Partenaires et organisations internationales
(Anglais)
|
L'Oréal SA, F-92583 Clichy; Beiersdorf AG. Hamburg (D), Rijksuniversiteit Leiden, Dept. of Dermatology, Leiden (NL)
|
Résumé des résultats (Abstract)
(Anglais)
|
The aim of this 3-year project was to pre-validate industrial (EpiDerm©, Episkin©, SkinEthic©) and 'in house' reconstructed human skin models in three main areas of cutaneous pharmacotoxicology: percutaneous absorption, cutaneous metabolism and irritation.
For compounds with largely differing physico-chemical properties (mannitol, caffeine, lauric acid) the human percutaneous permeability rank order was correctly predicted when applied at infinite as well as at finite dose to EpiDerm© or Episkin©. In contrast, smaller effects of different cosmetic formulations (w/o and o/w emulsionons, aqueous solution, hydrogel) on caffeine absorption were less predictable, which might be explained by the increased hydration of the superficial stratum corneum layers and/or less pronounced barrier properties due to the different composition and organization of the intercellular stratum corneum lipids, particularly ceramides 5, 6 and 7.
The activity of three key enzymes involved in epidermal xenobiotic metabolism has been detected in EpiDerm©, Episkin© and SkinEthic©: glutathione S-transferase (GST) and NAD(P)H: quinone reductase (NQR) as basic activities, and P450 IA1 (EROD) only after 3-methylcholanthrene induction. Episkin© provided the best results, i.e. practically the same enzyme activities as normal human epidermis.
Cutaneous irritation was studied by comparing in vitro parameters, such as e.g. viability (MTT), release of IL-1a and Lactate dehydrogenase (LDH), with the human in vivo irritancy parameters of a modified Frosch-Kligman Soap Chamber Test (repeated application for 4 days, final evaluation at day 5). Multivariate descriptive statistics have shown that MTT viability after 16 h of exposure and ET-50 (time of exposure required to reduce the viability of the epidermis by 50 %) were the best endpoints to discriminate between irritant and non-irritant products. The in vivo/in vitro comparison of the classification (mean total irritation score at day 5 versus %MTT viability after 16 h) of 22 cosmetic products (shampoos, emulsions, gels, oils, creams, mascaras, 12 irritant and 10 non irritant) resulted in an excellent concordance with all epidermal models tested (EpiDerm©, Episkin© and the Cosmital 'in-house' model; sensitivity = 92 %, specificity = 100 %, concordance = 95 %, kappa = 0.91). Other parameters such as the release of IL-1a or LDH can contribute for some models to the refinement of the prediction of the irritative potential of some substances or to the better understanding of their mechanisms of action. The protocol transferability from one laboratory to an other one was demonstrated with 6 cosmetic products and one standard surfactant (1 % sodium lauryl sulfate) by the excellent interlaboratory reproducibility of the ET-50 values and by an identical classification into 5 irritants and 2 non-irritants in both labs.
These results have been presented at the 'International Federation of the Society of Cosmetic Chemists' Conference in Stockholm (May 7-9, 2001) and are submitted to 'Toxicology in vitro' and to 'Skin Pharmacology and Applied Skin Physiology'.
|
