Journal of Ethnopharmacology 155 (2014) 1276–1283
Contents lists available at ScienceDirect
Journal of Ethnopharmacology journal homepage: www.elsevier.com/locate/jep
Research paper
High-resolution bacterial growth inhibition profiling combined with HPLC–HRMS–SPE–NMR for identification of antibacterial constituents in Chinese plants used to treat snakebites Yueqiu Liu, Mia Nielsen, Dan Staerk, Anna K. Jäger n Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen, Denmark
art ic l e i nf o
a b s t r a c t
Article history: Received 8 May 2014 Received in revised form 25 June 2014 Accepted 8 July 2014 Available online 17 July 2014
Ethnopharmacogical relevance: Bacterial infection is one of the main secondary infections caused by snakebite. The 88 plant species investigated in this study have been used as folk remedies for treatment of snakebite, and it is therefore the aim of this study to investigate whether the plants contain compounds with bacterial growth inhibition. Materials and methods: The water and ethanol extracts of 88 plant species were screened at 200 μg/mL against Bacillus subtilis, Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa for their antibacterial activity by micro-broth dilution assay. The most active extracts were fractionated into microplates using analytical-scale RP-HPLC, and subsequently growth inhibition was assessed for each well. The biochromatograms constructed from these assays were used to identify compounds responsible for antibacterial activity. The structures of five compounds were elucidated by HPLC–HRMS–SPE–NMR. Results: Crude extracts of Boehmeria nivea, Colocasia esculenta, Fagopyrum cymosum, Glochidion puberum, Melastoma dodecandrum, Polygonum bistorta, Polygonum cuspidatum and Sanguisorba officinalis showed MIC values below 200 μg/mL against either Bacillus subtilis, Staphylococcus aureus, Escherichia coli or Pseudomonas aeruginosa. The biochromatograms demonstrated that tannins play a main role for the bacterial growth inhibition observed for all above-mentioned plants except for Polygonum cuspidatum. Furthermore, the high-resolution bacterial growth inhibition profiling combined with HPLC–HRMS–SPE– NMR allowed fast identification of three non-tannin active compounds, i.e., piceid, resveratrol and emodin from ethanol extract of Polygonum cuspidatum. Conclusion: The high-resolution bacterial growth inhibition profiling allowed fast pinpointing of constituents responsible for the bioactivity, e.g., either showing tannins being the main bacterial growth inhibitors as observed for the majority of the active plants, or combined with HPLC–HRMS–SPE–NMR for fast structural identification of non-tannin constituents correlated with antibacterial activity. & 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Antibacterial Biochromatogram HPLC–SPE–NMR Polygonum cuspidatum Snakebite Tannins
1. Introduction On average 100,000 persons are bitten each year by venomous snakes in China, with a mortality rate of 5–10% (Nie, 2007). Snakebite envenomation leads to lesion formation at the bite site along with extensive tissue necrosis. The necrotic wound can furthermore acquire a secondary infection from bacteria coming from the snake's mouth at the time of the bite (Garg et al., 2009; Chen et al., 2011). Treatment of snakebite patients is usually performed with the corresponding antiserum as the first-choice medication. Antiserum can alleviate the lethal effect of venom to some extent, but it cannot prevent secondary bacterial infections
n
Corresponding author. Tel.: þ 45 3533 6339; fax: þ 45 3533 6041. E-mail address: [email protected] (A.K. Jäger).
http://dx.doi.org/10.1016/j.jep.2014.07.019 0378-8741/& 2014 Elsevier Ireland Ltd. All rights reserved.
caused by the snakebite. Therefore, besides antiserum, traditional medicines against snakebites are normally used in Chinese hospitals to prevent bacterial infections as well as other secondary injury (Zeng, 1996; Yue and Ni, 2000; Zeng et al., 2001; Yu et al., 2005; Gu and Yang, 2010; Liu et al., 2010). Furthermore, snakebites often happen in remote areas where there is lack of basic medical facilities and no access to antivenom. For these snakebite victims traditional medicines are important to heal the wound and resist infections. Although the usage of plants for wound healing and as an antibacterial remedy after snakebites can be traced back to a long time ago, the constituents responsible for the overall antibacterial activity have yet not been identified in all of these plants. Bioactivity-guided fractionation traditionally represents the first-choice method for isolation of bioactive analytes from plants. However, this method is very time-consuming and involves repetitive preparative-scale isolation processes without any structural
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
information until the bioactive constituents have been purified. This frequently leads to isolation of known natural products that have already been studied (Marston and Hostettmann, 2009; Johansen et al., 2011). The analysis time can be significantly decreased by using analytical-scale high-resolution bioassays for pharmacological profiling of complex extracts, aiming at pinpointing only bioactive constituents for subsequent structure elucidation by hyphenated analytical techniques. This has recently been demonstrated by Staerk and coworkers, using microplate-based high-resolution bioassay combined with hyphenated high-performance liquid chromatography–solid-phase extraction–nuclear magnetic resonance spectroscopy, i.e., HR-bioassay/HPLC–SPE– NMR, for expedited identification of radical scavengers (Agnolet et al., 2012; Wiese et al., 2013, Wubshet et al., 2013a), monoamine oxidase A inhibitors (Grosso et al., 2013), and α-glucosidase inhibitors (Schmidt et al., 2012, 2014; Wubshet et al., 2013b) in herbal remedies and food. Tannins, as a large polyphenol group, are widely distributed in plants, and approximately 70% of the traditional Chinese medicine contains different kinds of tannins (Liu and Wang, 2012). The antibacterial activity of tannins isolated from plants is well-known from several reports in the literature (Yoshida, 2000; Takuo, 2005; Buzzini et al., 2008). The current paper describes a rapid and efficient analysis for exploring antibacterial potential of 88 species of traditional Chinese plants against snakebites and searching for non-tannin bioactive constituents by high-resolution antibacterial profiling followed by HPLC–HRMS–SPE–NMR for structural identification.
1277
for an optical density reading of 0.072 absorbance units at 600 nm, equivalent to the 0.1 McFarland turbidity standard (McFarland, 1907). The bacterial growth inhibition assays were performed in 96-well microplates using the microdilution method. Thus, 100 μL plant extract (400 mg/mL in sterilized Mueller-Hinton nutrient broth with 2% DMSO) and 100 μL bacterial solution were added to each well to get a final crude extract concentration of 200 μg/mL. Wells with 100 μL bacterial solution and 100 μL nutrient broth with 2% DMSO were used as solvent control, wells with 100 μL bacterial solution and 100 μL nutrient broth were used as growth control, and wells with 200 μL nutrient broth were used as sterile control. Each experiment was run in triplicate. Bacterial growth was determined by measuring optical density at 600 nm at 0 h (t1) and 24 h (t2) following incubation at 37 1C. The tested extracts whose t2 value showed no significant difference with t1 (Po0.05) were considered as having promising bacterial growth inhibition. For the most potent crude plant extracts, MIC values were determined by the microdilution method. Crude extracts were dissolved in Mueller-Hinton broth with 2% DMSO to obtain a stock concentration of 200 μg/mL. Serial two-fold dilutions of each sample were made with Mueller-Hinton broth to obtain the final well concentrations of 200, 100, 50 and 25 μg/mL. Wells with crude extract concentrations for which inhibition percentages (being evaluated by the formula (1 test(t2 t1)/solvent control (t2 t1)) 100%) were higher than 99% were considered as full MIC values (Watkins et al., 2012). 2.4. High-resolution antibacterial profiles
A total of 88 species of Chinese medicinal plants have been collected based on the database about plants to treat snakebites worldwide (Molander et al., 2012). All the plant species and the parts used are presented in Table 1, and voucher specimens are deposited at Southwest Treasure Herbs (Chengdu, China). Powdered material (1 g) from each plant species was macerated in 10 mL Milli-Q water or 96% ethanol and subsequently sonicated for 30 min. Thereafter, the mixture was shaken for 1 h and left overnight. The next day, the mixture was shaken for another 10 min and filtered. The filtrate was evaporated to dryness under reduced pressure at 35 1C by a Savant SPD121P speed vacuum concentrator (Thermo Scientific, Waltham, MA).
Analytical-scale HPLC separation of the most active extracts was performed with an Agilent 1200 series instrument consisting of a pump, a column oven, a diode array detector, an autosampler, a fraction collector and a Phenomenex C18 Luna column (150 4.6 mm i.d., 5 μm, 100 Å; Phenomenex, Inc., Torrance, CA, USA). The column was operated at 40 1C and using a flow rate of 0.8 mL/min. The aqueous eluent (A) consisted of water/acetonitrile (95:5, v/v) and the organic eluent (B) consisted of acetonitrile/water (95:5, v/v), both added 0.1% formic acid. Thus, a scout separation of the extracts at a concentration of 12 mg/mL was fractioned into 96-well microplates by time-slicing the HPLC elute from 2 to 25 min in fractions of 0.3 min. The HPLC gradient (gradient I) was as follows: 0 min, 0% B; 27 min, 100% B; 33 min, 100% B. An additional separation of Polygonum cuspidatum was performed with gradient II: 0 min, 10% B; 15 min, 39% B; 22 min, 40% B; 23 min, 42% B; 35 min, 44% B; 95 min, 100% B; 105 min, 100% B. Microplates were dried overnight at 35 1C using a Savant SPD121P speed vacuum concentrator. All wells were added 100 μL MuellerHinton broth with 10% DMSO shaking for 1 h to dissolve collected material and thereafter 100 μL bacterial solution with a density equivalent to the 0.1 McFarland turbidity standard was added. After 24 h incubation at 37 1C, the growth inhibition percentages were calculated by the formula (1test(t2 t1)/solvent control(t2 t1)) 100%. High-resolution biochromatograms were constructed by representing bacterial growth inhibition percentages against chromatographic retention time.
