CN113759003A

CN113759003A - Licorice origin distinguishing method based on UPLC fingerprint spectrum and chemometrics method

Info

Publication number: CN113759003A
Application number: CN202010499895.3A
Authority: CN
Inventors: 张堂; 祝宇龙; 张静; 吴梦飞; 彭灿; 彭代银
Original assignee: Hefei Weiyuantang Traditional Chinese Medicine Co ltd; Anhui University of Traditional Chinese Medicine AHUTCM
Current assignee: Hefei Weiyuantang Traditional Chinese Medicine Co ltd; Anhui University of Traditional Chinese Medicine AHUTCM
Priority date: 2020-06-04
Filing date: 2020-06-04
Publication date: 2021-12-07
Anticipated expiration: 2040-06-04
Also published as: CN113759003B

Abstract

The invention discloses a liquorice producing area distinguishing method based on UPLC fingerprint spectrum and a chemical metering method, wherein UPLC is adopted to establish fingerprint spectrums of different producing areas and different extracts of liquorice, the fingerprint spectrums are combined with chemical metering methods such as SA, HCA, PCA, OPLS-DA and the like, a characteristic marker compound which has important influence on liquorice quality evaluation is screened out, then UPLC/Q-TOF-MS is adopted to identify the characteristic marker compound, and 7 peaks are quantitatively measured. And finally, simulating a discriminant function equation of the liquorice producing area by using a discriminant analysis method, and tracking the medicinal material producing area. The method not only provides valuable and potential reference for the quality, extraction method and development and utilization of the licorice root medicinal material or other related food and drugs, but also provides reference and basis for tracing the origin and quality identification of the licorice root medicinal material, can quickly and effectively identify the origin of the licorice root medicinal material through a discriminant function equation, has high positive judgment rate, and is suitable for popularization and application.

Description

Licorice origin distinguishing method based on UPLC fingerprint spectrum and chemometrics method

Technical Field

The invention relates to the technical field of traditional Chinese medicine component research, in particular to a liquorice producing area distinguishing method based on UPLC fingerprint spectrum and a stoichiometric method.

Background

Genuine herbs, also called genuine herbs, are the terms of high-quality Chinese herbs, meaning that the herbs have good effect. The genuine medicinal materials are preferably selected by long-term clinical application of the traditional Chinese medicine, are produced in specific regions, have better quality and curative effect, stable quality and higher popularity compared with the same Chinese medicinal materials produced in other regions, and are a unique comprehensive judgment standard for controlling the quality of the Chinese medicinal materials in the traditional Chinese medicine.

Licorice (Glycyrrhiza uralensis Fisch) is the earliest recorded and researched country in the world in China, and is loaded in Shen nong Ben Cao Jing of east Han. Has been used by traditional medicine for a variety of purposes, such as: invigorating qi, strengthening middle warmer, clearing away heat and toxic materials, eliminating phlegm, and stopping bleeding. Licorice root, radix Glycyrrhizae, the first herb in the Roots of the compendium of materia Medica, is the concoction for relieving seventy two toxins. The variation of the climatic environment and soil conditions of different producing areas can cause the difference of the internal quality of the liquorice. Licorice is mainly distributed in the aisles of Hexi in Xinjiang, inner Mongolia and Gansu, around Longxi, and in Ningxia. Compared with the same licorice medicinal materials produced in other areas, the licorice growing in the areas has better quality and curative effect, stable quality and higher popularity. However, the phenomena of secondary quality and counterfeit quality widely exist in the market, so that the reputation and the brand value of the liquorice genuine medicinal materials are seriously influenced, and the liquorice genuine medicinal materials are more likely to harm human health when used for treating disease conditions. However, a rapid analysis method capable of effectively identifying the production place of the licorice medicinal material is not available at present. Therefore, there is an urgent need to develop a method for identifying the origin and production area of licorice herbs, determining whether the herbs are genuine herbs, and controlling the quality of the genuine herbs.

Disclosure of Invention

In order to solve the problems, the invention aims to provide a liquorice producing area distinguishing method based on a UPLC fingerprint spectrum and a chemometric method, the method provides reference and basis for liquorice medicinal material producing area tracing and quality identification, the positive distinguishing rate is high, and the liquorice producing area distinguishing method is suitable for popularization and application.

