2.1 Reagents and materials

Seven batches of SXXTN were generously provided by Shandong Hongjitang Pharmaceutical Group Co., ltd. (Shandong, China) and listed in Table S1. A total of 101 reference standards and 3 internal standards were presented in Table S2. All standards were≥98 % by HPLC and 1H NMR analyses.

Ultrapure water (18.2 MΩ cm) for analysis was prepared by a Milli-Q water purification system (Millipore, Bedford, MA, USA). Methanol and acetonitrile (HPLC grade) were provided by Merck (Darmstadt, Germany), and formic acid (HPLC grade) were purchased from ROE (Newark, New Castle, DE, USA). Ethyl alcohol (HPLC grade) was offered by Sichuan Ruijingte Technology Co., ltd. (Sichuan, China).

2.2 Standard solution and samples preparations

The reference standards were solubilized by 75 % methanol-aqueous solution (v/v) to obtain 1.00 mg/mL reserve solution and diluted with appropriate solvent to a range of proper concentrations.

In qualitative analysis, for LC-MS, the SXXTN was ground into powder. SXXTN powder (0.3003 g) was accurately weighed and ultrasonic extracted (40 kHz, 500 W) with 5 mL 75 % methanol-aqueous solution (v/v) for 30 min. The extracts were centrifuged (13,000 rpm, 10 min, 4℃) before LC-MS analysis. For GC–MS, the powder (0.3000 g) was accurately weighed, then sonicated for 30 min at 40 kHz with 5 mL of ethanol. The filtrate was filtered by 0.45 μm filter membrane and centrifuged before sampling.

For quantitative analysis, to determine the non-volatile constituents, each batch of SXXTN powder (0.3 g) was weighed in three parallel times, then ultrasonic extracted (40 kHz, 500 W) with 5 mL 75 % methanol for 30 min. The filtrate was filtered by 0.45 μm filter membrane and centrifuged (13000 rpm, 10 min, 4℃). For alkaloids quantification (group A, 24 alkaloids), the supernatant was diluted after adding proper nitidine chloride (IS1, 1.11 μg/mL) as internal standard. For ginsenosides and acids quantification (group B, 14 ginsenosides, cinnamic acid and phenylalanine), the supernatant was diluted after proper saikosaponin C (IS2, 0.985 μg/mL) adding. To determine the volatile constituents, about 0.3 g the powder of SXXTN was extracted with 10 mL ethanol under ultrasonic conditions in ice-water bath for 20 min. The extraction was filtered through syringe filter (0.45 μm) and centrifuged. Isoborneol, borneol, 3-phenylpropyl cinnamate and cinnamyl cinnamate possessed significantly higher abundances in SXXTN comparing with other volatile components, indicating the large differences of contents among various compounds. Therefore, the supernatant was diluted 10 times before injecting to GC–MS for quantitative analysis of high-abundant volatile compounds (Group D), whereas directly injected for others with relatively low-abundances (Group C). And a certain amount of naphthalene (IS3, 23.2 μg/mL) was added to the supernatants as internal standard.

2.3 HPLC-QTOF MS analysis conditions

Agilent 1290 HPLC system (Agilent corporation, USA) was used to determine the non-volatile components of SXXTN. A ZORBAX Eclipse Plus C18 column (150 × 2.1 mm, 1.8 μm, Agilent Technologies, Santa Clara, USA) was used for sample separation. The mobile phase consisted of 0.1 % (v/v) formic acid in water (A) and acetonitrile (B) with the gradient elution set as follows: 0–3 min, 10 %-12 % B; 3–8 min, 12 %-17 % B; 8–20 min, 17 %–22 % B; 20–30 min, 22 %-35 % B; 30–45 min, 35 %-42 % B; 45–50 min, 42 %-60 % B; 50–57 min, 60 % B; 57–60 min, 60 %-80 % B; 60–68 min, 80 %-100 %. The flow rate was set at 0.4 mL /min, and the column temperature was maintained at 30 ℃. Sample volume was 1 μL for injection.

The Q-TOF mass spectrometer equipped with electrospray ionization (ESI) source was used to acquire data in positive and negative ion modes. The operation conditions were as below: drying gas (N₂) temperature, 300 ℃; drying gas flow, 8.0 L/min; nebulizer gas (N₂) pressure, 35 psig; sheath gas (N₂) temperature, 350 ℃; sheath gas flow, 11.0 L/min; capillary voltage positive ion mode, 4000 V; negative ion mode, 3500 V; fragmentor voltage, 120 V; skimmer voltage, 65 V. Full-scan MS and MS/MS data was collected over the m/z range of 50–1500 using extended dynamic range. Collision energy of secondary mass spectrometry was set as 15 eV, 30 eV and 45 eV.

