Method for screening fumonisins metabolites in to-be-detected sample based on general characteristic structure diagnosis ion-assisted metabolic pathway and application

Abstract

The invention discloses a method for screening fumonisins metabolites in a to-be-detected sample based on a general characteristic structure diagnosis ion-assisted metabolic pathway and application. In the screening method, a metabolite prediction database is constructed based on a metabolic pathway of the fumonisins metabolite, and a general characteristic structure diagnosis ion database of the fumonisins metabolite is constructed based on characteristic fragments of a fumonisins standard substance. Through combinatorial analysis of the metabolite prediction database and the general characteristic structure diagnosis ion database on the to-be-detected sample, the molecular formula of the fumonisins metabolite in the to-be-detected sample is rapidly and accurately determined, comprehensive screening of known fumonisins metabolite and unknown fumonisins metabolite is achieved, the screening leakage risk is reduced, and the detection accuracy is improved. Meanwhile, the false positive of the screening result is low, and the screening efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of analytical detection technology, and more specifically, to a method and application for screening fumonisin metabolites in a sample to be tested based on a common characteristic structure diagnosis ion-assisted metabolic pathway. Background Art

[0002] Fumonisins (FBs) are a class of water-soluble secondary metabolites produced by Fusarium wilt and Fusarium spp. under certain temperature and humidity conditions. They are also one of the most serious mycotoxins in global food contamination in recent years. They were first isolated from Fusarium culture by South African scientists such as Gelderblom in 1988 and named fumonisins. Subsequently, Bezuidenhout et al. studied the chemical structure of fumonisins and found that it is a class of structurally similar diester compounds composed of different polyhydric alcohols and tricarbamic acid, specifically including a fatty chain containing 20 carbon atoms and two identical hydrophilic ester side chains distributed on both sides of the fatty chain. With the continuous deepening of research, more and more fumonisins have been discovered and reported. So far, more than 40 fumonisins and their derivatives have been discovered, which can be divided into four categories: A, B, C, and P. Among them, fumonisins FB1, fumonisins FB2, and fumonisins FB3 are the main forms in category B. They are highly toxic and widely distributed in the fumonisin family. They are often reported to be detected in crops such as corn and rice, during processing, and in commercial products. However, among the more than 40 fumonisins and their derivatives reported so far, only 10 have been made into corresponding commercial standards, so the screening and identification of fumonisins and new derivatives without standards is still a major problem.

[0003] Fumonisins can cause a variety of toxic effects on humans and animals, including neurotoxicity, immunotoxicity, cytotoxicity, organ toxicity, reproductive toxicity, etc. The World Health Organization's International Agency for Research on Cancer has classified fumonisins FB1 and fumonisins FB2 as Class 2B carcinogens. Many studies have shown that fumonisins in complex matrices change the basic chemical structure of free fumonisins through changes in physical conditions, covalent interactions with matrix components, and complexation or physical encapsulation with matrix biopolymers to generate various modified fumonisins. These fumonisins produced in the matrix and fungal environment are often unknown and complex, and are easily overlooked in detection. In addition, some modified fumonisins are hidden and cannot be detected by conventional methods. They may be toxic and pose potential hazards to humans and animals, which brings great difficulties to food supervision. Therefore, the development of efficient and accurate methods for identifying and quantitatively analyzing fumonisins is of great significance for studying and evaluating food safety issues caused by fumonisin co-contamination.

[0004] Non-targeted screening technology based on liquid chromatography combined with high-resolution mass spectrometry (LC-HRMS) is an effective tool for qualitative screening and analysis of unknown metabolites. Non-targeted screening technology based on high-resolution mass spectrometry data has the characteristics of high throughput, high sensitivity, ability to analyze trace compounds, simultaneous provision of structural information of parent and metabolites, and no need for repeated pre-treatment and detection. It has been widely used in the fields of environment, food safety, and traditional Chinese medicine. However, although the screening method of high-resolution mass spectrometry can obtain rich data information, it is still challenging to minimize false positive results and obtain valuable data from the multi-components of complex matrices and the huge and complex raw data of LC-HRMS due to the excessive number of material signals interfering with the screening results and the multiple pseudo signals generated by a single analyte. Diagnostic ions (DIs) are fragment ions of the characteristic mass spectrometric fragmentation behavior of compounds. Nowadays, diagnostic ion filtering technology based on the characteristic structure driven by the cleavage of the compound skeleton has become an important means for screening and structural identification of similar unknown metabolites. LIANG et al. proposed a non-targeted screening method for unknown veterinary drugs and their metabolites based on the characteristic ion information of fragments in similar structures, established an MS / MS database containing 3710 veterinary drugs and their metabolites, and realized the non-targeted screening of unknown veterinary drugs and their metabolites in 33 eggs, screening out 4 veterinary drugs and 3 metabolites in total. Dongwei Cui et al. developed an unknown substance identification strategy based on diagnostic ion filtering technology, and identified the action pathways of 4 fungal toxins in 63 samples through comprehensive analysis of 4 diagnostic ions. Wang Dashan et al. studied the secondary mass spectrometry fragmentation behavior of 15 Euphorbia diterpenes and summarized the basic fragmentation rules and diagnostic ions, which is helpful for characterizing the structure of these compounds and the subsequent structural identification of unknown analogues. However, these diagnostic ion screening techniques generally screen suspicious compounds based on 3-5 fragment ions of common characteristic structures, which are prone to problems of redundant information and high false positives. In addition, due to the single MS 2 Spectral coverage of natural products is low, and MS 2 Spectra or spectral libraries are used to perform large-scale annotation of multiple components, and the annotation results are easily limited, which makes the identification of unknown metabolites incomplete and time-consuming and labor-intensive.

