Peking white pear producing area identification method based on metabonomics technology

Abstract

The invention provides a metabonomics technology-based Beijing white pear producing area identification method, which comprises the following steps: collecting Beijing white pear samples from different producing areas, detecting metabolites of the Beijing white pear samples from different producing areas by using an ultra-high performance liquid chromatography-tandem time-of-flight mass spectrometry method, analyzing to obtain differential metabolites among the Beijing white pears from different producing areas, and determining the production area of the Beijing white pears from different producing areas. And establishing a producing area distinguishing model based on orthogonal partial least squares discriminant analysis (OPLS-DA), and judging the producing area of an unknown sample according to the model. The method has the advantages of high accuracy, high sensitivity and high flux, can accurately distinguish products from small-range adjacent production places, and provides a simple, convenient and accurate solution for solving the problem that geographical indication products and other local characteristic products are pretended by the adjacent production places; and technical support is provided for protection of local characteristic products such as landmark products.

Description

Technical Field

[0001] The invention belongs to the technical field, and particularly relates to a method for identifying the origin of Beijing white pear based on metabolomics technology. Background Art

[0002] Jingbai Pear, also known as Beijing White Pear, is a specialty of Mentougou District, Beijing. It is a Chinese national geographical indication product and one of the best varieties in the Qiuzi Pear system. On March 13, 2012, the former General Administration of Quality Supervision, Inspection and Quarantine approved the implementation of geographical indication product protection for Jingbai Pear. Jingbai Pear is juicy, sweet and sour, with a strong aroma and high sugar content. The pear pulp contains a large amount of glucose, vitamins, amino acids and other nutrients. It has the functions of promoting body fluid, relieving heat, moisturizing, quenching thirst, clearing heat, resolving phlegm, strengthening the stomach, and sobering up. It is a health product suitable for all ages. Due to the small production area and large demand of Jingbai Pear, it has formed a clear price advantage over other varieties of pears.

[0003] At present, commonly used technologies for origin identification include stable isotope mass spectrometry, mineral element analysis, organic component analysis and spectral analysis technology, but these technologies are more suitable for distinguishing products from distant origins. Due to the long geographical distance and large differences in the origin environment, the differences in various indicators between products are relatively obvious, making it easier to achieve accurate distinction. However, for products from neighboring origins, especially between neighboring provinces and cities, since the origin environments are more similar, the quality differences of the products are relatively small, and the distinction effect of existing technologies is not ideal. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method for identifying the origin of Beijing white pear based on metabolomics technology.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for identifying the origin of Jingbai pear based on metabolomics technology comprises the following steps:

[0007] S1. Collect samples of Beijing white pear from different origins, extract them by ultrasonic using methanol-water solution, and then centrifuge to obtain the supernatant;

[0008] S2. Use ultra-high performance liquid chromatography-tandem time-of-flight mass spectrometry to detect the extract of each sample, and take an equal volume of the supernatant of each sample and mix them to prepare a quality control sample, recorded as QC. Insert a QC sample every few samples. QC is used to examine the stability of the instrument;

[0009] S3. Extract peaks, align peaks, and correct retention times of the raw data detected by the mass spectrometer; correct the peak area, and filter the peaks with a missing rate of >50% in each group of samples; obtain the identification information of metabolites by searching the public library and metDNA method after correction of the filtered peaks;

[0010] S4. Perform statistical analysis on the metabolites obtained using R program to obtain significantly different metabolites, and use the mass spectrometry response values ​​of the significantly different metabolites to construct a production area differentiation model;

[0011] S5. Extract and measure the unknown samples according to steps S1-S3 to obtain the information of significantly different metabolites, bring the corresponding mass spectrometry response values ​​into the model constructed in S4, and determine the origin of the sample according to the model sample area into which the sample falls.

[0012] As described above, preferably, in step S1, the methanol-water solution is a methanol-water solution with a volume concentration of 75% as the extract, and ultrasonic extraction is performed in an ice water bath, followed by centrifugation at 12000r / min for 10min at 4°C, and the supernatant is taken and filtered through a 0.22μm filter membrane for sample testing.

[0013] In the method as described above, preferably, in step S2, the mobile phase A of the liquid chromatography is 0.1% formic acid-water, the mobile phase B is 0.1% formic acid-acetonitrile, and gradient elution is adopted;

[0014] The conditions of the mass spectrometer are as follows: the ionization voltage is 2500V for positive ions and 1500V for negative ions; the positive and negative ion voltages of the curtain gas are both 10Psi; the positive and negative ion voltages of the collision gas are both 12Psi; the nebulization temperature is 325°C for positive ions and negative ions; and the positive and negative ion voltages of the nebulizer gas are both 40Psi.

