Bioorganic & Medicinal Chemistry Letters 21 (2011) 6884–6887
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl
Inhibitory effect on NO production of phenolic compounds from Myristica fragrans To Dao Cuong a, Tran Manh Hung a, MinKyun Na b, Do Thi Ha c, Jin Cheol Kim d, Dongho Lee e, SungWoo Ryoo f, Jeong Hyung Lee f, Jae Sue Choi g, Byung Sun Min a,⇑ a
College of Pharmacy, Catholic University of Daegu, Gyeongbuk 712-702, Republic of Korea College of Pharmacy, Yeungnam University, Gyeongbuk 712-749, Republic of Korea c National Institute of Medicinal Materials, 3B Quang Trung, Hoan Kiem, Hanoi, Viet Nam d Korea Research Institute of Chemical Technology, Daejeon 305-600, Republic of Korea e Korea University, Seoul 136-713, Republic of Korea f College of Natural Science, Kangwon National University, Kangwon, Republic of Korea g Faculty of Food Science and Biotechnology, Pukyung National University, Busan 608-737, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 13 June 2011 Revised 29 August 2011 Accepted 1 September 2011 Available online 17 September 2011 Keywords: Myristica fragrans Myristicaceae Phenolic Anti-inflammatory activity
a b s t r a c t Three new phenolics: ((7S)-80 -(benzo[30 ,40 ]dioxol-10 -yl)-7-hydroxypropyl)benzene-2,4-diol (1), ((7S)-80 (40 -hydroxy-30 -methoxyphenyl)-7-hydroxypropyl)benzene-2,4-diol (2) and ((8R,80 S)-7-(4-hydroxy-3methoxyphenyl)-80 -methylbutan-8-yl)-30 -methoxybenzene-40 ,50 -diol (3), along with four known compounds (4–7) were isolated from the seeds of Myristica fragrans. Their chemical structures were established mainly by 1D and 2D NMR techniques and mass spectrometry. Their anti-inflammatory activity was evaluated against LPS-induced NO production in macrophage RAW264.7 cells. Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Myristica fragrans Houtt (Myristicaceae) is an aromatic evergreen tree, which is cultivated in South Africa, India and other tropical countries. It has been reported to contain 25–30% fixed oils and 5–15% volatile oils, such as camphene, elemicin, eugenol, isoelemicin, isoeugenol, methoxyeugenol, pinene, sabinene, and safrol; and chemical substances, such as dihydroguaiaretic acid, myristicin, and lignans.1 The seeds of M. fragrans (nutmeg) were imported into Europe during the 12th century and they have long been used indigenously as a spice in many kinds of Western food.2 Nutmeg is also prescribed for medicinal purposes in Asia to treat many diseases, such as rheumatism, muscle spasm, decreased appetite, and diarrhea.3 Nutmeg has shown antioxidant, anti-inflammatory,4 protein tyrosine phosphatase 1B inhibitory,5 hepatoprotective,6 and acetylcholine esterase inhibitory activities.7 In the present paper, the isolation and structural elucidation of the isolates, as well as the evaluation of their inhibitory activity on LPS-induced NO production in macrophage RAW264.7 cells are described. Repeated column chromatography (silica gel, RP-18, and semipreparative HPLC) of the EtOAc soluble fraction of the seed of M. fragrans resulted in the isolation of three new (1–3) and four known ones (4–7). The chemical structures of the known com⇑ Corresponding author. Tel.: +82 53 850 3613; fax: +82 53 850 3602. E-mail address:
[email protected] (B.S. Min).
