A method for analyzing ginsenosides and identifying isomers thereof

Abstract

The application discloses a ginsenoside analysis and isomer identification method. The method adopts cyclic ion mobility mass spectrometry (cIMS-MS) to separate ginsenosides by multiple ion mobility separations, and realizes high-efficiency separation of difficult-to-separate isomers such as R / S configuration isomers by a "top-and-tail" IMS n strategy solves the wrap-around effect, and realizes high-efficiency separation of difficult-to-separate isomers such as R / S configuration isomers. By optimizing ion trap zone collision energy (110 eV) to generate aglycone fragment ions, a homologous aglycone characteristic collision cross section (CCS) library containing different configurations (R / S) and different aglycone types (PPD / PPT) is established, and the R / S configuration of complete ginsenosides is inferred by using aglycone fragment CCS values as molecular fingerprints. Meanwhile, five-dimensional information such as chromatographic retention time, accurate mass number, MS / MS fragments, complete molecular CCS values and aglycone fragment CCS values is integrated to construct a systematic identification workflow. The method solves the problem that R / S configuration isomers cannot be distinguished by mass spectrometry in the prior art, and has successfully identified 101 kinds of ginsenosides from ginseng extracts.

Description

Technical Field

[0001] This invention relates to the field of natural product analytical chemistry, specifically to a method for the analysis and isomer identification of ginsenosides. Background Technology

[0002] Ginsenosides are the main bioactive components of Panax spp., possessing a wide range of pharmacological activities, including antitumor, anti-inflammatory, antioxidant, neuroprotective, and cardiovascular protective effects. Dammarane-type ginsenosides (DTGs) constitute the vast majority of ginsenosides and can be classified into four main categories based on their aglycone structure: protopanaxadiol (PPD), protopanaxatriol (PPT), ocotillol (OT), and C-17 side-chain varied (C17SCV).

[0003] Ginsenosides exhibit high structural diversity, with various isomer types, including: (1) aglycone-type isomers; (2) glycosyl-type isomers (glucose, rhamnose, xylose, arabinose, etc.); (3) glycosidic bond linkage position isomers (1→2, 1→6, etc.); (4) glycosylation site isomers (C-3, C-6, C-20, C-28 positions); (5) double bond position isomers (Δ20(21) vs Δ20(22)); and (6) chiral center configuration isomers (R / S configuration, especially C-20 position). These isomers often co-elute on chromatography, and the mass spectrometric fragments are highly similar, posing a significant challenge to accurate identification.

[0004] Liquid chromatography-tandem mass spectrometry (LC-MS / MS) is currently the primary tool for ginsenoside analysis. In recent years, ion mobility spectrometry-mass spectrometry (IMS-MS) has significantly improved isomer identification capabilities by introducing the collision cross section (CCS) as a third dimension of information. Waters' cyclic ion mobility spectrometry-mass spectrometry (cIMS-MS) achieves resolutions far exceeding conventional linear ion mobility by extending the ion separation path and supports IMSn experiments—that is, the stepwise selection, fragmentation, and re-separation of specific ions.

[0005] However, conventional linear ion mobility spectrometry (such as SYNAPT G2-Si, Vion IM-QTOF, etc.) cannot achieve baseline separation for ginsenoside R / S configuration isomers with extremely similar CCS values ​​(such as 20(R)-Rg2 and 20(S)-Rg2, with a CCS difference of only about 3-4 Ų). For C-20 R / S configuration isomers and C-17 side-chain double bond position isomers (Δ20(21) vs Δ20(22)), the MS / MS spectra at different collision energies are almost identical, and cannot be distinguished by mass spectrometry information alone. Although differential ion mobility mass spectrometry (DMS-MS / MS) can achieve spatial separation of some isomers by gradually adjusting the concentration of gas modifier, its separation mechanism depends on the adjustment of compensation voltage, and its ability to distinguish R / S configuration isomers is still limited. Furthermore, while existing publicly available electron-induced fragmentation (EID) mass spectrometry methods can distinguish some ginsenoside isomers, they have specific instrument requirements and cannot universally solve the configuration determination problem of the C-20 chiral center. Existing multidimensional ginsenoside databases such as GinMIL, although containing CCS prediction values, only apply to intact molecules, and the predicted values ​​deviate significantly from experimental values. Specifically, GinMIL, based on the Vion ion mobility high-resolution LC-MS platform, uses machine learning (gradient booster model) to predict retention times and collision cross sections, covering the four-dimensional structural information (m / z, molecular formula, retention time, CCS, and some MS / MS fragments) of 579 ginsenosides, improving the isomer identification accuracy to approximately 88%. However, the core of this database lies in the CCS prediction of intact molecules, not in aglycone fragment ions; and there is a significant deviation between the predicted and experimentally measured values. For R / S configuration isomers and double bond position isomers, the predicted CCS values ​​cannot reliably distinguish them.

