Abstract
Tetrastigma hemsleyanum Diels et Gilg (T. hemsleyanums) is a kind of traditional folk medicinal plant which has been used widely in China for its antivirus, antitumor, and other clinical effects. In this study, ultra-high performance liquid chromatography coupled with hybrid quadrupole-orbitrap mass spectrometry (UPLC-Q-Exactive/MS) was utilized to analyze the chemical constituents of T. hemsleyanums. Fifty-one constituents were clarified, including flavonoids, anthraquinones, esters, fatty acids, phenols, and catechins. In the subsequent quantitative analysis, the contents of ten compounds of rutin, kaempferol, astragalin, quercitrin, quercetin, vitexin-rhamnoside, isorhamnetin, vitexin, emodin-8-O-β-D-glucoside, and isoquercetin in 18 batches of T. hemsleyanums collected from different places of cultivation were determined. Meanwhile, anti-influenza virus bioactivity in vitro of the above samples was detected with Gaussia Luciferase viral titer assay. It was found that the antiviral bioactivity varied from batches to batches in accordance with content difference of the chemical constituents in T. hemsleyanums. Correlation analysis was performed with SPSS software for the association between LC-MS chemometrics and bioactivity of influenza virus inhibition, and 8 constituents of flavonoids showed positive correlation coefficient, which may provide a valuable clue for searching potential antiviral components in T. hemsleyanums.
1. Introduction
Tetrastigma hemsleyanum Diels et Gilg (abbreviated. as T. hemsleyanums) is a kind of medicinal and edible plant wildly growing in south China, especially in Zhejiang, Guangxi, and Yunnan provinces [1]. It was firstly recorded in Zhong Hua Ben Cao [2] for its effects of clearing away heat, detoxifying, removing phlegm, promoting blood circulation, and relieving pain. Nowadays, it is used in Traditional Chinese Medicine (TCM) clinics for the treatment of children with high fever (convulsions), viral meningitis, pneumonia, and hepatitis. Many literatures confirmed that T. hemsleyanums belongs to the folk medicine; its root can be used to treat children’s wind-heat with water decoction which is recorded in local standardization, such as “Guangxi min jian chang yong shou yi cao yao” [3], “Kunming min jian chang yong cao yao” [4], and “Zhejiang min jian chang yong cao yao” [5]. The modern pharmacological studies had shown that it had effects of anti-inflammatory, analgesic, antipyretic, antivirus, antitumor [6–8], immunomodulatory [9], liver protection, etc.
In recent years researchers have paid more and more attention to flavonoids and anthraquinones which wildly exist in Chinese herbal medicine because of their outstanding bioactivities. Modern phytochemical studies have shown that flavonoids are the major active constituents in T. hemsleyanums [10]. Out of them, vitexin, isoquercetin, and astragalin showed the anticancer effect [11–13], quercitrin could improve the impaired mesenteric vascular activity in the colitis models [14], quercetin could yield an obvious mitigation of arthritic manifestations [15], kaempferol had effects of preventing and reversing ventricular fibrosis and cardiac dysfunction in vivo and in vitro experiments [16], and rutin played a therapeutic role in diabetic atherosclerosis through inhibiting premature senescence of vascular smooth muscle cells [17]. Because of their prominent pharmacological activities, determination for the contents of the above compounds in T. hemsleyanums and also the LC-MS characterizations of the plant are required for further quality control study.
It was reported recently that UPLC-Triple-TOF/MS had been applied for the qualitative analysis of the chemical compositions in T. hemsleyanums, and most of them were identified as flavonoids, esters, and benzene sulfonic acids [18]. UPLC-Q-Exactive/MS is a more powerful and sensitive analytical instrument for detecting constituents even in low abundance in complex plant extracts [19]. In this present study we use UPLC-Q-Exactive/MS technique firstly to analyze the components of T. hemsleyanums collected from different cultivation places in China. The data generated from UPLC-Q-Exactive/MS have distinguished the phytochemical differences among the 18 batches both in types of ingredients and in contents. Various factors, such as cultivation places, harvest season, and postharvest treatment, are responsible for the variation in contents of phytochemicals and hence are due to the variation of many bioactivities.
