Skin-mimetic chromatography for prediction of human percutaneous absorption of biologically active compounds occurring in medicinal plant extracts.

Abstract

The main aim of this study was to predict quantitatively human percutaneous absorption of chosen compounds commonly occurring in plants which can be used as medicinal extracts in the drug and beauty industries. The most important human percutaneous descriptors, i.e. logKp (logarithm of the water/skin partition coefficient) and logJmax (logarithm of the maximum flux of solutes penetrating the skin), of fatty acids and polyphenols were determined using both in vitro and in silico methods. For in vitro determination of human percutaneous absorption, micellar liquid chromatography based on hexadecyltrimethylammonium bromide, sodium dodecyl sulfate and polyoxyethylene (23) lauryl ether (Brij35) was used. Human percutaneous absorption was characterized by entirely new QSAR/QRAR models based on retention, lipophilic, steric and electronic data as well as on the linear free energy relationship parameters. Many different correlations between human skin absorption and different physicochemical parameters were performed, e.g. the in silico estimated logKp value was correlated with the retention parameter logkw (logarithm of the retention factor extrapolated to pure water) from the systems imitating a cutaneous environment (R2 = 0.92). Moreover, the influence of lipophilicity on percutaneous absorption was examined. The obtained correlation was excellent (R2 = 0.95).