[0006] More importantly, current technologies lack a characteristic CCS library based on aglycone fragment ions, making it impossible to infer R / S configurations from CCS information. Furthermore, while multi-loop separation in cIMS can improve resolution, it suffers from a "wrap-around" effect (fast ions catching up with slow ions). This effect occurs because, with increasing loop counts, faster-mobility ions gradually catch up with slower ions from the previous loop, causing temporal overlap between ions from different loops when they reach the detector. This effect limits the effective number of loops in multi-loop separation and is a recognized technical challenge in cIMS applications. Currently, there is no effective solution to this problem in the separation of ginsenoside isomers. Summary of the Invention

[0007] The purpose of this invention is to provide a method for the analysis and isomer identification of ginsenosides, in order to solve the technical problems in the prior art, such as insufficient separation resolution of ginsenoside isomers, inability to distinguish R / S configurations by mass spectrometry, and lack of a systematic identification strategy.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] A method for analyzing ginsenosides and identifying their isomers includes the following steps:

[0010] Step 1: The ion mobility of the sample containing ginsenosides was separated by multiple cycles of cyclic ion mobility mass spectrometry (cIMS-MS).

[0011] Step 2: When a circumferential effect occurs during multi-loop separation where fast ions catch up with slow ions, a "top-and-tail" IMS is employed. n The strategy involves selectively spraying target ions followed by continuous cyclic separation to achieve complete separation of ginsenoside isomers.

[0012] Step 3: By optimizing the collision energy in the ion trap region, ginsenosides are made to generate aglycone fragment ions after the ion source, and the collision cross section (CCS) value of the aglycone fragment ions is measured.

[0013] Step 4: Compare the CCS values ​​of the aglycone fragment ions measured in Step 3 with the characteristic CCS library of homologous aglycones, which contains the characteristic CCS values ​​of dammarane-type ginsenoside aglycone fragment ions with different configurations (R-type and S-type), thereby determining the R / S configuration of ginsenosides.

[0014] Step 5: Integrate chromatographic retention time, precise mass number, tandem mass spectrometry (MS / MS) fragment ion information (including mass-to-charge ratio m / z), intact molecule CCS value, and aglycone fragment CCS value to perform multidimensional identification of ginsenosides.

[0015] Preferably, the number of cycles for multi-circle ion mobility separation in step one is 15 to 45, and the ion mobility separation parameters are: 2 pushes per chamber, 200 chambers, 375 m / s traveling wave speed, and 15V traveling wave static height.

[0016] Preferably, the collision energy of the ion trap region in step three is 80~120eV, more preferably 110eV, and the aglycone fragment ions include: protopanaxadiol (PPD) type aglycone ions with a mass-to-charge ratio (m / z) of 459.4; and protopanaxadiol (PPT) type aglycone ions with a m / z of 475.4.

[0017] The C-17 side-chain variant aglycone fragment ions exhibit differentiated fracture behavior depending on the type of C-17 side-chain modification. For C-17 side-chain hydroxyl rearrangement type ginsenosides, such as pseudoginsenoside Rg2(E), the aglycone ion m / z 475.4 further loses C3H6O during collision-induced dissociation, producing a fragment of m / z 417.3, and this type does not produce a fragment of m / z 391.3; for C-17 side-chain dehydration to olefin type ginsenosides, such as ginsenoside F4, the aglycone ion is m / z 457.4. The above two C17SCV subtypes can be distinguished by the presence or absence of a fragment of m / z 417.3 and the difference in the mass of the aglycone ion, and the positions of the double bonds Δ20(22) and Δ20(21) can be further determined by combining the CCS value of the intact molecule.

[0018] Preferably, in the homologous aglycone characteristic CCS library described in step four, the characteristic CCS value of the PPD-type aglycone fragment ion is: S configuration 224.0-225.0 Å. 2 R configuration 226.0-228.0 Å 2 The characteristic CCS values ​​of PPT-type aglycone fragment ions are: S configuration 225.0-227.0 Å. 2 R configuration 227.5-229.0 Å 2 .

[0019] Preferably, the MS / MS fragment ion information in step five is used to distinguish the following isomer types:

[0020] Aglycone type isomers are distinguished based on the m / z of the characteristic aglycone ion;

[0021] Sugar type isomers are distinguished based on the characteristic sugar chain fragment ion m / z of 205.07, 221.07, or 191.05;

[0022] Glycosidic bond linkage position isomers are distinguished based on the characteristic disaccharide fragment ion m / z 323.08 or 293.08;

