CN108753855A - The method that WGCNA identifies D-ALPHA-Hydroxypropionic acid fermentation process notable module and Hubs metabolins - Google Patents

Abstract

The invention relates to a method for WGCNA to identify significant modules and Hubs metabolites in the D-lactic acid fermentation process; to conduct optimization research on the D-lactic acid fermentation process, including maintaining the anaerobic fermentation environment with nitrogen gas, regulating the fermentation pH value, replacing neutralizers, replacing Inexpensive nitrogen source, combined optimization of the above fermentation conditions, to obtain the optimal fermentation process route of D-lactic acid; dynamic detection of intracellular metabolites under various fermentation process conditions, to obtain dynamic changes in intracellular metabolism of D-lactic acid fermentation strains Characteristic curve: The mathematical model of WGCNA is used to statistically analyze the characteristics of intracellular metabolites determined in step 2), and obtain significant metabolic modules and Hubs metabolites highly correlated with each fermentation condition. The present invention reveals for the first time the intracellular metabolic modules and core metabolites of significant differences in the fermentation of D-lactic acid by Lactobacillus delbrueckii under different fermentation conditions by means of systems biology, and guides subsequent strain promotion and fermentation optimization.

Description

WGCNA's method for identifying significant modules and Hubs metabolites during D-lactic acid fermentation

technical field

The invention belongs to the field of lactobacillus systems biology, and relates to a kind of Lactobacillus delbrueckii (Lactobacillus delbrueckii, preserved in the laboratory, purchased from China General Microorganism Culture Collection Management Center, strain number CGMCC1.2624) under various fermentation process conditions, Using weighted correlation network analysis (WGCNA for short) to analyze metabolomics data, so as to identify significantly different metabolic modules and core (Hubs) metabolites.

Background technique

D-lactic acid has a molecular formula of CH ₃ CHOHCOOH, a relative molecular weight of 90.08, and a CAS number of 10326-41-7. It is the most common metabolite and is widely used in food, chemical, agricultural and pharmaceutical industries. It is one of the three major One of the traditional organic acids. As a chiral center, D-lactic acid is the precursor of various chiral substances, and is widely used in the chiral synthesis of pharmaceuticals, high-efficiency and low-toxicity pesticides and herbicides, cosmetics and other fields. Another more important use of D-lactic acid is as a raw material monomer for the synthesis of polylactic acid (PLA) materials. Polylactic acid, a renewable and recyclable emerging plastic that has received widespread attention in recent years, is a potential substitute for petroleum-based plastics.

At present, the main method of producing D-lactic acid is microbial fermentation. Since the production of D-lactic acid by biological fermentation has not been around for a long time, most of the research on D-lactic acid fermentation at home and abroad is still in the extracellular research stage, such as the evolution of strains, the improvement of medium and the optimization of fermentation operating conditions. - A systematic study of the intracellular metabolic mechanism of lactic acid fermentation has not been reported.

