Lycosides, Unusual Carotenoid-Derived Terpenoid Glycosides from a

Dec 9, 2017 - Vegetable juices, typical culture media for the plant pathogen Phytophthora, effectively induce its asexual reproduction (zoosporangia f...
1 downloads 9 Views 1MB Size
Download PDF
Article pubs.acs.org/JAFC

Cite This: J. Agric. Food Chem. 2018, 66, 163−169

Lycosides, Unusual Carotenoid-Derived Terpenoid Glycosides from a Vegetable Juice, Inhibit Asexual Reproduction of the Plant Pathogen Phytophthora Rika Iwai,† Chunguang Han,‡ Sudhakar V. S. Govindam,† and Makoto Ojika*,† †

Graduate School of Bioagricultural Sciences, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan Research Center for Materials Science, Nagoya University, Chikusa-ku, Nagoya 464-8602, Japan

‡

S Supporting Information *

ABSTRACT: Vegetable juices, typical culture media for the plant pathogen Phytophthora, effectively induce its asexual reproduction (zoosporangia formation). However, some chromatographic fractions from a vegetable juice were found to inhibit asexual reproduction. Bioassay-guided chromatographic steps led to the isolation of four novel compounds, named lycosides A− D, 1−4, that could be metabolic products from a carotenoid. They showed 50% inhibitory activity against the asexual reproduction of P. capsici at 2.1−7.6 μM. The structure−activity relationship and the universality of the inhibitory activity within the Phytophthora genus were also investigated. In addition, the quantitative analysis of lycosides in fresh vegetables and vegetable juices revealed that tomato is the source of these active substances. These food-derived chemicals could help provide safe agents to control the outbreak of the agricultural pest Phytophthora in fields. KEYWORDS: plant pathogen, Phytophthora, inhibitor, asexual reproduction, natural products, vegetable juice

■

INTRODUCTION The genus Phytophthora represents a group of filamentous fungus-like oomycetes and includes more than 100 species, most of which are deleterious to a broad range of economically and ecologically important plant species.1−3 P. infestans is notorious as the causal agent of the Great Famine during the mid-1840s that wiped out the potato crop in Ireland and eventually led to mass starvation.3 The potato late blight caused by this microorganism is one of the biggest worldwide problems in the agricultural sector, resulting in economic losses of several billion dollars annually.4 Farmers are forced to spray synthetic fungicides for about ten or more times to control the plant pest.5,6 The acute damages caused by this pathogen are primarily due to its asexual spores, zoosporangia, that produce thousands of motile zoospores as weapons.7 Instead of using synthetic fungicides, other approaches such as developing genetically resistant plant strains, chemically inducing plant resistance with elicitors and other compounds, and biocontrol could have fewer harmful effects on the environment. An additional possibility could involve using natural products that selectively suppress the reproduction of Phytophthora, especially zoosporangia formation, although such examples are rare.8 Phytophthora is generally cultured on vegetable juices (typically V8 juice) that promote hyphal growth and reproduction (both asexual and sexual ones, as indicated by zoosporangia and oospores formation, respectively). During our search for the vegetable juice components that promote Phytophthora reproduction, we found an interesting phenomenon. Some chromatographic fractions from an extract of V8 vegetable juice inhibited the formation of zoosporangia. This suggests that the vegetable juice includes not only promoters but also inhibitors of the Phytophthora reproduction. In this study, we report the isolation of the Phytophthora asexual reproduction © 2017 American Chemical Society

inhibitors named lycosides A−D, 1−4 (Figure 1), from V8 vegetable juice, as well as their structure−activity relationship

Figure 1. Structures of lycosides A−D, 1−4.

(SAR), the universality of their inhibitory activity within the Phytophthora genus, and distribution in vegetables, especially tomatoes.

■

MATERIALS AND METHODS

General Experimental Procedures. Specific rotations were recorded on a DIP-370 spectrometer (JASCO, Tokyo, Japan). Infrared (IR) spectra were measured using an FT-IR-7000S spectrometer (JASCO). Ultraviolet (UV) spectra were obtained using a V-530 spectrometer (JASCO). Circular dichroism (CD) spectra were obtained on a J-720WI spectrometer (JASCO). Mass spectra (MS) were recorded on a Mariner Biospectrometry Workstation (Applied Biosystems, Foster City, CA) in the positive electrospray ionization Received: Revised: Accepted: Published: 163

