polarity, synthesizing basement membrane components basally while generating a stratified epidermis apically. The overall polarity of the organotypic skin construct (SC) is maintained by the physical configuration which prevents the epidermal layer from encapsulating the dermal matrix. The constructs are also designed to permit long term culture at the air liquid interface. Culture at the air-liquid interface (ALI) enhances organization and differentiation of the air-exposed stratified epithelium (Asselineau et al., 1985). Culture at the ALI also results in a change in the proliferative index of the epidermal keratinocyte which declines to values more akin to it+ vivo. Engineering the construction of organotypic cultures to be feasible and reproducible has required a blend of conventional and more complex cell culture techniques. Assessment of organotypic cultures involves morphological, biochemical and functional endpoints as well as analysis of cell kinetics. The organotypic SC described is presently being used as a human in vitro skin model for dermatological (Elder et al., 1992), pharmaceutical (Ernesti et al., 1992) and toxicological research and testing (Gay et al., 1992). In addition, it is being tested clinically both as a wound dressing and a skin replacement.
Materials and m e t h o d s Collagen a n d m a s t e r cell b a n k s
Sterile, acid-extracted Type I collagen (Organogenesis Inc.) is derived from bovine tendon and used at 1.1 mg/ml in 0.05% acetic acid. Human dermal fibroblasts are derived from trypsin-collagenase dissociation of newborn foreskin and propagated to passage 4 - 6 in Dulbecco's modified Eagle's medium (DMEM) containing 10% newborn calf serum (NBCS, Hazleton). Human epidermal keratinocytes are derived from explant culture of neonatal foreskin and propagated on collagen-coated plastic in a defined Minimally Supplemented Basal Medium (MSBM) as described in detail elsewhere (Johnson et al., 1992). Keratinocyte master cells banks are established at passage 3. Epidermal cells used in organotypic cultures are cultivated from the frozen master cell bank for one passage in MSBM before seeding onto the collagen lattice. It is of practical interest to note that the cell yield from one foreskin is sufficient for the production of hundreds of square meters of SC (Parenteau et al., 1991). Potential foreskin donors are prescreened for pathogens via testing of maternal blood. Subsequently, established cell banks are screened for adventitious agents, abnormal karyology and tumorigenicity. F o r m a t i o n o f the delwlat lattice
Dermal lattices are made using a modification of the procedure described by Bell et al. (1979) employing a modified tissue culture insert containing a porous 3 p.m polycarbonate membrane (COSTAR). We have used inserts varying in size from a 2 cm 2 disc for the purpose of producing small SCs for in vitro testing purposes to a 4 x 8 inch rectangular insert for producing large SCs for grafting purposes. The procedure for fomaation of a SC has been described previously (Parenteau et al., 1991).
165 Molphological analysis
lmmunocytochemical analysis
SCs were fixed in the insert using 10% neutral buffered formalin, infiltrated with Tissue Prep II paraffin (Fisher Scientific, Pittsburgh, PA) and embedded in a second paraffin (Paraplast Xtra, Fisher Scientific). Four micrometer sections were stained with Harris' hematoxylin and eosin. For transmission electron microscopy (TEM), SCs were fixed in 2.5% glutaraldehyde - 2.0% paraformaldehyde and post-fixed in 1.0% osmium tetroxide with or without 1.5% potassium ferricyanide. Samples were stained en bloc with 2% uranyl acetate and embedded in Epox 812 resin (Ernest Fullam, Latham, NY). Thin sections were stained with uranyl acetate and lead citrate. Preservation and visualization of the extracellular stratum corneum lipid were enhanced by postfixing
The organotypic culture of human skin keratinocytes and fibroblasts to achieve form and function Nancy L. Parenteau, Patrick Bilbo, Cynthia J.M. Nolte, Valerie S. Mason and Mireille Rosenberg
Organogenesis Inc., Cambridge, MA, 02142, USA Received 3 April 1992; accepted in revised form 27 August 1992
Key words." organotypic, skin, skin culture, tissue engineering, morphology, immunocytochemistry
Abstract We describe an organotypic model of human skin comprised of a stratified layer of human epidermal keratinocytes and dermal fibroblasts within a contracted collagen lattice. Feasible and reproducible production of the skin construct has required the use of traditional as well as specialized culture techniques. The configuration of the construct has been engineered to maintain polarity and permit extended culture at the air-liquid interface. Morphological, biochemical and kinetic parameters were assessed and functional assays were performed to determine the degree of similarity to human skin. Light and ultrastructural morphology of the epidermis closely resembled human skin. The immunocytochemical localization of a number of differentiation markers and extracellular matrix proteins was also similar to human skin. Kinetic data showed a transition of the epidermal layer to a more in vivo-like growth rate during the development of the construct at the air-liquid interface. The barrier properties of the construct also increased with time reaching a permeability to water of less than 2%.h after approximately 2 weeks at the air-liquid interface which is still on average 30-fold more water-permeable than normal human skin. The construct is currently used for in vitro research and testing and is also being tested in clinical applications.
