Multi-technology Comparison For Systematic Characterization Of Ginsenosides And The Simultaneous Identification Of Multiple Panax-derived Herbal Medicines

Quality control is a necessary artifice to ensure the clinical effectiveness and safety of traditional Chinese medicine（TCM）,for which the systematic material basis elucidation,authentication,and quality evaluation,are the core issues.The Panax-derived TCMs,especially Panax ginseng C.A.Meyer,P.quinquefolius L.,and P.notoginseng（Burk.）F.H.Chen,have obvious tonifying effects receiving extremely wide consumption worldwide.Chinese Pharmacopoeia（2015 edition） collects a total of seven TCMs that are derived from the Panax species: Panax ginseng C.A.Meyer（PG）,red ginseng（RG）,P.quinquefolius L.（PQ）,P.notoginseng（Burk.）F.H.Chen（PN）,leaf of P.ginseng（PGL）,P.japonicus C.A.Meyer（PJ）,and P.japonicus C.A.Mey.var.major（Burk.）C.Y.Wu et K.M.Feng（PJM）.Saponins are the common components for various Panax species,which also act as the indicator for the purpose of quality control.In view of the containing of saponins as the common components,the precise quality control of Panax-derived TCMs is faced with great challenges.It is a feasible solution to achieving the precise identification of these Panax-derived TCMs by systematic clarification of the saponin compositions and the establishment of "Identification Markers".As a core research of the General Program of National Natural Science Foundation of China（“Monomethod Characterization of Structure Analogs-Based and Multi-technology Integrated New Mode of TCM Quality Standard Investigation”）,this thesis,based on a variety of data acquisition technologies available on ultra-high performance liquid chromatography/high resolution mass spectrometry（UHPLC/HRMS）,is designed to establish the efficient,highly sensitive,and high-coverage profiling and characterization approaches aiming to accomplish the simultaneously systematic characterization and differentiation analysis of the saponins from multiple Panax-derived TCMs,by a single method.Based on the flexible scan strategies facilitated by quadrupole-Orbitrap mass spectrometry（Q-Orbitrap MS）,this thesis established six untargeted or targeted scan approaches,in the first step.Taking the characterization of the saponins from the flower buds of P.ginseng（PGF） as a case for performance comparison,the results indicated the data-dependent acquisition approach by including a precursor ions list（PIL-DDA）and enabling the “If idle-pick other”（IIPO） function was highly sensitive and high-coverage for the TCM multicomponent characterization.On the basis of our previously elaborated in-house ginsenoside database（recording 499 ginsenosides,corresponding to 169 different masses）,a PIL containing the saponin target masses was thus constructed.Six data acquisition methods were established on the Q Exactive^TM Q-Orbitrap high-resolution LC/MS platform by optimizing the key parameters: 1)Full MS/dd-MS²（M1）;2)Full MS/dd-MS²/PIL/IIPO（M2,PIL-containing）;3)ISCID-Full MS/dd-MS²-Tags（M3）;4)Full MS/AIF/NL-dd-MS²（M4）;5)PRM（M5,PIL-containing）;6)Targeted SIM/dd-MS²（M6,PIL-containing）.A "two-step" approach,capable of comparing the characterization effects of different MS methods for PGF saponins,was developed.First,the theoretical mass-to-charge ratio values of the deprotonated molecules was calculated based on 169 different masses corresponding to the known ginsenosides,and the number of the targeted components scanned by each method was recorded as N_T by the use of mass defect filtering（MDF: ± 10 mDa variation）.On this basis,the number of the precursors recorded in N_T that triggered MS² fragmentation was noted as N_T2.As a result,the characterization results were as follows sorted in a descending order according to the numbers N_T/NT₂: M5（17670/17704）,M6（602/606）,M2（487/347）,M3（469/339）,M4（489/246）,and M1（459/339）.Comparatively,as the targeted MS² acquisition approaches,the ability to obtain the MS information of target components for M5 and M6 was much stronger than the other four untargeted manners,however,they were the least able to identify the unknown components when facing the complex chemical systems,and in lack of the full-scan MS¹ data.The number of the target masses scanned by M2 was comparable to that of M4,but the N_T2 indicator was higher than the other three.And accordingly,M2 was regarded as a highly sensitive,and simultaneously targeted/untargeted acquisition approach.It could enhance the coverage of target masses,in contrast to the conventional DDA strategies,the superiority of which may be even more significant when facing a extremely complex chemical matrix（such as the in vivo biological sample）.Aiming to further enhance the coverage of unknown saponins by the M2 method,in this thesis,a new strategy for constructing the PIL was presented,by the large-scale molecular design-based establishment of "Ginsenoside Virtual Library" and MDF.The newly developed Full MS/dd-MS²/PIL/IIPO method was able to identify 164 saponins from PGF.Molecular design rule for ginsenosides was formulated as follows: 27 aglycones,5 different glycosyls,24 substituents,≤ 6 glycosyl groups and ≤ 2 substituents contained in each molecule,A virtual library containing 13,536 ginsenoside molecules was constructed by C language programming.Moreover,combined with the fixed variation range MDF（± 10 m Da）,totally 1859 target masses were screened from the negative-mode full-scan MS1 data of PGF,generating a