[0015] As described above, preferably, in step S3, the raw data is converted into mzML format by ProteoWizard MSConvert software; the peak extraction adopts XCMS program; and the peak area correction adopts "SVR" method.

[0016] In the method as described above, preferably, in step S4, the R program includes univariate statistical analysis and multivariate statistical analysis, the univariate statistical analysis obtains candidate differential metabolites by Student's t-test and differential fold analysis, and the multivariate statistical analysis screens candidate differential metabolites by orthogonal partial least squares discriminant analysis (OPLS-DA);

[0017] The mass spectrometric response values ​​of the significantly different metabolites were analyzed by orthogonal partial least squares discriminant analysis (OPLS-DA) to establish a production area classification model.

[0018] As described above, preferably, the screening criteria for candidate differential metabolites are differences in metabolite expression abundance, and those with a difference multiple greater than or equal to 2 and less than or equal to 0.5 and a corrected significance level p value less than 0.05 are selected as candidate differential metabolites.

[0019] According to the method described above, preferably, the candidate differential metabolites simultaneously satisfy the projection variable importance value greater than 1.

[0020] As described above, preferably, in order to verify the effectiveness of the candidate differential metabolites screened for model construction, the R program is used to perform receiver operating characteristic curve analysis on the candidate differential metabolites, thereby obtaining the area between the ROC curve and the horizontal axis corresponding to each differential metabolite, which is the area under the curve AUC, and the candidate differential metabolites with an AUC value greater than 0.8 are screened as significant differential metabolites for model construction.

[0021] The beneficial effects of the present invention are:

[0022] The present invention provides a method for identifying Beijing white pears from different origins. The method uses ultra-high performance liquid chromatography tandem time-of-flight mass spectrometry to perform non-targeted metabolomics analysis on Beijing white pears from different origins, and uses an OPLS-DA analysis method to construct an origin identification model, which can clearly distinguish Beijing white pear samples from Mentougou District, Beijing from samples from Fangshan District, Beijing, Changli County, Hebei, and Gu'an County, Hebei. The present invention has the advantages of high accuracy, high sensitivity, and high throughput, and can accurately distinguish products from a small range of neighboring origins, providing a simple and accurate solution to the problem of geographical indication products and other local specialty products being impersonated by products from neighboring origins, and providing technical support for the protection of local specialty products such as landmark products. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 TIC overlap graph for mass spectrometry detection of QC samples;

[0024] Figure 2 This is the result of distinguishing samples from Beijing Mentougou District and Beijing Fangshan District;

[0025] Figure 3 This is the result of distinguishing samples from Mentougou District, Beijing, and Changli County, Hebei;

[0026] Figure 4 This is the result of distinguishing samples from Mentougou District, Beijing, and Gu'an County, Hebei;

[0027] Figure 5 This is the OPLS-DA model validation diagram for samples from Beijing Mentougou District and Beijing Fangshan District;

[0028] Figure 6 This is the OPLS-DA model validation diagram for samples from Mentougou District, Beijing and Changli County, Hebei;

[0029] Figure 7 This is the OPLS-DA model validation diagram for the samples from Mentougou District, Beijing and Gu'an County, Hebei. DETAILED DESCRIPTION

[0030] The metabolomics technology based on high-resolution mass spectrometry of the present invention separates and detects small molecule metabolites in samples, and then uses bioinformatics and statistical analysis methods to analyze the composition and concentration changes of metabolites, thereby accurately reflecting the metabolic conditions in different groups of samples. Even for plant samples of the same variety, differences in the production environment, artificial management methods, etc. will affect the quality of the product, which is reflected in the differences in the quality of products from different production areas. The metabolomics technology based on high-resolution mass spectrometry can accurately identify these tiny differences, thereby realizing the origin identification of products of the same variety from adjacent production areas.

[0031] The following examples are used to further illustrate the present invention, but should not be construed as limiting the present invention. Without departing from the spirit and substance of the present invention, modifications or substitutions made to the present invention all belong to the scope of the present invention.

[0032] Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art. Unless otherwise specified, the reagents used in the examples are of analytical grade or above.

[0033] Example 1

[0034] 1. Collection of Jingbai pear samples from different origins

[0035] Fresh ripe pear samples were collected in Beijing Mentougou District (MT), Beijing Fangshan District (FS), Hebei Changli County (CL), and Hebei Gu'an County (GA) to ensure their maturity, uniform size, and freedom from disease. Eight representative samples were collected from each production area, brought back to the laboratory, and stored in a -80°C refrigerator.