[0005] At present, the conventional method for identifying metabolic derivatives is to compare the primary and secondary spectra of suspected metabolic derivatives with commercially available standards to achieve the purpose of structural annotation and quantification of suspected compounds. However, this method is heavily dependent on standards and is not applicable to the identification and identification of metabolic derivatives without standards. With the development of computer metabolism, researchers have developed a series of molecular metabolite prediction tools based on metabolic knowledge and machine learning, such as BioTransformer 3.0, GLORYx, EAWAG-pathway prediction system, Mycotoxin DB, ToxinDB, etc., which can expand the types and number of possible metabolites by predicting suspected metabolites through metabolic reactions. These metabolite prediction platforms simulate the phase I metabolic reactions (such as hydrolysis, reduction or oxidation, etc.) and phase II metabolic reactions (such as acetylation, methylation, etc.) of the parent compound and its metabolites in plants and soil, and each forms a prediction database containing a large number of unknown metabolites. These metabolite databases greatly expand the scope of identification of unknown metabolic derivatives and improve the directionality of screening unknown metabolites. However, existing predictive metabolic databases are mainly suitable for the prediction of broad-spectrum / general / non-specific metabolites, and many different types of toxins in nature have different biotransformation reactions, so there is room for improvement in the current predictive databases in terms of the prediction coverage and accuracy of specific types of metabolites. Summary of the invention

[0006] Based on this, an object of the present invention is to provide a method for screening fumonisin metabolites in a sample to be tested based on a common characteristic structure diagnostic ion-assisted metabolic pathway. The method combines the metabolite prediction database and the common characteristic structure diagnostic ion database to analyze the sample to be tested, and quickly and accurately determines the molecular formula of the fumonisin metabolite in the sample to be tested, which not only realizes the comprehensive screening of known fumonisin metabolites and unknown fumonisin metabolites, and reduces the risk of missed screening, but also has a low false positive screening result, which is conducive to improving the screening efficiency.

[0007] Another object of the present invention is to provide an application of the above method in simulating the analysis of fumonisin types in food.

[0008] In order to achieve the above first object, the present invention adopts the following technical scheme:

[0009] The present invention discloses a method for screening fumonisin metabolites in a sample based on a common characteristic structure diagnostic ion-assisted metabolic pathway, which is characterized by comprising the following steps:

[0010] 1) Construct a metabolite prediction database based on the metabolic pathway of fumonisin metabolites, and construct a common characteristic structure diagnostic ion database of fumonisin metabolites based on the characteristic fragments of fumonisin standards;

[0011] 2) performing a first matching screening on the primary mass spectrum information of each compound obtained from the sample to be tested and the metabolite prediction database to determine the compound molecular formula corresponding to each mass spectrum peak in the primary mass spectrum, and obtaining a first compound molecular formula set;

[0012] Matching and screening the secondary mass spectrum information of each compound obtained from the sample to be tested with the common characteristic structure diagnostic ion database, collecting the compound molecular formulas with the characteristic fragment matching number greater than or equal to a preset threshold, and obtaining a second compound molecular formula set;

[0013] 3) performing metabolomics analysis on the compound molecular formulas in the intersection of the first compound molecular formula set and the second compound molecular formula set to determine the candidate structures of the fumonisin metabolites in the sample to be tested.