The invention is realized by the following technical scheme:

a liquorice producing area distinguishing method based on UPLC fingerprint spectrum and a stoichiometric method comprises the following steps:

(1) collecting Glycyrrhrizae radix materials of each production area, and establishing fingerprint of different Glycyrrhrizae radix production areas and different Glycyrrhrizae radix extracts by ultra high performance liquid chromatography; the Glycyrrhrizae radix is selected from Sinkiang, Gansu and inner Mongolia; the extract is a liquorice alcohol extract and a liquorice water extract;

(2) screening the characteristic marker compounds by utilizing a similarity analysis SA, a hierarchical clustering analysis HCA, a principal component analysis PCA and an orthogonal partial least squares discriminant analysis method OPLS-DA;

(3) identifying the characteristic marker compound by adopting ultra-high performance liquid chromatography-quadrupole time-of-flight mass spectrometry, and carrying out quantitative analysis by adopting ultra-high performance liquid chromatography;

(4) gradually performing discriminant analysis by adopting a Wilk's lambda method to obtain discriminant function equations of different production areas and different extracts of the liquorice, and discriminating the production areas of the liquorice medicinal materials;

the discriminant function equation of alcohol extracts of different licorice producing areas is as follows:

y Xinjiang-7.629S₁+3.029S₂-0.404S₃+663.007S₄+1765.007S₅+0.537S₆-82.594；

Y Gansu ═ 8.026S₁+3.36S₂-1.452S₃+830.744S₄+1857.135S₅+0.619S₆-95.093；

Y inner Mongolia 7.646S₁+3.706S₂-0.489S₃+649.922S₄+1793.914S₅+0.599S₆-94.193；

Wherein S is₁Is liquiritin apioside S₂Is liquiritin, S₃Is apiose isoliquiritigenin, S₄Is liquiritigenin, S₅Is echinocandin, S₆Is glycyrrhizic acid.

The discriminant function equation of water extracts of different production areas of liquorice is as follows:

y Xinjiang-3.783S₁+2.731S₂-0.137S₃+2418.876S₄-131.472S₅+0.697S₆-44.812；

Y Gansu ═ 3.698S₁+3.163S₂-1.26S₃+3107.573S₄-327.356S₅+0.792S₆-49.803；

Y inner Mongolia 3.423S₁+3.539S₂-1.159S₃+3713.539S₄-500.921S₅+0.957S₆-58.480；

Wherein S is₁Is liquiritin apioside S₂Is liquiritin, S₃Is apiose isoliquiritigenin, S₄Is echinocandin, S₅Is licochalcone, S₆Is glycyrrhizic acid.

Preferably, in the step (1), the preparation method of the licorice alcohol extract comprises the following steps: precisely weighing Glycyrrhrizae radix powder, precisely adding 30-50 times of 60-80% ethanol, weighing, ultrasonically extracting for 20-40min, cooling, weighing again, adding 60-80% ethanol for supplementing weight loss, shaking, standing, filtering with 0.22 μm organic filter membrane, and collecting filtrate.

Preferably, in the step (1), the preparation method of the licorice aqueous extract comprises the following steps: weighing Glycyrrhrizae radix, decocting with 12-16 times of water for 1-2 hr, pouring out extractive solution, adding 6-10 times of water, decocting for 0.5-1 hr, mixing the decoctions, diluting to content of 0.025g per 1mL, filtering with 0.22 μm water system filter membrane, and collecting filtrate.

Preferably, in the step (1) and the step (3), the chromatographic conditions in the ultra-high performance liquid chromatography are as follows: a C18 column; the detection wavelength is 254 nm; taking acetonitrile as a mobile phase A and 0.1% formic acid aqueous solution as a mobile phase B, and performing gradient elution, wherein the gradient elution procedure is as follows: 0-2 min, 80-75% B; 2-4 min, 75-73% B; 4-12 min, 73-60% B; 12-18 min, 60-40% B; 18-20 min, 40-80% B; 20-25 min, 80% B; the column temperature is 30 ℃; the flow rate was 0.3ml/min and the amount of sample was 1. mu.L.