As an important part of systems biology, metabolomics is another emerging discipline after genomics and proteomics. For industrial microorganisms, in the process of strain transformation and regulation of fermentation behavior, multiple metabolic functional areas or metabolic pathways of cells will change simultaneously, and the single analysis of any single point has limitations, and it is impossible to fully understand the bacteria mechanism of physiological metabolic changes. Only by analyzing from a systematic point of view can we have a clearer understanding of the metabolic behavior characteristics of microorganisms and clarify their metabolic mechanisms. Therefore, through metabolomics analysis, key metabolites or key metabolic pathways that are highly correlated with the synthesis of target products can be determined, thereby providing practical theoretical guidance for strain transformation, process optimization, and process regulation. Metabolomics technology is more and more widely used in agriculture, environmental science, human health, etc., and there are more and more related invention patents. For example, Liu Pengfei from China Agricultural University and others invented a method using metabolomics to reveal the mechanism of action of fungicides (a method for the study of fungicide action mechanisms based on metabolomics. Publication number: CN107796884A), based on GC-MS technology Metabolome detection was carried out on the mycelium of pathogenic bacteria cultured on the medium containing fungicides and no drug control medium, and the differential metabolites were found by comparative metabolomics research method, and the metabolites shared by different mechanisms of action were excluded, so as to obtain the specificity of the action mechanism of the target fungicide. Biomarkers, using changes in the content of biomarkers to realize the identification and high-throughput analysis of the mechanism of action of fungicides. The invention also discloses that the biomarker of the succinate dehydrogenase inhibitor is succinic acid, and the biomarker screening method provided by the invention is fast and reliable, and the obtained biomarker has strong specificity. Rapid identification and high-throughput analysis of the mechanism of action of active substances can be achieved by using specific changes in biomarkers. Jia Xiaoqiang from Tianjin University and others invented a method to improve the ability of Rhodococcus to degrade cycloaromatic pyrene by using metabolomics method (based on metabolomics to improve the degradation conditions of Rhodococcus to increase the degradation rate of polycyclic aromatic hydrocarbon pyrene. Publication number: CN107796906A ), using metabolomics methods combined with significant difference analysis to reveal the differences in the metabolites of Rhodococcus degrading PAH pyrene, and further explain the metabolic change mechanism related to the degradation of PAH pyrene. By analyzing the pyrene-degrading ability and metabolic level of Rhodococcus in different growth environments, metabolites related to pyrene degradation are found, thereby optimizing its degradation conditions and providing directions for improving the degradation rate of polycyclic aromatic hydrocarbon pyrene. This method can also be used It provides new ideas and methods for the optimization research of other microbial degradation conditions for the degradation of polycyclic aromatic hydrocarbons. Honkonen et al. of the American Procter & Gamble Company invented a method for analyzing the health status of human skin using metabolomic methods (Metabonomic methods to assess health of skin. Patent No. 7,761,242). Samples collected at different sites or at different times were used to diagnose skin conditions or evaluate the effects of various skin treatments, identifying biomarkers highly correlated with skin health. At present, there are no invention patents on the application of metabolomics in the production process of D-lactic acid.

For complex metabolome data analysis, in addition to traditional unsupervised methods and supervised methods, a newly emerging method called weighted correlation network analysis (WGCNA) has also been applied in metabolome data analysis. With the development of mass spectrometry detection technology, metabolomics analysis produces a large amount of detection data. When using traditional statistical analysis methods to analyze metabolomics data, we are faced with a huge challenge, that is, how to convert a large amount of complex group data into biological information of academic significance. First of all, it is difficult for traditional statistical analysis methods to identify metabolites with small multiples of change, resulting in deviations in the analysis results; second, traditional analysis methods are more suitable for pairwise analysis and comparison of samples, and cannot simultaneously analyze high levels of samples from multiple sources. Systematic comprehensive analysis of flux data. WGCNA is an algorithm for mining module information from high-throughput data, which breaks through the constraints of the above two aspects. WGCNA was originally used for transcriptome data analysis. In this method, a module is defined as a group of genes with similar expression profiles. If some genes always have similar expression changes in a physiological process or in different tissues, then We have reason to think that these genes are functionally related, and they can be defined as a module. This seems to be somewhat similar to the results obtained by cluster analysis, but the difference is that the clustering criteria of WGCNA have biological significance, rather than conventional clustering methods, so the results obtained by this method have higher credibility . At present, there are no invention patents regarding the application of WGCNA in metabolomics data analysis.

Therefore, the present invention adopts the method of metabolomics combined with weighted correlation network analysis model (WGCNA) for the first time to identify the significant metabolic modules and Hubs metabolites of L. The improvement of strains and the further optimization of fermentation process conditions.

Contents of the invention

The purpose of the present invention is to provide a method of applying weighted correlation network analysis (WGCNA) to metabolomics data analysis, thereby identifying significant difference metabolic modules and core (Hubs) metabolites; the specific steps are as follows:

1) Carry out research on the optimization of D-lactic acid fermentation process, including maintaining the anaerobic fermentation environment with nitrogen gas, adjusting the fermentation pH value, replacing neutralizers, replacing cheap nitrogen sources, and combining and optimizing the above fermentation conditions to obtain the optimal D-lactic acid Fermentation process route;

2) Dynamically detect the intracellular metabolites under various fermentation process conditions in step 1), and obtain the dynamic change characteristic curve of the intracellular metabolism of the D-lactic acid fermentation strain;

3) The mathematical model of WGCNA was used to statistically analyze the characteristics of intracellular metabolites determined in step 2), and significant metabolic modules and Hubs metabolites highly correlated with various fermentation conditions were obtained.