Abbreviations: ALI: Air-liquid interface; CK: Cultured keratinocytes; HS: human skin; HSPG: Heparan sulfate proteoglycan; LSE: Living skin equivalent; SC: skin construct
Introduction The objective of organotypic culture is to develop cellular models which exhibit as many properties of the organ from which they were derived as possible. In the case of skin, this entails the formation of a dermal component consisting of connective tissue cells embedded in extracellular matrix overlaid with a stratified, comified epidermis. In simplest terms, an organotypic culture
of skin can be "engineered" using its three major components, collagen, dermal fibroblasts and epidermal keratinocytes. Combination of these components to achieve organotypic form and function requires proper structural as well as biological support together with a permissive culture environment to maximize phenotypic expression. In the model presented, the structural support is provided by the collagen lattice which surrounds and is contracted
164 by the dermal fibroblasts. The collagen lattice provides a three dimensional framework for the dermal fibroblast and also provides biological support both for the dermal fibroblasts and the overlying keratinocytes. Fibroblasts within a collagen lattice are biosynthetic (Nakagawa et al., 1989) and responsive to cytokines (Dubertret, 1992). While they are proliferative (Parenteau et al., 1991), the fibroblasts are not hyperproliferative as they would be on a culture dish or other nonbiological substrate (Triglia et al., (1991), but exhibit controlled proliferation with the collagen providing biological feedback. Keratinocytes attach readily to the collagen lattice. Basal keratinocytes are able to maintain a cuboidal to columnar morphology and exhibit polarity, synthesizing basement membrane components basally while generating a stratified epidermis apically. The overall polarity of the organotypic skin construct (SC) is maintained by the physical configuration which prevents the epidermal layer from encapsulating the dermal matrix. The constructs are also designed to permit long term culture at the air liquid interface. Culture at the air-liquid interface (ALI) enhances organization and differentiation of the air-exposed stratified epithelium (Asselineau et al., 1985). Culture at the ALI also results in a change in the proliferative index of the epidermal keratinocyte which declines to values more akin to it+ vivo. Engineering the construction of organotypic cultures to be feasible and reproducible has required a blend of conventional and more complex cell culture techniques. Assessment of organotypic cultures involves morphological, biochemical and functional endpoints as well as analysis of cell kinetics. The organotypic SC described is presently being used as a human in vitro skin model for dermatological (Elder et al., 1992), pharmaceutical (Ernesti et al., 1992) and toxicological research and testing (Gay et al., 1992). In addition, it is being tested clinically both as a wound dressing and a skin replacement.
Materials and m e t h o d s Collagen a n d m a s t e r cell b a n k s
Sterile, acid-extracted Type I collagen (Organogenesis Inc.) is derived from bovine tendon and used at 1.1 mg/ml in 0.05% acetic acid. Human dermal fibroblasts are derived from trypsin-collagenase dissociation of newborn foreskin and propagated to passage 4 - 6 in Dulbecco's modified Eagle's medium (DMEM) containing 10% newborn calf serum (NBCS, Hazleton). Human epidermal keratinocytes are derived from explant culture of neonatal foreskin and propagated on collagen-coated plastic in a defined Minimally Supplemented Basal Medium (MSBM) as described in detail elsewhere (Johnson et al., 1992). Keratinocyte master cells banks are established at passage 3. Epidermal cells used in organotypic cultures are cultivated from the frozen master cell bank for one passage in MSBM before seeding onto the collagen lattice. It is of practical interest to note that the cell yield from one foreskin is sufficient for the production of hundreds of square meters of SC (Parenteau et al., 1991). Potential foreskin donors are prescreened for pathogens via testing of maternal blood. Subsequently, established cell banks are screened for adventitious agents, abnormal karyology and tumorigenicity. F o r m a t i o n o f the delwlat lattice
Dermal lattices are made using a modification of the procedure described by Bell et al. (1979) employing a modified tissue culture insert containing a porous 3 p.m polycarbonate membrane (COSTAR). We have used inserts varying in size from a 2 cm 2 disc for the purpose of producing small SCs for in vitro testing purposes to a 4 x 8 inch rectangular insert for producing large SCs for grafting purposes. The procedure for fomaation of a SC has been described previously (Parenteau et al., 1991).
165 Molphological analysis
lmmunocytochemical analysis
SCs were fixed in the insert using 10% neutral buffered formalin, infiltrated with Tissue Prep II paraffin (Fisher Scientific, Pittsburgh, PA) and embedded in a second paraffin (Paraplast Xtra, Fisher Scientific). Four micrometer sections were stained with Harris' hematoxylin and eosin. For transmission electron microscopy (TEM), SCs were fixed in 2.5% glutaraldehyde - 2.0% paraformaldehyde and post-fixed in 1.0% osmium tetroxide with or without 1.5% potassium ferricyanide. Samples were stained en bloc with 2% uranyl acetate and embedded in Epox 812 resin (Ernest Fullam, Latham, NY). Thin sections were stained with uranyl acetate and lead citrate. Preservation and visualization of the extracellular stratum corneum lipid were enhanced by postfixing
