Steroidal Saponins in Oat Bran - American Chemical Society

Feb 7, 2016 - Agricultural and Technical State University, North Carolina Research Campus, 500 Laureate Way, Kannapolis, North Carolina 28081,...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/JAFC

Steroidal Saponins in Oat Bran Junli Yang, Pei Wang, Wenbin Wu, Yantao Zhao, Emmanuel Idehen, and Shengmin Sang* Laboratory for Functional Foods and Human Health, Center for Excellence in Post-Harvest Technologies, North Carolina Agricultural and Technical State University, North Carolina Research Campus, 500 Laureate Way, Kannapolis, North Carolina 28081, United States ABSTRACT: Saponins are one type of widespread defense compound in the plant kingdom and have been exploited for the production of lead compounds with diverse pharmacological properties in drug discovery. Oats contain two unique steroidal saponins, avenacoside A, 1, and avenacoside B, 2. However, the chemical composition, the levels of these saponins in commercial oat products, and their health effects are still largely unknown. In this study, we directly purified 5 steroidal saponins (1−5) from a methanol extract of oat bran, characterized their structures by analyzing their MS and NMR spectra, and also tentatively identified 11 steroidal saponins (6−16) on the basis of their tandem mass spectra (MSn, n = 2−3). Among the five purified saponins, 5 is a new compound and 4 is purified from oats for the first time. Using HPLC-MS techniques, a complete profile of oat steroidal saponins was determined, and the contents of the two primary steroidal saponins, 1 and 2, were quantitated in 15 different commercial oat products. The total levels of these two saponins vary from 49.6 to 443.0 mg/kg, and oat bran or oatmeal has higher levels of these two saponins than cold oat cereal. Furthermore, our results on the inhibitory effects of 1 and 2 against the growth of human colon cancer cells HCT-116 and HT-29 showed that both had weak activity, with 2 being more active than 1. KEYWORDS: oat bran, steroidal saponin profile, avenocoside D, cytotoxic effect



fungi.21 To date there are only six steroidal saponins purified from oat brans, including avenacoside A, 1,22 avenacoside B, 2,23 and avenacoside C, 3,24 and 26-desglucoavenacosides A and B11,25,26 as well as one sulfated saponin.19 Saponins are one type of widespread defense compound in the plant kingdom,27 and they are mainly characterized for their antimicrobial effects and less frequently for insecticidal properties.28 Apart from their important role in plant defense systems, more and more saponins have been utilized for the production of lead compounds with diverse pharmacological properties; one such property is their anticancer effects.28−31 The chemical profile and the anticancer effects of oat steroidal saponins are still unknown. In addition, the levels of these saponins in commercial oat products have not been reported. In this regard, a systematic investigation on oat steroidal saponins was conducted here. The objective of the present study was to explore more avenacoside-type components from oats, give a profile of steroidal saponins in oats, quantitate their levels in commercial oat products, and evaluate their inhibitory effects on the growth of human colon cancer cells.

INTRODUCTION Oats (Avena sativa L.) have been considered as one of the healthiest foods worldwide.1,2 Oat grains are able to thrive in poor soil conditions.3 Most oat products are made from a hulled grain without stripping their bran and germ, and these parts retain large amounts of dietary fiber and bioactive phytochemicals, which display a serum cholesterol lowering effect and reduce the risk of heart disease and cardiovascular disease.4−7 Besides, consumption of oat products showed other health benefits, such as anticancer and antidiabetic effects,8 enhancing human immunity,9 and reducing the risk of high blood pressure.10 Oats produce a series of phytochemicals contributing to their health-related effects including steroidal saponins,11 avenanthramides,12 phenolic acids,13,14 tocols,15 and flavanoids.11 As the only saponin-accumulating cereal,16 oats contain two different saponin forms, avenasides and avenacosides, synthesized via two different biosynthetic pathways.17 Avenasides are triterpenoid saponins mainly stored in roots for inhibiting pathogens such as Gaeumannomyces graminis,17 whereas avenacosides belong to steroid glycosides and are mainly accumulated in oat leaves and grains.18,19 Avenacosides A and B, the two primary and also unique avenacosides in oats, are glycosylated at C-3 with a trisaccharide (one rhamnose and two glucose units) in the case of avenacoside A or a tetrasaccharide (one rhamnose and three glucose units) in the case of avenacoside B, and at C-26 with a glucose unit (Figure 1). Upon tissue disruption, the O-βglucosidic bond at C-26 is immediately hydrolyzed by a special β-glucosidase, named avenacosidase, to yield the bioactive 26-deglucoavenacosides,20 which possess strong antifungal activity.18 The sugar moieties at C-3 are essential for the antimicrobial effects of 26-deglucoavenacosides,21 and these saponins can be detoxified via sequential hydrolysis of the sugar units at C-3 by α-rhamnosidase and β-glucosidase secreted by pathogenic © 2016 American Chemical Society