October 14, 2017 December 7, 2017 December 9, 2017 December 9, 2017 DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry (ESI) mode. High-resolution MS was performed by infusion method with 50% MeCN/0.1% HCOOH as the solvent. For LC/MS measurements, the equipment was connected to an Agilent 1100 HPLC system (Hewlett-Packard, Palo Alto, CA) under the conditions described below. Nuclear magnetic resonance (NMR) spectra were recorded on an AMX2 600 (600 MHz for 1H) or ARX 400 (400 MHz for 1H) spectrometer (Bruker BioSpin, Yokohama, Japan). The chemical shifts (ppm) were referenced to the tetramethylsilane (TMS) peak. Preparative high-performance liquid chromatography (HPLC) was performed using a high-pressure gradient system (JASCO) composed of a PU-1586 pump, DG-1580-53 degasser, and UV 1570 detector. Strains and Culture. Phytophthora capsici NBRC 30696, P. cinnamomi NBRC 33180, and P. nicotianae NBRC 9049 were purchased from the Biological Resource Centre, National Institute of Technology and Evaluation (NBRC, Chiba, Japan). P. infestans PI 1234-1 (race 1.2.3.4) was provided by Prof. K. Kawakita at the Graduate School of Bioagricultural Sciences, Nagoya University. The strains were kept on PSA medium (soup stock from 200 g fresh sliced potato in 1 L water, 10 g sucrose, 20 g agar) at 15 °C and subcultured every 6 months. Prior to the zoosporangia formation test, the strains were precultured on the PSA medium in a 9 cm dish in an Eyela KCL-2000A incubator (Tokyo Rikakikai Co., Ltd., Tokyo, Japan) at 25 °C and 60% humidity except for P. infestans, which was precultured at 20 °C and 70% humidity. The preculture was continued until the surface was completely covered by the colony, which took 6−14 d depending on the strains. Zoosporangia Formation Test. A piece of precultured Phytophthora mycelium with agar (5 × 5 mm) was inoculated on one of the following media (5 mL in 6 cm dish): (a) 0.4% (v/v) potato/1% sucrose (w/v)/1.5% agar (w/v) (“0.4P medium”), (b) 4% or 3% V8 vegetable juice (v/v)/1% sucrose/2% agar (“4 or 3 V medium”, respectively), or (c) 5% V8 juice/0.02% (w/v) CaCO3/1% sucrose/2% agar (“5 V−Ca medium”). All the media contained 1% (v/v) DMSO or a sample solution in DMSO. The strains were incubated at 25 °C, 60% humidity (20 °C, 70% humidity for P. infestans) for 6−10 d depending on the strains and media. The optimized culture conditions are as follows: 0.4P (8 d culture) or 4 V (6 d culture) media for the tests of chromatographic fractions; 3 V medium (7 d culture) for the activity of the compounds against P. capsici; 3 V medium (6 and 8 d culture) for P. cinnamomi and P. nicotianae, respectively; and 5 V−Ca medium (10 d culture) for P. infestans. Three or four circular areas (12 mm diam.) at a distance of 1.5 cm from the colony center were cut out, and the number of zoosporangia was counted under a microscope. Preliminary Fractionation of Vegetable Juice and Zoosporangia Formation Activity. A can of V8 vegetable juice (340 mL) (Campbell Soup Company, Camden, NJ) was diluted with MeOH (220 mL) and stirred at room temperature for 1 h. The mixture was separated into supernatant and precipitate by centrifugation (2,000 rpm, 5 min). The latter was washed with 40% MeOH (100 mL) by mixing and centrifugation under the same conditions. The supernatant and washing were combined and concentrated. The aqueous residue was dissolved in 40% MeOH (200 mL in a final volume) and applied to an ODS column (100 g of Cosmosil 75C18-OPN) (Nacalai Tesque, Kyoto, Japan) that was eluted with 40, 60, and 80% MeOH (4 fr. × 100 mL each) and then MeOH (7 fr. × 100 mL) to produce 19 fractions. A portion (1.25 mL V8 juice equivalent) of each fraction was dissolved in 0.4P medium (5 mL) and tested for the zoosporangia formation of P. capsici, which was incubated for 8 d (Figure 2A). The fractions 7−9, which were eluted with 60% MeOH, were combined and concentrated to produce a solid (68.5 mg). The material was subjected to flash chromatography on a Hi-Flash size S column (silica gel 6 g) (Yamazen, Kyoto, Japan), which was eluted with MeOH/H2O/CHCl3 (9:1:90 to 45:5:50, 40 min linear gradient) at a flow rate of 3 mL/min to generate 5 fractions. A portion (2.5 mL V8 juice equivalent) of each fraction was dissolved in 4 V medium (5 mL) and tested for the zoosporangia formation of P. capsici, which was incubated for 6 d (Figure 2B). Isolation of 1−4. Twenty cans of V8 vegetable juice (6.8 L) were diluted with MeOH (4.4 L) and stirred at room temperature for 1 h. The mixture was filtered by suction, and the residue was washed with

