There is a clinical need for skin substitutes to replace full-thickness skin loss. Our group has developed a bilayered skin substitute produced from the patient's own fibroblasts and keratinocytes referred to as Self-Assembled Skin Substitute (SASS). After cell isolation and expansion, the current time required to produce SASS is 45 days. We aimed to optimize the manufacturing process to standardize the production of SASS and to reduce production time. The new approach consisted in seeding keratinocytes on a fibroblast-derived tissue sheet before its detachment from the culture plate. Four days following keratinocyte seeding, the resulting tissue was stacked on two fibroblast-derived tissue sheets and cultured at the air–liquid interface for 10 days. The resulting total production time was 31 days. An alternative method adapted to more contractile fibroblasts was also developed. It consisted in adding a peripheral frame before seeding fibroblasts in the culture plate. SASSs produced by both new methods shared similar histology, contractile behavior in vitro and in vivo evolution after grafting onto mice when compared with SASSs produced by the 45-day standard method. In conclusion, the new approach for the production of high-quality human skin substitutes should allow an earlier autologous grafting for the treatment of severely burned patients.

There is a high incidence of failure and recurrence for chronic skin wounds following conventional therapies. To promote healing, the use of skin substitutes containing living cells as wound dressings has been proposed. The aim of this study was to produce a scaffold-free cell-based bilayered tissue-engineered skin substitute (TES) containing living fibroblasts and keratinocytes suitable for use as wound dressing, while considering production time, handling effort during the manufacturing process, and stability of the final product. The self-assembly method, which relies on the ability of mesenchymal cells to secrete and organize connective tissue sheet sustaining keratinocyte growth, was used to produce TESs. Three fibroblast-seeding densities were tested to produce tissue sheets. At day 17, keratinocytes were added onto 1 or 3 (reference method) stacked tissue sheets. Four days later, TESs were subjected either to 4, 10, or 17 days of culture at the air–liquid interface (A/L). All resulting TESs were comparable in terms of their histological aspect, protein expression profile and contractile behavior in vitro. However, signs of extracellular matrix (ECM) digestion that progressed over culture time were noted in TESs produced with only one fibroblast-derived tissue sheet. With lower fibroblast density, the ECM of TESs was almost completely digested after 10 days A/L and lost histological integrity after grafting in athymic mice. Increasing the fibroblast seeding density 5 to 10 times solved this problem. We conclude that the proposed method allows for a 25-day production of a living TES, which retains its histological characteristics in vitro for at least two weeks.

Human keratinocyte culture has provided the means to treat burns, wounds and skin pathologies. To date, to efficiently culture keratinocytes, cells are cultured on an irradiated feeder layer (iFL), either comprising human (iHFL) or murine (i3T3FL) fibroblasts, and the culture medium is supplemented with a cyclic adenosine monophosphate (cAMP) accumulation inducing agent such as isoproterenol (ISO) or cholera toxin (CT). Previous studies have characterized how the feeder layer type and the cAMP inducer type influence epithelial cells’ phenotype independently from one another, but it is still unknown if an optimal combination of feeder layer and cAMP inducer types exists. We used sophisticated statistical models to search for a synergetic effect of feeder layer and cAMP inducer types on human keratinocytes’ proliferative potential. Our data suggests that, when culturing human keratinocytes, using iHFL over i3T3FL increases population doublings and colony-forming efficiency through signaling pathways involving Ak mouse strain thymoma (Akt, also known as protein kinase B) isoforms 1 to 3, signal transducer and activator of transcription 5 (STAT5), p53, and adenosine monophosphate activated protein kinase α1 (AMPKα1). Both tested cAMP inducers ISO and CT yielded comparable outcomes. However, no significant synergy between feeder layer and cAMP inducer types was detected. We conclude that, to promote human keratinocyte growth in the early passages of culture, co-culturing them with a human feeder layer is preferable to a murine feeder layer.