High-coverage non-targeted metabonomics breast milk analysis method and application thereof

Abstract

The invention discloses a breast milk analysis method with high coverage rate and non-targeted metabonomics and application thereof, and belongs to the field of detection and analys.The breast milk analysis method comprises the following steps that a milk sample is pretreated, dried and redissolved, UHPLC-QTOF-MS positive and negative ion mode analysis is conducted, and in the UHPLC-QTOF-MS positive and negative ion mode analysis, adopted chromatographic columns comprise a T3 chromatographic column and an HILIC chromatographic column. The analysis method provided by the invention is simple and convenient to operate, can quickly and efficiently treat the breast milk sample, is suitable for high-throughput analysis, and remarkably improves the detection efficiency. Meanwhile, through the step of combining protein precipitation with concentration, the matrix effect is effectively reduced, the detection sensitivity and reproducibility are improved, and the data reliability is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of detection and analysis, and in particular to a high-coverage non-targeted metabolomics breast milk analysis method and application thereof. Background Art

[0002] Breast milk is not only the most important source of nutrition for infants after birth, but also plays a key role in the development of the infant's immune system, the establishment of intestinal microbiota, and growth and development. Metabolomics helps reveal the biological functions of these metabolites and their relationship with infant health by comprehensively analyzing metabolites in breast milk, thus providing a scientific basis for the optimization of breastfeeding strategies and the improvement of infant formula.

[0003] Existing breast milk metabolomics research faces technical challenges in many aspects. First, due to the complex matrix of breast milk, there are many types of metabolites in breast milk, with large differences in concentration and chemical properties, ranging from polar molecules to non-polar molecules, which requires more flexible and efficient extraction methods. The complex changing characteristics of metabolites in breast milk between maternal, time and infant needs also indicate that data support for long-term dynamic monitoring is necessary, so it is of great significance to develop a fast, simple and suitable metabolomics method for high-throughput analysis.

[0004] Liquid chromatography-mass spectrometry (LC-MS) has become the main tool for breast milk metabolomics research due to its high sensitivity and wide detection range. It can simultaneously analyze polar and non-polar molecules and provide accurate metabolite identification and quantification for complex biological samples. However, the complexity of the breast milk matrix (such as high fat content, protein, and interference from trace elements) poses a challenge to the accuracy of analysis.

[0005] In LC-MS detection, according to the different separation principles, it can be divided into normal phase chromatography (NPLC), reverse phase chromatography (RPLC), ion exchange chromatography (IEC), size exclusion chromatography (SEC), etc. Among them, RPLC is the most widely used in metabolome analysis, accounting for more than 60% of all chromatographic applications. The stationary phase of reverse phase chromatography uses non-polar particulate matter, such as C18, phenyl, and C8. In addition to reverse phase chromatography, hydrophilic interaction chromatography (HILIC) is also commonly used in metabolomics. The metabolites analyzed by HILIC are often highly polar substances, and its stationary phase uses highly hydrophilic materials, such as silica gel, weak ion exchange adsorbents, etc. During elution, the elution order of HILIC and RPLC is opposite, so HILIC chromatography is also called reverse-reversed phase chromatography.

[0006] In practical applications, RPLC chromatography is the most widely used and can elute most substances, but breast milk contains extremely diverse metabolites. How to improve the detection coverage of the metabolome is also a technical problem that technicians in this field have always wanted to solve. Summary of the invention

[0007] Purpose of the Invention To overcome the above shortcomings, the purpose of the present invention is to provide a high-coverage non-targeted metabolomics breast milk analysis method and its application. The analysis method of the present invention is simple to operate, can quickly and efficiently process breast milk samples, and is suitable for high-throughput analysis, which significantly improves the detection efficiency. At the same time, by combining protein precipitation with a concentration step, the matrix effect is effectively reduced, the sensitivity and reproducibility of the detection are improved, and the reliability of the data is ensured.

[0008] Solution To achieve the purpose of the present invention, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a high coverage non-targeted metabolomics method for breast milk analysis, comprising the following steps: pre-treating the milk sample, drying, re-dissolving, and performing UHPLC-QTOF-MS positive and negative ion mode analysis, Among them, the chromatographic columns used in UHPLC-QTOF-MS positive and negative ion mode analysis include T3 column and HILIC column.

[0009] Furthermore, the reconstitution solution used for reconstitution includes isopropanol, methanol, acetonitrile and water, and the volume ratio of isopropanol, methanol, acetonitrile and water is (1-3): (0.5-1.5): (0.5-1.5): (0.5-1.5) or 2:1:1:1.

[0010] Further, the T3 chromatographic column comprises an ACQUITY UPLC HSS T3 chromatographic column; And / or, the HILIC chromatographic column comprises an ACQUITY UPLC BEH HILIC Column.

