Irradiated Human Fibroblasts as a Substitute Feeder Layer to Irradiated Mouse 3T3 for the Culture of Human Corneal Epithelial Cells: Impact on the Stability of the Transcription Factors Sp1 and NFI.

Because of the worldwide shortage of graftable corneas, alternatives to restore visual impairments, such as the production of a functional human cornea by tissue engineering, have emerged. Self-renewal of the corneal epithelium through the maintenance of a sub-population of corneal stem cells is required to maintain the functionality of such a reconstructed cornea. We previously reported an association between stem cell differentiation and the level to which they express the transcription factors Sp1 and NFI. In this study, we investigated the impact of replacing irradiated 3T3 (i3T3) murine fibroblast feeder cells by irradiated human corneal fibroblasts (iHFL) on the expression of Sp1 and NFI and evaluated their contribution to the proliferative properties of human corneal epithelial cells (hCECs) in both monolayer cultures and human tissue engineered corneas (hTECs). hCECs co-cultured with iHFL could be maintained for up to two more passages than when they were grown with i3T3. Western Blot and electrophoretic mobility shift assays (EMSAs) revealed no significant difference in the feeder-layer dependent increase in Sp1 at both the protein and DNA binding level, respectively, between HCECs grown with either i3T3 or iHFL. On the other hand, a significant increase in the expression and DNA binding of NFI was observed at each subsequent passage when hCECs were co-cultured along with i3T3. These changes were found to result from an increased expression of the NFIA and NFIB isoforms in hCECs grown with i3T3. Exposure of hCECs to cycloheximide revealed an increased stability of NFIB that likely resulted from post-translational glycosylation of this protein when these cells were co-cultured with i3T3. In addition, iHFL were as efficient as i3T3 at preserving corneal, slow-cycling, epithelial stem cells in the basal epithelium of the reconstructed hTECs. Furthermore, we observed an increased expression of genes whose encoded products promote hCECs differentiation along several passages in hCECs co-cultured with either type of feeder layer. Therefore, the iHFL feeder layer appears to be the most effective at maintaining the proliferative properties of hCECs in culture most likely by preserving high levels of Sp1 and low levels of NFIB, which is known for its gene repressor and cell differentiation properties.

Contribution of Sp1 to Telomerase Expression and Activity in Skin Keratinocytes Cultured With a Feeder Layer.

The growth of primary keratinocytes is improved by culturing them with a feeder layer. The aim of this study was to assess whether the feeder layer increases the lifespan of cultured epithelial cells by maintaining or improving telomerase activity and expression. The addition of an irradiated fibroblast feeder layer of either human or mouse origin (i3T3) helped maintain telomerase activity as well as expression of the transcription factor Sp1 in cultured keratinocytes. In contrast, senescence occurred earlier, together with a reduction of Sp1 expression and telomerase activity, in keratinocytes cultured without a feeder layer. Telomerase activity was consistently higher in keratinocytes grown on the three different feeder layers tested relative to cells grown without them. Suppression of Sp1 expression by RNA inhibition (RNAi) reduced both telomerase expression and activity in keratinocytes and also abolished their long-term growth capacity suggesting that Sp1 is a key regulator of both telomerase gene expression and cell cycle progression of primary cultured human skin keratinocytes. The results of the present study therefore suggest that the beneficial influence of the feeder layer relies on its ability to preserve telomerase activity in cultured human keratinocytes through the maintenance of stable levels of Sp1 expression.

Irradiated human dermal fibroblasts are as efficient as mouse fibroblasts as a feeder layer to improve human epidermal cell culture lifespan.