2.3. Antibacterial assay for initial screening
2.5. HPLC–HRMS–SPE–NMR
Bacterial growth inhibition assays of crude extracts were initially performed at a concentration of 200 μg/mL. The bacterial strains used were Gram-positive Bacillus subtilis (ATCC6633), Staphylococcus aureus (ATCC6530) and Gram-negative bacteria Escherichia coli (ATCC1229) and Pseudomonas aeruginosa (ATCC9027). All bacteria were cultured in sterile Mueller-Hinton nutrient broth and incubated at 37 1C overnight. The bacterial concentration was standardized using a spectrophotometer (UV-1800, Shimadzu Corp., Kyoto, Japan)
The bacterial growth inhibition profiles obtained from the scout-separation were used for setting up targeted HPLC–HRMS– SPE–NMR analysis of the bioactive analytes. The HPLC–HRMS– SPE–NMR system consisted of an Agilent 1100 chromatograph (quaternary pump, degasser, thermostatted column compartment, autosampler, and photodiode array detector) (Santa Clara, CA), a Bruker micrOTOF-Q II mass spectrometer equipped with an electrospray ionization source and operated via a 1:99 flow splitter
2. Materials and methods 2.1. Chemicals Ethanol (96%), dimethyl sulphoxide, HPLC grade acetonitrile, and methanol-d4 were from Sigma-Aldrich (St. Louis, MO, USA). Formic acid was from Merck (Darmstadt, Germany). MuellerHinton Bouillon was purchased from Carl Roth GmbH (Karlsruhe, Germany). Water was purified by 0.22 mm membrane filtration and deionization (Millipore, Billerica, MA, USA). 2.2. Plants material and extracts
1278
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Table 1 Plants tested for bacterial growth inhibition. Botanical family
Species
Voucher Specimen number
Part used
Anacardiaceae Apiaceae Apiaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Araceae Araceae Araceae Araceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Balsaminaceae Begoniaceae Campanulaceae Cannabaceae Clusiaceae Commelinaceae Cornaceae Crassulaceae Cucurbitaceae Cucurbitaceae Dioscoreaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Hypericaceae Lamiaceae Lamiaceae Lamiaceae Lamiaceae Lamiaceae Liliaceae Malvaceae Melanthiaceae Melastomataceae Menispermaceae Menispermaceae Moraceae Onagraceae Orchidaceae Oxalidaceae Poaceae Poaceae Polygonaceae Polygalaceae Polygonaceae Polygonaceae Polygonaceae Portulacaceae Primulaceae Primulaceae Primulaceae Ranunculaceae Rosaceae Rosaceae Rosaceae Rubiaceae
Rhus chinensis Mill. Angelica dahurica (Fisch.) Benth. & Hook. Oenanthe javanica (Bl.) DC. Cynanchum paniculatum (Bge.) Kitag. Rauvolfia verticillata (Lour.) Baill. Trachelospermum jasminoides Lem. Tylophora ovata Hook. ex Steud. Wrightia pubescens R. Br. Alocasia indica (Roxb.) Schott Amorphophallus rivieri Durieu ex Riviere Colocasia esculenta (L.) Schott Pinellia ternata (Thunb.) Breit. Arctium lappa L. Bidens bipinnata L. Bidens pilosa L. Emilia sonchifolia (L.) DC. Gynura segetum (Lour.) Merr. Inula helenium L. Ixeris debilis A. Grey Taraxacum mongolicum Hand.-Mazz. Senecio scandens Buch.-Ham. Siegesbeckia orientalis L. Youngia japonica DC. Impatiens balsamina L. Begonia evansiana Andr. Lobelia chinensis Lour. Humulus scandens (Lour.) Merr. Hypericum japonicum Thunb. Commelina communis L. Alangium chinense (Lour.) Rehd. Sedum erythrostictum Miq. Bolbostemma paniculatum (Maxim.) Fran. Trichosanthes cucumeroides Maxim. Dioscorea bulbifera L. Euphorbia lathyris L. Glochidion puberum Hutch. Phyllanthus urinaria L. Dolichos lablab L. Glycine max (L.) Merr. Lespedeza cuneata (Dum.) G.Don Pueraria lobata (Willd.) Ohwi Sophora subprostrata Chun & T.C.Chen Hypericum sampsonii Hance Ajuga decumbens Thumb. Leonurus heterophyllus Sweet Ocimum basilicum L. Prunella vulgaris L. Scutellaria barbata Don Tulipa edulis Baker Hibiscus mutabilis L. Paris polyphylla Sm. Melastoma dodecandrum Roxb. Cyclea hypoglauca (Schau.) Diels. Tinospora capillipes Gagnep. Morus alba L. Ludwigia adscendens (L.) Hara Cremastra variabilis (Bl.) Nakai Oxalis corniculata L. Dactylicapnos scandens (D. Don) Hutch.) Imperata cylindrica P.Beauv. Fagopyrum cymosum Meissn. Polygala japonica Houtt. Polygonum bistorta L. Polygonum cuspidatum Siebold & Zucc. Polygonum perfoliata L. Portulaca oleracea L. Androsace umbellata (Lour.) Merr Ardisia crenata Sims. Lysimachia paridiformis Franch. Semiaquilegia adoxoides (DC.) Makino Duchesnea indica (Andr.) Focke Potentilla kleiniana Wight & Arn. Sanguisorba officinalis L. Gardenia jasminoides Ellis
zyc003 zyc015 zyc022 zyc053 zyc054 zyc066 zyc057 zyc052 zyc101 zyc111 zyc112 zyc092 zyc183 zyc184 zyc185 zyc187 zyc190 zyc188 zyc195 zyc203 zyc182 zyc202 zyc210 zyc044 zyc215 zyc083 zyc005 zyc230 zyc235 zyc257 zyc263 zyc033 zyc036 zyc040 zyc265 zyc266 zyc268 sy001 sy002 zyc271 zyc272 zyc279 zyc259 zyc115 zyc114 zyc118 zyc125 zyc124 zyc133 zyc140 zyc135 zyc175 zyc144 zyc145 zyc152 zyc166 zyc303 zyc039 zyc167 zyc168 zyc088 zyc181 zyc085 zyc087 zyc086 zyc225 zyc280 zyc283 zyc284 zyc126 zyc245 zyc243 zyc244 zyc315
Root Root Whole plant Whole plant Root Leaf and stem Whole plant Leaf Leaf Tuber Stem Rhizome Fruit Whole plant Whole plant Whole plant Root Root Whole plant Whole plant Whole plant Whole plant Whole plant Flower Rhizome Whole plant Whole plant Whole plant Whole plant Root Whole plant Tuber Root Tuber Seed Root and leaf Whole plant Leaf Leaf Whole plant Root Root Whole plant Whole plant Whole plant Whole plant Whole plant Whole plant Bulb Leaf Rhizome Whole plant Root Rhizome Stem Whole plant Root Whole plant Root Spica Rhizome Whole plant Rhizome Root Whole plant Whole plant Whole plant Root Whole plant Root Whole plant Whole plant Root Fruit
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
1279
Table 1 (continued ) Botanical family
Species
Voucher Specimen number
Part used
Rubiaceae Rubiaceae Rubiaceae Rutaceae Rutaceae Rutaceae Rutaceae Sapindaceae Sapindaceae Saururaceae Solanaceae Thymelaeaceae Urticaceae Violaceae
Hedyotis diffusa Willd. Paederia scandens (Lour.) Merr. Rubia cordifolia L. Citrus grandis Hassk. Clausena excavata Burm.f. Ruta graveolens L. Zanthoxylum nitidum DC. Litchi chinensis Sonn. Litchi longan (Lour.) Steud. Houttoynia cordata Thunb. Atropa belladonna L. Wikstroemia indica (L.) C.A.Mey. Boehmeria nivea Gaudich. Viola diffusa Ging.
zyc308 zyc309 zyc322 zyc027 zyc025 zyc023 zyc031 zyc294 zyc295 zyc290 zyc285 zyc107 zyc105 zyc169
Whole plant Whole plant Whole plant Fruit Leaf Whole plant Root Fruit Seed Leaf and stem Root Root Root Whole plant
(Bruker Daltonik GmbH, Bremen, Germany), a Knauer Smartline 100 pump for post-column dilution (Knauer, Berlin, Germany), a Spark Holland Prospekt 2 SPE unit (Spark Holland, Emmen, The Netherlands), a Gilson 215 liquid-handler equipped with a 1-mm needle for automated filling of 1.7-mm NMR tubes, and a Bruker Avance III 600 MHz NMR spectrometer (1H operating frequency of 600.13 MHz) equipped with a Bruker SampleJet sample changer and a cryogenically cooled gradient inverse triple-resonance 1.7-mm TCI probe-head (Bruker Biospin GmbH, Rheinstetten, Germany). Mass spectra were acquired in positive-ion mode, and a solution of sodium formate clusters was automatically injected to enable internal mass calibration. Cumulative SPE trapping of selected peaks was performed after eight consecutive separations of 15 μL of 100 mg/mL solutions using the chromatographic conditions described above. The HPLC eluate was diluted with Milli-Q water at a flow rate of 1 mL/min prior to trapping on 10 2 mm i.d. Resin GP (general purpose, 5–15 μm, spherical shape, polydivinyl-benzene phase) SPE cartridges from Spark Holland (Emmen, The Netherlands), and analytes were trapped using absorption-thresholds (254 nm). 1D and 2D experiments (COSY, HSQC, HMBC and NOESY) were obtained from the standard Bruker pulse sequence library, and were recorded in methanol-d4 at 300 K. 1H and 13C chemical shifts were referenced to the residual solvent signal at δ 3.31 and δ 49.00, respectively. Separations were controlled with Bruker Hystar version 3.2 software, automated filling of NMR tubes was controlled by PrepGilsonST version 1.2 software, and automated NMR acquisition was controlled by Bruker IconNMR version 4.2 software. NMR data processing was performed using Bruker Topspin version 3.2 software.