[0011] Further, the mobile phase A used in the chromatography is an aqueous solution comprising ammonium acetate and ammonium fluoride; or, the mobile phase A used in the chromatography is an aqueous solution comprising ammonium acetate and ammonium fluoride in a molar ratio of 5: (0.1-0.5) or 5: (0.3-0.5); optionally, the concentration of ammonium acetate in the aqueous solution of mobile phase A is 5 mM; and / or, the mobile phase B used in the T3 chromatographic column includes acetonitrile; And / or, the mobile phase B used in the HILIC column is an acetonitrile solution comprising ammonium acetate and ammonium fluoride; or, the mobile phase B used in the HILIC column is a mixed solution comprising 95% by volume acetonitrile and 5% by volume water, the mixed solution comprising ammonium acetate and ammonium fluoride in a molar ratio of 5:(0.1-0.5) or 5:(0.3-0.5); optionally, the concentration of ammonium acetate in the aqueous solution of mobile phase B is 5 mM; and / or, the gradient elution program of the T3 column includes: 0 min: 0-10% mobile phase B, 1 min: 0-10% mobile phase B, 17 min: 90-100% mobile phase B, 19 min: 90-100% mobile phase B, 19.1 min: 0-10% mobile phase B, 22 min: 0-10% mobile phase B; and / or, the gradient elution program of the HILIC column includes: 0 min: 90-100% mobile phase B, 2 min: 90-100% mobile phase B, 8 min: 70-80% mobile phase B, 10 min: 45-55% mobile phase B, 12 min: 45-55% mobile phase B, 12.5 min: 90-100% mobile phase B, 15 min: 90-100% mobile phase B; and / or, in liquid chromatography, the flow rate is 0.1 to 0.5 mL / min; And / or, in liquid chromatography, the temperature is 30-45°C.

[0012] Furthermore, mass spectrometry was performed using electrospray ionization in both positive and negative ion modes; and / or, the mass spectrometer is AB SCIEX Triple TOF 6600; and / or, in mass spectrometry, the ion source temperature is controlled between 500 and 600°C; and / or, in mass spectrometry, primary scanning range: 50-1200 Da, secondary scanning range: 25-1200 Da; And / or, the secondary mass spectrum was acquired in data-dependent acquisition mode, and the peak intensity value screening mode was used, and the IDA settings were as follows: dynamic exclusion of isotope ion range: 4Da, and 10 fragment spectra were collected in each scan.

[0013] Furthermore, the mass spectrometry conditions are: .

[0014] Further, data analysis includes: preprocessing the overall analysis results using ME software, filtering characteristic variables based on the 80% rule, removing characteristic peaks with sample missing values ​​greater than 20%; filtering characteristic variables based on the relative standard deviation RSD of QC samples, screening out characteristic peaks with RSD less than 30%; deleting characteristic variables with missing values ​​exceeding 50%, and filling the missing values ​​with the minimum value in the variable. The preprocessed data is normalized by total amount.

[0015] Further, the milk sample includes breast milk, cow milk or goat milk.

[0016] Furthermore, the pretreatment method of the milk sample includes one or more of a protein precipitation method, a two-phase extraction method, and a single-phase extraction method.

[0017] Furthermore, the pretreatment method of the milk sample adopts a protein precipitation method, including: adding methanol to the milk sample at low temperature, mixing, ice bathing, high-speed centrifugation, taking the supernatant, drying, adding a complex solution, and using it for UHPLC-QTOF-MS positive and negative ion mode analysis; Alternatively, the pretreatment method of the milk sample adopts a single-phase extraction method, including: adding methanol to the milk sample at low temperature, mixing, adding methyl tert-butyl ether, ice bathing, high-speed centrifugation, taking the supernatant, drying, adding a complexing solution, and using it for UHPLC-QTOF-MS positive and negative ion mode analysis.

[0018] In a second aspect, a method for analyzing breast milk is provided for use in non-targeted metabolomics analysis.

[0019] Beneficial Effects (1) The analytical method of the present invention is easy to operate, can process breast milk samples quickly and efficiently, and is suitable for high-throughput analysis, significantly improving the detection efficiency. At the same time, by combining protein precipitation with a concentration step, the matrix effect is effectively reduced, the sensitivity and reproducibility of the detection are improved, and the reliability of the data is ensured.

[0020] (2) The present invention has a wider metabolite coverage: the combined detection strategy of T3 reverse phase chromatography column and HILIC hydrophilic interaction chromatography column is adopted to fully utilize the complementarity of the two chromatography columns to achieve a wide coverage of metabolites from polar to non-polar. Compared with the scheme of a single reverse phase chromatography column, the present invention can more comprehensively reflect the diversity of breast milk metabolites and provide a more complete metabolite spectrum for breast milk metabolomics analysis. The T3 chromatography column of the present invention is complementary to the HILIC chromatography column.

[0021] (3) The reconstitution solution of the present invention can meet the detection of breast milk samples at all stages and greatly improve the detection coverage. The protein precipitation technology used in the present invention can effectively reduce the matrix effect, and combined with the concentration step to increase the concentration of the target analyte, thereby improving the sensitivity and reproducibility of the detection. And it has better coverage and reliability. By optimizing the reconstitution solution, it can meet the needs of breast milk samples at all stages.

[0022] (4) The present invention can detect efficiently and is suitable for large-scale analysis: This method has fast detection speed, high sensitivity, and requires a small amount of sample, which is suitable for high-throughput analysis of large-scale breast milk samples. This has important practical value for large-scale research on breast milk metabolomics. The metabolite detection results of the present invention are more comprehensive: by combining efficient pretreatment methods and dual-column combined analysis, this method can detect more abundant metabolite types in breast milk samples, provide more comprehensive metabolite information, and lay a solid foundation for breast milk metabolomics research and maternal and child health research. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] One or more embodiments are exemplarily described by the pictures in the accompanying drawings, and these exemplary descriptions do not constitute limitations on the embodiments. The special word "exemplary" here means "used as an example, embodiment or illustrative". Any embodiment described as "exemplary" here is not necessarily interpreted as being superior or better than other embodiments.

