Resveratrol-Related Polymethoxystilbene Glycosides: Synthesis

Nov 5, 2015 - ... New Oxadiazole-Stilbene Hybrids against Phytopathogenic Fungi. Weilin Jian , Daohang He , Shaoyun Song. Scientific Reports 2016 6 (1...
1 downloads 0 Views 723KB Size
Download PDF
Article pubs.acs.org/jnp

Resveratrol-Related Polymethoxystilbene Glycosides: Synthesis, Antiproliferative Activity, and Glycosidase Inhibition Nunzio Cardullo,† Carmela Spatafora,*,† Nicolò Musso,‡ Vincenza Barresi,‡ Daniele Condorelli,‡ and Corrado Tringali† †

Dipartimento di Scienze Chimiche and ‡Dipartimento di Scienze Bio-Mediche, Sezione di Biochimica, Università degli Studi di Catania, Viale A. Doria 6, I-95125 Catania, Italy S Supporting Information *

ABSTRACT: A small library of polymethoxystilbene glycosides (20−25) related to the natural polyphenol resveratrol have been synthesized and subjected, together with their aglycones 17−19, to an antiproliferative activity bioassay toward Caco-2 and SHSY5Y cancer cells. Six of the compounds exhibit antiproliferative activity against at least one cell line. In particular, compounds 17 and 18 proved highly active on at least one of the two cell cultures. Compound 18 showed a GI50 value of 3 μM against Caco-2 cells, a value comparable to that of the anticancer drug 5-fluorouracil. The closely related compound 19 proved inactive, and its conjugates 22 and 25 showed weak cell growth inhibition. The results indicate that minimal differences in the structure of both polymethoxystilbenes and their glycosides can substantially affect the antiproliferative activity. The possible hydrolytic release of the aglycones 17−19 by β-glucosidase or β-galactosidase was also evaluated. Compounds 20−25 were also tested as potential βglucosidase, β-galactosidase, and α-glucosidase inhibitors. A promising inhibitory activity toward α-glucosidase was observed for 21 (IC50 = 78 μM) and 25 (IC50 = 70 μM), which might be indicative of their potential as lead compounds for development of antidiabetic or antiobesity agents.

S

better oral bioavailability, longer half-life, greater plasma exposure, and lower clearance than resveratrol.16 Polymethoxystilbenes are frequently much more antiproliferative than resveratrol14,17 and show other promising properties such as antiangiogenic activity and VEGF inhibition,9,15,18 inhibition of NF-kB activation,19 and inhibition of multidrug resistance (MDR) of tumor cells.20 Further studies indicate that 2 and other polymethoxystilbenes are able to interact with biomembrane models21−23 and have a higher bioavailability than resveratrol.24−27 Among the methoxylated stilbenes and related compounds, some cis-isomers have shown higher antiproliferative28 or antimetastatic14 activity than their trans-isomers, but they are prone to isomerization during storage and administration and during metabolism in liver microsomes.29 We expanded our studies on polymethoxystilbenes by exploring the antiproliferative properties of some of their glycosides, due to the fact that biological data on these glycosides have not been reported, although many natural

tilbenoids are an important family of natural products sharing with flavonoids shikimate as common biosynthetic precursor.1 Probably the most popular polyphenolic stilbenoid is resveratrol (1, Figure 1). This compound, originally identified as a phytoalexin produced by Vitis vinifera, has subsequently been found in red wine and in other plants and became an outstanding natural product for the high number of studies indicating 1 as possessing cardioprotective, cancer chemopreventive, antioxidative, and many other beneficial properties.2,3 As a continuation of our long-standing interest toward bioactive natural products and especially polyphenols,4,5 we have devoted part of our research activities to analogues and derivatives of resveratrol.6−9 Various in vivo studies have shown that 1 is impaired by a short half-life, rapid clearance, and low bioavailability, being rapidly metabolized.10,11 Thus, efforts have been made to obtain synthetic analogues with an increased metabolic stability and potentially enhanced antitumor activity. A simple resveratrol analogue, 3,5,4′-trimethoxystilbene (2), a natural product isolated from Virola elongatae,12 is significantly more potent than 1 in inhibiting growth or motility of human cancer cells7,13,14 and as an antiangiogenic15 agent and showed © XXXX American Chemical Society and American Society of Pharmacognosy

Received: July 15, 2015

A

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 2. Chemical structures of compounds 6 and 7.

subsequent mild regioselective hydroxylation at C-2 by mCPBA afforded the corresponding 2-hydroxypolymethoxystilbenes 17−19. We previously reported their synthesis, where especially 17 showed antiangiogenic activity.9,18 Compound 17 also proved to be an inhibitor of VEGF production in a granulosa cell model; compounds 17−19 were also able to inhibit the proliferation of SW480 colorectal tumor cells,6 and 19 proved more potent than both 17 and 18. The final step was the conjugation of compounds 17−19 via a nucleophilic substitution to the anomeric carbon of an activated sugar. The glycosylation can be troublesome with phenols because of the low reactivity of the phenolic group as a glycosyl acceptor. Glycosyl halides (especially bromides) are the most frequently used carbohydrate donors for aromatic Oglycosylation. Unlike other types of donors, glycosyl halides selectively give the O-glycosylation product with the inversion of anomeric configuration (SN2 mechanism) and with acceptable yields.42 Thus, to obtain, respectively, the 2-β-D-O-glucoand galactopyranosides, tetra-O-acetyl-α-D-glucopyranosyl bromide and tetra-O-acetyl-α-D-galactopyranosyl bromide were employed. In a first attempt, the glycosylation was carried out in dry ethanol at room temperature using basic conditions, but slow reactions, accompanied by the formation of a complex mixture of products, were observed. Better results were obtained in a biphasic system and in the presence of a phase transfer catalyst (tetrabutylammonium chloride), in basic medium. The procedure was further optimized by temperature modulation, and glycosides 20−25 were obtained in 30−46% yield through RP-18 reversed-phase PLC (see Experimental Section). No detectable byproduct was obtained, and the recovered unreacted substrate, when necessary, was recycled. Indeed, under these reaction conditions both glycosylation and acetate hydrolysis were achieved in one step. Compounds 20− 25 were subjected to full structural characterization, and NMR methods (COSY, HSQC, HMBC) allowed unambiguous assignment of all their 1H and 13C NMR signals (see Experimental Section). The glycoconjugates 20−25 were evaluated for antiproliferative activity toward Caco-2 and SH-SY5Y cell lines with the Mosmann bioassay, using 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyl tetrazolium bromide (MTT). The 2-hydroxypolymethoxystilbenes 17−19 were included in the panel, both for comparison with the related conjugates and for a further evaluation of their previously observed antiproliferative activity on SW480 colon carcinoma cells.6 5-Fluorouracil (5-FU) was used as positive control. In Table 1 only results for compounds showing a GI50 value lower than at least 10 μM are included.