[0023] Glycosylation site isomers are distinguished based on the relative abundance ratio of characteristic fragment ions. The identification of glycosylation site isomers is based on the systematic difference in the relative abundance ratio of characteristic fragment ions at the same collision energy. Taking protopanaxadiol-type ginsenosides as an example, ginsenoside CK, glycosylated at the C-3 position, and 20(S)-ginsenoside Rh2, glycosylated at the C-20 position, both produce an aglycone ion of m / z 459.4 and a deglycosylated fragment ion of m / z 621.4 at a collision energy of 40 eV. The abundance ratios of m / z 459.4 and m / z 621.4 are 1.6 and 0.51, respectively. 20(S)-ginsenoside Rh2 further produces m / z fragment ions at 60 eV. 537.3 characteristic fragments were observed, while ginsenoside CK lacked such fragments. Taking protopanaxadiol-type ginsenosides as an example, ginsenoside F1, glycosylated at C-20, showed a significantly higher abundance of fragments with broken glycosidic bonds at C-20 compared to ginsenoside Rh1, glycosylated at C-6. The glycosidic bond breaking sensitivity followed the order of C-20 being higher than C-6, and C-6 being higher than C-3, a difference that was identifiable at a collision energy of 40 eV. For the multi-site glycosylation mode, the disaccharide chain at C-3 of 20(S)-ginsenoside Rg3 remained intact at 40 eV, while the monosaccharide bond at C-20 of ginsenoside F2 was broken. The m / z values ​​for both were 621.4 and m / z. The abundance ratio of 783.5 showed significant differences and exhibited a differentiated evolution pattern with increasing collision energy. By measuring the relative abundance ratio of the above characteristic fragment ion pairs and the characteristic fragments at high collision energies, the isomers of glycosylation sites at C-3, C-6, and C-20 can be identified.

[0024] Double bond position isomers are distinguished based on the difference in CCS value of the complete molecule, with the CCS value of the Δ20(22) double bond configuration being greater than that of the Δ20(21) double bond configuration.

[0025] Preferably, the ginsenosides are dammarane-type ginsenosides, including PPD type, PPT type, OT type and C17SCV type.

[0026] Preferably, a chromatographic separation step is included before step one, using ultra-high performance liquid chromatography (UPLC) with an ACQUITY HSS T3 column (1.8 μm, 100 × 2.1 mm), a column temperature of 40 °C, a mobile phase of an aqueous solution containing 0.1% formic acid (phase A) and acetonitrile (phase B), a flow rate of 0.3 mL / min, and a gradient elution program as follows: 0–1.5 min 95% A, 1.5–3 min 95→75% A, 3–6 min 75→73% A, 6–9 min 73→67.5% A, 9–12 min 67.5→64% A, 12–18 min 64→53% A, 18–22 min 53→0% A, 22–24 min 0% A, 24–25 min 0→95% A, and 25–29 min equilibration with 95% A.

[0027] Preferably, the cIMS-MS operates in negative ion mode with a capillary voltage of 2.0 kV, a cone voltage of 40 V, a source temperature of 120 °C, a desolventizing temperature of 400 °C, a cone gas flow rate of 50 L / h, a desolventizing gas flow rate of 800 L / h, nitrogen as the collision gas and ion mobility gas, and a data acquisition range of m / z 50–2000.

[0028] Preferably, the method is used for the systematic identification of ginsenosides in extracts of ginseng plants, and can simultaneously identify PPD type, PPT type, OT type, C17SCV type and oleanolic acid type ginsenosides and their isomers.

[0029] This invention also provides a method for constructing a homologous aglycone characteristic CCS library, comprising the following steps:

[0030] Step 1: Select dammarane-type ginsenoside standards with known R / S configurations;

[0031] Step 2: Using cIMS-MS, aglycone fragment ions are generated under optimized collision energy conditions in the ion trap region;

[0032] Step 3: Determine the CCS values ​​of aglycone fragment ions generated by each standard, and establish a database of characteristic CCS values ​​of aglycone fragment ions with different configurations (R-type and S-type) and different aglycone types (PPD-type and PPT-type).

[0033] Step 4: Verify the stability of the characteristic CCS value: intra-day precision relative standard deviation (RSD) < 0.1%, inter-day precision RSD < 0.3%, cross-instrument relative deviation < 3.5%, and remain stable in the same type of aglycone fragments from different compound sources.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. This invention employs cIMS multi-loop separation technology to successfully achieve baseline separation of R / S configuration isomers (such as 20(R)-Rg2 and 20(S)-Rg2) that cannot be separated by conventional linear ion mobility; simultaneously, through "top-and-tail" IMS... n The strategy effectively solves the wrap-around effect in multi-loop separation and avoids peak overlap caused by fast ions catching up with slow ions. Compared with the existing technology where the GinMIL database can only achieve an isomer identification accuracy of about 88% and has limited ability to distinguish between R / S configuration isomers and double bond position isomers, this invention achieves baseline separation of the above-mentioned difficult-to-separate isomers by combining cIMS multi-loop separation with the "top-and-tail" strategy, which significantly improves the identification accuracy.