3. Results and discussion 3.1. Bacterial growth inhibition of crude extracts A total of 180 extracts from the 88 plant species in Table 1 were investigated. The initial assessment of growth inhibition showed that 9 extracts from 8 plants exhibited an antibacterial effect against at least one bacterium at a concentration of 200 μg/mL (i.e., optical density at incubation start showing no significant difference after 24 h incubation (po 0.05)). The most active extracts were ethanolic extracts of Boehmeria nivea, Colocasia esculenta, Fagopyrum cymosum, Glochidion puberum, Melastoma dodecandrum, Polygonum bistorta, Polygonum cuspidatum, Sanguisorba officinalis and the aqueous extract of Sanguisorba officinalis. All nine extracts showed the antibacterial effect against Pseudomonas aeruginosa (MIC given in values Table 2), and most of the extracts were active against Bacillus subtilis and Staphylococcus
aureus. However, only extracts of Sanguisorba officinalis, Polygonum bistorta and Polygonum cuspidatum showed inhibition of Escherichia coli growth. Since MIC values below 100 μg/mL are considered significantly active for crude extracts (Ríos and Recio, 2005), all extracts can be regarded as potential antibacterial remedies in traditional use against snakebite. 3.2. Coupling of antibacterial assay and HPLC–HRMS–SPE–NMR A microplate-based bacterial growth inhibition assay was combined with HPLC–HRMS–SPE–NMR analyses according to the following three steps: i) development of an HPLC separation method and time-based fractionation into 96-well microplates, ii) assessing bacterial growth inhibition assays in microplates, iii) constructing biochromatograms overlaying the HPLC chromatograms, iv) using biochromatograms to pinpoint constituents correlated with bacterial growth inhibition, and v) identification of antibacterial constituents only by HPLC–HRMS–SPE–NMR (Fig. 1). 3.2.1. High-resolution antibacterial assay An analytical-scale HPLC gradient was developed to secure that all analytes in all 9 extracts were included in a short-time chromatographic separation. This showed that the HPLC chromatograms of ethanol and water extracts were almost identical, and the latter was therefore excluded from further experiments. Thus, initial time-based fractionation of the remaining eight extracts was performed within 22.5 min. This leads to high-resolution bacterial growth inhibition profiles (biochromatograms) combined with the HPLC-chromatogram as shown in Fig. 2. The HPLCchromatograms showed that seven out of eight extracts (i.e., chromatogram A–E and G–H in Fig. 2) contained large amounts of tannins, which was not surprising considering that 70% of Chinese medicinal plants are tannin-rich (Liu and Wang, 2012). The biochromatograms clearly showed that the antibacterial activity of these seven extracts corresponds with the tannins, i.e., the hump of closely eluting peaks in the retention time range from 5 to 20 min (Adamson et al., 1999). Smaller differences in the growth inhibition profiles of the four species of bacteria were detected in the biochromatograms, e.g. for biochromatogram B, E, G. Tannins have the ability to bind to proteins, thereby forming insoluble complexes with enzymes (Samuelsson, 2004). Antibacterial activity of tannins has been reported for many plants, especially against Staphylococcus aureus (Akiyama et al., 2001). Tannins are proposed to target the bacterial cell wall, and e.g. complexation with cell wall proteins, disruption of the membrane, and complexation with metal ions are proposed as main
1280
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Table 2 MIC values for the active plant extracts. Species
Boehmeria nivea Colocasia esculenta Fagopyrum cymosum Glochidion puberum Melastoma dodecandrum Polygonum bistorta Polygonum cuspidatum Sanguisorba officinalis Sanguisorba officinalis
Extract type
Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Water
Part used
Root Stem Rhizome Root Whole plant Tuber Rhizome Root Root
Microorganism/MIC (μg/mL) Bacillus subtilis
Escherichia coli
Staphylococcus aureus
Pseudomonas aeruginosa
100 n.d 50 100 n.d 100 100 100 n.d
n.d n.d 100 n.d n.d 100 n.d 50 n.d
100 n.d 100 n.d 50 100 50 50 n.d
100 100 200 200 100 200 100 100 200
n.d: Not detected, MIC 4200 mg/mL.
Fig. 1. Schematic overview of the workflow for the combined use of highresolution bacterial growth inhibition profiling and HPLC–HRMS–SPE–NMR analysis: (i) HPLC separation and time-based microfractionation into 96-well microplates, (ii) bacterial growth inhibition assaying of microplates, (iii) construction of biochromatograms overlaying HPLC chromatograms, (iv) pinpointing constituents correlated with bacterial growth inhibition, and (v) structural identification of antibacterial constituents by HPLC–HRMS–SPE–NMR.
antibacterial mechanisms (Brownlee et al., 1990; Haslam, 1996; Cowan, 1999). The antimicrobial effect of tannins was reviewed by Buzzini et al. (2008) and Takuo (2005). Only the extract of Polygonum cuspidatum (chromatogram F in Fig. 2) showed distinct and well-separated peaks in the bacterial growth inhibition profile that corresponded with distinct peaks in the HPLC chromatogram. Furthermore, the observed bacterial growth inhibition of Bacillus subtilis, Staphylococcus aureus and Pseudomonas aeruginosa from 9 to 14 min was not correlated with a large tannin hump to the same extent as for the other plants. Therefore, a second HPLC separation with a higher sample loading and a less steep gradient was applied (gradient II). The eluate from 2.5 min to 93 min was collected in four 96-well microplates and submitted to bacterial growth inhibition assays. This resulted in a resolution of 3 data points per minute in the bacterial growth inhibition profile (Fig. 3). The high-resolution biochromatogram allowed fast pinpointing of active peaks 1, 2 and 5, and furthermore the increased sample loading disclosed tannins correlated with bacterial growth inhibition between 10 and 25 min.
3.2.2. HPLC–HRMS–SPE–NMR analysis of Polygonum cuspidatum extract The information obtained above was used for setting up UV threshold-based SPE trapping of active peaks 1, 2 and 5. Peaks 3 and 4 did not show bacterial growth inhibition, but their UV spectra were comparable with those observed for peaks 2 and 5. To investigate possible relationships between structure and
activity, peaks 3 and 4 were also included in the HPLC–HRMS– SPE–NMR analysis; thus a total of five peaks (1–5) were selectively trapped on SPE cartridges after 10 repeated separations using gradient II as described above. On the basis of HRMS as well as 1D and 2D NMR spectra obtained from the HPLC–HRMS–SPE–NMR analysis, peaks 1–5 were assigned to piceid (syn. resveratrol-3-O-β-D-glucoside) (1), C20H22O8, HR-ESIMS( þ) m/z 391.1371, 1H NMR (600 MHz, methanol-d4): δ 7.37 (2 H, BB0 -system, J ¼8.6 Hz, H-20 , H-60 ), 7.02 (1H, d, J¼ 16.4 Hz, H-b), 6.85 (1H, d, J ¼16.3 Hz, H-a), 6.79 (1H, t, J ¼1.6 Hz, H-2), 6.77 (2H, AA0 -system, J ¼8.6 Hz, H-30 , H-50 ), 6.45 (1H, t, J¼ 2.1 Hz, H-6), 6.22 (1H, t, J ¼1.7 Hz, H-4), 4.89 (1H, d, J ¼7.4 Hz, H-1″), 3.93 (1H, dd, J ¼12.1 Hz, 2.2, H-6″B), 3.71 (1H, dd, J¼ 12.1, 6.0 Hz, H-6″A), 3.46 (3H, overlapping, H-5″, H-3″, H-2″), 3.38 (1H, t, J ¼9.1 Hz, H-4″) (Jayatilake et al., 1993), resveratrol (2), C14H12O3, HR-ESIMS( þ) m/z 229.0866, 1H NMR (600 MHz, methanol-d4): δ 7.35 (2H, BB0 -system, J ¼8.6 Hz, H-20 , H-60 ), 6.96 (1H, d, J¼ 16.3 Hz, H-b), 6.80 (1H, d, J¼ 16.3 Hz, H-a), 6.77 (2H, AA0 -system, J ¼8.6 Hz, H-30 , H-50 ), 6.45 (2H, d, J ¼2.2 Hz, H-2, H-6), 6.16 (1H, t, J ¼2.1 Hz, H-4) (Jayatilake et al., 1993), torachrysone-1-O-β-D-glucoside (3), C20H24O9, HR-ESIMS( þ) m/z 409.1483, 1H NMR (600 MHz, methanol-d4): δ 7.05 (1H, s, H-4), 7.01 (1H, d, J ¼2.3 Hz, H-7), 6.85 (1H, d, J¼ 2.3 Hz, H-5), 5.11 (1H, d, J¼ 7.8 Hz, H-10 ), 3.94 (1H, dd, J¼ 12.2, 2.2Hz, H-60 B), 3.88 (3H, s, 6-OMe), 3.75 (1H, dd, J ¼12.2, 5.9 Hz, H-60 A), 3.55 (1H, dd, J ¼9.2, 7.9 Hz, H-20 ), 3.54 (1H, m, H-50 ), 3.50 (1H, t, J¼ 9.1, 8.8 Hz, H-30 ), 3.44 (1H, t, J ¼9.3, 8.9 Hz, H-40 ), 2.59 (3H, s, COCH3), 2.29 (3H, s, CH3) (Zhang et al., 2012), emodin-8-Oβ-D-glucoside (4), C21H20O10, HR-ESIMS( þ ) m/z 433.1131, 1H NMR (600 MHz, methanol-d4): δ 7.54 (1H, d, J ¼1.2 Hz, H-4), 7.36 (1H, d, J¼ 2.5 Hz, H-5), 7.13 (1H, d, J ¼2.5 Hz, H-7), 7.09 (1H, d, J ¼1.6 Hz, H-2), 5.01 (1H, d, J¼ 7.7 Hz, H-10 ), 3.96 (1H, dd, J ¼12.1, 2.2 Hz, H-60 B), 3.76 (1H, dd, J ¼12.1, 5.8 Hz, H-60 A), 3.65 (1H, m, H-20 ), 3.53 (2H, overlapping, H-30 , H-50 ), 3.46 (1H, t, J ¼9.3, 9.2 Hz, H-40 ), 2.45 (3H, s, CH3) (Hua et al., 2001), emodin (5), C15H10O5, HR-ESIMS( þ ) m/z 271.0605, 1H NMR (600 MHz, methanol-d4): δ 7.59 (1H, s, H-4), 7.21 (1H, d, J¼ 2.4 Hz, H-5), 7.14 (1H, s, H-2), 6.69 (1H, d, J ¼2.5 Hz, H-7), 2.44 (3H, s, CH3) (Cohen and Towers, 1995), respectively (see Fig. 4). Compounds 1 and 2, i.e. resveratrol-3-O-β-D-glucoside and the resveratrol aglycone, were both displaying bacterial growth inhibition. Similarly, compounds 4 and 5 are emodin-8-O-β-D-glucoside and the emodin aglycone, respectively. However, contrary to resveratrol and its 3-O-β-D-glucoside, the bacterial growth inhibition disappeared upon glucosylation at the 8-position of emodin. Thus, the 8-OH of emodin might be essential for the antibacterial activity. Piceid (1), resveratrol (2) and emodin (5) have previously been reported to possess antibacterial activity. The IC50 value of piceid against Staphylococcus aureus is 312.5 μg/mL (Shan et al., 2008), the IC50 value of resveratrol against Bacillus subtilis and Staphylococcus aureus are 2667 μg/mL and 2133 μg/mL,
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
1281
Fig. 2. High-resolution bacterial growth inhibition profiles of A: Sanguisorba officinalis, B: Glochidion puberum, C: Polygonum bistorta, D: Boehmeria nivea, E: Fagopyrum cymosum, F: Polygonum cuspidatum, G: Melastoma dodecandrum, and H: Colocasia esculenta.
1282
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
extract of Polygonum cuspidatum. Thus, microfractionation of an elongated separation gradient allowed construction of a highresolution bacterial growth inhibition profile covering more than 1.5 h long HPLC separation. The resolution of 3 data points per minute allowed direct correlation between bacterial growth inhibition in the biochromatogram and HPLC peaks 1, 2 and 5. Our study demonstrated that high-resolution bacterial growth inhibition profiling is a useful method for pinpointing the main active compounds in even very complex plant extracts, and that subsequent HPLC–HRMS–SPE–NMR analysis can be targeted bioactive constituents only, thereby avoiding the time-consuming and cumbersome preparative-scale analysis typically involved in bioactivity-guided fractionation.
Fig. 3. High-resolution bacterial growth inhibition profiles of Polygonum cuspidatum using HPLC gradient II for separation.
