SURFACTANT SWELLING OF STRATUM CORNEUM 135 Table IV Effects of Ethoxylated Alcohol Sulfates and Amphoterics on SLS-Induced Stratum Corneum Swelling Stratum Corneum Swelling** (% increase from hydrated length) Experiment No. Treatment* 1 hr. 3 hr. 6 hr. 1 1% SLS 31.0 ñ 3.9 43.5 ñ 3.2 46.1 _+ 3.6 1 1% SLS + 1%AEOS-3EO* 14.0 ñ 1.0 18.0 -+ 1.8 21.6 ñ 1.3 1 1% SLS + 1%AEOS-6EO* 13.5 ñ 3.7 16.7 ñ 3.4 19.1 ñ 3.6 2 1% SLS 29.4 -+ 4.4 39.0 - 4.9 46.8 ñ 4.1 2 1% SLS + 1% LDAO* 5.7 ñ 0.9 7.6 ñ 1.4 7.7 -+ 1.7 2 1% SLS + 0.5% LDAO* 13.0 ñ 3.5 21.0 --- 6.8 25.8 ñ 8.4 2 1% SLS + 1% Betaine* 10.9 - 2.4 17.0 ñ 3.8 21.0 ñ 4.4 2 1% SLS + 0.5% Betaine* 20.0 ñ 4.8 29.0 ñ 4.0 34.5 ñ 4.2 * Values are means ñ S.D. AEOS-3EO, AEOS-6EO, lauryl dimethyl amine oxide (LDAO), and coco amidopropyl betaine when tested alone at 1% w/w all produced small or negligible increases in swelling above the hydrated lengths (i.e. equal to or less than 5% increase in 6 hours). All treatments caused a significant reduction in swelling compared to 1% SLS treatment alone (P • 0.01). ** Times shown are incubation times after which readings were taken. surfactant, also suggests that the swelling response may be primarily due to surfactant monomer (or submicellar species) as discussed by others (3). This is reasonable since SLS micelies would presumably be too large (aggregation number is approximately 80) to penetrate the membrane. Observed increases in surfactant-induced swelling with time suggest continuous disruption of the secondary and tertiary structure of keratin proteins resulting in expo- sure of new binding sites. Reversibility of the swelling supports this view. Reversible interactions with skin lipids may occur, although one would expect that the lipids would be extracted by the surfactant and that this type of reaction should render the system irreversible, which is not the case under these conditions. Disruption of protein structure by surfactants as part of the swelling mechanism was also suggested by Scheu- plin and Ross (13) and Putterman eta/. (3) and may explain the edge curling phenom- enon observed by Tavss eta/. (4). Examination of the effects of systematic variation of surfactant structure on stratum corneum swelling allows some conclusions regarding the parameters controlling swelling and the mechanism of irritation by surfactants. These are discussed in the following sections. EFFECT OF INCREASING THE EXTENT OF ETHOXYLATION Increasing the extent of ethoxylation of alkyl (C12-C14 average) sulfate reduced swelling. This may be explained by several factors. Increasing the degree of ethoxyla- tion increases the molecular size of the penetrating species. Thus rate of penetration of the larger molecules into the matrix will be more difficult. Along with this, the critical micelie concentration decreases with increasing ethoxylation thus the submicellar levels of surfactant are lower for the higher ethoxylates. These factors taken together are con- sistent with the results.

136 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS THE C12 OR C14 MAXIMUM IN SWELLING For all the homologous surfactant series tested (except cationic surfactants which did not cause swelling), a maximum in swelling occurred for either the C•2 or C•4 homo- 1ogue. In vivo irritation also follows the same pattern for anionic surfactants (5,14). Our group (2) previously explained for similar results with alkyl sulfate homologues that this swelling maximum demonstrates the importance of hydrophobic interactions to the swelling mechanism. The hydrophobic bond energy increases with increasing chain length. However, at longer chain lengths, the larger molecules penetrate less efficiently into the matrix, and thus swelling is reduced for the longer chain alkyl sulfates. We also know that the submicellar concentrations of surfactant are lower for the higher alkyl chains due to the lower CMC. This would contribute to the lower relative amounts of swelling observed for the higher alkyl chain lengths. However, conversely, monomer levels would be higher for the shorter chain alkyl sulfates (since the CMC is relatively higher), yet swelling is depressed along with in vivo irritation for the shorter chain compared with the C•2 alkyl sulfate (5,14). This lends additional support for the importance of hydrophobic interactions in the swelling mechanism and probably for irritation. The importance of hydrophobic interactions to penetration of molecules through the epidermis has been clearly demonstrated (15,16). THE SWELLING MECHANISM Swelling is a two-step process: diffusion into the matrix followed by expansion against the elastic retractive forces of the substrate. Both steps are time-dependent. The swelling mechanism must involve a combination of ionic binding of the hydrophilic group as well as hydrophobic interactions of the alkyl chains with the substrate. We conclude this because nonionic surfactants don't promote swelling. We also conclude that the type of hydrophilic group is important to the swelling mechanism since, unlike anionic surfactants, cationic surfactants do not promote swelling and in fact cause shrinkage at longer incubation times (Figure 4). We speculate that cationic surfactants (at neutral pH) are attracted to and bind electrostatically to the excess of negative sites on the membrane (2, 17). This substitutes a hydrophobic cation for a small hydrophilic cation which converts a charged hydrophilic site into a hydrophobic site that repels water and may result in shrinking or tightening up of the structure. Anionic surfactants (at neutral pH), on the other hand, bind to hydrophobic sites by their hydrophobic tails, leaving the negative end group exposed. This type of binding creates additional anionic sites in the membrane, resulting in repulsive forces that separate the matrix, exposing more water-binding sites. This explanation is supported by our previous studies (2) of the effect of pH on surfac- tant-induced swelling. We demonstrated that at acidic pH anionic surfactants cause the least amount of swelling. At the lower pH, electrostatic binding of anionic surfactants would be enhanced to the new cationic sites produced. Increased ionic binding to the membrane would expose the hydrophobic tails which themselves associate and repel water, resulting in the reduced swelling at the lower pH compared to higher pH's. In the same manner, a cationic surfactant caused the least swelling at higher pH. This may be attributed again to enhanced ionic binding at higher pH for the cationic surfactant (which is known to be the case see for example reference 2), producing more association of hydrophobic tails, excluding water.