[0036] 2. This invention proposes a "homologous aglycone characteristic CCS library" strategy: by generating aglycone fragment ions under high-energy collision (110 eV) and measuring their CCS values; experiments showed that the S-configuration CCS of PPD-type aglycones is approximately 224.4 Å. 2 (Experimental average, range 224.0–225.0 Å) 2 The R configuration is approximately 226.2 Å. 2 (Experimental average, range 226.0–228.0 Å) 2 The S configuration CCS of the PPT-type aglycone is approximately 225.8 Å. 2 (Experimental average, range 225.0–227.0 Å) 2 The R configuration is approximately 228.2 Å. 2 (Experimental average, range 227.5–229.0 Å) 2 The CCS value remains stable (RSD < 0.5%) among the same type of aglycone from different sources, and can be used as a "molecular fingerprint" to infer the R / S configuration of intact ginsenosides, solving the long-standing technical problem that MS / MS cannot distinguish R / S configuration isomers. Experimental verification shows that this characteristic CCS value has good cross-compound stability: the CCS values ​​of PPD-type aglycone fragments from different ginsenoside standards (such as 20(S)-Rg3, F2, Rd, Rc, Rb2, Rb3, etc.) are all within 224.4 ± 0.3 Å. 2 Within the range (RSD < 0.5%); PPT-type aglycone fragments from different PPT-type standards (such as F1, 20(S)-Rh1, F5, 20(S)-Rg2, etc.) all had CCS values ​​of 225.8 ± 0.3 Å. 2 Within the range (RSD < 0.5%), this demonstrates the reliability and universality of the characteristic CCS value as a "molecular fingerprint";

[0037] 3. This invention integrates five dimensions of information, including LC retention time, precise mass number, MS / MS fragments, intact molecule CCS value, and aglycone fragment CCS value, to construct a complete strategy for identification from crude extract to monomers. This method has successfully identified 101 ginsenosides from ginseng extract, covering PPD, PPT, OT, C17SCV, and oleanolic acid type, significantly improving the accuracy and reliability of identification.

[0038] 4. The CCS determination in this invention has extremely high precision (intra-day RSD < 0.1%, inter-day RSD < 0.3%), and good comparability between different laboratories and different instruments (relative deviation < 3.5%). The matrix effect assessment shows that the influence of ginseng extraction matrix on CCS value is negligible, and the method is suitable for the analysis of complex biological samples. Attached Figure Description

[0039] Figure 1 To employ multi-turn cIMS-MS and "top-and-tail" IMS n Schematic diagram of the strategy for separating ginsenoside isomers;

[0040] Where A represents the multi-loop separation of the Rh1 isomer, and B represents the four R / S configuration isomers of Rg2 and Rg3 separated by the "top-and-tail" strategy;

[0041] Figure 2 The MS / MS spectra of ginsenosides Rg2(A), Rg3(B), pseudoginsenoside F11(C) and pseudoginsenoside Rg2(E)(D) under 60 eV CID conditions;

[0042] Figure 3 MS / MS spectra of ginsenosides Rh1(A), Rh2(B), CK(C) and F2(D) under 40 eV CID conditions;

[0043] Figure 4 Application of CCS in the identification of ginsenoside isomers;

[0044] Where A represents the intra-day and inter-day RSD of CCS determination, B represents the relative deviation of the CCS values ​​of ginsenosides in various ginseng matrices (sun-dried ginseng, red ginseng, black ginseng and ginseng wheat drink) compared with the control solution, C represents the relationship between the molecular weight of ginsenosides and the CCS value, D represents the comparison between the database predicted CCS and the experimentally determined CCS, and E represents the cross-validation of CCS values ​​between the two laboratories.

[0045] Figure 5 A flowchart of a comprehensive identification strategy for ginsenoside isomers based on MS / MS and CCS;

[0046] Figure 6Example of the identification process for a compound with a mass-to-charge ratio of 845.4913. Detailed Implementation

[0047] The present invention will be further described below with reference to specific embodiments.

[0048] Example 1: cIMS-MS analysis of ginsenoside standards

[0049] 1.1 Reagents and Materials

[0050] Twenty-five dammarane-type ginsenoside standards (see Table S1) were purchased from Chengdu Dest Biotechnology Co., Ltd. LC-grade acetonitrile, methanol (Merck, Germany), and formic acid (Aladdin, China) were used as mobile phase additives. Ultrapure water was prepared using a Milli-Q Integral 5 pure water system (Millipore, USA).

[0051] 1.2 Sample Preparation

[0052] Weigh 1 mg of each standard and dissolve it in 1 mL of methanol to prepare a single stock solution with a concentration of 1000 μg / mL. Mix appropriate amounts of each stock solution and dilute with methanol to a mixed working solution with a concentration of 10 μg / mL.

[0053] 1.3 Chromatographic conditions

[0054] A Waters Acquity Premier UPLC system equipped with an ACQUITY HSS T3 column (1.8 μm, 100 × 2.1 mm, Waters, UK) was used at a column temperature of 40 °C. Mobile phase A was an aqueous solution containing 0.1% (v / v) formic acid, and mobile phase B was acetonitrile, with a flow rate of 0.3 mL / min. The gradient elution program was as follows: 0–1.5 min 95% A; 1.5–3 min 95→75% A; 3–6 min 75→73% A; 6–9 min 73→67.5% A; 9–12 min 67.5→64% A; 12–18 min 64→53% A; 18–22 min 53→0% A; 22–24 min 0% A; 24–25 min 0→95% A; 25–29 min equilibration at 95% A. The injection volume was 4 μL.

[0055] 1.4 cIMS-MS conditions

[0056] A SELECT SERIES Cyclic IMS instrument (Waters Corporation, UK) was used. Negative ion mode: capillary voltage 2.0 kV; cone voltage 40 V; source compensation voltage 10 V; source temperature 120 °C; desolvation temperature 400 °C; cone gas flow rate 50 L / h; desolvation gas flow rate 800 L / h. Data acquisition range m / z 50–2000. Nitrogen was used as both the collision gas and the ion mobility gas.

