Application of High-Performance Liquid Chromatography Coupled with Linear Ion Trap Quadrupole Orbitrap Mass Spectrometry for Qualitative and Quantitative Assessment of Shejin-Liyan Granule Supplements

A method for high-performance liquid chromatography coupled with linear ion trap quadrupole Orbitrap high-resolution mass spectrometry (HPLC-LTQ-Orbitrap MS) was developed and validated for the qualitative and quantitative assessment of Shejin-liyan Granule. According to the fragmentation mechanism and high-resolution MS data, 54 compounds, including fourteen isoflavones, eleven ligands, eight flavonoids, six physalins, six organic acids, four triterpenoid saponins, two xanthones, two alkaloids, and one licorice coumarin, were identified or tentatively characterized. In addition, ten of the representative compounds (matrine, galuteolin, tectoridin, iridin, arctiin, tectorigenin, glycyrrhizic acid, irigenin, arctigenin, and irisflorentin) were quantified using the validated HPLC-LTQ-Orbitrap MS method. The method validation showed a good linearity with coefficients of determination (r2) above 0.9914 for all analytes. The accuracy of the intra- and inter-day variation of the investigated compounds was 95.0–105.0%, and the precision values were less than 4.89%. The mean recoveries and reproducibilities of each analyte were 95.1–104.8%, with relative standard deviations below 4.91%. The method successfully quantified the ten compounds in Shejin-liyan Granule, and the results show that the method is accurate, sensitive, and reliable.


Introduction
Pharyngitis is an inflammation of the pharyngeal mucous membrane and submucous lymphoid tissues that affects people of all ages around the world. Patients often report pain and irritation in the throat. In certain cases, the illness is not caused by an infection; in such a setting, antibiotics may be the wrong choice for treatment or they may only offer a modest improvement in symptoms [1,2]. Traditional Chinese medicines have been successfully used to treat pharyngitis for thousands of years because of their variety of multi-target, therapeutic, and synergistic effects [1,3,4].
Shejin-liyan Granule (SJLYKL) is a traditional Chinese herbal medicine that has been used to treat pharyngitis in clinical practice. It consists of six herbs: Rhizoma Belamcandae, Arctium lappa, radix Sophorae tonkinensis, Physalis alkekengi, radix Platycodonis, and radix Glycyrrhizae. These chemicals, such as the isoflavones from Rhizoma Belamcandae [5], physalins from Physalis alkekengi [6], ligands from Arctium lappa [7], and alkaloids from Sophorae tonkinensis [8] have been proven to have good anti-inflammatory, antibacterial, and antioxidant activities [9][10][11]. SJLYKL is an organic combination of complex and diverse chemical constituents from these six herbs, and its anti-pharyngitis effect may be closely related to these compounds. Therefore, systematically analyzing the constituents of SJLYKL could provide an interpretation of the material basis for its pharmacological effects.
The complete profiles and quantities of the bioactive ingredients in SJLYKL are not well understood. Several studies have focused on identifying the components in each of the herbs included in SJLYKL [5,6,[12][13][14]; however, no method has been developed to systematically analyze SJLYKL. Therefore, it is important to develop a systematic qualitative and quantitative evaluation method for the pharmacologically active compounds in SJLYKL.
Liquid chromatography-mass spectrometry (MS) has become increasingly common because of its high selectivity and sensitivity. The combination of a linear ion tap with a high-resolution Orbitrap analyzer has been used for qualitative and quantitative analyses in various applications, including bioactive compounds of traditional Chinese medicines, metabolites, and drug abuse [15][16][17]. In this study, we developed a simple and rapid method for the comprehensive qualitative and quantitative analyses of the major constituents of SJLYKL. This is the first time that a technique using high-performance liquid chromatography coupled with linear ion trap quadrupole Orbitrap high-resolution mass spectrometry (HPLC-LTQ-Orbitrap MS) has been applied to identify and quantify the components in SJLYKL.

Sample Preparation
SJLYKL (0.1 g) was accurately weighed and extracted with 10 mL of 75% methanol in an ultrasonic bath at room temperature. The weight loss was compensated by adding 75% methanol after extraction, and the extract was centrifuged at 17,000× g for 10 min. The resulting supernatant was diluted by a factor of 10 using pure water; then, a 10-µL aliquot of the supernatant was injected into the HPLC-LTQ-Orbitrap MS system for analysis. Samples that were above the upper limit of quantification were diluted and reanalyzed.