Preferably, in step (3), the mass spectrometry conditions are: an electrospray ionization (ESI) source, wherein the ESI source works at the positive and negative polarities simultaneously, the temperature of a capillary tube is 270 ℃, and the evaporation temperature is 300 ℃; the gas flow of the sheath layer is 35 mu L/min, the auxiliary gas flow is 5 mu L/min, the full scan analysis is carried out, the Q-TOF acquisition rate is 0.1s, and the mass range is 50-1200 m/z.

Compared with the prior art, the invention has the following beneficial effects:

the invention provides a method for distinguishing a liquorice producing area based on a UPLC fingerprint spectrum and a chemometrics method. Fingerprint spectrums of different producing areas and different extracts of liquorice are established by adopting Ultra Performance Liquid Chromatography (UPLC), the fingerprint spectrums are combined with a similarity evaluation system and chemometrics methods such as HCA, PCA and OPLS-DA, a characteristic marker compound which has important influence on the quality evaluation of the liquorice is screened out, then the characteristic marker compound is identified by adopting the UPLC-quadrupole time-of-flight mass spectrometry (UPLC/Q-TOF-MS), and 7 peaks are quantitatively determined. And finally, simulating a discriminant function equation of the liquorice producing area by using a discriminant analysis method, and tracking the medicinal material producing area. The method not only provides valuable and potential reference for the quality, extraction method and development and utilization of the licorice root medicinal material or other related food and drugs, but also provides reference and basis for tracing the origin and quality identification of the licorice root medicinal material, can quickly and effectively identify the origin of the licorice root medicinal material through a discriminant function equation, has high positive judgment rate, and is suitable for popularization and application.

Drawings

FIG. 1 is a fingerprint chart of 13 batches of alcohol extract of Glycyrrhiza glabra;

FIG. 2 is a common peak of 13 batches of alcohol extracts of Glycyrrhiza glabra;

FIG. 3 is a fingerprint chart of 13 batches of aqueous extract of Glycyrrhiza glabra;

FIG. 4 is a graph of the consensus peaks of 13 batches of aqueous licorice extracts;

FIG. 5 is a hierarchical clustering analysis dendrogram of 13 batches of licorice aqueous extracts;

FIG. 6 is a hierarchical clustering analysis dendrogram of 13 batches of alcohol extracts of Glycyrrhiza glabra;

FIG. 7 is a graph of Principal Component Analysis (PCA) scores of 13 batches of licorice alcohol extracts;

FIG. 8 is a graph of Principal Component Analysis (PCA) scores for 13 batches of licorice aqueous extracts;

FIG. 9 is a load graph of the Principal Component Analysis (PCA) of 13 batches of the licorice alcohol extract;

FIG. 10 is a graph of the Principal Component Analysis (PCA) load for 13 batches of licorice aqueous extracts;

FIG. 11 is a plot of the load S-plots of 13 batches of alcohol extracts of Glycyrrhiza glabra;

FIG. 12 is a graph of the loading S-plots of 13 batches of aqueous licorice extracts;

FIG. 13 is a mass spectrum of apioside liquiritin.

Detailed Description

The present invention is further illustrated by the following specific embodiments, which are not intended to limit the scope of the invention.

1. Medicinal material collection and fingerprint establishment

1.1 reagents and drugs

13 batches of licorice medicinal materials were collected in Xinjiang, Gansu and inner Mongolia, and divided into three groups: S1-S5 are medicinal materials in Xinjiang area, S6-S10 are medicinal materials in Gansu area, and S11-S13 are medicinal materials in inner Mongolia area.

Apiosomelin (batch number: PRF7061501) and apiosomelin (batch number: PRF8022729) are purchased from Chengdui scientific development Co., Ltd, and the purity is more than 98%; glycyrrhizin (batch No. 111610-201607, purity 93.1%), glycyrrhizic acid (batch No. 110731-201619, purity 93%) were purchased from China institute of food and drug testing; liquiritigenin (batch No. 17022104, purity 99.07%), licochalcone B (batch No. 17032202, purity 99.74%), and licochalcone A (batch No. 17032203, purity 99.77%) were purchased from Chinese Standard substance Web. Acetonitrile (chromatographically pure, batch: LCC0S85), ethanol (chromatographically pure, batch: LG40S76), formic acid (analytically pure, batch: 20140208); the water was ultrapure water prepared in a laboratory water purification system (PALL USA).

