A Fluorogenic Reagent, 4-Mercapto-7-methylthio-2,1,3

Anal. Chem. 2001, 73, 2165-2170

A Fluorogenic Reagent, 4-Mercapto-7-methylthio-2,1,3-benzoxadiazole for Carboxylic Acids, Designed by Prediction of the Fluorescence Intensity Seiichi Uchiyama, Tomofumi Santa, and Kazuhiro Imai*

Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan

During the course of our studies of the development of fluorogenic reagents having a 4,7-disubstituted benzofurazan structure, we previously proposed 7-acetylamino4-mercapto-2,1,3-benzoxadiazole (AABD-SH) as a fluorogenic reagent for carboxylic acids. Since then, progress has made it possible to estimate the fluorescence quantum yields of the 4,7-disubstituted benzofurazan compounds on the basis of the PM3 calculation of their S1T2 energies. Subsequently, a new fluorogenic reagent, 4-mercapto-7-methylthio-2,1,3-benzoxadiazole (MTBDSH) was designed and synthesized. In the presence of condensation reagents, triphenylphosphine (TPP) and 2,2′-dipyridyl disulfide (DPDS), MTBD-SH readily reacted with n-caprylic acid within 1 min at room temperature. The derivatives of five carboxylic acids (n-caprylic acid, n-capric acid, lauric acid, myristic acid, and palmitic acid) were well-separated on a reversed-phase column and were fluorimetrically detected at 519 nm with excitation at 391 nm. The detection limits (S/N ) 3) were 2.4-5.0 fmol. Thus, MTBD-SH had properties that were considered to be superior. For carboxylic acids, it was superior not only to AABD-SH, but also to many other conventional reagents. The superiority was examined in terms of its reactivity and sensitivity and the avoidance of interfering peaks that were derived from the reagent itself or degradation products in the chromatogram. Quantifying biologically active carboxylic acids such as fatty acids, bile acids, lactic acid, and prostaglandins is important in the bioanalytical, biological, and biomedical sciences. Because they exist in small quantities in bio-samples, the quantification of these acids requires a sensitive and selective detection method. Thus, various fluorescent derivatization reagents for high-performance liquid chromatography (HPLC) have been developed for the sensitive and selective determination of such carboxylic acids as the following: (I) bromomethanes,1-10 such as 4-bromomethyl7-methoxycoumarin (Br-Mmc)1; (II) diazomethanes,11-13 such as * Fax: 81-3-5841-4885. E-mail: [email protected]. (1) Dnges, W. Anal. Chem. 1977, 49, 442-445. (2) Tsuchiya, H.; Hayashi, T.; Naruse, H. Takagi, N. J. Chromatogr. 1982, 234, 121-130. (3) Farinotti, R.; Siard, P.; Bourson, J.; Kirkiacharian, S.; Valeur, B.; Mahuzier, G. J. Chromatogr. 1983, 269, 81-90. 10.1021/ac001232j CCC: $20.00 Published on Web 04/06/2001