Figure 2. Zoosporangia formation rate of P. capsici in the presence of chromatographic fractions from V8 vegetable juice. (A) Activity of the first step (ODS) column fractions. The strain was incubated on 0.4% potato medium for 8 d. The sample dose was a 25% juice equivalent. The values are the average of triplicate or duplicate data with ± s.d. The asterisks indicate the activity higher than 120%; (B) Activity of the second step (silica gel) column fractions from the ODS fractions 7−9. The strain was incubated on 4% V8 juice medium for 6 d. The sample dose was a 50% juice equivalent. The values are average of triplicate data with ± s.d. 40% MeOH (0.7 L). The supernatant and washing were combined and concentrated to a MeOH-free solution. The aqueous solution was dissolved in 40% MeOH (1 L in total) and chromatographed on ODS (1 kg of Cosmosil 75C18−OPN), which was eluted with 40, 60, 80% MeOH (4 L each). The 60% MeOH fraction (3.24 g) was chromatographed on silica gel (50 g of Wakogel C300) (Wako Pure Chemicals Industries, Osaka, Japan) eluted with MeOH/H2O/CHCl3 (27:3:70, 325 mL and then 45:5:50, 200 mL). A fraction (673 mg) eluted with MeOH/H2O/CHCl3 (27:3:70) was subjected to flash chromatography on a Hi-Flash size M column (silica gel 14 g) (Yamazen), which was eluted with MeOH/H2O/CHCl3 (4.5:0.5:95 to 40.5:4.5:55, 40 min linear gradient) at a flow rate of 6 mL/min. The active fraction (111 mg) eluted at 14−24 min was purified by HPLC. The column used was a 250 mm × 20 mm i.d., 5 μm, YMC-Pack DODS-5 (YMC, Kyoto, Japan), which was eluted with MeOH in water (50 to 75%, 75 min linear gradient) at a flow rate of 5 mL/min with detection at 210 nm. The active fraction (8.4 mg) eluted at 48−52 min was further purified by HPLC. The column used was a 250 mm × 10 mm i.d., 5 μm, Develosil ODS-UG-5 (Nomura Chemical, Seto, Aichi, Japan), which was eluted with 23% MeCN at a flow rate of 3 mL/min with detection at 234 nm to produce 1 (0.5 mg, tR = 48.6 min), 2 (1.2 mg, tR = 53.1 min), 3 (0.6 mg, tR = 54.9 min), and 4 (0.7 mg, tR = 57.5 min). Lycoside A, 1. Colorless oil, [α]30D +109 (c 0.021, MeOH); UV (MeOH) λmax(ε) 236 (24,000) nm; CD λext (Δε) 251 (+36), 227 (−39) nm; IR (KBr) νmax 3398, 1653, 1596, 1159, 1103, 1074 cm−1; HR ESIMS m/z 617.3178 [M + H]+ (calcd for C30H49O13 617.3168). NMR data are listed in Tables 1 and 2. Lycoside B, 2. Colorless oil; [α]30D +126 (c 0.034, MeOH); UV (MeOH) λmax(ε) 236 (22,000) nm; CD (MeOH) λext (Δε) 250 (+31), 226 (−33) nm; IR (film) νmax 3397, 1655, 1594, 1159, 1075, 1026 cm−1; HR ESIMS m/z 617.3203 [M + H]+ (calcd for C30H49O13 617.3168). NMR data are listed in Tables 1 and 2. Lycoside C, 3. Colorless oil, [α]30D + 133 (c 0.010, MeOH); UV (MeOH) λmax(ε) 236 nm (25,000) nm; CD λext (Δε) 249 (+35), 227 (−30) nm; IR (film) νmax 3419, 1650, 1593, 1077, 1024 cm−1; HR ESIMS m/z 455.2651 [M + H]+ (calcd for C24H39O8 455.2639). NMR data are listed in Tables 1 and 2. Lycoside D, 4. Colorless oil, [α]30D +144 (c 0.023, MeOH); UV (MeOH): λmax(ε) 236 nm (29,000) nm; CD λext (Δε) 250 (+30), 227 (−30) nm; IR (film) νmax 3396, 1655, 1596, 1076, 1024 cm−1; HR ESIMS m/z 455.2646 [M + H]+ (calcd for C24H39O8 455.2639). NMR data are listed in Tables 1 and 2. Preparation of Aglycons 5 and 6. β-D-Glucosidase (from almond, 11 units/mg) (Oriental Yeast, Tokyo, Japan) was dissolved in ammonium acetate buffer (0.1 M, pH 5) to prepare a stock solution of 20 units/mL. A portion (50 μL) of the enzyme stock solution was 164

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry Table 1. 1H NMR Data for Lycosides A−D, 1−4, in CD3OD (400 MHz) position

1

2

3

4

2 4 7 8 10 11 12 13 14 15 16 17 18 1’ 2’ 3′ 4’ 5′ 6’ 1’’ 2’’ 3′’ 4’’ 5′’ 6’’

2.17 (d, 16.2), 2.46 (d, 16.2) 5.91 (s) 5.77 (d, 15.6) 6.76 (d, 15.6) 5.47 (t, 7.4) 2.18, 2.34 (m) 1.51, 1.60 (m) 3.86 (m) 1.17 (d, 6.0) 1.81 (s) 1.92 (d, 1.2) 1.01 (s) 1.06 (s) 4.35 (d, 7.8) 3.23 (dd, 9.0, 7.8) 3.52 (t, 9.0) 3.57 (t, 9.0) 3.38 (m) 3.86 (m) 4.42 (d, 7.8) 3.22 (9.0, 7.8) 3.37 (m) 3.33 (m) 3.30 (m) 3.66 (dd, 11.7, 5.7), 3.87 (m)

2.19 (d, 16.8), 2.47 (d, 16.8) 5.90 (s) 5.77 (d, 15.8) 6.80 (d, 15.8) 5.45 (t, 7.6) 2.30 (m) 1.50, 1.65 (m) 3.80 (sext, 6.0) 1.24 (d, 6.0) 1.82 (s) 1.92 (d, 1.2) 1.00 (s) 1.05 (s) 4.35 (d, 7.6) 3.22 (dd, 9.0, 7.6) 3.51 (t, 9.0) 3.56 (t, 9.0) 3.38 (m) 3.86 (m) 4.41 (d, 8.0) 3.21 (dd, 8.8, 8.0) 3.36 (m) 3.33 (m) 3.31 (m) 3.65 (dd, 11.6, 5.2), 3.87 (dd, 11.6, 2.0)