MATERIALS AND METHODS

Materials. Silica gel (230−400 mesh) (Sorbent Technologies Inc., Atlanta, GA, USA) and Diaion HP-20 (Mitsubishi Chemical, Japan) were used for open column chromatography (CC). Chromatographic separations were monitored by analytical thin-layer chromatography (TLC) on 250 μm thick, 2−25 μm particle size, glass-backed silica gel plates, which were purchased from Sigma (Sigma-Aldrich, St. Louis, Received: Revised: Accepted: Published: 1549

December 22, 2015 February 4, 2016 February 7, 2016 February 7, 2016 DOI: 10.1021/acs.jafc.5b06071 J. Agric. Food Chem. 2016, 64, 1549−1556

Article

Journal of Agricultural and Food Chemistry

Figure 1. Structures of compounds 1−16 identified from oat bran. ∗∗, new compound; ∗, first purification from oat. MO, USA). All analytical and HPLC-MS grade solvents were obtained from Thermo Fisher Scientific (Waltham, MA, USA). All of the oat products were purchased online at Walmart and from a local supermarket, Harris Teeter (Kannapolis, NC, USA). HPLC-MS Analysis. HPLC-MS was performed with a ThermoFinnigan Spectra System consisting of an Ultimate 3000 degasser, an Ultimate 3000 RS pump, an Ultimate 3000 RS autosampler, an Ultimate 3000 RS column compartment, and an LTQ Velos Pro ion trap mass spectrometer (Thermo Electron, San Jose, CA, USA) incorporated with an electrospray ionization (ESI) interface. The column used was a 150 mm × 3.0 mm i.d., 5 μm, Gemini RP-18 (Phenomenex,

Torrance, CA, USA). The mobile phase consisted of water containing 0.2% formic acid (mobile phase A) and methanol with 0.2% formic acid (mobile phase B). The gradient elution was carried out for 60 min at a flow rate of 0.2 mL/min. A gradient eluting system was applied: 40% B from 0 to 3 min; 40−52% B from 3 to 35 min; 52−100% B from 35 to 45 min; 100% B from 45 to 50 min, and then to 40% B from 50 to 55 min. The column was then re-equilibrated with 0% B for 5 min. The injection volume was 10 μL for each sample. The HPLC eluent was introduced into the ESI interface. For mass spectrometric parameter optimization, the purified compound in methanol solution (10 μg/mL) was infused in ESI source and analyzed in negative ion 1550