1.2 preparation of reference stock solutions

Precisely weighing 10.0mg of each chemical standard of liquorice, respectively placing the chemical standards in a 1mL volumetric flask, and then diluting the chemical standards by 70% ethanol step by step to prepare a series of mixed reference solutions. All solutions were stored at 4 ℃ until needed for analysis.

1.3 preparation of test solutions

1.3.1 preparation of test solution of Glycyrrhiza alcohol extract

Precisely weighing 1.0g of licorice medicinal powder into a 50mL conical flask with a plug, precisely adding 40mL of 70% ethanol, weighing for a fixed mass, ultrasonically extracting for 30min, cooling, weighing for a fixed mass again, adding 70% ethanol for complementing weight loss, shaking uniformly, standing, filtering with a 0.22 μm organic filter membrane, and taking the subsequent filtrate to obtain the licorice alcohol extract (LEE).

1.3.2 preparation of test solution of aqueous extract of Glycyrrhiza

Weighing 50.0g of licorice root medicinal material, firstly adding 14 times of water, decocting for 2h, pouring out the extract, then adding 8 times of water, decocting for 1h, combining the two decoctions, and diluting until the medicine content per 1mL is 0.025 g. Filtering the diluted solution with 0.22 μm water system filter membrane, and collecting the filtrate to obtain Glycyrrhrizae radix water extract (LWE).

1.4 ultra high performance liquid UPLC chromatographic conditions

Waters-CORTECS C18 column (4.6X 50m m, 2.7 μm); the detection wavelength is 254 nm; taking acetonitrile as a mobile phase A and 0.1% formic acid aqueous solution as a mobile phase B, and performing gradient elution, wherein the gradient elution procedure is as follows: 0-2 min, 80-75% B; 2-4 min, 75-73% B; 4-12 min, 73-60% B; 12-18 min, 60-40% B; 18-20 min, 40-80% B; 20-25 min, 80% B; the column temperature is 30 ℃; the flow rate was 0.3ml/min and the amount of sample was 1. mu.L.

1.5 establishment of Liquorice UPLC fingerprint

In order to establish a characteristic fingerprint spectrum and a comprehensive fingerprint spectrum, 13 batches of licorice root samples with different producing areas and different extraction methods are analyzed by a UPLC method, and a traditional Chinese medicine chromatography fingerprint spectrum similarity evaluation system (2012A edition) is adopted for standardization. According to fingerprint analysis, 25 common peaks (FIG. 2) were isolated from 13 alcohol extract samples (FIG. 1) of licorice root crude material, and 23 common peaks (FIG. 4) were isolated from water extract samples (FIG. 3), which had high resolution and were used as reference chromatograms.

2. Screening for signature Compounds

2.1 Similarity Analysis (SA)

And (3) adopting professional software 'traditional Chinese medicine chromatography fingerprint similarity evaluation system' (2012A edition) released by the State pharmacopoeia Committee to calculate the similarity of alcohol extracts (LEE) and water extracts (LWE) of 13 batches of licorice medicinal materials in Xinjiang, Gansu and inner Mongolia. The correlation coefficients for the 13 LEE and LWE samples are shown in tables 1 and 2, respectively. When the correlation coefficient is close to 1.0, it indicates that the samples have complete similarity. The correlation coefficient is low, which indicates that the mathematical quality is poor when determining the relationship between the licorice samples. Thus, heterogeneity in similarity indicates that different extraction processes and diversity affect the quality of licorice samples.

TABLE 113 evaluation results of similarity of Glycyrrhiza alcohol extract samples

TABLE 213 evaluation results of the similarity of the Glycyrrhiza aqueous extract samples of batches

The result shows that the similarity of the fingerprint of the alcohol extract is more than 0.84, and the similarity of the fingerprint of the water extract is lower (more than 0.80). As can be seen from Table 1, the correlations between S1 and S3, S1 and S4, S11 and S7, S11 and S9, and S11 and S10 are higher than those of other origins of licorice. The results show that the content of

peaks

3, 4, 6, 14, 20 of S1 and S11 is significantly higher than the other peaks, possibly affecting the quality of the LEE sample. In contrast, the correlations between S3 and S6, S3 and S8 were lower than those of the other batches. The contents of the

peaks

4 and 20 of the S3 are obviously lower than those of the peaks S6 and S8. Therefore, characteristic peaks such as

peaks

3, 4, 6, 14, 20 and the like have important significance on the quality control of the LEE fingerprint.