[0024] Figure 1 The total ion current chromatograms (TIC diagrams) are collected using a T3 chromatographic column in positive and negative modes respectively after the protein precipitation method of the present invention; wherein a1, a2, and a3 are total ion current diagrams of the positive ion mode of the T3 chromatographic column; and b1, b2, and b3 are total ion current diagrams of the negative ion mode of the T3 chromatographic column.

[0025] Figure 2 The total ion current chromatograms (TIC diagrams) are collected using a HILIC column in positive and negative modes respectively after the protein precipitation method of the present invention; wherein c1, c2, and c3 are total ion current diagrams of the HILIC column in positive ion mode; and d1, d2, and d3 are total ion current diagrams of the HILIC column in negative ion mode.

[0026] Figure 3 It is the total ion current chromatogram (TIC diagram) of Example 1 and Comparative Example 3 of the present invention when different reconstitution solutions are used; wherein, a1, a2, and a3 are TIC diagrams and comparison diagrams of 1M breast milk samples using the reconstitution solutions of Example 1 and Comparative Example 3, respectively (a1 is the reconstitution solution of Example 1, a2 is the reconstitution solution of Comparative Example 3, and a3 is a comparison diagram); b1, b2, and b3 are TIC diagrams and comparison diagrams of 2M breast milk samples using the reconstitution solutions of Example 1 and Comparative Example 3, respectively (b1 is the reconstitution solution of Example 1, b2 is the reconstitution solution of Comparative Example 3, and b3 is a comparison diagram); c1, c2, and c3 are TIC diagrams and comparison diagrams of mixed breast milk samples at different stages using the reconstitution solutions of Example 1 and Comparative Example 3, respectively (c1 is the reconstitution solution of Example 1, c2 is the reconstitution solution of Comparative Example 3, and c3 is a comparison diagram).

[0027] Figure 4 The results of the inter-group difference analysis of Test Example 3 of the present invention; wherein, a is a Venn diagram of metabolites identified by T3 and HILIC columns; b is a score diagram of PCA analysis (A is the T3 group clustered on one side, and B is the HILIC group clustered on one side); c is a clustering heat map of metabolites (A is the T3 group, and B is the HILIC group).

[0028] Figure 5 This is the pathway enrichment analysis result of test example 4 of the present invention; the horizontal axis in the figure represents the pathway impact value (impact value), and the vertical axis represents the value after the P value is taken as -log10. The larger the circle and the darker the color, the higher the correlation between the group differences. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] In addition, in order to better illustrate the present invention, numerous specific details are provided in the specific embodiments below. It should be understood by those skilled in the art that the present invention can also be implemented without certain specific details. In some embodiments, raw materials, elements, methods, means, etc. well known to those skilled in the art are not described in detail, so as to highlight the main purpose of the present invention.

[0031] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.

[0032] The present invention has found through research that the selection of pretreatment methods and chromatographic columns is crucial to the reliability of the test results. Through continuous research, the present invention has obtained a fast, efficient, reliable and high-coverage method to improve the accuracy and efficiency of breast milk sample testing. Specifically: In the following examples, all reagents used are commercially available reagents, such as methanol, acetonitrile, formic acid, ammonium acetate, ammonium formate, isopropanol, methyl tert-butyl ether, and dichloromethane (all mass spectrometry grade), purchased from Thermo Fisher Scientific, Inc., USA; ammonium fluoride, purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.

[0033] In the following examples, the instruments used include: AB SCIEX Triple TOF 6600 LC-MS, AB SCIEX; 3K-15 refrigerated centrifuge, Sigma, Germany; KQ-5200E ultrasonic cleaner, Kunshan Ultrasonic Instrument Co., Ltd.; MIX-30S vortex mixer, Hangzhou Miou Instrument Co., Ltd.; MiLLi-Q, Advantage A10 ultrapure water system, Merck Millipore China Co., Ltd.; HM-D12 nitrogen blowdown instrument, Shandong Hengmei Electronic Technology Co., Ltd.; MC-8 cryoconcentrator, Beijing Ji Aim Instrument Co., Ltd.

[0034] In the following embodiments, the present invention uses MExplorer Ultimate (ME) software to process the data, and uses its cloud platform one-MAP (https: / / www.5omics.com) to identify and analyze the data. Its database includes a local database and public databases HMDB, KEGG, METLIN, etc., and different data processing requirements are achieved by establishing different algorithm chains.

[0035] In the following examples, the raw data (.wiff) file off the mass spectrometer was imported into MExplorer Ultimate (ME) software for data processing, and the ME software was used to perform peak alignment and chromatographic peak extraction on the spectrum of the raw data, and reliable qualitative identification results were obtained after secondary mass spectrometry denoising and screening, secondary mass spectrometry feature qualitative analysis, and matching of the secondary mass spectrometry qualitative results with the peak table.

[0036] In the following examples, the qualitative results of the positive and negative ion modes were integrated to obtain the overall analysis results, and the overall analysis results were preprocessed using ME software to filter the characteristic variables based on the 80% rule, and the characteristic peaks with sample missing values ​​greater than 20% were removed. The characteristic variables with missing values ​​exceeding 50% were deleted, and the missing values ​​were filled with the minimum value in the variable. The preprocessed data were normalized by total normalization (addition processing). The features with qualitative results in the screening peak table were subjected to subsequent analysis.

[0037] In the following examples, a cloud platform was used to perform statistical analysis, cluster analysis, pathway analysis, etc. on the identified metabolites, and a KEGG pathway diagram was drawn. The Maiwei cloud platform (https: / / cloud.metware.cn) was used to draw a score map and a cluster heat map of the PCA analysis. The Lianchuan Bio Cloud Platform (https: / / www.omicstudio.cn / home) was used to draw a Venn diagram.

