Aldose reductase inhibitors from the leaves of Myrciaria dubia (H. B. & K.) McVaugh.
Ellagic acid (1) and its two derivatives, 4-O-methylellagic acid (2) and 4-([alpha]-rhamnopyranosyl)ellagic acid (3) were isolated as inhibitors of aldose reductase (AR) from Myrciaria dubia (H. B. & K.) McVaugh. Compound 2 was the first isolated from the nature. Compound 3 showed the strongest inhibition against human recombinant AR (HRAR) and rat lens AR (RLAR). Inhibitory activity of compound 3 against HRAR (I[C.sub.50] value = 4.1 X [10.sup.-8] M) was 60 times more than that of quercetin (2.5 X [10.sup.-6] M). The type of inhibition against HRAR was uncompetitive.
[c] 2004 Published by Elsevier GmbH.
Keywords: Myrciaria dubia; Ellagic acid; 4-O-methylellagic acid; 4-([alpha]-rhamnopyranosyl)ellagic acid; Diabetes; Aldose reductase
Diabetes mellitus is a disease which contributes to lack of secretion or reaction of insulin. Lens, retina, nerves and kidney, which are insulin-insensitive, are the target organs for complications such as cataracts, retinopathy, neuropathy and nephropathy (Van-Heyningen, 1959; Robinson et al., 1983; Kador et al., 1985a,b; Beyer-Mears et al., 1986). Aldose reductase (AR) catalyzes glucose to sorbitol, and then sorbitol dehydrogenase converts sorbitol to fructose in the polyol pathway of glucose metabolism. Normally, AR has a low affinity for glucose, so that it has little catalytic activity for the conversion of glucose to sorbitol. However, in diabetes mellitus, the polyol pathway accelerates the formation of sorbitol in insulin-insensitive tissues. For these reasons, AR inhibitors are able to prevent the reduction of glucose to sorbitol and reduce diabetic complications (Terashima et al., 1984; Aida et al., 1990; Mizuno et al., 1999). During a search for possible AR inhibitors from Amazonian plants, an 80% MeOH extract of the leaves of Myrciaria dubia (H. B. & K.) McVaugh (Myrtaceae) was found to contain AR inhibitors, identified here as compounds 1, 2 and 3.
Materials and methods
Melting points were measured on a Buchi B-540 apparatus and are uncorrected. UV and IR spectra were measured on a Hitachi U-3010 spectrometer in MeOH and on a Shimadzu FTIR-8200 spectrometer in KBr, respectively. Optical rotation was measured on JASCO DIP-370 digital polarimeter. [.sup.1]H- and [.sup.13]C-NMR spectra were measured on a Varian VXR-500 instrument at 500 and 125 MHz, respectively, with TMS as internal standard. HMQC, HMBC and NOESY spectra were measured on a Varian VXR-500 instrument at 500 MHz. EIMS and HR-EIMS, SIMS and HR-SIMS (glycerol matrix) were measured on a Hitachi M-4100 instrument. Preparative medium pressure liquid chromatography (MPLC) was performed with a YFLC-10V equipped with a No.540-SY-2CSC pump, a DV-2 injector and a prep UV-10V.HC detector on a ODS-S-50C C18 (37 X 300 mm) (Yamazen Corp., Osaka, Japan). Preparative HPLC was performed with a Waters 600 pump, a 600E controller and a 996 photodiode array detector on a Cosmosil 5C18-AR II (20 X 250 mm) (Nakarai Tesque, Inc., Kyoto, Japan).
Extraction and isolation
Plant material: The dry leaves of M. dubia (H. B. & K.) McVaugh were collected in Iquitos, Peru in 1999, and identified by Dr. Franklin Ayala. A voucher specimen (No. 6720), has been deposited at Amazonian Natural Products, Peru.