2.17 (d, 16.8), 2.47 (d, 16.8) 5.91 (s) 5.77 (d, 16.2) 6.76 (d, 16.2) 5.48 (t, 7.2) 2.17, 2.35 (m) 1.51, 1.60 (m) 3.88 (m) 1.16 (d, 6.0) 1.81 (s) 1.92 (s) 1.01 (s) 1.06 (s) 4.32 (d, 7.8) 3.16 (dd, 9.0, 7.8) 3.35 (t, 9.0) 3.30 (m) 3.24 (m) 3.67 (dd, 12.0, 5.4), 3.85 (dd, 12.0, 2.0)

2.19 (d, 16.5), 2.48 (d, 16.5) 5.90 (s) 5.77 (d, 15.6) 6.79 (d, 15.6) 5.45 (t, 7.5) 2.28 (m) 1.49, 1.65 (m) 3.81 (sext, 6.0) 1.23 (d, 6.0) 1.82 (s) 1.92 (d, 0.6) 1.00 (s) 1.06 (s) 4.32 (d, 7.8) 3.16 (dd, 9.0, 7.8) 3.34 (t, 9.0) 3.28 (t, 9.0) 3.24 (m) 3.66 (dd, 12.0, 5.4), 3.84 (12.0, 1.8)

Table 2. 13C NMR Data for Lycosides A−D, 1−4, in CD3OD (100 MHz) position

1

2

3

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1’ 2’ 3′ 4’ 5′ 6’ 1’’ 2’’ 3′’ 4’’ 5′’ 6’’

42.7 50.8 201.2 127.1 167.6 80.0 130.1 128.8 132.4 132.5 24.5 38.7 75.4 19.9 20.8 19.7 24.6 23.6 102.1 75.0 76.5a 80.7 76.4a 62.1 104.7 74.8 77.9 71.4 78.2 62.4

42.6 50.8 201.2 127.1 167.3 80.7 130.3 128.8 132.6 132.4 24.3 37.9 77.3 22.0 20.8 19.7 24.7 23.6 103.9 75.1 76.4 80.7 76.4 62.0 104.6 74.9 77.9 71.4 78.1 62.4

42.7 50.8 201.2 127.1 167.6 80.7 130.1 128.8 132.4 132.6 24.6 38.7 75.1 19.8 20.8 19.7 24.6 23.6 102.2 75.2 77.8 71.8 78.2 62.9

42.6 50.8 201.3 127.1 167.5 80.7 130.3 128.8 132.6 132.4 24.3 37.9 77.2 22.0 20.8 19.7 24.7 23.6 104.0 75.4 77.8 71.7 78.1 62.8

added to a solution of lycoside D, 4 (0.6 mg) in the same buffer (0.2 mL). The mixture was incubated at 37 °C for 13.5 h. The mixture was extracted twice with EtOAc (0.5 mL). The extracted material was purified by HPLC. The column used was a 250 mm × 10 mm i.d., 5 μm, Develosil ODS-UG-5, which was eluted with 55% MeOH (55 min) and then 60% MeOH at a flow rate of 1.5 mL/min with detection at 235 nm to produce (13S)-aglycon, 6 (tR = 80.5 min, 0.083 mg). The same procedure was conducted for a mixture of lycoside A, 1 (0.4 mg) and C, 3 (0.5 mg) to produce (13R)-aglycon, 5 (tR = 77.3 min, 0.105 mg). The yields were calculated by UV measurement with a molar absorption coefficient of 25,000. (13R)-Aglycon, 5. Colorless powder, [α]28D +254 (c 0.0079, MeOH); UV (MeOH) λmax (ε) 236 nm (24,000) nm; CD (MeOH); λext (Δε) 249 (+28), 228 (−37) nm; IR (film) νmax 3421, 1655, 1126 cm−1; HR ESIMS m/z 293.2104 [M + H]+ (calculated for C18H29O3 293.2111). NMR data are listed in Table 3. (13S)-Aglycon, 6. Colorless powder, [α]30D +246 (c 0.0052, MeOH); UV (MeOH) λmax 236 nm (ε 24000); CD (MeOH) λext 247 (Δε +27), 225 (Δε −38) nm; IR (film) νmax 3421, 1652, 1594, 1125 cm−1; HR ESIMS m/z 293.2096 [M + H]+ (calculated for C18H29O3 293.2111). NMR data are listed in Table 3. Preparation of Mosher’s Esters of Aglycons. (13R)-Aglycon, 5 (0.1 mg, 0.35 μmol), was treated with (S)-methoxy(trifluoromethyl)phenylacetyl (MTPA) chloride (15 μL, 80 μmol) in dry pyridine (50 μL) at room temperature for 5 h. The reaction was quenched by adding water (10 μL) and dried. The resulting residue was chromatographed on alumina (1 g, aluminum oxide 90 activity II−III) (Merck, Kenilworth, NJ) using CHCl3 to produce (R)-MTPA ester 7r (ca. 0.1 mg). (13S)-Aglycon, 6 (0.1 mg, 0.35 μmol) was converted to (R)MTPA ester 8r (ca. 0.1 mg) and (S)-MTPA ester 8s (ca. 0.1 mg) by the same procedure with (S)-MTPA chloride and (R)-MTPA chloride, respectively. 7r. 1H NMR (CDCl3, 600 MHz) δ 7.53 (m, 2H, Ph), 7.41 (m. 3H, Ph), 6.70 (d, J = 15.6 Hz, 1H, H-8), 5.89 (s, 1H, H-4), 5.75 (d, J = 15.6 Hz, 1H, H-7), 5.38 (t, J = 7.5 Hz, 1H, H-10), 5.16 (m, 1H, H-13), 3.56 (s, 3H, OMe), 2.48 (d, J = 17.1 Hz, 1H, H-2), 2.27 (d, J = 17.1 Hz, 1H, H-2), 2.26 (m, 1H, H-11), 2.20 (m, 1H, H-11), 1.88 (s, 3H, H-16), 1.81 (s, 3H, H-15), 1.79 (m, 1H, H-12), 1.63 (m, 1H, H-12), 1.27 (d, J = 6.6 Hz, 3H, H-14), 1.10 (s, 3H, H-18), 1.00 (s, 3H, H-17).