[0036] 2. Sample extraction

[0037] After the sample is taken out, it is thawed on ice, 200 mg is weighed into a centrifuge tube, 1 ml of 75% methanol-water is added as the extract, ultrasonicated in an ice water bath for 10 min, vortexed for 1 min after being taken out, centrifuged at 12000 r / min for 10 min at 4°C, the supernatant is taken, filtered through a 0.22 μm filter membrane, and set aside.

[0038] 3. Metabolite analysis

[0039] Ultra-high performance liquid chromatography tandem time-of-flight mass spectrometer (QTOF / MS-6545, Aglient) was used for non-targeted metabolomics analysis. Waters ACQUITY UPLC HSS T3 C18 (1.8 μm, 2.1 mm*100 mm) column was selected, mobile phase A was 0.1% formic acid-water, mobile phase B was 0.1% formic acid-acetonitrile, gradient elution was used, and the mobile phase setting ratio was shown in Table 1. Column temperature was 40 °C, injection volume was 2 μL, and flow rate was 0.40 mL / min. Mass spectrometry acquisition conditions were set, as shown in Table 2. A certain volume of samples to be tested from all origins were mixed and then prepared to prepare quality control samples (QC) for testing the stability of instrument operation. During the instrument analysis, a QC sample was inserted for every 8 test analysis samples.

[0040] Table 1 Column mobile phase gradient conditions

[0041] Time (min) Flow rate (mL / min) A(％) B(％) 0 0.4 95 5 11 0.4 10 90 12 0.4 10 90 12.1 0.4 95 5 14 0.4 95 5

[0042] Table 2 Mass spectrometry conditions in positive and negative ion modes

[0043] parameter Positive ions Negative ions Ionization voltage 2500V 1500V Air curtain air 10Psi 10Psi Collision gas 12Psi 12Psi Atomization temperature 325℃ 325℃ Atomizing gas 40Psi 40Psi

[0044] 4. Data processing

[0045] The raw data from the mass spectrometer were converted into mzML format using ProteoWizard MSConvert software, and peak extraction, peak alignment, and retention time correction were performed using the XCMS program. The peak area was corrected using the "SVR" method, and peaks with a missing rate of >50% in each group of samples were filtered. After correction and screening, metabolite identification information was obtained by searching the public library and metDNA method.

[0046] 5. Data analysis

[0047] 5.1 Quality control sample analysis

[0048] The total ion current (TIC) graphs of mass spectrometry analysis of all QC samples were overlapped and analyzed, and the results were as follows: Figure 1 The results show that the curves of total ion current for metabolite detection have high overlap, that is, the retention time and peak intensity are consistent, indicating that the signal stability is good when the mass spectrometer detects the same sample at different times. The high stability of the instrument provides an important guarantee for the repeatability and reliability of the data.

[0049] 5.2 Screening of differential metabolites

[0050] In order to obtain the differentially expressed metabolites between the samples from Mentougou area and the other three production areas, we first used the R program to perform univariate statistical analysis including Student's t-test and fold change analysis. The samples from Mentougou area were used as the control samples. According to the difference in metabolite expression abundance, the fold change greater than or equal to 2 and less than or equal to 0.5 and the corrected significance level (p value) less than 0.05 were used as screening criteria to obtain differential metabolites.

[0051] Then, multivariate statistical analysis was performed using the R program. Orthogonal partial least squares discriminant analysis (OPLS-DA) was performed on the metabolites obtained from the samples of Mentougou District and Fangshan District, Mentougou District and Changli County, and Mentougou District and Gu'an County. The differential metabolites were further screened according to the projection variable importance (VIP) value of the metabolites in the model, and the differential metabolites with VIP values ​​greater than 1 were selected as candidate differential metabolites for the construction of the origin model.

[0052] In summary, we obtained an overall list of candidate metabolites that were differentially expressed between samples from Mentougou and the other three production areas, as shown in Table 3.

[0053] Table 3 Statistical table of the number of candidate differential metabolites between samples from Mentougou and the other three production areas

[0054] Group Total number of differential metabolites Down-regulated metabolites Upregulated metabolites FS_vs_MT 48 14 34 CL_vs_MT 219 109 110 GA_vs_MT 133 37 96

[0055] In order to further verify the effectiveness of the candidate differential metabolites screened for model construction, the R program was used to perform receiver operating characteristic curve (ROC curve) analysis on the candidate differential metabolites, thereby obtaining the area between the ROC curve and the abscissa corresponding to each candidate differential metabolite, which is the area under the curve (AUC). AUC is a quantitative evaluation index of the ROC curve. The value range of AUC is (0.5, 1]. The closer it is to 1, the better the prediction effect of the model established using the metabolite. Therefore, the present invention uses metabolites with an AUC value greater than 0.8 as significant differential metabolites for model construction. Table 4 shows the differential metabolites used for model construction between different groups and the corresponding AUC values.