[0014] Furthermore, according to the potential fumonisin metabolic pathways, the metabolic reaction information on each metabolic pathway is collected to construct a metabolite prediction database;

[0015] Wherein, the metabolic reaction includes a phase I metabolic reaction and / or a phase II metabolic reaction, the total number of reaction steps is 1-3 steps, and the maximum number of reaction steps of the phase II metabolic reaction is 2 steps.

[0016] Further, the phase I metabolic reaction includes one of a dehydroxylation reaction, a TCA thermal cracking reaction, a dehydration reaction, a desaturation reaction, a hydration reaction, an oxidative deamination to generate an alcohol, an oxidative deamination to generate a ketone, a nitro reduction reaction, an oxidation reaction, and a reduction reaction;

[0017] The phase II metabolic reactions include fructose acylation, sucrose acylation, carboxymethylation, amination, hydroxyoleylation, hydroxylinoleylation, aminooleylation, aminolinoleylation, aminopalmitoylation, demethylation, aminoamidation, carboxyacetylation, acetylation, arginine substitution, cysteine ​​substitution, glucoside substitution, glucuronide substitution, methoxyglucose substitution, glutamine substitution, glycine substitution, glutathione substitution, methylation, ornithine substitution, hexadecyl substitution, octadecyl substitution, sulfonation, taurination, diglucoside substitution, triglucoside substitution, tetraglucoside substitution, xylan-glucose substitution, malonyl-glucoside substitution, feruloyl glucose substitution, N-acetylcysteine ​​substitution, N-acetyl substitution, benzoyl substitution, hexitol substitution, and disodium anhydride substitution.

[0018] Further, the metabolic reaction includes 0-step phase I metabolic reaction and 1-2-step phase II metabolic reaction, or includes 1-step phase I metabolic reaction and 0-2-step phase II metabolic reaction.

[0019] Furthermore, the secondary mass spectrometry information of known fumonisin metabolite standards was used to determine fragment ions with common characteristic structures to construct a common characteristic structure diagnostic ion database.

[0020] Furthermore, the common characteristic structure diagnostic ion database includes characteristic fragments with m / z of 352.3210, 334.3104, 316.2999, 299.2733, 254.2115, 236.2009, 218.1903, 201.1638, 159.0288, 354.3367, 336.3261, 318.3155, 301.2890, 256.2271, 238.2165, 220.2060, and 203.1794.

[0021] Furthermore, the preset threshold is 4.

[0022] Furthermore, the method is performed by the data processing software Compound Discover 3.3.

[0023] Further, the first matching screening of the primary mass spectrometry information of each compound obtained from the sample to be tested with the metabolite prediction database includes:

[0024] A mass spectrometer is used to collect data for each compound in the sample to be tested, and data processing software is used to perform peak extraction, alignment and integration on the collected raw data. The primary mass spectrometry information of each compound in the sample to be tested is compared with the data information in the metabolite prediction database to determine the molecular formula of the compound corresponding to each mass spectrum peak in the primary mass spectrum.

[0025] Furthermore, the comparison is to identify the structure and metabolic pathway of the metabolite by combining information such as molecular formula, isotope distribution peak, accurate molecular weight, retention time, metabolic reaction, etc., and control the mass accuracy Δm≤5ppm, the isotope distribution peak matching degree (IP)≥70 points, and the response intensity≥e 5 , the expected retention time deviation is <2min.

[0026] Further, step 3) includes:

[0027] According to the compound molecular formula and the corresponding metabolic pathway, a false positive interference subtraction operation is performed on the compound molecular formula in the intersection of the first compound molecular formula set and the second compound molecular formula set to determine the candidate structure of the metabolite in the sample to be tested.

[0028] In order to achieve the above second purpose, the present invention adopts the following technical solutions:

[0029] The present invention discloses an application of the above method in simulating the analysis of fumonisin types in grains.

[0030] The beneficial effects of the present invention are as follows:

[0031] The present invention provides a method for screening fumonisin metabolites in a sample to be tested based on a common characteristic structure diagnostic ion-assisted metabolic pathway. Before the method is screened, it is necessary to construct a metabolite prediction database based on the metabolic pathway of fumonisin metabolites, and a common characteristic structure diagnostic ion database of fumonisin metabolites based on the characteristic fragments of fumonisin standards. The metabolic pathway of each metabolite can be learned through the constructed metabolite prediction database, which is helpful for judging whether the molecular formula preliminarily determined is correct and eliminating the interference of false positives. Through the constructed common characteristic structure diagnostic ion database, the mass spectrometry fragmentation behavior of fumonisin can be analyzed, and the accuracy of judging the metabolite structure in the sample to be tested can also be improved.