Figure 1. Chemical structures of compounds 1−5 and 26.

glycosides display antitumor activity or other promising biological properties.30,31 Bioactive glycostilbenoids have been found in many families of higher plants.30 Among these, resveratroloside (3, Figure 1) showed cytotoxic activity against human tumor cell lines; piceid32 (4, Figure 1) and 3,5,4′trihydroxystilbene-2-O-β-D-glucopyranoside (5, Figure 1) inhibited tumor growth and lung metastasis in mice bearing highly metastatic Lewis lung carcinoma and showed antiangiogenic activity.33 Angiogenesis inhibitors are intensively pursued as a promising approach to cancer therapy.34 Antiangiogenic activity through fibroblast growth factor-2 (FGF-2) inhibition has also been reported for a resveratrol-related natural glycoside bearing a disaccharide moiety.35 The growing attention to glycoconjugates as target-specific antiproliferative agents or effective prodrugs36 boosted our interest further. Indeed, glycoconjugates might target specific saccharide transporters and primarily glucose transporters; however, these cells exhibit an altered glucose metabolism, known as the “Warburg effect”, which has recently been indicated as an emerging hallmark of cancer.37 O-Glycosides should accumulate selectively in tumor cells, thus allowing dose reduction of the drug and consequently its general toxicity.38 As prodrugs, they might be hydrolyzed by specific glycosidases, thereby allowing the release of the active aglycone. In this regard it is worth citing a report on target-specific glycoconjugates evaluated for cytotoxic properties:39 the β-Dgalactoside of the synthetic drug diethylstilbestrol (6, Figure 2) showed an IC50 value of 75 μM on Caco-2 cells; the 4-O-β-Dglucoside of curcumin (7, Figure 2), the pigment found in Curcuma longa,40 proved the most potent, with an IC50 value of 10.3 μM on Caco-2 cells. Further glycosylated resveratrol prodrugs have been reported as anti-inflammatory agents, of which two were effective prodrugs in mice.41 This prompted the syntheses of gluco- and galactopyranosides of three polymethoxystilbenes bearing a 2-OH group, to be exploited as glycosyl acceptors.

■

RESULTS AND DISCUSSION The synthetic procedure for the glycoconjugate polymethoxystilbenes is shown in Scheme 1. The polymethoxystilbenes 11, 15, and 16 were synthesized via Arbuzov rearrangement followed by Horner−Emmons−Wadsworth olefination to afford the trans-isomer with high diastereoselectivity. A B

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Scheme 1. aSynthetic Procedures for Compounds 20−25

(a) P(OCH2CH3)3, 130 °C, 5 h; (b) CH3ONa, DMF, 0 °C, 30 min, rt, 1 h, 100 °C, 1 h, rt, 24 h; (c) m-CPBA, CH2Cl2, rt, 35 min; (d) K2CO3, TBACl, MeOH/H2O/CHCl3, tetra-O-acetyl-α-D-glucopyranosyl bromide, 24 h; (e) K2CO3, TBACl, MeOH/H2O/CHCl3, tetra-O-acetyl-α-Dgalactopyranosyl bromide.

a

Table 1. Antiproliferative Activity of Compounds 17−25a

cytotoxic effect up to the maximum tested concentration of 100 μM) is rather surprising due to its close structural similarity to 18. Hence, the substitution pattern of the ring devoid of the hydroxy group seems to be critical for the activity of these unconjugated stilbenoids. The interpretation of the biological data of the glycosides 20−25 is not straightforward. A reduction of the potency is observed due to conjugation of 17 and 18 with both glucose and galactose: their glycosides 20, 21 and 23, 24 are inactive toward Caco-2 cells, whereas a weak growth inhibition is observed only for 20 and 23 toward SH-SY5Y cells. This suggests that, at least in Caco-2 cells, these glycosides are not effectively taken up or are not adequately hydrolyzed and do not interact with the biological target. Something different is observed for the inactive aglycone 19: its galactoside (25) showed a weak antiproliferative effect toward Caco-2 cells and SH-SY5Y cells. We therefore subjected the glycosides 20−25 to hydrolytic reactions carried out in the presence of β-glucosidase or βgalactosidase, as an in vitro experimental model for a possible hydrolysis by intracellular enzymes (see Experimental Section). We hypothesized that the lack of antiproliferative activity of glucosides and galactosides of the active aglycones 17 and 18 could be due to poor hydrolysis by intracellular enzymes. Figure 3a shows the kinetic profile of compounds 20−25 in the

GI50 (μM) ± SD

b

c

compound

Caco-2

17 18 5-FU

5.8 ± 0.9 3.0 ± 0.4 2.2 ± 0.8

SH-SY5Yd 4.5 ± 0.5 14.3 ± 3.5 1.5 ± 0.1

a

Compounds 19−25 were inactive against both tested cancer cell lines or showed GI50 > 10 μM. bGI50 calculated after 72 h of continuous exposure relative to untreated controls; values are the mean (±SD) of four experiments. cCaco-2: human colon carcinoma. dSH-SY5Y: neuroblastoma.

Six of the tested compounds exhibit growth inhibition against at least one cell line, although the biological response shows significant variations for the different cell lines. In particular, compound 17 proved highly active on both Caco-2 and SHSY5Y cells, and compound 18 was potently active toward Caco2 (GI50 = 3.0 μM), with a GI50 value comparable to that of the anticancer drug 5-FU. Compounds 22 and 25 showed a weak growth inhibition of Caco-2 cells (GI50 = 44.2 and 16.0 μM, respectively), and a similar weak inhibition was observed for compounds 20, 23, and 25 toward SH-SY5Y cells (GI50 = 21.9, 48.4, and 22.5 μM, respectively). Compounds 19, 21, and 24 were inactive on both cell lines. The inactivity of 19 (no C

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

The glucosides 20 and 21 were progressively hydrolyzed by almond glucosidase after a few hours; a slower hydrolysis was observed for 22, with a half-life of 70 h. No significant hydrolysis was observed for the three galactosides 23−25 up to 96 h. Figure 3b shows that compounds 21−25 were not hydrolyzed by β-galactosidase up to 96 h. Compound 20 underwent slow hydrolysis, with a half-life of 77 h. The amount of 17 released from 20 after 72 h of β-galactosidase treatment is less than 50%. In summary, only 20 and 21 were significantly hydrolyzed by β-glucosidase, with a half-life of 11 and 20 h, respectively, much shorter than the 72 h duration of the biological assays. Nevertheless, these compounds are probably not adequately absorbed, with the exception of compound 20 by SH-SY5Y cells. In this latter case even a low uptake if coupled with a high hydrolytic rate by intracellular βglucosidases, affording the active aglycone 17, might justify the observed activity. The lack of hydrolytic effect by β-glucosidase and especially by β-galactosidase for most of the glycosides prompted us to investigate if these compounds might act as β-glucosidase or βgalactosidase inhibitors. This led to enzymatic inhibition assays for both enzymes in the presence of compounds 20−25, as detailed in the Experimental Section. In the analysis of βglycosidase activities p-nitrophenyl-β-glucopyranoside and onitrophenyl-β-galactopyranoside were employed as substrates. The results are summarized in Table 2, where the μM concentration required for 50% inhibition of the glycosidase activity (IC50) is reported.

Figure 3. Hydrolysis of 20−25 at 37 °C (pH = 7.2) in the presence of enzymes: (a) β-glucosidase from almonds, (b) β-galactosidase from A. oryzae. S = substrate.

Table 2. β-Glucosidase, β-Galactosidase, and α-Glucosidase Inhibitory Activities of Compounds 20−25 IC50 (μM) ± SDa

presence of β-glucosidase from almonds, whereas Figure 3b gives the results of the same experiment in the presence of βgalactosidase from Aspergillus oryzae. These results were confirmed by the quantification of the released aglycone, as reported in Figure 4.

β-glucosidase 20 21 22 23 24 25 acarbose

208 192 272 350 170 n.i.

± ± ± ± ±

47 50 16 6 60

β-galactosidase n.d. 258 297 298 299 296

± ± ± ± ±

42 68 68 53 12

α-glucosidase 91 78 133 201 70 332 65

± ± ± ± ± ± ±

17 21 19 29 3 70 9

a

Values are the mean (±SD) of three experiments; n.d.: not determined, n.i.: no inhibition.