We performed the antiviral determination assay using recombinant influenza virus PR8-NS1-Gluc to evaluate the anti-influenza virus bioactivity of the extracts from T. hemsleyanums. It is a high throughput screening protocol to identify entry inhibitors for influenza virus using a human immunodeficiency virus-based pseudotyping platform which can be performed in a BSL-2 facility. Conventional methods of active compound discovery from natural products involve bioactivity-guided isolation, which is laborious, costly, or impossible to source. In addition, the effects of underlying constituents with potent bioactivity present at very low or sometimes undetectable levels may be ignored. In this study, we process the data with SPSS correlation analysis to evaluate the association between chemometrics and bioactivities of the 18 batches of T. hemsleyanums from different districts. Content of 8 constituents of flavonoids showed positive correlation coefficient with anti-H1N1 influenza virus activity, which may facilitate the identification of potential antiviral components in T. hemsleyanums.
2. Materials and Methods
2.1. Chemicals and Plant Materials
HPLC-grade acetonitrile and formic acid were purchased from Thermo Fisher Scientific Company (USA). Ultra-pure water was purchased from Watsons Company (China). All other reagents were of analytical grade.
The reference standards such as kaempferol, isoquercetin, astragalin, rutin, quercetin, baicalein, and palmitic acid were purchased from Herbpurify Co., Ltd. (Chengdu, China); emodin-8-O-β-D-glucoside, quercitrin, vitexin, and vitexin-rhamnoside were purchased from Yuanye Bio-Technology Co., Ltd. (Shanghai, China); isorhamnetin, catechins and epicatechin were purchased from Institutes for Food and Drug control (Beijing, China).
The plant materials of T. hemsleyanums were collected from different provinces in China, as Zhejiang (ZJ1, ZJ2, and ZJ3), Guangxi (GX1, GX2, and GX3), Yunnan (YN1, YN2, and YN3), Fujian (FJ1, FJ2, and FJ3), Guizhou (GZ1, GZ2, and GZ3), and Hubei (HB1, HB2, and HB3). The botanical authentication was performed by Professor Lingchuan Xu from Department of Pharmacognosy, Shandong University of Traditional Chinese Medicine. A voucher specimen (no. TH20171201~TH20181118) was deposited in Key Lab for Natural Product, Shandong Province. The dried rhizome parts of T. hemsleyanums were used for further analysis.
2.2. Liquid Chromatography and Mass Spectrometry Analysis
In this study we employed a UPLC system tandem Q-Exactive/MS spectrometer (Thermo Fisher, CA, USA) equipped with a heated electrospray ionization (HESI) probe. Halo C18 (2.7 μm, 100 × 2.1 mm, Advanced materials technology, USA) column was used at the flow rate of 0.3 mL/min and column temperature of 30°C. The binary solvent system consisted of 0.05% aqueous formic acid (v/v) (A) and 0.05% formic acid in acetonitrile (B). Samples were eluted with the following linear gradients: 15% B at 0–15 min; 15–40% B at 15–22 min; 40–90% B at 22–50 min; 90–15% B at 50–60 min. Injection volume was 3 μL. Detection was performed using a Q-Exactive™ hybrid quadrupole-Orbitrap mass spectrometer in both positive and negative ionization modes. The optimal analysis conditions were set as follows: ion source, heated electrospray ionization probe; source temperature: 350°C; capillary temperature: 320°C; sheath gas: 45 arb; auxiliary gas: 10 arb; mass collecting range: m/z 100-1500. The full scan and fragment spectra were collected at the resolutions of 70000 and 17500, respectively. The collision energy was 30 eV, 50 eV, and 70 eV at negative mode and 10 eV, 30 eV, and 50 eV at positive mode.
2.3. Sample Preparation
The dried rhizome parts of T. hemsleyanums were ground to powder. For each sample, 15 g of rhizome powder was extracted with 75% ethanol reflux for 3 times, 1 h each time. The ethanol extracts were concentrated under reduced pressure evaporated to dryness and then dissolved with 50% acetonitrile of 25.0 mL as the stock sample solution. 1.0 mL of the above stock sample solution, adding baicalein with the final concentration of 4.781 μg/mL as the internal standard (IS), was filtered through a 0.22 μm syringe filter to obtain sample solution for qualitative and quantitative analysis. 20.0 mL of the stock sample solution was freeze-dried in vacuum to obtain a sort of powder extract for antiviral bioactivity testing.
2.4. Preparation of Standard Solutions
Rutin, kaempferol, astragalin, quercitrin, quercetin, vitexin-rhamnoside, isorhamnetin, vitexin, emodin-8-O-β-D-glucoside, and isoquercetin were dissolved in HPLC-grade acetonitrile to achieve standard stock solution with the concentration of 1.0 mg/mL and serially diluted to mixed working solution with 50% acetonitrile; baicalein was used as IS with concentration of 4.781 μg/mL in working solution. All the stock and working solutions were stored at 4°C.
