Using Ultra-Performance Liquid Chromatography Quadrupole Time of

Sep 19, 2017 - G. xanthochymus fruits were collected from the Fruit and Spice Park (Homestead, FL, U.S.A.) in June 2008. ..... jf7b02867_si_001.pdf (4...
2 downloads 15 Views 3MB Size
Article pubs.acs.org/JAFC

Using Ultra-Performance Liquid Chromatography Quadrupole Time of Flight Mass Spectrometry-Based Chemometrics for the Identification of Anti-angiogenic Biflavonoids from Edible Garcinia Species Ping Li,†,‡ Grace Gar-Lee Yue,§,|| Hin-Fai Kwok,§,|| Chun-lin Long,*,†,⊥ Clara Bik-San Lau,*,§,|| and Edward J. Kennelly*,†,# †

College of Life and Environmental Sciences, Minzu University of China, Beijing 100081, People’s Republic of China Key Laboratory of Agro-Environment in the Tropics, Ministry of Agriculture, South China Agricultural University, Guangzhou, Guangdong 510642, People’s Republic of China § Institute of Chinese Medicine and ||State Key Laboratory of Phytochemistry and Plant Resources in West China, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, People’s Republic of China ⊥ Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650201, People’s Republic of China # Department of Biological Sciences, Lehman College and The Graduate Center, City University of New York, New York City, New York 10468, United States ‡

S Supporting Information *

ABSTRACT: Garcinia xanthochymus fruits are edible and also used in traditional medicine. Our previous work showed that the isolated natural products from G. xanthochymus fruits have displayed antioxidant activity and cytotoxicity in the colon cancer cells. In this study, we developed a strategy to correlate a zebrafish angiogenesis assay with ultra-performance liquid chromatography quadrupole time of flight mass spectrometry-based chemometric analysis to identify potential anti-angiogenic activity compounds from G. xanthochymus fruits. Primary bioactivity results showed that the methanolic extracts from aril and pericarp but not from seed have significant inhibitory effects on the growth of subintestinal vessels (SIVs) in zebrafish embryos. A total of 13 markers, including benzophenones and biflavonoids, were predicted by untargeted principal component analysis and orthogonal partial least squares discriminate analysis, which were tentatively identified as priority markers for the bioactivity related in aril and pericarp. Amentoflavone, a biflavonoid, has been found to significantly inhibit the growth of SIVs at 10 and 20 μM and downregulate the expressions of Angpt2 and Tie2 genes of zebrafish embryos. Furthermore, seven biflavonoids, volkensiflavone, fukugetin, fukugeside, GB 1a, GB 1a glucoside, GB 2a, and GB 2a glucoside, isolated from Garcinia species were evaluated for their structure−activity relationship using the zebrafish model. Only fukugetin, which was previously shown to be anticancer, was active in inhibiting the SIV growth. In this report, both amentoflavone and fukugetin, for the first time, displayed anti-angiogenic effects on zebrafish, thus demonstrating an effective and rapid strategy to identify natural products for antiangiogenesis activity. KEYWORDS: anti-angiogenesis, Garcinia xanthochymus, UPLC−QTOF−MS, PCA, biflavonoids



INTRODUCTION The genus Garcinia belongs to the family Clusiaceae, mainly distributed in tropical Asia, Africa, and Polynesia. The fruits of many species in this genus are not only edible, such as Garcinia mangostana, Garcinia hanburyi, and Garcinia xanthochymus, but have also been used as medicine for a long time.1−3 Garcinia species are considered as potential anticancer or antitumor drugs as a result of their rich sources of polyprenylated benzophenones, xanthone derivatives, and biflavonoids.4,5 For instance, gambogic acid from Garcinia species possesses significant anti-angiogenesis activity both in vitro and in vivo.6,7 Other caged polyprenylated xanthones from G. hanburyi have been evaluated by the zebrafish model as well and exhibited anti-angiogenesis activities with less toxicity. 8 However, there are still limited reports about the antiangiogenic activity on other Garcinia species, such as G. © XXXX American Chemical Society

xanthochymus, that have not been tested. Our previous work demonstrated that the chemical constituents of G. xanthochymus fruits induced apoptosis in colon cancer cells and displayed antioxidant activity,9,10 leading to our further investigations on their potential anti-angiogenic activity using the zebrafish model. Angiogenesis plays an important role for tumor growth. The vascular endothelial growth factor (VEGF) signal pathway is the key to the regulation of physiological and pathological angiogenesis,11 and recombinant antibodies and targeting the VEGF pathway are proven strategies for antitumoral therapies Received: June 22, 2017 Revised: August 11, 2017 Accepted: September 6, 2017

