13116
J. Phys. Chem. B 2007, 111, 13116-13123
UV Protection and Singlet Oxygen Quenching Activity of Aloesaponarin I Shin-ichi Nagaoka,*,† Akiko Fujii, Megumi Hino, Mai Takemoto, Misaki Yasuda, Mariko Mishima, and Keishi Ohara† Department of Chemistry, Faculty of Science, Ehime UniVersity, Matsuyama 790-8577, Japan
Akane Masumoto and Hidemitsu Uno DiVision of Organic Synthesis and Analysis, Department of Molecular Science, Integrated Center of Sciences, Ehime UniVersity, Matsuyama 790-8577, Japan
Umpei Nagashima Research Institute for Computational Sciences, National Institute of AdVanced Industrial Science and Technology, 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan ReceiVed: July 5, 2007; In Final Form: September 3, 2007
The UV protection and singlet oxygen quenching of aloesaponarin I have been studied by means of laser spectroscopy. The excited-state intramolecular proton transfer that provides the UV protection takes place along only one of the molecule’s two intramolecular hydrogen bonds, and this can be understood by considering the nodal pattern of the wave function. The functional groups participating in the excited-state intramolecular proton transfer also play important roles in the singlet oxygen quenching. Aloesaponarin I has a quenching rate constant larger than that of vitamin E and has a long duration of action due to its resistance to UV degradation and chemical attacks by singlet oxygen and free radicals.
1. Introduction The cosmetic and healing properties of aloe have been valued for thousands of years.1-6 Legend has it that aloe was one of Cleopatra’s beauty secrets, and Aristotle is said to have persuaded Alexander the Great to capture the island of Socotra in the Indian Ocean simply to have a source of the aloe needed to heal his wounded soldiers. Aloe species are used all over the world to treat conditions ranging from dermatitis to HIVAIDS, and aloe gel is an active ingredient in hundreds of skin lotions, sun blocks, and cosmetics. One of the chemicals found in aloe is aloesaponarin I (methyl 3,8-dihydroxy-1-methyl-9,10-dioxo-9,10-dihydroanthracene-2carboxylate), which has two intramolecular hydrogen bonds.7-9 Intramolecularly hydrogen-bonded molecules are often UV protective because they prevent photosensitization by absorbing harmful UV radiation and dissipating its energy as heat through excited-state intramolecular proton transfer (ESIPT)10-13 along the intramolecular hydrogen bonds. ESIPT is a very interesting topic because it is a very simple chemical process readily accessible for both accurate measurement and quantitative theoretical analysis. Two of us (S.N. and U.N.) have constructed a simple theoretical model based on the nodal pattern of the wave function in ESIPT (refs 14-16 and Supporting Information), and this nodal-plane model is applicable to various photochemical reactions in various excited states of various molecules.17-21 The initial reaction caused by photoexcitation of spiropyran (spirooxazine) vapor, for example, was recently shown experimentally to be the φO-C bond rupture (Form B * Corresponding author. E-mail:
[email protected]. Phone: +81-89-927-9592. Fax: +81-89-927-9590. † Also at Graduate School of Science and Engineering.
Figure 1. Structures of aloesaponarin I (AS1) and related molecules used in the present experimental work.
in Figure 13 and reaction 1 of ref 22), but this had already been predicted by using our nodal-plane model (reactions 19 and 20 of ref 16). Another harmful effect of irradiation is the production of singlet oxygen (1O2,1∆g state),23 which is a reactive oxygen species that plays a key role in a variety of stress conditions, including photodegradation, photoaging, and photocarcinogenesis. 1O2 production accompanies the photoinihibition due to illumination of broad bean leaves with excess photosynthetically active radiation,24 and 1O2 is also generated in UVA-irradiated skin of live mice.25 Because aloesaponarin I has hydroxy groups at the 3- and 8-positions (Figure 1), it is expected to have 1O2 quenching activity in vivo, as do many natural 1O2 quenchers such as vitamin E,23,26 ubiquinol (the reduced form of coenzyme Q10),27,28 and flavonoids29,30 including tea catechins29,31,32 that have hydroxy groups. In the work reported here, we studied the ESIPT-based UV protective and 1O2 quenching functions of aloesaponarin I and related molecules experimentally by means of laser spectroscopy, and we took into consideration the results of ab initio molecular-orbital calculations when evaluating our experimental results. The stable ground-state (S0-state) structures of the molecules studied in the present work, aloesaponarin I (AS1), aloesaponarin I 3-O-methyl ether (methyl 8-hydroxy-3-methoxy1-methyl-9,10-dioxo-9,10-dihydroanthracene-2-carboxylate, AS2),
10.1021/jp075224j CCC: $37.00 © 2007 American Chemical Society Published on Web 10/19/2007
Activity of Aloesaponarin I SCHEME 1: 1O2 Generation for Measurement of Irreversible Chemical Quenching of 1O2