2.5. Identification of the Constituents
LC-MS data were acquired in both negative and positive ion modes and processed for target compounds identification using a combination of Xcalibur and Mass Frontier software packages (Thermo Scientific).
2.6. Validation of Quantitative Method
10 constituents of flavonoids were chosen for the content measurement. Internal standard method was used for quantitation. Peak area ratios of the analyte against IS were used for calculations and a weighted (1/concentration) regression analysis was used for standard curves preparation. Limit of detection (LOD) and limit of quantification (LOQ) for each constituent were detected. The intraday and interday precision of the determination for ten constituents were validated. Stability of sample solutions within 24 hours at 4°C was tested. Six replicates of samples were prepared to check the repeatability. The recoveries were analyzed by adding the analytical reference standards in the powder of T. hemsleyanums prior the extraction procedure and assessed at three levels of the amount which measured in analyte to standards added (1:0.8, 1:1, 1:1.2), using six replicates at each level.
2.7. Quantitative Analysis for Ten Compounds of Flavonoids
Contents measurement of ten compounds of flavonoids in 18 batches of T. hemsleyanums was performed under the condition of Section 2.2.
2.8. Anti-H1N1 Influenza Viral Determination
The antiviral determination assay was performed using recombinant influenza virus PR8-NS1-Gluc as previously described [20]. Briefly, MDCK cells grown in 24 well plates were inoculated with recombinant influenza virus PR8-NS1-Gluc at an moi of 0.01 PFU/cell. After incubation for 1 h at 37°C, virus inoclula were removed and cells were washed. Opti-MEM containing 2 μg/mL TPCK-trypsin was then added for virus propagation. At 36 hours after infection (hpi), 50 μl of culture medium was removed for luciferase assay using BioLux Gaussia Luciferase Assay kit (NEB, USA) according to the manufacturer’s instructions. For antiviral activity determination, extracts of T. hemsleyanums were added at indicated concentrations as shown in Figure 5 during virus propagation.
2.9. Statistics
In order to study the association between the chemical spectra and the anti-influenza viral bioactivity, correlation analysis was performed using the SPSS 17.0 software. The two variables in the correlation analysis were the ion intensity and the inhibitory effect against H1N1 influenza virus. All statistical analyses were two sided. p<0.05 was considered to be statistically significant.
3. Results and Discussion
3.1. Qualitative Analysis
Sample solution of T. hemsleyanums collected from Guangxi province was chosen for qualitative analysis. UPLC-Q-Exactive/MS base peak chromatograms of T. hemsleyanums were shown in Figure 1, and the identification results in negative ion mode and positive ion mode of mass spectrometry were shown in Tables 1 and 2. As shown in the tables, fifty-one compounds were clarified including flavonoids, anthraquinones, esters, fatty acids, phenols, and catechins according to accurate molecular weight calculation, ion fragmentation information, and some of them with confirmation of reference standards.
(a)
(b)
Flavonoids are mainly as follows: compounds 1, 8, 10, 12, 14, 18, 19, 20, 21, 24, and 25 were assigned to be procyanidin dimmer, isorhamnetin-3-pyranosearabinose-7-glucosyl-rhamnoside, kaempferol-3-O-furananose-7-O-rhamnosyl-glucoside, rutin, isoquercetin, kaempferol-3-O-rutone, astragalin, quercetin, vitexin, kaempferol, and isorhamnetin [18], and compound 17 was assigned to be quercitrin [21]. Compounds 11 and 21 were assigned to be vitexin-rhamnoside [21] and vitexin [22]. Compound 21 was also detected in positive ion mode. Compound 49 was the isomer of compound 21. Compounds 42 and 44 were assigned to be orientin [23] and isoorientin [24]. Peak 15, kaempferol-3-O-furanosine-7-O-rhamnose isomer, was detected from T. hemsleyanums for the first time.
Phenolic acids are mainly as follows: compounds 2 and 3 were assigned to be neochlorogenic acid and chlorogenic acid [22]. As we saw in Figure 2, the cleavage fragment strength of the neochlorogenic acid was m/z=191>179>135 (Figure 2(a)). And for chlorogenic acid, the strength of peak m/z=191 was the strongest; the strength of peaks m/z 179 and m/z 135 was equivalent and very weak (Figure 2(b)).