A fibroblast feeder layer is currently the best option for large scale expansion of autologous skin keratinocytes that are to be used for the treatment of severely burned patients. In a clinical context, using a human rather than a mouse feeder layer is desirable to reduce the risk of introducing animal antigens and unknown viruses. This study was designed to evaluate if irradiated human fibroblasts can be used in keratinocyte cultures without affecting their morphological and physiological properties. Keratinocytes were grown either with or without a feeder layer in serum-containing medium. Our results showed that keratinocytes grown either on an irradiated human feeder layer or irradiated 3T3 cells (i3T3) can be cultured for a comparable number of passages. The average epithelial cell size and morphology were also similar. On the other hand, keratinocytes grown without a feeder layer showed heavily bloated cells at early passages and stop proliferating after only a few passages. On the molecular aspect, the expression level of the transcription factor Sp1, a useful marker of keratinocytes lifespan, was maintained and stabilized for a high number of passages in keratinocytes grown with feeder layers whereas Sp1 expression dropped quickly without a feeder layer. Furthermore, gene profiling on microarrays identified potential target genes whose expression is differentially regulated in the absence or presence of an i3T3 feeder layer and which may contribute at preserving the growth characteristics of these cells. Irradiated human dermal fibroblasts therefore provide a good human feeder layer for an effective expansion of keratinocytes in vitro that are to be used for clinical purposes.

Dynamic mechanical stimulations induce anisotropy and improve the tensile properties of engineered tissues produced without exogenous scaffolding.

Mechanical strength and the production of extracellular matrix (ECM) are essential characteristics for engineered tissues designed to repair and replace connective tissues that are subject to stress and strain. In this study, dynamic mechanical stimulation (DMS) was investigated as a method to improve the mechanical properties of engineered tissues produced without the use of an exogenous scaffold, referred to as the self-assembly approach. This method, based exclusively on the use of human cells without any exogenous scaffolding, allows for the production of a tissue sheet comprised of cells and ECM components synthesized by dermal fibroblasts in vitro. A bioreactor chamber was designed to apply cyclic strain to engineered tissues in order to determine if dynamic culture had an impact on their mechanical properties and ECM organization. Fibroblasts were cultured in the presence of ascorbic acid for 35 days to promote ECM production and allow the formation of a tissue sheet. This sheet was grown on a custom-built anchoring system allowing for easy manipulation and fixation of the tissue in the bioreactor. Following the 35 day period, tissues were maintained for 3 days in static culture (SC), or subjected either to a static mechanical stimulation of 10% strain, or a dynamic DMS with a duty cycle of 10% uniaxial cyclic strain at 1Hz. ECM was characterized by histology, immunofluorescence labeling and Western blotting. Both static and dynamic mechanical stimulation induced the alignment of assessed cytoskeletal proteins and ECM components parallel to the axis of applied strain and increased the ECM content of the tissues compared to SC. Measurement of the tensile mechanical properties revealed that mechanical stimulation significantly increases both the ultimate tensile strength and tensile modulus of the engineered tissues when compared to the non-stimulated control. Moreover, we demonstrated that cyclic strain significantly increases these parameters when compared to a static-loading stimulation and that mechanical stimulation contributes to the establishment of anisotropy in the structural and mechanical properties of self-assembled tissue sheets.

Tissue engineering of skin and cornea: Development of new models for in vitro studies.

Human beings are greatly preoccupied with the unavoidable nature of aging. While the biological processes of senescence and aging are the subjects of intense investigations, the molecular mechanisms linking aging with disease and death are yet to be elucidated. Tissue engineering offers new models to study the various processes associated with aging. Using keratin 19 as a stem cell marker, our studies have revealed that stem cells are preserved in human skin reconstructed by tissue engineering and that the number of epithelial stem cells varies according to the donor's age. As with skin, human corneas can also be engineered in vitro. Among the epithelial cells used for reconstructing skin and corneas, significant age-dependent variations in the expression of the transcription factor Sp1 were observed. Culturing skin epithelial cells with a feeder layer extended their life span in culture, likely by preventing Sp1 degradation in epithelial cells, therefore demonstrating the pivotal role played by this transcription factor in cell proliferation. Finally, using the human tissue-engineered skin as a model, we linked Hsp27 activation with skin differentiation.