Most of the compounds showed moderate inhibitory activity toward both enzymes, with IC50 values in the approximate range 170−350 μM. All the determined IC50 values are substantially similar, and no significant difference was observed between glucosides and galactosides, except for the lack of βglucosidase inhibition by 25 up to 1 × 10−4 M. However, the data are noteworthy, if compared with previously reported IC50 values of other glycosidase inhibitors.43 This prompted the evaluation of these stilbenoid glycosides as α-glucosidase inhibitors, taking into account the growing interest toward new potential antidiabetic agents. We carried out an enzymatic inhibition assay employing p-nitrophenyl-α-glucopyranoside as the substrate. The results are reported in Table 2 as IC50 values (μM). The glycoconjugates 20−24 showed IC50 values in the approximate range 70−200 μM, whereas 25 showed an IC50 value higher than 300 μM. Compounds 20, 21, and 24 (IC50 = 91, 78, 70 μM, respectively) showed a promising inhibitory activity when compared with the antidiabetic drug acarbose (IC50 = 65 μM)44,45 or with many previously reported α-

Figure 4. (a) Percentage of aglycone released by β-glucosidase from 20−25; (b) percentage of aglycone released by β-galactosidase from 20−25. D

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

staining with a solution of cerium sulfate and phosphomolybdic acid followed by heating. The enzymatic reactions were incubated at 37 °C under shaking at 400 rpm; control reactions without enzyme were carried out under the same conditions. The progress of the reactions was monitored at regular time intervals by HPLC using an analytic reversed-phase column (Luna C18 column, 5 μm; 4.6 × 250 mm; Phenomenex) and the following solvent system at 1 mL/min: eluent A: H2O and HCOOH, 99:1 (v/v); eluent B: CH3CN and HCOOH, 99:1 (v/v). The elution gradient had the following profile: t0 min B = 45%, t7 min B = 80%, t10 min B = 90%, t12 min B = 100%. All chemicals were of reagent grade and were used without further purification. All chemicals, β-glucosidase from almonds, β-galactosidase from Aspergillus oryzae, α-glucosidase from Saccharomyces cerevisiae, p-nitrophenyl-β-D-glucopyranoside (pNP-β-Glc), o-nitrophenyl β-D-galactopyranoside (oNP-β-Gal), and p-nitrophenyl-α-Dglucopyranoside (pNP-α-Glc), were purchased from Sigma. Compounds 11 and 15−19 were synthesized as previously described by Spatafora et al.9 The stilbenoid 11 was obtained by reacting the diethyl (4-methoxybenzyl)phosphonate 9 (prepared from 4-methoxybenzyl chloride 8) with the aldehyde 10; analogously, 15 was synthesized by reacting the diethyl (3,5-dimethoxybenzyl)phosphonate 13 (prepared from 3,5-dimethoxybenzyl bromide 12) with the aldehyde 14; analogously, 16 was obtained by reaction of 13 with 10. Finally, compounds 11, 15, and 16 were subjected to a mild hydroxylation in the presence of m-chloroperbenzoic acid to give the 2-hydroxystilbenes 17−19. Synthesis of Compounds 20−25. Preliminary Experiments. Reaction conditions were optimized by subjecting compound 19 to preliminary experiments (below indicated as a−c). a. Synthesis in Homogeneous Phase. Compound 19 (0.0160 g, 0.017 mmol) was dissolved in KOH solution (1 mL of 1.25 M) in dry EtOH and stirred for 15 min. Tetra-O-acetyl-α-D-glucopyranosyl bromide (0.0102 g, 0.024 mmol) was added, and the mixture was stirred at room temperature for 3 days. The reaction was monitored by TLC (5% CH3OH/CHCl3); two further aliquots of tetra-O-acetyl-α-Dglucopyranosyl bromide (0.0050 g, 0.012 mmol) were added after 12 and 36 h. b. Synthesis in Heterogeneous Phase at Room Temperature. Compound 19 (0.0044 g, 0.014 mmol) was dissolved in MeOH/H2O (55:45, 2 mL) and stirred together with K2CO3 (0.0200 g, 0.145 mmol) for 15 min at room temperature. A mixture of tetrabutylammonium chloride (TBACl, 0.0042 g, 0.014 mmol) and tetra-Oacetyl-α-D-glucopyranosyl bromide (0.0145 g, 0.035 mmol) in CHCl3 (2 mL) were added. The biphasic system was stirred at room temperature for 20 h, and the reaction was monitored by TLC (5% CH3OH/CHCl3). c. Synthesis in Heterogeneous Phase under Reflux. Compound 19 (0.0057 g, 0.020 mmol) was dissolved in MeOH/H2O (45:55, 2.2 mL), and the solution was stirred with K2CO3 (0.0248 g, 0.179 mmol) at room temperature for 15 min. To the solution were added 1 mL of CHCl3, TBACl (0.0055 g, 0.018 mmol), and a chloroformic solution of tetra-O-acetyl-α-D-glucopyranosyl bromide (0.0185 g, 0.045 mmol, dissolved in 1.2 mL of CHCl3). The heterogeneous mixture was stirred at 60 °C for 24 h. After 10 h, a solution of tetra-O-acetyl-α-Dglucopyranosyl bromide (0.0098 g, 0.024 mmol/2 mL of CHCl3) was added. The reaction was monitored by TLC (5% CH3OH/CHCl3). The procedure c gave a higher yield of compound 25 and was also employed to synthesize compounds 21, 22, and 24. Under these conditions 20 and 23 were obtained in low yields; higher yields were obtained carrying out the reaction at room temperature (procedure b). (2R,3S,4S,5S)-2-[2-(4-Methoxystyryl)-4,6-dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5-triol (20). Compound 17 (0.0184 g, 0.064 mmol) was stirred with K2CO3 (0.0880 g, 0.637 mmol) in MeOH/H2O (45:55, 8.3 mL) for 15 min at room temperature. The catalyst TBACl (0.0178 g, 0.064 mmol), 3.6 mL of CHCl3, and a solution of tetra-O-acetyl-α-D-glucopyranosyl bromide (0.0396 g, 0.096 mmol in 4 mL of CHCl3) were added. The mixture was stirred at room temperature for 24 h. The reaction was monitored by TLC (5% CH3OH/CHCl3), and a solution of tetra-O-acetyl-α-D-

glucosidase inhibitors, whose IC50 values are higher than 100 μM.43 Resveratrol (1) and other related stilbenoids have previously been reported as α-glucosidase inhibitors46,47 or antidiabetic or antiobesity agents.48 In addition, the stilbenoid glucoside desoxyrhaponticin (5-hydroxy-4′-methoxystilbene-3O-β-D-glucopyranoside, 26), isolated from Rheum emodi49 and closely related to compounds 20−25, was reported as a yeast αglucosidase inhibitor. In conclusion the polymethoxystilbene glycosides 20−25 were synthesized and, together with the related 2-hydroxypolymethoxystilbenes 17−19, were subjected to an antiproliferative activity bioassay toward Caco-2 (human colon) and SHSY5Y (neuroblastoma) cancer cells. Compounds 17 and 18 proved highly active on both cells (Table 1); 18 showed a GI50 value of 3 μM (Caco-2 cells), a value comparable to that of 5FU. Compound 19 was inactive, although its glucoside 22 and galactoside 25 showed a weak growth inhibition of Caco-2 and SH-SY5Y cells. The possible hydrolytic release of the aglycones 17−19 by β-glucosidase or β-galactosidase was also evaluated; only 20 and 21 were significantly hydrolyzed by β-glucosidase. Collectively, the data show that minimal differences in the structure of both stilbenoids and conjugates may substantially affect their antiproliferative activity. An evaluation of the βglucosidase or β-galactosidase activity in the presence of compounds 20−25 revealed that all are moderate inhibitors at least toward one of these enzymes. Finally, compounds 21 and 24 showed a promising α-glucosidase inhibition. Thus, these resveratrol-related conjugates may be considered as lead compounds for future development of antidiabetic or antiobesity agents.

