Development of 3D gingival in vitro models using primary gingival cells.

Since March 2013, animal testing for toxicity evaluation of cosmetic ingredients is banned in Europe. This directive applies to all personal care ingredients including oral ingredients. Gingival in vitro 3D models are commercially available. However, it is essential to develop "in house model" to modulate several parameters to study oral diseases, determine the toxicity of ingredients, test biocompatibility, and evaluate different formulations of cosmetic ingredients. Our expertise in tissue engineering allowed us to reconstruct human oral tissues from normal human gingival cells (fibroblasts and keratinocytes). Indeed, isolation from surgical leftover was performed to culture these gingival cells. These cells keep their endogenous capacity to proliferate allowing reconstruction of equivalent tissue close to in vivo tissue. Reconstruction of gingival epithelium, chorion equivalent, and the combination of these two tissues (full thickness) using primary gingival cells displayed all characteristics of an in vivo gingival model.

Establishment and performance assessment of an in-house 3D Reconstructed Human Cornea-Like Epithelium (RhCE) as a screening tool for the identification of liquid chemicals with potential eye hazard.

The objective of this project was to develop an in-house 3D Reconstructed Human Cornea-Like Epithelium (RhCE) to be used as a screening tool in early stages of product development. This study reports the establishment of the experimental procedure and the performance assessment of the model to discriminate liquid chemicals classified as causing serious eye damage/irritation (UN-GHS Category 1/2) from liquid chemicals not requiring such hazard classification (UN-GHS No Category). Histological examination of this ocular equivalent model, based on Normal Human Keratinocytes (NHK) cultured in a chemically defined medium, revealed a stratified and well-organized tissue construct. Moreover, barrier robustness and functionality were demonstrated by the effective time-50 (ET50) of Triton X-100 measurement. The prediction model is based on cytotoxicity assessment following a test chemical exposure. When the mean tissue viability was over 60%, the chemical was defined as No Category; otherwise, it was classified as Category 1/2. In accordance with the applicable OECD guidance document (ENV/JM/MONO (2015)23), the performance of the model for eye hazard identification was evaluated with the minimum list of reference chemicals. As a result, the method scored 84.4% Accuracy, 70.8% Specificity, 100% Sensitivity and 93.3% for Concordance, demonstrating prediction performances close to those of Validated Reference Methods that are commonly used for regulatory purposes. Finally, these results suggest that this in-house RhCE based test method is a qualitative and accurate screening tool for eye hazard identification of liquid chemicals.

Property characterization of reconstructed human epidermis equivalents, and performance as a skin irritation model.

In recent years, in vitro skin models combining cell biology and tissue engineering have been developed in order to replace animal models for toxicological studies and to serve as research support to better understand skin biology. This study reports the development and characterization of a epidermal tissue equivalent meant to be used to develop and to evaluate the effect of applied cosmetic ingredients, and for alternative toxicological testing. This epidermis equivalent model was characterized relative to the morphological characteristics of short- and long-term maintained tissues by performing histological studies. We also studied the integrity of the epidermal barrier. Finally, with the goal of validating its use as a skin irritation test, we studied the irritation potential of 20 chemical references listed in OECD Test Guideline N°439 (In Vitro Skin Irritation: Reconstructed Human Epidermis Test Method). In 2015, OECD officially published the updated version of the Validated Reference Method (VRM) that uses reconstructed human epidermis models for irritation testing, thus offering the possibility for proposed putative similar test methods to obtain a validation agreement through Performance Standards-based validation. In this study, we observed that the epidermal equivalent we developed showed similarities to human in vivo skin, based on the analyzed parameters. Moreover, its performances as a skin irritation test were similar to the ones described in the OECD Test Guideline N°439.