A

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry in humans.12 During the past few decades, a few inhibitors of angiogenesis, such asvelcade, avastin, tarceva, and aflibercept, have been approved by the U.S. Food and Drug Administration as anti-angiogenesis drugs.13 Identification of novel antiangiogenesis compounds from natural products is a promising strategy. The zebrafish (Danio rerio) is a useful vertebrate model for angiogenesis as a result of the high fecundity of the adult, the optical transparency of embryos and larvae, short generation times, adaptability to high-throughput systems,14 and similar active effect on mammal models.15 Inhibition of the subintestinal vessels (SIVs) of the embryo is visualized in the primary screen;16 thus, zebrafish can serve as a rapid and useful assay for evaluating the effects of natural products on angiogenesis. Recently, zebrafish bioassay-guided identification of metabolite diversity from medicinal plants leading to the discovery of anti-angiogenesis activity compounds has been proven to be effective.17,18 The process of bioactivity-guided isolation usually requires a large amount of plant material and is timeconsuming, costly, or difficult to source. Liquid chromatography−mass spectrometry (LC−MS)-based chemical profiling combined with a chemometric approach to determine and identify potential biomarkers provides a reliable strategy to investigate medicinal and edible plants.19,20 In this study, the aril (the edible pulp of the fruit), pericarp (the rind of the fruit), and seed of G. xanthochymus fruits were investigated for their anti-angiogenesis effects on the zebrafish embryo for primary screening21 and then combined with ultraperformance liquid chromatography quadrupole time of flight mass spectrometry (UPLC−QTOF−MS)-based chemometric statistical analysis.



redissolved at a concentration of 1 mg/mL in 70% methanol aqueous solutions, filtered through a 0.22 μm polytetrafluoroethylene (PTFE) syringe filter. UPLC−QTOF−MS Analysis. UPLC analysis was conducted using a Waters ACQUITY UPLC Systems (Waters Corp., Milford, MA, U.S.A.), controlled by Masslynx, version 4.1. Chromatographic separation was performed with a 2.1 × 50 mm inner diameter, 1.7 μm UPLC BEH C18 reversed-phase column (Waters Corp., Milford, MA, U.S.A.) and kept at a temperature of 40 °C. The mobile phase consisted of 0.1% aqueous formic acid (A) and 0.1% formic acid in MeCN (B). The linear gradient elution was performed as follows: 0− 0.5 min, 20% B; 0.5−2.0 min, 20−75% B; 2.0−2.5 min, 75−80% B; 2.5−3.5 min, 80−82% B; 3.5−4.5 min, 82−95% B; 4.5−6.5 min, 95− 95% B; 6.5−6.8 min, 95−20% B; and 6.8−8.0 min, 20−20% B. The flow rate was set at 0.5 mL/min. The injection volume was 1 μL. Mass spectrometry was recorded using Xevo G2 QTOF (Waters MS Technologies, Manchester, U.K.) equipped with an ESI source and controlled by MassLynx software, version 4.1. The capillary voltages were set at 3000 V positive mode and 2500 V negative mode, and the cone voltage was 20 V. Nitrogen gas was used both for the nebulizer and in desolvation. The desolvation and cone gas flow rates were 800 and 40 L/h, respectively. The desolvation temperature was 300 °C, and the source temperature was 120 °C. All of the analyses were acquired using the LockSpray, and leucine enkephalin (1 μg/mL) in MeCN/water (50:50) with 1% formic acid was used as the lock mass at a flow rate of 10 μL/min, with the lock mass using [M + H]+ = m/z 556.2771 for positive-ion mode and [M − H]− = m/z 554.2615 for negative-ion mode. MS was full scan in centroid mode (both positive and negative), with the m/z range of 100−1200 Da, survey scan time of 0.5 s. Multivariate Analysis. Nine samples from three different collections of G. xanthochymus fruits were analyzed to provide statistical power. The chromatographic mass data in the negative mode were performed using MarkerLynx XS (version 4.1, Waters Corp., Milford, MA, U.S.A.) within the MassLynx software. Before analysis, the date preprocessing including peak peaking, alignment, peak integration, and retention time correction of all raw data were processed by MarkerLynx. The optimized parameters were a retention time range of 0.2−6.0 min, retention time window of 0.05 min, mass range of 100−1000 Da, and a mass tolerance of 50 mDa. The intensity threshold (counts) of collection parameters was set at 1000; the mass window was set as 0.05; the retention time tolerance was set as 0.20; the noise elimination level was set as 6.00; and isotopic peaks were excluded for analysis. All data were normalized to the total spectral intensity of the individual chromatogram, and the resulting data were imported to EZinfo 2.0 software (Umetrics) for principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA); S alphabet-like plot (S-plot) was calculated to visualize the relationship between covariance and correlation among the OPLS-DA results. Zebrafish Culture and Embryo Collection. Transgenic zebrafish line Tg( f li1:EGFP)y1 with endothelial cells expressing enhanced green fluorescent protein (EGFP) was purchased from the Zebrafish International Resource Centre, University of Oregon, Eugene, OR, U.S.A. Zebrafish were maintained at 28 °C on a 14 h (light)/10 h (dark) photoperiod and fed with brine shrimp and tropical fish food twice daily. Handling of zebrafish was in accordance to the animal license issued by the Department of Health of the Hong Kong Special Administrative Region and was approved by the Animal Experimentation Ethics Committee of The Chinese University of Hong Kong (reference 10/013/MIS). Zebrafish embryos were generated by natural pairwise mating of 3−4 pairs of zebrafish of 4−8 months old. Embryos were disinfected with 2 μg/mL methyl-blue solution and cultured in sterilized embryo medium (0.06 g/L Instant Ocean Salt, Aquarium Systems, Atlanta, GA, U.S.A.) prior to the experiments. Treatments. At 1 h postfertilization, the embryos were transferred to a 6-well plate and 20 embryos in each well were incubated with 5 mL of water with or without herbal extracts. At 72 h postfertilization, the SIVs of each embryo were observed under a fluorescence microscope.24