■

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were measured with a Rudolph Research Analytical (Autopol I) polarimeter at 25 °C and 589 nm. UV spectra were recorded on a Jasco V630 spectrometer. NMR spectra were run on a Varian Unity Inova spectrometer operating at 499.86 (1H) and 125.70 MHz (13C) and equipped with a gradient-enhanced, reverse-detection probe. Chemical shifts (δ) are indirectly referred to TMS using solvent signals. The 2D g-HSQCAD experiments were performed with matched adiabatic sweeps for coherence transfer, corresponding to a central 13C−1H Jvalue of 146 Hz. G-HMBCAD experiments were optimized for a longrange 13C−1H coupling constant of 8.0 Hz. All NMR experiments, including 2D spectra, i.e., g-COSY, g-HSQCAD, and g-HMBCAD, were performed using software supplied by the manufacturers and acquired at constant temperature (300 K). Methanol-d4 or pyridine-d5 were used as solvents. Mass spectra were acquired with an Agilent 6410 triple quadrupole (1200 Series) mass spectrometer equipped with a multimodal ionization source operating in MMI-ESI, in positive or negative mode. Samples infused were eluted on a cartridge (Zorbax Eclipse XDB-C18; 4.6 × 30 mm, 3.5 μm; Agilent) with MeOH/H2O/ HCOOH (98:2:0.1). The following parameters were used for sample ionization: gas temperature 300 °C; vaporizer temperature 250 °C; gas flow 10 L/min; nebulizer 60 psi; fragmentator 135 V; capillary voltage 3500 V; charging 2000 V. Elemental analyses were performed on a PerkinElmer 240B microanalyzer. High-performance liquid chromatography (HPLC) was carried out using an Agilent Series G1354A pump and an Agilent UV G1315D as diode array detector (DAD). An Agilent Series 1100 G1313A autosampler was used for sample injection. PLC was performed on LiChroprep Si 60 (0.025−0.040 mm; Merck) or on silica gel RP-18 (0.025−0.040 mm; Merck) using different solvent systems, as reported for each compound. Thin-layer chromatography (TLC) was carried out using precoated silica gel F254 plates (Merck); visualization of reaction components was achieved under UV light at a wavelength of 254 and 366 nm or by E

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

The aqueous phase was extracted with CHCl3 (3 × 6 mL), and the combined organic phases were washed with H2O (3 × 15 mL). The CHCl3 phase was dried over Na2SO4, filtered, and dried in vacuo. The residue was purified by liquid chromatography using RP-18 silica gel with a gradient of CH3CN in H2O (from 40% to 80%). Product 22 was obtained (0.0063 g, 38%): amorphous powder; Rf (TLC) = 0.37 (4% CH3OH/CHCl3); [α]25D = +55 (c 0.1, CH3OH); UV (CH3OH) λmax (log ε) 303 (4.30) nm. 1H NMR (500 MHz, methanol-d4, 300 K) δ 7.77 (d, J = 16.2 Hz, 1 H, 7-H), 7.02 (d, J = 16.2 Hz, 1 H, 8-H), 6.83 (d, J = 2.7 Hz, 1 H, 6-H), 6.77 (d, J = 2.0 Hz, 2 H, 2′-H and 6′-H), 6.55 (d, J = 2.7 Hz, 1 H, 4-H), 6.40 (t, J = 2.0 Hz, 1 H, 4′-H), 4.74 (d, J = 8.0 Hz, 1 H, 1″-H), 3.86 (s, 3 H, 3-OCH3), 3.84 (s, 3 H, 5-OCH3), 3.83 (s, 6 H, 3′-OCH3 and 5′-OCH3), 3.77 (dd, J = 11.0, J = 2.5 Hz,1 H, 6a″-H), 3.65 (dd, J = 11.0, J = 5.5 Hz, 1 H, 6b″-H), 3.56 (dd, J = 8.5, 8.0 Hz,1 H, 2″-H), 3.46−342 (m, 2 H, 3″-H and 4″-H), 3.18 (m, 1 H, 5″-H); 13C NMR (125 MHz, methanol-d4, 300 K) δ 162.5 (C-3′ and C-5′), 158.4 (C-5), 154.7 (C-3), 141.1 (C-1), 139.2 (C-2),133.5 (C-1′), 130.5 (C-8), 125.8 (C-7), 106.2 (C-1″), 105.7 (C-2′ and C6′), 101.11 (C-4), 101.66 (C-6), 101.08 (C-4′), 78.02 (C-5″), 77.92 (C-3″), 75.9 (C-2″), 68.7 (C-4″), 63.2 (C-6″), 56.7 (C-3-OCH3), 55.9 (C-3′-OCH3 and C-5′-OCH3), 56.1 (C-5-OCH3); ESIMS calcd for C24H30O10Na [M + Na]+ 501.17; found 501.2; anal. calcd for C24H30O10 C, 60.24; H, 6.32; found C, 60.30; H, 6.21. (2R,3S,4S,5R)-2-[2-(4-Methoxystyryl)-4,6-dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5-triol (23). Compound 17 (0.0232 g, 0.081 mmol) was stirred with K2CO3 (0.1123 g, 0.813 mmol) in MeOH/H2O (45:55, 10.2 mL) for 15 min at room temperature. To the aqueous solution were added CHCl3 (4.3 mL), TBACl (0.0226 g, 0.049 mmol), and a solution containing tetra-Oacetyl-α-D-galactopyranosyl bromide (0.0666 g, 0.162 mmol) in CHCl3 (4.8 mL). The heterogeneous mixture was stirred at room temperature for 24 h, and the reaction was monitored by TLC (5% CH3OH/CHCl3). Another addition of sugar was done after 10 h (0.0293 g/2.0 mL). The reaction was quenched by phase separation. The CHCl3 phase was dried over anhydrous Na2SO4, filtered, and evaporated in vacuo. The organic extract was purified by column chromatography with RP-18 silica gel with a gradient of CH3CN in H2O (from 10% to 70%). Amorphous powder (0.0069 g, 30%); Rf (TLC) = 0.23 (5% CH3OH/CHCl3); UV (CH3OH) λmax (log ε) 306 (4.01) nm; 1H NMR (500 MHz, pyridine-d5, 300 K) δ 8.60 (d, J = 16.5 Hz, 1 H, 7-H), 7.86 (d, J = 8.5 Hz, 2 H, 2′-H and 6′-H),7.49 (d, J = 16.5 Hz, 1 H, 8-H), 7.23 (d, J = 3.0 Hz, 1 H, 6-H), 6.99 (d, J = 8.5 Hz, 2 H, 3′-H and 5′-H), 6.76 (d, J = 3.0 Hz, 1 H, 4-H), 5.46 (d, J = 8.0 Hz, 1 H, 1″-H), 4.93 (dd, J = 9.0, 8.0 Hz, 1 H, 2″-H), 4.75 (bd, J = 3.5 Hz, 1 H, 4″-H), 4.63 (dd, J = 10.5, J = 6.5 Hz, 1 H, 6a″-H), 4.36 (dd, J = 10.5, J = 6.5, 1 H, 6b″-H), 4.32 (dd, J = 9.0, 3.5 Hz, 1 H, 3″H), 4.08 (bt, J = 6.5 Hz, 5″-H) 3.84 (s, 3 H, 5-OCH3), 3.83 (s, 3 H, 3OCH3), 3.70 (s, 3 H, 4′-OCH3). 13C NMR (125 MHz, pyridine-d5, 300 K) δ 159.9 (C-4′), 157.4 (C-5), 154.6 (C-3), 139.5 (C-2), 133.6 (C-1), 131.4 (C-1′), 129.3 (C-8), 128.7 (C-2′ and C-6′), 123.2* (C7), 114.7 (C-3′ and C-5′), 107.6 (C-1″), 101.2 (C-6), 100.7 (C-4), 77.0 (C-5″), 75.4 (C-3″), 73.6 (C-2″), 69.9 (C-4″), 61.9 (C-6″), 56.4 (C-3-OCH3), 55.6 (C-5-OCH3), 55.3 (C4′-OCH3); value with superscript (*) was assigned through HMBC correlations; the signal was partially overlapped with that of the residual undeuterated solvent; ESIMS calcd for C23H28O9Na [M + Na]+ 471.16; found 471.2; anal. calcd for C23H28O9 C, 61.60; H, 6.29; found C, 61.42; H, 6.23. (2R,3S,4S,5R)-2-[2-(3,4-Dimethoxystyryl)-4,6dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5triol (24). Compound 18 (0.0190 g, 0.060 mmol) was stirred with K2CO3 (0.0838 g, 0.606 mmol) in MeOH/H2O (45:55, 7.7 mL) at room temperature for 15 min. CHCl3 (4 mL), TBACl (0.0163 g, 0.060 mmol), and an organic solution of tetra-O-acetyl-α-D-galactopyranosyl bromide (0.0617 g, 0.150 mmol in 4 mL of CHCl3) were added to the aqueous phase. The mixture was heated at 60 °C for 24 h, and the reaction was monitored by TLC. After 10 h tetra-O-acetyl-α-Dgalactopyranosyl bromide (0.0301 g) in 2 mL of CHCl3 was added to the reaction mixture. The aqueous phase was extracted with CHCl3 (3 × 6 mL), and the combined organic phases were washed with H2O (3 × 15 mL), dried

