110 JOURNAL OF COSMETIC SCIENCE cytes (the cellular bricks), interlocked with the lipid lamellae (the intercellular mortar) (2-5). Compared to the porous structure of the viable epidermis and the porous-and­ hydrated structure of the dermis, the rigid and ordered structure of the stratum corneum makes it a very effective permeability barrier that is primarily responsible for the skin barrier function (2--4). The lipid lamellae of the SC consist of lipid bilayers alternating with aqueous, hydrophilic layers (1--4). Under passive skin permeation conditions, per­ meants traverse the SC through diffusion across the lipid lamellae. Although diffusion through the "oily" lipid lamellae can explain the permeation of hydrophobic molecules across the SC, it cannot explain the permeation of hydrophilic molecules across the SC, as observed in many earlier studies (6-9). Indeed, if no aqueous/hydrophilic transport pathways existed within the SC oily lipid domain, then aqueous/hydrophilic permeants, for example mannitol (6-9), could not traverse the SC solely through the lipoidal/hydrophobic pathways that exist in the lipid bilayer domains in the SC. The observation that hydrophilic solutes are able to permeate across the SC, even under passive skin permeation conditions, has led researchers to propose the existence of tortuous, aqueous porous pathways through the intercellular lipid lamellae in the SC. In fact, Menon and Elias (10) have established a morphological basis for the existence of a pore pathway in the mammalian SC. They applied hydrophilic and hydrophobic tracers in vivo to murine skin under passive skin permeation conditions, and also under enhanced skin permeation conditions, using chemical enhancers, a lipid synthesis inhibitor, sonophoresis, and iontophoresis, and following that, they utilized ruthenium tetroxide staining and microwave post fixation methods to visualize the resulting penetration pathways (10). Their results revealed that both the hydrophobic and the hydrophilic tracers localized to discrete lacunar domains embedded within the extracellular lipid lamellar domains (10). Menon and Elias also observed that under skin permeation enhancement conditions, the lacunar domains exhibited an increasing extent of structural continuity when compared to passive skin permeation conditions (10). Hence, structurally continuous lacunar domains have been considered by Menon and Elias as providing a physical basis for the existence of aqueous pores and polar pathways through the intercellular lipoidal mortar in the SC (10). These aqueous pores in the SC provide the primary skin barrier penetration and transport pathways for hydrophilic chemicals, which would otherwise not be able to penetrate into the skin barrier through the lipoidal, hydrophobic pathways that exist in the SC (6-11). In general, surfactants commonly encountered in skin care formulations are known to reduce the barrier properties of the skin (11-15). It is well-accepted that surfactants have to first penetrate into the skin barrier before they can reduce the skin barrier properties. Therefore, if a formulator can minimize surfactant skin penetration, this should also minimize the ability of the surfactant to reduce the skin barrier properties. Sodium dodecyl sulfate (SDS), an anionic surfactant and a model skin irritant, penetrates into and disrupts the skin barrier upon contacting it from an aqueous solution. The SDS mono­ mers self-assemble to form micelles at concentrations above the critical micelle concen­ tration (CMC). Moore et al. (11) and others (12,13) have observed, both in vitro and in vivoJ that the SDS-induced skin barrier disruption is dose-dependent, and that it in­ creases with an increase in the total SDS concentration above the CMC of SDS. This important observation contradicts the well-accepted monomer penetration model (MPM), which attempts to explain surfactant skin penetration by considering solely the role of the surfactant monomers that can penetrate the skin barrier through the aqueous

SDS MICELLES IN SKIN BARRIER PERTURBATION 111 pores in the SC (11-23). The MPM does not consider the possibility that surfactant micelles may also contribute to surfactant skin penetration, and consequently, to sur­ factant skin barrier disruption, since it considers the micelles to be too large to penetrate through the aqueous pores that exist in the SC. In her comprehensive review of surfac­ tant-skin interactions, Rhein (14) stated that the observed dose dependence of surfac­ tant-induced skin irritation beyond the CMC cannot be explained solely by the contri­ bution of the monomeric surfactant. Indeed, Agner and Serup (13) had earlier observed that the severity of the transepidermal water loss (TEWL) induced by SDS increased as the SDS concentration increased beyond the CMC of SDS (8.7 mM) (13). In separate studies, Ananthapadmanabhan et al. (15) and Faucher and Goddard (16) observed that as the SDS concentration increased beyond the CMC, the extent of SDS skin penetration also increased. Through in vitro SDS skin penetration studies, Moore et al. (11) provided substantial evidence that indicates that the amount of SDS that can penetrate into the skin barrier is dose-dependent, and furthermore, that the SDS surfactant in micellar form also contributes to SDS skin penetration. In addition, Moore et al. demonstrated conclusively that the contribution of the SDS micelles to SDS skin penetration dominates that of the SDS monomers at concentrations above the CMC, which are typically encountered in skin care formulations (11). In this paper, we have further investigated, from a mechanistic perspective, how SDS micelles may contribute to SDS skin penetration, thereby leading to the previously observed dose dependence of SDS-induced skin barrier perturbation (11-23). Specifi­ cally, we will provide new evidence, through in vitro transdermal permeability and skin electrical current measurements, in the context of a hindered-transport porous pathway model of the SC (6-9,42), that the aqueous pores in the SC increase both in size and in number density when skin is exposed to an aqueous SDS contacting solution, such that the average pore radius is larger than the SDS micelle radius. As a result, SDS micelles, contrary to the view put forward by the MPM, are not sterically hindered from pen­ etrating into the skin barrier through these pores. Inspired by our proposed mechanistic understanding of how SDS micelles may contrib­ ute to SDS-induced skin barrier perturbation, we have also investigated in vitro whether the addition of glycerol, a well-known humectant and skin beneficial agent, to the aqueous SDS contacting solution can minimize the observed contribution of the SDS micelles to SDS skin penetration. Although not within the scope of this paper, if shown to be valid in vivo1 such a strategy can also significantly reduce the amount of SDS that can penetrate into the skin barrier and induce skin barrier perturbation in vivo. Our approach considers exposing skin in vitro to aqueous mixtures of glycerol and SDS. The importance of glycerol (or glycerin) in skin care products is well established, and glycerol is widely used in cosmetic and pharmaceutical formulations (24-31). To explain its in vivo benefits, studies have focused on its humectant and smoothing effects (25) and on its protective functions in emulsion systems against skin irritation (26). Researchers have shown that glycerol diffuses into the SC, increases skin hydration, and relieves clinical signs of dryness (27-29). One of the views regarding the effect of glycerol on the skin held by researchers today is that it may influence the crystalline arrangement of the