MATERIALS AND METHODS

General Experimental Procedures. UPLC−QTOF−MS was operated by ACQUITY UPLC Systems (Waters, Milford, MA, U.S.A.) coupled with QTOF−MS (Xevo 2 QTOF, Waters MS Technologies, Manchester, U.K.), controlled by Masslynx 4.1 software. 1H and 13C nuclear magnetic resonance (NMR) were recorded on a Bruker Avance 300 NMR spectrometer (Bruker BioSpin GmbH, Rheinstetten/Karlsruhe, Germany). High-performance liquid chromatography (HPLC) separations used a Waters 2695 separation module, equipped with a Waters 2996 photodiode array detector, with a HPLC column (Synergi 4 μm Hydro-Rp 80A, 250 × 4.60 mm, 4 μm, Phenomenex, CA, U.S.A.) for analysis. HPLC−MS-grade MeCN, water, and formic acid were purchased from J.T. Baker (Philipsburg, NJ, U.S.A.), all reagent analytical grade. Reference compounds amentoflavone, volkensiflavone, fukugetin, fukugeside, and xanthochymol were obtained from a previous isolation.9 GB 1a,22 GB 1a glucoside,22 GB 2a,23 and GB 2a glucoside23 were isolated from G. paucinervis seeds, which have been identified on the basis of high-resolution electrospray ionization mass spectrometry (HRESIMS) and NMR and compared to the previous reports from Garcinia species. Plant Material. G. xanthochymus fruits were collected from the Fruit and Spice Park (Homestead, FL, U.S.A.) in June 2008. Fresh frozen fruits were stored at −20 °C until extraction. Collection identifications were 4006, 4007, and 1390. The voucher specimens were deposited at the William and Lynda Streere Herbarium, New York Botanical Garden (Bronx, NY, U.S.A.). Preparation of Extracts. Different collection identifications of G. xanthochymus fruits (fresh) were carefully separated into three parts (aril, pericarp, and seed), freeze-dried, and ground to a powder. Each of the powdered samples (10 g) were extracted with 70% methanol (3 × 30 mL), followed by ultrasonic extraction for 30 min in a water bath at room temperature, and the extract solutions were combined and concentrated under reduced pressure to obtain the crude extracts. Before UPLC−QTOF−MS analysis, the crude extracts were B

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 1. (A) Representative images of zebrafish embryos with normal SIV formation after incubation in the vehicle [0.001% dimethyl sulfoxide (DMSO)]. Upon treatments with pericarp extract at (B) 5 μg/mL or (C) 10 μg/mL or aril extract at (D) 10 μg/mL or (E) 15 μg/mL, the formation of SIVs was impaired. (F) SIV formation was not affected upon treatment with the extract from seed (20 μg/mL). (G) The average lengths of SIVs in the total of 14−21 zebrafish embryos were calculated and plotted. Data are expressed as the mean + standard error of the mean (SEM). (∗∗) p < 0.01 and (∗∗∗) p < 0.001 versus the control group [one-way analysis of variance (ANOVA)].

5−15 μg/mL compared to the control. Furthermore, the aril and pericarp showed the same effect on the growth of SIVs at 10 μg/mL, corresponding to an average SIV length at around 800 μm. These findings indicated that aril and pericarp of G. xanthochymus fruit extracts have similar effectiveness on inhibiting the growth of zebrafish SIVs under low concentrations, while the seed extract (even as high as 20 μg/mL) had no significant effect on the growth of SIVs. Chemometric Analysis of Different Parts of Fruits and Identification of Marker Compounds. Our initial results suggested that the aril and pericarp extracts demonstrated similar anti-angiogenesis activities in the zebrafish model, while the seed extract did not show a promising inhibitory effect. These results indicated the presence of a similar bioactivity chemical constituent in both aril and pericarp, which is different from that in seed. To rapidly identify marker compounds relative to their bioactivities, UPLC−QTOF−MS with positive and negative modes for determining the metabolic profiling of G. xanthochymus fruits was used and each sample was injected 3 times. However, the chromatographic peaks in negative mode (Figure 2A) were clearer with less background noise and fewer

The real-time polymerase chain reaction (PCR) was performed to investigate the molecular mechanisms of the herbal extract effects. At 48 or 72 h postfertilization, RNA was extracted from the embryos using Trizol reagent followed by real-time PCR using gene-specific primers to study the changes in gene expression.25