(a)
(b)
Compound 9 was assigned to be 4-hydroxy-3-methoxybenzaldehyde, of which fragmentation patterns was shown in Figure 3. Compounds 16 and 26 were assigned to be salicylic acid and rock acid [18].
Anthraquinones are mainly as follows: compound 22 was assigned to be emodin-8-O-β-D-glucopyranoside [25]. The [M-H]- ion of compound 22 was at m/z 431.1557. Glucose group was lost to produce ion of m/z 269.1029. The fragmentation patterns were summarized in Figure 4.
(a)
(b)
(c)
Catechins are mainly as follows: compounds 27 and 29 were assigned to be gallocatechin and epigallocatechin [23]. Compound 29 produced characteristic ions of m/z 125.0957 and m/z 137.0593 which was identified as epigallocatechin. Compound 27 and compound 29 were also detected in positive ion mode. Compounds 4, 5, 6, and 7 were assigned to be protocatechuic acid glucoside, gallic acid, catechins, and epicatechin [26, 27]. Catechins and epicatechin were confirmed by reference standard.
Esters are mainly as follows: compounds 28, 30, 32, 34, 35, and 51 were assigned to be gingergly eolipid A, gingergly eolipid B, lysophosphatidic acid, linoleic acid phosphatidic acid, 4-(2-dodecyl)-benzene-sulfonate, and methyl linolate [18].
Fatty acids are mainly as follows: compounds 39 and 50 were assigned to be palmitic acid and linoleic acid. Additionally, the loss of H2O group was considered as the representative fragmentation pathway in acids [28]. Palmitic acid was confirmed by reference standard.
The other compounds are mainly as follows: compounds 31, 33, and 36 were assigned to be 1-linoleoylglycero-2-phosphor-ethanolamine, soya-cerebroside I, and 4-(1-methyl-dodecyl)-benzene-sulfonic acid [18].
3.2. Quantification Method Validation
As shown in Table 3, standard curves of the ten compounds exhibited good linearity in the range of 1.6 to 1330 μg/mL, with coefficients of correlation ranging from 0.9961 for kaempferol to 0.9998 for quercetin. The parameters of LOD were from 0.1045 to 2.0781 pg, and LOQ were from 0.9290 to 15.5142 pg, which were sensitive enough to the detection of analytes. The intraday and interday precisions of present method were shown in Table 4, and the respective RSD values were from 0.03 to 2.79% and from 0.72 to 2.81%. The concentration stability of the constituents in samples kept at 4°C for 24 hours (n = 6) ranged from 0.28 to 3.42%. In addition, the sample solutions for T. hemsleyanums were prepared in parallel (n=6) to evaluate the repeatability and achieved the RSD of 0.26-2.69%. Based on the above methodology verification, the assay is reproducible and suitable for accurate and precise quantification of these ten chemical constituents in T. hemsleyanums (shown in Table 4). As shown in Table 5, the recoveries of ten constituents was from 94.86 to 99.9 % with the RSD value <3 %.
3.3. Quantitative Analysis
The validated UPLC-Q-Exactive/MS method was used for the content measurement of 10 constituents of rutin, kaempferol, astragalin, quercitrin, quercetin, vitexin-rhamnoside, isorhamnetin, vitexin, emodin-8-O-β-D-glucoside, and isoquercetin in T. hemsleyanums. The content of each constituent was calculated in terms of its respective calibration curve and the result was listed in Table 6. The content of vitexin was relative abundant, of which could reach the amount of 0.7079±0.0286 mg/g in that from YN1, whereas vitexin-rhamnoside may be the lowest in content since it could even not be detected in some batches. The result suggested that contents of these constituents fluctuated in different batches even for those that from the same production place. The total amounts of these ten compounds in T. hemsleyanums of YN1, GZ1, and GX3 were the top three highest, whereas for samples from Zhejiang province, amounts of flavonoids are shown to be the lowest. Many factors including growing environment of climate temperature, rainfall, sunlight, soil nutrient, etc. and harvest time and postharvest treatment may affect the content of bioactive compounds in T. hemsleyanums and thus may affect its efficacy.