[0057] Multiple acquisition modes were employed, including MSᴱ, HDMSᴱ, MS / MS, and HDMS / MS. To obtain the aglycone CCS value, the HDMSᴱ mode was used, with the collision energy of the ion trap region in the low-energy channel set to 110 eV.

[0058] TOF mass calibration uses sodium formate. Calibration was performed using the Major Mix IMS / Tof calibration kit (Waters), with single-charge calibrators covering a mass range of 150.0555–1965.9364 Da and a CCS range of 131.5–367.2 Å. 2 When %RMS A value below 2% is considered calibration qualified. Ten of the 26 calibrators are selected for calibration under multi-turn separation conditions. Calibration under multi-turn separation conditions employs a different calibration strategy than single-channel calibration: because multi-turn separation prolongs the ion path, the drift time of the calibrators increases accordingly. Therefore, this invention selects 10 calibrators from the 26 calibrators to cover the CCS range (150~370 Å) of the target compound. 2 The calibrator was calibrated using multiple passes. The relative deviation of the CCS values ​​measured under single-channel and multi-pass conditions was 0.02% to 1.27% (see Example 1.6), verifying the reliability of the multi-pass CCS values.

[0059] 1.5 Multi-loop separation and "top-and-tail" strategy

[0060] Multi-loop separation and "top-and-tail" IMS n The strategy is implemented manually using the Cyclic Sequence Control module in the software. In Cyclic Control mode, the number of pushes per bin is set to 2, and the number of bins is set to 200. Traveling wave parameters: both the cyclic traveling wave speed and the array traveling wave speed are 375 m / s, and the static height of the traveling wave is 15V. The number of cycles is changed by adjusting the separation time.

[0061] 1.6 Results

[0062] (1) Multi-ring separation of R / S configuration isomers

[0063] Figure 1To employ multi-turn cIMS-MS and "top-and-tail" IMS n A schematic diagram of the strategy for separating ginsenoside isomers; where A represents the multi-cycle separation of the Rh1 isomer, achieving baseline separation after 19 cycles, and this optimized method was subsequently applied to black ginseng extract; B represents the separation of four R / S configuration isomers of Rg2 and Rg3 using a "top-and-tail" strategy, selectively spraying 20(R)-Rg3 and 20(R)-Rg2 after 15 cycles, and continuing the cycle for the remaining isomers for another 30 cycles to achieve complete separation.

[0064] When four isomers (20(S)-Rg2, 20(R)-Rg2, 20(S)-Rg3, and 20(R)-Rg3) were simultaneously separated, after 15 cycles, a wrap-around phenomenon occurred where the 20(R)-Rg3 (fast ion) caught up with the 20(R)-Rg2 (slow ion), resulting in peak overlap. A top-and-tail IMS was used. n Strategy: After 15 cycles, 20(R)-Rg3 and 20(R)-Rg2 ions were selectively ejected within a short time window. Specific operating parameters were as follows: the ion selection window width was set to ±0.5 ms of the drift time (this window was set based on the approximately 1.2 ms difference in drift time between 20(R)-Rg3 and 20(R)-Rg2 to ensure selective ejection of fast ions without affecting slow ions); the ejection voltage was set to deflect the target ions into the pre-array store (i.e., the temporary ion storage area), while non-target ions continued to circulate in the annular channel. After ejection, the remaining ions (20(S)-Rg2 and 20(S)-Rg3) were circulated for another 30 cycles (a total of 45 cycles), ultimately achieving complete separation of the four R / S configuration isomers. Figure 1 B). Throughout the process, the cyclic traveling wave velocity and the static height of the traveling wave remain constant (375 m / s and 15 V, respectively).

[0065] (2) Consistency of CCS values ​​between single channel and multi-cycle

[0066] The relative deviation range between single-channel and multi-cycle CCS measurements was 0.02%–1.27%, indicating that extending the cycle had no significant effect on CCS measurements, and that multi-cycle CCS values ​​were reliable, as shown in Table 1.

[0067] Table 1 Comparison of TWCCS_N2 measurements between single-channel and multi-loop methods

[0068]

[0069] The above results indicate that extending the number of cycles to 15-45 has no significant effect on CCS determination.

[0070] Example 2: MS / MS identification of ginsenoside isomers

[0071] 2.1 Identification of Aglycone Types

[0072] Figure 2 MS / MS spectra of ginsenosides Rg2(A), Rg3(B), pseudoginsenoside F11(C), and pseudoginsenoside Rg2(E)(D) under 60 eV CID conditions show the characteristic aglycone ions of different dammarane isoforms. Figure 2 As shown, different dammarane isoforms exhibit characteristic aglycone ions under 60 eV CID conditions:

[0073] PPT type (Rg2): m / z 475.4 [C 30 H 52 O4–H] - and m / z 391.3[C 30 H 52 O4–H–C6H 12 ] -

[0074] PPD type (Rg3): m / z 459.4 [C 30 H 52 O3–H] - and m / z 375.3[C 30 H 52 O3–H–C6H 12 ] -

[0075] OT type (pseudoginsenoside F11): m / z 491.4 [C 30 H 52 O5–H] - and m / z 415.3[C 30 H 52 O5–H–C3H8O2] -

[0076] C17SCV type (pseudoginsenoside Rg2(E)): m / z 475.4 and m / z 417.3, none with m / z 391.3.