[0056] Table 4-1 Differential metabolites and AUC values ​​between groups in Mentougou and Fangshan districts

[0057] serial number Metabolites AUC value 1 Sucrose 0.953 2 (R)-S-Lactoylglutathione 0.969 3 Cytidine 0.875 4 3-Methylpyrrole-2,4-dicarboxylic acid 0.906 5 Prosapogenin A 0.891 6 DL-O-tyrosine 0.891 7 3-hydroxycinnamic acid 0.875 8 Pullulan 0.875 9 Trp-Pro-Asp 0.891 10 Thr-Phe-Leu-Thr-Gln 0.891 11 PE-NMe2(14:0 / 20:1(11Z)) 0.844 12 PC(18:1(11Z) / 15:0) 0.859 13 Leu-Leu-Trp 0.953 14 Avenestergenin A2 0.891 15 Arg-His-Thr 0.875 16 Ala-Ala-Gly 0.906 17 Quinic Acid 0.922 18 Estradiol acetate 0.859 19 L-Citrulline 0.906 20 Anthranoyllycoctonine 0.953 21 Thr-Ser-Phe-Asp 0.875

[0058] Table 4-2 Differential metabolites and AUC values ​​between the groups in Mentougou District and Changli County

[0059]

[0060]

[0061] Table 4-3 Differential metabolites and AUC values ​​between the groups in Mentougou District and Gu'an County

[0062] serial number Metabolites AUC value 1 Prunasin 0.906 2 N-Octadecyl-N'-propylsulfamide 0.922 3 Cytidine 0.922 4 Loganin 1 5 D-Ornithine 0.953 6 L-Cysteine-glutathione disulfide 0.922 7 Succinic Acid 0.906 8 cis-Ferulic acid[arabinosyl-(1->3)-[glucosyl-(1->6)]-glucosyl]ester 0.953 9 Ser-Tyr-Gln 0.922 10 Ser-Asn-Thr-Ile-Leu 0.938 11 Phosphatidylserine 0.938 12 Met-Glu-His 0.938 13 LysoPA(i-21:0 / 0:0) 0.984 14 Leu-Leu-Trp 0.938 15 Curcumenol 0.984 16 Asn-His-His 1 17 Anhydrorhodovibrin 0.969 18 Quinic Acid 1 19 D-Threonic Acid 0.969 20 Thr-Val-Ala-Lys 0.984

[0063] 5.3 Establishment and verification of production area classification model

[0064] The R program was used to construct an origin identification model based on orthogonal partial least squares discriminant analysis (OPLS-DA) using the mass spectrometry response values ​​of the significantly different metabolites screened out by 5.2. Figure 2-Figure 4 , which are the origin identification models of samples from Beijing Mentougou District and Beijing Fangshan District, Changli County, Hebei Province, and Gu'an County, Hebei Province. It can be seen that in each model, the samples from each origin are clustered together and are far apart without overlap, indicating that the OPLS-DA model can be used to distinguish the samples from Beijing Mentougou District and the other three origins, including Beijing Fangshan District, Changli County, Hebei Province, and Gu'an County, Hebei Province, and thus provide technical support for the protection of the geographical indication agricultural product - Beijing Mentougou Jingbai Pear.

[0065] To ensure the reliability of the model, we used 200 permutation tests, that is, the model conducted 200 random permutation and combination experiments on all samples in the two comparison groups to verify the grouping ability of the model. The horizontal axis represents the model accuracy, and the vertical axis is the frequency of the model classification effect. 2 Y represents the model's ability to explain the grouping, R 2 The closer Y is to 1, the more accurate the model grouping effect is. 2 Indicates the model's predictive ability for unknown samples, Q 2 A value greater than 0.5 is considered to be an effective model. 2 When >0.9, it is an excellent model. The verification results are as follows Figure 5-Figure 7 As shown, the three verification models R 2 Y is 0.990, 0.996, 0.994, and Q 2 They are 0.921, 0.923, and 0.802 respectively, indicating that all models are valid models.