and aloesaponarin I 8-O-methyl ether (methyl 3-hydroxy-8methoxy-1-methyl-9,10-dioxo-9,10-dihydroanthracene-2-carboxylate, AS3), are shown in Figure 1. The hydroxy protons at the 3- and 8-positions of AS2 and AS3 are methylated and thus unable to form intramolecular hydrogen bonds. 2. Experimental Section 2.1. Sample Preparation. The AS1 preparation method and analytical data and the AS3 preparation method were reported in previous papers.8,9 The analytical data of AS3 are as follows: yellow needles, mp 209.0-210.5 °C; 1H NMR (400 MHz, CDCl3, δ) 2.88 (s, 3H, 1-CH3), 4.02 (s, 3H, 8-OCH3), 4.04 (s, 3H, 2-OCH3), 7.31-7.33 (d, J ) 8.1 Hz, 1H, 7-H), 7.62-7.64 (d, J ) 8.8 Hz, 1H, 6-H), 7.66 (s, 1H, 4-H), 7.817.83 (d, J ) 7.6 Hz, 1H, 5-H), 10.5 (s, 1H, OH); EIMS (70 eV) m/z M+ 326, 311, 276; HRMS-EI (m/z) M+ calculated for C18H14O6, 326.0790; found, 326.0789. AS2 was prepared by dissolving 0.44 mmol of AS1 in a 1:1 mixture of dry toluene and dry methanol (29 mL, 1 L ) 1 dm3) under nitrogen, adding 5.6 mL of 10% (trimethylsilyl)diazomethane in hexane, and stirring the mixture for 2 h. Evaporation of the solvent left yellow powdery crystals that we purified by recrystallization. The yield was 88%: mp 209.5-211.0 °C; 1H NMR (400 MHz, CDCl , δ) 2.75 (s, 3H, 1-CH ), 3.98 (s, 3 3 3H, 3-OCH3), 4.02 (s, 3H, 2-OCH3), 7.30-7.32 (dd, J ) 1.2 and 9.5 Hz, 1H, 7-H), 7.61-7.65 (t, J ) 7.8 Hz, 1H, 6-H), 7.77 (s, 1H, 4-H), 7.79-7.80 (dd, J ) 1.2 and 9.5 Hz, 1H, 5-H), 12.8 (s, 1H, OH); EIMS (70 eV) m/z M+ 326, 268, 254, 236, 152, 104, 77, 64; HRMS-EI (m/z) M+ calculated for C18H14O6, 326.0790; found, 326.0789. 3-(1,4-Epidioxy-4-methyl-1,4-dihydro-1-naphthyl)propionic acid (EP, Scheme 1) and 2,6-di-t-butyl-4-(4-methoxyphenyl)phenoxyl (ArO•, Scheme 2) were prepared as reported in previous papers.33,34 Commercial rose bengal (Tokyo Chemical Industry) was used as received. The commercial cyclohexane (Nacalai Tesque) used for spectral measurements of AS1, AS2, and AS3 had been specially prepared for luminescence and was used without further purification. The ethanol (Nacalai Tesque) was dried and purified by distillation. 2.2. Measurements. Unless otherwise noted, the measurements were performed at room temperature. In the spectroscopic SCHEME 2: Reaction between AS1 and ArO•
J. Phys. Chem. B, Vol. 111, No. 45, 2007 13117 measurements, the sample concentrations were about 10-510-4 M (1 M ) 1 mol/dm3). Before all measurements except the measurements of the 1O2 quenching, the sample solutions were bubbled with nitrogen gas. The absorption and fluorescence spectra were measured with a 1-cm2 quartz cell. The absorption spectra were measured with a Shimadzu UV mini-1240 or UV-2100S spectrophotometer, and the fluorescence spectra were measured with a Shimadzu RF-5000 spectrofluorophotometer. The fluorescence-spectrum signals were transferred to a PC and analyzed by using curvefitting.35 The fluorescence quantum yields (φf’s) were determined by comparing the fluorescence-emission spectra of the samples with that of 9,10-diphenylanthracene in cyclohexane (φf ) 0.90)36 after they had been corrected for the spectral sensitivity of the detector. The fluorescence and phosphorescence decay rate constants (lifetimes) were measured with a temporal resolution of ≈1 ns by a Hamamatsu C7990-01 near-infrared fluorescence-lifetime measurement system in which the second or third harmonic of a Nd:YAG laser (Hamamatsu C8597-02, 532 or 355 nm) was used as the excitation light. The transient absorption spectra were obtained with a Unisoku USP-500 pulse-flash spectrophotometer or a Unisoku TSP-1000 time-resolved spectrophotometer in which the excitation light was the third harmonics of a Nd:YAG laser (Continuum Surelite I). A rectangular quartz cell (2 cm in light-path length) or a 1-cm2 quartz cell was used in the measurements. If necessary, the sample solutions were bubbled with nitrogen gas during the measurements to remove dissolved oxygen. The second-order rate constants of the 1O2 quenching by AS1, AS2, and AS3 were determined by measuring the phosphorescence decay rate constants of 1O2 in air-saturated ethanol solutions including rose bengal (sensitizer S) and the quencher Q. O2(1∆g) emitting the phosphorescence was generated by energy transfer (photosensitization) to O2(3Σg-) from the lowest excited triplet state of the sensitizer (3S*), which had been formed by absorption of laser light at 532 nm and the subsequent intersystem crossing (reaction 1): intersystem crossing
S + hν (laser light) f 1S* 98 3S* S* + 3O2 f S + 1O2
3 k0
O2 98 3O2 + hν (phosphorescence-emission at ≈
1
1270 nm)23 (1) where 1S* and k0 respectively denote the lowest-excited singlet state of the sensitizer and the natural decay rate constant of 1O2. The concentration of the sensitizer was about 2 × 10-4 M. When a quencher is added to the solution, 1O2 is quenched as follows
13118 J. Phys. Chem. B, Vol. 111, No. 45, 2007 1
kQ
O2 + Q 98 3O2 + quenching product
Nagaoka et al.
