STRUCTURE AND PERMEABILITY OF HUMAN NAIL 373 the dorsal and ventral layers. Permeation fluxes for both model compounds were found to follow the rank order: dorsal-and-ventral-filed nail plate dorsal-filed nail plate ventral-filed nail plate. The permeation parameters obtained from this study are shown in Table II. The permeability coefficient of 5-FU was found to be highest in the intermediate layer, while that for FB was highest in the ventral layer. The diffusion coefficients for both model compounds were lowest in the dorsal layer. The concentra- tions of the drugs in each layer corresponded to their water solubility and flux. The results from this study indicate that the upper dorsal layer of the nail plate constitutes the main barrier to the diffusion of drugs. This finding contradicts the earlier study using shaved nails (35), where the dorsal layer was thought not to be the main diffusional barrier. This discrepancy could possibly be due to the difference in methodology used by the researchers. The latest study with 5-FU and FB (38) appears to have carefully controlled conditions, wherein the thickness of each layer of the nail is precisely mea- sured and differences in permeability properties of each layer have been quantitatively demonstrated. EFFECT OF pH Various studies have been conducted to assess the effect ofpH on nail permeability using ionizable drugs. In order to determine whether vehicle pH affected nail plate perme- ability, the in vitro penetration of an antifungal agent, miconazole (radiolabeled), through the nail was studied at various pH values from 3.1 to 8.2. Miconazole, a weak base with a pKa of 6.65, was thus present in varying degrees of ionization at different pH values. The results of this study indicated that the flux of miconazole was nearly invariant at low and high pH conditions, suggesting that the degree of ionization had little effect on nail permeability. Thus, it was concluded that for this drug, water solubility could be increased by lowering the pH of the formulation without compro- mising the permeability characteristics (39). Although the study described above suggests that pH has little effect on nail perme- ability, other investigations have shown that pH is indeed a factor in altering the nail permeability characteristics of compounds. Soong (35) investigated the effect of pH on the permeation of benzoic acid through the nail plate. In these experiments, the donor contained saturated solutions of the permeant (by maintaining an excess of drug) at various pH conditions. The pH of the receptor compartment corresponded to that of the donor. It was found that as the pH increased from 2.0 to 8.5, the permeability coefficient of benzoic acid decreased by 95.5% and the lag time increased appreciably. The highest permeability coefficient and shortest lag times were observed at the lowest pH (2.0) when benzoic acid (a weak acid) was undissociated. Hence, this study shows that un- charged molecules preferentially permeate the nail barrier. Mertin and Lippold (36) also investigated the effect of pH on nail permeability. Their study involved the evaluation of permeation of benzoic acid (a weak acid) and pyridine (a weak base) at pH values of 2.0 and 7.4 through a bovine hoof membrane. Thus, benzoic acid is in the dissociated state at pH 7.4, while pyridine is dissociated at pH 2.0. Figure 4 shows the change in permeability coefficient with change in pH for these two ionizable drugs. Figure 4 indicates that dissociation of the therapeutic agent hinders its permeation (as reflected by a decreased permeability coefficient). Since keratin is a protein with an isoelectric point of approximately 5, it is positively charged at pH 2.0 and negatively charged at

374 JOURNAL OF COSMETIC SCIENCE