DOI: 10.1021/acs.jafc.5b06071 J. Agric. Food Chem. 2016, 64, 1549−1556

Article

Journal of Agricultural and Food Chemistry

Preparation of Standards of 1 and 2 and the Extracts of Commercial Oat Products. The stock solutions (0.1 mg/mL) of 1 and 2 were prepared in 50% (v/v) aqueous methanol solution. Stock solutions were stored at −20 °C before use. The above stock solutions were diluted with 50% methanol to prepare 0.5, 0.75, 1.5, 2.5, 5.0, and 10.0 μg/mL 1 and 0.25, 0.375, 0.75, 1.25, 2.5, and 5.0 μg/mL 2, respectively. All of the samples were freshly prepared before use. Three independent samples of each oat product were used in this study. One gram of each oat product was accurately weighed and extracted three times with 50 mL of methanol under sonication for 30 min and then cooled and centrifuged at 8000 rpm for 10 min. Supernatants from the three extractions were combined and concentrated to dryness under vacuum at 35 °C. The residue was reconstituted in 2.0 mL of 50% methanol and centrifuged for 10 min at 16000 rpm. Before injection, the supernatant of each sample was diluted 10 times (for oat cereal) or 20 times (for oat bran and oatmeal) with 50% methanol. Each sample was analyzed in triplicate. Growth Inhibitory Effects of 1 and 2 on Human Colon Cancer Cells. Cell growth inhibition was determined by a 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay. Human colon cancer cells HCT-116 and HT-29 were plated in 96-well microplates with 5000 cells/well and allowed to attach for 24 h at 37 °C. The test compounds (in DMSO) were added to cell culture medium to desired final concentrations (25, 50, 75, 100, 150, and 200 μM, the final DMSO concentrations for control and treatments were 0.1%, n = 8−16). After the cells had been cultured for 72 h, the medium was aspirated, and cells were treated with 200 μL of fresh medium containing 2.41 mmol/L MTT. After incubation for 3 h at 37 °C, the medium containing MTT was aspirated, 100 μL of DMSO was added to solubilize the formazan precipitate, and the plates were shaken gently for 1 h at room temperature. Absorbance values were derived from the plate reading at 550 nm on a Biotek microtiter plate reader. The reading reflected the number of viable cells and was expressed as a percentage of viable cells in the control. Both HCT-116 and HT-29 cells were cultured in McCoy’s 5A medium. All of the above media were supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 1% glutamine, and the cells were kept in a 37 °C incubator with 95% humidity and 5% CO2. IC50 values were obtained using GraphPad Prism (GraphPad Software, San Diego, CA, USA).