[0032] The metabolite prediction database and the common characteristic structure diagnostic ion database are matched and screened with the primary mass spectrometry information and the secondary mass spectrometry information of each compound in the sample to be tested, and then the candidate structure of the metabolite is obtained by reverse deduction with the help of metabolic reactions and metabolic pathways.

[0033] The screening method provided by the present invention is not only suitable for the structural screening of known fumonisin metabolite standards, but also can be used to reversely infer the structure of unknown fumonisin metabolites through metabolic reactions and metabolic pathways. Compared with the screening methods disclosed in the prior art, it is more targeted and has higher screening efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0035] Figure 1 A flow chart showing the operation of the screening method of the present invention.

[0036] Figure 2a A comparison of the secondary mass spectra of 9 standards that screen out toxins and actual corn samples in a specific embodiment is shown.

[0037] Figure 2b The secondary mass spectra of six toxins screened out without standards in actual corn samples in a specific embodiment are shown. DETAILED DESCRIPTION

[0038] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the content specifically described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0039] In order to design a method for detecting the types of fumonisin metabolites in food, a method for screening fumonisin metabolites in samples to be tested based on common characteristic structure diagnosis ion-assisted metabolic pathways was developed. The method uses the software Compound Discover 3.3 for data processing and specifically includes the following steps:

[0040] Construction of a metabolite prediction database based on the metabolic pathways of fumonisin metabolites

[0041] Based on literature reports and research experience, the applicant screened out 50 common metabolic reactions that may be suitable for toxins such as fumonisins from the plant metabolic pathway database, such as esterification and acetylation. After classification, the metabolic reaction types were divided into phase I metabolic reactions and phase II metabolic reactions. The specific metabolic reaction classification is shown in Table 1.

[0042] Table 1

[0043]

[0044]

[0045] In order to make the constructed metabolite prediction database cover all fumonisin metabolites as much as possible, the present invention limits the number of steps of the metabolic pathway: the maximum number of reaction steps of the phase II metabolic reaction is set to 2 steps, and the maximum number of steps of the total metabolic reaction is set to 3 steps.

[0046] Based on the different metabolic pathways of fumonisin, fumonisin B1 was used as the starting material to collect the names of metabolic reactions, molecular formulas of lost groups (or added groups), structural formulas, etc., and sort them into CSV format files to form a list of metabolic structure modifications. The modification list was then imported into the transformation database of Compound Discover 3.3 software, and a total of 18,050 metabolite molecular formulas were obtained. After deleting isomers, the construction of the metabolite prediction database was completed. According to the following statistics, a total of 7,121 fumonisin metabolites were included:

[0047] When the number of metabolic reaction steps in phase I is 0 and the number of metabolic reaction steps in phase II is 1, there are 40 molecular formulas M1 of fumonisin metabolites that can be produced through mathematical statistics;

[0048] When the number of metabolic reaction steps in phase I is 0 and the number of metabolic reaction steps in phase II is 2, the number of molecular formulas M2 of fumonisin metabolites that can be produced through mathematical statistics is 1600;

[0049] When the number of metabolic reaction steps in phase I is 1 and the number of metabolic reaction steps in phase II is 0, there are 10 molecular formulas M3 of fumonisin metabolites that can be produced through mathematical statistics;

[0050] When the number of metabolic reaction steps in phase I is 1 and the number of metabolic reaction steps in phase II is 1, the number of molecular formulas M4 of fumonisin metabolites that can be produced through mathematical statistics is 400;

[0051] When the number of steps of phase I metabolic reactions is 1 and the number of steps of phase II metabolic reactions is 2, there are 16,000 molecular formulas M5 of fumonisin metabolites that can be produced through mathematical statistics.

[0052] After evaluation, the metabolite prediction database provided by the present invention contains 44 fumonisin metabolites that have been reported in the literature, and also contains the molecular formulas of fumonisin metabolite standards such as fumonisin B1-B3 (i.e. FB1, FB2, FB3), hydrolyzed fumonisin B1-B3 (i.e. HFB1, HFB2, HFB3), fumonisin A1-A2 (i.e. FA1, FA2), semi-hydrolyzed fumonisin B1a, B1b (i.e. PHFB1a, PHFB1b), indicating that the metabolite prediction database of the present invention has a high coverage rate and can be used to predict metabolite molecular formulas.