RESULTS Effects of Extracts from Different Parts of Fruit. We are interested in exploring how differences in metabolic profiles from different fruit parts may correlate with biological activities and help to predict specific bioactive markers identified from UPLC−QTOF−MS-based analysis. The zebrafish embryos with fluorescent blood vessels were treated with different concentrations of crude extracts, and results showed that the extracts from pericarp (5 and 10 μg/mL, panels B and C of Figure 1) or aril (10 and 15 μg/mL, panels D and E of Figure 1) but not from seed (20 μg/mL, Figure 1F) showed a promising inhibitory effect on the growth of SIVs in zebrafish embryos. The average length of inhibition on SIVs was calculated for comparative analysis between the treatment of aril and pericarp extracts, as shown in Figure 1G; both pericarp and aril extracts have significantly impaired the embryos SIVs at C

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 2. Chemometric analysis of G. xanthochymus fruits. (A) UPLC−QTOF−MS total ion chromatograms (negative mode) of the aril, pericarp, and seed extracts. (B) Score plot from PCA, with R2X[1] = 0.87 and R2X[2] = 0.06, drawn with Hotelling’s 95% confidence ellipse. (C) Corresponding loading plot for the PCA score plot. The highlighted triangles represent the major marker ions distinguishing seed from aril and pericarp. (D) S-plot from OPLS-DA. Covariance and correlation plotted on the x and y axes, respectively. The ions labeled with red boxes are significant marker ions for aril and pericarp and have been tentatively identified in Table 1.

which ions contributed most significantly to sample clustering or difference, supervised OPLS-DA was used to compare the two groups (bioactive group, aril and pericarp; inactive group, seed). Internal cross-validation was performed in the building of PCA and OPLS-DA models (Figure S2 of the Supporting Information), with a R2 value of 0.98 and Q2 value of 0.99. The resulting scatter plot (S-plot) was used to recognize important marker ions that distinguish the two groups (Figure 2D). Each point provides an exact mass ion and its retention time. Covariance, w*, and correlation value, p[corr], for each marker are variables plotted on the x and y axes, respectively. These marker ions at the top of the S-plot may be correlated to the biological activity. As a result, 13 ions were selected for the markers for aril and pericarp, considered as priority compounds to determine if they correlate with anti-angiogenic activity. The detailed information on the 13 ions is listed in Table 1. The molecular ions from both positive and negative modes were conducted to identify the selected marker ions. Two reference compounds (xanthochymol and amentoflavone) were co-injected by UPLC and compared by retention time and fragmentation patterns. Additional compounds were tentatively identified by comparing molecular formulas and key fragment ions to compounds previously identified in other Garcinia species. On the basis of the S-plot, these 13 compounds were targeted as the high-priority markers, and we analyzed these

fragmentation ions compared to positive mode (Figure S1 of the Supporting Information). Therefore, all of the collected raw data in negative mode, including retention time, exact mass, and ion intensity, were used as variables in the multivariate analysis. The aril and pericarp samples were clearly clustered into a group in the PCA score plot (Figure 2B), and 1375 ions were detected in the analysis. The goodness-of-fit (R2) and predictability (Q2) values of PCA were 0.94 and 0.92, respectively, indicating that there are similar metabolic constituents shared between aril and pericarp. The seed samples clustered distinctly away from aril and pericarp samples in the PCA score plot, showing that the chemical constituents of the seeds differed significantly from both aril and pericarp. This result has a trend similar to the total ions chromatography (TIC) of aril, pericarp, and seed (Figure 2A). Besides, the PCA found that the metabolites of the three different seed collections had differences. The loading plot for the PCA score plot (Figure 2C) highlights marker ions that distinguish seed from aril and pericarp. OPLS-DA with a S-plot is used to select statistically significant and potentially biologically active compounds. The S-plot provides covariance and correlation between metabolites and the modeled class to allow for easier visualization. The variables that changed significantly are plotted at the top and bottom of the S-plot, and those that do not significantly contribute are plotted in the middle.26 To better understand D

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry Table 1. Dereplication of Marker Compounds for the Aril and Pericarp Extracts from G. xanthochymus Fruits molecular ions [M + H]+/[M − H]− number

RT (min)

mass (m/z)

formula

error (ppm)

molecular formula

1

1.59

C27H32O6

gambogenone9,33

3

2.62

C38H52O8

18-hydroxygarcimultiflorone D or isogarcimultiflorone F34

4

2.80

C38H50O8

GDPPH 3 or its isomers35

5

3.10

C38H50O6

isoxanthochymol, guttiferone H or E, or cycloxanthochymol9

6

3.24

C33H42O6

aristophenone A9

7

3.32

C38H50O6

isoxanthochymol, guttiferone H or E, or cycloxanthochymol9

8

3.39

C38H48O6

garcicowin C or D or oblongifolin F or G33

9

3.54

C38H52O7

garcimultiflorone C or D or its isomers34,36

10

4.38

C38H50O6

isoxanthochymol, guttiferone H or E, or cycloxanthochymol9

11

4.45

C38H50O6

xanthochymols (co-injection)

