Rapid Determination of Lignans in Schisandra chinensis by Supramolecular Solvent (SUPRAS)-Based Extraction and High-Performance Liquid Chromatography–Tandem Mass Spectrometry (HPLC-MS)

Abstract A supramolecular solvent (SUPRAS)-based sample preparation method was proposed for the determination of lignans in Schisandra chinensis (S. chinensis) fruit. The optimal conditions were: proportion of hexafluoroisopropanol in SUPRAS: 35% (v/v), solid-to-liquid ratio: 20:1 mg/mL, and vortex time: 30 s. Additionally, simple, rapid, and sensitive high-performance liquid chromatography coupled with triple quadrupole mass spectrometry and multiple reaction monitoring was employed to simultaneously quantify seven lignans in S. chinensis. The validation showed that the linearity from 0.01 to 20.83 µg/mL was excellent (all R2 ≥ 0.99). The lowest limits of quantification were from 0.14 to 17.79 ng/mL. The relative standard deviation (RSD) of the intra-day retention time and peak area were from 0.12% to 0.21% and 2.35% to 6.41%. The RSD of retention time and peak area for the inter-day precision ranged from 0.17% to 0.32% and 3.83% to 7.39%. The average recoveries (80.21%–114.53%) demonstrated good accuracy for all analytes. The developed procedure was employed to determine lignans in processed and raw S. chinensis fruit. The lignans were concentrated in the seeds and the wine-processed material had the highest content. This study provides a reference for the analysis of S. chinensis fruit for lignans.


Introduction
Schisandra chinensis (Turcz.)Baill. is a traditional Chinese medicine that grows in North China, Northeastern China, Japan, and Korea (Szopa, Ekiert, and Ekiert 2017).The red fruit of S. chinensis is kidney-shaped and widely applied for the treatment of chronic cough, wheezing, enuresis, frequent urination, dyspnea, diarrhea, and diabetes (Kandhasamy et al. 2018;Onay, Hofer, and Ganzera 2020).According to the literature, lignans are essential bioactive compounds of S. chinensis fruit, especially schizandrol A, schizandrol B, schizantherin A, schizantherin B, schisandrin A, schisandrin B, and schisandrin C (Figure S1), which exhibit anti-inflammatory, anticancer, antioxidant, antifibrosis, and metabolic regulation activities (Gao et al. 2019;Kandhasamy et al. 2018).Interestingly, S. chinensis fruit possesses different pharmacological activities after processing.Vinegar-processed S. chinensis fruit is generally used to treat cough, residual essence, and diarrhea because of its ability to enhance acerbity and astringency.Wineprocessed S. chinensis fruit may benefit the kidney and strengthen its essence.Honeyprocessed S. chinensis fruit has astriction properties and tonifies.
Therefore, it is of great importance to establish an efficient method for the determination of lignans in S. chinensis fruit.High-performance liquid chromatography (HPLC), ultrahigh-performance liquid chromatography (UHPLC), liquid chromatography-tandem mass spectrometry (LC-MS), and gas chromatography (GC) have been employed for this analysis (Liu et al. 2013;Zhang et al. 2016).However, these methods provide long run times, low sensitivity, and the inability to quantify the lignans at low levels (Jiao et al. 2019).
The use of HPLC with triple quadrupole mass spectrometry (HPLC-QQQ-MS/MS) provides excellent accuracy and specificity for the simultaneous determination of multiple targets in complex mixtures (Gai et al. 2021;Wang et al. 2019).Particularly, the multiple reaction monitoring (MRM) mode of HPLC-QQQ-MS/MS (HPLC-QQQ-MS/MS selectively quantifies targets by rapidly scanning the ions of precursors to generate product transitions.This method significantly reduces the analysis time without peak overlap and with low quantitation limits (Jiao et al. 2019;Wu et al. 2016).HPLC-QQQ-MS/MS (MRM) has been frequently used in phytochemistry, drug metabolism, pharmacokinetics, and environmental studies (Gai et al. 2021;Jiao et al. 2019;Wang et al. 2019;Wu et al. 2016).
Typically, sample preparation for lignans in S. chinensis fruit uses ethanol, methanol, or ethyl acetate, which is time-consuming and requires large volumes.Supramolecular solvents (SUPRAS) are nanostructured liquids formed by self-assembly of amphiphiles in colloidal suspensions.These compounds are valuable in sample preparation since they may provide hydrophobic, hydrogen-bonding, electrostatic, and p-cation interactions with analytes (Torres-Valenzuela, Ballesteros-G omez, and Rubio 2020).Hexafluoroisopropanol (HFIP), an amphiphilic perfluoro alcohol, has a high density and is a strong hydrogen bond donor with high hydrophobicity.HFIP has been demonstrated to induce amphiphiles to coacervate in aqueous media for the formation of SUPRAS (Khaledi, Jenkins, and Liang 2013).Recently, SUPRAS have been used for multiple substances efficiently in beverages, urine, serum, canned food, sweat, hair, and fingernails for LC/MS analysis (Accioni et al. 2018).
Here, HFIP-SUPRAS-based sample preparation is described for the determination of lignans in S. chinensis fruit by HPLC-QQQ-MS/MS.The sample treatment conditions were optimized to achieve the highest extraction efficiency.A rapid and sensitive HPLC-QQQ-MS/MS strategy for the quantification of seven lignans in S. chinensis fruit was validated for the analysis of S. chinensis whole fruit, pulp, and seeds

