122 � 10 1 9 ... Q) 'E a C: 7 Q) 6 .c: -= 5 � 4 � 3 � 2 JOURNAL OF COSMETIC SCIENCE CMC of SDS = 8.7 mM .I Micelle Contribution - C: 8 1 } Monomer g Contribution o o---=------.._______.._______________......, 0 50 100 150 200 Total SDS Concentration in the Aqueous Contacting Solution (mM) Figure 4. Comparison of SDS skin penetration in vitro induced by aqueous contacting solutions of SDS (diamonds) and of SOS + 10 wt% glycerol (triangles). 'fhe dotted vertical line at an SDS concentration of 8.7 mM denotes the CMC of SDS. 'fhe dashed line passing through the diamonds is drawn as a guide to the eye. 'fhe error bars represent a standard error based on 6-10 p-F'fS samples. barrier increases significantly. In Figure 4, the contribution of the SDS monomers to SDS skin penetration above the CMC remains approximately constant above 8.7 mM (the CMC value), and corresponds to the horizontal solid line. On the other hand, the total SDS contribution to SDS skin penetration increases above the CMC, and corre­ sponds to the dashed line, drawn as a guide to the eye. Clearly, the difference between the dashed and the solid lines at any given total SDS concentration corresponds to the contribution of the SOS micelles to SDS skin penetration. Note that below the CMC, only the SDS monomers are available for penetration into the skin. Consequently, the diamonds and the triangles overlap below the CMC (see Figure 4). These results are in excellent agreement with the SDS skin penetration results reported by Moore et al. (11). Indeed, these authors showed earlier that: (i) there is a significant SDS micellar contri­ bution to SDS skin penetration, and (ii) the SDS micellar contribution increases with an increase in the total SDS concentration above the CMC. However, in this paper, we have demonstrated in vitroJ for the first time, that the significant SDS micellar contribution to SDS skin penetration also leads to a large extent of SDS skin barrier perturbation, as quantified by the observed increases in the skin electrical currents and in the mannitol transdermal permeabilities (see Figures 2 and 3, respectively). These in vitro results suggest, from a practical, formulation design point of view, that any strategy designed to minimize skin barrier perturbation induced by surfactants like SDS, in addition to minimizing the penetration of the surfactant monomers into the skin, as was done in the past, may also benefit from minimizing the penetration of the surfactant micelles into the skin. In this paper, we have investigated in vitro such a simple and useful practical strategy by using mixtures of SDS and glycerol, which we discuss next. Specifically, we conducted skin radioactivity assays using 14 C-SDS in the presence of 10 wt% added glycerol in aqueous solution to measure the amount of SDS that may

SDS MICELLES IN SKIN BARRIER PERTURBATION 123 penetrate into the skin barrier in the presence of glycerol (corresponding to the triangles in Figure 4). It is interesting to observe that the triangles and the diamonds overlap below the CMC in Figure 4. At an SDS concentration below the CMC of SDS (8.7 mM), the SDS aqueous contacting solution essentially consists of SDS monomers contacting the skin. Therefore, upon adding 10 wt% glycerol to the SDS aqueous contacting solution, one can observe that the SDS monomers are not hindered from penetrating into the skin. However, the addition of 10 wt% glycerol to the SDS aqueous contacting solution at concentrations above the CMC significantly impacts SDS skin penetration. Indeed, as can be seen, the presence of 10 wt% glycerol in the SDS contacting solution eliminates almost completely the amount of SDS that can penetrate into the skin barrier from the high SDS concentration contacting solutions. The significant difference be­ tween the diamonds (or the dashed line) and the triangles (which lie very close to the SDS monomer contribution corresponding to the solid line) clearly shows that SDS micelles, which would have contributed to skin penetration in the absence of 10 w% glycerol, cannot do so in the presence of 10 wt% glycerol in the contacting solution. These in vitro results suggest that the addition of 10 wt% glycerol to the SDS contacting solutions may also represent a simple, yet very useful, practical strategy to mitigate SDS-induced skin barrier perturbation in vivo by preventing the SDS micelles from penetrating into the skin barrier. In the following section, we put forward several hypotheses to explain, from a mecha­ nistic viewpoint, why glycerol, without affecting the skin penetration ability of the SDS monomers, is able to significantly reduce the ability of the SDS micelles to contribute to SDS skin penetration in vitro. POSSIBLE HYPOTHESES TO EXPLAIN THE EFFECT OF GLYCEROL ON THE OBSERVED IN VITRO DOSE INDEPENDENCE OF SDS SKIN PENETRATION Using micelle stability arguments put forward by Patist et al. (43), Moore et al. (11) have shown that the kinetics of micelle dissolution cannot be invoked to explain the observed dose dependence of SDS skin penetration. Moore et al. have also compared the time constant for the breakup of SDS micelles to replenish the decreased SDS monomer supply to the SC as the SDS molecules penetrate into the skin with the time constant for SDS diffusion across the skin. This comparison has unambiguously shown that the rate-determining step for SDS skin penetration is governed by the diffusion, or the penetration, through the SC and not by the micelle kinetics (11). Furthermore, Moore et al. have shown that micelle disintegration upon impinging on the SC and subsequent absorption by the skin barrier also does not seem to be a plausible mechanism to explain the observed dose dependence of SDS skin penetration (11,44,45). With all of the above in mind, according to Moore et al. J a consistent hypothesis to explain the observed dose dependence of SDS skin penetration considers the ability of SDS micelles to penetrate into the SC, based on a size limitation (11). Without directly measuring the skin aqueous pore radius, r pore • and the pore number density, e/rr, Moore et al. hypothesized6 6 Note that Moore et al. (11), to their credit, compared micelle sizes for free SDS micelles and PEO-bound SDS micelles, using DLS measurements similar to those reported here, and found that the PEO-bound SDS mirPlli h:1d R larger hydrodynamic radius than the free SDS micelle. This observation, along with the observation that the PEO-bound SDS micelle, unlike the free SDS micelle, did not contribute to SDS skin (continued on p. 124)