Influence of Glucuronidation and Reduction

CHEMBIOCHEM FULL PAPERS DOI: 10.1002/cbic.201300080

Influence of Glucuronidation and Reduction Modifications of Resveratrol on its Biological Activities Dong-Liang Lu, De-Jun Ding, Wen-Jing Yan, Ran-Ran Li, Fang Dai, Qi Wang, Sha-Sha Yu, Yan Li, Xiao-Ling Jin, and Bo Zhou*[a] Resveratrol (3,5,4’-trihydroxystilbene, RES), a star among dietary polyphenols, shows a wide range of biological activities, but it is rapidly and extensively metabolized into its glucuronide and sulfate conjugates as well as to the corresponding reduced products. This begs the question of whether the metabolites of RES contribute to its in vivo biological activity. To explore this possibility, we synthesized its glucuronidation (3-GR and 4’-GR) and reduction (DHR) metabolites, and evaluated the effect of these structure modifications on biological activities, including binding ability with human serum albumin (HSA), antioxidant activity in homogeneous solutions and heterogeneous media, anti-inflammatory activity, and cytotoxicity against various cancer cell lines. We found that 1) 4’-GR, DHR and RES show nearly equal binding to HSA, mainly through hydrogen bonding, whereas 3-GR adopts a quite different orientation

mode upon binding, thereby resulting in reduced ability; 2) 3GR shows comparable (even equal) ability to RES in FRAP- and AAPH-induced DNA strand breakage assays; DHR, 3-GR, and 4’GR exhibit anti-hemolysis activity comparable to that of RES; additionally, 3-GR and DHR retain some degree activity of the parent molecule in DPPHC-scavenging and cupric ion-initiated oxidation of LDL assays, respectively; 3) compared to RES, 4’GR displays equipotent ability in the inhibition of COX-2, and DHR presents comparable activity in inhibiting NO production and growth of SMMC-7721 cells. Relative to RES, its glucuronidation and reduction metabolites showed equal, comparable, or some degree of activity in the above assays, depending on the specific compound and test model, which probably supports their roles in contributing to the in vivo biological activities of the parent molecule.

Introduction Resveratrol (3,5,4’-trihydroxystilbene, RES), was isolated first from the roots of white hellebore (Veratrum grandiflorum) and subsequently from the roots of Japanese knotweed (Polygonum cuspidatum); it has been found in about 70 diverse plant species, including grapes, mulberries, and peanuts.[1e] During the past two decades, it has undoubtedly become a star molecule among dietary polyphenols and exhibits a wide range of biological activities, including antioxidant, anti-inflammatory, cardioprotective, anti-aging, and cancer-chemopreventive activities.[1] To translate these impressive observations to clinical application, studies on its bioavailability, pharmacokinetics, and metabolites are continuously undertaken: in vitro cell, ex vivo (small intestine), in vivo animal models, and in humans.[2] These studies demonstrate convincingly that RES is well absorbed but rapidly metabolized, predominantly into glucuronide and sulfate conjugates.[2c–j] Additionally, hydrogenation of the aliphatic double bond in the stilbene scaffold of RES is also suggested to occur in both glucuronide and sulfate metabolites.[2i,j] For example, Walle and co-workers have shown that after administration of a 25 mg oral dose (dietary amount), RES [a] D.-L. Lu, Dr. D.-J. Ding, W.-J. Yan, R.-R. Li, Dr. F. Dai, Q. Wang, S.-S. Yu, Y. Li, Dr. X.-L. Jin, Prof. Dr. B. Zhou State Key Laboratory of Applied Organic Chemistry, Lanzhou University 222 Tianshui Street S., Lanzhou 730000 (China) E-mail: [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201300080.

2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

absorption was at least 70 %, but systemic bioavailability was very low because of rapid and extensive metabolism.[2j] Notably, plasma concentration of unchanged RES was in only the nanomolar range—far below the micromolar range of its metabolites.[2j] These facts inevitably lead to doubt about the physiological relevance of the high RES concentrations usually used in in vitro models,[3a] and also invite the question: do RES metabolites contribute to the in vivo biological activities attributed to RES?[1e, 3b] The first step to answer this question is to investigate the biological activities of RES metabolites. However, research into the metabolites has been impeded by the paucity of commercial sources. Only a few studies have been conducted to test the biological activity of its metabolites and their ability to modulate the function of some known targets in vitro.[4] Specifically, Miksits et al. have reported that, in contrast to RES, its sulfated metabolites show poor cytotoxicity against human malignant and nonmalignant breast cancer cells;[4e] Calamini and co-workers have compared the action of RES and its metabolites (3- and 4’-O-sulfate, and 3-O-glucuronide conjugates) on different enzyme targets including cyclooxygenases (COX1 and -2), NAD + -dependent histone deacetylase, and quinone reductase 2 (QR2), and concluded that these metabolites might be responsible for the beneficial health effects attributed previously to the parent molecule;[4c] Hosino and collaborators synthesized RES sulfates and found that most of them were less active than the parent molecule, but that 3-O-sulfate ChemBioChem 2013, 14, 1094 – 1104

1094

CHEMBIOCHEM FULL PAPERS displayed comparable or even greater QR1 induction, 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical-scavenging, and COX1 inhibition;[4b] and, Storniolo and Moreno demonstrated that RES metabolites could significantly inhibit the growth of intestinal epithelial cancer (Caco-2 cells) by induction of G0/G1 phase arrest.[4a] However, relative to the numerous publications on activities and pleiotropic mechanisms of RES, data on its metabolites (especially its glucuronides and reduced product) are still scarce. Therefore, as a logic extension of our ongoing research into developing resveratrol-directed antioxidants, prooxidants, and cancer chemopreventive agents,[5] we synthesized its 3-O-b-d- and 4’-O-b-d-glucuronides (3-GR and 4’-GR), and its reduced product dihydroresveratrol (DHR, Scheme 1). We investigated how structure modifications affect biological activities, including binding ability to human serum albumin (HSA), antioxidant activity in homogeneous solutions and heterogeneous media, anti-inflammatory activity, and cytotoxicity against various cancer cells. To our knowledge, this is the first report of the binding abilities of RES metabolites with HSA and systematic evaluation their antioxidant activities.

www.chembiochem.org Results and Discussion Synthesis of the glucuronidation and reduction metabolites of resveratrol Synthesis of the 3-GR and 4’-GR resveratrol glucuronides has been achieved by Heck coupling of iodo-O-b-d-glucuronate derivatives to the corresponding styrenes,[6a] direct coupling of RES with the bromo-glucuronide donor, or coupling of silylprotected RES with trifluoroacetamidite or trichloroacetimidate glucuronide donors.[4f] We used a similar route to that reported by Lucas and colleagues to complete the synthesis of 3-GR and 4’-GR, but with slight difference in the deprotection process (Scheme 1).[6b] BF3·OEt2 promoted selective coupling of tert-butylmethylsilyl (TBS)-protected RES with a common trichloroacetimidate glucuronide donor to give 3 and 4, which were subsequently deprotected to furnish 3-GR and 4’-GR in the presence of NaOH and methanol. DHR was easily prepared by hydrogenation over Pd/C catalysts (Scheme 1).

Scheme 1. Molecular structures and synthesis of the glucuronidation and reduction metabolites of RES. Reagents and reaction conditions: a) TBSCl (2.1 equiv), imidazole (2.5 equiv), DMF, 5 8C to RT, 6 h; b) H2, Pd/C, EtOH or AcOEt; c) BF3·Et2O, CH2Cl2, 1 h; d) aqueous NaOH, MeOH, overnight.