pounds were identified to be erythro-(7S,8R)-7-(4-hydroxy-3methoxypheny1)-8-[20 -methoxy-40 -(E)-propenyl)phenoxy]propan-7-ol (4),8 1-(2,6-dihydroxyphenyl)-9-(3,4-dihydroxyphenyl)0 1-nonanone (malabaricone C) (5),9 (+)-erythro-(7S,8R)-D8 -7-acet10 0 0 0 oxy-3,4,3 ,5 -tetramethoxy-8-O-4 -neolignan (6), and (+)-ery0 thro-(7S,8R)-D8 -7-hydroxy-3,4,30 ,50 -tetramethoxy-8-O-40 -neolignan (7),10 by comparing the physicochemical and spectroscopic data (IR, UV, MS, 1D, and 2D NMR) with those reported in the literatures (Fig. 1).11 Compound 1 was isolated as colorless oil, with the molecular formula C16H16O5, as determined by the HRTOFMS at m/z 287.0930 for the [MH] ion (calcd for C16H15O5, 287.0919). The IR spectrum of 1 suggested the presence of hydroxy group (3449 cm1), and aromatic absorption (1428 cm1).12a The 1H NMR spectrum showed six aromatic protons at dH 6.28–6.89, one methine proton at dH 4.89, four methylene protons at dH 2.84, 2.64, 2.05, and 1.90, and one methylenedioxy proton at dH 5.93 (Table 1). The 13C NMR spectrum of 1 showed 16 signals, including twelve aromatic carbons, two methylene carbons, one methine carbon signals at dC 79.1 (C-7) and one methylenedioxy carbon at dC 102.4 (Table 1). The 1H and 13C NMR spectra of 1 were very similar to those of a-(3-hydroxyphenyl)-1,3-benzodioxole-5-propanol, which is a production of synthesis.13 However, the 13C NMR spectrum showed a signal that could be assignable to an oxygenated
0960-894X/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bmcl.2011.09.021
6885
T. D. Cuong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6884–6887
OH OH
3
HO
2
1
4
8
7S
8'
6
1'
2'
R1
4'
R2
5 6' 5'
1 2
3'
H3CO HO
3
2
4
1
7
6
CH3 8R
8'S
CH3
5
1'
2'
3' 4'
6'
5'
OCH3 OH
OH
R1 + R2 = OCH2O R1 = OCH3 R2 = OH
3
OH H3CO
OCH3
O
HO
HO
OH OH
O 4
OH
5 OR H3CO O
H3CO
OCH3
H3CO R 6 7
COMe H
Figure 1. Chemical structure of isolated compounds (1–7).
Table 1 1 H and 13C NMR (400 MHz) spectroscopic data of compounds 1–3 Position
1 dH (ppm)
1 2 3 4 5 6 7 8 10 20 30 40 50 60 80 OCH2O 8-CH3 80 -CH3 3-OCH3 30 -OCH3
6.28 (d, 2.4) 6.33 6.87 4.89 2.05
(dd, 2.4, 8.4) (d, 8.4) (dd, 2.4, 10.4) (m), 1.90 (m)
6.89 (d, 1.2)
6.79 6.88 2.84 5.93
(d, 8.0) (dd, 1.2, 8.0) (m), 2.64 (m) (d, 2.4)
2 dC (ppm) 137.5 157.1 104.2 157.7 109.2 131.0 79.1 31.6 114.3 107.6 149.2 148.6 109.0 120.6 25.4 102.4
dH (ppm)
6.27 (s) 6.32 6.87 4.90 2.12
(d, 8.4) (d, 8.4) (dd, 2.0, 10.0) (m), 2.01 (m)
6.99 (s)
6.80 (d, 8.0) 6.86 (d, 8.0) 2.88 (m), 2.70 (m)
3.86 (s)
aromatic carbon at C-4, which was confirmed by the HMBC correlation between H-5 (dH 6.33)/H-6 (dH 6.87) and C-4 (dC 157.7). Furthermore, the long-range correlations between dH 6.28 (H-3)/4.89 (H-7) and dC 157.1 (C-2) were also observed in the HMBC, suggesting that the oxygenated aromatic carbon was located at the C-2 position (Fig. 2). The CD spectrum of 1 showed one positive cotton effect at De207 +1.00, and two negative effects at De215 5.29 and De290 4.10, suggesting a 7S configuration by comparison with (S)-malic acid dinaphthylester (showed one positive cotton effect at De215 +6.8, and one negative effect at De225 5.5).14 In addition, it was also confirmed by comparison with those of (S)-3-(2methoxyphenyl)-1-phenylpropan-1-ol, a production of synthesis.15 Based on the above data analyses, 1 was established to be
3 dC (ppm) 135.0 157.3 104.2 157.7 109.2 131.1 79.3 31.5 114.4 111.0 149.0 147.3 116.1 120.0 25.6
56.6
dH (ppm) 6.61 (d, 2.0)
6.74 6.59 2.75 1.51
(d, 8.0) (dd, 2.0, 8.0) (dd, 4.0, 16.0)2.59 (dd, 11.6, 16.0) (m)
6.12 (s)
6.60 (s) 1.61 (m) 0.87 1.10 3.78 3.81