[0038] Example 1: Sample pretreatment method In metabolomics studies of complex matrices such as breast milk, different pretreatment methods (such as protein precipitation, single-phase extraction, and two-phase extraction) and reconstitution methods have a significant impact on the tendency and coverage of metabolite detection results.

[0039] 1.1. Dual-phase extraction method The biphasic extraction method refers to the extraction method of patent number CN110243956A: 200 μL breast milk sample was extracted, and the aqueous phase was retained during the extraction process. The aqueous phases were combined together, mixed and centrifuged, the supernatant was collected and concentrated to dryness, and then re-dissolved in equal volume and centrifuged for loading onto the machine, which was marked as the Lipid-out group.

[0040] 1.2. Single-phase extraction method Take 100 μL of breast milk sample in a 1.5 mL centrifuge tube, add 600 μL of methanol solution, mix well, add 600 μL of methyl tert-butyl ether (MTBE), mix well, place in an ice bath for 10 min, centrifuge at 10000 r / min at 4 °C for 10 min, transfer 1100 μL of supernatant to a new 1.5 mL centrifuge tube, blow nitrogen to dryness, re-dissolve in equal volume, centrifuge and wait for loading, marked as MTBE group.

[0041] 1.3 Protein precipitation method Take 100 μL of breast milk sample in a 1.5 mL centrifuge tube, add 600 μL of methanol solution, mix well, place in an ice bath for 10 min, centrifuge at 10000 r / min at 4 °C for 10 min, transfer 1100 μL of supernatant to a new 1.5 mL centrifuge tube, blow nitrogen to dryness, re-dissolve in equal volume, centrifuge and wait for loading, marked as PPT group.

[0042] In the above pretreatment method, the reconstitution solution is isopropanol:methanol:acetonitrile:water (2:1:1:1, v / v / v / v).

[0043] In this embodiment, two chromatographic columns, T3 and HILIC, were used for sample injection after reconstitution, and samples were collected in positive and negative ion modes, respectively. The corresponding chromatographic conditions are as follows: 1) Mobile phase A: an aqueous solution containing 5 mM ammonium acetate (NH4AC) and 0.3 mM ammonium fluoride (NH4F); Mobile phase B: acetonitrile; Flow rate: 0.3 mL / min; Column temperature: 40°C; Injection volume: 5 μL; T3 chromatographic column model: ACQUITY UPLC HSS T3 (1.8µm, 2.1×100mm), Waters Company, USA; its gradient elution program: 0min: 2%B, 1min: 2%B, 17min: 98%B, 19min: 98%B, 19.1min: 2%B, 22min: 2%B.

[0044] 2) Mobile phase A: aqueous solution containing 5 mM ammonium acetate (NH4AC) and 0.3 mM ammonium fluoride (NH4F); Mobile phase B: a mixed solution of acetonitrile: water (95:5, v / v), containing 5 mM ammonium acetate (NH4AC) and 0.3 mM ammonium fluoride (NH4F); Flow rate: 0.3 mL / min; Column temperature: 40°C; Injection volume: 5 μL; HILIC column model: ACQUITY UPLC BEH HILIC Column (1.7µm, 2.1×100mm), Waters Company, USA; its gradient elution program: 0min: 100%B, 2min: 100%B, 8min: 75%B, 10min: 50%B, 12min: 50%B, 12.5min: 100%B, 15min: 100%B.

[0045] The mass spectrometry conditions used in this example are as follows: The mass spectrometer was AB SCIEX Triple TOF 6600, and electrospray ionization (ESI) positive and negative ion modes were used for detection. The mass spectrometry parameters were set as shown in Table 1: Table 1. Mass spectrometry parameters .

[0046] The primary scanning range was 50-1200Da, and the secondary scanning range was 25-1200Da. The secondary mass spectrum was acquired in data-dependent acquisition mode (IDA), and the peak intensity value screening mode was adopted. The IDA settings were as follows: dynamic exclusion of isotope ion range: 4Da, and 10 fragment spectra were collected for each scan.

[0047] In this embodiment, seven breast milk samples were mixed and treated by the three pretreatment methods of 1.1, 1.2 and 1.3 above, namely protein precipitation method (PPT group), single-phase extraction method (MTBE group) and two-phase extraction method to extract the aqueous phase (Lipid-out group), and then concentrated and re-dissolved. After re-dissolving, the samples were injected. During the injection, two chromatographic columns, T3 and HILIC, were used to collect samples in positive and negative ion modes, respectively. The results of the protein precipitation method are shown in FIG. Figure 1 , Figure 2 shown.

[0048] Figure 1 The a1, a2, and a3 results show that the acquisition performance of the T3 column in the positive ion mode is relatively stable and has good repeatability.

[0049] Figure 1 The results of b1, b2, and b3 show that the acquisition performance of the T3 column in negative ion mode is relatively stable and has good repeatability.

[0050] Figure 2 The c1, c2, and c3 results show that the acquisition performance of the HILIC column in positive ion mode is relatively stable and has good repeatability.

[0051] Figure 2 The d1, d2, and d3 results show that the acquisition performance of the HILIC column in negative ion mode is relatively stable and has good repeatability.