mode to give the following optimized parameters: spray voltage, 3.6 kV; sheath gas (nitrogen) flow rate, 34 (arbitrary units); capillary voltage, −13 V; capillary temperature, 300 °C; tube lens offset, −60 V. For the quantification of the two major saponins, target ions at m/z 1061.7 [M − H]− for 1 and at m/z 1223.9 [M − H]− for 2 were monitored using the selected ion monitoring (SIM) mode. For the identification of steroidal saponins, the collision-induced dissociation (CID) was conducted with an isolation width of 1.2 Da, and the normalized collision energy was set to 35% for MSn analysis. The mass range was measured from m/z 50 to 1400. Data acquisition and analysis were performed with Xcalibur 2.0 version (Thermo Electron). Nuclear Magnetic Resonance (NMR) Analysis. 1H (600 MHz), 13 C (150 MHz), heteronuclear single-quantum correlation (HSQC), and heteronuclear multiple-bond correlation (HMBC) NMR spectra were recorded on a Bruker 600 MHz NMR instrument. All samples were dissolved in methanol-d4 containing tetramethylsilane (TMS) as the internal standard. Extraction and Enrichment of Steroidal Saponins. Oat bran (50 kg) purchased from Kalyx (www.kalyx.com) was continuously extracted by 100% methanol (Voat:Vmethanol = 1:5) at room temperature three times for 4 days each time. After filtration using cotton, the methanol extract was concentrated under reduced pressure to yield a crude residue (2136 g). This residue was reconstituted in water and partitioned against n-hexane, ethyl acetate (EtOAc), and n-butanol (n-BuOH). After concentration in vacuo, the n-BuOH fraction (242.8 g) was suspended in water and applied to a Diaion HP-20 column (7.5 cm i.d. × 60 cm) eluted with water, 30% ethanol in water, 70% ethanol in water, and ethanol successively (5 L each) to afford four fractions, F1−F4, respectively. Fraction F3 (70% ethanol elution) was evaporated in vacuo and kept as the steroidal-saponin enriched sample (8.9 g) at −80 °C. Fraction monitoring was by TLC (chloroform/methanol/ water, 70:35:5.5, v/v/v). The spots on TLC were visualized by spraying with a H2SO4/ethanol (5:95, v/v) solution followed by heating. Purification of Steroidal Saponins 1−5. Repeated purification of fraction F3 by silica gel open column (5.0 cm i.d. × 30 cm) eluted with a chloroform/methanol/water system (70:30:5.5 to 70:35:5.5, v/v/v, 2 L each) afforded five steroidal saponins: 1 (1.2 g), AA; 2 (609 mg), AB; 3 (10.5 mg), AC; 4 (10.9 mg), chamaedroside E2; and a new saponin, avenacoside D, 5 (5.8 mg). Chamaedroside E2 (4): yellow amorphous powder; 1H NMR (600 MHz, in methanol-d4) δH 5.37 (1H, br s, H-6), 4.45 (1H, m, H-16), 4.41 (1H, d, J = 7.8 Hz, H-1″), 4.40 (1H, d, J = 7.8 Hz, H-1‴), 4.29 (1H, d, J = 7.7 Hz, H-1′), 3.80 (1H, d, J = 12.0 Hz, H-26a), 3.50 (1H, dd, J = 11.8, 2.6 Hz, H-3), 3.47 (1H, d, J = 11.2 Hz, H-26b), 1.22 (3H, s, H-27), 1.04 (3H, s, H-19), 0.98 (3H, d, J = 6.8 Hz, H-21), 0.81 (3H, s, H-18); 13C NMR (150 MHz, in methanol-d4) δC 142.0 (C-5), 122.6 (C-6), 121.7 (C-22), 105.0 (C-1′), 104.6 (C-1‴), 102.3 (C-1″), 63.2/62.8/61.9 (C-6′/6″/6‴), 84.2 (C-25), 81.1 (C-16), 78.9 (C-3), 77.9 (C-26), 61.4 (C-17), 56.7 (C-14), 50.7 (C-9), 41.5 (C-20), 40.9 (C-12), 39.7 (C-4), 39.4 (C-13), 38.5 (C-1), 37.0 (C-10), 32.7 (C-24), 32.6 (C-8), 32.6 (C-7), 32.2 (C-15), 31.8 (C-23), 30.0 (C-2), 23.3 (C-27), 21.0 (C-11), 16.9 (C-19), 15.6 (C-18), 14.1 (C-21); negative ESI/MS, m/z 915.7 [M − H]− and 961.7 [M + HCOOH − H]−. Avenacoside D (5): yellow amorphous powder; 1H NMR (600 MHz, in methanol-d4) δH 5.33 (1H, br s, H-6), 5.20 (1H, br s, H-1″″), 4.51 (1H, d, J = 7.8 Hz, H-1″″″), 4.47 (1H, d, J = 7.8 Hz, H-1″), 4.42 (1H, d, J = 7.8 Hz, H-1″″′), 4.40 (1H, overlap, H-16), 4.35 (1H, d, J = 7.8 Hz, H-1‴), 4.25 (1H, d, J = 7.7 Hz, H-1′), 3.81 (1H, d, J = 11.2 Hz, H-26a), 3.59 (1H, br d, J = 11.6 Hz, H-3), 3.50 (1H, d, J = 11.2 Hz, H-26b), 1.19 (3H, d, J = 7.6 Hz, H-6″″), 1.18 (3H, s, H-27), 1.00 (3H, s, H-19), 0.90 (3H, d, J = 6.8 Hz, H-21), 0.76 (3H, s, H-18); 13C NMR (150 MHz, in methanol-d4) δC 140.9 (C-5), 121.6 (C-6), 120.7 (C-22), 104.2 (C-1′), 104.0 (C-1″″″), 103.6 (C-1″″′), 103.2 (C-1‴), 101.0 (C-1″″), 99.4 (C-1″), 84.2 (C-25), 81.1 (C-16), 78.4 (C-3), 77.1 (C-26), 62.2/61.7/ 61.6/61.5/60.9 (C-6′/6″/6‴/6″″′/6″″″), 61.4 (C-17), 56.7 (C-14), 50.7 (C-9), 40.5 (C-20), 40.5 (C-12), 39.9 (C-4), 39.8 (C-13), 38.5 (C-1), 37.0 (C-10), 32.7 (C-24), 32.7 (C-7), 32.6 (C-8), 32.2 (C-15), 31.8 (C-23), 29.7 (C-2), 23.3 (C-27), 20.9 (C-11), 18.8 (C-6″″), 16.9 (C-19), 15.6 (C-18), 14.1 (C-21); negative ESI/MS, m/z 1385.8 [M − H]− and 1431.8 [M + HCOOH − H]−.