a

Interchangeable signals. 165

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry

nebulizer gas 0.2 L/min, nozzle potential 200 V, flow rate 5 μL/min (split ratio of LC/MS, 20:1). The lycosides were detected on an extracted ion chromatograph (XIC) by using m/z 639.4 [M + Na]+ for 1 and 2 or 477.3 [M + Na]+ ion for 3 and 4 at approximately 8.8 and 9.7 min, respectively. The ionization efficiency of 1 and 2 (and 3 and 4) was assumed to be the same in this analysis. The peak areas of XICs obtained by 15 and 47 pmol/injection of 1 and 4 were used to obtain the standard curves for quantitative analysis.

Table 3. 1H and 13C NMR Data for Aglycons, 5 and 6, in CDCl3a (13R)-aglycon, 5

(13S)-aglycon, 6

position

δC

δH (mult., J in Hz)

δC

δH (mult., J in Hz)

1 2

41.4 49.8

41.0 49.8

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

198.1 126.7 163.1 79.8 128.6 127.7 131.5 131.5 23.8 39.2 67.4 23.7 20.6 19.0 24.1 23.0

2.26 (d, 17.2), 2.47 (d, 17.2) 5.92 (s) 5.73 (d, 15.8) 6.78 (d, 15.8) 5.45 (t, 7.4) 2.29 (m) 1.52 (m) 3.77 (m) 1.20 (d, 6.0) 1.82 (s) 1.91 (d, 1.2) 1.01 (s) 1.10 (s)

2.25 (d, 17.0), 2.46 (d, 17.0) 5.93 (s) 5.71 (d, 15.6) 6.77 (d, 15.6) 5.44 (t, 7.8) 2.30 (m) 1.50 (m) 3.77 (m) 1.20 (d, 6.0) 1.82 (s) 1.91 (d, 1.2) 1.02 (s) 1.10 (s)

198.0 126.7 163.0 79.7 128.7 127.8 131.6 131.6 23.7 39.0 67.3 23.7 20.6 19.0 24.2 23.0

■

RESULTS AND DISCUSSION Isolation. The inhibitory activity against the zoosporangia formation was evaluated with P. capsici NBRC 30696. The strain was inoculated on a potato-sucrose-agar medium that contained a sample to be tested and incubated for 6−10 d. The number of zoosporangia formed on the mycelia was compared to a control. In a small-scale fractionation, one can of V8 vegetable juice was extracted with 40% MeOH, and the extract was chromatographed on ODS to produce 19 fractions, a few of which showed significant inhibitory activity (fractions 7−9) (Figure 2A). The active fractions were then subjected to silica gel column chromatography to generate one inhibitory fraction (fraction 1) (Figure 2B). This preliminary result prompted us to identify the natural inhibitors of the Phytophthora zoosporangia formation. For the separation of the target compounds, V8 vegetable juice (6.8 L) was extracted with MeOH, and the soluble part was chromatographed on ODS. The inhibitory fraction was then subjected to further chromatographic steps by using the bioassay described above. The final HPLC produced four components named lycosides A−D, 1−4, in the yields of 0.5, 1.2, 0.6, and 0.7 mg, respectively. Structural Elucidation. The most abundant compound, lycoside B, 2 was first utilized to analyze its structure. The molecular formula C30H48O13 was determined by the [M + H]+ ion peak in high-resolution electrospray ionization mass spectra (ESI MS). A broad and intense infrared (IR) absorption at 3397 cm−1 suggested the presence of a number of hydroxy groups. Two IR bands at 1655 and 1594 cm−1 and the ultraviolet (UV) maximum at 236 nm could indicate conjugated diene and/or enone functionalities. 1H NMR data (Table 1) and a 1H−1H correlation spectroscopy (DQF-COSY) experiment revealed the following substructures: CH3−CH(O)−CH2−CH2− CH, −CHCH−, and some sugar-related connectivities (Figure 3). 13C NMR data (Table 2) indicated the presence of five quaternary carbons due to a ketone (δ 201.2, C3), two sp3 carbons (δ 42.7, C1; 80.7, C6), and two olefinic carbons (δ 154.6, C5; 132.5, C9). The connectivities of these substructures, as well as two sugar moieties, were finally revealed by heteronuclear multiple bond coherence (HMBC) experiments (Figure 3). The disaccharide structure was determined to be 4O-β-D-glucopyranosyl-β-D-glucopyranose (β-cellobiose) based on the J values of 7.6 (H1′-H2′) and 8.0 Hz (H1″-H2″), nuclear Overhauser effects correlation spectroscopy (NOESY) correla-