ChemBioChem 2013, 14, 1094 – 1104

1095

CHEMBIOCHEM FULL PAPERS

www.chembiochem.org

Binding of resveratrol and its glucuronidation and reduction metabolites to HSA HSA, a principal extracellular protein with a high concentration in blood plasma, not only plays an important role in maintaining colloidal osmotic pressure and the pH of blood, but also serves as a carrier for many drugs to different molecular targets.[7] The interaction of drugs with HSA is one of the major determinants for their metabolism and the transportation process, and can also be used as a model for drug–protein binding.[7] Therefore, it is of interest to study the interaction of HSA with RES in aqueous solution, especially in view of the fact that (due to low water solubility) it has to be bound to HSA to retain a high concentration in serum. Although a few studies have investigate this interaction,[8] whether the RES metabolites possess HAS-binding ability, and how the glucuronidation and reduction modifications affect the ability of the parent molecule, are not known. Thus, we first characterized the interaction of RES and its metabolites with HSA by isothermal titration calorimetry (ITC), saturation transfer difference NMR (STD NMR), and molecular docking simulation. The ITC instrument can relatively easily provide the thermodynamic parameters of a ligand–protein interaction in a single experiment, based on a direct and precise measurement of the enthalpy change of the association process. Figure 1 shows the regular ITC profiles for the binding of RES and its metabolites with HSA in phosphate buffer (pH 7.4) at 25 8C. All the binding interactions are exothermic; dissociation constants (Kd), enthalpy changes (DH), and entropy changes (TDS) derived from the profiles are listed in Table 1. A comparison of Kd values clearly indicates that 4’-GR and DHR have binding abilities nearly identical to that for RES, whereas 3-GR is about six times less active. This also demonstrates that 4’-OH glucuronidation of RES and hydrogenation of the aliphatic double bond do not influence its ability to interact with HSA, but 3- or 5-OH glucuronidation can significantly decrease it. It is well known that hydrogen bonds, van der Waals forces, and hydrophobic and electrostatic interactions contribute to ligand–protein interaction, and the thermodynamic parameters (DH, DS, and free energy change, DG) provide the main evidence for the binding modes.[9] From a thermodynamic point of view, negative DH is frequently taken as evidence of hydrogen-bond formation at the binding site, whereas positive DS is characteristic of hydrophobic interactions.[9] Specifically, when DH < 0 and DS > 0 the forces are principally hydrogen-bond

Figure 1. Calorimetric profiles. Raw exothermic heat (above) and corresponding integrated data (below) of the binding of A) RES, B) DHR, C) 3-GR, and D) 4’-GR to HSA at 25 8C. HSA (50 mm) was titrated with increasing amounts of RES, DHR, 3-GR, or 4’-GR (stock solutions: 8, 4, 8, and 3 mm, respectively).

formation and hydrophobic interactions; when DH < 0 and DS < 0, this indicates hydrogen-bond formation and van der Waals forces; DH > 0 and DS > 0 reflects hydrophobic interactions; and DH 0 and DS > 0 reveals electrostatic interactions.[9] From Table 1 it is clear that for RES and its metabolites, the interactions with HSA are all accompanied by negative DH, DS, and DG, thus suggesting that hydrogen bonds and van der Waals forces play key roles in the binding process, which is

Table 1. Antioxidant and anti-inflammatory activities of RES and its metabolites and their thermodynamic parameters for binding to HSA. Compound

Kd[a] [mm]

RES DHR 3-GR 4’-GR

80.6 5 87.7 5 490 15 65.8 17

DH [kcal mol 1] 10.8 9.96 30.5 7.18

TDS [kcal mol 1] 5.22 4.41 26.0 1.47

FRAP[a] [n]

DPPH[a] EC50 [mm]

1.45 0.01 0.55 0.01 1.06 0.01 0.42 0.01

26.5 0.7 ~ 200 62.1 0.6 > 200

teff[b] [min]

C logP[c]

COX-2[a] IC50 [mm]

NO[a] IC50 [mm]

74 63 56 52

2.833 2.587 1.038 0.858

37.5 2.5 57.1 4.0 161 16 40.6 4.1

46.1 0.6 64.4 2.0 284 22 328 14

[a] Data are mean SD from three determinations. [b] Data are the average of three determinations, which were reproducible with deviation less than 10 %. [c] Calculated by Bio-Loom software.[16]


ChemBioChem 2013, 14, 1094 – 1104

1096


www.chembiochem.org

spontaneous and enthalpy driven. A similar conclusion was reached in a study by N’soukpo-Kossi et al: binding of RES to HSA is mainly through hydrogen bonds with polypeptide C=O, C N, and NH groups.[8c] STD NMR has emerged as a powerful tool in the study of ligand–protein interaction: comparison of the STD responses for different protons within a ligand reveals the orientation of a ligand when bound to its targets in solution.[10] Protons within a ligand that are closer to the target protein will have more-intense signals in 1H STD NMR because of more-efficient saturation transfer.[10] Consequently, Figure 3. Binding epitopes as determined by 1H STD NMR for resveratrol and its glucurowe conducted an STD NMR experiment to further innidation and reduction metabolites. Higher proton percentage indicates closer proximity vestigate the interaction of RES and its metabolites to the surface of HSA. a: overlapping signals (H-7, 8 and H-3’, 5’ of 4’-GR). with HSA. A representative STD NMR spectrum for RES is shown in Figure 2. Other compounds also exhibited the STD responses (see Figures S1–S3 in the Supportresponse of H-2’ and H-6’ at aromatic ring B. This could be ating Information), thus implying that RES and its metabolites tributable to the fact that free rotation of the carbon–carbon could efficiently interact with HSA. The relative STD responses single bond results in a slight difference in orientation befor different protons of RES and its metabolites were calculated tween RES and DHR upon binding. Additionally, 4’-GR and RES and normalized (largest proton signal set at 100 %). The resultemployed a similar pattern upon binding, as judged by their ing binding epitopes of RES and its metabolites (for protons epitopes. However, a weak STD response (30 %) was assigned that are close to HSA upon binding) are summarized in to the three protons (H-2’’, H-3’’, and H-4’’) of the glucuronosyl Figure 3. group, thus suggesting that this hydrophilic group is oriented Strong STD responses (69–100 %) were observed for all of to the outside of the binding pocket (towards the solvent). the aromatic protons of RES and H-4 on the aromatic ring A Strikingly, a clear distinction was observed between the bindhad the largest intensity, whereas a relatively weak response ing epitopes of 3-GR and RES. Specifically, glucuronidation of (49 %) was obtained for the two protons (H-7 and H-8) of the 3-OH reduced significantly the STD intensities of H-4, H-2(6), Haliphatic double bond. This reveals that the aromatic A and B 7, H-8, and H-2’(6’), and H-3’(5’) on aromatic ring B (rather than rings are in more intimate contact with the protein surface H-4 on the aromatic ring A) exhibited the largest STD response. than the double bond. DHR had a very similar binding epitope This probably reflects the fact that 3-GR (compared to RES, to RES, but hydrogenation of the double bond decreased the DHR, and 4’-GR) employs a significantly different orientation STD intensity of H-7 and H-8 along with an increase in the STD binding mode, hence impairing its binding ability to HSA, as evidenced by the ITC experiment. Molecule docking was also performed to further investigate the potential binding modes between RES or its metabolites and HSA (Figures 4, 5, and Figure S4). Crystallographic analysis of the HSA molecule showed that it is a heart-shaped monomer of 585 amino acid residues with three homologous a-helices in domains I (residues 1–195), II (196–383), and III (384– 585).[7b] Each domain is divided into subdomains “A” and “B”. The binding site called “site 1” (warfarin-binding site) in subdomain IIA contains a large hydrophobic cavity and is known to be an interaction site for many bioactive substances, including RES.[7, 8b] A 2D plot for the RES–HSA interaction emphasizes the nonbonded contacts of RES with the residues of site 1, including Tyr148, Pro147, Leu250, Phe149, Tyr150, Gln196, Cys200, and Cys245 (Figure 4). More importantly, there are five hydrogen bonds between phenolic hydroxy groups of RES and residues Gln29, Lys106, Ala151, and His242. Figure 5 and Figure S4 display 3D docking for RES and its metabolites to HSA. All the compounds make a number of hydrogen-bond interactions with amino acid residues and appear particularly well adapted to the pocket of site 1 (Figure 5). Although RES and DHR bind to HSA at essentially the same position, they exhibit a slight Figure 2. A) 1H NMR and B) 1H STD NMR spectra of RES (5 mm) in the presdifference in the orientation because of the free rotation of the ence of HSA (2 mg mL 1). 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemBioChem 2013, 14, 1094 – 1104