© 2001 American Chemical Society

9-anthryldiazomethane (ADAM);11,12 (III) amines,14-26 such as 9-aminophenanthrene (9-AP);14 (IV) hydrazines,27-36 such as 6,7(4) Yamaguchi, M.; Hara, S.; Matsunaga, R.; Nakamura, M.; Ohkura Y. Anal. Sci. 1985, 1, 295-296. (5) Takadate, A.; Masuda, T.; Tajima, C.; Murata, C.; Irikura, M.; Goya, S. Anal. Sci. 1992, 8, 663-668. (6) Gatti, R.; Cavrini, V.; Roveri, P. Chromatographia 1992, 33, 13-18. (7) Takadate, A.; Masuda, T.; Murata, C.; Haratake, C.; Isobe, A.; Irikura, M.; Goya, S. Anal. Sci. 1992, 8, 695-697. (8) Takadate, A.; Masuda, T.; Murata, C.; Tanaka, T.; Miyahara, H.; Goya, S. Chem. Lett. 1993, 811-814. (9) Takechi, H.; Kamada, S.; Machida, M. Chem. Pharm. Bull. 1996, 44, 793799. (10) Katoh, A.; Fujimoto, T.; Takahashi, M.; Ohkanda, J. Heterocycles 1999, 50, 299-308. (11) Nimura, N.; Kinoshita, T. Anal. Lett. 1980, 13, 191-202. (12) Barker, S. A.; Monti, J. A.; Christian, S. T.; Benington, F.; Morin, R. D. Anal. Biochem. 1980, 107, 116-123. (13) Nimura, N.; Kinoshita, T.; Yoshida, T.; Uetake, A.; Nakai, C. Anal. Chem. 1988, 60, 2067-2070. (14) Ikeda, M.; Shimada, K.; Sakaguchi, T.; Matsumoto, U. J. Chromatogr. 1984, 305, 261-270. (15) Yanagisama, I.; Yamane, M.; Urayama, T. J. Chromatogr. 1985, 345, 229240. (16) Lee, Y. M.; Nakamura, H.; Nakajima, T. Anal. Sci. 1989, 5, 681-685. (17) Nakajima, R.; Shimada, K.; Fujii, Y.; Yamamoto, A.; Hara, T. Bull. Chem. Soc. Jpn. 1991, 64, 3173-3175. (18) Tod, M.; Prevot, M.; Chalom, J.; Farinotti, R.; Mahuzier, G. J. Chromatogr. 1991, 542, 295-306. (19) Toyo’oka, T.; Ishibashi, M.; Takeda, Y.; Nakashima, K.; Akiyama, S.; Uzu, S.; Imai, K. J. Chromatogr. 1991, 588, 61-71. (20) Toyo’oka, T.; Ishibashi, M.; Takeda, Y.; Imai, K. Analyst 1991, 116, 609613. (21) Toyo’oka, T.; Ishibashi, M.; Terao, T. Analyst 1992, 117, 727-733. (22) Kondo, J.; Imaoka, T.; Kawasaki, T.; Nakanishi, A.; Kawahara, Y. J. Chromatogr. 1993, 645, 75-81. (23) Kondo, J.; Suzuki, N.; Imaoka, T.; Kawasaki, T.; Nakanishi, A.; Kawahara, Y. Anal. Sci. 1994, 10, 17-23. (24) Matsumoto, K.; Ichitani, Y.; Ogasawara, N.; Yuki, H.; Imai, K. J. Chromatogr. A 1994, 678, 241-247. (25) Prados, P.; Fukushima, T.; Santa, T.; Homma, H.; Tsunoda, M.; Al-Kindy, S.; Mori, S.; Yokosu, H.; Imai, K. Anal. Chim. Acta 1997, 344, 227-232. (26) Inoue, H.; Ikeno, M.; Ishii, Y.; Tsuruta, Y. J. Chromatogr. A 1998, 816, 137-143. (27) Yamaguchi, M.; Iwata, T.; Inoue, K.; Hara, S.; Nakamura, M. Analyst 1990, 115, 1363-1366. (28) Iwata, T.; Nakamura, M.; Yamaguchi, M. Anal. Sci. 1992, 8, 889-892. (29) Nakashima, K.; Taguchi, Y.; Kuroda, N.; Akiyama, S.; Duan, G. J. Chromatogr. 1993, 619, 1-8. (30) Iwata, T.; Hirose, T.; Nakamura, M.; Yamaguchi, M. Anayst 1994, 119, 1747-1751. (31) Saito, M.; Ushijima, T.; Sasamoto, K.; Ohkura, Y.; Ueno, K. J. Chromatogr. B 1995, 674, 167-175. (32) Saito, M.; Ushijima, T.; Sasamoto, K.; Yakata, K.; Ohkura, Y.; Ueno, K. Anal. Chim. Acta 1995, 300, 243-251.