In the step 1), optimize the fermentation process conditions for the production of D-lactic acid by L.delbrueckii (purchased from China General Microorganism Culture Collection and Management Center, strain number CGMCC1.2624) on a 7.5L tank, and feed sterile nitrogen 0.5h, remove the initial dissolved oxygen in the fermentation medium; replace the neutralizer from calcium carbonate to calcium hydroxide and sodium hydroxide to control the fermentation pH, and optimize the fermentation pH to 5.9; replace the nitrogen source from beef extract to Complex nitrogen source of peptone and yeast powder.

In the step 2), GC-MS and LC-MS/MS techniques are used to dynamically sample the fermentation broth under different fermentation process conditions in the step 1) to detect intracellular metabolites, and perform processing and multivariate statistical analysis.

The specific instructions are as follows:

1) Carry out research on the optimization of D-lactic acid fermentation process, including continuously feeding sterile nitrogen for 0.5h to maintain the anaerobic fermentation environment, adjusting the pH value of the fermentation to 5.9, replacing the neutralizer with calcium hydroxide or sodium hydroxide, and reducing expensive The nitrogen source beef extract replaces the cheap nitrogen source peptone and the compound nitrogen source of yeast powder, and the above fermentation conditions are combined and optimized to obtain the optimal fermentation process route of D-lactic acid. Detailed fermentation process conditions description is shown in Table 1;

2) Using gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS/MS) technology platforms, dynamic detection of intracellular metabolites under various fermentation process conditions in 1) to obtain D-lactic acid fermentation The characteristic curve of the dynamic change of the intracellular metabolism of the strain;

3) The mathematical model of weighted correlation network analysis (WGCNA) was used to statistically analyze the characteristics of intracellular metabolites determined in 2), and significant metabolic modules and Hubs metabolites highly correlated with various fermentation conditions were obtained.

Table 1 Description of fermentation process conditions

The present invention relates to a weighted correlation network analysis (WGCNA for short) analysis of metabolomics data to identify significantly different metabolic modules and core (Hubs) metabolites. The present invention reveals for the first time that Lactobacillus delbrueckii (Lactobacillus delbrueckii, preserved in the laboratory, purchased from the China General Microorganism Culture Collection and Management Center, strain number CGMCC1.2624) can efficiently ferment D - Intracellular significantly different metabolic modules and core (Hubs) metabolites of lactic acid to achieve the purpose of guiding subsequent strain improvement and further optimization of fermentation process conditions.

Description of drawings

Under the optimal fermentation process condition of Fig. 1, the apparent kinetic characteristic curve of D-lactic acid fermentation;

Fig. 2 The analysis diagram of Hubs metabolites and their metabolic correlation under the optimal fermentation process conditions.

Detailed ways

In order to make the purpose, technical scheme and effect of the present invention clearer, the present invention is described in conjunction with accompanying drawing and specific embodiment:

As a preferred mode of the present invention, a method for identifying significant metabolic modules and Hubs metabolites in the D-lactic acid fermentation process in L. delbrueckii is provided by using metabolomics combined with network analysis.

The first step is to optimize the fermentation process conditions for the production of D-lactic acid by L.delbrueckii on a 7.5L tank. The optimized conditions include: feeding sterile nitrogen for 0.5h, excluding the initial dissolved oxygen in the fermentation medium, and creating anaerobic fermentation Environment; replace the neutralizer from calcium carbonate with calcium hydroxide and sodium hydroxide, which can precisely control the fermentation pH, and optimize the fermentation pH to 5.9; replace the nitrogen source from beef extract to a composite nitrogen source of peptone and yeast powder, The cost of fermentation raw materials is greatly reduced; the above fermentation process conditions are combined and optimized to obtain the best D-lactic acid fermentation process route. In the second step, GC-MS and LC-MS/MS technologies are used to dynamically sample and detect intracellular metabolites from the fermentation broth under different fermentation process conditions in the first step, and perform processing and multivariate statistical analysis (PCA); the third In the first step, WGCNA analysis is performed on the metabolome data processed in the second step to identify significant metabolic modules and Hubs metabolites.