1097


Figure 4. LIGPLOT diagram for the binding of RES to active site 1 of HSA. Dotted lines show hydrogen bond interactions between the ligand and protein. The spoked arcs represent residues making nonbonded contacts with the ligand.

carbon–carbon single bond derived form the hydrogenation (Figure 5, see also Figure S4). As expected, 4’-GR (additional glucuronosyl group on aromatic ring B) binds with HSA in a conformation that places the group at the bottom of the polar pocket, where it can interact with the bulk solvent (Figures 5 and S4). Interestingly, the bound conformation of 3-GR is rotated by about 1808 with respect to that of RES, thus resulting in placing the glucuronosyl group of the aromatic ring A in the same hydrophilic subchamber to that of 4’-GR (Figure 5). The completely different orientation mode of 3-GR (compared to the other compounds) probably explains its lower binding ability. Indeed, these results from the binding-mode docking analysis are in agreement with those from the ITC and STD NMR data. Antioxidant activity of resveratrol and its glucuronidation and reduction metabolites One of the most-extensively investigated biological activities of RES is its cancer chemopreventive potential, since Pezzuto and co-workers demonstrated its ability to inhibit carcinogenesis at multistep stages.[1g] Taking into account the causative role of oxidative stress by free radicals (or reactive oxygen species) in cancer,[11] we next evaluated systematically the antioxi 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chembiochem.org dant activity of RES and its metabolites in a set of assays: ferric reducing antioxidant power (FRAP), 2,2-diphenyl-1-picrylhydrazyl free radical (DPPHC)-scavenging activity, inhibition of DNA strand breakages, oxidation hemolysis of human red blood cells (RBCs) induced by 2,2’-azobis(2-amidinopropane hydrochloride, AAPH), and protection of human low-density lipoprotein (LDL) oxidation by cupric ions. FRAP is a classic method to test the electron-donating ability of antioxidants by measuring the reduction of a ferric tripyridyltriazine FeIII(tptz)2 complex to the ferrous tripyridyltriazine FeII(tptz)2.[12] The results of FRAP analysis of RES and its metabolites are expressed as the number of donated electrons per molecule (n, Table 1). Based on the determined n values, their FRAP ability follows the order of RES > 3-GR > DHR > 4’-GR. In the second position, 3-GR (with the retained 4’-OH) exhibited a electron-donating capability comparable to that of RES (n = 1.06 and 1.45, respectively), thus highlighting the importance of 4’-OH in the electron-donating capability of RES. This result also supports the previous conclusion that 4’-OH is more active than 3-OH and 5-OH in the antioxidant reaction of RES.[5e,g, 13] In contrast, impairment of the conjugated links, such as hydrogenation of the aliphatic double bond, could significantly decrease its electron-donating ability, as exemplified by DHR and RES (n = 0.55 and 1.45, respectively). This is in agreement with our previous finding that extension by conjugation for the stilbene scaffold of RES is an efficient strategy to improve its antioxidant activity.[5d,e,f] Next, we used the DPPHC-scavenging method to estimate the hydrogen-transfer ability of RES and its metabolites by determination of EC50 (concentration to eliminate 50 % of the radical). Among the three metabolites, only 3-GR retained some activity in this assay, but potency was reduced approximately 2.3-fold relative to RES, which is in line with the FRAP data. Because DNA is a primary target for free-radical oxidative damage, we also employed the AAPH-induced DNA strandbreakage model to probe the compounds’ activities against DNA oxidative damage, by using agarose gel electrophoresis analysis. Upon addition of 5 mm AAPH, supercoiled pBR322 plasmid DNA was completely converted into its circular and linear forms, indicative of single- and double-strand breakages, respectively (lane 2 in Figure 6 A). Notably, all the compounds exhibited significant protection against DNA strand breakage, with an activity order of RES 3-GR > DHR 4’-GR (Figure 6 B). The nearly equal activities of 3-GR and RES again suggests an indispensable role of 4’-OH in antioxidant reactions. The above three models are essentially homogeneoussystem activity tests; lipids are another target for free radicalmediated injury, so, the AAPH-induced RBC hemolysis model was applied, to determine antioxidant effect of RES and its metabolites in a heterogeneous environment. As RBC membranes are rich in polyunsaturated fatty acids, free-radical-mediated lipid peroxidation causes fissure of membranes and ultimately leads to leakage of hemoglobin (hemolysis); the extent is easily detected by measuring the absorbance of hemolysate at 540 nm.[14] When 50 mm AAPH was added to an aqueous suspensions of RBCs and incubated at 37 8C, it induced fast hemolysis (88 min; line a in Figure 7) in the presence of the enChemBioChem 2013, 14, 1094 – 1104

1098


www.chembiochem.org

Figure 5. Docking pose of RES and its metabolites with HSA. A) RES; B) DHR; C) 3-GR; D) 4’-GR. Residues of interest and compounds are represented as sticks. Protein chains are truncated for clarity.

Figure 7. Inhibitory effect of RES and its metabolites (40 mm) against 50 mm AAPH-induced hemolysis of 5 % human RBCs in 0.10 mm PBS (pH 7.4) under an aerobic atmosphere at 37 8C. a) AAPH alone; b) AAPH + 4’-GR; c) AAPH + 3-GR; d) AAPH + DHR; e) AAPH + RES. Data are expressed as the mean of three RBC samples.

Figure 6. Protection of RES and its metabolites (20 mm) against the 5 mm AAPH-induced strand breakage of plasmid pBR322 DNA (100 ng/25 mL) in PBS (pH 7.4) at 37 8C for 60 min. A) Agarose gel electrophoresis: lane 1, control; lane 2, AAPH alone; lanes 3–6, AAPH + RES, AAPH + DHR, AAPH + 3-GR, and AAPH + 4’-GR, respectively. B) Quantitative analysis of this protection. DNA damage is represented as percentage of supercoiled DNA relative to native DNA.


dogenous antioxidants in RBC membranes.[15] However, addition of 40 mm RES (or its metabolites) prolonged the inhibition time (i.e., significantly delayed hemolysis; lines b–e). The antihemolysis efficacy (“effective inhibition”, teff) was determined as the difference between two time-points (teff = t tinh, where tinh is the hemolysis time in the absence of exogenous inhibitor). ChemBioChem 2013, 14, 1094 – 1104

1099

CHEMBIOCHEM FULL PAPERS According to the teff values (Table 1), the efficacy of the compounds decreases in the order RES > DHR > 3-GR 4’-GR, and there is a clear positive correlation between this activity order and the lipophilicity of the compounds as determined by their octanol/water partition coefficients (C logP) by using Bio-Loom software (BioByte, Claremont, CA).[16] However, it should be pointed out that the differences in activity among the compounds in this assay were small; for instance, only a 1.4-fold difference existed between the best and worst (RES and 4’-GR). This result shows that the glucuronidation and reduction metabolites of RES remain active and contribute significantly to the antioxidant activity of the parent molecule in plasma. This is of interest given that the level of free-form RES (relative to those of its metabolites) is very low in human plasma.[2j] Converging evidence supports the theory that free-radicalinduced peroxidation of LDL critically contributes to the risk of human antherosclerosis.[15, 17] Thus, LDL oxidation mediated by cupric ions was selected as the last model for antioxidant activity testing, to measure protective effect of RES and its metabolites in heterogeneous media by monitoring the formation of malondialdehyde as a thiobarbituric acid reactive substance (TBARS). RES was the most active, followed by DHR; 3-GR and 4’-GR were inactive (Figure 8).

Figure 8. Effects of RES and its metabolites (100 mm) on oxidative damage of LDL (1 mg mL 1) by cupric ions (10 mm). Each value of malondialdehyde (MDA) represents the mean of triplicate determination.