[0077] The order of glycosidic bond cleavage sensitivity is: PPT > PPD >> OT ≈ C17SCV.

[0078] 2.2 Identification of Sugar Type and Glycosidic Bond Position

[0079] Glucose-rhamnose (Re, Rg2): m / z 205.07

[0080] Glucose-glucose (Rb1, Rc, Rd, etc.): m / z 221.07

[0081] Glucose-arabinose / xylose (Rc, Rb2, Rb3): m / z 191.05

[0082] 1→6 linked disaccharide (Glc-Glc): m / z 323.08

[0083] 1→6 linked disaccharide (Ara / Xyl-Glc): m / z 293.08

[0084] 2.3 Identification of Glycosylation Sites

[0085] Figure 3 MS / MS spectra of ginsenosides Rh1(A), Rh2(B), CK(C), and F2(D) under 40 eV CID conditions are shown, illustrating the effect of differences in glycosylation sites on fragment abundance ratios. Figure 3 As shown, the abundance ratios of CK (C-3 glycosylation) and 20(S)-Rh2 (C-20 glycosylation) at 40 eV (m / z 459.4 / 621.4) were 1.6 and 0.51, respectively; at 60 eV, Rh2 exhibited a characteristic peak at m / z 537.3 (84 Da loss of the C-17 side chain), while CK did not show this peak. Glycosidic bond cleavage sensitivity: C-20 > C-6 > C-3.

[0086] Example 3: Methodological Validation of CCS Measurement

[0087] 3.1 Precision

[0088] Figure 4 Application of CCS in the identification of ginsenoside isomers: Figure (A) shows the intra-day and inter-day RSDs determined by CCS; (B) compares the CCS values ​​of ginsenosides in the control solution and ginseng matrix; (C) cross-validation of CCS values ​​between the two laboratories; (D) the relationship between ginsenoside molecular weight and CCS value; (E) compares the CCS values ​​predicted by the database with the experimentally determined CCS; (F) shows the CCS values ​​of aglycone fragments from different ginsenoside sources. Table S1 lists 25 ginsenosides. Intraday and interday precision data.

[0089] Table S1. 25 kinds of ginsenosides Intra-day and inter-day precision

[0090]

[0091] like Figure 4 As shown in A and Table S1, the intra-day precision RSD is <0.1%; the inter-day precision RSD of representative ginsenosides is 0.05%~0.3%.

[0092] 3.2 Cross-validation

[0093] Table S2 shows the relationship between the two laboratories. Value cross-validation data.

[0094] Table S2. Between the two laboratories Value cross-validation

[0095]

[0096] Cross-validation of CCS values ​​between the two laboratories showed that, Figure 4 E and Table S2: 88% of compounds have a relative deviation of <1%, and 12% of compounds have a relative deviation of <3.5%.

[0097] 3.3 Matrix effect

[0098] Table S3 shows the concentrations of ginsenosides in the reference solution and ginseng matrix. Value comparison situation.

[0099] Table S3. Ginsenosides in reference solution and ginseng matrix Value comparison

[0100]

[0101] like Figure 4 As shown in B and Table S3, the effect of the ginseng extraction matrix on the CCS value of ginsenosides is negligible.

[0102] 3.4 Molecular weight-CCS relationship

[0103] like Figure 4 As shown in D, the CCS value of ginsenosides increases with increasing molecular weight, and the more sugar substitutions, the larger the CCS value.

[0104] 3.5 Database Prediction Comparison

[0105] Table S4 shows the database prediction and experimental measurements. Statistical analysis of value comparison.

[0106] Table S4. Comparison of Database Prediction and Experimental Measurement of CCS

[0107]

[0108] like Figure 4 As shown in E and Table S4, the predicted values ​​from the CCSbase and AllCCS databases deviate significantly from the experimental values, and the predicted values ​​cannot correctly sort the R / S configuration isomers.

[0109] Example 4: Construction and application of a homologous aglycone characteristic CCS library

[0110] 4.1 Library Construction

[0111] Aglycone fragment ions were generated from different PPD and PPT type ginsenoside standards using an ion trap collision energy of 110 eV, and their CCS values ​​were determined. Results are as follows: Figure 4 And as shown in Table S5:

[0112] Table S5. Characteristics of homologous aglycones Library

[0113]

[0114] The CCS value is highly stable (RSD < 0.5%) among aglycones of the same type from different sources, and can serve as a characteristic "molecular fingerprint". Specifically, the CCS values ​​of aglycone fragments from 12 different PPD types of ginsenosides (including 20(S)-Rg3, F2, Rd, Rc, Rb2, Rb3, CK, 20(S)-Rh2, etc.) ranged from 224.19 to 224.69 Å. 2 (Average 224.45Å) 2 (RSD=0.08%); derived from aglycone fragments of eight different PPT-type ginsenosides (including F1, 20(S)-Rh1, 20(R)-Rh1, F5, 20(S)-Rg2, 20(R)-Rg2, etc.), with CCS values ​​ranging from 225.58 to 226.3 Å. 2 (Average 225.8Å) 2 (RSD=0.12%). This result indicates that the CCS value of aglycone fragments depends primarily on the aglycone type (PPD / PPT) and the configuration at the C-20 position (R / S), and is independent of the type, number, and linkage position of the glycosyl group.