a

Observed at 400 MHz for 1H and 100 MHz for 13C.

8r. 1H NMR (CDCl3, 400 MHz) δ 7.52 (m, 2H, Ph), 7.41 (m. 3H, Ph), 6.66 (d, J = 15.9 Hz, 1H, H-8), 5.91 (s, 1H, H-4), 5.73 (d, J = 15.9 Hz, 1H, H-7), 5.35 (t, J = 7.2 Hz, 1H, H-10), 5.17 (m, 1H, H-13), 3.56 (s, 3H, OMe), 2.45 (d, J = 16.5 Hz, 1H, H-2), 2.26 (d, J = 16.5 Hz, 1H, H-2), 2.18 (m, 1H, H-11), 2.10 (m, 1H, H-11), 1.90 (s, 3H, H-16), 1.79 (s, 3H, H-15), 1.72 (m, 1H, H-12), 1.51 (m, 1H, H-12), 1.36 (d, J = 6.6 Hz, 3H, H-14), 1.10 (s, 3H, H-18), 1.00 (s, 3H, H-17). 8s. 1H NMR (CDCl3, 600 MHz) δ 7.52 (m, 2H, Ph), 7.41 (m. 3H, Ph), 6.69 (d, J = 15.3 Hz, 1H, H-8), 5.90 (s, 1H, H-4), 5.74 (d, J = 15.3 Hz, 1H, H-7), 5.39 (t, J = 7.2 Hz, 1H, H-10), 5.15 (m, 1H, H-13), 3.56 (s, 3H, OMe), 2.46 (d, J = 17.1 Hz, 1H, H-2), 2.28 (m, 1H, H-11), 2.26 (d, J = 17.1 Hz, 1H, H-2), 2.19 (m, 1H, H-11), 1.89 (d, J = 1.2 Hz, 3H, H-16), 1.81 (s, 3H, H-15), 1.79 (m, 1H, H-12), 1.64 (m, 1H, H-12), 1.28 (d, J = 6.6 Hz, 3H, H-14), 1.09 (s, 3H, H-18), 0.99 (s, 3H, H-17). Quantitative Analysis of Lycosides in Vegetables and Juices. A fresh vegetable (100 g) was homogenized in MeOH (100 mL) in a blender, and the volume was adjusted to 200 mL with 50% MeOH. Vegetable juice (100 mL) was mixed with MeOH (100 mL). In both cases, the mixture was stirred at room temperature for 1 h. The solid part was separated by centrifugation and washed with 10% MeOH (50 mL). The supernatant and washing solutions were combined and concentrated to a MeOH-free aqueous solution that was dissolved in 30% MeOH (50 mL in total) and chromatographed on ODS (Cosmosil 75C18−OPN, 10 g). The fraction eluted with 60% MeOH was concentrated to dryness and was thoroughly suspended in MeOH (0.5 mL) and centrifuged at 30,000 rpm for 3 min. A portion (1 μL) of the supernatant was analyzed by LC/MS. The analytical conditions were as follows: LC: Unison UK-C8 column (2 × 75 mm, Imtakt, Kyoto Japan), 50% MeOH (0.3 mM HCOOH/0.3 mM HCOONa), flow rate 0.1 mL/min. MS: ionization and detection ESI-TOF, positive mode,

Figure 3. Key 2-dimensional NMR correlations of 2. Bold lines: DQF-COSY; arrows: HMBC; dotted arrows: NOESY. 166

