Differential effects of detergents on keratinocyte gene expression

We have studied the effect of various detergents on keratinocyte gene expression in vitro, using an anionic detergent (sodium dodecyl sulfate), a cationic detergent cetyltrimethylammoniumbromide (CTAB), and two nonionic detergents, Nonidet P-40 and Tween-20. We measured the effect of these detergents on direct cellular toxicity (lactate dehydrogenase release), on the expression of markers for normal differentiation (cytokeratin 1 and involucrin expression), and on disturbed keratinocyte differentiation (SKALP) by northern blot analysis. As reported in other studies, large differences were noted in direct cellular toxicity. In a culture model that mimics normal epidermal differentiation we found that low concentrations of sodium dodecyl sulfate could induce the expression of SKALP, a proteinase inhibitor that is not normally expressed in human epidermis but is found in hyperproliferative skin. Sodium dodecyl sulfate caused upregulation of involucrin and downregulation of cytokeratin 1 expression, which is associated with the hyperproliferative/inflammatory epidermal phenotype found in psoriasis, wound healing, and skin irritation. These changes were not induced after treatment of cultures with CTAB, Triton X-100, and Nonidet-P40. This effect appeared to be specific for the class of anionic detergents because sodium dodecyl benzene sulfonate and sodium laurate also induced SKALP expression. These in vitro findings showed only a partial correlation with the potential of different detergents to induce clinical, biophysical, and cell biologic changes in vivo in human skin. Both sodium dodecyl sulfate and CTAB were found to cause induction and upregulation of SKALP and involucrin at low doses following a 24 h patch test, whereas high concentrations of Triton X-100 did not. Sodium dodecyl sulfate induced higher rates of transepidermal water loss, whereas CTAB treated skin showed more signs of cellular toxicity. We conclude that the action of anionic detergents on epidermal keratinocytes is qualitatively different from the other detergents tested, which might have implications for in vitro toxicology studies that use cell biologic parameters as a read-out. We would hypothesize that detergents cause skin injury by several mechanisms that include direct cellular toxicity, disruption of barrier function, and detergent specific effects on cellular differentiation, as demonstrated here for sodium dodecyl sulfate, sodium dodecyl benzene sulfonate, and sodium laurate.