(d, 6.4) (d, 6.0) (s) (s)
dC (ppm) 134.6 113.8 149.1 145.2 115.9 123.4 40.2 48.5 129.4 112.2 147.2 139.8 145.9 117.5 37.1 17.7 20.4 56.5 55.6
((7S)-80 -(benzo[30 ,40 ]dioxol-10 -yl)-7-hydroxypropyl)benzene-2,4diol. Compound 2 was obtained as colorless oil, with absorption at 3447 cm1 (OH) in its IR spectrum. The molecular formula of 2 was determined to be C16H18O5 from the molecular ion peak at m/z 289.1076 for the [MH] ion (calcd for 289.1075) in the HRTOFMS. Its UV spectrum showed maximum absorption at 281 nm.12b The 1H and 13C NMR spectra of 2 were very similar with those of 1. However, the 1H and 13C NMR spectra showed a signal that could be assignable to a methoxy group at the C-30 position of the benzene ring, which was confirmed by the HMBC correlation between dH 3.86 (30 -OCH3)/6.99 (H-20 )/6.80 (H-50 ) and dC 149.0 (C-30 ) (Fig. 2). In addition, the configuration at the 7-position of 2
6886
T. D. Cuong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6884–6887
OH OH
Table 2 Inhibition of NO Production in macrophage RAW264.7 cells by compounds 1–7
OH OH O
HO
OCH3
HO
O 2
1
H3CO HO
CH3
OH
OCH3
CH3
Compound
IC50 valuea (lM)
1 2 3 4 5 6 7
> 50 > 50 32.5 ± 2.2 25.0 ± 3.1 2.3 ± 0.4 24.5 ± 2.5 > 50
Celastrolb
1.0 ± 0.1
a
OH OH
3
The inhibitory effects are represented as the molar concentration (lM) giving 50% inhibition (IC50) relative to the vehicle control. These data represent the average values of three repeated experiments. b Positive control for NO production.
Figure 2. Selected HMBC correlations (H ? C) for new compounds 1–3.
was also determined to be S from its CD spectrum, which showed one positive cotton effect at De208 +1.25, and two negative effects at De216 2.63 and De290 1.15, suggesting a 7S configuration.14 Thus, 2 was established to be ((7S)-80 -(40 -hydroxy-30 -methoxyphenyl)-7-hydroxypropyl)benzene-2,4-diol.Compound 3 was isolated as colorless oil, with the molecular formula C19H24O5, as determined by the HRTOFMS at m/z 331.1531 for the [MH] ion (calcd for C19H23O5, 331.1545). The IR spectrum of 3 suggested the presence of hydroxy group at 3442 cm1, 2919 (C–C) and 1425 (aromatic absorption) cm1.12c The 1H and 13C NMR spectra of 3 were similar with those of meso-dihydroguaiaretic acid isolated from Myristica argentea (Table 1).16 However, the 13C NMR of 3 possessed a signal that could be assignable to an oxygenated aromatic carbon at the C-50 position, which was confirmed by the HMBC correlation between dH 6.60 (H-60 ) and dC 145.9 (C-50 ) (Fig. 2). In addition, the long-range correlations between two methyl protons at dH 0.87 (8-CH3)/dH 1.10 (80 -CH3) and carbon signals at dC 48.5 (C-8)/dC 37.1 (C-80 ) were also observed, suggesting that two the methyl groups were located at the C-8 and C-80 positions. Comparison of the 1H and 13C NMR spectra of 3 with those of meso-dihydroguaiaretic acid,16 3 showed no signal for the methylene group at the C-70 position in those of the former, suggesting 3 was a norlignan. It was also further confirmed by HMBC correlations between dH 1.51 (H8)/1.61 (H-80 ) and dC 129.4 (C-10 ), and between the signal at dH 6.12 (H-20 ) and dC 37.1 (C-80 ) (Fig. 2 and Supplementary data). Together with the optical rotation ½a25 D –8.0 (c 0.1 MeOH), the CD spectrum of 3 further showed two negative effects at De239 10.26 and De275 7.79, and one positive cotton effects at De292 +10.71, suggesting the absolute configuration of 3 was 8R and 80 S by comparison with 0 those of (+)-erythro-(7S,8R)-D8 -7-acetoxy-3,4,30 ,50 -tetramethoxy8-O-40 -neolignan [CD (c 0.13, MeOH): De202 +8.48, De244 2.56 