Acknowledgment HPLC equipment used for obtaining high-resolution bacterial growth inhibition profiles was obtained via a grant from The Carlsberg Foundation (2006-01-0453). The 600 MHz HPLC–HRMS– SPE–NMR system used in this work was acquired through a grant from “Apotekerfonden af 1991”, The Carlsberg Foundation, and the Danish Agency for Science, Technology and Innovation (2136-080023) via the National Research Infrastructure funds. Y.L. acknowledges the Chinese Scholarship Council for a scholarship. References
Fig. 4. Structures of compounds 1–5.
respectively (Taguri et al., 2006), while IC50 values of emodin against Bacillus subtilis, Staphylococcus aureus and Pseudomonas aeruginosa are 7.8, 3.9 and 1.5 μg/mL, respectively (Chukwujekwu et al., 2006; Beattie et al., 2010). Thus, the results in our work are in agreement with the previous reports. However, resveratrol showed no activity against Pseudomonas aeruginosa in this work.
4. Conclusions In conclusion, bacterial growth inhibition of 180 crude extracts from 88 species used against snakebites in traditional Chinese medicine was measured. Subsequently, high-resolution bacterial growth inhibition profiles of the eight extracts with highest bacterial growth inhibition revealed tannins to play a main role in the observed antibacterial activity. From the view of preventing secondary bacterial infections, our research gives a scientific support for those plants in traditional use against snakebite. This study also allowed fast pinpointing of three active non-tannin compounds, i.e., piceid (1), resveratrol (2), and emodin (5) from
Adamson, G.E., Lazarus, S.A., Mitchell, A.E., Prior, R.L., Cao, G., Jacobs, P.H., Kremers, B.G., Hammerstone, J.F., Rucker, R.B., Ritter, K.A., 1999. HPLC method for the quantification of procyanidins in cocoa and chocolate samples and correlation to total antioxidant capacity. Journal of Agricultural and Food Chemistry 47, 4184–4188. Agnolet, S., Wiese, S., Verpoorte, R., Staerk, D., 2012. Comprehensive analysis of commercial willow bark extracts by new technology platform: combined use of metabolomics, high-perfomance liquid chromatography–solid-phase extraction–nuclear magnetic resonance spectroscopy and high-resolution radical scavenging assay. Journal of Chromatography A 1262, 130–137. Akiyama, H., Fuji, K., Yamasaki, O., Oono, T., Iwatski, K., 2001. Antibacterial action of several tannins against Staphylococcus aureus. Journal of Antimicrobial Chemotherapy 4, 87–491. Beattie, K.D., Rouf, R., Gander, L., May, T.W., Ratkowsky, D., Donner, C.D., Gill, M., Grice, I.D., Tiralongo, E., 2010. Antibacterial metabolites from Australian macrofungi from the genus Cortinarius. Phytochemistry 71, 948–955. Brownlee, H.E., McEuen, A.R., Hedger, J., Scott, I.M., 1990. Anti-fungal effects of cocoa tannin on the witches' broom pathogen Crinipellis perniciosa. Physiological and Molecular Plant Pathology 36, 39–48. Buzzini, P., Arapitsas, P., Goretti, M., Branda, E., Turchetti, B., Pinelli, P., Ieri, F., Romani, A., 2008. Antimicrobial and antiviral activity of hydrolysable tannins. Mini-Reviews in Medicinal Chemistry 8, 1179–1187. Chen, C.M., Wu, K.G., Chen, C.J., Wang, C.J., 2011. Bacterial infection in association with snakebite: a 10-year experience in a northern Taiwan medical center. Journal of Microbiology, Immunology and Infection 44, 456–460. Chukwujekwu, J.C., Coombes, P.H., Mulholland, D.A., Van Staden, J., 2006. Emodin, an antibacterial anthraquinone from the roots of Cassia occidentalis. South African Journal of Botany 72, 295–297. Cohen, P.A., Towers, G.H.N., 1995. The anthraquinones of Heterodermia obscurata. Phytochemistry 40, 911–915. Cowan, M.M., 1999. Plant products as antimicrobial agents. Clinical Microbiology Reviews 12, 564–582. Garg, A., Sujatha, S., Garg, J., Acharya, N.S., Parija, S.C., 2009. Wound infections secondary to snakebite. Journal of Infection in Developing Countries 3, 221–223. Grosso, C., Jäger, A.K., Staerk, D., 2013. Coupling of a high-resolution monoamine oxidase-A inhibitor assay and HPLC–SPE–NMR for advanced bioactivityprofiling of plant extracts. Phytochemical Analysis 24, 141–147. Gu, H.L., Yang, W.F., 2010. 100 cases of using detoxicating decoction coupled with western medicines to treat bite by Agkistrodon Halys (Pallas). Journal of Shanxi Traditional Chinese Medicine 31, 458–459. Haslam, E., 1996. Natural polyphenols (vegetable tannins) as drugs: possible modes of action. Journal of Natural Products 59, 205–215. Hua, Y., Zhou, J.Y., Ni, W., Cheng, C.X., 2001. Studies on the constituents of Reymoutria Japonica Houtt. Natural Products Research and Development 13, 16–18. Jayatilake, G.S., Jayasuriya, H., Lee, E.S., Koonchanok, N.M., Geahlen, R.L., Ashendel, C.L., McLaughlin, J.L., Chang, C.J., 1993. Kinase inhibitors from Polygonum cuspidatum. Journal of Natrural Products 56, 1805–1810.
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Johansen, K.T., Wubshet, S.G., Nyberg, N.T., Jaroszewski, J.W., 2011. From retrospective assessment to prospective decisions in natural product isolation: HPLC–SPE–NMR analysis of Carthamus oxyacantha. Journal of Natural Products 74, 2454–2461. Liu, X., Wang, Q.A., 2012. Natural Products Chemistry. Chemical Industry Press, Beijing, China, pp. 154–156. Liu, H.Y., Zhou, W.J., Zhu, L.Y., Ou Yang, W.G., Li, X.X., Xu, B.X., 2010. An analysis of causes for local tissue necrosis by viper bites and treatment of internal and external medicine. Journal of Snake 22, 347–348. Marston, A., Hostettmann, K., 2009. Natural product analysis over the last decades. Planta Medica 75, 672–682. McFarland, M.D., 1907. The nephelometer: an instrument for estimating the number of bacteria in suspensions used for calculating the opsonic index and for vaccines. Journal of the American Medical Association 49, 1176–1178. Molander, M., Saslis-Lagoudakis, C.H., Jäger, A.K., Rønsted, N., 2012. Cross-cultural comparison of medicinal floras used against snakebites. Journal of Ethnopharmacology 139, 863–872. Nie, H., 2007. Progress in treatment of snakebites. Journal of Modern Clinical Medicine 33, 234–236. Ríos, J.L., Recio, M.C., 2005. Medicinal plants and antimicrobial activity. Journal of Ethnopharmacology 100, 80–84. Samuelsson, G., 2004. Drug of Natural Origin–A Textbook of Pharmacognosy. Swedish Pharmaceutical Press, Stockholm, Sweden. Schmidt, J.S., Lauridsen, M.B., Dragsted, L.O., Nielsen, J., Staerk, D., 2012. Development of a bioassay-coupled HPLC–SPE–ttNMR platform for identification of αglucosidase inhibitors in apple peel (Malus domestica Borkh.). Food Chemistry 135, 1692–1699. Schmidt, J.S., Nyberg, N.T., Staerk, D., 2014. Assessment of constituents in Allium by multivariate data analysis, high-resolution α-glucosidase inhibition assay and HPLC–SPE–NMR. Food Chemistry 161, 192–198. Shan, B., Cai, Y.Z., Brooks, J.D., Corke, H., 2008. Antibacterial properties of Polygonum cuspidatum roots and their major bioactive constituents. Food Chemistry 109, 530–537. Taguri, T., Tanaka, T., Kouno, I., 2006. Antibacterial spectrum of plant polyphenols and extracts depending upon hydroxyphenyl structure. Biological & Pharmaceutical Bulletin 29, 2226–2235.
1283
Takuo, O., 2005. Systematics and health effects of chemically distinct tannins in medicinal plants. Phytochemistry 66, 2012–2031. Watkins, F., Pendry, B., Sanchez-Medina1, A., Corcoran, O., 2012. Antimicrobial assays of three native British plants used in Anglo-Saxon medicine for wound healing formulations in 10th century England. Journal of Ethnopharmacology 144, 408–415. Wiese, S., Wubshet, S.G., Nielsen, J., Staerk, D., 2013. Coupling HPLC–SPE–NMR with a microplate-based high-resolution antioxidant assay for efficient analysis of antioxidants in food—validation and proof-of-concept study with caper buds. Food Chemistry 141, 4010–4018. Wubshet, S.G., Nyberg, N.T., Tejesvi, M.V., Pirttilä, A.M., Kajula, M., Mattila, S., Staerk, D., 2013a. Targeting high-performance liquid chromatography–high-resolution mass spectrometry–solid-phase extraction–nuclear magnetic resonance analysis with high-resolution radical scavenging profiles—bioactive secondary metabolites from the endophytic fungus Penicillium namyslowskii. Journal of Chromatography A 1302, 34–39. Wubshet, S.G., Schmidt, J.S., Wiese, S., Staerk, D., 2013b. High-resolution screening combined with HPLC–HRMS–SPE–NMR for identification of potential healthpromoting constituents in sea aster and searocket—new nordic food ingredients. Journal of Agricultural and food Chemistry 61, 8616–8623. Yu, W.Q., Xiong, S.X., Shen, L.M., Wang, W.C., Gong, L.P., Li, J.E., Qiu, G.R., Wei, M., Xu, L.B., Zhang, G.P., Ren, H., Li, Y.F., Wang, D., Ding, X.F., Jin, L., Lin, C.S., Zou, G.M., 2005. Clinical and experimental studies on Jiuweixiaozhongbadusan for treatment of Agkistrodon Halys (Pallas) bite by external use. Journal of Snake 17, 153–156. Yue, S.Y., Ni, L.J., 2000. Pathogenesis and its therapy after bitten by Agkistrodon Halys (Pallas). Journal of Snake 12, 59–63. Yoshida, T., 2000. Structure and function of natural polyphenols: focusing on tannins. Farumashia 36, 1044–1048. Zeng, Z.Y., Qu, J.L., Lin, S., 2001. The mechanism of local tissue necrosis after snakebites and its prevention. Journal of Snake 13, 72–75. Zeng, S.Y., 1996. 18 cases of using traditional Chinese medicines to treat snakebites. Journal of Hunan College of Traditional Chinese Medicine 16, 38–39. Zhang, H., Guo, Z., Wu, N., Xu, W., Han, L., Li, N., Han, Y., 2012. Two novel naphthalene glucosides and an anthraquinone isolated from Rumex dentatus and their antiproliferation activities in four cell lines. Molecules 17, 843–850.