As can be seen from Table 2, the correlations between S4 and S1, S4 and S5, and S11 and S7 are also higher than those of other origins of Glycyrrhiza uralensis. The results show that the

peaks

5, 6, 7, 11, 19 of S4 and S11 are significantly higher in content than the other peaks, and may affect the quality of the LWE samples. In contrast, the correlations between S3 and S6, S3 and S7, S3 and S13 are lower than those of other batches. The peak 6 content of S3 was significantly lower than S6, S7 and S13. Therefore, the

characteristic peaks

5, 6, 7, 11 and 19 have important significance on the quality control of the LWE fingerprint.

2.2 Hierarchical Cluster Analysis (HCA)

HCA is a multivariate analysis technique that groups samples according to their internal characteristics. The invention adopts SIMCA14.1 software to carry out HCA, and calculates the similarity by the square of the hierarchical distance and the Euclidean distance. The degree of correlation between samples depends on the distance in the tree, with the shortest distance representing the highest degree of relationship. Thus, these objects are treated as attributes of the same group. This figure is shown as a tree diagram, which serves as a tool to explain the closest distance between clusters of samples. Through the classification of different batches of licorice samples, some quality evaluation indexes with statistical significance are screened out. Hierarchical clustering analysis results show that 13 batches of LWE samples are divided into two categories, the liquorice produced in Xinjiang is divided into a category I, one group is S1 and S2, and the other group is S3, S4 and S5. Gansu and inner Mongolia liquorice are divided into II groups, S11 and S13 are one group, and S6-S10 and S12 are one group (fig. 5). LEE samples are divided into two main classes, wherein S1-S5 are I classes, and S6-S13 are II classes. Gansu licorice root, S6, S7, S8, S9 and S10, and inner Mongolia licorice root, S11, S12 and S13, are in one group (FIG. 6).

2.3 Principal Components Analysis (PCA)

Principal component analysis is used to alleviate the multiple collinearity problem and explore the relationship between the independent variables. As a traditional and typical data analysis technique, it is a major component of many fingerprint studies. In addition, the licorice fingerprint needs to be evaluated to determine whether the licorice fingerprint can effectively group samples in different regions and identify a representative peak in the chromatogram. Therefore, the invention uses the correlation between the multivariate analysis score obtained by Principal Component Analysis (PCA) in SIMCA14.1 software and the load to describe the relationship between the feature markers (25 feature markers in LEE sample, 23 feature markers in LWE sample) selected by the similarity evaluation software and 13 batches of licorice samples from different sources and different extraction methods. In principal component analysis, a large amount of data is replaced by a Principal Component Score (PCS), which contains the same information as the original data. And (3) calculating the average central relative peak area of the common peak of the liquorice sample to generate a main component analysis. The first principal component (PC1) contains the largest variance in the data, while the second principal component (PC2) represents the largest variance unexplained by PC 1. According to the Principal Component Analysis (PCA) score plot of LEE (fig. 7), PC1 and PC2 account for 45.2% and 29.7%, respectively, of the total variation of the original observations, accounting for 75.1% of the total variation. Similarly, in the PCA score plot of LWE (fig. 8), PC1 and PC2 account for 67.8% and 14.3%, respectively, of the total variation of the original observations, and for 82.1% of the total variation.

As shown in the LEE principal component analysis score chart (fig. 7), three groups can be distinguished depending on the sample source. S2, S3 and S5 can be easily grouped, with S1 being located near S4 and their chemical compositions being related to each other. Thus, S2, S3, S5, S1 and S4 tend to fall into the same class according to the similarity of their chemical compositions. S8, S9, S10, S7 and S6 are easily grouped, and their chemical compositions are related to each other. Meanwhile, S11 is adjacent to S13, and therefore, S12, S11 and S13 tend to be grouped into the same cluster according to the similarity of their chemical compositions. Also, the score plot (fig. 8) shows that the 13 LWE samples are divided into three groups. Clustering analysis was performed on S1 and S2 with positive PC2 values and on S3, S4 and S5 with negative PC1 values. Thus, S1, S2, S3, S4 and S5 tend to fall into the same class according to the similarity of their chemical compositions. And S6, S7, S11, S12, S13 were clustered by PC1 positive values. According to PC2, S3, S4, S5 and S8, S9, S10 are clearly separated. S8, S9, S10 were clustered with negative values for PC 2. These groups are also consistent with the results of some of the relationships and geographic differences described by HCA between licorice origins.