0 and De280 1.19] and (+)-erythro-(7S,8R)-D8 -7-hydroxy-3,4,30 ,50 0 tetramethoxy-8-O-4 -neolignan [CD (c 0.25, MeOH): De208 +5.58, De244 6.04 and De273 1.54].12 Based on the above data analyses, 3 was established to be ((8R,80 S)-7-(4-hydroxy-3-methoxyphenyl)80 -methylbutan-8-yl)-30 -methoxybenzene-40 ,50 -diol. The cytotoxic effects of the isolated compounds (1–7) were evaluated in the presence or absence of LPS using the MTT assay. These compounds did not affect the cell viabilities of RAW 264.7 cells in either the presence or absence of LPS, even at a dose of 50 lM after a period of 24 h (data not shown). The amount of NO produced was determined by the amount of nitrite, a stable metabolite of NO. To assess the effects of these compounds on the LPS-induced production of NO in RAW 264.7 cells, a cell culture medium was harvested, with the production of nitrite measured using the Griess reaction.17 As shown in Table 2 and 5 showed inhibitory potency, with an IC50 value of 2.3 lM, 3, 4, and 6 displayed moderate effects with IC50 values of 32.5, 25.0, and 24.5 lM, respectively, but
LPS (1 g/ml) C5 ( M)
-
+
-
0
+ 1
+ 3
+ 10
iNOS COX-2 Tubulin Figure 3. Inhibition of LPS-induced iNOS and COX-2 expression in RAW264.7 cells by compound 5 (C5). RAW264.7 cells were pretreated for 30 min with the indicated concentrations of 5, and followed by the stimulation with LPS (1 lg/ml) for 18 h. Total lysates were prepared and the expression levels of iNOS and COX-2 were determined by immunoblot analysis. Histograms show densitometric analyses of relative iNOS and COX-2 expression levels normalized against a-tubulin. bar., the mean ± S.D. Asterisk indicates a significant difference (**, P <0.01) compared the control.
the others (1, 2, and 7) were weak or inactive. Because 5 exhibited a strong inhibitory effect on the LPS-induced production of NO in macrophage RAW264.7 cells, the effect of this compound on the LPS-induced COX-2 and iNOS expressions were investigated. RAW264.7 cells were stimulated with 1 lg/mL of LPS for 18 h in the presence of increasing concentrations of 5, and the expression levels of COX-2 and iNOS protein were determined by immunoblot analyses (Fig. 3).18 Compound 5 (0.1–10 lM) dose-dependently reduced the LPS-induced COX-2 and iNOS expressions, but did not change the a-tubulin expression. The results showed that 5 (malabaricone C) inhibited, not only the iNOS and COX-2 mRNA expressions, but also the iNOS and COX-2 promoter activities in LPS stimulated RAW264.7 cells, suggesting that this compound could suppress LPS-induced iNOS and COX-2 expressions at the transcription level. Acknowledgment This research was supported by the National Research Foundation of Korea Grant funded by the Korean Government (MEST) (KRF-2009-0084675). References and notes 1. (a) Forrest, J. E.; Heacock, R. A. J. Chromatogr., A 1972, 69, 115; (b) Isogai, A.; Suzuki, A.; Tamura, S. Agar. Biol. Chem. 1973, 37, 193; (c) Kuo, Y. H. Gaoxiong Yi Xue Ke Xue Za Zhi 1989, 5, 621; (d) Janssen, J.; Laekeman, G. M.; Peiters, L. A.; Totte, J.; Herman, A. G.; Vlietinck, A. J. J. Ethnopharmacol. 1990, 29, 179. 2. Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and Medicine, 3rd ed.; Oxford University Press, 2000. 3. Kim, S. H.; Lee, D. H.; Kwon, S. H.; Lim, B. H.; Lee, S. H.; Min, B. C. Korean J. Pharmacogn. 2007, 38, 19. 4. Jin, D. Q.; Lim, C. S.; Hwang, J. K.; Ha, I.; Han, J. S. Biochem. Biophys. Res. Commun. 2005, 331, 1264. 5. Yang, S.; Na, M. K.; Jang, J. P.; Kim, K. A.; Kim, B. Y.; Sung, N. J.; Oh, W. K.; Ahn, J. S. Phytother. Res. 2006, 20, 680.