glucopyranosyl bromide (0.0175 g, 0.042 mmol/2 mL of CHCl3) was added after 10 h. The reaction was quenched by phase separation. The organic phase was dried over anhydrous Na2SO4, and evaporated to dryness, and the crude residue was purified by flash chromatography on RP-18 silica gel using a gradient of CH3CN/H2O (from 10% to 70%). Amorphous powder (0.0056 g, 33% yield): Rf (TLC) = 0.25 (5% CH3OH/CHCl3); [α]25D = +73 (c 0.1, CH3OH); UV (CH3OH) λmax (log ε) 306 (4.32) nm; 1H NMR (500 MHz, methanol-d4, 300 K) δ 7.67 (d, J = 16.5 Hz, 1 H, 7-H), 7.54 (d, J = 8.7 Hz, 2 H, 2′-H and 6′H), 7.04 (d, J = 16.5 Hz, 1 H, 8-H), 6.92 (d, J = 8.7 Hz, 2 H, 3′-H and 5′-H), 6.82 (d, J = 2.5 Hz, 1 H, 6-H), 6.53 (d, J = 2.5 Hz, 1 H, 4-H), 4.72 (d, J = 8.0 Hz, 1 H, 1″-H), 3.86 (s, 3 H, 3-OCH3), 3.84 (s, 3 H, 5OCH3), 3.82 (s, 3 H, 4′-OCH3), 3.77 (dd, J = 12.0, 3.0 Hz, 1 H, 6a″H), 3.67 (dd, J = 12.0, J = 5.0 Hz, 1 H, 6b″-H), 3.56 (dd, J = 9.0, 8.0 Hz, 1 H, 2″-H), 3.48−3.42 (m, 2 H, 3″-H and 4″-H), 3.20 (ddd, J = 6.5, 5.0, 3.0 Hz, 1H, 5″-H); 13C NMR (125 MHz, methanol-d4, 300 K) δ 160.9 (C-4′), 158.4 (C-5), 154.5 (C-3), 139.1 (C-2), 134.1 (C-1), 131.9 (C-1′), 130.1 (C-8), 129.1 (C-2′ and C-6′), 123.1 (C-7), 115.1 (C-3′ and C-5′), 106.4 (C-1″), 101.5 (C-6), 100.5 (C-4), 78.1 (C-5″), 77.9 (C-3″), 75.9 (C-2″), 71.5 (C-4″), 62.7 (C-6″), 56.7 (C-3-OCH3), 56.1 (C-5-OCH3), 55.7 (C4′-OCH3); ESIMS calcd for C23H28O9Na [M + Na]+ 471.16; found 471.2; anal. calcd for C23H28O9 C, 61.60; H, 6.29; found C, 61.24; H, 6.27. (2R,3S,4S,5S)-2-[2-(3,4-Dimethoxystyryl)-4,6-dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5-triol (21). Compound 18 (0.0212 g, 0.067 mmol) was dissolved in MeOH/H2O (45:55, 8.5 mL), and the solution was stirred with K2CO3 (0.0926 g, 0.670 mmol) at room temperature for 15 min. To the aqueous solution was added TBACl (0.0182 g, 0.067 mmol) in CHCl3 (4 mL), and subsequently, a solution of tetra-O-acetyl-α-D-glucopyranosyl bromide (0.0688 g, 0.168 mmol in 4.5 mL of CHCl3) was added. The heterogeneous mixture was heated at 60 °C for 24 h. The reaction was monitored by TLC (4% CH3OH/CHCl3), and after 10 h tetra-O-acetyl-α-Dglucopyranosyl bromide in CHCl3 (0.0301 g/2 mL) was added. The aqueous phase was extracted with CHCl3 (3 × 8 mL); the combined organic phases were washed with H2O (3 × 16 mL), dried over Na2SO4, filtered, and evaporated to dryness. The crude extract was purified by column chromatography with RP-18 silica gel with a gradient of CH3CN in H2O (from 10% to 50%), affording 21 (0.0087 g, 43%): amorphous powder; Rf (TLC) = 0.16 (4% CH3OH/CHCl3); [α]25D = 22 (c 0.2, CH3OH); UV (CH3OH) λmax (log ε) 320 (3.94) nm; 1H NMR (500 MHz, methanol-d4, 300 K) δ 7.68 (d, J = 16.5 Hz, 1 H, 7-H), 7.27 (d, J = 1.5 Hz, 1 H, 2′-H), 7.13 (dd, J = 8.0 Hz, J = 1.5 Hz, 1 H, 6′-H), 7.03 (d, J = 16.5 Hz, 1 H, 8-H), 6.94 (d, J = 8.0 Hz, 1 H, 5′-H), 6.82 (d, J = 3.0 Hz, 1 H, 6-H), 6.53 (d, J = 3.0 Hz, 1 H, 4H), 4.72 (d, J = 8.5 Hz, 1 H, 1″-H), 3.91 (s, 3 H, 3′-OCH3), 3.86 (s, 3 H, 3-OCH3), 3.85 (s, 3 H, 4′-OCH3), 3.84 (s, 3 H, 5-OCH3), 3.76 (dd, J = 12.0, J = 3.5 Hz, 1 H, 6a″-H), 3.66 (dd, J = 12.0, 3JH,H = 5.0 Hz, 1 H, 6b″-H), 3.57 (dd, J = 8.5, 7.0 Hz, 1 H, 2″-H), 3.43* (dd, J = 7.5, 7.0 Hz, 1 H, 3″-H), 3.45* (dd, J = 7.0, 4.0, 1 H, 4″-H), 3.18 (m, 1 H, 5″-H); the signals with identical superscript (*) are partially overlapped; 13C NMR (125 MHz, methanol-d4, 300 K) δ 158.4 (C-5), 154.7 (C-3), 150.7 (C-3′), 150.5 (C-4′), 139.1 (C-2), 133.9 (C-1), 132.6 (C-1′), 130.3 (C-8), 123.4 (C-7), 121.5 (C-6′), 112.9 (C-5′), 110.8 (C-2′), 106.4 (C-1″), 101.5 (C-6), 100.7 (C-4), 78.01 (C-5″), 77.9 (C-3″), 75.9 (C-2″), 71.6 (C-4″), 62.7 (C-6″), 56.72 (C-4′OCH3), 56.63 (C-3-OCH3), 56.49 (C-3′-OCH3), 56.07 (C-5-OCH3); ESIMS calcd for C24H30O10Na [M + Na]+ 501.17; found 501.2; anal calcd for C24H30O10 C, 60.24; H, 6.32; found C, 60.25; H, 6.19. (2R,3S,4S,5S)-2-[2-(3,5-Dimethoxystyryl)-4,6-dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5-triol (22). Compound 19 (0.0162 g, 0.051 mmol) was dissolved in MeOH/H2O (45:55, 7.5 mL), and the solution was stirred with K2CO3 (0.0714 g, 0.517 mmol) at room temperature for 15 min. CHCl3 (4 mL), TBACl (0.0142 g, 0.051 mmol), and the solution of tetra-O-acetyl-α-D-glucopyranosyl bromide previously prepared (0.0524 g, 0.127 mmol in 3 mL of CHCl3) were added to the aqueous phase. The reaction mixture was stirred at 60 °C for 24 h, and tetra-O-acetyl-α-D-glucopyranosyl bromide (0.025 g) in CHCl3 (1.5 mL) was added after 10 h. The reaction was monitored by TLC (4% CH3OH/CHCl3). F

