HAIR PHOTODAMAGE 113 Hair 300 3•0 460 4{i0 500 Wavelength (nm) Figure 2. Normalized fluorescence emission spectra of virgin brown hair (noisy spectrum) and solid L-tryptophan (solid line). Excitation wavelength was 295 nm, and the excitation and emission slits were 1 and 5 nm, respectively. composition. An additional feature of these spectra is the gradual shift of the emission maximum to longer wavelengths as a function of the irradiation time. This might be related to the fact that the observed emission is a superposition of at least two bands with the maxima at 338 nm and 357 nm, as can be demonstrated by spectral subtraction (data not shown). Although Trp fluorescence life-time and, therefore, the emission maximum is known to be sensitive to the polarity of the chromophore environment (24), we are, as yet, unsure of the precise structural significance of this finding. A similar experiment in which powdered L-tryptophan was exposed to 295-nm light also resulted in a gradual decrease in emission intensity with irradiation time (data not shown), albeit at a different rate and without any shift in the emission maximum. This further confirms that the data obtained for hair under these experimental conditions, indeed, reflect Trp photodamage. The kinetic experiments of hair photodecomposition performed at various wavelengths in the UVB region indicate that the 295-nm light is most effective in destroying hair Trp, as shown in Figure 4. This most likely results from the fact that this wavelength corresponds to the absorption maximum of Trp. The kinetics of hair Trp photodamage, as determined by the fluorescence measurements, are complex, and can be affected by a variety of factors. For example, the nonuniformity of exposure of hair protein as a function of distance from the fiber surface to the interior is caused by the attenuation of the exciting (and photodamaging) light beam. Similarly, the intensity of the emitted light is also reduced as it traverses out of the hair. Thus the

114 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS 3 0 400 450 500 Wavelength (nm) Figure 3. Effect of 295 ----- 5 nm irradiation on the tryptophan emission spectrum of Piedmont hair. The excitation wavelength was 295 --- 0.5 nm. The cumulative sample irradiation time for spectra 1-8 from the top was 0, 3, 8, 15, 25, 40, 60, and 105 min, respectively. surface Trp will be damaged relatively fast, followed by a slower decomposition of Trp inside the fiber, which is exposed to light of lower intensity. The kinetics of photode- composition of Trp might be further influenced by the differences in the polarity of the local environment of Trp residues (24). In view of these complications, the kinetic data cannot be subjected to rigorous quantitative analysis required to yield mechanistic information. However, the experimental technique is sufficiently precise for compara- tive purposes. While hair Trp photodamage can be affected by exposure to artificial light, it was also important to demonstrate the significance of this phenomenon in the context of natural sunlight irradiation. Figure 5 shows the kinetics of Trp photodamage in yak hair upon exposure to sunlight during June 1991 in Stamford, Connecticut. A hair tress was weathered outdoors and the fluorescence measurements were performed at indicated times. The data clearly show that hair Trp damage, indeed, occurs under sunlight exposure. Similar results were obtained for Piedmont (human) hair, suggesting that keratins, in general, are vulnerable to such damage. It should be noted that the con- ditions of the kinetic experiment of Trp photodamage by natural light cannot be controlled and thus exactly reproduced. The primary reasons for this are the seasonal and time-of-the-day variations in spectral distribution of light, as well as weather patterns such as clouds. For example, the maximum intensity of UV light in Stamford occurs in summer (June and July) and at noontime. Because of the uncertainty in the determi-