Chromatographic and Mass Conditions
Chromatographic separation was performed on a Dikma ODS C 18 column (150 × 4.6 mm, 5 µm) maintained at 30 • C using a Dionex HPLC system (Thermo Fisher Scientific, Sunnyvale, CA, USA). The gradient elution was 30 min at a flow rate of 1.0 mL/min using solvent A (water with 0.1% acetic acid and 2 mM ammonium acetate in water) and solvent B (acetonitrile) as the mobile phase. The elution was run on the following schedule: 10-20% B at 0-5 min, 20 For qualitative analysis, the Orbitrap resolution of the survey scan was set to 30,000 and that for the MSn scan was set to 15,000. The data-dependent MS2 scanning was performed to trigger fragmentation spectra of the target ions and to prevent repetition by the dynamic exclusion settings. Peaks were identified by comparison with those of the standards. For those peaks that did not correspond to the standards, a database including about 200 major compounds was established by collecting information from the literature on the six herbs in SJLYKL, including their names, formulas, accurate molecular weights, and MS2 information. The accurate masses of the additive ions, such as [M + H] + , [M + Na] + , [M − H] − , and [M + HCOO] − , were also calculated. The MS detection was performed in selected ion monitoring mode to quantify the ten compounds, including matrine, galuteolin, tectoridin, iridin, arctiin, tectorigenin, glycyrrhizic acid, irigenin, arctigenin, and irisflorentin.

Method Validation for Quantitative Analysis
Stock standard solutions of these ten compounds were separately prepared in methanol and kept at 4 • C. A mixed working solution was prepared before each use and diluted to the appropriate concentration to create calibration curves. The calibration curve of each compound was prepared using at least five different concentrations. The external standard method was constructed using the area with respect to known concentrations of the test compound (C, µg/mL) and weighting of reciprocal concentrations (1/C 2 ). The limits of detection (LOD) under the present chromatographic conditions were determined at a signal-to-noise ratio of 3.
The intra-day and inter-day accuracies (deviation from the nominal concentration (%)) and precisions (relative standard deviation, RSD%) were analyzed at different concentrations on one day, and this experiment was repeated for three consecutive days.
To evaluate the recovery of the method, known amounts of these compounds were added to SJLYKL, and the samples were then quantified as described above. The recovery of each analyte was calculated according to the following equation: where A det is the total detected amount of each compound, A orig is the original amount of each compound in SJLYKL, and A spi is the spiked known amount of each component. The amount of the target compound was calculated using the corresponding calibration curve.
To investigate the repeatability, five samples from the same batch of SJLYKL were accurately weighed and treated as described above. The sample stability was assessed by analyzing SJLYKL samples stored at 4 • C, after 0, 4, 8, and 24 h.

Development of the Extraction Method
The factors that affect the extraction efficiency, including the extraction solvents, extraction method, and extraction times, were investigated to optimize the sample extraction efficiency. The ultrasonic bath extraction method was convenient and effective for the examined components; therefore, further experiments were performed using ultrasonic bath extraction. Several components, such as arctiin, arctigenin, tectoridin, and tectorigenin, were not completely extracted using pure methanol as a solvent; thus, different water-methanol ratios (0:100, 25:75, and 50:50, v/v) were screened. The yield of arctiin, arctigenin, tectoridin, and tectorigenin increased significantly when extractions were performed with 75% or 50% methanol, and fewer interfering peaks were found when 75% methanol was used. The duration of the extraction (30 min, 40 min, 50 min, or 60 min) was also investigated to optimize the extraction procedure. The optimal extraction of SJLYKL (0.1 g of powder) was obtained using 10 mL of 75% methanol in an ultrasonic water bath for 30 min.