Example

1. Optimization of the fermentation process conditions for the production of D-lactic acid by L.delbrueckii

Lactobacillus delbrueckii (Lactobacillus delbrueckii, laboratory preservation, purchased from China Common Microorganisms

Collection Management Center, strain number CGMCC1.2624)

First, inoculate mature L.delbrueckii shake flask seeds into a 7.5L fermenter with 4.5L fermentation medium at a rate of 10% (v/v) (Tween 80 1g/L, peptone 22.5g/L , yeast extract powder 7.5g/L, diammonium hydrogen citrate 2g/L, dipotassium hydrogen phosphate 2g/L, anhydrous sodium acetate 3g/L, anhydrous magnesium sulfate 0.2g/L, manganese sulfate 0.1g/L, Glucose 80g/L, pH=5.9). 42°C, 150rpm fermentation for 48h. The remaining fermentation conditions are shown in Table 1. The fermentation results showed that under the combined fermentation process conditions, the D-lactic acid fermentation yield was as high as 133g/L (Figure 1).

2. Preparation of intracellular metabolite samples

(1) Extraction of intracellular metabolites

1) Take 20mL fermentation broth sample, quickly add an equal volume of -40°C pre-cooled 60% cold methanol-water solution, vortex for 2s, place in -20°C, quench the cells for 5min, and terminate the intracellular metabolic reaction, 4 Centrifuge at 5000 rpm for 10 min at 4 °C, discard the supernatant, wash the cell pellet three times with 0.9% NaCl solution at 4 °C, and centrifuge at 5000 rpm at 4 °C for 5 min after washing, discard the supernatant, retain the cells, and remove the medium components.

2) Grind the washed bacteria with liquid nitrogen in a mortar pre-cooled at -20°C to break the cell wall, weigh about 200 mg of the bacteria in a 1.5 mL EP tube, add 1 mL of extract (50% (v/v ), -40°C methanol solution), after vortex mixing, repeated freezing and thawing with liquid nitrogen 3 times, 1 min each time.

3) Centrifuge the above crude extract at 8000rpm for 5min at 4°C, take the supernatant into a new EP tube, and re-use 0.5mL extract (50% (v/v), -40°C methanol solution), and centrifuged at 8000rpm for 5min at 4°C to take the supernatant, combined the two supernatants, centrifuged at 8000rpm for 5min at 4°C again, and took the supernatant.

4) For the same sample, take 100 μL of cell extract into two clean EP tubes, add 10 μL of succinic d4acid (0.1 mg/mL) and 10 μL of D-sorbitol-13C6 (0.1 mg/mL) respectively as GC-MS and LC- Internal standard for MS/MS detection, after mixing thoroughly, vacuum freeze (-48°C) until the sample is completely dry.

5) Freeze-dried samples were stored in a -80°C ultra-low temperature freezer for later use or directly subjected to sample derivatization.

(2) Derivatization of samples

Prepare 20 mg/mL methoxyammonium hydrochloride pyridine solution, take 50 μL and add it to the completely dried sample, vortex and oscillate to mix, and fully react in the metal bath at 40°C for 90 minutes, then add 80 μL trimethyl base silyl trifluoroacetamide, mix well, and fully react in a metal bath at 40°C for 30 minutes. The fully derivatized sample was filtered through a 0.22 μm oil membrane, and 150 μL was taken into a sample vial for GC-MS and LC-MS/MS analysis.

3. Detection of intracellular metabolites

(1) GC-MS detection

GC-MS system: gas chromatography (Agilent 6890N), mass spectrometry (Agilent 9575C MSD) and autosampler (Agilent 5183-2085). The chromatographic column adopts DB-5MS capillary column (30mm×0.25mm, 0.25μm, Agilent Technologies). The chromatographic conditions were as follows: injection volume 1 μL, split ratio 1:1, inlet temperature and interface temperature 280 °C, carrier gas (helium, constant pressure), flow rate 1.0 mL/min. Column temperature rise program: keep at 70°C for 2 minutes; raise the temperature to 290°C at 5°C/min and keep for 3 minutes. Mass spectrometry parameters: Electron bombardment ion source is used, the electron bombardment energy is 70eV, the ion source temperature is 250°C, the current is 40μA, and the scanning range is 50-650m/z.