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry

effect could reflect the spacial arrangement of the two conjugated systems and not be affected by the sugar moiety and the C9 geometry. This result strongly indicates the 6S configuration of the lycosides. The absolute configuration at C-13 was next examined by the modified Mosher’s method.11 The aglycon 6 was converted to (S)- and (R)-MTPA esters, 8s and 8r, respectively, and the chemical shift difference, δ(8s) − δ(8r), was calculated (Figure 5). The result clearly demonstrated the 13S configuration of 6. To confirm the 13R configuration of the aglycon 5, 5 was also converted to (R)-MTPA ester 7r. The 1H NMR data of 7r was almost identical to that of 8s (Figure 5), indicating that the C7− C14 portions of 7r and 8s are mirror images of each other. To summarize the above analyses, the absolute stereochemistry of the lycosides is shown in Figure 1. The aglycons possess a rare carbon skeleton, and only two examples are known.12 Glycosides of this type of aglycon have not been reported previously. The skeleton is possibly biosynthesized by the degradation of a carotenoid such as zeaxanthin in the manner in which abscisic acid (ABA) biosynthesis occurs in higher plants, although the oxidative cleavage occurs at C13−C14 in this case but not at C11−C12 for the ABA biosynthesis. Inhibition of Zoosporangia Formation. The inhibitory activity of the lycosides 1−4 and aglycons 5 and 6 against the zoosporangia formation of P. capsici was evaluated by the biological test described above. The inhibitory curves resulted in IC50 values of 2.1, 3.5, 4.1, 7.6, 1.2, and 1.8 μM for 1−6, respectively (Figure 6A). Among the lycosides, 1 was most active, and the two β-cellobiosides 1 and 2, were more active than the β-glucosides 3 and 4. Interestingly, the aglycons, 5 and 6, were more active than the lycosides, suggesting that the sugar moiety of the lycosides is not necessarily important for the inhibitory activity. The compounds with the 13R configuration always showed slightly higher activity than the corresponding 13S isomers. However, the SAR is not very clear because the difference was not significant, and another bioassay test produced somewhat different activity (data not shown). The universality of the activity within the Phytophthora genus was also examined by using other species. Lycoside D, 4, was administered to three additional strains, P. cinnamomi NBRC 30697, P. nicotianae NBRC 9049, and P. infestans PI 1234-1. Their asexual reproduction was inhibited by 4 at the IC50 values of 34, 5.5, and 12 μM, respectively (Figure 6B). Although P. cinnamomi was slightly susceptible, the lycosides could be universal inhibitors against Phytophthora. Distribution of Lycosides in Vegetables. The distribution of the lycosides was then examined by using several vegetable juices and fresh vegetables. To quantitate the lycosides, lycosides A, 1, and D, 4, were analyzed by LC/MS. These compounds were clearly detected at 15 pmol/injection (Figure 7). Since lycoside B, 2, and C, 3, showed the same retention times as those of 1 and 4, respectively, the lycoside contents were obtained as the sum of 1 + 2 and the sum of 3 + 4. Several commercial vegetable juices and fresh vegetables were extracted with MeOH, and the extracts were roughly separated by the ODS column. The 60% MeOH fraction was used for the quantitative analysis of the lycosides. Among the six fresh vegetables (tomato, lettuce, egg plants, carrot, parsley, celery, and paprika) that are used to prepare V8 vegetable juice, only tomato contained the lycosides (Figure 8A). Several commercial tomato cultivars were also analyzed, revealing that the lycoside contents largely depended on the cultivars (Figure 8A). In

tions of H1′-H5′, H1“-H3″, and H1“-H5” (Figure 3), and the comparison of 13C chemical shifts (Table 2) with a database.9 The Z geometry of the trisubstituted double bond at C9 was concluded by the NOESY correlations of H8−H11 and H10− H15 (Figure 3). Based on these analyses, the planar structure of 2 was determined as shown in Figure 3. Lycoside A, 1, C30H48O13, was found to be an isomer of 2 based on the high-resolution ESI MS data. The NMR data (Tables 1 and 2) were very similar to those of 2 except for the small 1H chemical shift difference through H11 to H14, suggesting that 1 is the 13-epimer of 2. The 2D NMR data also confirmed the planar structure of 1. Lycoside C, 3, has the molecular formula of C24H38O8 as determined by high-resolution ESI MS. The NMR data indicated that 3 possessed the same aglycon as that of 1 and only one β-D-glucose (Tables 1 and 2). This was strongly supported by the distinct high field shifts at position 4′ of the glucose moiety: δ 3.57 to 3.30 (H4′) and δ 80.7 to 71.8 (C4′). The 2D NMR analysis confirmed the structure of 3. Lycoside D, 4, C24H38O8 was found to be an isomer of 3 based on the high-resolution ESI MS data. The NMR data (Tables 1 and 2) were very similar to those of 3 except for the small chemical shift difference through H11 to H14, indicating that 4 is the 13-epimer of 3, similar to the relationship between 1 and 2. The 2D NMR analyses including the NOESY experiment confirmed the structure of 4. To confirm the D-glucoside structures, lycosides A, 1, and C, 3, were treated with a β-D-glucosidase that produced the same aglycon 5. The structure of 5 was then confirmed by 1D and 2D NMR analysis. Lycoside D, 4, was also treated with the same enzyme to produce aglycon 6. The NMR data for these aglycons are indistinguishable, as summarized in Table 3. The reactivity of the lycosides against β-D-glucosidase supports that the sugar moieties consist of D-glucose. The aglycons, 5 and 6, share the major part of their structure with abscisic acid (Figure 4). To determine the absolute

Figure 4. Structures of aglycons, 5 and 6, and (+)-abscisic acid.

configuration at position C-6, the circular dichroism (CD) spectra of all the lycosides, 1−4, and their aglycons, 5 and 6, were first obtained and compared with a reference CD data of abscisic acid. All the compounds showed highly similar positive (∼250 nm) and negative (∼227 nm) Cotton effects that were similar to that of (+)-abscisic acid10 (Table 4). This Cotton Table 4. CD Data for Lycosides and Aglycons in MeOH compounds

