PHOTOCHEMISTRY OF UROCANIC ACID 153 MATERIALS AND METHODS REAGENTS AND SAMPLE PREPARATION All experiments were accomplished in a quartz cuvette using 10 -3 M urocanic acid solutions that were buffered. trans-urocanic acid was purchased from Aldrich (Milwau- kee, WI) and used without further purification. Bromocresol purple was purchased from Fischer. All solutions were buffered in either 0.1 M acetate buffer, pH 5.6, the average pH of the stratum corneum and sweat (14), or 0.1 M potassium phosphate, pH 7.2, the average pH of living cells. c-UA was isolated following the procedure of Anglin and Everett (15). ABSORPTION SPECTRA AND PROBABILITY OF UV ABSORPTION Absorption spectra were recorded on an HP 8245 A diode array spectrophotometer. The probability of absorption of solar radiation was determined from the product of the extinction coefficient of t-UA at pH 7.2 and the solar flux penetrating to the earth's surface at 30 ø (16). PHOTOACOUSTIC CALORIMETRY Experiment and theory (17). In this paper, we do not present our photoacoustic (a.k.a. ultrasonic) data, as it has been published previously (13) however, because we refer to the data in the following discussion, we give a brief description of the technique and the information it provides for those who are unfamiliar with this type of spectroscopy. Photoacoustic calorimetry provides fundamental energetic information on a chromo- phore. As a result, this technique can be used as a spectroscopic tool to determine if the absorbing molecule dissipates the absorbed energy through a singlet state or through potentially photosensitizing or photoreactive triplet state(s), with radical, biradical, or reaction intermediates following absorption into the excited state (18). The latter are of concern in photobiology because reactive intermediates can lead to damage on the cellular level. The technique works as follows: Following absorption of light by a molecule, the energy absorbed can be released through different pathways (both radiative and nonradiative) or retained by the molecule in intermediate state(s). The total energy absorbed must be equal to the sum of the energy released back to the solvent and the energy retained by the molecule in intermediate state(s). The energy released by the molecule to the surrounding solvent is determined as an ultrasonic wave by a transducer connected to the cuvette containing a solution of the molecule and the solvent. The timescale of the energy dissipation processes measured is determined, in part, by the frequency response of the transducer. In our experiment, the heat released on a timescale of 10 -9 s or faster can be recorded. By comparing the amplitude of the ultrasonic wave of the molecule (tram-urocanic acid) to a reference (bromocresol purple) known to release all of the absorbed energy to the solvent, the amount of energy retained by the molecule can be determined. If any energy is retained by the molecule, it is then determined that the molecule forms a long-lived intermediate state, which could be potentially reactive as discussed above.

154 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS RESULTS AND DISCUSSION Figure 2 displays the absorption spectrum of t-UA at pH 7.2 and the probability of photon absorption based upon the solar flux penetrating to the earth's surface (16). The probability of absorption is given by the product of E(}t) * F(}0, where E(}t) is the absorption cross-section of t-UA and F(}t) is the solar flux (16). As the graph shows, the probability of absorption of solar ultraviolet radiation by UA maximizes near 300 nm. UA's absorptivity is small in the UV-A, and little UV-C penetrates to the earth's surface because of absorption by stratospheric ozone, which explains why the probability of absorption of a photon is greatest in the UV-B near 300 nm. Following absorption of solar UV, the isomerization of trans-urocanic acid to cis-urocanic acid is wavelength- dependent (19). Near 300 nm, t-UA isomerizes efficiently to c-UA, as the isomerization quantum yield value is 0.31 at 302 nm, and 0.49 at 313 nm. Near the absorption maximum, however, isomerization from t-UA to c-UA is less efficient, with the isom- erization quantum yield ranging from 0.044 at 264 nm to 0.079 at 289 nm (19). The wavelength-dependent photochemistry exhibited by t-UA results from the presence of multiple electronic states under the structureless absorption spectrum that have • 1.5 ß ,• 1.0 ß .• 0.5 . ?l , 2.' i -., , m m m I 260 280 300 320 340 360 380 Wavelength (nm) Figure 2. The absorption spectrum of t-UA (solid line), pH 7.2, and the probability of absorption of the UV solar flux reaching the earth's surface by t-UA (dashed line) as a function of wavelength. The data show that the greatest probability of UV absorption by t-UA is near 300 nm, where isomerization dominates the photochemistry. The striped region marks the wavelengths that penetrate to the earth's surface and induce reaction intermediates following absorption by either UA isomer, whereas the dotted region indicates the wavelengths where such intermediates are not generated following absorption.