Contents lists available at ScienceDirect
Journal of Ethnopharmacology journal homepage: www.elsevier.com/locate/jep
Research paper
High-resolution bacterial growth inhibition profiling combined with HPLC–HRMS–SPE–NMR for identification of antibacterial constituents in Chinese plants used to treat snakebites Yueqiu Liu, Mia Nielsen, Dan Staerk, Anna K. Jäger n Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen, Denmark
art ic l e i nf o
a b s t r a c t
Article history: Received 8 May 2014 Received in revised form 25 June 2014 Accepted 8 July 2014 Available online 17 July 2014
Ethnopharmacogical relevance: Bacterial infection is one of the main secondary infections caused by snakebite. The 88 plant species investigated in this study have been used as folk remedies for treatment of snakebite, and it is therefore the aim of this study to investigate whether the plants contain compounds with bacterial growth inhibition. Materials and methods: The water and ethanol extracts of 88 plant species were screened at 200 μg/mL against Bacillus subtilis, Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa for their antibacterial activity by micro-broth dilution assay. The most active extracts were fractionated into microplates using analytical-scale RP-HPLC, and subsequently growth inhibition was assessed for each well. The biochromatograms constructed from these assays were used to identify compounds responsible for antibacterial activity. The structures of five compounds were elucidated by HPLC–HRMS–SPE–NMR. Results: Crude extracts of Boehmeria nivea, Colocasia esculenta, Fagopyrum cymosum, Glochidion puberum, Melastoma dodecandrum, Polygonum bistorta, Polygonum cuspidatum and Sanguisorba officinalis showed MIC values below 200 μg/mL against either Bacillus subtilis, Staphylococcus aureus, Escherichia coli or Pseudomonas aeruginosa. The biochromatograms demonstrated that tannins play a main role for the bacterial growth inhibition observed for all above-mentioned plants except for Polygonum cuspidatum. Furthermore, the high-resolution bacterial growth inhibition profiling combined with HPLC–HRMS–SPE– NMR allowed fast identification of three non-tannin active compounds, i.e., piceid, resveratrol and emodin from ethanol extract of Polygonum cuspidatum. Conclusion: The high-resolution bacterial growth inhibition profiling allowed fast pinpointing of constituents responsible for the bioactivity, e.g., either showing tannins being the main bacterial growth inhibitors as observed for the majority of the active plants, or combined with HPLC–HRMS–SPE–NMR for fast structural identification of non-tannin constituents correlated with antibacterial activity. & 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Antibacterial Biochromatogram HPLC–SPE–NMR Polygonum cuspidatum Snakebite Tannins
1. Introduction On average 100,000 persons are bitten each year by venomous snakes in China, with a mortality rate of 5–10% (Nie, 2007). Snakebite envenomation leads to lesion formation at the bite site along with extensive tissue necrosis. The necrotic wound can furthermore acquire a secondary infection from bacteria coming from the snake's mouth at the time of the bite (Garg et al., 2009; Chen et al., 2011). Treatment of snakebite patients is usually performed with the corresponding antiserum as the first-choice medication. Antiserum can alleviate the lethal effect of venom to some extent, but it cannot prevent secondary bacterial infections
n
Corresponding author. Tel.: þ 45 3533 6339; fax: þ 45 3533 6041. E-mail address: [email protected] (A.K. Jäger).
http://dx.doi.org/10.1016/j.jep.2014.07.019 0378-8741/& 2014 Elsevier Ireland Ltd. All rights reserved.
caused by the snakebite. Therefore, besides antiserum, traditional medicines against snakebites are normally used in Chinese hospitals to prevent bacterial infections as well as other secondary injury (Zeng, 1996; Yue and Ni, 2000; Zeng et al., 2001; Yu et al., 2005; Gu and Yang, 2010; Liu et al., 2010). Furthermore, snakebites often happen in remote areas where there is lack of basic medical facilities and no access to antivenom. For these snakebite victims traditional medicines are important to heal the wound and resist infections. Although the usage of plants for wound healing and as an antibacterial remedy after snakebites can be traced back to a long time ago, the constituents responsible for the overall antibacterial activity have yet not been identified in all of these plants. Bioactivity-guided fractionation traditionally represents the first-choice method for isolation of bioactive analytes from plants. However, this method is very time-consuming and involves repetitive preparative-scale isolation processes without any structural
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
information until the bioactive constituents have been purified. This frequently leads to isolation of known natural products that have already been studied (Marston and Hostettmann, 2009; Johansen et al., 2011). The analysis time can be significantly decreased by using analytical-scale high-resolution bioassays for pharmacological profiling of complex extracts, aiming at pinpointing only bioactive constituents for subsequent structure elucidation by hyphenated analytical techniques. This has recently been demonstrated by Staerk and coworkers, using microplate-based high-resolution bioassay combined with hyphenated high-performance liquid chromatography–solid-phase extraction–nuclear magnetic resonance spectroscopy, i.e., HR-bioassay/HPLC–SPE– NMR, for expedited identification of radical scavengers (Agnolet et al., 2012; Wiese et al., 2013, Wubshet et al., 2013a), monoamine oxidase A inhibitors (Grosso et al., 2013), and α-glucosidase inhibitors (Schmidt et al., 2012, 2014; Wubshet et al., 2013b) in herbal remedies and food. Tannins, as a large polyphenol group, are widely distributed in plants, and approximately 70% of the traditional Chinese medicine contains different kinds of tannins (Liu and Wang, 2012). The antibacterial activity of tannins isolated from plants is well-known from several reports in the literature (Yoshida, 2000; Takuo, 2005; Buzzini et al., 2008). The current paper describes a rapid and efficient analysis for exploring antibacterial potential of 88 species of traditional Chinese plants against snakebites and searching for non-tannin bioactive constituents by high-resolution antibacterial profiling followed by HPLC–HRMS–SPE–NMR for structural identification.
1277
for an optical density reading of 0.072 absorbance units at 600 nm, equivalent to the 0.1 McFarland turbidity standard (McFarland, 1907). The bacterial growth inhibition assays were performed in 96-well microplates using the microdilution method. Thus, 100 μL plant extract (400 mg/mL in sterilized Mueller-Hinton nutrient broth with 2% DMSO) and 100 μL bacterial solution were added to each well to get a final crude extract concentration of 200 μg/mL. Wells with 100 μL bacterial solution and 100 μL nutrient broth with 2% DMSO were used as solvent control, wells with 100 μL bacterial solution and 100 μL nutrient broth were used as growth control, and wells with 200 μL nutrient broth were used as sterile control. Each experiment was run in triplicate. Bacterial growth was determined by measuring optical density at 600 nm at 0 h (t1) and 24 h (t2) following incubation at 37 1C. The tested extracts whose t2 value showed no significant difference with t1 (Po0.05) were considered as having promising bacterial growth inhibition. For the most potent crude plant extracts, MIC values were determined by the microdilution method. Crude extracts were dissolved in Mueller-Hinton broth with 2% DMSO to obtain a stock concentration of 200 μg/mL. Serial two-fold dilutions of each sample were made with Mueller-Hinton broth to obtain the final well concentrations of 200, 100, 50 and 25 μg/mL. Wells with crude extract concentrations for which inhibition percentages (being evaluated by the formula (1 test(t2 t1)/solvent control (t2 t1)) 100%) were higher than 99% were considered as full MIC values (Watkins et al., 2012). 2.4. High-resolution antibacterial profiles
A total of 88 species of Chinese medicinal plants have been collected based on the database about plants to treat snakebites worldwide (Molander et al., 2012). All the plant species and the parts used are presented in Table 1, and voucher specimens are deposited at Southwest Treasure Herbs (Chengdu, China). Powdered material (1 g) from each plant species was macerated in 10 mL Milli-Q water or 96% ethanol and subsequently sonicated for 30 min. Thereafter, the mixture was shaken for 1 h and left overnight. The next day, the mixture was shaken for another 10 min and filtered. The filtrate was evaporated to dryness under reduced pressure at 35 1C by a Savant SPD121P speed vacuum concentrator (Thermo Scientific, Waltham, MA).
Analytical-scale HPLC separation of the most active extracts was performed with an Agilent 1200 series instrument consisting of a pump, a column oven, a diode array detector, an autosampler, a fraction collector and a Phenomenex C18 Luna column (150 4.6 mm i.d., 5 μm, 100 Å; Phenomenex, Inc., Torrance, CA, USA). The column was operated at 40 1C and using a flow rate of 0.8 mL/min. The aqueous eluent (A) consisted of water/acetonitrile (95:5, v/v) and the organic eluent (B) consisted of acetonitrile/water (95:5, v/v), both added 0.1% formic acid. Thus, a scout separation of the extracts at a concentration of 12 mg/mL was fractioned into 96-well microplates by time-slicing the HPLC elute from 2 to 25 min in fractions of 0.3 min. The HPLC gradient (gradient I) was as follows: 0 min, 0% B; 27 min, 100% B; 33 min, 100% B. An additional separation of Polygonum cuspidatum was performed with gradient II: 0 min, 10% B; 15 min, 39% B; 22 min, 40% B; 23 min, 42% B; 35 min, 44% B; 95 min, 100% B; 105 min, 100% B. Microplates were dried overnight at 35 1C using a Savant SPD121P speed vacuum concentrator. All wells were added 100 μL MuellerHinton broth with 10% DMSO shaking for 1 h to dissolve collected material and thereafter 100 μL bacterial solution with a density equivalent to the 0.1 McFarland turbidity standard was added. After 24 h incubation at 37 1C, the growth inhibition percentages were calculated by the formula (1test(t2 t1)/solvent control(t2 t1)) 100%. High-resolution biochromatograms were constructed by representing bacterial growth inhibition percentages against chromatographic retention time.