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

with anhydrous Na2SO4, filtered, and evaporated to dryness in vacuo. The crude extract was purified by column chromatography with RP-18 silica gel eluting with a gradient of CH3CN in H2O (from 10% to 30%). By PLC 0.0115 g of galactoside 24 was recovered (46%): Rf (TLC) = 0.23 (4% CH3OH/CHCl3); [α]25D = +80 (c 0.2, CH3OH); UV (CH3OH) λmax (log ε) 322 (4.28) nm; 1H NMR (500 MHz, pyridine-d5, 300 K) δ 8.64 (d, J = 16.5 Hz, 1 H, 7-H), 7.72 (d, J = 2.0 Hz, 1 H, 2′-H), 7.52 (d, J = 16.5 Hz, 1 H, 8-H), 7.40 (dd, J = 8.5, J = 2.0 Hz, 1 H, 6′-H), 7.26 (d, J = 3.0 Hz, 1 H, 6-H), 6.94 (d, J = 8.5 Hz, 1 H, 5′-H), 6.77 (d, J = 3.0 Hz, 1 H, 4-H), 5.49 (d, J = 7.5 Hz, 1 H, 1″H), 4.92 (bdd, J = 7.5 Hz, 1H, 2″-H), 4.72 (bs, 1 H, 4″-H), 4.59 (bdd, J = 10.5, J = 7.0 Hz, 1H, 6a″-H), 4.36* (dd, J = 10.5, J = 5.5 Hz, 1 H, 6b″-H), 4.32* (dd, J = 9.3, 5.5 Hz, 1 H, 3″-H), 4.09 (bdd, J = 7.0, 5.5 Hz, 1 H, 5″-H), 3.94 (s, 3 H, 3′-OCH3), 3.84 (s, 6 H, 3-OCH3 and 5OCH3), 3.79 (s, 3 H, 4′-OCH3); the signals with identical superscript (*) are partially overlapped; 13C NMR (125 MHz, pyridine-d5, 300 K) δ 157.2 (C-5), 154.5 (C-3), 150.1# (C-3′), 149.6# (C-4′), 139.2 (C-2), 133.4 (C-1), 131.7 (C-1′), 129.4 (C-8), 123.8 (C-7), 120.8 (C-6′), 112.3 (C-5′), 110.0 (C-2′), 107.2 (C-1″), 100.9 (C-6), 100.6 (C-4), 76.7 (C-5″), 75.2 (C-3″), 73.3 (C-2″), 69.6 (C-4″), 61.7 (C-6″), 56.2 (C-3-OCH3), 55.8 (C-3′-OCH3), 55.7 (C-4′-OCH3), 55.4 (C-5OCH3); values with superscript (#) were attributed through HMBC correlations; the signals were partially overlapped with that of the residual undeuterated solvent; ESI-MS calcd for C24H30O10Na [M + Na]+ 501.17; found 501.2; anal. calcd for C24H30O10 C, 60.24; H, 6.32; found C, 60.17; H, 6.31. (2R,3S,4S,5R)-2-[2-(3,5-Dimethoxystyryl)-4,6dimethoxyphenoxy]tetrahydro-6-hydroxymethyl-2H-pyran-3,4,5triol (25). Compound 19 (0.0280 g, 0.088 mmol) was stirred in MeOH/H2O (45:55, 11.3 mL) with K2CO3 (0.1226 g, 0.887 mmol) at room temperature for 15 min. After this time, CHCl3 (5 mL), TBACl (0.0244 g, 0.088 mmol), and a solution of tetra-O-acetyl-α-Dgalactopyranosyl bromide (0.0904 g, 0.220 mmol) in CHCl3 (6.5 mL) were added to the reaction mixture. The heterogeneous mixture was stirred at 60 °C for 24 h, and at 10 h, 0.0396 g/3 mL of sugar in CHCl3 was added. The aqueous phase was extracted with CHCl3 (3 × 10 mL), and the combined CHCl3 phases were washed with H2O (3 × 20 mL), dried over anhydrous Na2SO4, filtered, and evaporated to dryness. The crude extraction was purified by column chromatography with RP-18 silica gel with a gradient of CH3CN in H2O (from 40% to 70%). From PLC 25 was recovered (0.0109 g, 40%): Rf (TLC) = 0.38 (4% CH3OH/CHCl3); [α]25D = +70 (c 0.03, CH3OH); UV (CH3OH) λmax (log ε) 305 (4.46) nm; 1H NMR (500 MHz, pyridine-d5, 300 K) δ 8.75 (d, JH,H = 16.4 Hz, 1 H, 7-H), 7.54 (d, J = 16.4 Hz, 1 H, 8-H), 7.27 (d, J = 2.3 Hz, 2 H, 2′-H and 6′-H), 7.25 (d, J = 2.5 Hz, 1 H, 6-H), 6.78 (d, J = 2.5 Hz, 1 H, 4-H), 6.71 (t, J = 2,3 Hz, 1 H, 4′-H), 5.48 (d, J = 8.0 Hz, 1 H, 1″-H), 4.91 (dd, J = 8.3, 8.0 Hz, 1 H, 2″-H), 4.70 (bd, J = 3.3 Hz, 1 H, 4″-H), 4.59 (dd, J = 10.5, 3JH,H = 7.0 Hz, 1 H, 6a″-H), 4.35 (dd, J = 10.5, J = 5.5 Hz, 1 H, 6b″-H), 4.32 (dd, J = 8.3, 3.3 Hz, 1 H, 3″-H), 4.08 (bdd, J 7.0, 5.5 Hz, 1 H, 5″-H), 3.84 (s, 3 H, 3-OCH3), 3.83 (s, 3 H, 5-OCH3), 3.82 (s, 6 H, 3′-OCH3 and 5′-OCH3); 13C NMR (125 MHz, pyridine-d5, 300 K) δ 161.6 (C3′ and C-5′), 157.3 (C-5), 154.6 (C-3), 140.6 (C-1), 139.4 (C-2), 133.0 (C-1′), 129.6 (C-8), 123.0* (C-7), 107.2 (C-1″), 105.1 (C-2′ and C-6′), 101.13 (C-4 and C-4′), 101.06 (C-6), 76.7 (C-5″), 75.1 (C3″), 73.2 (C-2″), 69.5 (C-4″), 61.5 (C-6″), 56.2 (C-3-OCH3), 55.4 (C-5-OCH3), 55.3 (C-3′-OCH3 and C-5′-OCH3); value with superscript (*) was assigned through HMBC correlations; the signal was partially overlapped with that of the residual undeuterated solvent; ESIMS calcd for C24H30O10Na [M + Na]+ 501.17; found 501.2; anal. calcd for C24H30O10 C, 60.24; H, 6.32; found C, 60.19; H, 6.29. Antiproliferative Activity Assay. Human Cell Cultures. The colorectal adenocarcinoma Caco-2 cells (ATCC number: HTB-37) were obtained from the American Type Culture Collection (ATCC, Teddington, UK). The Caco-2 cell line was grown in RPMI 1640 medium (cat. 61870-010, Gibco by Life Technologies, Italy). The SHSY5Y (ATCC CRL-2266) was kindly provided by Prof. Maria Angela Sortino (University of Catania, Italy). The SH-SY5Y cells were grown in DMEM-F12 (cat. no. 21331-046, Gibco by Life Technologies, Italy). All media were supplemented with 10% (v/v) heat-inactivated