[0052] This example uses the above three pretreatment methods to treat the same mixed breast milk, and performs T3 and HILIC chromatography columns respectively, and analyzes and compares the peak extraction number and metabolite quantity, and the chemical classification and attribution of metabolites. Some of the results are as follows: Comparison of peak extraction number and metabolite quantity: After merging the data of T3 and HILIC columns using the above three pretreatment methods, the number of metabolites obtained after ME software library search and identification is shown in Table 2. The metabolites identified in positive ion mode (POS) and negative ion mode (NEG) were counted respectively, and the number of metabolites after the positive and negative ion mode data were merged and deduplicated was counted. Among them, the PPT group is the data obtained after protein precipitation treatment of mixed breast milk, the MTBE group is the data obtained after single-phase extraction of mixed breast milk, and the Lipid-out group is the data obtained by removing the organic phase in the two-phase extraction.

[0053] Table 2. Metabolite quantity statistics in three pretreatment methods .

[0054] According to Table 2, the maximum number of metabolites in the PPT group under positive ion mode (POS) is 2738, the maximum number of metabolites in the MTBE group under negative ion mode (NEG) is 1492, and after the positive and negative ion modes are combined and deduplicated, the maximum number of metabolites in the PPT group is 3883. Overall, there is not much difference in the number of metabolite identifications after the three pretreatment methods. All three methods can stably identify a certain number of metabolites, which can meet the data analysis of metabolomics.

[0055] Chemical classification and statistics of metabolites: After merging the data from the T3 and HILIC columns, the chemical classification of each group of metabolites detected was also statistically analyzed. The results are shown in Table 3 below.

[0056] Table 3 Metabolite classification and proportion of three pretreatment methods .

[0057] The results in Table 3 show that the PPT group has the most types, 3883, and the number of lipid and lipid molecules is also the largest, 2206. Lipids and lipid molecules account for the majority in each group. The compound types of lipid and lipid molecules were counted, and a total of 8 major categories were identified, including glycerophospholipids, fatty acyl, glycerolipids, sphingolipids, pregnenolone lipids, steroids and steroid derivatives, endocannabinoids, and glycolipids. The number of each identified and its proportion in lipid and lipid molecules are shown in Table 4: Table 4. Statistics of compound types of lipid and lipid-like molecules .

[0058] From Table 4 above, we can see that the number of MTBE group in the eight categories is less than that of PPT group, which to some extent indicates that the single-phase extraction used by MTBE group is not as comprehensive as that of PPT group in lipid detection. There is a slight difference in the number of major categories between PPT group and Lipid-out group. The difference in lipid types is further counted, and Table 5 shows the relatively large number of lipid types.

[0059] Table 5. Lipid species statistics .

[0060] It can be seen from Table 5 that the number of detections in the PPT group is still higher than that in the MTBE group. Compared with the Lipid-out group, the number of SM and TG in the PPT group is higher than that in the Lipid-out group. The number of Cer, PE, PC, LPE, CL, MG, DG, and FAHFA in the PPT group is higher than that in the Lipid-out group. There is no significant difference in other lipids between the two methods. Therefore, in the metabolomics detection method, the PPT group is more comprehensive among lipids and lipid molecules. Unless the experiment focuses more on SM and TG, the two-phase extraction method can be used.

[0061] This example also summarizes and counts the subclasses and compound types of other compounds in the PPT group and the MTBE group except lipids and lipid-like molecules, wherein the PPT group has 229 compound subclasses and the MTBE group has 216 compound subclasses, and Table 6 shows the categories with more than 8 compound subclasses in the PPT group. As can be seen from Table 6, the PPT group has more types than the MTBE group.

[0062] Table 6. Subclasses and compound types of compounds other than lipids .

[0063] The amino acid and sugar compounds were further summarized. About 40 kinds of amino acids were detected in the PPT group and the MTBE group, including L-valine, L-tyrosine, L-serine, L-glutamine, D-tyrosine, and D-aspartic acid. Ten kinds of sugars were detected in the PPT group, including maltotriose, xylitol, and D-xylulose. Five kinds of sugars were detected in the MTBE group, including α-D-glucose and fructan, which were less than those detected in the PPT group.

[0064] In summary, the metabolite coverage of the PPT group based on the protein precipitation method is more comprehensive, the detection effect of each category of compounds is very ideal, the operation is relatively simple, and high-throughput batch operation can be achieved.

[0065] Comparative Example 1 The difference lies in the different reconstitution solutions. The reconstitution solution used in this comparative example is 50% methanol aqueous solution. Taking the protein precipitation method as an example, when 50% methanol aqueous solution is added, mixed and centrifuged, white or yellow substances appear on the upper layer and cannot be injected.

[0066] Comparative Example 2 The difference lies in the different reconstitution solutions. The reconstitution solution used in this comparative example is methanol: dichloromethane: water (2:1:1, v / v / v). Taking the protein precipitation method as an example, when a reconstitution solution of methanol: dichloromethane: water (2:1:1, v / v / v) is added, stratification occurs in some samples with high oil content, such as breast milk within 1 month. The upper clear liquid can be sampled, but when only the aqueous phase is sampled after stratification, the metabolites dissolved in the organic phase cannot be detected, resulting in a decrease in metabolite coverage.

[0067] Comparative Example 3 The difference lies in the different reconstitution solutions. The reconstitution solution used in this comparative example is methanol: methyl tert-butyl ether: water (2:1:1, v / v / v).