Notably, 3-GR was a better antioxidant than 4’-GR in both FRAP and DPPH assays, yet these metabolites showed similar (minimal) effect on LDL oxidation. As mentioned above, the FRAP and DPPH assays assess the electron-donating and formal hydrogen-transfer abilities of the compounds, respectively; 4’-OH of RES plays a more important role than 3-OH and 5-OH, as characterized by ionization potential (IP) and bond dissociation energy (BDE) of the phenolic OH.[13a] Theoretical calculations showed that the IP and BDE for 4-hydroxytrans-stilbene are 156.78 and 80.59 kcal mol 1, respectively, which are lower than those (162.49 and 86.03 kcal mol 1) of 3hydroxy-trans-stilbene. However, in the case of LDL, introduction of a glucuronosyl group at either 3- or 4’-position probably decreases the chelation ability of cupric ions, or prevents penetration into LDL and the subsequent reaction with the propagating lipid peroxyl radicals within LDL, thereby resulting in the same inactivity of 3-GR and 4’-GR toward LDL oxidation. 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chembiochem.org Anti-inflammatory activity of resveratrol and its glucuronidation and reduction metabolites The relationship between cancer and inflammation is well established,[18] and RES has been demonstrated to modulate a variety of molecular processes in the development of inflammation.[19] One of its targets is cyclooxygenase-2 (COX-2), which is responsible for the production of prostaglandins at the site of inflammation; its activity is increased in malignant tissue.[20] Additionally, aberrant induction of inducible nitric oxide synthase (iNOS) results in excessive production of nitric oxide (NO) and can cause chronic inflammatory diseases and cancer.[21] Consequently, we used inhibition of COX-2 activity and NO production as two key indexes for evaluating the anti-inflammatory activity of RES and its metabolites. COX-2 activity was measured by using screening assay kit that monitors the appearance of oxidized N,N,N’,N’-tetramethyl-p-phenylenediamine (TMPD) at 590 nm.[22] According to the determined IC50 values (Table 1), COX-2 inhibitory ability of RES and its metabolites follows the order of RES 4’-GR > DHR > 3GR. A striking feature of our data is that 4’-GR and RES showed equipotent inhibitory ability toward COX-2 (IC50 = 40.6 and 37.5 mm, respectively). Additionally, 3-GR was 4.3-fold weaker than RES. Similarly, Calamini and co-workers have observed that RES and its 4’-O-sulfate are almost equally effective in COX-2 inhibition.[4c] These results demonstrate the importance of the 3,5-di-OH moiety of RES in COX-2 inhibitory activity, and that its modification (e.g., 3-GR) leads to loss of activity. Notably, substantially different IC50 values for inhibition of COX-2 by RES have been reported (from 0.75 to 85 mm).[1g, 4b,c, 23] These discrepancies can be attributed to differences in kinetic assays. It has been observed that the IC50 values for Celecoxib in COX-2 inhibition vary considerably among different assay methods:[24] values of 0.006, 0.04 and 0.2 mm were obtained by prostaglandin E2-ELISA, peroxidaseTMPD, and oxygen-uptake assays, respectively.[24] In the case of RES, IC50 was about 1 mm based on the prostaglandin E2-ELISA assay,[4b,c, 23a–d] but 45 mm[23e] and 85 mm[1g] by the peroxidaseTMPD method for ovine and human recombinant COX-2, respectively, thus suggesting that the source of the enzyme is another key determinant for IC50. Our result (IC50 = 37.5 mm) is very close to that reported by Johnson and Maddipati (45 mm),[23e] based on the same peroxidase-TMPD method and the same ovine COX-2. Next, the effect of RES and its metabolites on nitrite (an oxidation product of NO) accumulation from lipopolysaccharide (LPS)-activated macrophages (leukaemic monocyte cell line RAW 264.7) was examined by using Griess reagent. For inhibition of NO production (Table 1), RES was still the most potent (IC50 = 46.1 mm) and DHR was somewhat weaker (IC50 = 64.4 mm); this is consistent with results previously reported by Kageura et al.[25d] Only a small degree of ability was shown by 3-GR and 4’-GR. In previous studies there was some discrepancy in RES IC50 values for inhibiting NO production, depending on cell type, concentration of LPS, and the action time.[25] For example, Matsuda et al.[25d,e] reported an IC50 of 68 mm with 10 mg mL 1 LPSChemBioChem 2013, 14, 1094 – 1104

1100

CHEMBIOCHEM FULL PAPERS challenged peritoneal exudate cells for 20 h; Kim et al.[25a] and Djoko et al.[25b] obtained IC50 values of 46.5 and 29.9 mm when using the same RAW 264.7 cells but different concentrations of LPS (1 and 0.1 mg mL 1, respectively) and different action times (24 and 20 h, respectively). In contrast, the study of Cho et al.[25c] revealed a low value (18.5 mm) with the RAW 264.7 cells. Our value (46.1 mm) is consistent with that reported by Kim et al. (46.5 mm),[25a] when using the same cell model, concentration, and LPS action time.

Cytotoxicity of resveratrol and its glucuronidation and reduction metabolites against various cancer cells Finally, we examined the cytotoxic effect of RES and its metabolites against human colon (HCT116), sarcoma (HT1080), lung (NCI-H460), liver (SMMC-7721), and breast carcinoma (MDAMB-231) cells, by using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay.[26] RES exhibited a clear dose-dependent cytotoxic effect for all cell types (Figure 9). It was the most active, followed by DHR, which displayed comparable activity only with SMMC-7721 cells; 3-GR and 4’-GR were inactive. Similarly, it has been demonstrated that, in comparison with RES, its sulfated metabolites show low cytotoxicity against human breast cancer cells.[4b,e] Thus, it is reasonable to conclude that glucuronidation and sulfation of RES reduce its absorption or penetration into cells and, thereby, its cytotoxicity. However, interestingly, RES me-

www.chembiochem.org tabolites showed a similar cytotoxic effect to the parent molecule on Caco-2 cells, by induction of cell cycle arrest.[4a] This suggests that RES metabolites can permeate Caco-2 cells, and are probably converted back to RES, given the relatively high expression of b-glucuronidase[27] in Caco-2 cells relative to other cancer cell lines. Currently, there is no direct evidence that RES metabolites are able to cross the plasma membrane. Nevertheless, the evidence with Caco-2 cells[4a] implies that although glucuronidation and sulfation typically reduce the cell permeability of RES and aid its excretion, its metabolites can still enter cells and probably act as a pool from which free RES can be regenerated. Additionally, it has been reported that the 4’- and 3-O-sulfates of RES can moderately suppress expression of iNOS in LPS-treated RAW 264.7 cells,[4b] and presumably, certain sulfate metabolites can be transported intact into kidney cells by anionic transporters.[28] These data and analysis support the possibility that RES metabolites permeate into certain cell types and thereby gain access to targets, including COX-2 and DNA. Additionally, it should be stressed that in terms of pharmacokinetics studies of RES, the concentrations used in in vitro experiments (10 5–10 4 m, including our experiments) cannot be achieved in vivo. When RES is administered orally at 25 mg, the concentration of the free form in human plasma is less than 5 ng mL 1.[2j] Even with oral administration of 5 g, the peak plasma level of RES is only 539 ng mL 1 (2.4 mm).[2e] However, the peak levels of RES metabolites are much higher.[2e,g,j] For instance, after a moderate consumption of red wine, RES, 3-GR, and 4’-GR have been detected at up to 26, 190, and 2.2 mm, respectively.[2g] Noticeably, the study of Marier et al. showed that oral administration of RES (50 mg kg 1) to rats led to peak plasma levels of the free and glucuronide forms at 6.6 and 105 mm, respectively.[2m] In addition, the concentration of 3-GR in normal colorectal tissue (proximal to tumor and obtained from the right side of the colon) in a colorectal cancer patient who received 0.5 g of RES daily for eight days was reported to be 86 nmol g 1 (~ 86 mm).[2c] Therefore, as far as peak levels are concerned, RES metabolites seem to easily attain higher levels in vivo than the free form. This provides a powerful impetus for our current study. Our in vitro results (and those of others)[4a–c] suggest that RES metabolites might be responsible the in vivo biological activity previously attributed only to the parent molecule. However, in vitro data cannot be simply extrapolated to the in vivo situation. Thus, characterization of the in vivo biological activities of RES metabolites is still needed.

Conclusions

Figure 9. Cytotoxicity of RES and its metabolites against human carcinoma cell lines: A) colon (HCT116), B) sarcoma (HT1080), C) lung (NCI-H460), D) liver (SMMC-7721), and E) breast (MDA-MB-231). a) RES; b) DHR; c) 4’-GR; d) 3-GR.