12

4.50

C38H48O6

garcicowin C or D or oblongifolin F or G33

13

4.74

−2.8 0 0.2 0 0.5 −0.3 −0.2 −0.6 2.5 −1.2 0.6 −0.4 0.3 −2 0.7 −0.5 1.6 −0.8 2.5 −1.5 0.3 −1.0 −0.8 −0.8 1.0 −0.9

amentoflavone (co-injection)

2.03

C30H19O10 C30H17O10 C27H33O6 C27H31O6 C38H53O8 C38H51O8 C38H51O8 C38H49O8 C38H51O6 C38H49O6 C33H43O6 C33H41O6 C38H51O6 C38H49O6 C38H49O6 C38H47O6 C38H53O7 C38H51O7 C38H51O6 C38H49O6 C38H51O6 C38H49O6 C38H49O6 C38H47O6 C38H51O5 C38H49O5

C30H18O10

2

539.0954 537.0822 453.2278 451.2121 637.3743 635.3582 635.3583 633.3423 603.3701 601.3522 535.3063 533.2901 603.3688 601.3521 601.3533 599.3370 621.3801 619.3630 603.3701 601.3520 603.3688 601.3523 601.3524 599.3368 587.3742 585.3575

identification (reference)

C38H50O5

6-epi-guttiferone J10

Figure 3. Representative images of zebrafish embryos treated with (A) vehicle, (B) xanthochymol at 20 μM, (C) amentoflavone at 5 μM, (D) amentoflavone 10 μM, or (E) amentoflavone at 20 μM.

concentrations of amentoflavone (5−20 μM; panels C−E of Figure 3). The average length of SIVs of embryos was 650 μm, treated with amentoflavone at 10 μM. With the higher concentration of treatment at 20 μM, the greater the effect in SIVs and the average length was close to 600 μm (Figure 4A), i.e., reduced the length by ∼33%. Although the inhibition of amentoflavone was not as strong as the positive control eriocalyxin B (at 15 μM), which could reduce the SIV length by around 50% in the same zebrafish assay,27 the average length

aril/pericarp compounds to determine if any of them possessed anti-angiogenic activity. Effects of the Selected Biflavonoids. We selected available reference compounds based on the list of the 13 identified makers. Two representative compounds, xanthochymol and amentoflavone, were further examined for their potential effects in zebrafish embryos. Xanthochymol (20 μM) had no effect on the growth of SIVs (Figure 3B), whereas a significant inhibitory effect in SIVs was observed under a series E

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 4. (A) Average length of SIVs of embryos treated with amentoflavone were calculated and plotted. (B) Quantitative RT-PCR analyses of Angpt2 and Tie-2 gene mRNA. Data were normalized to corresponding GAPDH expressions in control cells. mRNA expression results are expressed as fold of the control [mean + standard deviation (SD) from three independent experiments]. Data are expressed as the mean + SEM. (∗) p < 0.05 and (∗∗∗) p < 0.001 versus the control group (one-way ANOVA).

Figure 5. (A) Chemical structure of biflavonoids and (B and C) representative images of zebrafish embryos treated with (B) vehicle or (C) fukugetin at 10 μM.

was significantly decreased after amentoflavone (10 and 20 μM) treatment. The gene expression in angiogenesis signaling pathways of compound-treated zebrafish embryos were examined by quantitative reverse transcription (RT)-PCR analysis. As shown in Figure 4B, amentoflavone also downregulated the expressions of Angpt2 and Tie2 genes of zebrafish embryos. It had significant inhibition at 20 μM. Our results demonstrate clearly that amentoflavone from the aril and

pericarp extracts may account for at least part of the antiangiogenesis activity in zebrafish model. In an effort to identify additional bioactive biflavonoids and better understand their structure−activity relationship, we selected seven biflavonoids, volkensiflavone, fukugetin, fukugeside, GB 1a, GB 1a glucoside, GB 2a, and GB 2a glucoside (Figure 5), from Garcinia species to test their anti-angiogenesis activity in the zebrafish model. However, only fukugetin (10 μM) exhibited anti-angiogenic activity (panels B and C of F

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry Figure 5), while the other biflavonoids were inactive in the zebrafish assay.

support the significant anti-angiogenic activity of biflavonoids found in edible and medicinal Garcinia species.