[0053] Construction of a diagnostic ion database of common characteristic structures of fumonisin metabolites

[0054] Fumonisin contains propane-1,2,3-tricarboxylic acid (TCA) side chains, amino and hydroxyl groups in its structure, and is easy to react with groups of other compounds in the matrix to form various derivatives. Their typical common characteristics are the skeleton structure with C2-amino (primary amino), C3 / C5 / C10 / C14 / C15-hydroxyl and C14 / C15-TCA groups. Currently, the commercially available fumonisin metabolite standards include FB1, FB2, FB3, HFB1, HFB2, HFB3, FA1, FA2, PHFB1a, PHFB1b. According to the mass spectrometry fragments and structures of the fumonisin metabolite standards combined with the theoretical fragments provided by Mass Frontier, the fragmentation pathways and characteristic fragments of fumonisin metabolites can be known. By analyzing the secondary high-resolution mass spectra of 8 existing standard substances, the main characteristic fragments of FB1, HFB1, PHFB1a and PHFB1b are 352.3210, 334.3104, 316.2999, 299.2733, 254.2115, 236.2009, 218.1903, 201.1638, 159.0288; the main characteristic fragments of FB2, FB3, HFB2 and HFB3 are 354.3367, 336.3261, 318.3155, 301.2890, 256.2271, 238.2165, 220.2060, 203.1794. The m / z of these two types of standards always differ by 2. It can be found that except for the parent ion fragments of fully hydrolyzed fumonisin metabolites (HFB1, HFB2, HFB3), which do not contain the characteristic fragment 159.0288, other standards all have the characteristic fragment 159.0288. This is because the characteristic fragment 159.0288 is mainly obtained by the cleavage of the TCA group, while the fully hydrolyzed fumonisin structure does not contain the propane-1,2,3-tricarboxylic acid group. Therefore, according to the analysis of characteristic fragments, the characteristic fragments of the following 17 fumonisin metabolites were determined, namely 352.3210, 334.3104, 316.2999, 299.2733, 254.2115, 236.2009, 218.1903, 201.1638, 159.0288, 354.3367, 336.3261, 318.3155, 301.2890, 256.2271, 238.2165, 220.2060, 203.1794, and the construction of the common characteristic structure diagnostic ion database was completed. Technicians can use the common characteristic structure diagnostic ion database to screen fumonisin metabolites, which not only expands the screening range but also increases the accuracy of the screening method compared with the traditional characteristic fragment (generally 2-5) comparison screening.

[0055] Accuracy evaluation

[0056] A mixed solution of 8 common fumonisin metabolites was added to blank corn, and the metabolite prediction database and the common characteristic structure diagnostic ion database were used to screen the test data. Among them, the common characteristic structure diagnostic ion database screened out 33 candidate fumonisin metabolites, and the metabolite prediction database screened out 411 candidate fumonisin metabolites. The 8 added fumonisin metabolites can be screened out by the above two databases, and the number of false positive compounds screened out is 25 and 403 respectively. The false positive screening of the metabolite prediction database is high. If the two databases are combined, a total of 12 candidate fumonisin metabolites can be screened out, and the number of false positive compounds is 4, indicating that when the two databases are combined for screening, the false positive rate is reduced by 6-100 times, greatly reducing the workload of screening unknown substances.

[0057] The following steps are combined Figure 1 Provide explanation.

[0058] Data collection of samples to be tested

[0059] Screening and identification of fumonisin metabolites in 30 samples of corn with high fumonisin content collected from the main producing areas of my country were performed as follows:

[0060] After the sample is crushed (90% through a 40-mesh sieve) and mixed, 5.00 g is accurately weighed, and 20 mL of acetonitrile-water-acetic acid (70:29:1, volume ratio) extract is added. After vortex mixing at 2500 rpm for 20 minutes, the solid-liquid separation is performed at 7000 r / min for 5 minutes. Accurately transfer 0.5 mL of the supernatant to a 1.5 mL centrifuge tube, add 0.5 mL of water to dilute, vortex mix for 1 minute, and then centrifuge at 12000 r / min for 10 minutes. Take the supernatant and filter it with a 0.22 μm PTFE filter membrane, mix it and wait for testing. The primary and secondary mass spectrometry information of each compound in the sample was obtained by simultaneous primary and secondary mass spectrometry acquisition using ultra-performance liquid chromatography (UltiMate 3000, Thermo Fisher Scientific, Rockford, IL, USA)-high-resolution mass spectrometry (QExactive, Thermo Fisher Scientific, Rockford, IL, USA). The chromatograms and mass spectra obtained by the test are shown in Table 2.

[0061] Compound Discover 3.3 was used to perform peak extraction, alignment, integration, background subtraction and other operations on the collected raw data.