[0115] Table 2. CCS library of homologous aglycone characteristics (based on the summary of determination data of 16 standards)

[0116]

[0117] The determination in Table 2 is based on the following: each value is the average ± standard deviation of multiple independent measurements. Based on the statistical distribution of the measured values ​​of each standard and with a margin of error, the characteristic range of PPD-type R-configuration aglycone fragments (CCS) can be determined to be 226.0–228.0 Å. 2 This range covers all measured values ​​and their fluctuation ranges.

[0118] 4.2 Application of R / S Configuration Determination

[0119] For ginsenosides with unknown configurations, the CCS values ​​of their aglycone fragment ions were measured and compared with the aforementioned characteristic CCS library:

[0120] If the CCS of the aglycone fragment is approximately 224.4 Å 2 (PPD type) or 225.8Å2 (PPT type), determined to be S configuration;

[0121] If the CCS of the aglycone fragment is approximately 226.2 Å 2 (PPD type) or 228.2Å 2 (PPT type), determined to be R configuration.

[0122] Example 5: Systematic identification of ginsenosides in ginseng extract

[0123] 5.1 Sample Preparation

[0124] Accurately weigh 0.5 g of ginseng powder (80 mesh), add 50 mL of water / methanol (50:50, v / v), extract by sonication (40 kHz) for 40 min, centrifuge at 13500 rpm for 10 min, and take the supernatant for UPLC-cIM-QTOF-MS analysis.

[0125] 5.2 Identification Process

[0126] Figure 5 This is a flowchart illustrating a comprehensive identification strategy for ginsenoside isomers based on MS / MS and CCS. Figure 5 As shown, the UNIFI workflow is used, and the following steps are followed for identification:

[0127] (1) Screening candidate compounds by precise mass number and isotopic distribution;

[0128] (2) Determine the aglycone type by identifying characteristic aglycone ions;

[0129] (3) Determine the sugar composition and linkage mode by glycan fragment ions;

[0130] (4) Determine glycosylation sites by fragment abundance ratio;

[0131] (5) Differentiate double bond positional isomers by the CCS value of the intact molecule;

[0132] (6) Determine the R / S configuration by the CCS value of the aglycone fragment.

[0133] 5.3 Identification Results

[0134] One hundred and one hundred and one ginsenosides were successfully identified from ginseng extracts, as shown in Table S6, covering PPD, PPT, OT, C17SCV, and oleanolic acid types. Statistically, by aglycone type: 46 PPD type, 28 PPT type, 5 OT type, 14 C17SCV type, and 8 oleanolic acid (OA) type. Among them, this invention method for the first time clearly distinguished the R / S configuration at the C-20 position in ginseng extracts, including R / S isomer pairs of important active ingredients such as 20(R) / 20(S)-Rg2, 20(R) / 20(S)-Rg3, 20(R) / 20(S)-Rh1, and 20(R) / 20(S)-Rh2 (see Table S6 for details).

[0135]

[0136] 5.4 Typical Identification Examples

[0137] Figure 6 This example illustrates the identification process for a compound with a mass-to-charge ratio of 845.4913, demonstrating the complete reasoning path from 15 candidate compounds to their final identification as ginsenoside Rg1, ginsenoside Rf, and notoginsenoside U. Figure 6 As shown, three peaks at m / z 845.4913 were detected at retention times of 6.16, 10.53, and 10.83 min, indicating 15 candidate compounds. These were further narrowed down to 7 compounds by m / z 475.3794 and 391.2851, classifying them as PPT-type compounds.

[0138] Peak at 6.16 min: No disaccharide characteristic fragments, m / z 475.3889 > m / z 637.4451 at 60 eV, identified as ginsenoside Rg1 (disaccharide monosubstituted at C-20).

[0139] Peaks at 10.53 min and 10.83 min: containing m / z 221.0717 (Glc-Glc disaccharide), with aglycone CCS of 225.58 and 225.64 Å, respectively. 2 The peak at 10.83 min was determined to be of S configuration; the peak at 10.83 min contained an additional m / z of 323.1165 (1→6Glc-Glc), which was identified as Panax notoginseng saponin U.