(2)
where kQ is the second-order rate constant for reaction 2. Under our experimental conditions, the concentration of the quencher was much larger than that of 1O2. According to Merkel and Kearns,37 the rate constant k for the phosphorescence decay due to 1O2 quenching is then given by
k ) k0 + kQ[Q]
(3)
where [Q] is the concentration of the quencher. We can determine the kQ value from the slope of the k versus [Q] plot. It should be noted that neither AS1, AS2, nor AS3 was efficiently brought to an excited state by laser irradiation at 532 nm. The yield of the irreversible chemical reaction between 1O2 and AS1 (Scheme 3) was estimated from the bleaching of AS130,32 by generating 1O2 at 35 °C through thermal decomposition of EP (Scheme 1).33 The susceptibility of AS1 to free radicals was examined by using the reaction with ArO•, which can be regarded as a model for active oxygen radicals in biological systems (Scheme 2).38 The second-order reaction rate constant ks in ethanol was determined with a Unisoku RSP1000-03F stopped-flow spectrophotometer as reported in a previous paper.38
Figure 2. Absorption (Abs.) and fluorescence-emission (Fluor.) spectra of (a) AS1, (b) AS2, and (c) AS3 in cyclohexane. denotes the absorption coefficient. The fluorescence-emission spectra of AS1 and AS2 were obtained by photoexcitation at 380 nm and have not been corrected for the spectral sensitivity of the detector.
SCHEME 3: Irreversible Chemical Reaction between AS1 and 1O2
3. Computational Method and Procedure The experimental results on ESIPT in AS1 were interpreted in light of ab initio molecular-orbital calculations39 of energy levels of models for the normal and intramolecularly protontransferred forms. These calculations were done at the CIS/ 6-31+G**//HF/6-31+G** level using the Gaussian 03 program.40 The experimental results on the 1O2 quenching action were interpreted in consideration of the stabilization energies of various encounter complexes of 1O2 and the AS1 normalspecies model. These energies were calculated at the MP2/631G**//HF/6-31G** level. As shown in ref 23, such computational results are often useful in the interpretation of the experimental results on the 1O2 quenching action. 4. Results and Discussion 4.1. ESIPT. The absorption and fluorescence-emission spectra of AS1, AS2, and AS3 in cyclohexane are shown in Figure 2, and the corresponding φf’s and fluorescence lifetimes (τf’s) are listed in Table 1. The absorption spectra of AS1 and AS2 agree well with their fluorescence-excitation spectra. The hydroxy group at the 8-position in AS2 is intramolecularly hydrogen-bonded to the carbonyl oxygen at the 9-position, but the hydroxy proton at the 3-position is methylated and cannot hydrogen-bond to the adjacent carbonyl oxygen (Figure 1). AS2 shows fluorescence-emission around 600 nm with an unusually large red-shift (Stokes-shift) from the absorption spectrum (Figure 2). As in many previous reports,10-13 this observation can be explained in terms of ESIPT (Figure 3): photoexcitation of the normal form (stable S0-state species) produces the lowestexcited 1(π,π*) state (S1 state), in which ESIPT takes place rapidly, stabilizing the S1 state. The proton-transferred form in the S1 state emits fluorescence and is left in the S0 state. Then, the reverse proton transfer takes place, and the stable normal form is regenerated. Because a significant amount of absorption photon energy is dissipated as heat in this cycle, the protontransferred form fluoresces at a lower energy with an unusually large Stokes shift. The Stokes-shifted fluorescence-emission of
TABLE 1: Of’s, τf’s, and kQ’s of AS1, AS2, and AS3 τf/ps
φf molecule cyclohexane AS1 2.7 × 10-3 AS2 2.0 × 10-3 AS3
