JOURNAL OF COSMETIC SCIENCE 192 The observation that dry skin grades were signifi cantly lower in summer was expected. The observation that erythema grades were not signifi cantly different between seasons is in agreement with the results reported by Visscher et al. who compared dry skin grades on the hands of health-care workers in winter and late spring and found signifi cantly higher dryness grades in the winter with no signifi cant difference in erythema grades between seasons (6). SC COHESION In addition to being a convenient way to sample the SC for analysis, tape stripping can be used to assess the cohesiveness of the SC. Each strip removes less protein from SC that is more cohesive. An advantage of the SquameScan™ system is that it accurately measures SC removed nondestructively (25,36,37), and the tapes can still be used for further anal- ysis. Figure 3 shows protein removed by each tape strip in winter and summer. Each tape removed signifi cantly more SC in winter compared to summer (p 0.001 for each strip level). The difference was especially marked for the fi rst fi ve tape strips. Total protein in strip 4 measured by BCA protein analysis in addition to SquameScan™ is shown in Figure 4. Signifi cantly, more protein was seen on the winter tape 4 strips and the winter/summer ratio for tape 4 was 1.9 by BCA protein analysis and 2.0 by SquameScan™ (Figure 3). The fi nding of reduced SC cohesion in winter dry skin is consistent with the results of Lu et al. (38) who reported results comparing “cosmetic dry skin” to normal skin on the legs analyzed by tape stripping with D-Squame discs. More protein was removed by each of the fi rst 10 strips of dry skin with the differences being largest in the fi rst strips. In order for SC to desquamate properly, the desmosomes (corneodesmosomes) that bind individual corneocytes together must break down, and this process has been shown to be impeded in winter dry skin (2,39). It is probable that each tape strip removes more SC, especially from the less compact outer layers of the SC (40,41), which may be more prominent in dry skin because the desmosomes are not completely broken down in the outer layers of dry skin (2,39,42). Figure 3. SC cohesiveness was evaluated by measuring protein removed by each of the 10 sequential tape strips nondestructively using the SquameScan™ 850A infrared densitometer. Signifi cantly ( p 0.0001) more protein was removed by each strip in winter compared to summer.

EFFECTS OF SEASON ON STRATUM CORNEUM 193 The higher sensitivity of skin to irritants in winter may also be at least partly related to SC cohesiveness. Tumor necrosis factor-α (TNF-α) gene polymorphism makes skin more susceptible to surfactant (19,43) and sensory irritation (44). Davis et al. (45) reported that D-Squame tape stripping removed signifi cantly more SC from subjects with this poly- morphism indicating that they had reduced SC cohesion compared to subject without TNF-α polymorphism. NMF COMPONENTS The NMFs of the SC are primarily free amino acids derived from the breakdown of fi lag- grin (46–49) though lactate may also play a role in keeping the SC hydrated (14). NMF in SC has been investigated by several different workers in various contexts, including body site variations (9,50), effects of age (51), dry skin (51,52), treatments such as soak- ing and lipid extraction (53,54), and barrier perturbation and repair (55). One of the most important NMF components is the very hygroscopic compound PCA formed by the nonenzymatic cyclization of free glutamine produced during fi laggrin hydrolysis (56). PCA is reduced in the lesioned skin of patients with atopic dermatitis (AD), along with reduced skin hydration (57), and in the skin of the elderly (42). Feng et al. reported that PCA levels are reduced in the SC of subjects with “cosmetic dry skin” compared to normal subjects, and PCA levels have been reported to be higher in subjects with lower visual grades for dry skin (52). Tapes 3 and 10 were analyzed for the NMF components (free amino acids, t-UA, and PCA) that arise primarily from the breakdown of fi laggrin (46,48,49). Figure 5A shows PCA levels normalized to protein in tapes 3 and 10 in summer and winter. The difference between strips was not signifi cant but the difference between summer and winter was highly signifi cant at both strips 3 and 10. Other amino acids in strip 10 are shown in Figure 5B. There were small but statistically signifi cant increases in glycine, proline, serine, and the sum of amino acids in summer compared to winter, and t-UA and histi- dine were lower in summer compared to winter. The lower level of t-UA in summer is consistent with the level reported by Egawa and Tagami (9), and may be due to cis–trans isomerization of UA under the effect of ultraviolet Figure 4. Total protein in tape 4 in summer and winter measured by BCA protein analysis.