RESULTS AND DISCUSSION Structural Elucidation of Steroidal Saponins 1−5. As part of our efforts to complete the chemical profile of oat bran, five steroidal saponins 1−5 (Figure 1) were isolated by means of chromatographic methods, including silica gel and Diaion HP-20 chromatography. Compounds 1−3 were identified as avenacosides A, B, and C, respectively, according to literature data.22,32,33 Avenacoside A, 1, and avenacoside B, 2, have been reported as the primary saponins in oat bran and analyzed by HPLC-TQ-MS and HPLC with UV detection.32,33 Compound 3 was first discovered from fresh bulbs of Lilium brownii34 and subsequently isolated from oat bran and named avenacoside C by Lu et al.24 Compound 4 gave a deprotonated ion at m/z 915.7 [M − H]− in an HPLC-MS spectrum. The MS/MS spectrum of the precursor ion at m/z 915.7 displayed a fragment ion at m/z 753.5, generated by the loss of a glucose unit. 1H and 13C NMR spectra of this compound showed glycone signals similar to those of AA, 1, and AB, 2. The difference lies in the number of sugar units. There are three sugar units found in the NMR spectra of 4 [δH 4.41 (1H, d, J = 7.9 Hz), 4.40 (1H, d, J = 7.9 Hz), 4.29 (1H, d, J = 7.7 Hz); δC 105.0 (CH), 104.6 (CH), 102.3 (CH)], whereas AA, 1, and AB, 2, have four and five sugar units, respectively. On the basis of HSQC and HMBC data, 4 was determined as chamaedroside E2, which has been reported as a chemical component from Veronica chamaedrys L. 1551

DOI: 10.1021/acs.jafc.5b06071 J. Agric. Food Chem. 2016, 64, 1549−1556

Article

Journal of Agricultural and Food Chemistry

Figure 2. HMBC correlations (from H to C) of compound 5.

Figure 3. ESI/MSn (n = 2−3) spectra and fragmentation pattern of (A) compound 3 and (B) compound 1.

and 1.19 (3H, d, J = 6.5 Hz); δC 101.0 and 18.8]. The NMR spectra also showed diagnostic signals for one trisubstituted olefinic bond [δH 5.33 (1H, br s); δC 140.9 and 121.6], one methyl doublet [δH 0.91 (3H, d, J = 6.8 Hz)], and three methyl singlets [δH 1.18 (3H, s), 1.00 (3H, s), and 0.76 (3H, s)]. Finally, the linkage patterns of sugars and the whole planar structure of 5 were constructed by HSQC and HMBC techniques (Figure 2). The similarity of NMR chemical shifts between 5 and AA and AB was used to determine relative configurations of 5 as shown in Figure 1. Compound 5 is a new steroidal saponin and was named here avenacoside D. Fragmentation Patterns of Steroidal Saponins. When there is one −Glc−Rha unit at C-3 and one −Glc unit at C-26, the fragmentation priority of the sugar units is the loss of the

