Application Of High Performance Liquid Chromatography-High Resolution Mass Spectrometry With New Data Mining Techniques In Rapid Screening Of Active

There are a variety of components in Traditional Chinese Medicines（TCMs）,and the index components are usually used for the quality evaluation.The present analytical methods include the biological methods（biochromatography,biomicroarray,and enzyme-linked immunosorbent assay,etc.）and chromatographic methods（liquid chromatography,gas chromatography and capillary electrophoresis coupled with various detectors）.Among them,high performance liquid chromatography coupled with high resolution mass spectrometry（HPLC-HRMS）shows high sensitivity and efficiency,and it can obtain rich structural information.However,the traditional data processing of HPLC-HRMS is labor-intensive,and effective data mining techniques are therefore needed to simplify the data processing.Commonly used data mining tools include background subtraction（BS）and mass defect filtering（MDF）.BS can significantly remove majority of background signals;however,the HRMS data were largely overlapped.The efficiency of MDF relied on the templates,which can result in displaying more false-positives in MDF-processed chromatograms.This study presented an HPLC-HRMS method in combination with a novel data-mining tool,i.e.,background subtraction tandem ring double binding（BS-RDB）,for rapid profiling of index components of two representative herbs（Artemisia annua L.and Radix Dichroae）.For comparison,two commonly used data-mining tools（BS and MDF）were also applied to the detection of index components.To evaluate the efficiency of the present techniques,the traditional purification techniques were used to separate index components from the herb Radix Dichroae.In addition,the PK-PD of artemisinin in the herb Artemisia annua L.was also performed.1.Screening of index components and metabolites of Artemisia annua L.by HPLC-HRMSCompared to pure artemisinin（QHS）,the same amount of QHS in dried leaves of Artemisia annua L.（AAL）showed higher antimalarial activity.Two representative types of index components（sesquiterpene lactones and flavonoids）probably showed synergistic antimalarial effect on QHS,but which components and the mechanism of action deserve further study.In this chapter,a new analytical strategy（i.e.,HPLC-HRMS coupled with BS-RDB）was used to rapidly screen the index components of methanol AAL extracts（MAE）,as well as their metabolites in rats.Two commonly used data-mining techniques（BS and MDF）were used to evaluate the efficiency of the present data mining tool.By comparing with reference standards,structure of major components was confirmed.Structures of other detected components or metabolites were characterized based on MS information（elemental compositions,RDB values and fragmentation patterns）,the biotransformation network of sesquiterpene lactones（SL）and flavonoids（FL）established in AAL and the biosynthesis pathways of sesquiterpene lactones in AAL.The results showed that a total of 38 prototypes of SL and 35 prototypes of FL were detected in MAE.A total of 20 metabolites of SL（included 5 phase I metabolites and 15 phase II metabolites）and 10 metabolites of FL（included 1 phase I metabolite and 9 phase II metabolites）were detected in biological samples of rats after an oral administration of MAE.After the BS processing,all prototypes of SL and prototypes of FL were detected,but were not differentiated in MAE;prototypes of two types of components and their metabolites in biological samples of rats were not detected.After the MDF processing,all prototypes of SL,some prototypes of FL and their metabolites were detected in MAE and biological samples of rats（only 18 out of 35 prototypes of FL were found in the MAE）;two types of components were differentiated well,but included many false-positives.After the BS-RDB processing,all prototypes of SL,prototypes of FL and their metabolites were detected in MAE and biological samples of rats;two types of components were differentiated well with fewer false-positives.HPLC-HRMS coupled with the novel data mining tool BS-RDB can rapidly detect two representative types of components in MAE and their metabolites in biological samples of rats.In addition,two types of components were differentiated well.A total of 20 prototypes of SL and 6 prototypes of FL were not previously reported in MAE.The main metabolic pathways for SL included hydroxylation and glucuronidation.The main metabolic pathways for FL included hydroxylation,methylation,deglycosylation,glucuronidation and sulfation.2.Screening of index components of Radix Dichroae by HPLC-HRMSRadix Dichroae extracts（MRE）showed remarkable antimalarial activity,and the main antimalarial active component is dichroine.Whether other minor or trace alkaloids（AL）analogues exist in Radix Dichroae deserves further investigation.In this chapter,HPLC-HRMS coupled with a new analytical strategy BS-RDB was used to rapidly screen the alkaloids in MRE.Two commonly used data-mining techniques（BS and MDF）were used to evaluate the advantages of BS-RDB.The traditional separation of Radix Dichroae was performed to verify the effectiveness of the present new analytical strategy.By comparing with reference standards,structure of major alkaloids was confirmed.Structures of other detected components were characterized based on MS information（elemental compositions,RDB values and fragmentation patterns）.A total of 7 prototypes of AL were detected in MRE,among which dichroine and isodichroine were found in the highest MS abundance.All ALs as well as many interferences were observed in the one-step BS-processed chromatograms of MRE.After the MDF processing,5 out of 7 AL were found in MRE.After BS-RDB processing,all 7 AL were found in MRE,with few false-positives.Two AL（dichroine and isodichroine）and 7-OH courmarin were separated based on the traditional separation technique,and other trace AL were not obtained.HPLC-HRMS coupled with the new data mining tool BS-RDB can rapidly profile the alkaloids in Radix Dichroae,and 4 prototypes of AL were not previously reported.3.Synergistic anti-malarial index components of Artemisia annua L.on QHSSL and FL are two main types of index components in AAL.Ethyl acetate and petroleum ether can extract components of SL and FL in AAL,respectively.Compared to pure artemisinin（QHS）,the synergistic anti-malarial effects of QHS in two extracts were studied,in terms of its PD-PK profile.HPLC-HRMS coupled with BS-RDB was applied to screen the differential index components in two extracts.The pharmacokinetics of QHS in combination with the differential index components in high content were performed to reveal contributors in AAL leading to enhanced antimalarial activity of QHS.Compared to pure QHS,remarkably synergistic antimalarial activity of QHS was observed in PAE,and a 2.1-fold increase in exposures to both QHS and its major metabolite DQHS exposure（AUC）was found in healthy rats after oral consumption of QHS in PAE（P<0.05）.These results suggested that the differential matrices in PAE contributed to the absorption of QHS in vivo.Prototypes of SL were the main index components in PAE.The components with high content were artemisinin B（AB）and artemisinic acid（AAC）.Compared to QHS,a 1.8-fold increase in QHS exposure（AUC）（P<0.05）and a 1.7-fold increase in DQHS exposure（AUC）（P<0.05）was found in healthy rats after oral consumption of QHS in combination with AB and AAC;however,the QHS and DQHS exposure did not change significantly after oral consumption of QHS in combination with either AB or AAC（P>0.05）.These results showed that AB plus AAC can contribute to the absorption of QHS in vivo.Application of HPLC-HRMS coupled with new strategy BS-RDB revealed that AB and AAC were contributors in AAL leading to enhanced antimalarial activity of QHS.The present results indicated the efficiency of the present new analytical strategy.Based on the results above,the HPLC-HRMS-based technique in combination with the new data mining tool BS-RDB can screen index components in herbs rapidly and effectively.In addition,rich structural information can be obtained based on more sensitive MS datasets.However,LC-MS based analysis alone is never sufficient to the unambiguous identification of unknown compounds.Combining with other powerful analytical tools such as LC-NMR can be expected to acquire complementary structural information and thus ensure global qualitative analysis of complicated herbal components.