Extraction and isolation: The dry leaves of M. dubia (800 g) were extracted with 80% MeOH (5000 ml X 3). Each filtrate was combined and evaporated to obtain the extract (100 g), which was dissolved successively in CH[Cl.sub.3] (500 ml X 3, 28g) and EtOAc (500 ml X 3, 5g). The residue (43 g) was treated with 20% MeOH to obtain soluble (40 g) and insoluble (2.5 g) fractions. The soluble fraction was chromatographed on a Sephadex LH-20 column (Amersham Biosciences, Uppsala, Sweden) (5 X 20 cm) using a stepwise gradient of MeOH-[H.sub.2]O (2:8 [right arrow] 10:0) to yield 7 fractions (numbered 1-7). Compound 1 (2.4 g) was isolated by recrystallization of fraction 7 from MeOH. Furthermore, fractions 3 and 5 were separately subjected to reversed-phase preparative MPLC using a gradient of MeOH-[H.sub.2]O (2:8 [right arrow] 8:2), and then reversed-phase preparative HPLC using MeCN-[H.sub.2]O (15:85 [right arrow] 40:60) containing 0.1% trifluoroacetic acid. Compounds 2 (80 mg) and 3 (40 mg) were isolated from fraction 3 and fraction 5, respectively.
Materials: Ellagic acid was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan) and used after recrystallization from methanol.
Compound 1: Ellagic acid: Pale yellow powder; mp > 300 [degrees]C; UV (MeOH): [[lambda].sub.max] (log [epsilon]) = 254 (4.41), 366 nm (3.71); IR (KBr): [v.sub.max] = 3100 (OH), 1699 (ArC[O.sub.2]), 1618 [cm.sup.-1] (Ar); EIMS (positive-ion mode): m/z (rel. int.) = 302 [M][.sup.+] (6.6); HR-EIMS (positive-ion mode): m/z = 302.0078 [M][.sup.+] (Calcd for [C.sub.14][H.sub.6][O.sub.8], 302.0062); [.sup.1]H-NMR (500 MHz, C[D.sub.3]OD): [delta] = 7.55 (2H, s, H-5 and H-5') (Khac et al., 1990; Nawwar et al., 1994; Ueda et al., 2001).
Compound 2: 4-O-Methylellagic acid: Light brown powder; mp > 300 [degrees]C; UV (MeOH): [[lambda].sub.max] (log [epsilon]) = 255 (4.38), 363 nm (3.70); IR (KBr): [v.sub.max] = 3421 (OH), 1717 (ArC[O.sub.2]), 1635 [cm.sup.-1] (Ar); EIMS (positive-ion mode): m/z (rel. int.) = 316 [M][.sup.+] (6.1); HR-EIMS (positive-ion mode): m/z = 316.0229 [M][.sup.+] (calcd for [C.sub.15][H.sub.8][O.sub.8], 316.0218); [.sup.1]H-NMR (500 MHz, C[D.sub.3]OD): [delta] = 4.04 (3H, s, 4-OC[H.sub.3]), 7.55 (1H, s, H-5'), 7.65 (1H, s, H-5); [.sup.13]C-NMR (125 MHz, C[D.sub.3]OD): [delta] = 57.3 (4-OC[H.sub.3]), 108.0 (C-5), 109.3 (C-6), 110.0 (C-6'), 112.0 (C-5'), 113.8 (C-1'), 115.7 (C-1), 138.2 (C-2), 139.6 (C-2'), 140.7 (C-3'), 141.8 (C-3), 150.0 (C-4'), 151.4 (C-4), 161.4 (C-7'), 161.5 (C-7) (Takahashi et al., 1977).
Compound 3: 4-([alpha]-Rhamnopyranosyl)ellagic acid (eschweilenol C): White powder; mp > 300[degrees]C; [[alpha]][.sup.20]; -87.0[degrees] (c = 0.02, MeOH); UV (MeOH): [[lambda].sub.max] (log [epsilon]) = 255 (4.82), 359 nm (3.88); IR (KBr): [v.sub.max] = 3377 (OH), 1724 (ArC[O.sub.2]), 1622 [cm.sup.-1] (Ar); SIMS (negative-ion mode): m/z (rel. int.) = 447 [M-H][.sup.-] (8.7); HR-SIMS (negative-ion mode) m/z = 447.0581 [M-H][.sup.-] (Calcd for [C.sub.20][H.sub.15][O.sub.12], 447.0562); [.sup.1]H-NMR (500 MHz, C[D.sub.3]OD): [delta] = 1.27 (1H, d, J = 6.5Hz, H-6"), 3.51 (1H, t, J = 9.5 Hz, H-4"), 3.74 (1H, m, H-5"), 3.99 (1H, dd, J = 3.5, 9.5 Hz, H-3"), 4.18 (1H, br s, H-2"), 5.57 (1H, br s, H-1"), 7.56 (1H, s, H-5'), 7.93 (1H, s, H-5); [.sup.13]C-NMR (125 MHz, C[D.sub.3]OD): [delta] = 18.1 (C-6"), 71.3 (C-5"), 71.8 (C-2"), 72.2 (C-3"), 73.8 (C-4"), 101.8 (C-1"), 109.3 (C-6), 110.1 (C-6'), 111.9 (C-5'), 113.7 (C-1'), 113.9 (C-5), 116.5 (C-1), 138.1 (C-2), 138.2 (C-2'), 141.9 (C-3'), 142.9 (C-3), 148.0 (C-4), 150.3 (C-4'), 161.3 (C-7'), 161.4 (C-7) (Yang et al., 1998).