Profiles of Ingredients in the SJLYKL Extract
In the analysis of the obtained chromatographic peaks by the HPLC-LTQ-Orbitrap MS method, we excluded the peaks which were the obtained parent accurate molecular weights without product ion. Then, to prevent misunderstanding caused by interference peaks, some peaks with an absolute intensity lower than 104 were removed. As a result, by comparing the information collected from the literature and standards with the data obtained by the HPLC-LTQ-Orbitrap MS method mass spectrometry, a total of 54 compounds were identified from the SJLYKL extract, including fourteen isoflavones, eleven ligands, eight flavonoids, six physalins, four triterpenoid saponins, six organic acids, two xanthones, two alkaloids, and one licorice coumarin ( Table 1). The chemical structures of the 54 compounds are available online as Supplementary Materials. Typical peak chromatograms in positive and negative ion modes are shown in Figure 1. Fourteen isoflavones, including seven isoflavone O-glucosides, four aglycones, and three isoflavones with methylenedioxy groups, were unambiguously or tentatively identified [18]. By comparing the retention times and the MS spectra of the SJLYKL extract with those of the standards, five peaks were unambiguously characterized as tectoridin (peak 1), iridin (peak 2), tectorigenin (peak 3), irigenin (peak 4), and irisflorentin (peak 5). Peaks 6, 7, 8, and 9 exhibited a high intensity [M + H − 162] + ion, and were tentatively characterized as iristectorin B, iristectorin A, isoiridin, and 3,5-dimethoxyirisolone-4-O-glucoside, respectively, based on daughter ions [19]. Peak 10 showed a loss of 162 Da at m/z 463.1236 and an aglycon ion [M + H − 162 − 162] at m/z 301.0707; therefore, this compound was inferred as tectorigenin-7-O-glucosyl-4-O-glucoside [5]. Besides known aglycones, peaks 11 and 12 shared the same molecular ions and fragment pathways, and they were deduced as iristectorigenin A and iristectorigenin B, respectively, based on reported data [5]. Peaks 13 and 14 yielded diagnostic ions at [M + H − CH 3 ] + and [M + H − 2CH 3 − CO] + , so they were tentatively characterized as noririsflorentin and dichotomitin, respectively [13].  15,18,19,20, and 21 were identified as matairesinol, matairesinoside, lappaol H, lappaol A, and lappaol F, respectively [12,20]. Peaks 22,23,24, and 25 were of isomers with the adduct ions of [M + Na] + at 577.2048 in positive mode, and they yielded fragment ions at [M + H − CH 3 ] + and [M + H − OCH 3 ] + . These four peaks were ambiguously assigned as lappaol C, isolappaol C, lappaol E, and arctignan A, respectively [7,12].
Six physalins were detected. Peaks 34 and 35, with the same deprotonated molecular ion at m/z 543.1866 in negative ion mode, represent a pair of stereoisomers. Their parent and product ions were in agreement with physalin D and physalin D', and the polarity of physalin D' was stronger than that of physalin D; the compound with the shorter retention time should be physalin D' [6]. Using the same method, the other two pairs of stereoisomers, peaks 36 and 37, and peaks 38 and 39, were identified as physalin F and physalin A, and physalin O and physalin L, respectively [6,13,21].
The four triterpene saponins were acidic saponins, and glycyrrhizic acid (peak 40) and glycyrrhetinic acid (peak 41) were identified by comparing their retention times and accurate masses with those of the standards. Peak 42 showed [M + H] + ions at m/z 839.4070, which is 16 Da greater than that of glycyrrhizic acid, so it could be assigned to licorice saponin G2 [14].  [23].
Organic acids and alkaloids were identified by comparing their retention times and accurate masses with those of the standards. Peaks 47, 48, and 49 were identified as chlorogenic acid, neochlorogenic acid, and cryptochlorogenic acid, respectively, by comparison with reference compounds. Similarly, because of available standards, peaks 50, 51, 52, 53, and 54 were unequivocally identified as isochlorogenic acid B, isochlorogenic acid A, sochlorogenic acid C, oxymatrine, and matrine, respectively.

Quantitative Analysis
Extracted ion chromatograms of the standard sample and SJLYKL samples are shown in Figure 2. The results of regression analysis and LOD values for the ten compounds are shown in Table 2.
All calibration curves showed good linearity (r 2 > 0.9914) between the peak area and concentration. The accuracy of the intra-and inter-day variation of these investigated compounds was 95.0-105.0% and the precision values were less than 5.0% (Table 3). The results of the recovery and repeatability test are shown in Table 4. The recovery of all analytes was 95.1-104.8%, and the RSD values of the repeatability results were less than 4.94%. The sample solution was stable for 24 h at 4 • C (RSD < 4.31%, data not shown). Then, the proposed method was applied to analyze ten compounds in five SJLYKL samples. The identified levels of these compounds are summarized in Figure 3. Arctiin was present in the highest concentration, followed by tectoridin. These two components provide good anti-inflammatory effects [24,25] and may be active ingredients in SJLYKL. Therefore, these compounds could be marker compounds for quality control of SJLYKL.

Conclusions
In this study, a simple, accurate, and reliable HPLC-LTQ-Orbitrap MS method was established to qualitatively determine the 54 components of SJLYKL. The method successfully quantified ten major components in five batches of SJLYKL samples. This novel approach was useful to identify constituents and control the quality of SJLYKL. These results offer useful information for understanding the material basis of the therapeutic effects of SJLYKL and for its clinical application.

Conflicts of Interest:
The authors declare no conflict of interest.