2.3. Antibacterial assay for initial screening
2.5. HPLC–HRMS–SPE–NMR
Bacterial growth inhibition assays of crude extracts were initially performed at a concentration of 200 μg/mL. The bacterial strains used were Gram-positive Bacillus subtilis (ATCC6633), Staphylococcus aureus (ATCC6530) and Gram-negative bacteria Escherichia coli (ATCC1229) and Pseudomonas aeruginosa (ATCC9027). All bacteria were cultured in sterile Mueller-Hinton nutrient broth and incubated at 37 1C overnight. The bacterial concentration was standardized using a spectrophotometer (UV-1800, Shimadzu Corp., Kyoto, Japan)
The bacterial growth inhibition profiles obtained from the scout-separation were used for setting up targeted HPLC–HRMS– SPE–NMR analysis of the bioactive analytes. The HPLC–HRMS– SPE–NMR system consisted of an Agilent 1100 chromatograph (quaternary pump, degasser, thermostatted column compartment, autosampler, and photodiode array detector) (Santa Clara, CA), a Bruker micrOTOF-Q II mass spectrometer equipped with an electrospray ionization source and operated via a 1:99 flow splitter
2. Materials and methods 2.1. Chemicals Ethanol (96%), dimethyl sulphoxide, HPLC grade acetonitrile, and methanol-d4 were from Sigma-Aldrich (St. Louis, MO, USA). Formic acid was from Merck (Darmstadt, Germany). MuellerHinton Bouillon was purchased from Carl Roth GmbH (Karlsruhe, Germany). Water was purified by 0.22 mm membrane filtration and deionization (Millipore, Billerica, MA, USA). 2.2. Plants material and extracts
1278
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Table 1 Plants tested for bacterial growth inhibition. Botanical family
Species
Voucher Specimen number
Part used
Anacardiaceae Apiaceae Apiaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Apocynaceae Araceae Araceae Araceae Araceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Balsaminaceae Begoniaceae Campanulaceae Cannabaceae Clusiaceae Commelinaceae Cornaceae Crassulaceae Cucurbitaceae Cucurbitaceae Dioscoreaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Hypericaceae Lamiaceae Lamiaceae Lamiaceae Lamiaceae Lamiaceae Liliaceae Malvaceae Melanthiaceae Melastomataceae Menispermaceae Menispermaceae Moraceae Onagraceae Orchidaceae Oxalidaceae Poaceae Poaceae Polygonaceae Polygalaceae Polygonaceae Polygonaceae Polygonaceae Portulacaceae Primulaceae Primulaceae Primulaceae Ranunculaceae Rosaceae Rosaceae Rosaceae Rubiaceae
Rhus chinensis Mill. Angelica dahurica (Fisch.) Benth. & Hook. Oenanthe javanica (Bl.) DC. Cynanchum paniculatum (Bge.) Kitag. Rauvolfia verticillata (Lour.) Baill. Trachelospermum jasminoides Lem. Tylophora ovata Hook. ex Steud. Wrightia pubescens R. Br. Alocasia indica (Roxb.) Schott Amorphophallus rivieri Durieu ex Riviere Colocasia esculenta (L.) Schott Pinellia ternata (Thunb.) Breit. Arctium lappa L. Bidens bipinnata L. Bidens pilosa L. Emilia sonchifolia (L.) DC. Gynura segetum (Lour.) Merr. Inula helenium L. Ixeris debilis A. Grey Taraxacum mongolicum Hand.-Mazz. Senecio scandens Buch.-Ham. Siegesbeckia orientalis L. Youngia japonica DC. Impatiens balsamina L. Begonia evansiana Andr. Lobelia chinensis Lour. Humulus scandens (Lour.) Merr. Hypericum japonicum Thunb. Commelina communis L. Alangium chinense (Lour.) Rehd. Sedum erythrostictum Miq. Bolbostemma paniculatum (Maxim.) Fran. Trichosanthes cucumeroides Maxim. Dioscorea bulbifera L. Euphorbia lathyris L. Glochidion puberum Hutch. Phyllanthus urinaria L. Dolichos lablab L. Glycine max (L.) Merr. Lespedeza cuneata (Dum.) G.Don Pueraria lobata (Willd.) Ohwi Sophora subprostrata Chun & T.C.Chen Hypericum sampsonii Hance Ajuga decumbens Thumb. Leonurus heterophyllus Sweet Ocimum basilicum L. Prunella vulgaris L. Scutellaria barbata Don Tulipa edulis Baker Hibiscus mutabilis L. Paris polyphylla Sm. Melastoma dodecandrum Roxb. Cyclea hypoglauca (Schau.) Diels. Tinospora capillipes Gagnep. Morus alba L. Ludwigia adscendens (L.) Hara Cremastra variabilis (Bl.) Nakai Oxalis corniculata L. Dactylicapnos scandens (D. Don) Hutch.) Imperata cylindrica P.Beauv. Fagopyrum cymosum Meissn. Polygala japonica Houtt. Polygonum bistorta L. Polygonum cuspidatum Siebold & Zucc. Polygonum perfoliata L. Portulaca oleracea L. Androsace umbellata (Lour.) Merr Ardisia crenata Sims. Lysimachia paridiformis Franch. Semiaquilegia adoxoides (DC.) Makino Duchesnea indica (Andr.) Focke Potentilla kleiniana Wight & Arn. Sanguisorba officinalis L. Gardenia jasminoides Ellis
zyc003 zyc015 zyc022 zyc053 zyc054 zyc066 zyc057 zyc052 zyc101 zyc111 zyc112 zyc092 zyc183 zyc184 zyc185 zyc187 zyc190 zyc188 zyc195 zyc203 zyc182 zyc202 zyc210 zyc044 zyc215 zyc083 zyc005 zyc230 zyc235 zyc257 zyc263 zyc033 zyc036 zyc040 zyc265 zyc266 zyc268 sy001 sy002 zyc271 zyc272 zyc279 zyc259 zyc115 zyc114 zyc118 zyc125 zyc124 zyc133 zyc140 zyc135 zyc175 zyc144 zyc145 zyc152 zyc166 zyc303 zyc039 zyc167 zyc168 zyc088 zyc181 zyc085 zyc087 zyc086 zyc225 zyc280 zyc283 zyc284 zyc126 zyc245 zyc243 zyc244 zyc315
Root Root Whole plant Whole plant Root Leaf and stem Whole plant Leaf Leaf Tuber Stem Rhizome Fruit Whole plant Whole plant Whole plant Root Root Whole plant Whole plant Whole plant Whole plant Whole plant Flower Rhizome Whole plant Whole plant Whole plant Whole plant Root Whole plant Tuber Root Tuber Seed Root and leaf Whole plant Leaf Leaf Whole plant Root Root Whole plant Whole plant Whole plant Whole plant Whole plant Whole plant Bulb Leaf Rhizome Whole plant Root Rhizome Stem Whole plant Root Whole plant Root Spica Rhizome Whole plant Rhizome Root Whole plant Whole plant Whole plant Root Whole plant Root Whole plant Whole plant Root Fruit
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
1279
Table 1 (continued ) Botanical family
Species
Voucher Specimen number
Part used
Rubiaceae Rubiaceae Rubiaceae Rutaceae Rutaceae Rutaceae Rutaceae Sapindaceae Sapindaceae Saururaceae Solanaceae Thymelaeaceae Urticaceae Violaceae
Hedyotis diffusa Willd. Paederia scandens (Lour.) Merr. Rubia cordifolia L. Citrus grandis Hassk. Clausena excavata Burm.f. Ruta graveolens L. Zanthoxylum nitidum DC. Litchi chinensis Sonn. Litchi longan (Lour.) Steud. Houttoynia cordata Thunb. Atropa belladonna L. Wikstroemia indica (L.) C.A.Mey. Boehmeria nivea Gaudich. Viola diffusa Ging.
zyc308 zyc309 zyc322 zyc027 zyc025 zyc023 zyc031 zyc294 zyc295 zyc290 zyc285 zyc107 zyc105 zyc169
Whole plant Whole plant Whole plant Fruit Leaf Whole plant Root Fruit Seed Leaf and stem Root Root Root Whole plant
(Bruker Daltonik GmbH, Bremen, Germany), a Knauer Smartline 100 pump for post-column dilution (Knauer, Berlin, Germany), a Spark Holland Prospekt 2 SPE unit (Spark Holland, Emmen, The Netherlands), a Gilson 215 liquid-handler equipped with a 1-mm needle for automated filling of 1.7-mm NMR tubes, and a Bruker Avance III 600 MHz NMR spectrometer (1H operating frequency of 600.13 MHz) equipped with a Bruker SampleJet sample changer and a cryogenically cooled gradient inverse triple-resonance 1.7-mm TCI probe-head (Bruker Biospin GmbH, Rheinstetten, Germany). Mass spectra were acquired in positive-ion mode, and a solution of sodium formate clusters was automatically injected to enable internal mass calibration. Cumulative SPE trapping of selected peaks was performed after eight consecutive separations of 15 μL of 100 mg/mL solutions using the chromatographic conditions described above. The HPLC eluate was diluted with Milli-Q water at a flow rate of 1 mL/min prior to trapping on 10 2 mm i.d. Resin GP (general purpose, 5–15 μm, spherical shape, polydivinyl-benzene phase) SPE cartridges from Spark Holland (Emmen, The Netherlands), and analytes were trapped using absorption-thresholds (254 nm). 1D and 2D experiments (COSY, HSQC, HMBC and NOESY) were obtained from the standard Bruker pulse sequence library, and were recorded in methanol-d4 at 300 K. 1H and 13C chemical shifts were referenced to the residual solvent signal at δ 3.31 and δ 49.00, respectively. Separations were controlled with Bruker Hystar version 3.2 software, automated filling of NMR tubes was controlled by PrepGilsonST version 1.2 software, and automated NMR acquisition was controlled by Bruker IconNMR version 4.2 software. NMR data processing was performed using Bruker Topspin version 3.2 software.