Assay for aldose reductase inhibition activity
Materials: Human recombinant AR (HRAR) and DL-glyceraldehyde were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan), quercetin from Nakalai Tesque, Inc. (Kyoto, Japan) and [beta]-NADPH from Oriental Yeast Co., Ltd. (Osaka, Japan).
Animals: Male Wistar rats (older than 14 weeks) were purchased from Nihon SLC Co., Ltd., Hamamatsu, Japan.
Purification of rat lens AR (RLAR): All steps were performed at 4[degrees]C. RLAR was prepared with some modifications for the procedure developed by Kador (Kador et al., 1986). Lenses from male Wistar rats were homogenized with 10 vol of 10 mM phosphate buffet (pH 7.0) containing 1 mM 2-mercaptoethanol and 1 mM ethylenediamine tetraacetic acid, and centrifuged at 18,000g for 20 min. Ammonium sulfate was added slowly to the stirring supernatant fraction to yield a 30% saturated solution. After stirring for 3h, the solution was centrifuged at 18,000g for 20 min, and the supernatant was added gradually to 80% saturated ammonium sulfate. The solution was stirred for 3h and centrifuged at 18,000g for 20 min to obtain the precipitate. It was dissolved in a minimal amount of the phosphate buffer described above and desalted on a Sephadex G-75 (Amersham Biosciences, Uppsala, Sweden) column (2.5 X 50 cm) with the same buffer. The fractions with AR activity were then combined, concentrated by Amicon 8MC Micro-ultrafiltration system (Amicon Inc., Beverly, USA) with YM-30 membrane (Millipore Corp., Bedford, USA) and stored at -40 [degrees]C.
Enzyme assay: Enzyme activities were assayed spectrophotometrically on a Hitachi U-3010 spectrophotometer. The activities of HRAR and RLAR were according to the procedure of Nishimura (Nishimura et al., 1991). The reaction mixture contained 0.15 mM [beta]-NADPH, 10 mM DL-glyceraldehyde, 5 [micro]l of HRAR or RLAR and 5 [micro]l of test sample solution or dimethyl sulfoxide (DMSO) in a total volume of 1 ml of 100 mM sodium phosphate buffer (pH 6.2). After the reaction mixtures were incubated at 25 [degrees]C for 3 min in advance, the reaction was started by addition of the enzyme, and then the decrease of absorbance was measured at 340 nm for 1 min using a Hitachi U-3010 spectrophotometer. The inhibitory activity (%) was estimated as follows: [1 - ([DELTA]A sample/min - [DELTA]A blank/min)/([DELTA]A control/min - [DELTA]A blank/min)] X 100.
[DELTA]A sample/min showed a decrease of absorbance for 1 min with a sample, [DELTA]A blank/min with DMSO and water instead of a sample and a substrate, respectively, and [DELTA]A control/min with DMSO instead of a sample. The kinetic studies of inhibitory activity against AR among the substrate, and compounds 2 and 3 were analyzed using the Lineweaver-Burk plot.
Results and discussion
Isolation of AR inhibitors
The 80% MeOH extract of the leaves of M. dubia was separated into three fractions: CH[Cl.sub.3]- and EtOAc-soluble fractions and the residue. The residue showed the highest inhibitory activity against HRAR in all three fractions. The residue was further chromatographed on Sephadex LH-20 followed by preparative MPLC. Three active compounds were finally isolated by preparative HPLC.
Compound 1, [C.sub.14][H.sub.6][O.sub.8] determined by HR-EIMS, was identified as ellagic acid on the basis of spectral analyses (Khac et al., 1990; Nawwar et al., 1994; Ueda et al. 2001).