[0062] The high-resolution mass spectrometer is a quadrupole / electrostatic field orbital trap high-resolution mass spectrometer Q-Exactive, purchased from ThermoFisher Scientific. The mass spectrometry conditions are: the temperature of the heated electrospray ion source is 300°C, the capillary voltage is 3.2 kV, the temperature of the ion transfer tube is 320°C, the sheath gas is 35 units, and the auxiliary gas is 10 units. Full scan / ddms2 scanning mode: the acquisition range is 100-1200Da, ESI+ positive ion mode acquisition, the primary mass spectrometry resolution is 70000FWHM, the secondary mass spectrometry resolution is 17500FWHM, and the collision energy is 20eV, 40eV, and 60eV.

[0063] Screening and analysis

[0064] Compare the primary mass spectrometry information in the sample to be tested with the data information in the metabolite prediction database, combine the molecular formula, isotope distribution peak, accurate molecular weight, retention time, metabolic reaction and other information to identify the structure and metabolic pathway of the metabolite, control the mass accuracy Δm≤5ppm, the isotope distribution peak matching score (IP)≥70 points, and the response intensity≥e 5 , the expected retention time deviation is <2min, and the first compound molecular formula set is obtained.

[0065] The secondary mass spectrometry information in the sample to be tested is matched and screened with the common characteristic structure diagnostic ion database obtained above, and the compound molecular formulas with characteristic fragment matching numbers greater than or equal to 4 are obtained by controlling the mass accuracy Δm≤5ppm to obtain a second compound molecular formula set.

[0066] The compound formulas in the intersection of the first compound formula set and the second compound formula set were subjected to a false positive interference subtraction operation. The basis was that the technicians screened out 16 fumonisin metabolite formulas in 30 corn samples based on the compound formulas and the corresponding metabolic pathways. There were two identical fumonisin metabolite formulas, totaling 15 fumonisin metabolites, as shown in Table 2. Compounds 1-9 were confirmed to be FB1, FB2, FB3, HFB2, HFB3, PHFB1a, PHFB1b, FA1, and FA2 by comparison with the standard. The remaining 6 were derivatives of unknown structures of fumonisin. The predicted molecular formulas given by the metabolite prediction database based on their primary mass spectrometry information were C 22 H 45 NO4, C 22 H 45 NO4, C 28 H 51 NO9, C 28 H 51 NO9, C 28 H 53NO9, C 28 H 53 NO9. The following describes the discovery and identification process of fumonisin by combining accurate mass data, mass spectrometry information, and structural information such as the metabolic pathway of fumonisin.

[0067] Take compound 1C34H59NO15 as an example to illustrate the identification of compounds with existing standards. The m / z of compound C34H59NO15 is 722.3960 (Δm = 0.33ppm), and the isotope matching score is 84 points. From the mass spectrum, it can be analyzed (see Figure 2a ), the compound has 8 characteristic fragments that match the characteristic fragments in the common characteristic structure diagnostic ion database, namely m / z 159.02869, m / z 201.16368, m / z 236.20049, m / z 254.21149, m / z299.27216, m / z316.29953, m / z 334.31000, and m / z 352.32062. According to the metabolic reaction analysis in the metabolite prediction database, the compound was preliminarily analyzed as fumonisin B1. By comparing the primary chromatography and secondary mass spectrometry with the fumonisin B1 standard, the compound C34H59NO15 was determined to be fumonisin B1.

[0068] Compound 10 (m / z 388.3418, [M+H]) and unknown compound 11 (m / z 388.3419, [M+H]) were found in the metabolite prediction database. Their predicted molecular formulas were both C22H45NO4, but their retention times in the chromatogram were different, 16.70min and 17.21min, respectively. Therefore, the relationship between the two unknown compounds was determined to be isomers. Through their mass spectra analysis, we can see that (see the secondary mass spectra) Figure 2b), compound 10 has 4 secondary fragment ions (i.e., m / z 236.20030, m / z 254.2110, m / z 334.31015, and m / z 352.3208) that match the characteristic fragments in the common characteristic structure diagnosis ion database, and compound 11 has 2 secondary fragment ions (i.e., m / z 334.30984 and m / z 352.32050) that match the characteristic fragments in the common characteristic structure diagnosis ion database. In the range of m / z 100-300, the fragment ions of compounds 10 and 11 are highly similar to the characteristic fragments of FB1, and it is preliminarily judged that they are all FB1 metabolites. Further analysis showed that the parent ion 722.39602 of FB1 lost two 158.0215s to generate HFB1 (m / z 406.3530, [M+H]), and then lost one 18.0106 (H2O) to form a fragment ion of 388.3424. That is, the metabolic pathways of the two unknown compounds were that the TCA groups at the C14 and C15 sites on the FB1 carbon chain were hydrolyzed and lost, and then a water molecule was lost. Therefore, it was inferred that compounds 10 and 11 were two isomers of dehydration hydrolysis fumonisin B1 (Dehydration HFB1), namely DehydrationHFB1a and Dehydration HFB1b.