However, the literature did not give its full NMR assignment. Here we report the full 1H and 13C NMR data for the first time. This is the first report of this compound from oats. Compound 5 showed a molecular formula of C63H102O33 by analyzing its deprotonated ions at m/z 1385.8 [M − H]− and 1431.8 [M + HCOOH − H]−. In the MS/MS spectrum, the precursor ion at m/z 1385.8 [M − H]− gave fragment ions at m/z 1223.8, 1061.7, 899.6, 753.5, and 591.4, which were generated by the loss of the sugar units in its structure in sequence. 1H and 13C NMR data of 5 demonstrated the existence of five glucose units [δH 4.51 (1H, d, J = 7.8 Hz), 4.47 (1H, d, J = 7.8 Hz), 4.42 (1H, d, J = 7.8 Hz), 4.35 (1H, d, J = 7.8 Hz), and 4.25 (1H, d, J = 7.7 Hz); δC 104.2, 104.0, 103.6, 103.2, and 99.4] and one rhamnose unit [δH 5.20 (1H, br s) 1552

DOI: 10.1021/acs.jafc.5b06071 J. Agric. Food Chem. 2016, 64, 1549−1556

Article

Journal of Agricultural and Food Chemistry Table 1. ESI-MS and ESI-MSn (n = 2−3) Fragment Ions of Compounds 1−16 in Oat Bran tR (min)

[M − H]−

[M + HCOOH − H]−

MS2 −

MS3 −

1

27.58

1061.7

1108.0

1061.7/899.6 [M − Glc − H] (B), 753.5 [M − Glc − Rha − H]

2

27.10

1223.9

1269.7

1223.9/1061.7 [M − Glc − H]− (B), 899.7 [M − 2Glc − H]−, 753.6 [M − 2Glc − Rha − H]−

3 4 5

28.17 28.45 26.93

899.7 915.7 1385.8

945.9 961.7 1431.8

899.7/753.5 [M − Rha − H]− (B) 915.7/753.5 [M − Glc − H]− (B) 1385.8/1223.8 [M − Glc − H]− (B), 1061.7 [M − 2Glc − H]−, 899.6 [M − 3Glc − H]−, 753.5[M − 3Glc − Rha − H]−

6 7 8 9 10

30.87 32.78 22.74 27.62 21.10

899.7 899.7 915.6 915.6 1061.8

945.9 945.6 961.7 961.7 1107.8

899.7/737.5 [M − Glc − H]− (B) 899.7/737.5 [M − Glc − H]− (B) 915.6/753.5 [M − Glc − H]− (B) 915.6/753.5 [M − Glc − H]− (B) 1061.8/899.6 [M − Glc − H]− (B), 753.6 [M − Glc − Rha − H]−

11

20.15

1223.9

1269.9

1223.9/1061.7 [M − Glc − H]−, 899.6 [M − 2Glc − H]− (B), 753.5 [M − 2Glc − Rha − H]−

12

25.64

1223.9

1269.7

1223.9/1061.7 [M − Glc − H]− (B), 899.7 [M − 2Glc − H]−, 753.6 [M − 2Glc − Rha − H]−

13

25.64

1223.9

1269.8

1223.9/1077.7 [M − Rha − H]− (B), 915.7 [M − Glc − Rha − H]−, 753.6 [M − 2Glc − Rha − H]−

14

27.35

1223.9

1269.7

1223.9/1077.7 [M − Rha − H]− (B), 915.7 [M − Glc − Rha − H]−

15

25.18

1385.8

1431.8

1385.8/1223.8 [M − Glc − H]−, 1061.7 [M − 2Glc − H]− (B), 899.6 [M − 3Glc − H]−, 753.5 [M − 3Glc − Rha − H]−

16

26.63

1385.8

1431.7

1385.8/1223.8 [M − Glc − H]−, 1061.7 [M − 2Glc − H]− (B), 899.6 [M − 3Glc − H]−, 753.5 [M − 3Glc − Rha − H]−