fetal bovine serum, 2 mM L-alanyl-L-glutamine, and penicillin− streptomycin (50 units−50 mg per mL), and cell cultures were incubated at 37 °C under a humidified atmosphere of 5% CO2/95% air. The culture media were changed twice a week. Treatment with Cytotoxic Compounds and MTT Colorimetric Assay. Human cancer cell lines ((2.5−3.0) × 103 cells/0.33 cm2) were plated in Nunclon Microwell 96-well plates (Nunc) and were incubated at 37 °C. After 24 h, cells were treated with the compounds (final concentration 0.01−100 μM). Cells treated with 0.5−1% of DMSO were used as controls. Microplates were incubated at 37 °C in humidified atmosphere of 5% CO2/95% air for 3 days, and then cytotoxicity was measured with a colorimetric assay based on the use of a tetrazolium salt [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide].50 The results were read on a multiwall scanning spectrophotometer (Multiscan reader), using a wavelength of 570 nm. Each value was the average of 6−8 wells. The GI50 value was calculated according to the NCI:51 thus, GI50 is the concentration of test compound where 100(T − T0)/(C − T0) = 50 (where T is the optical density of the test well after a 72 h period of exposure to test compound, T0 is the optical density at time zero, C is the DMSO control optical density). The cytotoxicity effect was calculated according to NCI when the optical density of treated cells was lower than the T0 value using the following formula: 100(T − T0)/T0 < 0. Enzymatic Cleavage Assay. Compounds 20−25 were dissolved in DMSO and diluted with phosphate buffer at pH = 7.2 (Na2HPO4/ KH2PO4 0.050 M) to a final concentration of 5.0 × 10−5 M (the final concentration of DMSO did not exceeded 1.0%). β-Glucosidase from almonds (1.2 U/mL) or β-galactosidase from A. oryzae (0.7 U/mL) was added, and the mixtures were incubated at 37 °C. Samples (200 μL) were taken at regular time intervals up to 96 h, and the enzymatic hydrolysis was monitored by HPLC on an RP-18 column at 305 nm (see General Experimental Procedures). The activity of the enzymes does not change when the solution contains 1.0% DMSO. All assays were performed in triplicate. In order to quantify the released aglycone by the cleavage assay, calibration curves were created for compounds 17−19: four solutions of the aglycones prepared within the concentration range of the test (10−5−10−6 M) were eluted by HPLC on an RP-18 column under the same conditions of the assay. Glycosidase Enzyme Inhibition Assay. A slight modification of the method of Tsujii and co-workers52 was used to evaluate the inhibition of β-glucosidase from almonds, β-galactosidase from A. oryzae, and αglucosidase from S. cerevisiae by compounds 20−25. The enzymatic activity was determined spectrophotometrically at 400 nm by monitoring the release of p-nitrophenol and o-nitrophenol from the substrates pNP-β-Glc or pNP-α-Glc (for glucosidases) and oNP-β-Gal (for galactosidase), respectively. Each substrate was dissolved in DMSO/phosphate buffer (Na2HPO4/KH2PO4 0.050 M, pH = 7.2) in order to have a final concentration of 6.5 × 10−5 M. In each assay, 3 mL of substrate solution were incubated with 0.1, 0.2, 0.4, 0.6, or 1.0 mL of the stock solution of each glycoside (2.6−4.7 × 10−4 M; for details see the Supporting Information) in the presence of the corresponding enzyme (2.9 U/mL for β-glucosidase, 4.0 U/mL for βgalactosidase, and 2.0 U/mL for α-glucosidase) at 25 °C. The final concentration of DMSO of the solutions did not exceed 1.0%. In each set of experiments, the assay was performed in triplicate with five different concentrations. Preliminary kinetic studies of the three enzymes in the presence of pertinent substrate have provided the following incubation times: 30 min for β-glucosidase and βgalactosidase and 2 h for α-glucosidase. The percentage inhibition was determined for each compound from the residual activity by comparing the enzyme activity with and without compounds. The concentration of inhibition required for 50% of glycosidase activity under the assay conditions was defined as the IC50 value. G