[0068] Test Example 1 1M (breast milk at 1 month postpartum), 2M (breast milk at 2 months postpartum) and mixed breast milk at various stages were pre-treated with PPT, and re-dissolved with the re-solutions of Example 1 and Comparative Example 3, respectively, and centrifuged. The supernatant was clarified and loaded onto the machine, and each sample was repeated three times. The reproducibility of each sample was good. The total ion current chromatogram (TIC graph) of the sample in the negative mode is shown in the figure. Figure 3 shown.

[0069] Figure 3 The results show that there is little difference in the TIC graphs under negative mode conditions after running the samples using the re-solution systems of Example 1 and Comparative Example 3, but methyl tert-butyl ether (MTBE) is highly volatile, and the re-solution of Example 1 (isopropanol: methanol: acetonitrile: water (2:1:1:1, v / v / v / v)) is preferred.

[0070] Test Example 2: Statistics on the number of metabolites identified and their chemical classification ME software was used to identify and analyze the differential metabolites of breast milk samples collected on the T3 column and the HILIC column. After the positive ion mode data and the negative ion mode data were merged and deduplicated, 4593 metabolites were identified on the T3 column, and a total of 1711 metabolites were identified on the HILIC column in positive and negative ion modes. All identified metabolites were classified and counted according to their chemical classification information, and the number of different metabolites detected on the HILIC and T3 columns and the supplement of HILIC to T3 were summarized. The proportion of each type of metabolite in the primary classification is shown in Table 7.

[0071] Table 7 Metabolite classification and proportion .

[0072] As shown in Table 7, among the metabolites detected on the T3 and HILIC columns, lipids and lipid-like molecules accounted for the largest proportion in the primary classification, of which 2762 were detected on the T3 column and 786 on the HILIC column, of which 529 metabolites were detected on the HILIC column alone, that is, the HILIC column supplemented 529 lipid and lipid-like molecules. Similarly, the differences between the HILIC column and the T3 column were also reflected in the categories of organic acids and their derivatives, organic heterocyclic compounds, organic oxygen-containing compounds, benzoic compounds, phenylacetone and polyketone compounds.

[0073] Test case 3: Inter-group difference analysis The connections and differences between the metabolites of the T3 column and the HILIC column were displayed and explained using Venn diagrams, principal component analysis (PCA), and heat maps.

[0074] A Venn diagram was made for the metabolites identified by the T3 column and the HILIC column, such as Figure 4 The a shows that the two columns identified a total of 5830 metabolites, of which 427 metabolites were detected on both the T3 and HILIC columns, 4130 metabolites were detected on the T3 column alone, and 1273 metabolites were detected on the HILIC column alone, indicating that the T3 column was the main source of all metabolites, accounting for about 78.16%. In addition to the 7.32% detected on both columns, HILIC alone detected 21.84%, accounting for about one-third of the metabolites identified by the T3 column. Therefore, in order to obtain more comprehensive metabolite information, it is necessary to use both T3 and HILIC columns when collecting samples.

[0075] In order to intuitively show the difference between the metabolites of T3 and HILIC columns, principal component analysis (PCA) was performed on the compounds identified on the two columns. PCA is an unsupervised learning method that projects high-dimensional data into low-dimensional space through linear transformation, retaining the main variance information in the data and facilitating data analysis and visualization. The PCA method was used to observe the overall distribution trend between the two groups of samples. Figure 4 As can be seen in b, the same sample has obvious differences on the two columns, with the T3 group clustered on one side (e.g. Figure 4 b) of Figure A), the HILIC groups are clustered on one side (e.g. Figure 4 The difference is obvious. The HILIC column can detect metabolites that T3 cannot detect.

[0076] In the cluster heat map, each vertical column represents a single sample, and each horizontal row represents a metabolite. The closer the color is to red, the more positively correlated and highly expressed the sample is, while the closer it is to blue, the more negatively correlated and less expressed the sample is. The expression patterns of samples or metabolites in the same branch will be more similar. Figure 4 c is a heat map analysis of all metabolites identified on the T3 and HILIC columns, and the top 50 metabolites with the greatest differences are displayed. The color gradient (red to blue) in the heat map intuitively shows the differences between metabolites in different groups. It can be seen that the difference in metabolites between the T3 column and the HILIC column is obvious. The main categories in the figure are divided into two clusters. The first cluster of metabolites is the high content collected on the T3 column (such as Figure 4 The second cluster of metabolites is the high content collected on the HILIC column (such as Figure 4 (c in Figure B).

[0077] Test Example 4: Path Analysis In metabolomics research, pathway analysis is an important tool for understanding the dynamic changes of metabolites in biological systems. Analysis based on KEGG pathways helps to further reveal their biological functions. In order to deeply understand the functions and roles of differential metabolites in pathways, we used the relative-betweeness centrality method to perform pathway topology analysis and evaluate the importance of these differential metabolites in the pathway network. The hypergeometric distribution test method was used for enrichment analysis to evaluate the enrichment of these differential metabolites in specific pathways.

[0078] KEGG annotation was performed on the 597 differential metabolites screened, of which 164 had KEGG IDs. After enrichment analysis, 90 metabolites were located in 38 metabolic pathways, of which 18 metabolic pathways had impact values ​​(pathway impact values) > 0.05. The results are shown in Table 8; the pathway enrichment analysis results are shown in Table 8. Figure 5 In the figure, the horizontal axis represents the impact value of the pathway, and the vertical axis represents the value after the P value is taken as -log10. The larger the circle and the darker the color, the higher the correlation between the differences between the groups. The results show that the pathways where the differential metabolites on the T3 column and the HILIC column are located are mainly 18 pathways such as glycerophospholipid metabolism, D-glutamine and D-glutamate metabolism, galactose metabolism, arginine and proline metabolism, starch and sucrose metabolism, amino sugar and nucleotide sugar metabolism. The changes in metabolites in these pathways have important biological significance under experimental conditions. Among them, glycerophospholipid metabolism is located in the upper right corner, the point is large and the color is red, indicating that it has a high impact and significance in the study. It shows that the metabolites detected by the T3 column and the HILIC column have certain complementarity in the metabolic pathways.