λext, nm (Δε)

lycoside A (1) lycoside B (2) lycoside C (3) lycoside D (4) (13R)-aglycon (5) (13S)-aglycon (6) (+)-abscisic acid10

251 (+36), 227 (−39) 250 (+31), 226 (−33) 249 (+35), 227 (−30) 250 (+30), 227 (−30) 249 (+28), 228 (−37) 247 (+27), 228 (−38) 261 (+34.5), 229 (−28.0) 167

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry

Figure 5. Determination of the C-13 configuration by a modified Mosher’s method. (A) Chemical shift difference of (S)-MTPA ester 8s and (R)MTPA ester 8r derived from aglycon 6. (B) Chemical shifts of (R)-MTPA ester 7r derived from 5 were almost identical to those of 8s.

Figure 6. Inhibition of Phytophthora zoosporangia formation by lycosides and aglycons. (A) Inhibitory activity of lycosides A−D, 1−4, and their aglycons, 5 and 6, against P. capsici NBRC 30697. (B) Inhibitory activity of lycoside D, 4, against four Phytophthora strains. The curves are drawn by the sigmoid curve fitting. Error bars are omitted here.

Figure 7. LC/MS analysis of lycosides A, 1, and D, 4. (A) ESI-TOF mass spectrum of 1 and extracted ion chromatogram (XIC, inset) using the m/z 639.4 [M + Na]+ ion at a 15 pmol injection. (B) Mass spectrum of 4 and XIC (inset) using the m/z 477.3 [M + Na]+ ion at a 15 pmol injection.

Figure 8. Lycoside contents in fresh vegetables and vegetable juices. (A) Lycosides in fresh vegetables; (B) Lycosides in Japanese vegetable juices. Each data point is obtained from a single experiment. 168

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169

Article

Journal of Agricultural and Food Chemistry

(12) Seger, C.; Hofer, O.; Vajrodaya, S.; Greger, H. Two new norditerpenes from Glycosmis cf. cyanocarpa. Nat. Prod. Lett. 1998, 12, 117−124.

addition, several commercial vegetable juices were found to universally contain the lycosides at approximately 0.2−0.8 μM in total (Figure 8B). These results suggest that the lycosides are widely distributed at least in tomatoes and stable enough to be preserved in a can for foods. Since the lycosides are food chemicals, they or their derivatives could be a promising candidate to be ecologically friendly pesticides. In addition, the quantitative analysis of the lycosides in pathogen-resistant tomatoes will be valuable to know the correlation between the resistance and the lycoside productivity in tomatoes.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.7b04766.

■

Key 2D NMR correlations, spectral data for new compounds, and inhibition data of zoosporangia formation by compounds (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +(81) 527894116. Fax: +(81) 527894118. ORCID

Makoto Ojika: 0000-0002-6671-8598 Funding

This work was supported by Japan Society for the Promotion of Science (JSPS) for KAKENHI (No. 26252015). Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Fry, W. Phytophthora infestans: the plant (and R gene) destroyer. Mol. Plant Pathol. 2008, 9, 385−402. (2) Hansen, E. M.; Reeser, P. W.; Sutton, W. Phytophthora beyond agriculture. Annu. Rev. Phytopathol. 2012, 50, 359−378. (3) Erwin, D. C.; Ribeiro, O. K. Phytophthora diseases worldwide. American Phytopathological Society, MN, USA, 1996. (4) Haverkort, A. J.; Struik, P. C.; Visser, R. G. F.; Jacobsen, E. Applied biotechnology to combat late blight in potato caused by. Potato Res. 2009, 52, 249−264. (5) Forbes, G. Global overview of late blight. Proceedings regional workshop on potato late blight for east and southeast Asia and the Pacific; Yezin Agricultural University, 2004; pp 3−10. (6) Liljeroth, E.; Lankinen, A.; Wiik, L.; Burra, D. D.; Alexandersson, E.; Andreasson, E. Potassium phosphite combined with reduced doses of fungicides provides efficient protection against potato late blight in large-scale field trials. Crop Prot. 2016, 86, 42−55. (7) Judelson, H. S.; Blanco, F. A. The spores of Phytophthora: weapons of the plant destroyer. Nat. Rev. Microbiol. 2005, 3, 47−58. (8) Vedenyapina, E. G.; Safir, G. R.; Niemira, B. A.; Chase, T. E. Low concentrations of the isoflavone genistein influence in vitro asexual reproduction and growth of Phytophthora sojae. Phytopathology 1996, 86, 144−148. (9) Bock, K.; Pedersen, C.; Pedersen, H. Carbon-13 nuclear magnetic resonance data for oligosaccharides. Adv. Carbohydr. Chem. Biochem. 1984, 42, 193−225. (10) Koreeda, M.; Weiss, G.; Nakanishi, K. Absolute configuration of natural (+)-abscisic acid. J. Am. Chem. Soc. 1973, 95, 239−240. (11) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. High-field FT NMR application of Mosher’s method. The absolute configurations of marine terpenoids. J. Am. Chem. Soc. 1991, 113, 4092−4096. 169

DOI: 10.1021/acs.jafc.7b04766 J. Agric. Food Chem. 2018, 66, 163−169