Analytical Chemistry, Vol. 73, No. 10, May 15, 2001 2165

dimethoxy-1-methyl-2(1H)-quinoxalinone-3-propionylcarboxylic acid hydrazide;27 and (V) trifluoromethanesulfonates,37-39 such as 2-(2,3naphthalimino)ethyl trifluoromethanesulfonate (NE-OTf).37 However, these reagents were not always satisfactory with respect to the detectability of carboxylic acids. This was because these reagents fluoresced themselves and were thus called “fluorescent labeling reagents”. The fluorescence of the reagent and the degradation byproducts generated in their reaction can interfere with the quantification of small amounts of the analytes. Furthermore, the excitation and emission wavelengths of their derivatives could be so short that the quantification was often interfered with by the biomatrixes containing the fluorescence compounds. Therefore, the development of a reagent for carboxylic acids with the following characteristics was desired. It should be (I) fluorogenic (reagent itself is not fluorescent), (II) reactive under mild conditions, and (III) its derivatives should have long excitation and emission wavelengths. With these requirements in mind, we have investigated the effects of the substituent groups at the 4- and 7-positions of the benzofurazan (2,1,3-benzoxadiazole) compounds on their fluorescence intensities. From this study, we evaluated the relationship between the fluorescence intensities of these compounds and the Hammett substituent constants (σp)40 of the substituent groups (see Experimental Section below) and developed a method for predicting the approximate fluorescence intensities from their chemical structures.41 On the basis of this method, we have developed 7-acetylamino-4-mercapto-2,1,3-benzoxadiazole (AABDSH),42 the fluorogenic reagent for carboxylic acids, as well as 7-phenylsulfonyl-4-(2,1,3-benzoxadiazolyl) isocyanate (PSBDNCO)43 for alcohols and 7-methylthio-4-(2,1,3-benzoxadiazolyl) isothiocyanate (MTBD-NCS)44 for amino acid sequencing analysis. The detection limits of the derivatives of the carboxylic acids with AABD-SH were 10-20 fmol (S/N ) 3). At the same time, we also investigated, on a theoretical basis, the relationship between the fluorescence intensities of the benzofurazan compounds and their chemical structures. The semiempirical PM345 calculation clarified that the benzofurazan compounds with the smaller S1-T2 energies had larger fluorescence quantum yields,46,47 which suggested that a fluorogenic (33) Saito, M.; Ushijima, T.; Sasamoto, K.; Ohkura, Y.; Ueno, K. Anal. Sci. 1995, 11, 103-107. (34) Santa, T.; Kimoto, K.; Fukushima, T.; Homma, H.; Imai, K. Biomed. Chromatogr. 1996, 10, 183-185. (35) Santa, T.; Takeda, A.; Uchiyama, S.; Fukushima, T.; Homma, H.; Suzuki, S.; Yokosu, H.; Lim, C. K.; Imai, K. J. Pharm. Biomed. Anal. 1998, 17, 10651070. (36) Hara, S.; Inoue, K.; Fukuzawa, M.; Ono, N.; Kuroda, T.; Yamaguchi, M. Anal. Sci. 1998, 14, 1173-1176. (37) Yasaka, Y.; Tanaka, M.; Shono, T.; Tetsumi, T.; Katakawa, J. J. Chromatogr. 1990, 508, 133-140. (38) Akasaka, K.; Ohrui, H.; Megro, H. Analyst 1993, 118, 765-768. (39) Yasaka, Y.; Ono, Y.; Tanaka, M. J. Chromatogr. A 1998, 810, 221-225. (40) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165-195. (41) Uchiyama, S.; Santa, T.; Fukushima, T.; Homma, H.; Imai, K. J. Chem. Soc., Perkin Trans. 2 1998, 2165-2173. (42) Santa, T.; Okamoto, T.; Uchiyama, S.; Mitsuhashi, K.; Imai, K. Analyst 1999, 124, 1689-1693. (43) Uchiyama, S.; Santa, T.; Suzuki, S.; Yokosu, H.; Imai, K. Anal. Chem. 1999, 71, 5367-5371. (44) Toriba, A.; Adzuma, K.; Santa, T.; Imai, K. Anal. Chem. 2000, 72, 732739. (45) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209-220. (46) Uchiyama, S.; Santa, T.; Imai, K. J. Chem. Soc., Perkin Trans. 2 1999, 25252532.

2166

Analytical Chemistry, Vol. 73, No. 10, May 15, 2001

Figure 1. Design of fluorogenic reagents for carboxylic acids using the relationship between the fluorescence intensities of the 4,7disubstituted benzofurazan compounds and the Hammett substituent constants (σp) at the 4- and 7-positions.

benzofurazan reagent more sensitive than AABD-SH could be designed for carboxylic acids, because the quantum yields are related, though not identical, to the fluorescence intensities. Therefore, in this article, we report the results of such an attempt to design and synthesize a more ideal reagent. EXPERIMENTAL SECTION Method for Using the Relationship between the Fluorescence Intensities of 4,7-Disubstituted Benzofurazan Compounds and the σp Values at the 4- and 7-Positions. As described in a previous paper,41 we found a relationship between the fluorescence intensities and the σp values at the 4- and 7-positions of the 4,7-disubstituted benzofurazan compounds (Figure 1). In this figure, the abscissa and the ordinate were the sum of the σp values at the 4- and 7-positions

x ) σp(4) + σp(7)

(1)

and the difference in the σp values between the 4- and 7-positions,

y ) |σp(4) - σp(7)|

(2)

respectively. Seventy 4,7-disubstituted benzofurazan compounds were plotted in the figure and classified into three groups according to their relative fluorescence intensity [RFI; fluorescence intensity of 4-amino-7-N,N-dimethylaminosulfonyl-2,1,3-benzoxadiazole (DBD-NH2) was arbitrarily taken as 1.0. RFI ) 0-1, having no or weak fluorescence (O); RFI ) 1-5, having moderate fluorescence (2); RFI > 5, having strong fluorescence (9)]. It was interesting to note that the fluorescent compounds, represented as closed squares and triangles, were concentrated in two areas (A and B), but the nonfluorescent compounds were scattered outside of these two areas. (47) Uchiyama, S.; Santa, T.; Okiyama, N.; Azuma, K.; Imai, K. J. Chem. Soc., Perkin Trans. 2 2000, 1199-1207.