T. D. Cuong et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6884–6887 6. Morita, T.; Jinno, K.; Kawagishi, H.; Arimoto, Y.; Suganuma, H.; Inakuma, T.; Sugiyama, K. J. Agric. Food Chem. 2003, 51, 1560. 7. Mukherjee, P. K.; Kumar, V.; Houghton, P. J. Phytother. Res. 2007, 21, 1142. 8. Hada, S.; Hattori, M.; Tezuka, Y.; Kikuchi, T.; Namba, T. Phytochemistry 1988, 27, 563. 9. (a) Purushothaman, K. K.; Sarada, A.; Connolly, J. D. J. Chem. Soc. 1977, 5, 587; (b) Pham, V. C.; Jossang, A.; Sevenet, T.; Bodo, B. Tetrahedron 2000, 56, 1707. 10. Lin, D.; Hong, W. T.; Xiao, J. H.; Qian, Q. G.; Wei, M. Z. Planta Med. 2009, 75, 1241. 11. The seeds of M. fragrans were purchased in Dong Xuan herbarium market, Hanoi, Vietnam, in January 2010 and identified by Professor Pham Thanh Ky, Department of Pharmacognosy, Hanoi University of Pharmacy. A voucher specimen (0182) was deposited in the herbarium of the Hanoi College of Pharmacy. The seeds of M. fragrans (2 kg) were extracted three times (3 h 3 L) with refluxing methanol. After the solvent was removed under reduced pressure, the residue was suspended in H2O and then partitioned with n-hexane, EtOAc, and n-BuOH, successively. The EtOAc soluble fraction (54 g) was chromatographed on a silica gel column (10 60 cm; 63–200 lm particle size, Merck) using a stepwise gradient of CHCl3/MeOH (50:1 to 0:1, each 2 L) to yield nine fractions (Fr.1–Fr.9) according to their TLC profiles. Fraction 5 (2.4 g) was subjected to reversed phase (ODS-A) column chromatography (6.0 60 cm; 150 lm particle size) and eluted with MeOH/H2O (from 4:1 to 4:0, 3 L for each step) to afford five sub-fractions (Fr.5-1 to Fr.5-5). Further purification of Fr.5-3 (430 mg) by semi-preparative Waters HPLC systems [using an isocratic solvent system of 70% MeOH in H2O + 0.1% trifluoroacetic acid (flow rate 5 mL/min) over 90 min; UV detection at 210 nm; YMC Pak ODSA column (20 250 mm, 5 lm particle size] resulted in the isolation of compounds 4 (14 mg; tR = 38.5 min), 6 (12 mg; tR = 44.8 min) and 7 (14 mg; tR = 48.2 min), respectively. Fraction 6 (3.5 g) was also subjected to reversed phase (ODS-A) column chromatography (6.0 60 cm; 150 lm particle size) and eluted with MeOH/H2O (from 5:2 to 5:0, 2 L for each step) to afford seven sub-fractions (Fr.6-1 to Fr.6-7). Further purification of Fr.6-2 (410 mg) by semipreparative Waters HPLC systems [using an isocratic solvent system of 65% MeOH in H2O + 0.1% trifluoroacetic acid (flow rate 5 mL/min) over 90 min; UV detection at 210 nm; YMC Pak ODS-A column (20 250 mm, 5 lm particle size] resulted in the isolation of compounds 1 (5.7 mg, tR = 37.2 min), 2 (3.6 mg; tR = 40.8 min), and 5 (50.3 mg, tR = 44.2 min), respectively. Subfraction F.6-3 (254 mg) was further purified by semi-preparative HPLC [using an isocratic solvent system of 75% MeOH in H2O + 0.1% trifluoroacetic acid (flow rate 5 mL/min) over 90 min; UV detection at 210 nm; YMC Pak ODS-A column (20 250 mm, 5 lm particle size] resulted in the isolation of compound 3 (3.0 mg, tR = 46.2 min). 12. Physical and spectroscopic data of new compounds: (a) ((7S)-80 (benzo[30 ,40 ]dioxol-10 -yl)-7-hydroxypropyl)benzene-2,4-diol (1): Colorless oil;
13. 14. 15. 16. 17.