3. Results and discussion 3.1. Bacterial growth inhibition of crude extracts A total of 180 extracts from the 88 plant species in Table 1 were investigated. The initial assessment of growth inhibition showed that 9 extracts from 8 plants exhibited an antibacterial effect against at least one bacterium at a concentration of 200 μg/mL (i.e., optical density at incubation start showing no significant difference after 24 h incubation (po 0.05)). The most active extracts were ethanolic extracts of Boehmeria nivea, Colocasia esculenta, Fagopyrum cymosum, Glochidion puberum, Melastoma dodecandrum, Polygonum bistorta, Polygonum cuspidatum, Sanguisorba officinalis and the aqueous extract of Sanguisorba officinalis. All nine extracts showed the antibacterial effect against Pseudomonas aeruginosa (MIC given in values Table 2), and most of the extracts were active against Bacillus subtilis and Staphylococcus
aureus. However, only extracts of Sanguisorba officinalis, Polygonum bistorta and Polygonum cuspidatum showed inhibition of Escherichia coli growth. Since MIC values below 100 μg/mL are considered significantly active for crude extracts (Ríos and Recio, 2005), all extracts can be regarded as potential antibacterial remedies in traditional use against snakebite. 3.2. Coupling of antibacterial assay and HPLC–HRMS–SPE–NMR A microplate-based bacterial growth inhibition assay was combined with HPLC–HRMS–SPE–NMR analyses according to the following three steps: i) development of an HPLC separation method and time-based fractionation into 96-well microplates, ii) assessing bacterial growth inhibition assays in microplates, iii) constructing biochromatograms overlaying the HPLC chromatograms, iv) using biochromatograms to pinpoint constituents correlated with bacterial growth inhibition, and v) identification of antibacterial constituents only by HPLC–HRMS–SPE–NMR (Fig. 1). 3.2.1. High-resolution antibacterial assay An analytical-scale HPLC gradient was developed to secure that all analytes in all 9 extracts were included in a short-time chromatographic separation. This showed that the HPLC chromatograms of ethanol and water extracts were almost identical, and the latter was therefore excluded from further experiments. Thus, initial time-based fractionation of the remaining eight extracts was performed within 22.5 min. This leads to high-resolution bacterial growth inhibition profiles (biochromatograms) combined with the HPLC-chromatogram as shown in Fig. 2. The HPLCchromatograms showed that seven out of eight extracts (i.e., chromatogram A–E and G–H in Fig. 2) contained large amounts of tannins, which was not surprising considering that 70% of Chinese medicinal plants are tannin-rich (Liu and Wang, 2012). The biochromatograms clearly showed that the antibacterial activity of these seven extracts corresponds with the tannins, i.e., the hump of closely eluting peaks in the retention time range from 5 to 20 min (Adamson et al., 1999). Smaller differences in the growth inhibition profiles of the four species of bacteria were detected in the biochromatograms, e.g. for biochromatogram B, E, G. Tannins have the ability to bind to proteins, thereby forming insoluble complexes with enzymes (Samuelsson, 2004). Antibacterial activity of tannins has been reported for many plants, especially against Staphylococcus aureus (Akiyama et al., 2001). Tannins are proposed to target the bacterial cell wall, and e.g. complexation with cell wall proteins, disruption of the membrane, and complexation with metal ions are proposed as main
1280
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Table 2 MIC values for the active plant extracts. Species
Boehmeria nivea Colocasia esculenta Fagopyrum cymosum Glochidion puberum Melastoma dodecandrum Polygonum bistorta Polygonum cuspidatum Sanguisorba officinalis Sanguisorba officinalis
Extract type
Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Water
Part used
Root Stem Rhizome Root Whole plant Tuber Rhizome Root Root
Microorganism/MIC (μg/mL) Bacillus subtilis
Escherichia coli
Staphylococcus aureus
Pseudomonas aeruginosa
100 n.d 50 100 n.d 100 100 100 n.d
n.d n.d 100 n.d n.d 100 n.d 50 n.d
100 n.d 100 n.d 50 100 50 50 n.d
100 100 200 200 100 200 100 100 200
n.d: Not detected, MIC 4200 mg/mL.
Fig. 1. Schematic overview of the workflow for the combined use of highresolution bacterial growth inhibition profiling and HPLC–HRMS–SPE–NMR analysis: (i) HPLC separation and time-based microfractionation into 96-well microplates, (ii) bacterial growth inhibition assaying of microplates, (iii) construction of biochromatograms overlaying HPLC chromatograms, (iv) pinpointing constituents correlated with bacterial growth inhibition, and (v) structural identification of antibacterial constituents by HPLC–HRMS–SPE–NMR.
antibacterial mechanisms (Brownlee et al., 1990; Haslam, 1996; Cowan, 1999). The antimicrobial effect of tannins was reviewed by Buzzini et al. (2008) and Takuo (2005). Only the extract of Polygonum cuspidatum (chromatogram F in Fig. 2) showed distinct and well-separated peaks in the bacterial growth inhibition profile that corresponded with distinct peaks in the HPLC chromatogram. Furthermore, the observed bacterial growth inhibition of Bacillus subtilis, Staphylococcus aureus and Pseudomonas aeruginosa from 9 to 14 min was not correlated with a large tannin hump to the same extent as for the other plants. Therefore, a second HPLC separation with a higher sample loading and a less steep gradient was applied (gradient II). The eluate from 2.5 min to 93 min was collected in four 96-well microplates and submitted to bacterial growth inhibition assays. This resulted in a resolution of 3 data points per minute in the bacterial growth inhibition profile (Fig. 3). The high-resolution biochromatogram allowed fast pinpointing of active peaks 1, 2 and 5, and furthermore the increased sample loading disclosed tannins correlated with bacterial growth inhibition between 10 and 25 min.
3.2.2. HPLC–HRMS–SPE–NMR analysis of Polygonum cuspidatum extract The information obtained above was used for setting up UV threshold-based SPE trapping of active peaks 1, 2 and 5. Peaks 3 and 4 did not show bacterial growth inhibition, but their UV spectra were comparable with those observed for peaks 2 and 5. To investigate possible relationships between structure and
activity, peaks 3 and 4 were also included in the HPLC–HRMS– SPE–NMR analysis; thus a total of five peaks (1–5) were selectively trapped on SPE cartridges after 10 repeated separations using gradient II as described above. On the basis of HRMS as well as 1D and 2D NMR spectra obtained from the HPLC–HRMS–SPE–NMR analysis, peaks 1–5 were assigned to piceid (syn. resveratrol-3-O-β-D-glucoside) (1), C20H22O8, HR-ESIMS( þ) m/z 391.1371, 1H NMR (600 MHz, methanol-d4): δ 7.37 (2 H, BB0 -system, J ¼8.6 Hz, H-20 , H-60 ), 7.02 (1H, d, J¼ 16.4 Hz, H-b), 6.85 (1H, d, J ¼16.3 Hz, H-a), 6.79 (1H, t, J ¼1.6 Hz, H-2), 6.77 (2H, AA0 -system, J ¼8.6 Hz, H-30 , H-50 ), 6.45 (1H, t, J¼ 2.1 Hz, H-6), 6.22 (1H, t, J ¼1.7 Hz, H-4), 4.89 (1H, d, J ¼7.4 Hz, H-1″), 3.93 (1H, dd, J ¼12.1 Hz, 2.2, H-6″B), 3.71 (1H, dd, J¼ 12.1, 6.0 Hz, H-6″A), 3.46 (3H, overlapping, H-5″, H-3″, H-2″), 3.38 (1H, t, J ¼9.1 Hz, H-4″) (Jayatilake et al., 1993), resveratrol (2), C14H12O3, HR-ESIMS( þ) m/z 229.0866, 1H NMR (600 MHz, methanol-d4): δ 7.35 (2H, BB0 -system, J ¼8.6 Hz, H-20 , H-60 ), 6.96 (1H, d, J¼ 16.3 Hz, H-b), 6.80 (1H, d, J¼ 16.3 Hz, H-a), 6.77 (2H, AA0 -system, J ¼8.6 Hz, H-30 , H-50 ), 6.45 (2H, d, J ¼2.2 Hz, H-2, H-6), 6.16 (1H, t, J ¼2.1 Hz, H-4) (Jayatilake et al., 1993), torachrysone-1-O-β-D-glucoside (3), C20H24O9, HR-ESIMS( þ) m/z 409.1483, 1H NMR (600 MHz, methanol-d4): δ 7.05 (1H, s, H-4), 7.01 (1H, d, J ¼2.3 Hz, H-7), 6.85 (1H, d, J¼ 2.3 Hz, H-5), 5.11 (1H, d, J¼ 7.8 Hz, H-10 ), 3.94 (1H, dd, J¼ 12.2, 2.2Hz, H-60 B), 3.88 (3H, s, 6-OMe), 3.75 (1H, dd, J ¼12.2, 5.9 Hz, H-60 A), 3.55 (1H, dd, J ¼9.2, 7.9 Hz, H-20 ), 3.54 (1H, m, H-50 ), 3.50 (1H, t, J¼ 9.1, 8.8 Hz, H-30 ), 3.44 (1H, t, J ¼9.3, 8.9 Hz, H-40 ), 2.59 (3H, s, COCH3), 2.29 (3H, s, CH3) (Zhang et al., 2012), emodin-8-Oβ-D-glucoside (4), C21H20O10, HR-ESIMS( þ ) m/z 433.1131, 1H NMR (600 MHz, methanol-d4): δ 7.54 (1H, d, J ¼1.2 Hz, H-4), 7.36 (1H, d, J¼ 2.5 Hz, H-5), 7.13 (1H, d, J ¼2.5 Hz, H-7), 7.09 (1H, d, J ¼1.6 Hz, H-2), 5.01 (1H, d, J¼ 7.7 Hz, H-10 ), 3.96 (1H, dd, J ¼12.1, 2.2 Hz, H-60 B), 3.76 (1H, dd, J ¼12.1, 5.8 Hz, H-60 A), 3.65 (1H, m, H-20 ), 3.53 (2H, overlapping, H-30 , H-50 ), 3.46 (1H, t, J ¼9.3, 9.2 Hz, H-40 ), 2.45 (3H, s, CH3) (Hua et al., 2001), emodin (5), C15H10O5, HR-ESIMS( þ ) m/z 271.0605, 1H NMR (600 MHz, methanol-d4): δ 7.59 (1H, s, H-4), 7.21 (1H, d, J¼ 2.4 Hz, H-5), 7.14 (1H, s, H-2), 6.69 (1H, d, J ¼2.5 Hz, H-7), 2.44 (3H, s, CH3) (Cohen and Towers, 1995), respectively (see Fig. 4). Compounds 1 and 2, i.e. resveratrol-3-O-β-D-glucoside and the resveratrol aglycone, were both displaying bacterial growth inhibition. Similarly, compounds 4 and 5 are emodin-8-O-β-D-glucoside and the emodin aglycone, respectively. However, contrary to resveratrol and its 3-O-β-D-glucoside, the bacterial growth inhibition disappeared upon glucosylation at the 8-position of emodin. Thus, the 8-OH of emodin might be essential for the antibacterial activity. Piceid (1), resveratrol (2) and emodin (5) have previously been reported to possess antibacterial activity. The IC50 value of piceid against Staphylococcus aureus is 312.5 μg/mL (Shan et al., 2008), the IC50 value of resveratrol against Bacillus subtilis and Staphylococcus aureus are 2667 μg/mL and 2133 μg/mL,
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
1281
Fig. 2. High-resolution bacterial growth inhibition profiles of A: Sanguisorba officinalis, B: Glochidion puberum, C: Polygonum bistorta, D: Boehmeria nivea, E: Fagopyrum cymosum, F: Polygonum cuspidatum, G: Melastoma dodecandrum, and H: Colocasia esculenta.
1282
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
extract of Polygonum cuspidatum. Thus, microfractionation of an elongated separation gradient allowed construction of a highresolution bacterial growth inhibition profile covering more than 1.5 h long HPLC separation. The resolution of 3 data points per minute allowed direct correlation between bacterial growth inhibition in the biochromatogram and HPLC peaks 1, 2 and 5. Our study demonstrated that high-resolution bacterial growth inhibition profiling is a useful method for pinpointing the main active compounds in even very complex plant extracts, and that subsequent HPLC–HRMS–SPE–NMR analysis can be targeted bioactive constituents only, thereby avoiding the time-consuming and cumbersome preparative-scale analysis typically involved in bioactivity-guided fractionation.
Fig. 3. High-resolution bacterial growth inhibition profiles of Polygonum cuspidatum using HPLC gradient II for separation.