Figure 1 demonstrated the procedure in selecting compounds that were fluorescent or nonfluorescent. The 4,7-disubstituted benzofurazan compounds having a certain substituent group (e.g., R1) at the 4-position were located on the line represented by the equation

y ) |x - 2σp(4)|

(3)

This equation was obtained by the elimination of σp(7) from eqs 1 and 2. Similarly, the compounds having a certain substituent group (e.g., R2) at the 7-position were located on the line

y ) |x - 2σp(7)|

(4)

The point of intersection (σp(4) + σp(7), |σp(4) - σp(7)|) of the two equations (eqs 3 and 4) corresponds to the 4,7-disubstituted benzofurazan compound having the substituent groups R1 and R2. The compounds located in area A or B should be fluorescent, whereas the compounds outside areas A and B should be nonfluorescent. Computational Methods. The method described in our previous report47 was adopted for all of the PM3 calculations that were used in the present study, with C.I. ) 6 in this study. Materials. Water was purified using a Milli-Q reagent system (Millipore, Bedford, MA). Acetonitrile was of HPLC grade (Kanto Chemicals, Tokyo, Japan). All other reagents were of guaranteed reagent grade and used without further purification. Apparatus. Melting points were measured on a Yanagimoto Micro Point Apparatus (Tokyo, Japan) and are uncorrected. Proton nuclear magnetic resonance (1H NMR) spectra were obtained using a JEOL LA-500 spectrometer (Tokyo, Japan) and tetramethylsilane as the internal standard in CDCl3. Mass spectra were measured using a Hitachi M-1200H mass spectrometer (atmospheric pressure chemical ionization (APCI) system) (Tokyo, Japan). UV-vis absorption spectra (30 µM) were measured using a JASCO Ubest-50 spectrometer (Tokyo, Japan). Fluorescence spectra (MTBD-SH, 30 µM; MTBD-SAc, AABD-SAc, and MTBDSCOC7H15, 1 µM) were measured using a Hitachi F-4010 fluorescence spectrometer (Tokyo, Japan). The fluorescence quantum yields (Φ) were determined using quinine sulfate in 0.1 M sulfuric acid (Φ ) 0.55; λex., 355 nm) as the standard. High-Performance Liquid Chromatography. The highperformance liquid chromatograph consisted of a Hitachi L-6300 pump, a Hitachi L-1080 fluorescence detector, and a Hitachi D-2500 integrator. The separation of the derivatives was studied using an analytical column, TSKgel ODS-80Ts (150 × 4.6 mm; i.d., 5 µm) (TOSOH, Tokyo, Japan) and linear gradient elution from 0 to 35 min (acetonitrile, 60 to 100% in water, v/v), and isocratic elution from 35 min (acetonitrile, 100%) at a flow rate of 1.0 mL/ min. The eluate was monitored by fluorescence detection (λex., 391 nm; λem., 519 nm). Synthesis of 4-Mercapto-7-methylthio-2,1,3-benzoxadiazole (MTBD-SH). 4-Methylthio-2,1,3-benzoxadiazole47 (670 mg) was dissolved in CH2Cl2 (20 mL). After the addition of chlorosulfuric acid (4 mL) in CH2Cl2 (20 mL) at 0 °C, the mixture was stirred for 10 min. The solution was poured into ice water (100 g) and extracted using AcOEt (200 mL). The organic layer was dried over anhydrous Na2SO4 and concentrated in vacuo. The