Compound 2, [C.sub.15][H.sub.8][O.sub.8] determined by HR-EIMS was obtained as a light brown powder, for which UV and IR spectra were similar to compound 1. The [.sup.1]H- and [.sup.13]C-NMR spectra showed that compound 2 consisted of the ellagic acid skeleton and one the methoxyl group (3 H, [delta]4.04), which was in agreement with the molecular formula. The position of the methoxyl group was determined at C-4 by the correlation between protons of methoxyl group ([delta]4.04) and H-5 ([delta]7.65) on NOESY. Based on these results, compound 2 was determined to be 4-O-methylellagic acid (Fig. 1). This was the first isolation from nature (Takahashi et al., 1977).
Compound 3 was obtained as a white powder with the molecular formula [C.sub.20][H.sub.15][O.sub.12], determined by HR-SIMS. The [.sup.1]H- and [.sup.13]C-NMR spectra revealed that compound 3 consisted of ellagic acid and a sugar. The type of sugar was determined as rhamnose by analysis of the chemical shifts and coupling patterns of its proton signals, analyzed by the COSY, HMQC and HMBC. According to the observation of a NOE effect between H-5 ([delta]7.93) and H-1" ([delta]5.57) in the NOESY, rhamnose was attached to C-4-O at ellagic acid. The configuration of rhamnose was determined to be [alpha] by comparing the chemical shifts of both [alpha]- and [beta]-rhamnoses in [.sup.13]C-NMR spectra (Tanaka, 1985). Thus, the structure of compound 3 was established as 4-([alpha]-rhamnopyranosyl)ellagic acid (Fig. 1) (Yang et al., 1998).
Inhibitory activity of compounds 1, 2 and 3 against AR
Inhibitory activities against HRAR and RLAR, of compounds 1, 2 and 3, and quercetin as a positive control, were evaluated. For HRAR and RLAR, DL-glycelaldehyde and [beta]-NADPH were used as a substrate and a cofactor, respectively. Compound 3 inhibited both HRAR (I[C.sub.50] value = 4.1 X [10.sup.-8]M) and RLAR (2.9 X [10.sup.-8] M) most strongly of all three compounds, and its inhibitory activity against HRAR was 60 times more than that of quercetin (2.5 X [10.sup.-6] M) (Table 1).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Compound 1 has been reported to inhibit HRAR in a uncompetitive manner (Ueda et al. 2001). The effects of compounds 2 and 3 in the Lineweaver-Burk plot of HRAR-inhibitory activity with DL-glyceraldehyde as a substrate were shown in Fig. 2. The Lineweaver-Burk plot of 1/velocity vs. 1/[DL-glyceraldehyde] in the presence of each compound was parallel to that in the absence of each compound. Thus compounds 2 and 3 inhibited HRAR activity in a uncompetitive manner. We speculated that these three compounds combined neither with the free enzyme nor with the normal substrate, but rather with the enzyme-substrate complex to produce the inactive enzyme-substrate-inhibitor complex, and then the reaction did not proceed to produce sorbitol.
During a search for possible AR inhibitors from Amazonian plants, ellagic acid (1), 4-O-methylellagic acid (2) and 4-([alpha]-rhamnopyranosyl)ellagic acid (3) were identified as AR inhibitors from M. dubia. Compound 3 showed the strongest inhibition of HRAR and RLAR of the three compounds, and its inhibitory activity against HRAR was 60 times more than that of quercetin. Their inhibition against HRAR was uncompetitive.
Table 1. Inhibitory activity of compounds 1-3, and quercetin against AR. I[C.sub.50] (M) value Compound HRAR RLAR 1 2.7 X [10.sup.-7] 4.7 X [10.sup.-8] 2 2.4 X [10.sup.-7] 1.4 X [10.sup.-7] 3 4.1 X [10.sup.-8] 2.9 X [10.sup.-8] Quercetin 2.4 X [10.sup.-6] n.t. n.t.: not tested. HRAR: human recombinant AR; RLAR: rat lens AR.
Received 20 September 2003; accepted 19 December 2003
Abbrevations: AR -- aldose reductase; HRAR -- human recombinant AR; RLAR -- rat lens AR
Aida, K., Tawata, M., Shindo, H., Onaya, T., Sasaki, H., Yamaguchi, T., Chin, M., Mitsuhashi, H., 1990. Isoliquiritigenin: a new aldose reductase inhibitor from Glycyrrhizae radix. Planta Med. 56, 254-258.