899.6/881.6, 753.5 (B), 737.6, 591.5 753.5/591.5 (B), 573.5, 429.4 1061.7/899.7 (B), 881.6, 753.6, 591.6, 573.6 899.7/753.5(B), 737.7, 591.5, 573.4 753.5/591.6 (B), 429.4 753.5/591.4 (B), 573.5, 429.3 753.5/591.5 (B), 573.6, 429.4 1061.7/915.6, 899.6 (B), 753.5 899.6/753.5 (B), 737.6, 591.4 753.6/591.4 (B), 573.4, 429.4 737.5/591.5 (B), 429.3 737.5/591.5 (B) 753.5/591.4 (B) 753.5/591.5 (B), 573.5 899.6/753.6 (B), 737.5, 591.5 753.6/591.5 (B), 429.4 1061.7/899.7 (B), 881.6, 753.6, 591.6, 573.6 899.6/753.5 (B), 737.6, 591.5, 573.4 753.5/591.6 (B), 573.5, 429.4 1061.7/899.7 (B), 881.6, 753.6, 591.6, 573.6 899.7/753.5 (B), 737.6, 591.5, 573.4 753.6/591.6 (B), 573.5, 429.4 1077.7/915.7 (B), 573.6 915.7/753.6 (B), 591.5, 573.4 753.6/591.4 (B), 1077.7/915.6 (B) 915.7/753.6 (B), 735.6, 591.5, 573.4 1061.7/915.6, 899.6 (B), 753.5 899.6/753.5 (B), 737.6, 591.5 753.5/591.4 (B), 429.4 1061.7/915.7, 899.6 (B), 753.5 899.6/753.5 (B), 737.6, 591.4 753.6/591.5 (B), 573.5, 429.4

Figure 4. Total ion chromatogram (TIC) of compounds 1−16 in oat bran extract generated from negative HPLC-ESI/MSn.

-Rha unit instead of the −Glc unit at C-26. This is supported by the observation of m/z 753.5 [M − Rha − H]− as the base ion of the tandem mass of 3 (m/z 899.7 [M − H]−) (Figure 3A).

However, if there are additional glucose units on the -Glc-Rha unit, the additional glucose units will have priority over the −Rha unit to be cleavaged from the side chain at C-3. For example, 1553

DOI: 10.1021/acs.jafc.5b06071 J. Agric. Food Chem. 2016, 64, 1549−1556

Article

Journal of Agricultural and Food Chemistry

we hypothesized that 13 and 14 had unique side chains with −Glc−Glc−Glc−Rha or −Glc−Glc−Glc−Glc−Rha at C-3, respectively (Figure 1). The MS2 spectra of 15 and 16 had three major fragment ions at m/z 1223.8 [M − Glc − H]−, 1061.7 [M − 2Glc − H]−, and 899.6 [M − 3Glc − H]− and a major fragment ion at m/z 753.5 [M − 3Glc − Rha − H]− in the tandem mass spectra of m/z 899.6 (MS3, 899.6/1385.8), which were similar to those of 5 (Table 1). Therefore, the structures of 15 and 16 were similar to that of 5 except for the linkage between the glucose and rhamnose units of the side chain at C-3 (Figure 1). Validation of the Quantitative HPLC-MS Method. The quantitative HPLC-MS method was validated in terms of linearity, precision, and accuracy. Calibration curves were constructed by plotting the integrated peak areas (x) of chromatography versus the corresponding concentrations of the injected standard solutions (y). The calibration curves were obtained over the concentration ranges from 0.5 to 10 μg/mL for 1 and from 0.25 to 5 μg/mL for 2 with good linearity (R2 > 0.999). The limit of quantification was 0.12 μg/mL for 1 and 0.1 μg/mL for 2. The intraday variation was determined by analyzing the known concentrations of 1 and 2 in six replicates during a single day, whereas interday variation was determined in duplicate on three consecutive days, respectively. The overall intra- and interday variations were