18.
6887
½a25 D 11.0 (c 0.1, MeOH); IR mmax (KBr): 3449, 2923, 1428, 1263, 1162, 1033 cm1; UV kmax (MeOH): 284 nm; CD (c 0.1, MeOH): De207 +1.00, De215 5.92, De290 4.10; HRTOFMS m/z 287.0930 [MH] (calcd for C16H15O5, 287.0919). 1H NMR (400 MHz, CD3OD) and 13C NMR (100 MHz, CD3OD) spectroscopic data, see Table 1. (b) ((7S)-80 -(40 -hydroxy-30 -methoxyphenyl)-7hydroxypropyl)benzene-2,4-diol (2): Colorless oil; ½a25 D 28.0 (c 0.05, MeOH); IR mmax (KBr): 3447, 2928, 1421, 1262, 1161, 1035 cm1; UV kmax (MeOH): 281 nm; CD (c 0.09, MeOH): De208 (nm) +1.17, De216 2.63, De290 1.45; HRTOFMS m/z 289.1076 [MH] (calcd for C16H17O5, 289.1075). 1H NMR (400 MHz, CD3OD) and 13C NMR (100 MHz, CD3OD) spectroscopic data, see Table 1. (c) ((8R,80 S)-7-(4-hydroxy-3-methoxyphenyl)-80 -methylbutan-8-yl)-30 methoxybenzene-40 ,50 -diol (3): Colorless oil; ½a25 D 8.0 (c 0.1, MeOH); IR mmax (KBr): 3442, 2919, 1425, 1264 1165, 1031 cm1; UV kmax (MeOH): 283 nm; CD (c 0.1, MeOH): De239 10.26, De275 7.79, De292 +10.71; HRTOFMS m/z 331.1531 [MH] (calcd for C19H23O5, 331.1545). 1H NMR (400 MHz, CD3OD) and 13C NMR (100 MHz, CD3OD) spectroscopic data, see Table 1. Andrade, C. K. Z.; Silva, W. A. Lett. Org. Chem. 2006, 3, 39. Fischbeck, A.; Bartke, N.; Humpf, H. U. Monatshefte Für Chemie 2005, 136, 397. Liu, Y.; Da, C. S.; Yu, S. L.; Yin, X. G.; Wang, J. R.; Fan, X. Y.; Li, W. P.; Wang, R. J. Org. Chem. 2010, 75, 6869. Nakatani, N.; Ikeda, K.; Kikuzaki, H.; Kido, M.; Yamaguchi, Y. Phytochemistry 1988, 27, 3127. Determination of NO production and the cell viability assay. The level of NO production was determined by measuring the amount of nitrite from the cell culture supernatants as described previously. Briefly, the RAW264.7 cells (1 105 cells/well) were stimulated with or without 1 lg/mL of LPS (Sigma Chemical Co., St. Louis, MO) for 24 h in the presence or absence of the test compounds (0.5–25 lM). The cell culture supernatant (100 lL) was then reacted with 100 lL of Griess reagent. The remaining cells after the Griess assay were used to test their viability using a MTT (Sigma Chemical Co., St. Louis, MO)-based colorimetric assay as previously described. Immunoblot analysis. Proteins were extracted from cells in ice-cold lysis buffer (50 mM Tris–HCl, pH 7.5, 1% Nonidet P-40, 1 mM EDTA, 1 mM phenylmethyl sulfonyl fluoride, 1 lg/ml leupeptin, 1 mM sodium vanadate, 150 mM NaCl). Fifty microgram of protein per lane was separated by sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis (PAGE) and followed by transferring to a polyvinylidene difluoride membrane (Millipore, Bedford, MA, USA). The membrane was blocked with 5% skim milk, and then incubated with the corresponding antibody. Antibodies for COX-2, iNOS were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibody for a-tubulin was from Sigma. After binding of an appropriate secondary antibody coupled to horseradish peroxidase, proteins were visualized by enhanced chemiluminescence according to the instructions of the manufacturer (Amersham Pharmacia Biotec, Buckinghamshire, UK).