FIGS. 9 and 10 show that the load graph of principal component analysis is important for evaluating the quality of 13 licorice samples. The loading plots show multivariate effects on the differences between samples, and it can be seen that these common peaks have an effect on the quality of licorice. From the LEE load graph (fig. 9), peak 14, peak 8, peak 15, peak 4 and peak 3 have the greatest impact on the quality of the geographic differences. In addition, peak 6 and peak 20 also had an effect on the quality evaluation. Finally, 7 major compounds were screened according to statistical data. Also, from the load graph of LWE (fig. 10), peak 10, peak 14, peak 21, peak 7, and peak 19 have the greatest impact on the quality of geographic discrepancies. Peak 11 and peak 5 also had an effect on the quality assessment. Finally, 7 major compounds were selected from the population according to statistical data. The similarity evaluation result is consistent with HCA and PCA, and can be used for the quality evaluation of the licorice fingerprint.

2.4 orthogonal partial least squares discriminant analysis (OPLS-DA)

Orthogonal partial least squares discriminant analysis (OPLS-DA) is a supervised classification technique that improves interpretability. From the LEE load S-plots diagram (FIG. 11), peak 14, peak 19, peak 10 have the greatest quality impact on geographic variation; from the S-plots plot of load for LWE (fig. 12), peaks 10, 11 have the greatest impact on the quality of geographic discrepancies. However, peaks 19, 15, 8 in LEE and 21, 14, etc. in LWE do not provide a resolvable maximum wavelength representing the ultraviolet spectrum of the unknown compound, which is the presence of the unknown compound. Finally, by combining SA, HCA and PCA analysis, the important functions of apioside liquiritin, apioside isoliquiritin, echinacolone, glycyrrhizic acid and liquiritigenin in LEE quality evaluation are clarified. Similarly, apigenin, liquiritin, apigenin isoliquiritin, echinocandin, licochalcone B, glycyrrhizic acid also play an important role in the quality evaluation of LWE.

3. Identification and quantitative analysis of chemical components

3.1 identification of Components

The mass spectrometer detector was equipped with an electrospray ionization (ESI) source. The ESI sources operate simultaneously in positive and negative polarity. The optimal mass spectrum parameters are as follows: the capillary temperature is 270 ℃, and the evaporation temperature is 300 ℃; the gas flow of the sheath layer is 35 mu L/min, the auxiliary gas flow is 5 mu L/min, the full scan analysis is carried out, the Q-TOF acquisition rate is 0.1s, and the mass range is 50-1200 m/z. Data-dependent acquisition was also performed in order to obtain a complete molecular fragmentation pattern.

In general, compounds can be identified by comparison of retention time, MS and MS/MS cleavage pattern with a control. In addition, the invention also utilizes a novel data acquisition post-processing software UNIFI to realize the automatic, rapid and accurate qualitative analysis of the chemical components of the liquorice. By these methods, unknown signature compounds (25 signatures in LEE and 23 signatures in LWE) were screened from the SA and HCA results. The characteristic marker compounds have obvious influence on the quality evaluation of the liquorice, and highlight the difference of the liquorice components under different extraction methods. The mass spectral characteristics and identification of these compounds are shown in tables 3 and 4.

TABLE 3 MS data for characteristic components of alcohol extract of Glycyrrhiza glabra

TABLE 4 characteristic composition MS data of aqueous extract of Glycyrrhiza

The fragment ion identification is illustrated by using Violanthin (pansy flavonoid glycoside) and Liquiritin apioside (apigenin glycoside) as examples. As can be seen in FIG. 13, the exact mass value of compound (m/z 577.1563[ m-H ] -) and fragment ion (m/z 415.1024[ m-H-Glc ] -), molecular weight and chemical formula demonstrate the presence of pansy flavonoid glycosides. Furthermore, the error between the experimental and theoretical mass values is within the allowable range, and it is clear that the fragment ion peak (m/z 415.1024) is from the parent ion peak [ m-H ] + (577.1563). In addition, there are also exact mass data (m/z 551.1757, [ m + H ] +), (m/z 549.1648, [ m-H ] -) and fragment ions (m/z 257.0828), a known compound, named apigenin, and thus, peak 3 was initially determined.