[0069] Compound 12 (m / z 546.3636, RT=17.56) and compound 13 (m / z 546.3635, RT=18.49) are a group of isomers that can be found in the metabolite prediction database, and the predicted molecular formula is C28H51NO9. Figure 2bMass spectrum analysis showed that compound 12 had two secondary fragment ions (m / z 352.32080, m / z 354.33685) that matched the characteristic fragments in the common characteristic structure diagnosis ion database, and compound 13 had three secondary fragment ions (m / z 318.31512, m / z336.32578, m / z 354.33615) that matched the characteristic fragments in the common characteristic structure diagnosis ion database. In the range of m / z100-300, the fragment ions of compounds 12 and 13 were highly similar to the FB2 fragment ions, and they were preliminarily judged to be FB2 metabolites. Further analysis combined with the metabolite prediction database showed that the parent ion of FB1 (m / z 722.39602) lost a 158.0215 to generate PHFB1 (m / z 564.3745, [M+H]), and then continued to lose a 18.0106 (H2O) to form a 546.3639 fragment ion, namely compound 12 (m / z 546.3636) and compound 13 (m / z 546.3639). Combined with the analysis of candidate compounds with the same molecular formula in the metabolite prediction database, the metabolic pathways of compound 12 and compound 13 were that the TCA group of C14 or C15 on FB1 was lost by hydrolysis, and then a water molecule was lost. Therefore, it was inferred that compound 12 and compound 13 were two isomers of dehydrated hemihydrolysis fumonisin B1 (Dehydration PHFB1), namely dehydrated hemihydrolysis fumonisin B1-RT1 (Dehydration PHFB1-RT1) and dehydration hemihydrolysis fumonisin B1-RT2 (Dehydration PHFB1-RT2).

[0070] Compound 14 (m / z 548.3791, RT = 17.92) and compound 15 (m / z 548.3791, RT = 18.38) are a group of isomers that can be found in the metabolite prediction database, and the predicted molecular formula is C28H53NO9. Figure 2b) analysis showed that compound 14 had two secondary fragment ions (m / z 336.32581, m / z 354.33646) that matched the characteristic fragments in the common characteristic structure diagnosis ion database, and compound 15 had five secondary fragment ions (m / z 159.0288, m / z 203.0288, m / z 318.31531, m / z 336.32562, m / z 354.33633) that matched the characteristic fragments in the common characteristic structure diagnosis ion database. In the range of m / z 100-300, the fragment ions of compounds 14 and 15 were highly similar to the FB2 fragment ions, and they were judged to be FB2 metabolites. Further analysis of the metabolic pathways in the metabolite prediction database showed that the parent ion 706.40083 of FB2 lost one TCA group (m / z 158.0215) to generate semi-hydrolyzed fumonisin B2 (PHFB2, m / z 548.37933, [M+H]), which had the same molecular formula as unknown compound 12 (m / z 546.3636) and unknown compound 13 (m / z 546.3635). Combined with the analysis of candidate compounds with the same molecular formula in the FBs metabolite prediction database, the metabolic pathways of compounds 14 and 15 were that FB2 lost the TCA group, that is, the TCA group at C14 or C15 on the carbon chain of FB2 was hydrolyzed and lost, and then both sites had hydroxyl groups (-OH). Therefore, it was inferred that compounds 14 and 15 were semi-hydrolyzed fumonisin B2 (Dehydration Fumonisin B2). Two isomers of PHFB1, hemihydrolyzed fumonisin B2a (PHFB2a) and hemihydrolyzed fumonisin B2b (PHFB2b), have been reported in other literatures, and the secondary fragments of the compounds in this sample are consistent with those reported in the literature ([Use of liquid chromatography–high-resolution mass spectrometry for isolation and characterization of hydrolyzed fumonisinsand relevant analysis in maize-based products]).

[0071] Therefore, the molecular formula of the unknown metabolite can be inferred through the prediction database and common characteristic structural fragment diagnostic ions, and the structure of the unknown metabolite can be finally determined through the analysis of the molecular formula and metabolic pathway of the unknown metabolite by technicians.