Beyer-Mears, A., Cruz, E., Edelist, T., Varagiannis, E., 1986. Diminished proteinuria in diabetes mellitus by sorbinil an aldose reductase inhibitor. Pharmacology 32, 52-60.
Kador, P.F., Robinson Jr., W.G., Kinoshita, J.H., 1985a. The pharmacology of aldose reductase inhibitors. Ann. Rev. Pharmacol. Toxicol. 25, 691-714.
Kador, P.F., Kinoshita, J.H., Sharpless, N.E., 1985b. Aldose reductase inhibitors: a potential new class of agents for the pharmacological control of certain diabetic complications J. Med. Chem. 28, 841-849.
Kador, P.F., Kinoshita, J.H., Brittain, D.R., Mirrlees, D.J., Sennitt, C.M., Stribling, D., 1986. Purified rat lens aldose reductase. Polyol production in vitro and its inhibition by aldose reductase inhibitors. Biochem. J. 240, 233-237.
Khac, D.D., Tran-Van, S., Campos, A.M., Lallemand, J.-Y., Fetizon, M., 1990. Ellagic compounds from Diplopanax stachyanthus. Phytochemistry 29, 251-256.
Mizuno, K., Kato, N., Makino, M., Suzuki, T., Shindo, M., 1999. Continuous inhibition of excessive polyol pathway flux in peripheral nerves by aldose reductase inhibitor fidarestat leads to improvement of diabetic neuropathy. J. Diabet Complications 13, 141-150.
Nawwar, M.A.M., Hussein, S.A.M., Merfort, I., 1994. NMR spectral analysis of polyphenols from Punica granatum. Phytochemistry 36, 793-798.
Nishimura, C., Yamaoka, T., Mizutani, M., Yamashita, K., Akera, T., Tanimoto, T., 1991. Purification and characterization of the recombinant human aldose reductase expressed in baculovirus system. Biochim. Biophys. Acta 1078, 171-178.
Robinson Jr., W.G., Kador, P.F., Kinoshita, J.H., 1983. Retinal capillaries: basement membrane thickening by galactosemia prevented with aldose reductase inhibitor. Science 221, 1177-1179.
Takahashi, M., Ueda, J., Sasaki, J., 1977. The components of the plants of Lagerstroemia genus V Synthesis of the aglycon of lagertannin 3,4-di-O-methylellagic acid from gallic acid. Yakugaku Zasshi 97, 1236-1239.
Tanaka, O., 1985. Application of [.sup.13]C-nuclear magnetic resonance spectrometry to structural studies on glycosides; saponins of Panax spp. and natural sweet glycosides. Yakugaku Zasshi 105, 323-351.
Terashima, H., Hama, K., Yamamoto, R., Tsuboshima, M., Kikkawa, R., Hatanaka, I., Shigeta, Y., 1984. Effects of a new aldose reductase inhibitor on various tissues in vitro. J. Pharmacol. Exp. Ther. 229, 226-230.
Ueda, H., Tachibana, Y., Moriyasu, M., Kawanishi, K., Alves, S.M., 2001. Aldose reductase inhibitors from the fruits of Caesalpinia ferrea Mart. Phytomedicine 8, 377-381.
Van-Heyningen, R., 1959. Formation of polyols by the lens of the rat with 'sugar' cataract. Nature 184, 194-195.
Yang, S-W., Zhou, B-N., Wisse, J.H., Evans, R., Van-der-Werff, H., Miller, J.S., Kingston, D.G.I., 1998. Three new ellagic acid derivatives from the bark of Eschweilera coriacea from the Suriname rainforest. J. Nat. Prod. 61, 901-906.
H. Ueda (a), E. Kuroiwa (a), Y. Tachibana (a), K. Kawanishi (a,*), F. Ayala (b), M. Moriyasu (a)
(a) Kobe Pharmaceutical University, 4-19-1 Motoyamakitamachi, Higashinada-ku, Kobe, 658-8558 Japan
(b) Amazonian Natural Products, Iquitos, Peru
*Corresponding author. Tel./fax: +81-78-441-7576.
E-mail address: firstname.lastname@example.org (K. Kawanishi).