DISCUSSION In this study, we have presented an available and effective strategy to discover anti-angiogenesis natural products from edible fruits. The strategy is particularly useful because only a small amount of plant material (ca. 10 g) is needed for both LC−MS and zebrafish studies, providing a strategic method to discover new anti-angiogenic natural products. Zebrafish assays have been developed and applied in high-throughput screening approaches, and they are also successfully used in bioactivityguided isolation of several anti-angiogenesis natural products.18,28 We have combined zebrafish assays with a UPLCQTOF-MS-based chemometric statistical method, which is useful to readily identify potential biomarkers from crude extracts. We found one marker amentoflavone corresponding to anti-angiogenic activity in zebrafish. Although amentoflavone has shown inhibitory effects on angiogenesis in breast cancer cells,29 endothelial cells, and tumor-bearing mice models,30 the anti-angiogenic activity of this compound in zebrafish embryo has not been reported by other research groups. Although the inhibitory effect of amentoflavone was not as potent as eriocalyxin B (positive control), the reduced length of SIV in the embryos was observable in zebrafish embryos (Figure 3). Other priority markers (Table 1) might be potential antiangiogenesis active compounds, which need to be further investigated. Biflavonoids have a somewhat limited distribution in plant species but have significant antioxidant, anti-inflammatory, antibacterial, and anticancer activities.31 Nevertheless, studies regarding the structure−activity relationship of biflavonoids and their anti-angiogenic effect are not well studied. We selected seven biflavonoids to search for additional anti-angiogenic biflavonoids, but only fukugetin significantly inhibited the growth of zebrafish SIVs. This finding is similar to the previous study that reported that fukugetin inhibited tumor growth and tumor angiogenesis in a mouse tumor model in vivo.32 In comparison of the molecular structure of C-3/C-8″-linked Garcinia biflavonoids in the current study (Figure 5A), it shows that their anti-angiogenesis activities are dependent upon the following: (a) The glucoside link on C-7″ of the D ring will lose activity; for example, fukugeside has one glucoside on C-7″ and showed no inhibition of SIV growth. (b) The stereochemistry of the F ring and/or hydroxyl of the E ring are important for bioactivity, such as fukugetin, as compared to either GB 2a or volkensiflavone. The selected biflavonoids in the present study are C-3/C-8″-linked biflavonoids, and further studies have to be conducted to elucidate the structure−activity relationship of other analogous biflavonoids. In conclusion, in this work, anti-angiogenesis activity results combined with UPLC−QTOF−MS-based chemometric statistical analysis were successfully applied to efficiently identify potential bioactive compounds from edible G. xanthochymus fruits. From OPLS-DA analysis, 13 metabolites were prioritized as potential anti-angiogenic constituents from aril and pericarp extracts. One of the markers, amentoflavone, displayed antiangiogenic activity and proved to affect Angpt2 and Tie-2 gene mRNA by quantitative RT-PCR analyses. Further investigation on the structure−activity relationship of seven biflavonoids from Garcinia species showed that fukugetin has similar antiangiogenic effects. Our results provided scientific evidence to

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.7b02867. UPLC−QTOF−MS total ion chromatograms of aril, pericarp, and seed extracts of G. xanthochymus (positive mode) (Figure S1) and supervised OPLS-DA used to compare the two groups (bioactive group, aril and pericarp; inactive group, seed) (Figure S2) (PDF)



AUTHOR INFORMATION

Corresponding Authors

*Telephone: +86-10-68930381. Fax: 10-68930381. E-mail: [email protected]. *Telephone: 852-39436109. Fax: 852-26035248. E-mail: [email protected]. *Telephone: +1-718-960-1105. Fax: +1-718-960-8236. E-mail: [email protected]. ORCID

Edward J. Kennelly: 0000-0002-1682-2696 Funding

This work was supported by the Ministry of Education of China through its 111 Program and Discipline Development Program for Minzu University of China numbered B08044, YLDX01013, and 2015MDTD16C. It was also supported by National Natural Science Foundation of China (3171101235 and 31161140345), Natural Science Foundation of Guangdong Province of China (4001020121), the Science and Technology Program of Guangdong Province (2015B09093077). Edward J. Kennelly was supported by the United States Fulbright Fellowship for his work at the Chinese University of Hong Kong (2014−2015). Notes

The authors declare no competing financial interest.



REFERENCES

(1) Pedraza-Chaverri, J.; Cardenas-Rodriguez, N.; Orozco-Ibarra, M.; Perez-Rojas, J. M. Medicinal properties of mangosteen (Garcinia mangostana). Food Chem. Toxicol. 2008, 46, 3227−3239. (2) Kumar, S.; Sharma, S.; Chattopadhyay, S. K. The potential health benefit of polyisoprenylated benzophenones from Garcinia and related genera: Ethnobotanical and therapeutic importance. Fitoterapia 2013, 89, 86−125. (3) Patil, B. P. Everything you wanted to know about Kokum (Garcinia family)Botany forum. Proceedings of the Information Material on the Occasion of 2nd National Seminar on Kokum, March 4−5, 2005, Kokum Brochure; Western Ghats Kokum Foundation: Panaji, Goa, India, 2005. (4) Hemshekhar, M.; Sunitha, K.; Santhosh, M. S.; Devaraja, S.; Kemparaju, K.; Vishwanath, B. S.; Niranjana, S. R.; Girish, K. S. An overview on genus garcinia: Phytochemical and therapeutical aspects. Phytochem. Rev. 2011, 10, 325−351. (5) Wu, S. B.; Long, C.; Kennelly, E. J. Structural diversity and bioactivities of natural benzophenones. Nat. Prod. Rep. 2014, 31, 1158−1174. (6) Yi, T.; Yi, Z.; Cho, S. G.; Luo, J.; Pandey, M. K.; Aggarwal, B. B.; Liu, M. Gambogic acid inhibits angiogenesis and prostate tumor growth by suppressing vascular endothelial growth factor receptor 2 signaling. Cancer Res. 2008, 68, 1843−1850. (7) Lu, N.; Yang, Y.; You, Q. D.; Ling, Y.; Gao, Y.; Gu, H. Y.; Zhao, L.; Wang, X. T.; Guo, Q. L. Gambogic acid inhibits angiogenesis