Claims

1. A method for the analysis and isomer identification of ginsenosides, characterized in that, Includes the following steps: Step 1: The sample containing ginsenosides was separated by multiple cycles of ion mobility mass spectrometry (cIMS-MS). The number of cycles for separation was set according to the difference in the collision cross section (CCS) of the isomers to be separated. Step 2: When a circumduction effect occurs during multi-loop separation where fast ions catch up with slow ions, top-and-tail IMS is used. n The strategy involves selecting the target ion and then continuing the cyclical separation to achieve complete separation of ginsenoside isomers. Step 3: By optimizing the collision energy in the ion trap region, ginsenosides are made to generate aglycone fragment ions after the ion source, and the CCS value of the aglycone fragment ions is measured. Step 4: Compare the CCS values ​​of the aglycone fragment ions measured in Step 3 with the characteristic CCS library of homologous aglycones, which contains the characteristic CCS values ​​of dammarane-type ginsenoside aglycone fragment ions with different configurations, R and S, to determine the R / S configuration of ginsenosides. Step 5: Integrate chromatographic retention time, precise mass number, tandem mass spectrometry (MS / MS) fragment ion information, intact molecule CCS value, and aglycone fragment CCS value to perform multidimensional identification of ginsenosides.

2. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The number of cycles for multi-cycle ion mobility separation in step one is 15 to 45. The ion mobility separation parameters are: 2 pushes per chamber, 200 chambers, 375 m / s traveling wave velocity, and 15 V traveling wave static height.

3. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The collision energy of the ion trap region in step three is 80~120 eV, preferably 110 eV, and the aglycone fragment ions include: Protopanaxadiol PPD type aglycone ion mass-to-charge ratio m / z 459.4; Protopanaxadiol PPT type aglycone ion m / z 475.

4.

4. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, In the homologous aglycone characteristic CCS library described in step four, the characteristic CCS values ​​of PPD-type aglycone fragment ions are: S configuration 224.0-225.0 Å. 2 R configuration 226.0-228.0 Å 2 The characteristic CCS values ​​of PPT-type aglycone fragment ions are: S configuration 225.0-227.0 Å. 2 R configuration 227.5-229.0 Å 2 .

5. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The MS / MS fragment ion information mentioned in step five is used to distinguish the following isomer types: Aglycone type isomers, based on characteristic aglycone ions m / z To distinguish; Sugar type isomers, based on characteristic sugar chain fragment ions m / z Distinguish between 205.07, 221.07, or 191.05; Glycosidic bond linkage position isomers, based on characteristic disaccharide fragment ions m / z Distinguish between 323.08 and 293.08; Glycosylation site isomers are distinguished based on the relative abundance ratio of characteristic fragment ions; Double bond position isomers are distinguished based on the difference in CCS value of the complete molecule, with the CCS value of the Δ20(22) double bond configuration being greater than that of the Δ20(21) double bond configuration.

6. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The ginsenosides are dammarane-type ginsenosides, including protopanaxadiol (PPD), protopanaxadiol (PPT), octyltriol (OT), and C-17 side chain variant (C17SCV).

7. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The process begins with a chromatographic separation step, performed using ultra-high performance liquid chromatography (UPLC). The column used is an ACQUITY HSS T3 column (1.8 μm, 100 × 2.1 mm), with a column temperature of 40 °C. The mobile phase consists of an aqueous solution (phase A) containing 0.1% formic acid and acetonitrile (phase B). The flow rate is 0.3 mL / min, and the gradient elution program is as follows: 0–1.5 min 95% A, 1.5–3 min 95→75% A, 3–6 min 75→73% A, 6–9 min 73→67.5% A, 9–12 min 67.5→64% A, 12–18 min 64→53% A, 18–22 min 53→0% A, 22–24 min 0% A, 24–25 min 0→95% A, and 25–29 min equilibration with 95% A.

8. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The cIMS-MS was operated in negative ion mode with a capillary voltage of 2.0 kV, a cone voltage of 40 V, a source temperature of 120 °C, a desolventizing temperature of 400 °C, a cone gas flow rate of 50 L / h, a desolventizing gas flow rate of 800 L / h, nitrogen as the collision gas and ion mobility gas, and a data acquisition range of m / z 50–2000, acquired in negative ion mode.

9. The method for analyzing ginsenosides and identifying isomers according to claim 1, characterized in that, The method is used for the systematic identification of ginsenosides in extracts of ginseng plants, and can simultaneously identify PPD type, PPT type, OT type, C17SCV type ginsenosides and their isomers.

10. A method for constructing a CCS library featuring homologous aglycones, characterized in that, Includes the following steps: Step 1: Select dammarane-type ginsenoside standards with known R / S configurations; Step 2: Aglycone fragment ions were generated using cyclic ion mobility mass spectrometry (cIMS-MS) under optimized collision energy conditions in the ion trap region. Step 3: Measure the collision cross section (CCS) value of aglycone fragment ions generated by each standard, and establish a database of characteristic CCS values ​​of aglycone fragment ions with different configurations (R-type and S-type) and different aglycone types (PPD-type and PPT-type). Step 4: Verify the stability of the characteristic CCS values: intra-day precision relative standard deviation RSD < 0.1%, inter-day precision RSD < 0.3%, cross-instrument relative deviation < 3.5%, and remain stable in the same type of aglycone fragments from different compound sources.

Images