residue was chromatographed on silica gel using AcOEt-nhexane (1:2) to afford 7-chlorosulfonyl-4-methylthio-2,1,3-benzoxadiazole (355 mg, 33%) as a yellow powder: mp, 115-116 °C; δH 8.08 (1H, d, J ) 7.65), 7.07 (1H, d, J ) 7.65), 2.75 (3H, s). Anal. Calcd for C7H5ClN2O3S2: C, 31.76; H, 1.90; N, 10.58. Found: C, 31.71; H, 2.16; N, 10.51. 7-Chlorosulfonyl-4-methylthio-2,1,3-benzoxadiazole (355 mg) was dissolved in CH2Cl2 (8 mL). After the addition of acetic acid (12 mL), concentrated HCl (4 mL), and SnCl2‚2H2O (2 g), the mixture was stirred at room temperature for 30 min. The solution was poured into ice water (100 g) and extracted using AcOEt (200 mL). The organic layer was dried over anhydrous Na2SO4 and concentrated in vacuo. The residue was chromatographed on silica gel using CH2Cl2-n-hexane (1:1) to afford 4-mercapto-7-methylthio-2,1,3-benzoxadiazole (61 mg, 23%) as a yellow powder: mp, 161-162 °C; δH 7.15 (1H, d, J ) 7.30), 6.94 (1H, d, J ) 7.30), 4.13 (1H, s), 2.61 (3H, s). APCI-MS: m/z 197 [M - H]-. Anal. Calcd for C7H6N2OS2: C, 42.40; H, 3.05; N, 14.13. Found: C, 42.70; H, 2.76; N, 14.33. Synthesis of 4-Acetylthio-7-methylthio-2,1,3-benzoxadiazole (MTBD-SAc). 4-Mercapto-7-methylthio-2,1,3-benzoxadiazole (8.0 mg) was dissolved in CH2Cl2 (2 mL). After the addition of acetic anhydride (0.5 mL), the mixture was stirred at room temperature for 8 h. The reaction mixture was evaporated to dryness under reduced pressure, and the residue was chromatographed on silica gel using CH2Cl2-n-hexane (1:2) to afford 4-acetylthio-7-methylthio-2,1,3-benzoxadiazole (6.5 mg, 67%) as a yellow powder: mp, 92-93 °C. δH 7.41 (1H, d, J ) 7.30), 7.01 (1H, d, J ) 7.30), 2.63 (3H, s), 2.49 (3H, s). APCI-MS: m/z 241 [M + H]+. Anal. Calcd for C9H8N2O2S2: C, 44.98; H, 3.36; N, 11.66. Found: C, 45.10; H, 3.62; N, 11.53. Synthesis of 4-(n-Caprylylthio)-7-methylthio-2,1,3-benzoxadiazole (MTBD-SCOC7H15). 4-Mercapto-7-methylthio-2,1,3benzoxadiazole (20 mg) was dissolved in n-caprylic acid (0.5 mL) containing triphenylphosphine (TPP, 500 mg). After the addition of 2,2′-dipyridyl disulfide (DPDS, 500 mg), the mixture was stirred at room temperature for 30 min. The reaction mixture was then chromatographed on silica gel using CH2Cl2-n-hexane (1:3) to afford 4-(n-caprylylthio)-7-methylthio-2,1,3-benzoxadiazole (27 mg, 83%) as a yellow oil: δH 7.36 (1H, d, J ) 7.35), 6.96 (1H, d, J ) 7.35), 2.67 (2H, t), 2.60 (3H, s), 1.68 (2H, m), 1.24-1.33 (8H, m), 0.81 (3H, t). APCI-MS: m/z 325 [M + H]+. Time Course of the Reaction of MTBD-SH with n-Caprylic Acid. To a 400-µL vial was added 20 µL of MTBD-SH in CH2Cl2 (20 mM), 20 µL of n-caprylic acid in MeCN (100 µM), 20 µL TPP in MeCN (20 mM), and 20 µL DPDS in MeCN (20 mM). At specific intervals, an aliquot (5 µL) of the reaction mixture was subjected to HPLC. The reaction yield was determined by comparison to the peak height of authentic MTBD-SCOC7H15. Derivatization of Carboxylic Acids with MTBD-SH. To a 400-µL vial was added 20 µL of MTBD-SH in CH2Cl2 (20 mM), 20 µL of mixed carboxylic acids (80 µM each of n-caprylic, n-capric, lauric, myristic, and palmitic acids) in MeCN, 20 µL TPP in MeCN (20 mM), and 20 µL DPDS in MeCN (20 mM). The vial was capped and allowed to stand at room temperature for 3 min. Calibration Curve for the Derivatives of MTBD-SH with Carboxylic Acids. To a 400-µL vial was added 20 µL of MTBDSH in CH2Cl2 (20 mM), 20 µL of mixed carboxylic acids (0.078, Analytical Chemistry, Vol. 73, No. 10, May 15, 2001

2167

Figure 2. Synthesis of MTBD-SH and its derivatizing reaction with carboxylic acids.