[0079] Table 8. Pathway analysis results .

[0080] The present invention uses three pre-treatment methods, namely protein precipitation method (PPT), single-phase extraction method (MTBE) and two-phase extraction method (Lipid-out), to simultaneously treat the same mixed breast milk. By comparing the number of metabolites identified in the three groups of data, the number of metabolites in the PPT group after the positive and negative ions are combined and deduplicated is 3883 at most. Overall, there is not much difference in the number of metabolite identifications after treatment by the three methods. All three methods can stably identify a certain number of metabolites, which can meet the data analysis of metabolomics.

[0081] The present invention also optimizes the reconstitution solution. The preferred reconstitution solution is isopropanol:methanol:acetonitrile:water (2:1:1:1, v / v / v / v). The volume ratio can also be (1-3):(0.5-1.5):(0.5-1.5):(0.5-1.5).

[0082] Secondly, after merging the data of T3 and HILIC columns of the three pretreatment methods, the chemical classification of the metabolites detected in each group was statistically analyzed. By counting the number of compound superclasses, it was determined that lipids and lipid-like molecules accounted for the largest proportion in the three groups. Among the eight major types of compounds, the number of MTBE group was less than that of PPT group, and there was a slight difference in the number of major categories between PPT group and Lipid-out group. Among lipids and lipid-like molecules, PPT group was more comprehensive.

[0083] Finally, studies have shown that although the single-phase extraction method provides a high metabolome coverage, it cannot capture HMOs, amino acids, sugars, and other polar metabolites. Based on this, the subclasses and compound types of other compounds other than lipids and lipid molecules in the PPT group and the MTBE group were counted, among which the PPT group had more types than the MTBE group. The amino acid and sugar compounds were further summarized. The PPT group and the MTBE group detected and identified about 40 kinds of amino acids, the PPT group detected 10 kinds of sugars such as maltotriose, xylitol, and D-xylulose, and the MTBE group detected 5 kinds of sugars such as α-D-glucose and fructan, which were less than those detected by PPT.

[0084] The comparison results of the three pre-treatment methods showed that the metabolite coverage of the PPT group based on the protein precipitation method was more comprehensive, the detection effect of each category of compounds was very ideal, the operation was relatively simple, and high-throughput batch operation could be achieved.

[0085] The optimization of the pre-treatment steps of the present invention, such as protein removal, lipid separation and concentration steps, is crucial to minimize the matrix effect and improve the sensitivity of detection and the reproducibility of data. The optimized pre-treatment process is a key link to ensure the reliability of metabolomics data and interpret the biological significance.

[0086] The present invention uses two chromatographic columns, T3 and HILIC, and can detect non-polar metabolites and polar metabolites at the same time. 4593 metabolites were identified on the T3 column and 1711 metabolites were identified on the HILIC column. In the primary classification, lipids and lipid-like molecules accounted for the largest proportion, of which 2762 were detected on the T3 column and 786 were detected on the HILIC column. Among them, 529 metabolites were detected on the HILIC column alone, that is, the HILIC column supplemented 529 lipids and lipid-like molecules. Similarly, the differences between the detection on the HILIC column and the T3 are also manifested in the categories of organic acids and their derivatives, organic heterocyclic compounds, organic oxygen-containing compounds, benzoic compounds, phenylacetone and polyketone compounds.

[0087] The connection and difference of metabolites between T3 column and HILIC column were analyzed by using Venn diagram, principal component analysis and heat map. The Venn diagram showed that 5830 metabolites were identified by the two columns, of which 427 metabolites were detected on both T3 and HILIC columns, 4130 metabolites were detected on T3 column alone, and 1273 metabolites were detected on HILIC column alone, indicating that the T3 column was the main metabolite detected, accounting for about 78.16%. In addition to 7.32% detected on both columns, HILIC detected 21.84% alone, accounting for about one-third of the metabolites identified by T3 column. Therefore, in order to obtain more comprehensive metabolite information, T3 and HILIC columns should be used simultaneously when collecting samples. Both PCA score map and heat map show significant differences in metabolites on the two columns. Breast milk contains extremely diverse metabolites, so when combined with mass spectrometry, a single liquid chromatography system is not enough to achieve reasonable metabolome coverage. Hydrophilic interaction liquid chromatography (HILIC) provides complementary information to reversed-phase liquid chromatography (RPLC) by retaining polar metabolites, further improving metabolome coverage.

[0088] Pathway analysis of these differential metabolites revealed that the pathways in which the differential metabolites on the T3 column and the HILIC column were located were mainly 18 pathways, including glycerophospholipid metabolism, D-glutamine and D-glutamate metabolism, galactose metabolism, arginine and proline metabolism, starch and sucrose metabolism, amino sugar and nucleotide sugar metabolism. The changes in metabolites in these pathways had important biological significance under experimental conditions. This showed that the metabolites detected by the T3 column and the HILIC column had certain complementarity in metabolic pathways, especially in the glycerophospholipid metabolic pathway.