The established diverse biological activities of RES and its rapid and extensive metabolism prompted our interest in synthesizing its glucuronidation and reduction metabolites and investigating the influence of structural modification on biological activity. Based on the results obtained from this work, the following conclusions can be drawn: 1) binding of 4’-GR and DHR to HSA, inhibition by 3-GR against AAPH-induced DNA strand ChemBioChem 2013, 14, 1094 – 1104

1101

CHEMBIOCHEM FULL PAPERS breakages, and inhibition by 4’-GR of COX-2, are equal to those for RES; 2) electron-donating ability of 3-GR, anti-hemolysis activity of DHR, 3-GR, and 4’-GR, and inhibition by DHR of NO production and SMMC-7721 proliferation are comparable to those of RES; 3) in comparison with RES, 3-GR and DHR retained some degree of activity in the DPPHC-scavenging and cupric ion-initiated oxidation of LDL assays, respectively. The above observations support the view that the glucuronidation and reduction metabolites of RES could be contributing to the in vivo biological activities attributed to the parent molecule.

Experimental Section Materials: DPPHC free radical, pBR322 DNA, AAPH, LPS, HSA, RPMI 1640, and Dulbecco’s modified Eagle’s medium (DMEM) were purchased from Sigma–Aldrich. MTT and 2,4,6-tri(2-pyridyl)-s-triazine (TPTZ) were obtained from Amresco (Solon, OH) and Alfa Aesar (Ward Hill, MA), respectively. Other chemicals were of analytical grade. Synthesis of 3-GR, 4’-GR and DHR: 3-GR and 4’-GR were synthesized by published methods;[6, 29] DHR was prepared by hydrogenation over Pd/C catalysts. Their structures were confirmed by 1H and 13 C NMR spectroscopy and ESI-MS by using Avance 400 (400 MHz) and Esquire 6000 (Bruker Daltonics) spectrometers, respectively. (For synthesis details, 1H and 13C NMR and ESI-MS spectra of 3-GR, 4’-GR, and DHR, see the Supporting Information.) (E)-Resveratrol-3-O-b-d-glucuronide (3-GR): Yellowish solid; m.p. 153–158 8C; 1H NMR (400 MHz, DMSO): d = 3.39–3.21 (m, 3 H; H-2’’, H-3’’, H-4’’), 3.87 (d, J = 9.6 Hz, 1 H; H-5’’), 4.98 (d, J = 7.6 Hz, 1 H; H1’’), 5.24 (d, J = 4.8 Hz, 1 H; -OH), 5.41 (d, J = 5.2 Hz, 1 H; -OH), 6.32 (br s, 1 H; H-4), 6.58 (s, 1 H; H-6), 6.66 (s, 1 H; H-2), 6.87 (d, J = 8.4 Hz, 2 H; H-3’, H-5’), 6.93 (d, J = 16.4 Hz, 1 H; H-8), 7.01 (d, J = 16.4 Hz, 1 H; H-7), 7.40 (d, J = 8.4 Hz, 2 H; H-2’, H-6’), 9.50 (br s, 1 H; phenolic-OH), 9.59 ppm (br s, 1 H; phenolic-OH); 13C NMR (100 MHz, DMSO): d = 71.8 (C-2’’, C-3’’ or C-4’’), 73.4 (C-2’’, C-3’’ or C-4’’), 75.9 (C-2’’, C-3’’ or C-4’’), 76.2 (C-5’’), 100.5 (C-1’’), 103.0 (C-4), 105.2 (C2), 107.6 (C-6), 116.0 (C-3’, 5’), 125.5 (C-7), 128.3 (C-2’, 6’), 128.4 (C8), 129.0 (C-1’), 139.8 (C-1), 157.8 (C-4’), 158.8 (C-3), 158.9 (C-5), 170.6 ppm (C-6’’); MS: m/z 403 [M H] . (E)-Resveratrol-4’-O-b-d-glucuronide (4’-GR): Yellowish solid; m.p. 105–109 8C; 1H NMR (400 MHz, DMSO): d = 3.26–3.42 (m, 3 H; H-2’’, H-3’’, H-4’’), 3.90 (d, J = 9.6 Hz, 1 H; H-5’’), 5.06 (d, J = 7.2 Hz, 1 H; H1’’), 5.21 (br s, 1 H; -OH), 5.42 (d, J = 4.8 Hz, 1 H; -OH), 6.13 (br s,1 H; H-4), 6.41 (d, J = 1.6 Hz, 2 H; H-2, H-6), 6.92 (d, J = 16.4 Hz, 1 H; H-8), 6.98 (d, J = 16.4 Hz, 1 H; H-7), 7.01 (d, J = 8.8 Hz, 2 H; H-3’, H-5’), 7.52 (d, J = 8.8 Hz, 2 H; H-2’, H-6’), 9.20 ppm (s, 2 H; phenolic -OH); 13 C NMR (100 MHz, DMSO): d = 73.1 (C-2’’, C-3’’ or C-4’’), 74.7 (C-2’’, C-3’’ or C-4’’), 76.7 (C-5’’), 77.5 (C-2’’, C-3’’ or C-4’’), 102.6 (C-1’’), 103.1 (C-4), 106.1 (C-2, C-6), 118.2 (C-3’, C-5’), 128.7 (C-2’, C-6’), 128.8 (C-7), 128.9 (C-8), 133.7 (C-1’), 141.1 (C-1), 158.6 (C-4’), 159.8 (C-3, C-5), 171.7 ppm (C-6’’); MS: m/z 403 [M H] . Dihydroresveratrol (DHR): White powder; m.p. 161–164 8C; H NMR (400 MHz, (CD3)2CO): d = 2.67–2.71 (m, 2 H; H-8), 2.74–2.78 (m, 2 H; H-7), 6.19 (t, J = 2.0 Hz, 1 H; H-4), 6.22 (d, J = 2.0 Hz, 2 H; H2, H-6), 6.74 (d, J = 8.4 Hz, 2 H; H-3’, H-5’), 7.04 (d, J = 8.4 Hz, 2 H; H2’, H-6’), 8.05 (s, 2 H; OH-3, OH-5), 8.07 ppm (s, 1 H; OH-4’); 13C NMR (100 MHz, (CD3)2CO): d = 37.1 (C-8), 38.7 (C-7), 100.7 (C-4), 107.4 (C2, C-6), 115.5 (C-3’, C-5’), 129.8 (C-2’, C-6’), 133.2 (C-1’), 144.8 (C-1), 155.9 (C-4’), 158.9 ppm (C-3, C-5); MS: m/z 231 [M+H] + . 1