0.3125, 1.25, 5, 20, or 80 µM each of n-caprylic, n-capric, lauric, myristic, and palmitic acids) in MeCN, 20 µL TPP in MeCN (20 mM), and 20 µL DPDS in MeCN (20 mM). The vial was capped and allowed to stand at room temperature for 3 min. An aliquot (2 µL) of reaction mixture was then subjected to HPLC. RESULTS AND DISCUSSION Design and Synthesis of the Fluorogenic Derivatization Reagent for Carboxylic Acids. The new fluorogenic reagents for carboxylic acids were designed, in the first stage, using the relationship shown in Figure 1. First, thiol (SH; σp ) 0.15) was selected as the reacting group at the 4-position of the objective reagents among the SH, NH2, and OH groups, because the sulfur atom is generally the most reactive to electrophilic compounds. The derivatives were tentatively considered as acetylated ones (its substituent was SAc). In Figure 1, the compounds having an SH group at the 4-position should be located on the equation, y ) |x - 0.30| [line 1 in Figure 1; only y ) - x + 0.30 (x < 0.30) was drawn because y ) x - 0.30 (x > 0.30) was unnecessary for this discussion]. The 4,7-disubstituted benzofurazan compounds having an SAc group (σp ) 0.44) at the 4-position are to be located on the equation, y ) |x - 0.88| [line 2 in Figure 1; only y ) - x + 0.88 (x < 0.88) was drawn]. These data suggested that the compounds on line 1 were derivatized and converted to the compounds on line 2. Next, the selection of the substituent groups at the 7-position of the new fluorogenic reagents was made. Namely, we searched the adequate σp parameters for the equations, y ) x - 2σp(7) [x > 2σp(7)], with which the points of intersection of line 1 were out of the fluorescent areas (A and B) and the points of intersection of line 2 were in the fluorescent area. Because the equations in the range of 3 (from y ) x + 0.54 to y ) x - 0.06, which represents the B area borders) were suited for the above condition, the substituent groups with σp values from -0.27 to 0.03 were appropriate at the 7-position. Among the substituent groups having σp values in these ranges, OMe (σp ) -0.27), Me (σp ) -0.17), NHAc (σp ) 0.00) and SMe (σp ) 0.00) were selected. Because 7-acetylamino-4mercapto-2,1,3-benzoxadiazole (AABD-SH) was already reported,42 4-mercapto-7-methoxy-2,1,3-benzoxadiazole (MOBD-SH), 4-mercapto-7-methyl-2,1,3-benzoxadiazole (MBD-SH), and 4-mercapto7-methylthio-2,1,3-benzoxadiazole (MTBD-SH) were considered as candidates for the new fluorogenic reagent of carboxylic acids. During the second stage, to obtain the most sensitive reagent, the fluorescence intensities of the acetyl derivertives of the above four reagents were estimated using the PM3 calculations.46,47 The calculation revealed that the S1-T2 energy of the derivative of MTBD-SH (-0.969 eV) was the smallest among those of AABDSH (-0.557 eV), MOBD-SH (-0.399 eV) and MBD-SH (-0.106 eV). Because it was shown that the benzofurazan compound 2168