Acknowledgment HPLC equipment used for obtaining high-resolution bacterial growth inhibition profiles was obtained via a grant from The Carlsberg Foundation (2006-01-0453). The 600 MHz HPLC–HRMS– SPE–NMR system used in this work was acquired through a grant from “Apotekerfonden af 1991”, The Carlsberg Foundation, and the Danish Agency for Science, Technology and Innovation (2136-080023) via the National Research Infrastructure funds. Y.L. acknowledges the Chinese Scholarship Council for a scholarship. References
Fig. 4. Structures of compounds 1–5.
respectively (Taguri et al., 2006), while IC50 values of emodin against Bacillus subtilis, Staphylococcus aureus and Pseudomonas aeruginosa are 7.8, 3.9 and 1.5 μg/mL, respectively (Chukwujekwu et al., 2006; Beattie et al., 2010). Thus, the results in our work are in agreement with the previous reports. However, resveratrol showed no activity against Pseudomonas aeruginosa in this work.
4. Conclusions In conclusion, bacterial growth inhibition of 180 crude extracts from 88 species used against snakebites in traditional Chinese medicine was measured. Subsequently, high-resolution bacterial growth inhibition profiles of the eight extracts with highest bacterial growth inhibition revealed tannins to play a main role in the observed antibacterial activity. From the view of preventing secondary bacterial infections, our research gives a scientific support for those plants in traditional use against snakebite. This study also allowed fast pinpointing of three active non-tannin compounds, i.e., piceid (1), resveratrol (2), and emodin (5) from
Adamson, G.E., Lazarus, S.A., Mitchell, A.E., Prior, R.L., Cao, G., Jacobs, P.H., Kremers, B.G., Hammerstone, J.F., Rucker, R.B., Ritter, K.A., 1999. HPLC method for the quantification of procyanidins in cocoa and chocolate samples and correlation to total antioxidant capacity. Journal of Agricultural and Food Chemistry 47, 4184–4188. Agnolet, S., Wiese, S., Verpoorte, R., Staerk, D., 2012. Comprehensive analysis of commercial willow bark extracts by new technology platform: combined use of metabolomics, high-perfomance liquid chromatography–solid-phase extraction–nuclear magnetic resonance spectroscopy and high-resolution radical scavenging assay. Journal of Chromatography A 1262, 130–137. Akiyama, H., Fuji, K., Yamasaki, O., Oono, T., Iwatski, K., 2001. Antibacterial action of several tannins against Staphylococcus aureus. Journal of Antimicrobial Chemotherapy 4, 87–491. Beattie, K.D., Rouf, R., Gander, L., May, T.W., Ratkowsky, D., Donner, C.D., Gill, M., Grice, I.D., Tiralongo, E., 2010. Antibacterial metabolites from Australian macrofungi from the genus Cortinarius. Phytochemistry 71, 948–955. Brownlee, H.E., McEuen, A.R., Hedger, J., Scott, I.M., 1990. Anti-fungal effects of cocoa tannin on the witches' broom pathogen Crinipellis perniciosa. Physiological and Molecular Plant Pathology 36, 39–48. Buzzini, P., Arapitsas, P., Goretti, M., Branda, E., Turchetti, B., Pinelli, P., Ieri, F., Romani, A., 2008. Antimicrobial and antiviral activity of hydrolysable tannins. Mini-Reviews in Medicinal Chemistry 8, 1179–1187. Chen, C.M., Wu, K.G., Chen, C.J., Wang, C.J., 2011. Bacterial infection in association with snakebite: a 10-year experience in a northern Taiwan medical center. Journal of Microbiology, Immunology and Infection 44, 456–460. Chukwujekwu, J.C., Coombes, P.H., Mulholland, D.A., Van Staden, J., 2006. Emodin, an antibacterial anthraquinone from the roots of Cassia occidentalis. South African Journal of Botany 72, 295–297. Cohen, P.A., Towers, G.H.N., 1995. The anthraquinones of Heterodermia obscurata. Phytochemistry 40, 911–915. Cowan, M.M., 1999. Plant products as antimicrobial agents. Clinical Microbiology Reviews 12, 564–582. Garg, A., Sujatha, S., Garg, J., Acharya, N.S., Parija, S.C., 2009. Wound infections secondary to snakebite. Journal of Infection in Developing Countries 3, 221–223. Grosso, C., Jäger, A.K., Staerk, D., 2013. Coupling of a high-resolution monoamine oxidase-A inhibitor assay and HPLC–SPE–NMR for advanced bioactivityprofiling of plant extracts. Phytochemical Analysis 24, 141–147. Gu, H.L., Yang, W.F., 2010. 100 cases of using detoxicating decoction coupled with western medicines to treat bite by Agkistrodon Halys (Pallas). Journal of Shanxi Traditional Chinese Medicine 31, 458–459. Haslam, E., 1996. Natural polyphenols (vegetable tannins) as drugs: possible modes of action. Journal of Natural Products 59, 205–215. Hua, Y., Zhou, J.Y., Ni, W., Cheng, C.X., 2001. Studies on the constituents of Reymoutria Japonica Houtt. Natural Products Research and Development 13, 16–18. Jayatilake, G.S., Jayasuriya, H., Lee, E.S., Koonchanok, N.M., Geahlen, R.L., Ashendel, C.L., McLaughlin, J.L., Chang, C.J., 1993. Kinase inhibitors from Polygonum cuspidatum. Journal of Natrural Products 56, 1805–1810.
Y. Liu et al. / Journal of Ethnopharmacology 155 (2014) 1276–1283
Johansen, K.T., Wubshet, S.G., Nyberg, N.T., Jaroszewski, J.W., 2011. From retrospective assessment to prospective decisions in natural product isolation: HPLC–SPE–NMR analysis of Carthamus oxyacantha. Journal of Natural Products 74, 2454–2461. Liu, X., Wang, Q.A., 2012. Natural Products Chemistry. Chemical Industry Press, Beijing, China, pp. 154–156. Liu, H.Y., Zhou, W.J., Zhu, L.Y., Ou Yang, W.G., Li, X.X., Xu, B.X., 2010. An analysis of causes for local tissue necrosis by viper bites and treatment of internal and external medicine. Journal of Snake 22, 347–348. Marston, A., Hostettmann, K., 2009. Natural product analysis over the last decades. Planta Medica 75, 672–682. McFarland, M.D., 1907. The nephelometer: an instrument for estimating the number of bacteria in suspensions used for calculating the opsonic index and for vaccines. Journal of the American Medical Association 49, 1176–1178. Molander, M., Saslis-Lagoudakis, C.H., Jäger, A.K., Rønsted, N., 2012. Cross-cultural comparison of medicinal floras used against snakebites. Journal of Ethnopharmacology 139, 863–872. Nie, H., 2007. Progress in treatment of snakebites. Journal of Modern Clinical Medicine 33, 234–236. Ríos, J.L., Recio, M.C., 2005. Medicinal plants and antimicrobial activity. Journal of Ethnopharmacology 100, 80–84. Samuelsson, G., 2004. Drug of Natural Origin–A Textbook of Pharmacognosy. Swedish Pharmaceutical Press, Stockholm, Sweden. Schmidt, J.S., Lauridsen, M.B., Dragsted, L.O., Nielsen, J., Staerk, D., 2012. Development of a bioassay-coupled HPLC–SPE–ttNMR platform for identification of αglucosidase inhibitors in apple peel (Malus domestica Borkh.). Food Chemistry 135, 1692–1699. Schmidt, J.S., Nyberg, N.T., Staerk, D., 2014. Assessment of constituents in Allium by multivariate data analysis, high-resolution α-glucosidase inhibition assay and HPLC–SPE–NMR. Food Chemistry 161, 192–198. Shan, B., Cai, Y.Z., Brooks, J.D., Corke, H., 2008. Antibacterial properties of Polygonum cuspidatum roots and their major bioactive constituents. Food Chemistry 109, 530–537. Taguri, T., Tanaka, T., Kouno, I., 2006. Antibacterial spectrum of plant polyphenols and extracts depending upon hydroxyphenyl structure. Biological & Pharmaceutical Bulletin 29, 2226–2235.
1283
Takuo, O., 2005. Systematics and health effects of chemically distinct tannins in medicinal plants. Phytochemistry 66, 2012–2031. Watkins, F., Pendry, B., Sanchez-Medina1, A., Corcoran, O., 2012. Antimicrobial assays of three native British plants used in Anglo-Saxon medicine for wound healing formulations in 10th century England. Journal of Ethnopharmacology 144, 408–415. Wiese, S., Wubshet, S.G., Nielsen, J., Staerk, D., 2013. Coupling HPLC–SPE–NMR with a microplate-based high-resolution antioxidant assay for efficient analysis of antioxidants in food—validation and proof-of-concept study with caper buds. Food Chemistry 141, 4010–4018. Wubshet, S.G., Nyberg, N.T., Tejesvi, M.V., Pirttilä, A.M., Kajula, M., Mattila, S., Staerk, D., 2013a. Targeting high-performance liquid chromatography–high-resolution mass spectrometry–solid-phase extraction–nuclear magnetic resonance analysis with high-resolution radical scavenging profiles—bioactive secondary metabolites from the endophytic fungus Penicillium namyslowskii. Journal of Chromatography A 1302, 34–39. Wubshet, S.G., Schmidt, J.S., Wiese, S., Staerk, D., 2013b. High-resolution screening combined with HPLC–HRMS–SPE–NMR for identification of potential healthpromoting constituents in sea aster and searocket—new nordic food ingredients. Journal of Agricultural and food Chemistry 61, 8616–8623. Yu, W.Q., Xiong, S.X., Shen, L.M., Wang, W.C., Gong, L.P., Li, J.E., Qiu, G.R., Wei, M., Xu, L.B., Zhang, G.P., Ren, H., Li, Y.F., Wang, D., Ding, X.F., Jin, L., Lin, C.S., Zou, G.M., 2005. Clinical and experimental studies on Jiuweixiaozhongbadusan for treatment of Agkistrodon Halys (Pallas) bite by external use. Journal of Snake 17, 153–156. Yue, S.Y., Ni, L.J., 2000. Pathogenesis and its therapy after bitten by Agkistrodon Halys (Pallas). Journal of Snake 12, 59–63. Yoshida, T., 2000. Structure and function of natural polyphenols: focusing on tannins. Farumashia 36, 1044–1048. Zeng, Z.Y., Qu, J.L., Lin, S., 2001. The mechanism of local tissue necrosis after snakebites and its prevention. Journal of Snake 13, 72–75. Zeng, S.Y., 1996. 18 cases of using traditional Chinese medicines to treat snakebites. Journal of Hunan College of Traditional Chinese Medicine 16, 38–39. Zhang, H., Guo, Z., Wu, N., Xu, W., Han, L., Li, N., Han, Y., 2012. Two novel naphthalene glucosides and an anthraquinone isolated from Rumex dentatus and their antiproliferation activities in four cell lines. Molecules 17, 843–850.
