Identification and Screening of Natural Neuraminidase Inhibitors from Reduning Injection via One-Step High-Performance Liquid Chromatography-Fraction Collector and UHPLC/Q-TOF-MS

Neuraminidase plays an essential role in the spread of influenza viruses via cleaving sialic acids from the host cell receptors and virions. Neuraminidase has been regarded as an essential target for prevention and treatment of influenza infection. ,e one-step high-performance liquid chromatography-fraction collector (HPLC-FC) was selected to prepare fractions from Reduning (RDN) injection, while ultra-high-performance liquid chromatography/quadrupole-time-of-flight mass spectrometry (UHPLC/Q-TOFMS) was used to identify fractions depending on their retention time and molecular ion. As a result, 75 fractions were prepared and 28 fractions out of them exhibited NA inhibitory effects with the dose-effect relationship. Exploring it further, six components including neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C were the main components that accounted for almost 80% of inhibitory activity of RDN injection. Accordingly, these results demonstrated that this strategy could not only rapidly identify but also accurately screen active components from traditional Chinese medicine.


Introduction
Influenza virus is still a threat to human lives. It possesses two major glycoproteins enzymes, hemagglutinin (HA) and neuraminidase (NA) [1,2]. HA binds to the host cell receptors to facilitate viral entry, while NA facilitates the spread of the infection by releasing newly synthesized virions into neighboring cells [3]. Replication, infection, and spreading play a key role in the pathogenesis of the influenza virus, especially for spreading. Accumulative evidences demonstrated that NA played an essential role in the spread of influenza viruses by cleaving sialic acids from the host cell receptors and virions [4,5].
erefore, NA has been regarded as an essential target for prevention and treatment of influenza infection. Besides, NA could also be used as an important candidate antigen for universal influenza vaccines [6]. Zanamivir and oseltamivir, as NA inhibitors (NAIs), are the most common drugs for combating the influenza virus. However, drug-resistant virus of these drugs has emerged [7]. Moreover, oseltamivir is expensive and zanamivir cannot to be taken orally. Accordingly, with an increasing appearance on the drug resistance of antiviral drug [8][9][10], screening novel NAIs from Traditional Chinese Medicine (TCM) has attracted great attention towards the treatment of influenza.
Traditional Chinese Medicines (TCMs) played important roles in clinical practice over the past years. It has been widely accepted by most countries owing to its long-term clinical use and reliable efficacy. However, there are still some challenges for exploring the pharmacology function of TCM because the chemical structures of active ingredients are largely undefined. In general, traditional methods for screening active components have relied on animal and cell models, which was complex to operation and time-consuming. Hence, it was essential to establish a rapid, convenient, and accurate method to screen and identify the bioactive components in TCM.
Reduning injection (RDN) contains some effective ingredients extracted from three herbs (Flos Lonicerae, Herba Artemisiae annuae, and Gardenia jasminoides). It has been widely used as an antipyretic and anti-inflammatory drug in China [11]. In clinic, RDN injection not only has been used to treat common cold, cough, acute upper respiratory infection, and acute bronchitis, but also has a potential therapeutic effect on influenza. A novel online DPPH-CE-DAD method [12] was developed for quality assessment of RDN injection and a reliable pharmacology model [13] was constructed for understanding the molecular mechanisms of multiple components from RND injection, respectively. e pharmacological target at signaling pathways level and the potential target proteins and the key pathways of RDN injection have also been investigated [14,15]. However, there is no report on screening and identification of NAIs from RDN injection. Furthermore, a great number of researchers paid attention to screening active components from TCM. Ge et al. [16] established an activity-integrated strategy to screen glucosidase inhibitors from Coptis chinensis and Lu et al. [17] developed a peak fractionation approach to screen direct thrombin inhibitors from Radix Salvia miltiorrhiza, respectively [18]. Taking these methods into consideration, there was little literature to screen NA inhibitors from RDN injection and summarize the main active components for elucidating the effective substances of RDN injection.
In this study, an accurate strategy for preparing, screening, and identification of NAIs from RDN injection was developed based on NA inhibitory activity. e 75 fractions were prepared by high-performance liquid chromatography-fraction collector (HPLC-FC) in one run depending on the retention time. Following that, ultrahigh-performance liquid chromatography (UHPLC) and ultra-high-performance liquid chromatography/quadrupole-time-of-flight mass spectrometry (UHPLC/Q-TOF-MS) were used for the precise identification of each fraction and NA inhibitory activities of all fractions were performed with the same time. In total, 67 components were identified from RDN injection and 15 active components with an inhibitory effect against NA were screened and identified. It was found that neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C were the main active components that accounted for 80% of inhibitory activity among the RDN injection. Accordingly, the one-step work can not only identify 67 components from RDN injection, but also screen natural NAIs from TCM and provides a promising strategy for elucidating the effective substances of TCM.