2.4 GC–MS analysis conditions

Compound identification was performed by Agilent 7890B GC system combined with Agilent 5977 Mass Selection Detector. Samples were separated by Agilent HP-5MS (30 m × 0.25 mm, 0.25 μm) column. The carrier gas was high purity helium, and the flow rate was 1 mL/min. Initial column temperature was 60 ℃, and programmed to rise at 20 °C/min to 85 °C (1 min held), 5 ℃/min to 100 ℃ (5 min held), 15 ℃/min to 150 ℃ (6 min held), 5 ℃/min to 200 ℃ (4 min held), rising at 5 ℃/min to 280 ℃ (5 min held). The injection volume was 1 μL and the splitting ratio was 30:1. The temperature of injector and aux heaters was controlled at 250℃ and 280℃, respectively. MS quadrupole and ion source temperature were maintained at 150℃ and 230℃, severally. MS data were recorded at 70 eV and acquired in full scan mode over the range of m/z 40–600.

2.5 HPLC-QQQ MS analysis conditions

The quantitative analysis was performed on the Shimadzu LCMS-8050 triple quadrupole tandem mass spectrometry detector (Shimadzu, Kyoto, Japan) with an Agilent Zorbax Eclipse Plus C18 column (2.1 × 150 mm, 1.8 µm, Agilent Technologies, Santa Clara, USA). For group A, 0.1 % (v/v) formic acid water (A) and acetonitrile (B) were used as mobile phases, and the gradient elution procedure was as follows: 0–12 min, 19 %-20 % B; 12–14 min, 20 %-35 % B; 14–16 min, 35 %-90 % B; 16–19 min, 90 %-100 % B. For group B, the mobile phase was water (A) and acetonitrile (B), with the following gradient elution: 0–3 min, 10 %-12 % B; 3–6 min, 12 %-35 % B; 6–14 min, 35 %-36.5 % B 14–15 min, 36.5 %-90 % B; 15–19 min, 90 %-100 % B. The flow rate was maintained at 0.4 mL/min, with the injection volume 2 μL for all samples. The MS conditions were as below: capillary voltage, 4000 V; drying gas temperature, 300 °C. The flow rate of drying gas (N₂) and nebulizer gas (N₂) was 10.0 L/min and 3.0 L/min, severally. Analytes were determined in MRM modes, and the optimized parameters were shown in Table S3 and S4.

2.6 GC-QQQ MS analysis conditions

The quantitative analysis of volatile components was operated on an Agilent 7890B gas chromatography coupling to Agilent 5977A mass spectrometry (Agilent, Santa Clara, CA, USA). For group C, the initial column temperature was 60 ℃, and programmed to rise at 20 °C/min to 85 °C, 5 ℃/min to 100 ℃ (5 min held), 15 ℃/min to 150 ℃, 5 ℃/min to 180 ℃ (1 min held), finally rising at 15 ℃/min to 280 ℃ (2 min held). For group D, the initial column temperature was set at 100 ℃, and programmed to rise at 10 °C/min to 110 °C, 3 ℃/min to 120 ℃, 55 ℃/min to 265 ℃, finally rising at 18 ℃/min to 280 ℃ (2 min held). The injection volume was 1 μL and the splitting ratio was 10:1. The MRM parameters for all analytes are presented in Table S5 and S6. Other analytical conditions refer to Section 2.4.

3.1 Qualitative analysis of SXXTN based on diagnostic ion strategy by HPLC-QTOF MS

The HPLC-QTOF MS conditions of the mobile phase systems (methanol-aqueous, acetonitrile-aqueous, and acetonitrile-aqueous with 0.1 % formic acid), gradient program, column temperature (25 °C, 30 °C, and 35 °C) and the flow rate (0.2, 0.3 and 0.4 mL/min) were optimized in order to obtain overall constituents of SXXTN with good resolution within a short analysis. The total peak area was adopted as a criterion for optimization. Ultimately, the optimum conditions mentioned in Section 2.3 were preferred.

