154 JOURNAL OF COSMETIC SCIENCE Figure 2. Conventional TEM findings in hair shafts: (A) Before irradiation. (B) After UVA irradiation. (C) After UVB irradiation. Arrowhead: holes in the endocuticles. Arrow: cleavage along the endocuticles by confluence of holes. Asterisk: confluence of holes. Figure 3. Lipid TEM findings in hair shafts: (A) Before irradiation. (B) After UV A irradiation. (C) After UVB irradiation. Arrowhead: bulging portions in the intercellular lipid layers. Arrow: small focal lacunae along the inrercellular spaces.

HAIR SHAFT DAMAGE AFTER UV IRRADIATION 155 25hD 2011D UVA 151tD IOlcD 251tD 20110 1511D UVB lOltD Figure 4. Western blot analysis of UV-light-irradiated hair samples with ubiquitin antibodies. Upper panel: after UV A irradiation. Lower panel: after UVB irradiation. parison to the hair shafts after UV A irradiation, there are a few relatively weak positive findings around the 1 0kDa area after UVB irradiation. We think that the positive finding before UV A irradiation may result from the preceding natural weathering processes. There have been several reports on the evaluation of photoaging and photoprotection of hair fibers with microscopy. Braida et al. (9) reported TEM findings on hair shafts after using two kinds of sunlight stimulators. On the other hand, Bousquet et al. ( 10) reported analysis of patterns of cuticles for evaluation of hair photoprotection with confocal microscopy. The former revealed the difference in melanin particles in irradiated hair samples with TEM, and the latter reported morphological findings in hair cuticles with confocal microscopy. In this study, we carried out investigations with four detailed tools having different strong points: SEM findings reveal superficial cuticular changes, con- ventional TEM findings reveal deep cuticular changes, and TEM findings using Lee's fixative reveal intercellular lipid layer changes more precisely. We also performed a protein analysis using ubiquitin antibodies. Ubiquitin-mediated degradations of regu- latory proteins are known to play important roles in numerous processes, including cell-cycle progression, signal transduction, transcriptional regulation, receptor down- regulation, and endocytosis (11). In a recent study, Inoue et al. have suggested that ubiquitin could be an indicator of hair damage. It was reported that the main wavelength of UV light absorbed by keratin was UVB, which could do damage mainly by the cleaving of the protein bonds and protein crosslinking (12). In the hair shaft, UVB is known to penetrate about 5 µm in depth, and the intensity decreases exponentially after that (12). Considering that intact hair cuticles are 6-10 layers, each with a thickness of 0.3-0.5µm, UVB seems mainly to affect the cuticles. This suggests that hair damage by UVB irradiation is mainly confined to the superficial layers of the hair shaft, the cuticle layers. We think our findings are related to the penetration depth of UV light mentioned above. UV A irradiation can penetrate deeply into the cortex, and so biochemical changes, including cuticles and