Plant material
The S. chinensis were collected from Wuchang City, Heilongjiang Province on September 2020, and identified as Schisandra chinensis (Turcz.)Baill (Storage voucher No.: BJFC00082657) by Professor Xian Yunmu of College of Conservation Area, Beijing Forestry University.All S. chinensis fruits were dried at room temperature.S. chinensis pulp and seeds were separated from the fruit.The whole fruit, pulp, and seeds were stored at 4 C until analysis.

Preparation of standard solutions
Standards for calibration were prepared in methanol, passed through a 0.22 mm membrane, and stored at À20 C until analysis.

Supramolecular solvent preparation
n-Octanol (5% v/v) was dissolved in HFIP (5%-35%, v/v), water was added to the coacervation-inducing agent, vortex mixed for 30 s, and divided into upper hydrophilic and lower hydrophobic phases by centrifugation at 5000 rpm for 5 min.

Sample preparation
The extraction of lignans in S. chinensis was conducted by SUPRAS.S. chinensis powder (20 mg) was placed on a 2 mL microfuge tube, 1 mL of 5% (v/v) octanol, 35% (v/v) HFIP, and 60% (v/v) water were added, vortex mixed for 30 s, and centrifuged at 5000 rpm for 5 min to separate the S. chinensis residues from the solution.All samples were passed through syringe filters with a 0.22 mm membrane before HPLC-QQQ-MS/MS analysis.
The proportion of HFIP in the SUPRAS, solid-to-liquid ratio, and vortex time (15-120 s) were optimized for the maximum yield of the lignans.

Apparatus and conditions
HPLC-QQQ-MS/MS was performed on an Agilent 1290 Infinity system equipped with a vacuum degasser, autosampler, thermostat, and binary pump (Agilent Technologies, USA).The Agilent 6460 triple quadrupole mass spectrometer included an electrospray ionization source (ESI).Analytes were separated by an Agilent Eclipse Plus C 18 (Rapid Resolution High Definition) column (2.1 mm Â 50 mm, 1.8 mm) at 30 C. The isocratic mobile phase was 30% 0.1% aqueous formic acid (A) and 70% acetonitrile (B) at 0.4 mL/min for 4 min.The injection volume was 2 mL.
Mass spectra were acquired by a positive ESI source using a capillary voltage of 3500 V (ESIþ), nebulizer pressure of 50 psi, gas temperature of 300 C, and gas flow of 10 L/min.The fragmentation voltage and collision energy of each analyte were further optimized to obtain the strongest quantitative transition as shown in Table 1.Agilent Mass Hunter workstation software (version b.07.00) was used to determine the retention times and peak areas.

Validation
The protocol was validated for linearity range, correlation coefficients (R 2 ), the lowest limit of quantification (LLOQ), repeatability, precision, stability, and recovery according to ICH Q2 Validation of Analytical Procedures: Text and Methodology.The calibration curves were constructed by standard solutions with concentrations from 0.01 to 20.83 mg/mL.The sensitivity was evaluated by the LLOQ as 10 times the residual standard deviation of the calibration curve divided by its slope (Wang et al. 2019).
Repeatability expresses the precision across a short time interval by continuous injection of the same standard solution six times and calculation of the relative standard deviation (RSD).The intra-and interday precision experiments were conducted by detecting the sample every 8 h and every 24 h, respectively.The precision was evaluated by evaluating the RSD of the retention time and peak area.
The stability was evaluated by monitoring a standard solution every 6 h within 48 h and evaluation of the RSD.The recovery experiments were conducted with 10 mg spikes of S. chinensis in triplicate to evaluate the accuracy.

Processing of S. chinensis
The processed S. chinensis products were prepared according to the Chinese Pharmacopoeia 2020 as follows.
Fried S. chinensis (FS).200 g of dry S. chinensis were fried until they turned black.
Vinegar-processed S. chinensis (VS).200 g of dry S. chinensis were soaked with vinegar (40 g) for 2 h and fried until they turned black.
Salt-processed S. chinensis (SS).200 g of dry S. chinensis were soaked with 4 g of water and fried until they turned black.Wine-processed S. chinensis (WS).200 g of dry S. chinensis were treated with yellow wine (40 g) for 2 h and fried until they turned black.
Honey-processed S. chinensis (HS).200 g of dry S. chinensis were soaked in 40 g honey in boiling water and fried until they turned black.
The processed S. chinensis was extracted by SUPRAS before HPLC-QQQ-MS/MS (MRM) analysis.