(2) LC-MS/MS detection of intracellular metabolites

In GC-MS detection, some intracellular phosphate sugar metabolites in the pentose phosphate pathway (PP pathway) and glycolysis pathway (EMP pathway) cannot be detected, such as glucose 6-phosphate (G6P), 6- Fructose phosphate (F6P), fructose 1,6-diphosphate (F-1,6-P), glyceraldehyde 3-phosphate (3PG), phosphoenolpyruvate (PEP), ribulose 5-phosphate (RiBP ), ribose 5-phosphate (R5P), xylulose 5-phosphate (X5P), sedoheptulose 7-phosphate (S7P) and erythrose 4-phosphate (E4P). Therefore, for the detection of the above metabolites, the LC-MS/MS system method was used for detection in this study. Leu, The Netherlands) and API-3000 tandem mass spectrometer (PE-Sciex) equipped with an electronic ion spray source (Turbo Ion Spray) for mass spectrometry identification, where the ion source parameters are: ion spray voltage -2500V, ion source temperature 400 ° C , the spray airflow and collision airflow were set to 10 and 4, respectively.

4. Metabolite data processing and multivariate statistical analysis

(1) The mass spectrogram detected by GC-MS adopts AMDIS (Version3.2, National Institute of Standards and Technology, Gaithersburg, MD, USA), NIST (National Institute of Standards and Technology mass spectral library, 2005) and GMD (Golm Metabolome Database) Agilent software such as Agilent first performs deconvolution, noise reduction, peak area integration and metabolite identification, and then performs derivatization group processing on each identified metabolite to obtain the data of each metabolite. 80 intracellular metabolites were co-identified and semi-quantified by GC-MS.

(2) LC-MS/MS detection data is quantitatively/qualitatively analyzed using the corresponding metabolite standard measurement data, and each metabolite and internal standard are analyzed by Window NT software (Ver.1.3.1), Finally the concentration of each metabolite was obtained. 10 intracellular metabolites were identified and semi-quantified by LC-MS/MS.

(3) All metabolites detected by GC-MS were normalized by dividing the metabolite peak area by the dry weight of the sample and the peak area of the corresponding internal standard, and the absolute concentration of the metabolites detected by LC-MS/MS The same method was used for standardization. The two sets of data were combined for normalization and scaling, and the SIMCA-P package (Ver 11.5; Umetrics, Umea, Sweden) was used for principal component analysis. Five independent biological repetitions were performed for each sample, and the data were expressed as the mean plus standard deviation. The results of principal component analysis showed that there were huge differences in intracellular metabolism under different fermentation conditions.

5. Construction of WGCNA method

According to WGCNA standard operating procedures, a correlation network analysis model was constructed for the metabolome data detected by GC-MS and LC-MS/MS and normalized. The specific process is as follows: first calculate the weighted Pearson correlation coefficient matrix of the metabolome data, and then convert it into a metabolite connection strength matrix, and use topological overlap (TO) to calculate the degree of correlation between metabolites to obtain metabolite similarity Based on TO, hierarchical clustering of metabolite dissimilarity was carried out, a hierarchical clustering tree was established, and highly correlated metabolites were gathered together by dynamic cutting tree algorithm, and finally 9 metabolites with biological Significant differences in the significance of the metabolic modules are shown in Table 2. The identification of Hub metabolites (Hubs) is based on the significant modules determined by the WGCNA analysis, and the data of the significant metabolic modules are further input into VisANT and Cytoscape software for visualization, and 8 are screened out by sorting the connectivity of metabolites. A metabolite with high connectivity is Hub metabolites (Figure 2).

Table 2: Significant difference mode

6. Seed culture conditions

Get the bacterial strain preserved in 2mL glycerol tube, inoculate in 200mL seed medium ((Tween 80 1g/L, peptone 10g/L, yeast extract powder 5g/L, diammonium hydrogen citrate 2g/L, dipotassium hydrogen phosphate 2g /L, anhydrous sodium acetate 3g/L, anhydrous magnesium sulfate 0.2g/L, manganese sulfate 0.1g/L, glucose 20g/L, pH=5.9), 42 ℃ static culture to logarithmic growth phase, generally At 11-13 hours, the bacterial cells just began to precipitate. The stratification between the bacterial cells and the supernatant was not obvious, and most of the bacterial cells were short rod-shaped during microscopic examination, and some were long rod-shaped in a split state. At this time, the bacterial cells Body OD600 = between 1.5 and 2.1.