G

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry through suppressing vascular endothelial growth factor-induced tyrosine phosphorylation of KDR/Flk-1. Cancer Lett. 2007, 258, 80− 89. (8) Yang, J.; He, S.; Li, S.; Zhang, R.; Peng, A.; Chen, L. In vitro and in vivo antiangiogenic activity of caged polyprenylated xanthones isolated from Garcinia hanburyi Hook. f. Molecules 2013, 18, 15305− 15313. (9) Baggett, S.; Protiva, P.; Mazzola, E. P.; Yang, H.; Ressler, E. T.; Basile, M. J.; Weinstein, I. B.; Kennelly, E. J. Bioactive benzophenones from Garcinia xanthochymus fruits. J. Nat. Prod. 2005, 68, 354−360. (10) Einbond, L.; Mighty, J.; Kashiwazaki, R.; Figueroa, M.; Jalees, F.; Acuna, U.; LeGendre, O.; Foster, D.; Kennelly, E. Garcinia benzophenones inhibit the growth of human colon cancer cells and synergize with sulindac sulfide and turmeric. Anti-Cancer Agents Med. Chem. 2013, 13, 1540−1550. (11) Hicklin, D. J.; Ellis, L. M. Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J. Clin. Oncol. 2005, 23, 1011−1027. (12) Ferrara, N. Vascular endothelial growth factor: Basic science and clinical progress. Endocr. Rev. 2004, 25, 581−611. (13) Langer, R. Biomaterials and biotechnology: From the discovery of the first angiogenesis inhibitors to the development of controlled drug delivery systems and the foundation of tissue engineering. J. Biomed. Mater. Res., Part A 2013, 101A, 2449−2455. (14) Camus, S.; Quevedo, C.; Menendez, S.; Paramonov, I.; Stouten, P. F. W.; Janssen, R. A. J.; Rueb, S.; He, S.; Snaar-Jagalska, B. E.; Laricchia-Robbio, L.; Izpisua Belmonte, J. C. Identification of phosphorylase kinase as a novel therapeutic target through highthroughput screening for anti-angiogenesis compounds in zebrafish. Oncogene 2012, 31, 4333−4342. (15) Lieschke, G. J.; Currie, P. D. Animal models of human disease: Zebrafish swim into view. Nat. Rev. Genet. 2007, 8, 353−367. (16) Serbedzija, G. N.; Flynn, E.; Willett, C. E. Zebrafish angiogenesis: A new model for drug screening. Angiogenesis 1999, 3, 353−359. (17) Zon, L. I.; Peterson, R. T. In vivo drug discovery in the zebrafish. Nat. Rev. Drug Discovery 2005, 4, 35−44. (18) Crawford, A. D.; Liekens, S.; Kamuhabwa, A. R.; Maes, J.; Munck, S.; Busson, R.; Rozenski, J.; Esguerra, C. V.; de Witte, P. A. Zebrafish bioassay-guided natural product discovery: Isolation of angiogenesis inhibitors from East African medicinal plants. PLoS One 2011, 6, e14694. (19) Kulakowski, D. M.; Wu, S. B.; Balick, M. J.; Kennelly, E. J. Merging bioactivity with liquid chromatography-mass spectrometrybased chemometrics to identify minor immunomodulatory compounds from a Micronesian adaptogen, Phaleria nisidai. J. Chromatogr. A 2014, 1364, 74−82. (20) Wu, S. B.; Wu, J.; Yin, Z.; Zhang, J.; Long, C.; Kennelly, E. J.; Zheng, S. Bioactive and marker compounds from two edible darkcolored Myrciaria fruits and the synthesis of jaboticabin. J. Agric. Food Chem. 2013, 61, 4035−4043. (21) Yue, G. G. L.; Kwok, H. F.; Li, P.; Senthilkumar, H. A.; Long, C. l.; Kennelly, E. J.; Bik-San Lau, C. A biflavonoid amentoflavone from Garcinia xanthochymus fruit exhibited antiangiogenic activities in zebrafish embryos. World J. Trad. Chin. Med. 2015, 1, 94. (22) Lin, Y. M.; Anderson, H.; Flavin, M. T.; Pai, Y. H.; MataGreenwood, E.; Pengsuparp, T.; Pezzuto, J. M.; Schinazi, R. F.; Hughes, S. H.; Chen, F. C. In vitro anti-HIV activity of biflavonoids isolated from Rhus succedanea and Garcinia multif lora. J. Nat. Prod. 1997, 60, 884−888. (23) Li, Y. F.; Chen, Y.; Xiao, C. Y.; Chen, D.; Xiao, Y. X.; Mei, Z. A. Rapid screening and identification of alpha-amylase inhibitors from Garcinia xanthochymus using enzyme-immobilized magnetic nanoparticles coupled with HPLC and MS. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2014, 960, 166−173. (24) Yue, G. G.; Lee, J. K.; Kwok, H. F.; Cheng, L.; Wong, E. C.; Jiang, L.; Yu, H.; Leung, H. W.; Wong, Y. L.; Leung, P. C.; Fung, K. P.; Lau, C. B. Novel PI3K/AKT targeting anti-angiogenic activities of 4-