Diagnostic ion strategy is regarded as a powerful approach for rapid characterization of chemicals in TCMs based on the principle that similar chemical constituents have similar cleavage rules and the fragmentation information, which is applicable for the identification of structural analogues in complex TCMs and formulae (Wang et al., 2017). In our study, by comparing with the reference standards, the known compounds were marked. On the basis of MS/MS analysis of authentic compounds, the characteristic fragmentation pathways of compounds with the same carbon skeleton were presented, the obtained rules were further applied to the structural characterization of its derivatives. For other unknown compounds, identification based on MS/MS spectra and relevant literature or online databases, including PubChem search ( https://pubchem.ncbi.nlm.nih.gov/) and the Human metabolome database ( https://www.hmdb.ca/). The typical total ion chromatograms (TICs) of SXXTN by HPLC-QTOF MS in both of positive and negative ion modes were displayed in Fig. 1. Totally, 140 compounds were identified based on diagnostic ion strategy, including 60 alkaloids, 34 ginsenosides, 21 organic acids, 12 phthalides, 10 triterpenes, 2 esters and 1 amino acid. The chemical structures and detailed information of compounds could be viewed in Fig. 2 and Table 1, respectively. The MS/MS spectra and fragmentation pathways of the representative chemicals were shown in Fig S1 and Fig S2.

Close

Fig. 1 Total ion current chromatograms of SXXTN in positive ion mode (A) and negative ion mode (B) by HPLC-QTOF MS.

Export to PPT

Close

Fig. 2 Structures of chemical constituents from SXXTN.

Export to PPT

Close

Fig. 2 Structures of chemical constituents from SXXTN.

Export to PPT

Close

Fig. 2 Structures of chemical constituents from SXXTN.

Export to PPT

3.2 GC–MS qualitative analysis of SXXTN

In preceding reports, the volatile components in SXXTN such as Artificial Musk, Chuanxiong Rhizoma, Styrax and Borneolum Syntheticum have been revealed (Ding et al., 2022; He et al., 2018; Gurbuz et al., 2013; Sun et al., 2014), whereas little attention was paid to the volatile components in the intact SXXTN prescription. In this work, we supplemented the information of volatile chemicals and improved the global characterizations of complicated ingredients in SXXTN.

The GC–MS conditions of the temperature program, splitting ratio (10:1, 30:1 and 50:1) and the injector temperature (250℃, 280℃ and 300℃) were optimized in the direction of analyzing comprehensive volatile constituents of SXXTN with well separation performance in a short analysis. The total peak area was calculated as a criterion for optimization. The final conditions were described in Section 2.4.

The TICs of SXXTN by GC–MS can be viewed in Fig. 3. There were 44 volatile compounds tentatively identified from SXXTN, based on the mass spectrometric data of reference standards, the mass spectral library (NIST17) and the literature, including 12 organic acid esters, 6 monoterpenes, 6 phthalides, 3 organic acids, 3 sesquiterpenes, 3 alcohols, 3 alkanes, 3 hydrocarbons, 2 steroids, 1 macrocyclic ketone, 1 alkaloid and 1 anhydride (Table 2). The chemical structures were illustrated in Fig. 2. Among them, seven compounds (66, 68, 110, 108, 111, 138, 46) have been identified in the previous LC-MS analysis. According to the comparison with authentic standards, 25 components were clearly marked. Isoborneol (1 4 6), borneol (1 4 8), cinnamyl alcohol (1 5 1), (E)-ligustilide (1 1 1), muscone (1 6 3), 3-phenylpropyl cinnamate (1 7 3) and cinnamyl cinnamate (1 7 4) exhibited relatively high abundances during GC–MS analysis of SXXTN.

Close

Fig. 3 Total ion current chromatograms of SXXTN by GC–MS.

Export to PPT

3.3 Quantification of 40 non-volatile compounds in SXXTN by HPLC-QQQ MS

Forty confirmed non-volatile chemicals were further quantified by the optimized HPLC-QQQ MS method (Table S3-S4) to evaluate the quality of SXXTN. According to the difference in the response of the components in positive and negative ion modes, two MRM methods with different polarity were established for Quantification. The typical MRM chromatograms of analytes were illustrated in Fig S3.

Nice linearity with coefficients of determination (R² > 0.9900) were obtained for the 40 analytes. Limit of detection (LOD) and limit of quantitation (LOQ) tests were performed and listed in Table S7. As exhibited in Table S8, relative standard deviations (RSD) of repeatability, intra- and inter-day precision were 1.27 % − 4.79 %, 1.07 % − 5.41 % and 1.18 % − 9.43 %, respectively. Besides, all analytes could remain stable within 24 h under 4℃, with the RSD ranging 0.60 % − 6.14 % and recoveries of 40 compounds were ranged from 80.36 % − 117.13 % with the RSD ranging 2.38 % − 12.61 %. Consequently, the established HPLC-QQQ MS approach was proved as a sensitive, repeatable and accurate tool for the quantification of non-volatile compounds in SXXTN.