Chemicals and Materials.
Reduning injections were offered from Jiangsu Kanion Pharmaceutical Co., Ltd. (Lianyungang, China). Acetonitrile (Dikma Technologies Inc., USA), methanol (Tianjin Concord Science Co., Ltd, Tianjin, China), and formic acid (Tianjin Kermel Science Company) were of HPLC grade. Deionized water was purified with a Milli-Q Academic ultrapure water system (Millipore, Milford, MA, USA) and reference standards including neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C were purchased from Chengdu Must Biotechnology Co., Ltd. (Chengdu, China). Neuraminidase inhibitors screen kits were also purchased from Beyotime Biotechnology.
A Waters ACQUITY UHPLC System (Waters Co., Milford, MA) equipped with a photodiode array detector was used in the range of 190-400 nm. e system was controlled by Empower 2 workstation. An ACQUITY UHPLC BEH C 18 (1.7 μm, 2.1 × 50 mm) column was used for separation with the mobile phase 0.1% (v/v) formic acid aqueous solution (A) and acetonitrile (B) using a gradient elution: 0-4 min, 4-4% B; 4-6 min, 4-5% B; 6-9 min, 5-7% B; 9-11 min, 7-9% B; 11-12 min, 9-12% B; 12-13 min, 12-14% B; 13-15 min, 14 e identification of each fraction of RDN injection was performed by an Agilent UHPLC-Q-TOF/MS system. e ESI-MS spectra were acquired in the negative ion mode. e capillary voltage was set to 3.0 kV for the negative ion mode, high-purity nitrogen was set as the nebulization, and auxiliary gas with the gas pressure was 35 psig. e velocity of drying gas was 9.0 L min −1 with the temperature of 350°C, the fragmentor voltage was at 175 V, skimmer voltage was 65 V, and the collision energy (CE) was at 30, respectively.

2
International Journal of Analytical Chemistry Data was collected at the scan range of 100-1500 Da for MS and 50-1500 Da for MS/MS. Data acquisition, exact mass, and elemental composition analyses were controlled by Mass Hunter software (Agilent Technologies). e mobile phase, column, column temperature, flow rate, and gradient program used in the UHPLC-Q-TOF/MS analysis were the same as the UHPLC analysis.

Preparation of Sample
Solutions. An Agilent series 1260 HPLC system with Fraction collector (BSZ-100, Shanghai Qingpu Huxi Instrument, Shanghai, China) was used for the fractions collection. e RDN injection was injected into the HPLC system for the separation of each compound. Fractions were automatically collected by using a fixed time of 1 min. As a result, a total of 75 fractions were prepared. ey were evaporated to dryness by the nitrogen gas and the residues were redissolved and diluted for the activity assay. us, the sample solutions for the bioassays were obtained. In order to screen the NA inhibition effects with dose-effect relationship, three different concentrations (low, medium, and high concentration) were prepared for activity assay.
erefore, the dried residue was reconstituted in 150 μL water, 75 μL was for activity assay (high concentration), 75 μL was diluted with 75 μL water (medium concentration), and finally the low concentration was diluted with 75 μL water.

Bioassay of NA Inhibitors. 70 μL buffer solution was
added to a 96-well plate and different doses (0, 1, 2, 5, 7.5, and 10 μL) of NA were mixed with it. en different volumes (20, 19, 18, 15, 12.5, and 10) μL of Milli-Q water were added to reach the total volume to 90 μL, respectively. It was then incubated for 2 min at 37°C and then 10 μL of NA fluorescence substrate was added into the plate, incubating for 30 min before detection. e fluorescence value (FV) of wells was measured at 360 nm of excitation wavelength and 450 nm of emission wavelength by Flex Station 3 Microplate Reader. e relationship between the dose of NA and FV is stated below: 70 μL buffer of solution was added to a 96-well plate. 10 μL of NA and 10 μL of each fraction were added into it. e volume was then adjusted to 90 μL with 10 μL of Milli-Q water. After incubated for 2 min at 37°C, 10 μL of NA fluorescence substrate was then added into the plate, incubating for 30 min before detection. e inhibition rate of each fraction was calculated by the following formula: where FVfm represented the FV of each fraction and FVfb represented the blank background of each fraction.

Optimization of Chromatographic
Conditions. e HPLC conditions were considered for obtaining chromatograms with better resolution. e different mobile phases (acetonitrile-water, methanol-water), concentrations of additive (0.1%, 0.2%, and 0.3% formic acid and phosphoric acid), and column temperatures (20, 30, and 40°C) were optimized to achieve better separation. By comparing resolutions and the peak shapes of the investigated components, the best separation was achieved when acetonitrilewater was selected as mobile phase and 0.1% formic acid was chosen as an additive to the water phase. e column temperature and flow rate were optimal at 40°C and 1 mL min −1 , respectively. e HPLC chromatogram is shown in Figure 1.
e UHPLC conditions were also considered as a requirement for obtaining chromatograms with better peak resolution in the same way as the HPLC. But the flow rate was set as 0.3 mL min −1 .
e UHPLC chromatogram is displayed in Figure 2.