having a smaller S1-T2 energy fluoresced more strongly,46,47 MTBD-SH (4-mercapto-7-methylthio-2,1,3-benzoxadiazole) should be the most sensitive fluorogenic reagent for carboxylic acids, and AABD-SH should be the next most sensitive one. With this prediction, MTBD-SH, its acetyl derivative (MTBDSAc), and its derivative of n-caprylic acid (MTBD-SCOC7H15) were synthesized, and the fluorescence intensity of MTBD-SAc was compared to that of AABD-SAc (the acetyl derivative of AABDSH). MTBD-SH was synthesized by the reduction of 4-methylthio7-chlorosulfonyl-2,1,3-benzoxadiazole with SnSl2/HCl/CH3COOH48 as a yellow powder (Figure 2). MTBD-SAc was synthesized by the acetylation of MTBD-SH using acetic anhydride. MTBDSCOC7H15 was also synthesized from MTBD-SH and n-caprylic acid using triphenylphosphine (TPP) and 2,2′-dipyridyl disulfide (DPDS). Their chemical structures were identified by NMR, elemental analysis, and mass spectrometry. Fluorescence Characteristics of MTBD-SH and Derivatives of MTBD-SH with Carboxylic Acids. The absorption and fluorescence spectra of MTBD-SH, the acetyl derivative, MTBDSAc, and the derivative of n-caprylic acid, MTBD-SCOC7H15 were measured and are summarized in Table 1. As expected, MTBDSH did not fluoresce, but MTBD-SAc and MTBD-SCOC7H15 strongly fluoresced. Furthermore, the fluorescence intensity of MTBD-SAc was much higher than that of AABD-SAc (e.g., 7.4fold in methanol) as depicted in Table 1. These results indicated that MTBD-SH, as predicted, was a more sensitive fluorogenic reagent than AABD-SH. The excitation and emission wavelengths of MTBD-SAc (λex/ λem ) 387-394 nm/498-550 nm) were longer than those of the derivatives of the carboxylic acids with the conventional fluorescent labeling reagents,1-18,22,23,26-33,36-39 and fluorogenic AABDSH42 except for other labeling reagents having the benzofurazan skeleton for carboxylic acids19-21,24,25,34,35 (λex/λem ) 440-470 nm/ 530-580 nm), and the Stokes shift of MTBD-SAc (108-156 nm) was also greater than those of the conventional fluorescent labeling reagents, except for (S)-(+)-1-methyl-2-(6,7-dimethoxy2,3-naphthalimido)ethyl trifluoromethanesulfonate [(S)-(+)MDNEOtf] (184 nm),39 monodansyl cadaverine (MDC) (178 nm),16 2-bromoacetyl-6-methoxynaphthalene (Br-AMN) (160 nm),6 and 2-(2,3-anthracenedicarboximido)ethyl trifluoromethanesulfonate (AE-Otf) (158 nm).38 These results indicate that MTBD-SH might be prominent with respect to the avoidance of interference from biomatrixes. Derivatization of n-Caprylic Acid with MTBD-SH. A time course study on the derivatizing reaction of n-caprylic acid (25 µM) with MTBD-SH (5 mM) was performed at room temperature in the presence of the condensation agents [Mukaiyama A,49 (48) Marvel, C. S.; Caesar, P. D. J. Am. Chem. Soc. 1951, 73, 1097-1099. (49) Mukaiyama, T.; Matsueda, R.; Suzuki, M. Tetrahydron Lett. 1970, 19011904.

Table 1. Fluorescence Characteristics of MTBD-SH (reagent), MTBD-SAc, and MTBD-SCOC7H15 (Derivatives) compounds

λab (nm)

(103)

λex (nm)a

Φ

RFIb

benzene

MTBD-SH MTBD-SAc MTBD-SCOC7H15

411 390 393

7.58 10.07 9.61

519 498 497

0.0017 0.31 0.29

0.36 90 77

dichloromethane

MTBD-SH MTBD-SAc (cf. AABD-SAc42) MTBD-SCOC7H15

410 389 369 392

9.04 10.54 8.11 9.99

528 510 507 510

0.0018 0.52 0.20 0.34

0.45 160 41 94

acetonitrile


411 391 391

8.00 9.51 9.24

526 519 519

0.00050 0.27 0.28

0.10 76 73

ethanol


406 390 390

8.17 9.93 9.68

514 525 524

0.00030 0.14 0.14

0.054 38 36

methanol


409 387 367 390

8.10 10.40 7.96 9.06

515 529 522 529

0.00026 0.095 0.020 0.089

0.045 28.0 3.80 22.0

water


409 394 366 398

5.81 8.75 6.75 8.51

530 550 532 517

0.00028 0.0047 0.0033 0.014

0.036 1.0 0.24 3.0

solvents

a

Excited at λab (nm). b RFI ) Relative fluorescence intensity. Fluorescence intensity of DBD-NH2 was arbitrarily taken as 1.0.41

triphenylphosphine (TPP, 5 mM) and 2,2′-dipyridyl disulfide (DPDS, 5 mM)]. As a result, the derivatization reaction proceeded to completion within 1 min. The conventional fluorescent labeling reagents and the fluorogenic AABD-SH for carboxylic acids1-39,42 required a higher temperature (37-100 °C) or longer reaction time (5 min to 3 h) for the derivitization. These data indicated that MTBD-SH should be the most reactive reagent among all of the reagents for carboxylic acids developed so far. It was noteworthy that MTBD-SH did not react with n-caprylic acid in the presence of certain types of condensation agents, such as 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC),50 diethylphosphoryl cyanide (DEPC),51 or phenyl N-phenylphosphoramidchloridate,52 but it did react in the presence of tributylphosphine (TBP)/DPDS as well as TPP/DPDS. These results indicated that the SH group of MTBD-SH was not reactive to carboxylic acids but was activated only by the trialkyl- or triaryl- phosphine/ DPDS. Second, the order of the addition of DPDS was important for the derivatization reaction. Namely, DPDS must be added last to the reaction mixture of n-caprylic acid, MTBD-SH, and TPP; otherwise, the derivatization reaction did not proceed. Although the first paper,49 in which TPP/DPDS was used as the condensation agents, speculated about the occurrence of the activated complex of carboxylic acid/TPP/DPDS, it seemed that another unknown active complex of the carboxylic acid/TPP/MTBD-SH was produced in the present study. Separation of the Derivatives of Carboxylic Acids with MTBD-SH on a Reversed-Phase Column. The chromatogram of the derivatives thus obtained is shown in Figure 3. As expected, the interfering peaks from MTBD-SH itself and the degradation (50) Sheehan, J. C.; Cruickshank, P. A.; Boshart, G. L. J. Org. Chem. 1961, 26, 2525-2528. (51) Yamada, S.; Kasai, Y.; Shioiri, T. Tetrahydron Lett. 1973, 1595-1598. (52) Mestres, R.; Palomo, C. Synthesis 1982, 288-291.