3.2 quantitative analysis

3.2.1 Linear relationship

Accurately sucking appropriate amount of each reference substance solution under item "1.2" to prepare a series of mixed reference substance solutions which are continuously diluted. The peak areas were recorded according to the chromatographic conditions under "1.5". Linear regression was performed with peak area as ordinate and mass concentration as abscissa. The standard curve and related information for each component are shown in table 5.

TABLE 5 Standard curves for 7 ingredients in licorice

3.2.2 degree of accuracy, repeatability, stability

The fine density calculation comprises taking the same batch of Glycyrrhrizae radix samples, preparing LWE and LEE according to item "1.3", continuously sampling for 6 times according to item "1.4" chromatographic condition, and determining fingerprint of LWE and LEE. The Relative Standard Deviation (RSD) is less than 3.0 percent, which indicates that the precision of the instrument meets the requirement.

And performing repeated calculation on 6 samples in the same batch respectively, preparing LWE and LEE under the same condition, injecting samples respectively, and recording chromatographic peaks. Taking the same batch of licorice samples, preparing LWE and LEE according to item 1.3, and determining fingerprint spectra according to chromatographic conditions under item 1.4 at 0, 2, 4, 8, 12 and 24 hours. RSD were all less than 3.0%, indicating that the licorice sample solution was stable within 24 hours, and the results are summarized in table 6.

TABLE 6 precision, reproducibility, stability of 7 ingredients in licorice

3.2.3 quantitative Studies

Finally, 6 chemical components with statistically significant components were selected as the main components of LWE and LEE, respectively. The comprehensive exploratory analysis method is suitable for application of 13 batches of liquorice with different producing areas and selection of an extraction method. Meanwhile, the content of 7 characteristic compounds under different extraction methods (5 chemical components are the same in LEE and LWE, and 1 chemical component is different from each other) was determined, and summarized in tables 7 and 8. The content of 6 compounds in the extracts of different producing areas is inconsistent, which indicates that different cultivation conditions, such as climate, sunshine of different producing areas and the like, may affect the quality of the liquorice. In addition, the contents of 6 components in LEE and LWE are different, which shows that the extraction process has significant influence on the extraction of the effective components of licorice. Taking the areas of Xinjiang and inner Mongolia as examples, the content of the triterpenes is obviously higher than that of the areas of Gansu, and the content of the LEE characteristic components is obviously higher than that of LWE. In addition, the types of the effective components of the liquorice obtained by different extraction methods are different, for example, licochalcone B exists in LWE but not in LEE, which shows that the different extraction methods have obvious influence on the dissolution of the effective components of the traditional Chinese medicine.

TABLE 7 contents of characteristic components of 13 batches of alcohol extracts of licorice in different production areas

TABLE 8 content of characteristic components of 13 batches of aqueous extracts of licorice in different producing areas

4. Discriminant analysis

Discriminant analysis is a multivariate statistical analysis method for determining a classification of a type of a study object from various feature values of the study object under a condition of determining the classification. In the research, U ═ X1, X2, … and X13 structural domains representing 13 licorice samples are established, and the contents of 7 characteristic peaks such as echinacoside, glycyrrhizic acid and apioside liquiritin are selected as discrimination factors to form a 13 × 7 matrix as shown in formula (1).

Xi＝(Xi1,Xi2,...,Xi7) (1)

In the formula: i is a sample number; x is a sample.