[0089] The present invention uses ultra-high performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UHPLC-QTOF-MS) technology to establish a non-targeted metabolomics analysis method for breast milk, and compares the extraction effects of three pretreatment methods on metabolites in breast milk, and determines that protein precipitation combined with concentration steps is the best pretreatment method. At the same time, a method combining a reverse phase chromatography column and a hydrophilic interaction chromatography column is used to collect and analyze samples, and a simple, highly sensitive, reliable, and high metabolite coverage breast milk metabolomics detection method is developed. It is of great significance for understanding the nutritional composition of breast milk and evaluating infant health and development.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high coverage non-targeted metabolomics method for breast milk analysis, characterized in that: The steps include: pre-treatment of milk samples, drying, re-dissolving, and UHPLC-QTOF-MS positive and negative ion mode analysis. Among them, the chromatographic columns used in UHPLC-QTOF-MS positive and negative ion mode analysis include T3 column and HILIC column.

2. The analysis method according to claim 1, characterized in that The reconstitution solution used for reconstitution includes isopropanol, methanol, acetonitrile and water. The volume ratio of isopropanol, methanol, acetonitrile and water is (1~3):(0.5~1.5):(0.5~1.5):(0.5~1.5) or 2:1:1:

1.

3. The analysis method according to claim 1, characterized in that T3 columns include ACQUITY UPLC HSS T3 columns; And / or, the HILIC chromatographic column comprises an ACQUITY UPLC BEH HILIC Column.

4. The breast milk analysis method according to claim 1, characterized in that: The mobile phase A used in the chromatography is an aqueous solution comprising ammonium acetate and ammonium fluoride; or, the mobile phase A used in the chromatography is an aqueous solution comprising ammonium acetate and ammonium fluoride in a molar ratio of 5:(0.1-0.5) or 5:(0.3-0.5); and / or, the mobile phase B used in the T3 chromatographic column includes acetonitrile; And / or, the mobile phase B used in the HILIC chromatographic column is an acetonitrile solution comprising ammonium acetate and ammonium fluoride; or, the mobile phase B used in the HILIC chromatographic column is a mixed solution comprising 95% by volume acetonitrile and 5% by volume water, wherein the mixed solution comprises ammonium acetate and ammonium fluoride in a molar ratio of 5:(0.1-0.5) or 5:(0.3-0.5); and / or, the gradient elution program of the T3 column includes: 0 min: 0-10% mobile phase B, 1 min: 0-10% mobile phase B, 17 min: 90-100% mobile phase B, 19 min: 90-100% mobile phase B, 19.1 min: 0-10% mobile phase B, 22 min: 0-10% mobile phase B; and / or, the gradient elution program of the HILIC column includes: 0 min: 90-100% mobile phase B, 2 min: 90-100% mobile phase B, 8 min: 70-80% mobile phase B, 10 min: 45-55% mobile phase B, 12 min: 45-55% mobile phase B, 12.5 min: 90-100% mobile phase B, 15 min: 90-100% mobile phase B; and / or, in liquid chromatography, a flow rate of 0.1 to 0.5 mL / min or 0.3 mL / min; And / or, in liquid chromatography, the temperature is 30-45°C or 40°C.

5. The method for analyzing breast milk according to any one of claims 1 to 4, characterized in that: Mass spectrometry was detected using electrospray ionization in positive and negative ion modes; and / or, the mass spectrometer is AB SCIEX Triple TOF 6600; and / or, in mass spectrometry, the ion source temperature is controlled between 500 and 600°C; and / or, in mass spectrometry, primary scanning range: 50-1200 Da, secondary scanning range: 25-1200 Da; And / or, the secondary mass spectrum was acquired in data-dependent acquisition mode, and the peak intensity value screening mode was used, and the IDA settings were as follows: dynamic exclusion of isotope ion range: 4Da, and 10 fragment spectra were collected in each scan.

6. The method for analyzing breast milk according to claim 5, characterized in that: The mass spectrometry conditions were: 。 7. The method for analyzing breast milk according to any one of claims 1 to 4, characterized in that: Data analysis includes: using ME software to preprocess the overall analysis results, filtering characteristic variables based on the 80% rule, and removing characteristic peaks with sample missing values ​​greater than 20%; filtering characteristic variables based on the relative standard deviation RSD of QC samples, and screening out characteristic peaks with RSD less than 30%; deleting characteristic variables with missing values ​​exceeding 50%, and filling the missing values ​​with the minimum value in the variable; and applying total normalization to the preprocessed data; And / or, the milk sample comprises breast milk, cow milk or goat milk.

8. The method for analyzing breast milk according to any one of claims 1 to 4, characterized in that: The pretreatment method of the milk sample includes one or more of a protein precipitation method, a two-phase extraction method, and a single-phase extraction method.

9. The method for analyzing breast milk according to claim 8, characterized in that: The pretreatment method of milk samples uses protein precipitation method, including: adding methanol to milk samples at low temperature, mixing, ice bath, high-speed centrifugation, taking supernatant, drying, adding complex solution, and using UHPLC-QTOF-MS positive and negative ion mode analysis; Alternatively, the pretreatment method of the milk sample adopts a single-phase extraction method, including: adding methanol to the milk sample at low temperature, mixing, adding methyl tert-butyl ether, ice bathing, high-speed centrifugation, taking the supernatant, drying, adding a complexing solution, and using it for UHPLC-QTOF-MS positive and negative ion mode analysis.

10. Use of the breast milk analysis method according to any one of claims 1 to 9 in non-targeted metabolomics analysis.

Images