[0066] 5.4 Unknown Sample Verification

[0067] When there are unknown samples that need to be identified, first perform metabolite extraction and determination according to the above steps 2 and 3 to obtain the mass spectrometry response values ​​of the significantly different metabolites used to build the model, and bring the corresponding values ​​into each model. According to the model area where the sample falls, the origin of the sample can be determined. The experiment collected 8 samples, and after metabolite extraction and analysis, the mass spectrometry response values ​​of the significantly different metabolites were obtained, which were brought into the three models accordingly. The determination results are shown in Table 5.

[0068] Table 5 Model prediction results of unknown samples

[0069] Sample No. Origin Accuracy X1 MT 98.0％ X2 MT 97.9％ X3 FS 98.2％ X4 FS 98.4％ X5 CL 98.1％ X6 CL 98.2％ X7 GA 98.9％ X8 GA 99.1％

[0070] The eight samples were respectively judged to be from Beijing Mentougou District, Beijing Fangshan District, Hebei Changli County, and Hebei Gu'an County, and the discrimination accuracy rates were all greater than 97%, indicating that the model constructed by the method of the present invention has a high accuracy rate.

Claims

1. A method for identifying the origin of Beijing white pear based on metabolomics technology, characterized in that: It includes the following steps: S1. Collect samples of Beijing white pear from different origins, extract them by ultrasonic using methanol-water solution, and then centrifuge to obtain the supernatant; S2. Use ultra-high performance liquid chromatography-tandem time-of-flight mass spectrometry to detect the extract of each sample, and take an equal volume of the supernatant of each sample and mix them to prepare a quality control sample, recorded as QC. Insert a QC sample every few samples. QC is used to examine the stability of the instrument; S3. Extract peaks, align peaks, and correct retention times of the raw data detected by the mass spectrometer; correct the peak area and filter the peaks with a missing rate of >50% in each group of samples; obtain metabolite identification information by searching the public library and metDNA method after correction of the filtered peaks; and establish a production area classification model; S4. Perform statistical analysis on the metabolites obtained using R program to obtain significantly different metabolites, and use the mass spectrometry response values ​​of the significantly different metabolites to construct a production area differentiation model; S5. Extract and measure the unknown samples according to steps S1-S3 to obtain the information of significantly different metabolites, bring the corresponding mass spectrometry response values ​​into the model constructed in S4, and determine the origin of the sample according to the model sample area into which the sample falls.

2. The method according to claim 1, characterized in that In step S1, a methanol-water solution with a volume concentration of 75% methanol-water solution is used as the extract, and ultrasonic extraction is performed in an ice water bath. After that, the solution is centrifuged at 12000r / min for 10min at 4°C, and the supernatant is taken and filtered through a 0.22μm filter membrane for sample detection.

3. The method according to claim 1, characterized in that In step S2, the mobile phase A of the liquid chromatography is 0.1% formic acid-water, the mobile phase B is 0.1% formic acid-acetonitrile, and gradient elution is adopted; The conditions of the mass spectrometer are as follows: the ionization voltage is 2500V for positive ions and 1500V for negative ions; the positive and negative ion voltages of the curtain gas are both 10Psi; the positive and negative ion voltages of the collision gas are both 12Psi; the nebulization temperature is 325°C for positive ions and negative ions; and the positive and negative ion voltages of the nebulizer gas are both 40Psi.

4. The method according to claim 1, characterized in that: In step S3, the raw data were converted into mzML format using ProteoWizardMSConvert software; peak extraction was performed using the XCMS program; and peak area correction was performed using the "SVR" method.

5. The method according to claim 1, characterized in that In step S4, the R program includes univariate statistical analysis and multivariate statistical analysis. The univariate statistical analysis obtains candidate differential metabolites through Student's t-test and differential fold analysis. The multivariate statistical analysis screens candidate differential metabolites through orthogonal partial least squares discriminant analysis. The mass spectrometric response values ​​of the obtained significantly different metabolites were used to construct a production area differentiation model based on orthogonal partial least squares discriminant analysis.

6. The method according to claim 5, characterized in that The screening criteria for candidate differential metabolites were differences in metabolite expression abundance, and those with a difference fold greater than or equal to 2 and less than or equal to 0.5 and a corrected significance level p value less than 0.05 were selected as candidate differential metabolites.

7. The method according to claim 5, characterized in that The significantly different metabolites also satisfy the projection variable importance value greater than 1.

8. The method according to claim 5, characterized in that In order to verify the effectiveness of the candidate differential metabolites screened for model construction, the R program was used to perform receiver operating characteristic curve analysis on the candidate differential metabolites, thereby obtaining the area between the ROC curve and the horizontal axis corresponding to each candidate differential metabolite, which is the area under the curve AUC. Differential metabolites with an AUC value greater than 0.8 were screened as significant differential metabolites for model construction.

Images