www.chembiochem.org Isothermal titration calorimetry (ITC): All calorimetric experiments were conducted by using a MicroCal iTC200 instrument (GE Healthcare) at 25 8C, with RES or its metabolites in the syringe (titrant) and protein in the cell (titrand). Phosphate buffer (10 mm, pH 7.4) was degassed by filtration (pore size 0.22 mm) before use. The ITC experiment using a 40 mL automatic rotating syringe stirring (1000 rpm, 19 2 mL injections, 120 s (3-GR, 4’-GR, RES) or 180 s (DHR) intervals). Origin 7.0 software provided with the MicroCal instrument was used to acquire and process ITC data. The dissociation constants (Kd), enthalpy (DH), and entropy (DS) were obtained from curve fitting, with the stoichiometry setting at n = 1 for c < 1 (c = nKd 1 [protein]). Saturation transfer difference spectrum (STD NMR): NMR spectra were recorded on an Avance 400 (Bruker) at 25 8C. Spectra processing was performed with topspin 2.1 software (Bruker). Selective saturation of HSA was achieved by a train of Gaussian-shaped pulses (1 % truncation, 50 ms each, 1 ms separation, total presaturation time 3 s). The protein signals were suppressed by application of a 30 ms spin-lock pulse. On-resonance and off-resonance for HSA irradiation were performed at 0 and 17 ppm, respectively. Subtraction spectra were performed internally by phase cycling after every scan. Molecular docking: The crystal structure of HSA (PDB ID: 1H9Z) in complex with warfarin was downloaded from the RCSB Protein Data Bank (http://www.rcsb.org).[30] Molecular docking was performed by using AutoDock 4.2 (http://autodock.scripps.edu), and the Lamarckian genetic algorithm (LGA) was used as the search tool.[31] In our docking calculation, the grid maps were constructed with 60 60 60 points and a grid spacing of 0.375 (roughly a quarter of the length of a carbon–carbon single bond). The docking parameters were as follow:[31a] LGA population size 100, number of energy evaluations 2.5 million, all other parameters as default.[31b] LIGPLOT,[32] a program to automatically plot protein–ligand interactions, was used to calculate the hydrogen-bond and hydrophobic interactions between HSA and RES or its metabolites. Visualization of the docked pose was rendered by PyMOL 0.99 (http://www. pymol.org/). Assay for ferric reducing antioxidant power (FRAP): FRAP assays were used to evaluate the reducing capability of RES and its metabolites, according to the procedure of Benzie and Strain;[12] details are described in our previous report.[5e] DPPHC -scavenging assay: The EC50 values of RES and its metabolites in the scavenging of DPPHC were determined by monitoring the absorbance of DPPHC (60 mm) at 517 nm in methanol, according to the method of Gaulejac[33] with minor changes. A TU-1901 UV/ Vis spectrometer (Beijing Purkinje, Beijing, China) was used after the solution had been allowed to stand 120 min in the dark. Assay for oxidative DNA strand breakages induced by AAPH: Inhibition by RES or its metabolites against the AAPH-induced strand breakages of supercoiled pBR322 plasmid DNA, was assessed by measuring its conversion into the circular and linear forms as described previously.[34] Assay for hemolysis of RBCs: Human RBCs were provided by the Red Cross Center for Blood (Gansu, China). The extent of hemolysis was determined spectrophotometrically as described previously.[34] RES or its metabolites was dissolved in dimethyl sulfoxide (DMSO) and added to a suspension RBCs (5 %) and incubated (5 min) before addition of AAPH. The final concentration of DMSO was less than 0.1 % (v/v). ChemBioChem 2013, 14, 1094 – 1104

1102

CHEMBIOCHEM FULL PAPERS Cu2 + -induced LDL oxidation and TBARS assay: LDL was separated from the plasma of healthy volunteers by the discontinuous density gradient centrifugation procedure, as previously described.[35] Before oxidation, LDL was dialyzed against PBS several times (4 8C) to remove EDTA, and the concentration of LDL protein was determined by using a bicinchoninic acid (BCA) assay.[36] RES or its metabolites (100 mm) was added to an LDL sample (final protein concentration 1 mg mL 1), then the mixture was incubated with CuSO4 (10 mm) in a shaking water bath (37 8C). After 7 h, the oxidation was terminated by addition of excessive EDTA to the solution, which was then treated with a twofold volume of TBA reagent (TBA (0.4 %), SDS (0.5 %), acetic acid (9.4 %)) in a boilingwater bath for 1 h. After cooling, the precipitate was removed by centrifugation (8000 g, 10 min), the amount of malondialdehyde (TBARS) in the supernatant was monitored by using an Infinite 200 multifunctional microplate reader (532 nm; Tecan, Mnnedorf, Switzerland), and a standard curve of malonaldehyde was plotted according to the procedure of Esterbauer and Cheeseman.[37] Measurement of nitrite: Nitrite production (an indicator of NO synthesis) was measured in the supernatant of cultured macrophages, as described previously.[38] Briefly, murine RAW 264.7 macrophage cells (1 105 cells per well) in DMEM with FBS (10 %) and NaHCO3 (3.7 g L 1) were seeded in 96-well plates under an atmosphere of 5 % CO2 at 37 8C. After 24 h, cells were treated with LPS (1 mg mL 1) in the absence or presence of compounds at various concentrations for 24 h. All compounds were dissolved in DMSO, and the final concentration of DMSO was adjusted to 0.1 % (v/v). NO concentration was determined by measuring the accumulation of nitrite (an oxidation product of NO) in the RAW 264.7 cell-culture supernatant. Supernatant medium (75 mL) was mixed (1:1, v/v) with Griess reagent (75 mL, sulfanilamide (1 %) in phosphoric acid (5 %) and N-(1-naphthyl)ethylenedimine dihydrochloride (0.1 %) in distilled water) in the 96-well plates and incubated at room temperature for 10 min. The optical density was measured at 570 nm in an Infinite 200 microplate reader. The inhibition activity of RES and its metabolites against NO production is expressed as IC50 (concentration of test compound to inhibit 50 % of NO release induced by LPS in RAW 264.7 cells). Inhibition of COX-2 enzymatic activity: COX-2 activity was evaluated using colorimetric COX (ovine) inhibitor screening assay kit (#760111; Cayman Chemical, Cleveland, OH) according to manufacturer’s instruction. The peroxidase activity was assayed colorimetrically by monitoring the appearance of oxidized N,N,N’,N’-tetramethyl-p-phenylenediamine (TMPD) at 590 nm.[22] Assay buffer (150 mL), heme (10 mL), COX-2 (10 mL) and inhibitor solutions (10 mL) were added to each well. The mixture was shaken for a few seconds and preincubated (25 8C, 5 min). Colorimetric substrate solution (20 mL) was added, and the reaction was started by addition of arachidonic acid (20 mL). After carefully shaking the plate for a few seconds and incubating (25 8C, 5 min), the absorbance at 590 nm was measured with a multifunctional microplate reader (Tecan infinite 200). All test and control (without tested compounds) assays were corrected for nonenzymic hydrolysis by use of blanks. All data are the arithmetic mean of triplicate determinations. Cell culture: Human colon (HCT116), sarcoma (HT1080), lung (NCIH460), liver (SMMC-7721), and breast (MDA-MB-231) carcinoma cell lines were purchased from the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, and incubated in a humidified atmosphere (37 8C, 5 % CO2/95 % air). Cell lines were grown in RPMI 1640 medium supplemented with heat-inactivated 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chembiochem.org FBS (10 %), penicillin (100 U mL 1), and streptomycin (100 mg mL 1). Exponentially growing cells were used throughout. Assay for cytotoxicity: Compounds were dissolved in DMSO to the desired stock solution concentrations and stored at 20 8C. Added DMSO in cell suspension culture medium was less than 0.1 % (v/v). The effects of RES and its metabolites on in vitro growth of cell lines was evaluated by the MTT colorimetric assay.[26] Cells were seeded in 96-well plates (2000 cells per well), allowed to attach to the plate (24 h), and then treated with various amounts of RES or its metabolites (72 h; three replicates). Then, MTT solution was added to each well (final protein concentration 0.5 mg mL 1) and plates were incubated (4 h, 37 8C). Absorbance was measured at 570 nm in an M680 microplate reader (Bio-Rad, Hercules, CA). Percentage cell viability was calculated relative to control wells (100 %).

Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant Nos. 20972063, 21172101 and 31100607), the 111 Project, Program for New Century Excellent Talents in University (NCET-06-0906) and the Fundamental Research Funds for the Central Universities. Keywords: activity · reduction · resveratrol