Optimization of the SUPRAS extraction
The extraction was optimized for SUPRAS composition, vortex time, and solid-liquid ratio.First, the influence of SUPRAS composition upon the yield of lignans from S. chinensis fruit was investigated.In particular, the ratio of HFIP is important for the preparation of HFIP-octanol-SUPRAS (Li et al. 2020).Figure 1A shows the yields of the lignans were maximum at 35% HFIP because the SUPRAS provides high solubilization of lignans.Li et al. (2020) reported that the structure of supramolecular aggregates in the HFIP-octanol SUPRAS phase was reversed micellar whose size increases significantly following the addition of HFIP.
Next, the vortex time was optimized.Figure 1B shows the maximum yield of lignans achieved at 30 s (Caballero-Casero, Beza, and Rubio 2022).SUPRAS are widely used for sample treatment because of the short extraction time.Cai et al. (2022) reported the highest extraction efficiency of hydroxy-a-sanshool in Zanthoxylum bungeanum Maxim peels was achieved at 60 s by vortex-assisted octanoic acid-ethanol-SUPRAS extraction.The analytes from urine were efficiently extracted by vortex-assisted HFIP-octanoic acid-SUPRAS extraction in 10 s (Zong et al. 2018).For parabens, HFIP/Brij-35 SUPRAS-based extraction provided the most rapid analysis (20 s; Chen et al. 2019).
The solid-liquid ratio was also optimized as shown in Figure 1C.The extraction yields of lignans increased with the sample volume up to 20:1 mg/mL.At higher ratios, the yields of the lignans decreased with the S. chinensis mass due to the reduction in transfer efficiency.
The yields of lignans isolated by HFIP-octanol-SUPRAS using the optimal conditions were compared with the isolation by organic solvents in Figure 1D.The highest yields of the analytes were obtained by SUPRAS and this approach was employed for the S. chinensis samples.

Optimization of mass spectrometry conditions
The MRM parameters, such as the gas flow, nebulizer pressure, ionization mode, transitions, fragmentor voltage, and collision energy, were optimized for the simultaneous quantitation of the analytes.All analytes were analyzed in both positive and negative ionization modes with an ESI full scan; better results were obtained using.positive ionization.After optimizing the atomizer pressure (50 psi) and gas flow (10 L/min), the protonated analyte molecular ions were employed as the precursors for the MRM transitions.Subsequently, the fragmentor voltage and collision energy of each component were optimized using the instrument software automatically to obtain the maximum response as shown in Table 1.
The product ion mass spectra were analyzed in detail under the optimal conditions.For example, the precursor ion of schizandrol A ( at m/z 409.00.Thus, the MRM transition of m/z 455.20!409.00 was selected to be the specific transition.The precursor ion of schizantherin A ([M þ Na] þ , m/z 559.00) produced the same fragmentation pattern as schizantherin B at m/z 414.9.Hence, a different product ion was selected as the MRM transition of schizantherin A (m/z 559.00!436.90,ESIÀ).
After the optimization, representative MRM chromatograms are shown in Figure 2. The optimized conditions were used to determine the analytes in S. chinensis fruit by HPLC-QQQ-MS/MS.

Validation
The validation of HPLC-QQQ-MS/MS was performed in terms of linearity, repeatability, sensitivity, precision, stability, and recovery as shown in Tables 2 and 3.The calibration of each analyte was plotted by the peak area ratio (y) to the concentration (x) using unweighted least-square linear regression.Excellent linearity was observed for all analytes (R 2 > 0.99) across a relatively wide range of concentrations (0.01-20.83 mg/mL).Table 2 shows the reported procedure exhibited better sensitivity with LLOQs from 0.14 to 17.79 ng/mL which were superior to the literature for the determination of lignans in S. chinensis fruit (Liu et al. 2013;Onay, Hofer, and Ganzera 2020).
The results for the repeatability of the developed method show that the RSDs were from 0.99% to 5.47% (Table 3).The stability of this method was analyzed through triplicate injections of mixed standards every 6 h within 48 h.The stability RSD for the lignans within 48 h was less than 7%, suggesting no analyte degradation during the analysis (Table 3).The intra-and interday RSDs for all analytes by retention time were less than 0.21% and 0.32%, respectively.The intra-and interday RSDs for peak areas were less than 6.41% and 7.39%, respectively.
These results demonstrate that the developed HPLC-QQQ-MS/MS protocol has suitable precision.The average recoveries of the lignans were between 80.21% and 114.53%, within the required range.Hence, the established procedure is suitable for the quantification of the lignans in S. chinensis fruit on the bases of linearity, sensitivity, repeatability, precision, stability, and recovery.