[0072] Table 2 Screening results of fumonisin in corn samples

[0073]

[0074] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made on the basis of the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A method for screening fumonisin metabolites in a sample based on a common characteristic structure diagnosis ion-assisted metabolic pathway, characterized in that: The steps include: 1) Construct a metabolite prediction database based on the metabolic pathway of fumonisin metabolites, and construct a common characteristic structure diagnostic ion database of fumonisin metabolites based on the characteristic fragments of fumonisin standards; 2) performing a first matching screening on the primary mass spectrum information of each compound obtained from the sample to be tested and the metabolite prediction database to determine the compound molecular formula corresponding to each mass spectrum peak in the primary mass spectrum, and obtaining a first compound molecular formula set; Matching and screening the secondary mass spectrum information of each compound obtained from the sample to be tested with the common characteristic structure diagnostic ion database, collecting the compound molecular formulas with the characteristic fragment matching number greater than or equal to a preset threshold, and obtaining a second compound molecular formula set; 3) performing metabolomics analysis on the compound molecular formulas in the intersection of the first compound molecular formula set and the second compound molecular formula set to determine the candidate structures of the fumonisin metabolites in the sample to be tested.

2. The method according to claim 1, characterized in that According to the potential fumonisin metabolic pathway, the metabolic reaction information on each metabolic pathway is collected to build a metabolite prediction database; Wherein, the metabolic reaction includes a phase I metabolic reaction and / or a phase II metabolic reaction, the total number of reaction steps is 1-3 steps, and the maximum number of reaction steps of the phase II metabolic reaction is 2 steps.

3. The method according to claim 2, characterized in that The phase I metabolic reaction includes one of a dehydroxylation reaction, a TCA thermal cracking reaction, a dehydration reaction, a desaturation reaction, a hydration reaction, an oxidative deamination reaction to generate an alcohol, an oxidative deamination reaction to generate a ketone, a nitro reduction reaction, an oxidation reaction, and a reduction reaction; The phase II metabolic reactions include fructose acylation, sucrose acylation, carboxymethylation, amination, hydroxyoleylation, hydroxylinoleylation, aminooleylation, aminolinoleylation, aminopalmitoylation, demethylation, aminoamidation, carboxyacetylation, acetylation, arginine substitution, cysteine ​​substitution, glucoside substitution, glucuronide substitution, methoxyglucose substitution, glutamine substitution, glycine substitution, glutathione substitution, methylation, ornithine substitution, hexadecyl substitution, octadecyl substitution, sulfonation, taurination, diglucoside substitution, triglucoside substitution, tetraglucoside substitution, xylan-glucose substitution, malonyl-glucoside substitution, feruloyl glucose substitution, N-acetylcysteine ​​substitution, N-acetyl substitution, benzoyl substitution, hexitol substitution, and disodium anhydride substitution.

4. The method according to claim 2, characterized in that: The metabolic reaction includes 0-step phase I metabolic reaction and 1-2-step phase II metabolic reaction, or includes 1-step phase I metabolic reaction and 0-2-step phase II metabolic reaction.

5. The method according to claim 1, characterized in that The secondary mass spectrometry information of known fumonisin metabolite standards was used to identify fragment ions with common characteristic structures and to construct a diagnostic ion database with common characteristic structures.

6. The method according to claim 5, characterized in that The common characteristic structure diagnostic ion database includes characteristic fragments with m / z of 352.3210, 334.3104, 316.2999, 299.2733, 254.2115, 236.2009, 218.1903, 201.1638, 159.0288, 354.3367, 336.3261, 318.3155, 301.2890, 256.2271, 238.2165, 220.2060, and 203.1794.

7. The method according to claim 1, characterized in that The preset threshold is 4.

8. The method according to claim 1, characterized in that The first matching screening of the primary mass spectrometry information of each compound obtained from the sample to be tested with the metabolite prediction database includes: A mass spectrometer is used to collect data for each compound in the sample to be tested, and data processing software is used to perform peak extraction, alignment and integration on the collected raw data. The primary mass spectrometry information of each compound in the sample to be tested is compared with the data information in the metabolite prediction database to determine the molecular formula of the compound corresponding to each mass spectrum peak in the primary mass spectrum.

9. The method according to claim 1, characterized in that: Step 3) includes: According to the compound molecular formula and the corresponding metabolic pathway, a false positive interference subtraction operation is performed on the compound molecular formula in the intersection of the first compound molecular formula set and the second compound molecular formula set to determine the candidate structure of the metabolite in the sample to be tested.

10. Use of the method according to any one of claims 1 to 9 in simulating the analysis of fumonisin types in food.

Images