glucuronidation

·

metabolism

·

[1] a) S. Quideau, D. Deffieux, L. Pouysgu, Angew. Chem. 2012, 124, 6930 – 6932; Angew. Chem. Int. Ed. 2012, 51, 6824 – 6826; b) S.-J. Park, F. Ahmad, A. Philp, K. Baar, T. Williams, H. Luo, H. Ke, H. Rehmann, R. Taussig, A. L. Brown, M. K. Kim, M. A. Beaven, A. B. Burgin, V. Manganiello, J. H. Chung, Cell 2012, 148, 421 – 433; c) P. Saiko, A. Szakmary, W. Jaeger, T. Szekeres, Mutat. Res. 2008, 658, 68 – 94; d) J. M. Pezzuto, J. Agric. Food Chem. 2008, 56, 6777 – 6784; e) J. A. Baur, D. A. Sinclair, Nat. Rev. Drug Discovery 2006, 5, 493 – 506; f) K. T. Howitz, K. J. Bitterman, H. Y. Cohen, D. W. Lamming, S. Lavu, J. G. Wood, R. E. Zipkin, P. Chung, A. Kisielewski, L.-L. Zhang, B. Scherer, D. A. Sinclair, Nature 2003, 425, 191 – 196; g) M. Jang, L. Cai, G. O. Udeani, K. V. Slowing, C. F. Thomas, C. W. W. Beecher, H. H. S. Fong, N. R. Farnsworth, A. D. Kinghorn, R. G. Mehta, R. C. Moon, J. M. Pezzuto, Science 1997, 275, 218 – 220; h) E. N. Frankel, A. L. Waterhouse, J. E. Kinsella, Lancet 1993, 341, 1103 – 1104. [2] a) A. Amri, J. C. Chaumeil, S. Sfar, C. Charrueau, J. Controlled Release 2012, 158, 182 – 193; b) C.-H. Cottart, V. Nivet-Antoine, C. Laguillier-Morizot, J.-L. Beaudeux, Mol. Nutr. Food Res. 2010, 54, 7 – 16; c) K. R. Patel, V. A. Brown, D. J. L. Jones, R. G. Britton, D. Hemingway, A. S. Miller, K. P. West, T. D. Booth, M. Perloff, J. A. Crowell, D. E. Brenner, W. P. Steward, A. J. Gescher, K. Brown, Cancer Res. 2010, 70, 7392 – 7399; d) A. Burkon, V. Somoza, Mol. Nutr. Food Res. 2008, 52, 549 – 557; e) D. J. Boocock, G. E. S. Faust, K. R. Patel, A. M. Schinas, V. A. Brown, M. P. Ducharme, T. D. Booth, J. A. Crowell, M. Perloff, A. J. Gescher, W. P. Steward, D. E. Brenner, Cancer Epidemiol. Biomarkers Prev. 2007, 16, 1246 – 1252; f) D. J. Boocock, K. R. Patel, G. E. S. Faust, D. P. Normolle, T. H. Marczylo, J. A. Crowell, D. E. Brenner, T. D. Booth, A. Gescher, W. P. Steward, J. Chromatogr. B 2007, 848, 182 – 187; g) P. Vitaglione, S. Sforza, G. Galaverna, C. Ghidini, N. Caporaso, P. P. Vescovi, V. Fogliano, R. Marchelli, Mol. Nutr. Food Res. 2005, 49, 495 – 504; h) E. Wenzel, T. Soldo, H. Erbersdobler, V. Somoza, Mol. Nutr. Food Res. 2005, 49, 482 – 494; i) D. Wang, T. Hang, C. Wu, W. Liu, J. Chromatogr. B 2005, 829, 97 – 106; j) T. Walle, F. Hsieh, M. H. DeLegge, J. E. Oatis, Jr., U. K. Walle, Drug Metab. Dispos. 2004, 32, 1377 – 1382; k) A. LanÅon, D. Delmas, H. Osman, J.-P. Thnot, B. Jannin, N. Latruffe, Biochem. Biophys. Res. Commun. 2004, 316, 1132 – 1137; l) M. I. Kaldas, U. K. Walle, T. Walle, J. Pharm. Pharmacol. 2003, 55, 307 – 312; m) J.-F. Marier, P. Vachon, A. Gritsas, J. Zhang, J.-P. Moreau, M. P. Ducharme, J. Pharmacol. Exp. Ther. 2002, 302, 369 – 373.

ChemBioChem 2013, 14, 1094 – 1104

1103

CHEMBIOCHEM FULL PAPERS [3] a) R. Corder, A. Crozier, P. A. Kroon, Nature 2003, 426, 119; b) A. J. Gescher, W. P. Steward, Cancer Epidemiol. Biomarkers Prev. 2003, 12, 953 – 957. [4] a) C. E. Storniolo, J. J. Moreno, Food Chem. 2012, 134, 1385 – 1391; b) J. Hoshino, E.-J. Park, T. P. Kondratyuk, L. Marler, J. M. Pezzuto, R. B. van Breemen, S. Mo, Y. Li, M. Cushman, J. Med. Chem. 2010, 53, 5033 – 5043; c) B. Calamini, K. Ratia, M. G. Malkowski, M. Cuendet, J. M. Pezzuto, B. D. Santarsiero, A. D. Mesecar, Biochem. J. 2010, 429, 273 – 282; d) A. A. Gakh, N. Y. Anisimova, M. V. Kiselevsky, S. V. Sadovnikov, I. N. Stankov, M. V. Yudin, K. A. Rufanov, M. Y. Krasavin, A. V. Sosnov, Bioorg. Med. Chem. Lett. 2010, 20, 6149 – 6151; e) M. Miksits, K. Wlcek, M. Svoboda, O. Kunert, E. Haslinger, T. Thalhammer, T. Szekeres, W. Jger, Planta Med. 2009, 75, 1227 – 1230; f) L.-X. Wang, A. Heredia, H. Song, Z. Zhang, B. Yu, C. Davis, R. Redfield, J. Pharm. Sci. 2004, 93, 2448 – 2457. [5] a) G.-Y. Liu, J. Yang, F. Dai, W.-J. Yan, Q. Wang, X.-Z. Li, D.-J. Ding, X.-Y. Cao, B. Zhou, Chem. Eur. J. 2012, 18, 11100 – 11106; b) X.-Y. Cao, J. Yang, F. Dai, D.-J. Ding, Y.-F. Kang, F. Wang, X.-Z. Li, G.-Y. Liu, S.-S. Yu, X.-L. Jin, B. Zhou, Chem. Eur. J. 2012, 18, 5898 – 5905; c) X.-Z. Li, X. Wei, C.-J. Zhang, X.-L. Jin, J.-J. Tang, G.-J. Fan, B. Zhou, Food Chem. 2012, 135, 1239 – 1244; d) D.-J. Ding, X.-Y. Cao, F. Dai, X.-Z. Li, G.-Y. Liu, D. Lin, X. Fu, X.-L. Jin, B. Zhou, Food Chem. 2012, 135, 1011 – 1019; e) J.-J. Tang, G.-J. Fan, F. Dai, D.-J. Ding, Q. Wang, D.-L. Lu, R.-R. Li, X.-Z. Li, L.-M. Hu, X.-L. Jin, B. Zhou, Free Radical Biol. Med. 2011, 50, 1447 – 1457; f) J. Yang, G.-Y. Liu, D.-L. Lu, F. Dai, Y.-P. Qian, X.-L. Jin, B. Zhou, Chem. Eur. J. 2010, 16, 12808 – 12813; g) Y.-J. Shang, Y.-P. Qian, X.-D. Liu, F. Dai, X.-L. Shang, W.Q. Jia, Q. Liu, J.-G. Fang, B. Zhou, J. Org. Chem. 2009, 74, 5025 – 5031. [6] a) D. A. Learmonth, Bioconjugate Chem. 2003, 14, 262 – 267; b) R. Lucas, D. Alcantara, J. C. Morales, Carbohydr. Res. 2009, 344, 1340 – 1346. [7] a) S. Naveenraj, S. Anandan, J. Photochem. Photobiol. C 2013, 14, 53 – 71, and references cited therein; b) V. T. C. Chuang, M. Otagiri, Chirality 2006, 18, 159 – 166. [8] a) Y. L. Jiang, Bioorg. Med. Chem. 2008, 16, 6406 – 6414; b) Z. Lu, Y. Zhang, H. Liu, J. Yuan, Z. Zheng, G. Zou, J. Fluoresc. 2007, 17, 580 – 587; c) C. N. N’soukpo-Kossi, C. St-Louis, M. Beauregard, M. Subirade, R. Carpentier, S. Hotchandani, H. A. Tajmir-Riahi, J. Biomol. Struct. Dyn. 2006, 24, 277 – 283. [9] P. D. Ross, S. Subramanian, Biochemistry 1981, 20, 3096 – 3102. [10] a) A. Viegas, J. Manso, F. L. Nobrega, E. J. Cabrita, J. Chem. Educ. 2011, 88, 990 – 994; b) B. Meyer, T. Peters, Angew. Chem. 2003, 115, 890 – 918; Angew. Chem. Int. Ed. 2003, 42, 864 – 890; c) M. Mayer, B. Meyer, Angew. Chem. 1999, 111, 1902 – 1906; Angew. Chem. Int. Ed. 1999, 38, 1784 – 1788. [11] a) J. E. Klaunig, L. M. Kamendulis, Annu. Rev. Pharmacol. Toxicol. 2004, 44, 239 – 267; b) S. P. Hussain, L. J. Hofseth, C. C. Harris, Nat. Rev. Cancer 2003, 3, 276 – 285. [12] I. F. F. Benzie, J. J. Strain, Anal. Biochem. 1996, 239, 70 – 76. [13] a) A. N. Queiroz, B. A. Q. Gomes, W. M. Moraes Jr, R. S. Borges, Eur. J. Med. Chem. 2009, 44, 1644 – 1649; b) F. Caruso, J. Tanski, A. VillegasEstrada, M. Rossi, J. Agric. Food Chem. 2004, 52, 7279 – 7285; c) H. Cao, X. Pan, C. Li, C. Zhou, F. Deng, T. Li, Bioorg. Med. Chem. Lett. 2003, 13, 1869 – 1871; d) S. Stojanovic´, O. Brede, Phys. Chem. Chem. Phys. 2002, 4, 757 – 764; e) S. Stojanovic´, H. Sprinz, O. Brede, Arch. Biochem. Biophys. 2001, 391, 79 – 89; f) M. Wang, Y. Jin, C.-T. Ho, J. Agric. Food Chem. 1999, 47, 3974 – 3977. [14] E. Niki, E. Komuro, M. Takahashi, S. Urano, E. Ito, K. Terao, J. Biol. Chem. 1988, 263, 19809 – 19814. [15] H. Esterbauer, P. Ramos, Rev. Physiol. Biochem. Pharmacol. 1996, 127, 31 – 64. [16] a) C. D. Selassie, S. Kapur, R. P. Verma, M. Rosario, J. Med. Chem. 2005, 48, 7234 – 7242; b) C. Hansch, A. Leo, D. H. Hoekman, Exploring QSAR: Fundamentals and Applications in Chemistry and Biology, American Chemical Society, Washington, D.C., 1995. [17] D. Steinberg, S. Parthasarathy, T. E. Carew, J. C. Khoo, J. L. Witztum, N. Engl. J. Med. 1989, 320, 915 – 924.