According to the established HPLC-QQQ MS quantitative analysis method, the content of 40 compounds in 7 batches of SXXTN provided by the enterprise was determined as shown in Table 3 and Fig. 4. The total content of 40 analytes in each batch was 2.03 % − 2.30 %. Among them, the components with higher content (>1.00 mg/g) were norglaucine (43), glaucine (46), dehydrocorydaline (75) and ginsenoside Rb₁ (86). In the previous reported, aforementioned compounds presented promising effects for myocardial protection (Wen et al., 2022; Han et al., 2012; Zheng et al., 2017; Kong et al., 2018). For example, studies have found that intraperitoneal injection of dehydrocorydine in ApoE^-/- mice can not only inhibit the development of atherosclerosis, but also improve aortic compliance and plaque stability (Wen et al., 2022). Tetrahydropalmatine can activate PI3K/Akt/eNOS/NO pathway, increase the expression of HIF-1a and VEGF, and inhibit iNOS-derived NO production in myocardium. This effect may reduce the accumulation of inflammatory factors (including TNF-a and MPO) and reduce the degree of apoptosis (Han et al., 2012). It has also been reported that ginsenoside Rb₁ can improve heart failure, which may be achieved by regulating the mitochondrial membrane in cardiomyocytes (Kong et al., 2018).

Close

Fig. 4 The content distribution heatmap of 40 non-volatile compounds in 7 batches of SXXTN.

Export to PPT

3.4 Quantification of 17 volatile compounds in SXXTN by GC-QQQ MS

We established rapid and accurate quantitative methods for detecting the contents of the major volatile compounds in SXXTN. Seventeen confirmed compounds including 5 phthalides, 3 organic acid esters, 3 alcohols, 3 monoterpenes, 1 sesquiterpene, 1 phenol and 1 macrocyclic ketone were determined by GC-QQQ MS with naphthalene (IS3) as internal standards. The optimized conditions were shown in Table S5-S6 and typical MRM chromatograms of 17 analytes were illustrated in Fig S4.

The optimized GC-QQQ MS method was validated in the aspect of linearity, LODs, LOQs, precision, repeatability, stability and recovery and the results were presented in Table S9 and Table S10. Reasonable correlation coefficient values (R2 > 0.9904) indicated good correlations between investigated standards concentrations and their peak areas within the ranges tested. The ranges of LODs and LOQs for all the analytes were 0.002 - 2.642 μg/mL, and 0.013 – 6.653 μg/mL, respectively. The RSDs of repeatability, intra- and inter-day precision were 1.63% - 9.66%, 0.44% – 9.06%, 0.80% – 8.06%, respectively. All analytes could remain stable within 24 h under 4℃, with the RSD ranging 1.14% - 7.97%. The developed method had good accuracy with the recoveries were between 90.20% and 123.51%. These results provided that the established method was accurate, reproducible, and reliable for assessing the quality of volatile compounds in SXXTN.

According to the established GC-QQQ MS quantitative analysis method, the contents of main volatile components in 7 batches of SXXTN provided by the enterprise were determined as displayed in Table 3 and Fig. 5. The total content of analytes in each batch was 3.34 %-4.26 % in SXXTN. Borneol (1 4 6), isoborneol (1 4 8), cinnamyl cinnamate (1 7 4), 3-phenylpropyl cinnamate (1 7 3) and muscone (1 6 3) were the predominant components and were closely related to the anti-coronary heart disease and angina pectoris effect of SXXTN (Liu et al., 2017; Wu et al., 2011; Wang et al., 2020).

Close

Fig. 5 The content distribution heatmap of 17 volatile compounds in 7 batches of SXXTN.

Export to PPT

Combined with the quantitative analysis of non-volatile components, the total contents of 57 main components in 7 batches of SXXTN were 5.50 % − 6.49 %. The percentages of different structural types of chemicals in the total 57 analytes were as follows: monoterpenes and sesquiterpenes accounted for the largest proportion (42 %), followed by alkaloids (26 %), organic acids and esters (18 %), ginsenosides (6 %) and phthalide (5 %). Both volatile and non-volatile components should be taken into consideration for quality evaluation of SXXTN. More batches of samples are more conducive to assessing the consistency and stability of SXXTN.