And (4) carrying out stepwise discriminant analysis by using SPSS 22.0 software and adopting a Walk's lambda method to obtain a discrimination function of the liquorice producing area. And evaluating the advantages and the disadvantages of the discriminant function by a back-substitution estimation method. Finally, the discriminant function equation for LEE is as follows (S1: apigenin, S2: liquiritin, S3: apigenin, S4: glycyrrhizin, S5: echinocandin, S6: glycyrrhizic acid):

Similarly, the discriminant function equation for the aqueous extract LWE of Glycyrrhiza uralensis was obtained as follows (S1: apigenin, S2: glycyrrhizin, S3: apigenin, S4: Acyrrhizachalcone, S5: Acyrchalcone B, S6: glycyrrhizic acid):

And substituting the content of each characteristic peak after screening into a function equation, and comparing the Y values of the function equations of different producing areas, wherein the sample belongs to the source producing area represented by the equation if the Y value is maximum. By checking the back substitution estimation method, we checked another 10 batches of licorice with known origin, and compared the discriminant analysis of LEE and LWE origin with the actual results, with 90% and 80% accuracy (tables 9 and 10), respectively. This shows that the established discriminant function equation is relatively stable, and the liquorice producing area can be predicted and identified.

TABLE 9 evaluation of the discriminant function equation for alcohol extracts of Glycyrrhiza uralensis

TABLE 10 evaluation of the discriminant function equation for aqueous extracts of Glycyrrhiza

Claims

1. A liquorice origin distinguishing method based on UPLC fingerprint spectrum and stoichiometry is characterized by comprising

The method comprises the following steps:

y xinjiang =7.629S₁+3.029S₂-0.404S₃+663.007S₄+1765.007S₅+0.537S₆-82.594；

Y gansu =8.026S₁+3.36S₂-1.452S₃+830.744S₄+1857.135S₅+0.619S₆-95.093；

Y inner mongolia =7.646S₁+3.706S₂-0.489S₃+649.922S₄+1793.914S₅+0.599S₆-94.193；

y xinjiang =3.783S₁+2.731S₂-0.137S₃+2418.876S₄-131.472S₅+0.697S₆-44.812；

Y gansu =3.698S₁+3.163S₂-1.26S₃+3107.573S₄-327.356S₅+0.792S₆-49.803 ；

Y inner mongolia =3.423S₁+3.539S₂-1.159S₃+3713.539S₄-500.921S₅+0.957S₆-58.480 ；

2. The method for distinguishing the production area of liquorice based on UPLC fingerprint and stoichiometry as claimed in claim 1, wherein in step (1), the preparation method of the liquorice alcohol extract comprises the following steps:

precisely weighing Glycyrrhrizae radix powder, precisely adding 30-50 times of 60-80% ethanol, weighing, ultrasonically extracting for 20-40min, cooling, weighing again, adding 60-80% ethanol for supplementing weight loss, shaking, standing, filtering with 0.22 μm organic filter membrane, and collecting filtrate.

3. The method for distinguishing the production area of liquorice based on UPLC fingerprint and stoichiometry as claimed in claim 1, wherein in step (1), the preparation method of the liquorice aqueous extract comprises the following steps:

weighing Glycyrrhrizae radix, decocting with 12-16 times of water for 1-2 hr, pouring out extractive solution, decocting with 6-10 times of water for 0.5-1 hr, mixing the decoctions, diluting to content of 0.025g per 1mL, filtering with 0.22 μm water system filter membrane, and collecting filtrate.

4. The method for distinguishing the production area of liquorice based on UPLC fingerprint and stoichiometry as claimed in claim 1, wherein in the step (1) and the step (3), the chromatographic conditions in the ultra high performance liquid chromatography are as follows:

a C18 column; the detection wavelength is 254 nm; taking acetonitrile as a mobile phase A and 0.1% formic acid aqueous solution as a mobile phase B, and performing gradient elution, wherein the gradient elution procedure is as follows: 0-2 min, 80-75% B; 2-4 min, 75-73% B; 4-12 min, 73-60% B; 12-18 min, 60-40% B; 18-20 min, 40-80% B; 20-25 min, 80% B; the column temperature is 30 ℃; the flow rate was 0.3ml/min and the amount of sample was 1. mu.L.

5. The method for distinguishing liquorice producing areas based on UPLC fingerprint and stoichiometry as claimed in claim 1, wherein in step (3), the mass spectrum conditions are as follows: an electrospray ionization ESI source, wherein the ESI source works at the positive polarity and the negative polarity simultaneously, the temperature of a capillary is 270 ℃, and the evaporation temperature is 300 ℃; the gas flow of the sheath layer is 35 mu L/min, the auxiliary gas flow is 5 mu L/min, the full scan analysis is carried out, the Q-TOF acquisition rate is 0.1s, and the mass range is 50-1200 m/z.