3.4. Anti-H1N1 Influenza Virus Activity
Since T. hemsleyanums from Zhejiang province is popular on market, we took sample of ZY2 as the representative sample for anti-influenza virus pretest. As shown in Figure 5(a), extract of ZJ2 inhibited influenza virus replication in a dose-response manner, and the IC50 is 27.4μg/mL, while no cytotoxicity was observed at a concentration of as high as 200μg/mL (Figure 5(b)). Ribavirin at 100μM was used as positive control. Antiviral activities of the other 17 batches were shown in Figure 5(c) that all of them showed inhibitory effect against H1N1 influenza virus replication as good as ZJ2 at the concentration of 50μg/mL. All three sample batches from Guizhou province (GZ1, GZ2, and GZ3, 94.0-98.0%) and Hubei province (HB1, HB2, and HB3, 93.5-97.3%), two batch from Guangxi province (GX1 and GX2, 94.3-94.5%), one batch from Fujian (FJ1, 96.2%), and one batch from Yunnan (YN1, 98.6%) exhibited higher anti-influenza virus activities than the others, which were probably due to the higher values of flavonoids in these samples.
3.5. Statistical Analysis
In order to determine the contribution of various constituents for the antiviral activity, the spectrum-effect relationships between LC-MS fingerprints and inhibitory effect for influenza virus were evaluated with correlation analysis statistical method. Since normal distribution was not satisfied, Spearman was used for all the analyses. When the correlation coefficient is greater than 0, it indicates that a component is positively correlated with antiviral activity, and the larger the value, the stronger the correlation. When the correlation coefficient is less than 0, it means that it is negatively related to antiviral activity. Results were shown in Table 7 which revealed that 10 peaks had significant correlationship with anti-influenza virus activity. 8 of them, which were identified as rutin, kaempferol, astragalin, quercitrin, quercetin, kaempferol-3-o-rutinoside, procyanidin dimmer, and epicatechin, were positively related to antiviral activity, of which the highest correlationship was with astragalin (r = 0.711), followed by epicatechin (0.641), quercitrin (0.614), and quercetin (0.617). 2 of the 10 peaks, identified as vitexin (-0.5370) and isorhamnetin-3-pyranose arabinose-7-glucosyl rhamnoside (-0.6630), indicated negative correlationship with antiviral activity. That is, for constituents with positive coefficient, which may have underlying anti-H1N1 influenza effect, the higher the content, the better the bioactivity. This result was in agreement with literatures, in which quercetin, quercitrin, rutin, kaempferol, and isoquercitrin were reported to have anti-H1N1 influenza virus activity [29–32]. Thereby it may provide valuable clues for further discovery of active anti-influenza virus compounds. LC-MS chemometrics are a kind of potent method for analyzing the complex system of TCM herbs, but lacking of bioactive characteristics. Therefore, chemical fingerprints combined with bioactivity evaluation to construct a fingerprint-efficacy relationship is a good way to explore the bioactive constituent of TCM herb and may lend material basis supporting to the quality control of T. hemsleyanums
4. Conclusion
In our study, an efficient UPLC-Q-Exactive/MS method was established for qualitative analysis of up to 51 constituents identified in T. hemsleyanums for the first time and also for quantitative analysis of 10 constituents in 18 batches collected from six cultivation places. Method validation showed the ideal sensitivity, stability, and reliability of the analysis method. It was also the first time for UPLC-Q-Exactive/MS and Gaussia Luciferase viral titer assay being combined to reveal the underlying anti-influenza virus bioactive compounds in T. hemsleyanums. 8 possible active constituents were discovered showing positive contribution for antivirus effect, and five of them are consistent with reports published in literatures. This analytical method led to the identification of compounds with anti-H1N1 influenza virus activity in T. hemsleyanums which may be difficult to be extracted and identified by traditional bioactivity-guided fractionation procedures and may provide an available reference mode for revealing the material basis in traditional Chinese herb of which thereby could also provide reference for quality assessment of TCM herbs.
Data Availability
Most of the data (Figures 1 to 5 and Tables 1 to 7) used to support the findings of this study are included within the article. Some of the data used to support the findings of this study are included within the supplementary information file.
Conflicts of Interest
The authors declare no conflicts of interest.
Authors’ Contributions
FuJuan Ding and JiangTing Liu contributed equally to this work.
Acknowledgments
The research was financially supported by Key Technology Innovation Project of Important Industries in Shandong Province (no. 2016CYJS08A01-5), National Natural Science Foundation of China (no. 81774167), Key Research and Development Program in Shandong Province (no. 2018CXGC1307), and College Science and Technology Plan Project of Shandong Province (J16LM12).
Supplementary Materials
Table 1: the information of T. hemsleyanums. Table 2: parameters of LC-Q-Exactive /MS analysis for 10 constituents. Figure 1: total ion chromatogram of 18 batches of T. hemsleyanum. (Supplementary Materials)