new PIL（PIL-2）for PGF,based on which a new M2 method was developed.Compared with the known saponin database-derived PIL,the DDA method,by including a PIL based on the "Ginsenoside Virtual Library",could newly acquire the MS² spectra of 17 components,and the structures of 9 components thereof were identified.We could characterize 164 ginsenosides,by analyzing the high-energy collision-induced dissociation-MS²（HCD-MS²）data of PGF,29 of which were confirmatively identified by comparison the t R and MS information with the reference compounds,and 34 therein were with unknown masses.To further explore the differences between DDA and DIA in the multicomponent characterization of TCM,this dissertation using PJ as a case,established two data acquisition approaches on the Vion^TM IMS-QTOF high-resolution LC/MS instrument,DDA（MS-MS/MS,without PIL）and HDMS^E,in combination with the efficient automated peak annotation workflows by UNIFI.A comprehensive qualitative analysis of the chemical constituents of PJ finally led to the identification or tentative characterization of 178 saponins,which could indicate the complementarity of DDA and DIA modes.By means of the optimization of chromatography-MS parameters,the profiling methods utilized a BEH Shield RP18 column,acetonitrile-0.1 % formic acid as the mobile phase,and the negative-mode MS detection by DDA and HDMS^E facilitated by the Vion IMS-QTOF instrument.In particular,a "three-step" approach was proposed,for the first time,to optimize the mass-dependent ramp collision energy（MDRCE） in setting DDA.Standardized workflows were elaborated for the efficient annotation of both the DDA and HDMS^E data by UNIFI and searching the in-house ginsenoside database（containing 504 known ginsenosides and 60 reference compounds）.We could identify or tentatively characterize 178 saponins from PJ,of which 75 may have not ever been isolated from the Panax genus.Amongst them,168 components were analyzed based on the DDA data,while 10 ones were supplemented from the HDMS^E data.In comparison,the fragmentation information of ginsenosides acquired by DDA was more reliable with fewer false positive results,while HDMS^E could provide the CCS（collision cross section） values with the highest coverage.The integration of these two different data acquisition modes can obtain complementary information supporting the structural identification of metabolites.On the basis of systematic ginsenosides characterization,a novel characteristic chromatogram（CC）,based on the selective ion monitoring（SIM） of the sapogenin product ions due to the positive-mode in-source fragmentation of ginsenosides（pISF-G）,was established to realize the authentication of multiple Panax varieties.By selectively monitoring four sapogenin product ions,a group-target profiling characteristic chromatogram was elaborated,which,by combining chemometrics,could enable the authentication of seven Ginseng drugs that are recorded in Chinese Pharmacopoeia（2015 edition）,as well as identify the presence of PG/PQ/PN in 15 different TCM formulae,on both the Vion IMS-QTOF and QTrap 4500 mass spectrometers.By applying six high-resolution mass spectrometers（Agilent 6520 QTOF and 6545 QTOF,Waters Xevo G2-S QTOF and Vion IMS-QTOF,Thermo Fisher Q Exactive Q-Orbitrap and LTQ-Orbitrap Velos Pro）,the susceptibility of p ISF-G was primarily demonstrated.By analyzing 58 ginsenoside reference compounds,the abundant ions in the low-mass region of the positive full-scan MS¹ spectra were generated from the protonated sapogenins by the successive elimination of H₂ O,and showed specificity for ginsenosides classification.By the selective monitoring of four characteristic sapogenin fragments（PPD-type: m/z 407.37/CCS 206.24 ?²;PPT-type: m/z 423.36/CCS 211.26 ?²;OA-type: m/z 439.36/CCS 209.60 ?²;OT-type: m/z 457.37/CCS 217.81 ?²）,untargeted characterization of ginsenosides was achieved,and the characteristic chromatograms for seven Ginseng drugs were elaborated.Chemometric analysis of the multi-batch CC data resulted the discovery of 35 marker compounds.More significantly,we could readily discriminate among PG,PQ and PN,from 15 TCM formulae.These results could be repeatable on the QTrap 4500 mass spectrometer.The characteristic chromatogram technology has initially verified the feasibility of the "Characterization of Analogous Structures with the Same Approach" strategy,and is a new technology that supports the "Monomethod-for-Multiple Use" strategy for TCM quality control.In this thesis,an improved DDA approach based on a virtual ginsenoside database-oriented PIL was established on the Q-Orbitrap mass spectrometer,for the first time,with high sensitivity,high coverage,and simultaneous targeted/untargeted features.It was first confirmed that,on the Vion IMS-QTOF mass spectrometer,the integration of DDA and DIA was able to obtain complementary MS information favoring the comprehensive multicomponent characterization of TCM.A systematic research was performed,for the first time,to probe into the genesic mechanism of the positive-mode in-source fragmentation of ginsenosides,which resulted in the proposing of a novel characteristic chromatogram of ginsenosides capable of the authentication of Ginseng herbal medicines,especially to identify those Panax species from the TCM formulae.These results and findings obtained in this thesis provide a methodological reference for the comprehensive chemical basis elucidation and precise quality control of TCM.