vinylphenol, a new therapeutic potential of a well-known styrene metabolite. Sci. Rep. 2015, 5, 11149. (25) Yue, G. G.-L.; Kwok, H.-F.; Lee, J. K.-M.; Jiang, L.; Chan, K.-M.; Cheng, L.; Wong, E. C.-W.; Leung, P.-C.; Fung, K.-P.; Lau, C. B.-S. Novel anti-angiogenic effects of aromatic-turmerone, essential oil isolated from spice turmeric. J. Funct. Foods 2015, 15, 243−253. (26) Wiklund, S.; Johansson, E.; Sjöström, L.; Mellerowicz, E. J.; Edlund, U.; Shockcor, J. P.; Gottfries, J.; Moritz, T.; Trygg, J. Visualization of GC/TOF-MS-based metabolomics data for identification of biochemically interesting compounds using OPLS class models. Anal. Chem. 2008, 80, 115−122. (27) Zhou, X.; Yue, G. G.; Liu, M.; Zuo, Z.; Lee, J. K.; Li, M.; Tsui, S. K.; Fung, K. P.; Sun, H.; Pu, J.; Lau, C. B. Eriocalyxin B, a natural diterpenoid, inhibited VEGF-induced angiogenesis and diminished angiogenesis-dependent breast tumor growth by suppressing VEGFR2 signaling. Oncotarget 2016, 7, 82820−82835. (28) Crawford, A. D.; Esguerra, C. V.; de Witte, P. A. Fishing for drugs from nature: Zebrafish as a technology platform for natural product discovery. Planta Med. 2008, 74, 624−632. (29) Chen, J. H.; Chen, W. L.; Liu, Y. C. Amentoflavone induces anti-angiogenic and anti-metastatic effects through suppression of NFκB activation in MCF-7 cells. Anticancer Res. 2015, 35, 6685−6693. (30) Guruvayoorappan, C.; Kuttan, G. Inhibition of tumor specific angiogenesis by amentoflavone. Biochemistry (Moscow) 2008, 73, 209− 218. (31) Gontijo, V. S.; Dos Santos, M. H.; Viegas, C., Jr. Biological and chemical aspects of natural biflavonoids from plants: A brief review. Mini-Rev. Med. Chem. 2017, 17, 834−862. (32) Pang, X.; Yi, T.; Yi, Z.; Cho, S. G.; Qu, W.; Pinkaew, D.; Fujise, K.; Liu, M. Morelloflavone, a biflavonoid, inhibits tumor angiogenesis by targeting rho GTPases and extracellular signal-regulated kinase signaling pathways. Cancer Res. 2009, 69, 518−525. (33) Zhou, Y.; Lee, S.; Choi, F. F. K.; Xu, G.; Liu, X.; Song, J. Z.; Li, S. L.; Qiao, C. F.; Xu, H. X. Qualitative and quantitative analysis of polycyclic polyprenylated acylphloroglucinols from Garcinia species using ultra performance liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry. Anal. Chim. Acta 2010, 678, 96−107. (34) Liu, X.; Yu, T.; Gao, X. M.; Zhou, Y.; Qiao, C. F.; Peng, Y.; Chen, S. L.; Luo, K. Q.; Xu, H. X. Apoptotic effects of polyprenylated benzoylphloroglucinol derivatives from the twigs of Garcinia multif lora. J. Nat. Prod. 2010, 73, 1355−1359. (35) Sang, S.; Liao, C. H.; Pan, M. H.; Rosen, R. T.; Lin-Shiau, S. Y.; Lin, J. K.; Ho, C. T. Chemical studies on antioxidant mechanism of garcinol: Analysis of radical reaction products of garcinol with peroxyl radicals and their antitumor activities. Tetrahedron 2002, 58, 10095− 10102. (36) Chen, J. J.; Ting, C. W.; Hwang, T. L.; Chen, I. S. Benzophenone derivatives from the fruits of Garcinia multif lora and their anti-inflammatory activity. J. Nat. Prod. 2009, 72, 253−258.

H

DOI: 10.1021/acs.jafc.7b02867 J. Agric. Food Chem. XXXX, XXX, XXX−XXX