Figure 3. Chromatogram of carboxylic acids derivatized with MTBDSH: (1) 10 pmol n-caprylic acid, (2) 10 pmol n-capric acid, (3) 10 pmol lauric acid, (4) 10 pmol myristic acid, and (5) 10 pmol palmitic acid; column, TSK gel ODS-80 Ts (150 × 4.6 mm; i.d., 5µm); eluent, acetonitrile-water (gradient, see Experimental Section); 1.0 mL min-1 flow rate; detection, excitation 391 nm, emission 519 nm.

products were extremely small. The calibration curves for the carboxylic acids were linear over the range from 39 fmol (19.5 nM) to 40 pmol (20 µM) per injection (r > 0.999). The detection limits (signal-to-noise ratio of 3) of the derivatives were 2.4-4.0 fmol per injection, which are equivalent to the derivatives of 3-bromomethyl-6,7-dimethoxy-1-methyl-2(1H)-quinoxalinone (BrDMEQ) (0.3-1 fmol),4 AE-OTf (0.8-2.7 fmol),38 4-(5,6-dimethoxy2-benzimidazoyl)benzohydrazide (DMBI-hydrazide) (1-3 fmol),28 6,7-methylenedioxy-1-methyl-2-oxo-1,2-dihydroquinoxalin-3-ylpropionohydrazide (MMEQ-hydrazide) (1-4 fmol),36 N-(4-nitro-2,1,3benzoxadiazoyl-7-yl)-N-methyl-2-aminoacetohydrazide (NBD-COHz) (2-4 fmol),35 4-(N,N-dimethylaminosulfonyl)-7-N-piperazino2,1,3-benzoxadiazole (DBD-PZ) (3-5 fmol),19 6,7-dimethoxy-1Analytical Chemistry, Vol. 73, No. 10, May 15, 2001

2169

methyl-2(1H)-quinoxalinone-3-propionylcarboxylic acid hydrazide (3-6 fmol),27 and 4-(N-hydrazinoformylmethyl-N-methyl)-amino7-N,N-dimethylaminosulfonyl-2,1,3-benzoxadiazole (DBD-CO-Hz) (3-9 fmol)34 and are lower than those of the other conventional fluorescent labeling reagents1-3,5-18,20-26,29-33,37,39 and fluorogenic AABD-SH.42 It should be pointed out that because the conventional fluorescent labeling reagents1-39 strongly fluoresce themselves, the fluorescence of the excess reagents and degradation products interfered with the quantification of a small amount of intended derivatives. Taken all together, these findings suggested that MTBD-SH was superior to the other reagents with regard to the detectability of carboxylic acids and was useful for the derivatization of a trace amount of biologically important carboxylic acids. In conclusion, the findings of the present study suggested that the PM3 calculation of the S1-T2 energy was useful for the development of a new highly sensitive benzofurazan reagent. The PM3 calculation showed that the S1-T2 energy of 4-methylamino-

2170


7-cyano-2,1,3-benzoxadiazole (-0.367 eV) was smaller than that of 4-methylamino-7-nitro-2,1,3-benzoxadiazole (-0.192 eV). Thus, we concluded that 7-fluoro-4-cyano-2,1,3-benzoxadiazole is a more sensitive reagent for amines than the very sensitive benzofurazan reagent proposed before, 7-fluoro-4-nitro-2,1,3-benzoxadiazole (NBDF).53 Further studies are currently in progress in our laboratory and these results will be reported soon. ACKNOWLEDGMENT We thank Dr. Chiho Lee for his valuable suggestions and discussion. Received for review October 18, 2000. Accepted February 14, 2001. AC001232J (53) Imai, K.; Watanabe, Y. Anal. Chim. Acta 1981, 130, 377-383.

A Fluorogenic Reagent, 4-Mercapto-7-methylthio-2,1,3

Recommend Documents