www.chembiochem.org [18] L. M. Coussens, Z. Werb, Nature 2002, 420, 860 – 867. [19] C. C. Udenigwe, V. R. Ramprasath, R. E. Aluko, P. J. Jones, Nutr. Rev. 2008, 66, 445 – 454, and references therein. [20] J. R. Vane, Y. S. Bakhle, R. M. Botting, Annu. Rev. Pharmacol. Toxicol. 1998, 38, 97 – 120. [21] H. Kleinert, A. Pautz, K. Linker, P. M. Schwarz, Eur. J. Pharmacol. 2004, 500, 255 – 266. [22] R. J. Kulmacz, W. E. M. Lands, Prostaglandins 1983, 25, 531 – 540. [23] a) S. S. Kang, M. Cuendet, D. C. Endringer, V. L. Croy, J. M. Pezzuto, M. A. Lipton, Bioorg. Med. Chem. 2009, 17, 1044 – 1054; b) N. Handler, G. Brunhofer, C. Studenik, K. Leisser, W. Jaeger, S. Parth, T. Erker, Bioorg. Med. Chem. 2007, 15, 6109 – 6118; c) M. Murias, N. Handler, T. Erker, K. Pleban, G. Ecker, P. Saiko, T. Szekeres, W. Jger, Bioorg. Med. Chem. 2004, 12, 5571 – 5578; d) A. M. Rimando, M. Cuendet, C. Desmarchelier, R. G. Mehta, J. M. Pezzuto, S. O. Duke, J. Agric. Food Chem. 2002, 50, 3453 – 3457; e) J. L. Johnson, K. R. Maddipati, Prostaglandins Other Lipid Mediators 1998, 56, 131 – 143. [24] J. K. Gierse, C. M. Koboldt, M. C. Walker, K. Seibert, P. C. Isakson, Biochem. J. 1999, 339, 607 – 614. [25] a) M. H. Kim, J. S. Shin, K. T. Lee, Y. S. Lee, Bull. Korean Chem. Soc. 2011, 32, 299 – 302; b) B. Djoko, R. Y.-Y. Chiou, J.-J. Shee, Y.-W. Liu, J. Agric. Food Chem. 2007, 55, 2376 – 2383; c) D.-I. Cho, N.-Y. Koo, W. J. Chung, T.S. Kim, S. Y. Ryu, S. Y. Im, K.-M. Kim, Life Sci. 2002, 71, 2071 – 2082; d) T. Kageura, H. Matsuda, T. Morikawa, I. Toguchida, S. Harima, M. Oda, M, Yoshikawa, Bioorg. Med. Chem. 2001, 9, 1887 – 1893; e) H. Matsuda, T. Kageura, T. Morikawa, I. Toguchida, S. Harima, M. Yoshikawa, Bioorg. Med. Chem. Lett. 2000, 10, 323 – 327. [26] R. F. Hussain, A. M. E. Nouri, R. T. D. Oliver, J. Immunol. Methods 1993, 160, 89 – 96. [27] a) B. Sperker, C. Tomkiewicz, O. Burk, R. Barouki, H. K. Kroemer, Mol. Pharmacol. 2001, 59, 177 – 182; b) K. Shimoi, T. Nakayama, Methods Enzymol. 2005, 400, 263 – 272. [28] a) N. Mizuno, T. Takahashi, Y. Iwase, H. Kusuhara, T. Niwa, Y. Sugiyama, Drug Metab. Dispos. 2007, 35, 1429 – 1434; b) J.-O. Kwak, H.-W. Kim, K.-J. Oh, C.-B. Ko, H. Park, S. H. Cha, J. Steroid Biochem. Mol. Biol. 2005, 97, 369 – 375. [29] Z. Zhang, B. Yu, R. R. Schmidt, Synthesis 2006, 1301 – 1306. [30] I. Petitpas, A. A. Bhattacharya, S. Twine, M. East, S. Curry, J. Biol. Chem. 2001, 276, 22804 – 22809. [31] a) G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart, R. K. Belew, A. J. Olson, J. Comput. Chem. 1998, 19, 1639 – 1662; b) R. Huey, G. M. Morris, A. J. Olson, D. S. Goodsell, J. Comput. Chem. 2007, 28, 1145 – 1152. [32] a) A. C. Wallace, R. A. Laskowski, J. M. Thornton, Protein Eng. 1995, 8, 127 – 134; b) R. A. Laskowski, M. B. Swindells, J. Chem. Inf. Model. 2011, 51, 2778 – 2786. [33] N. Saint-Cricq de Gaulejac, C. Provost, N. Vivas, J. Agric. Food Chem. 1999, 47, 425 – 431. [34] Y.-P. Qian, Y.-J. Shang, Q.-F. Teng, J. Chang, G.-J. Fan, X. Wei, R.-R. Li, H.-P. Li, X.-J. Yao, F. Dai, B. Zhou, Food Chem. 2011, 126, 241 – 248. [35] B. H. Chung, T. Wilkinson, J. C. Geer, J. P. Segrest, J. Lipid Res. 1980, 21, 284 – 291. [36] P. K. Smith, R. I. Krohn, G. T. Hermanson, A. K. Mallia, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson, D. C. Klenk, Anal. Biochem. 1985, 150, 76 – 85. [37] H. Esterbauer, K. H. Cheeseman, Methods Enzymol. 1990, 186, 407 – 421. [38] a) E.-M. Choi, J.-K. Hwang, Fitoterapia 2005, 76, 194 – 203; b) C.-L. Chang, L.-J. Zhang, R. Y. Chen, L.-M. Y. Kuo, J.-P. Huang, H.-C. Huang, K.-H. Lee, Y.-C. Wu, Y.-H. Kuo, J. Nat. Prod. 2010, 73, 1482 – 1488.

Received: February 15, 2013 Published online on May 23, 2013

ChemBioChem 2013, 14, 1094 – 1104

1104