7. Determination of D-lactic acid yield and chiral purity

(1) The production of D-lactic acid was detected by high performance liquid chromatography (Agilent 1200, USA). The specific operation is as follows:

1) Take 1mL fermentation broth and centrifuge at 10000rpm for 2min.

2) Take 100 μL of the supernatant, add it to 900 μL of 0.1M dilute sulfuric acid solution, mix well and centrifuge at 10,000 rpm for 5 min.

3) Take the supernatant and pass it through a 0.22 μm water filter membrane for HPLC analysis: A: Chromatographic column HPX-87H (250mm×4.6mm, BioRad, America), mobile phase: 2.5mM sulfuric acid solution, flow rate 0.4 ml/min, column temperature 65°C, detection by parallax monitor, injection volume 20 μL; B: chromatographic column ZOBAX SB-C18 (250mm×4.6mm, Agilent, America), mobile phase: 5mM sulfuric acid solution, flow rate 0.5mL/ min, column temperature 30°C, UV detection wavelength 210nm, injection volume 20μL.

(2) The chiral purity of D-lactic acid was separated and determined by high performance liquid chromatography. The specific operation is as follows:

1) Take 1mL fermentation broth and centrifuge at 10000rpm for 2min.

2) Take 100 μL of the supernatant, add it into 900 μL of water, mix well and centrifuge at 10,000 rpm for 5 min.

3) Take the supernatant and pass it through a 0.22 μm water-based filter membrane for HPLC analysis: Chromatographic column Chiralseparation columns CRS10W (50mm×4.6mm, MCI GEL, Japan), mobile phase: 2mM copper sulfate solution, flow rate 0.5ml /min, the column temperature is 25°C, the UV detection wavelength is 254nm, and the injection volume is 20μL.

The method disclosed and proposed by the present invention using weighted association network analysis to identify significant metabolic modules and Hubs metabolites in the D-lactic acid fermentation process can be realized by those skilled in the art by referring to the content of this article and appropriately changing the conditional route. Although the method of the present invention The technologies and technologies have been described through preferred implementation examples, and those skilled in the art can obviously modify or recombine the methods and technical routes described herein without departing from the content, spirit and scope of the present invention to realize the final preparation technology. In particular, it should be pointed out that all similar substitutions and modifications will be obvious to those skilled in the art, and they are all considered to be included in the spirit, scope and content of the present invention.

Claims (3)

1. The method that 1.WGCNA identifies D-ALPHA-Hydroxypropionic acid fermentation process notable module and Hubs metabolins；It is characterized in that steps are as follows：

   1) D-ALPHA-Hydroxypropionic acid fermentation technology optimization research is carried out, including logical nitrogen maintains anaerobic fermentation environment, regulation and control fermentation pH value, replaces Neutralizer replaces cheap nitrogen source, and above-mentioned fermentation condition is combined optimization, obtains the optimal zymotechnique route of D-ALPHA-Hydroxypropionic acid；

   2) dynamic detection is carried out to the intracellular metabolin under various technological condition for fermentation in step 1), obtains D-ALPHA-Hydroxypropionic acid fermentation strain Intracellular is metabolized dynamic change characterization curve；

   3) mathematical model of WGCNA is used to carry out statistics parsing to the intracellular metabolites characteristic measured in step 2), obtain with The highly relevant notable metabolism module of each fermentation condition and Hubs metabolins.
2. 2. the method as described in claim 1 carries out L.delbrueckii productions it is characterized in that in step 1) on 7.5L tanks The optimization of fermentation condition of D-ALPHA-Hydroxypropionic acid is passed through sterile nitrogen 0.5h, excludes the initial dissolution oxygen in fermentation medium；It will neutralize Agent is replaced with the calcium hydroxide and sodium hydroxide of control fermentation pH by calcium carbonate respectively, and it is 5.9 to optimize fermentation pH；By nitrogen source by Beef extract is changed to the compound nitrogen source of peptone and yeast powder.
3. 3. the method as described in claim 1, it is characterized in that using GC-MS and LC-MS/MS technologies in step 1) in step 2) Zymotic fluid under different fermentations process conditions carries out Dynamic sampling and detects intracellular metabolin, and carries out processing and multivariate statistics point Analysis.