Identification of 67 Components in RDN Injection Using UHPLC/Q-TOF-MS.
e peak number, retention time, MS/MS fragmentation ions information, and identification of 67 components are listed in Table 1 [29], and 3,5-di-O-Caffeoyl-4-O-(3-hydroxy-3-methyl)glutaroylquinic acid [22] were therefore identified. e same pattern was with peak 60 weighed 723.2134 (C 33 H 40 O 18 ) and therefore it was suspected to be Jasmigeniposide A [22]. Peaks   us, artemisitene was detected at peak of 51, and peak 48 was identified as artemisitene [38].     [39]. e total ion chromatography of 67 components from RDN injection is shown in Figure 3. In order to further confirm the main components from RDN injection, the reference standards were injected to UHPLC-Q-TOF/MS. As shown in Table 2, the precursor ions and MS/MS fragmentation ions of neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C were the same as those in RDN injection (peaks 5, 20, 25, 57, 58, and 59), which gave the definite evidence to support the results of identification. e total ion chromatography of reference standards is shown in Figure 4.

Screening the Potential Active Fractions of
NIs. An in vitro activity assay was performed to screen the potential NAIs. Only fractions with dose-effect relationships were thought of as potential active components, which were identified and validated as the NAIs. Some components had NA inhibition just for only one concentration but did not exhibit dose-effect relationship and therefore could not be selected as active components. On the other hand, some fractions showed the inhibition with dose-effect relationships, but the inhibition rates of each concentration were just below the limit activity at 6%. Hence, they were not selected as active components. e reason why the rate was chosen at 6% was that the systematic error was 6% obtained by the calibration curve for screening the NAIs. In total, the inhibitory ratios of 28 fractions which were truly active fractions among the 75 fractions (fractions 3,4,5,6,8,15,18,19,20,21,22,23,25,29,30,31,32,38,39,40,46,47,66,67,68,69,70, and 71) increased with an increasing concentration, which exhibited obvious relativity between the dose and the effect as shown in Figure 5. ese fractions have been indicated as the potential NAIs.

Verifying the Potential Active Fractions via UHPLC and UHPLC-Q-TOF/MS. A total of 75 fractions were prepared.
After screening the inhibitors, each fraction was injected to the UHPLC again. Comparing each fraction with RDN injection, the components from each fraction can be primarily identified depending on their retention time and corresponding peak number. e results can be shown in Figure 6. e fractions (3,4,5,6,8,15,18,19,20,21,23,25,27,29,30,31,32,38,39,40,46,47,66,67,68,69,70, and 71) contained peaks which have the same retention time with RDN injection. Finally, the corresponding relationship between fraction number and peak number of the potential active components with NA inhibition is shown in Table 3.

Confirmation of the Main Components with NA
Inhibition. In order to confirm the bioactivity of the components, the main active components with NAIs rates reaching 50% including 6 components, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C, were selected to be further validated. IC 50 value was determined to be the effective concentration at which the activity of NA was inhibited by 50%. As a result, the IC 50 value of neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C were calculated as 23.79 μg mL −1 , 17.40 μg mL −1 , 231.7 μg mL −1 , 40.33 μg mL −1 , less than 10 μg mL −1 , and 14.52 μg mL −1 , respectively. Studying it further, the total inhibition ratio of these 6 components has reached almost 80% of RDN injection. As shown in Figure 7, comparing the inhibition ratio of RDN injection and 6 components whose concentrations were the same with RDN injection, it was found that the inhibition ratios of screening 6 active components and RDN injection were almost the   same, indicating that the screening 6 active components could reflect the total inhibition ratio of RDN injection against NA. As shown in Figure 7, the inhibition ratios of RDN injection with different dilution times and 6 components seemed to have similar property. Moreover, these components contained the original nucleus of quininic acid and caffeic acid, from which it could be concluded that chlorogenic acids were main active components against NA.

Comparison of the Developed Method and Others.
Nowadays, many researchers focused on the screening active components from TCM. It is therefore essential to establish the direct, rapid, and accurate methods for screening these active components. A great number of screening methods were based on only one concentration of components for evaluating the biological activity. However, these methods have not been proven to evaluate the dose-effect activity of the selected components. In our study, the developed method combined the isolation, screening, and identification of bioactive compounds using the HPLC-FC for the preparation and isolation while using UHPLC and UHPLC/ Q-TOF-MS for identifying. More importantly, three different concentrations (low, medium, and high concentration) were prepared by one-step HPLC-FC method for screening the active components, making the selected active compounds more credible via dose-effect relationship. What is more, the developed method can also identify other components from RDN injection. erefore, the developed method was not only accurate to identify 67 compounds   International Journal of Analytical Chemistry 9