Comparison with the literature
A comparison of the developed method with the literature is shown in Table S1.The SUPRAS-based sample preparation achieved complete extraction in 30 s which is faster than the previously published methods.HPLC-QQQ-MS/MS required only 3.8 min to determine the lignans, while the literature procedures were considerably longer.The LLOQs were comparable to the literature, especially for supercritical fluid chromatography (SFC) , LC-MS, and HPLC-DAD-MS (Deng et al. 2008;Liu et al. 2013;Onay, Hofer, and Ganzera 2020).Hence, the reported approach provides analytical capabilities for the determination of lignans in S. chinensis fruit.Lignans in S. chinensis tissues HPLC-QQQ-MS/MS was employed to determine the lignans in S. chinensis fruit, pulp, and seeds (Figure 3).The lignans were predominantly in the seeds while the contents in the pulp were low.According to the literature, polysaccharides and tannins are the main components in the pulp (Li et al. 2021).Wang et al. (2015) characterized the lignans in the seed kernel, seed coat, and pulp of S. chinensis fruit.The lignans were the most abundant in the seed kernel, followed by the seed coat and the pulp.Similarly, Wei et al. (2011) show the lignan concentrations in S. chinensis seeds were higher than in the fruit.In addition, the quantities of lignans in S. chinensis oil from the seeds were much higher than in the fruit (Gao et al. 2019).Our results confirm the lignans are concentrated in the seeds.
Schizandrol A, schizantherin B, and schisandrol B (10.51 mg/g) were the most abundant lignans in the fruit.These compounds provide tumor and anti-oxidative activities, anti-inflammatory properties, and liver protection (Szopa, Ekiert, and Ekiert 2017).Moreover, they protect effect against neurodegenerative diseases (Kandhasamy et al. 2018).However, the literature has reported that other lignans were most abundant in S. chinensis fruits.For example, Gao et al. (2019) reported that the schisandrin A was the major lignan, followed by deoxyhawthorne.Huang et al. (2007) observed that schisandrin A and schisandrin B had the highest contents.Deng et al. (2008) showed that Schisandrin and gomisin N were present at the highest concentrations.Liu et al. (2013) reported that schisandrol A and schisandrin B were the most abundant.These variations are due to the cultivation conditions and collection time of S. chinensis.

Comparison of raw and processed S. chinensis fruit
The Chinese Pharmacopoeia 2020 reports that the traditional processing for S. chinensis fruit includes stir-frying, vinegar-processing, salt-processing, wine-processing, and honey-processing (Figure 4).Table 4 shows the lignan concentrations in processed S. chinensis were lower than in raw S. chinensis (RS) except for the wine-processed samples as reported by Pang et al. (2011).The high lignan content was observed following wine processing because this approach promotes the dissolution of lignans.The honeyprocess fruit had the lowest lignin content because honey is not conducive to lignan dissolution.These results provide a further understanding of the composition of unprocessed and processed S. chinensis fruit and provided a reference for quality control.

Conclusions
HFIP-octanol-SUPRAS was used for the determination of lignans in S. chinensis fruit with higher efficiency compared to traditional approaches.A sensitive, reliable, and rapid method simultaneously quantified seven lignans.The mass spectrometry conditions were systematically optimized and the linearity, sensitivity, precision, stability, and recovery were validated.
The lignans were determined in the tissues of the S. chinensis fruit.The lignans were concentrated in the seeds.Schizandrol A, schisandrol B, and schisantherin B were present at the highest concentrations.The lignan content was affected by the processing technique.The wine-processed S. chinensis values were higher than for untreated S. chinensis.
A rapid and reliable approach is reported for the determination of lignans in S. chinensis fruit suitable for quality control that may allow further additional applications in food and medicine.The processing methods are described in the Materials and Methods section.

Figure 1 .
Figure 1.Optimization of the conditions for the extraction of the lignans: (A) HFIP composition in the SUPRAS, (B) solid-to-liquid ratio, (C) vortex time, and (D) extraction solvent.

Figure 2 .
Figure 2. Mass spectra and MRM chromatograms of the lignans using the optimal conditions.

Figure 4 .
Figure 4. Morphology of processed S. chinensis.The processing methods are described in the Materials and Methods section.

Table 1 .
Optimized MRM parameters for the lignans.

Table 2 .
Analytical figures of merit for multiple reaction monitoring mass spectrometry.

Table 3 .
Validation of multiple reaction monitoring mass spectrometry.
b Peak area.