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

■

Article

(19) Heynekamp, J. J.; Weber, W. M.; Hunsaker, L. A.; Gonzales, A. M.; Orlando, R. A.; Deck, L. M.; Vander Jagt, D. L. J. Med. Chem. 2006, 49, 7182−7189. (20) Valdameri, G.; Rangel, L. P.; Spatafora, C.; Guitton, J.; Gauthier, C.; Arnaud, O.; Ferreira-Pereira, A.; Falson, P.; Winnischofer, S. M. B.; Rocha, M. E. M.; Tringali, C.; Di Pietro, A. ACS Chem. Biol. 2012, 7, 322−330. (21) Sarpietro, M. G.; Ottimo, S.; Giuffrida, M. C.; Spatafora, C.; Tringali, C.; Castelli, F. Int. J. Pharm. 2010, 388, 144−150. (22) Sarpietro, M. G.; Spatafora, C.; Accolla, M. L.; Cascio, O.; Tringali, C.; Castelli, F. J. Nat. Prod. 2013, 76, 1424−1431. (23) Sarpietro, M. G.; Spatafora, C.; Tringali, C.; Micieli, D.; Castelli, F. J. Agric. Food Chem. 2007, 55, 3720−3728. (24) Lin, H.-S.; Spatafora, C.; Tringali, C.; Ho, P. C. J. Pharm. Biomed. Anal. 2012, 57, 94−98. (25) Lin, H. S.; Tringali, C.; Spatafora, C.; Choo, Q. Y.; Ho, P. C. Chromatographia 2010, 72, 827−832. (26) Lin, H. S.; Tringali, C.; Spatafora, C.; Wu, C.; Ho, P. C. J. Pharm. Biomed. Anal. 2010, 51, 679−684. (27) Lin, H. S.; Zhang, W.; Go, M. L.; Tringali, C.; Spatafora, C.; Ho, P. C. J. Agric. Food Chem. 2011, 59, 1072−1077. (28) Cardile, V.; Chillemi, R.; Lombardo, L.; Sciuto, S.; Spatafora, C.; Tringali, C. Z. Naturforsch., C: J. Biosci. 2007, 62, 189−195. (29) Lee, J.; Kim, S. J.; Choi, H.; Kim, Y. H.; Lim, I. T.; Yang, H. M.; Lee, C. S.; Kang, H. R.; Ahn, S. K.; Moon, S. K.; Kim, D. H.; Lee, S.; Choi, N. S.; Lee, K. J. J. Med. Chem. 2010, 53, 6337−6354. (30) Dembitsky, V. M. Lipids 2005, 40, 869−900. (31) La Ferla, B.; Airoldi, C.; Zona, C.; Orsato, A.; Cardona, F.; Merlo, S.; Sironi, E.; D’Orazio, G.; Nicotra, F. Nat. Prod. Rep. 2011, 28, 630−648. (32) Du, Q. H.; Peng, C.; Zhang, H. Pharm. Biol. 2013, 51, 1347− 1354. (33) Kimura, Y.; Okuda, H. J. Pharm. Pharmacol. 2000, 52, 1287− 1295. (34) Carmeliet, P.; Jain, R. K. Nature 2000, 407, 249−257. (35) Hussain, S.; Slevin, M.; Ahmed, N.; West, D.; Choudhary, M. I.; Naz, H.; Gaffney, J. BMC Cell Biol. 2009, 10, 3010.1186/1471-212110-30. (36) Calvaresi, E. C.; Hergenrother, P. J. Chem. Sci. 2013, 4, 2319− 2333. (37) Hanahan, D.; Weinberg, R. A. Cell 2011, 144, 646−674. (38) Young, C. D.; Anderson, S. M. Breast Cancer Res. 2008, 10, 307−308. (39) Arafa, H. M. M. Invest. New Drugs 2010, 28, 306−317. (40) Aggarwal, B. B.; Sundaram, C.; Malani, N.; Ichikawa, H. Molecular Targets and Therapeutic Uses of Curcumin in Health and Disease; Springer Science+Business Media, LLC: New York, 2007; p 595. (41) Larrosa, M.; Tome-Carneiro, J.; Yanez-Gascon, M. J.; Alcantara, D.; Selma, M. V.; Beltran, D.; Garcia-Conesa, M. T.; Urban, C.; Lucas, R.; Tomas-Barberan, F.; Morales, J. C.; Espin, J. C. J. Med. Chem. 2010, 53, 7365−7376. (42) Jacobsson, M.; Malmberg, J.; Ellervik, U. Carbohydr. Res. 2006, 341, 1266−1281. (43) de Melo, E. B.; Gomes, A. D.; Carvalho, I. Tetrahedron 2006, 62, 10277−10302. (44) Scheen, A. J. Diabetes Metab. 1998, 24, 311−320. (45) Du, Z. Y.; Liu, R. R.; Shao, W. Y.; Mao, X. P.; Ma, L.; Gu, L. Q.; Huang, Z. S.; Chan, A. S. C. Eur. J. Med. Chem. 2006, 41, 213−218. (46) Lam, S. H.; Chen, J. M.; Kang, C. J.; Chen, C. H.; Lee, S. S. Phytochemistry 2008, 69, 1173−1178. (47) Kerem, Z.; Bilkis, I.; Flaishman, M. A.; Sivan, U. J. Agric. Food Chem. 2006, 54, 1243−1247. (48) Kim, S.; Jin, Y.; Choi, Y.; Park, T. Biochem. Pharmacol. 2011, 81, 1343−1351. (49) Babu, K. S.; Tiwari, A. K.; Srinivas, P. V.; Ali, A. Z.; Raju, B. C.; Rao, J. M. Bioorg. Med. Chem. Lett. 2004, 14, 3841−3845. (50) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.5b00619. ESI mass, 1D and 2D NMR, and UV spectra of compounds 20−25; α-glycosidase inhibition assay (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel: +39 095 7385015. Fax: +39 095 580138. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by grants of Regione Siciliana (FERS 4.1.2.A 2007−2013 “Piattaforma regionale di ricerca traslazionale per la salute”) and of the University of Catania (FIR, “Finanziamento della Ricerca” 2014).

■

REFERENCES

(1) Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach, 2nd ed.; John Wiley & Sons, Ltd: Chichester, 2002. (2) Aggarwal, B. B.; Shishodia, S. Resveratrol in Health and Disease; CRC Taylor & Francis: Boca Raton, 2006. (3) Pezzuto, J. M. Pharm. Biol. 2008, 46, 443−573. (4) Latruffe, N.; Delmas, D.; Lizard, G.; Tringali, C.; Spatafora, C.; Vervandier-Fasseur, D.; Meunier, P. Resveratrol against Major Pathologies: From Diet Prevention to Possible Alternative Chemotherapies with New Structural Analogues. In Bioactive Compounds from Natural Sources, 2nd ed.; Natural Products as Lead Compounds in Drug Discovery; Tringali, C., Ed.; CRC Press-Taylor & Francis: Boca Raton, 2012; pp 339−378. (5) Spatafora, C.; Tringali, C. Anti-Cancer Agents Med. Chem. 2012, 12, 902−918. (6) Mazue, F.; Colin, D.; Gobbo, J.; Wegner, M.; Rescifina, A.; Spatafora, C.; Fasseur, D.; Delmas, D.; Meunier, P.; Tringali, C.; Latruffe, N. Eur. J. Med. Chem. 2010, 45, 2972−2980. (7) Cardile, V.; Lombardo, L.; Spatafora, C.; Tringali, C. Bioorg. Chem. 2005, 33, 22−33. (8) Bhusainahalli, V. M.; Spatafora, C.; Chalal, M.; VervandierFasseur, D.; Meunier, P.; Latruffe, N.; Tringali, C. Eur. J. Org. Chem. 2012, 5217−5224. (9) Spatafora, C.; Basini, G.; Baioni, L.; Grasselli, F.; Sofia, A.; Tringali, C. Nat. Prod. Commun. 2009, 4, 239−246. (10) Baur, J. A.; Sinclair, D. A. Nat. Rev. Drug Discovery 2006, 5, 493− 506. (11) Walle, T. Ann. N. Y. Acad. Sci. 2011, 1215, 9−15. (12) MacRae, W. D. T.; Towers, G. H. N. Phytochemistry 1985, 24, 561−566. (13) Sale, S.; Verschoyle, R. D.; Boocock, D.; Jones, D. J. L.; Wilsher, N.; Ruparelia, K. C.; Potter, G. A.; Farmer, P. B.; Steward, W. P.; Gescher, A. J. Br. J. Cancer 2004, 90, 736−744. (14) Morris, V. L.; Toseef, T.; Nazumudeen, F. B.; Rivoira, C.; Spatafora, C.; Tringali, C.; Rotenberg, S. A. Mol. Cell. Biochem. 2015, 402, 83−91. (15) Belleri, M.; Ribatti, D.; Nicoli, S.; Cotelli, F.; Forti, L.; Vannini, V.; Stivala, L. A.; Presta, M. Mol. Pharmacol. 2005, 67, 1451−1459. (16) Lin, H. S.; Ho, P. C. J. Pharm. Biomed. Anal. 2009, 49, 387−392. (17) Chillemi, R.; Sciuto, S.; Spatafora, C.; Tringali, C. Nat. Prod. Commun. 2007, 2, 499−513. (18) Basini, G.; Tringali, C.; Baioni, L.; Bussolati, S.; Spatafora, C.; Grasselli, F. Mol. Nutr. Food Res. 2010, 54, S236−S243. H

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

(51) Scherf, U.; Ross, D. T.; Waltham, M.; Smith, L. H.; Lee, J. K.; Tanabe, L.; Kohn, K. W.; Reinhold, W. C.; Myers, T. G.; Andrews, D. T.; Scudiero, D. A.; Eisen, M. B.; Sausville, E. A.; Pommier, Y.; Botstein, D.; Brown, P. O.; Weinstein, J. N. Nat. Genet. 2000, 24, 236− 244. (52) Tsujii, E.; Muroi, M.; Shiragami, N.; Takatsuki, A. Biochem. Biophys. Res. Commun. 1996, 220, 459−466.

I

DOI: 10.1021/acs.jnatprod.5b00619 J. Nat. Prod. XXXX, XXX, XXX−XXX