Diagnostic marker for metabolic syndrome and preclinical stage thereof and application of diagnostic marker

Abstract

The invention provides a diagnostic marker for metabolic syndrome and preclinical stage thereof, and the diagnostic marker or a plasma small molecule metabolism biomarker consists of the following substances: pyrimidine, (S)-beta-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidine acetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, piperidine acid, hypotaurine and creatinine. The beverage is prepared from phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, lemonic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid and D-glucose. The invention also provides a kit for detecting the plasma small molecule metabolism biomarker composition, the spectrum peak intensity of the obtained plasma small molecule metabolism biomarker is detected, the risk score of whether each subject is MetS or pre-MetS is calculated, and then whether the subject is MetS or pre-MetS is predicted. According to the invention, efficient and rapid diagnosis of the metabolic syndrome and the preclinical stage thereof is realized.

Description

[0001] This application is a divisional application of an invention patent application with an application date of February 14, 2023, application number 202310109934.8, and invention name “A plasma small molecule metabolic marker for the diagnosis of metabolic syndrome and its preclinical stage and its application”. Technical Field

[0002] The present invention belongs to the field of biomedicine, and specifically relates to diagnostic markers of metabolic syndrome and its preclinical stage and their application. Background Art

[0003] In recent years, with the development of my country's social economy and the changes in people's lifestyle and dietary structure, the incidence of metabolic syndrome (MetS) and its preclinical stage (pre-MetS) has continued to rise. The prevalence of MetS in men and women increased from 25.8% and 18.0% in 2008 to 31.0% and 36.8% in 2010, respectively. [1] As a chronic progressive metabolic disorder, MetS and pre-MetS are considered to be independent risk factors for acute cardiovascular and cerebrovascular events. [2] and is associated with increased incidence of several cancers, cardiovascular disease, and neurodegenerative diseases [3-5] The number of people suffering from MetS and pre-MetS and their harm are constantly expanding, seriously threatening national health and increasing social costs, and have become one of the important public issues in my country.

[0004] The current diagnosis of MetS is based on the simultaneous identification of at least three of a group of known risk factors, including obesity, hypertension, hyperglycemia, and dyslipidemia. [6] At present, the most widely used diagnostic criteria in Chinese population studies are those recommended by the Diabetes Branch of the Chinese Medical Association in 2004 (CDS2004). MeS can be diagnosed if three or more of the following four items are present: ① Overweight and / or obesity: Body mass index (BMI) ≥ 25.0 kg / m 2;② Hyperglycemia: fasting blood glucose (FPG) ≥ 6.1mmol / L (110mg / dl) and (or) two-hour postprandial blood glucose (2hPG) ≥ 7.8mmol / L (140mg / dl), and (or) diagnosed with diabetes and treated;③ Hypertension: systolic blood pressure (SBP) / diastolic blood pressure (DBP) ≥ 140 / 90mmHg, and (or) confirmed with hypertension and treated;④ Dyslipidemia: fasting serum triglycerides (TG) ≥ 1.7mmol / L (150mg / dl), and (or) fasting blood high-density lipoprotein (HDL-C) < 0.9mmol / L (35mg / dl) (male) or < 1.0mmol / L (39mg / dl) (female). The diagnosis of pre-MetS refers to the presence of any 1-2 of the above 4 items.

[0005] The above guidelines recommend that there are as many as 8 clinical test indicators involved, and the range of test items is very complicated. In addition to physical examinations such as height, weight, and blood pressure, multiple blood biochemical tests are also required (including: blood glucose metabolism tests such as FPG and 2hPG, and blood lipid metabolism tests such as TG and HDL-C). Physical examination indicators are often limited by measurement accuracy and manual testing techniques, and have high measurement errors. In addition, multiple biochemical test items have different requirements for the fasting / eating status of the subjects, and their cutoff values ​​for jointly judging whether MetS and pre MetS occur are relatively complex. Therefore, although metabolic screening of community populations has received widespread attention, due to the complexity of MetS and pre MetS diagnosis, there is currently a lack of high-throughput, convenient and efficient MetS and pre MetS diagnostic tool platforms to promote precision medicine and management of MetS and pre MetS.

[0006] References:

[0007] [1]LU J,WANG L,LI M,et al.Metabolic Syndrome Among Adults in China:The 2010 China Noncommunicable Disease Surveillance[J].J Clin EndocrinolMetab,2017,102(2):507-15.

[0008] [2]WANG Y, LI J, ZHENG X, et al. Risk Factors Associated With Major Cardiovascular Events 1Year After Acute Myocardial Infarction[J]. JAMA NetwOpen, 2018, 1(4):e181079.

[0009] [3]ESPOSITO K, CHIODINIP, COLAO A, et al. Metabolic syndrome and risk of cancer: a systematic review and meta-analysis[J]. Diabetes Care, 2012, 35(11):2402-11.

[0010] [4]CHEN H, ZHENG

[0011] [5]WIECKOWSKA-GACEK A, MIETELSKA-POROWSKA A, WYDRYCH M, et al. Westerndiet as a trigger of Alzheimer's disease: From metabolic syndrome and systemicinflammation to neuroinflammation and neurodegeneration[J]. Ageing Res Rev, 2021,70:101397.

[0012] [6] Du Guoli, Xu Jing, Zhang Li, et al. Epidemiological survey of metabolic syndrome in Xinjiang Kazakhs and comparative study of three diagnostic criteria [J]. Chinese General Practice, 2012, 15(24): 2762-4,9.

[0013] In recent years, metabolomics based on comprehensive analysis of plasma small molecule metabolites has made revolutionary progress in the field of in vitro diagnosis of diseases. In the framework of systems biology of genome-transcriptome-proteome-metabolome, metabolome is the stage closest to phenotype in the biological dynamic regulation system, and is the essential characteristic and material basis of life. MetS and pre-MetS patients show different metabolic models from healthy people due to the disorder of their body's metabolic state. There are currently no reports on plasma small molecule metabolic biomarkers for the diagnosis of pre-MetS patients. The plasma small molecule metabolite detection methods currently available for MetS diagnosis include two categories: nuclear magnetic resonance (NMR) technology and mass spectrometry (MS) technology. NMR technology is limited by its detection sensitivity, and it is difficult to distinguish the overlap of signals and noise, and it is difficult to quantify metabolic signals in the detection and extraction of plasma metabolite peaks. Therefore, it is difficult to achieve plasma metabolite peak extraction of MetS and pre-MetS using NMR technology. Among MS technologies, the most commonly used gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS) require complex sample pretreatment steps before metabolic signal detection, including desalting, protein removal, derivatization, concentration, etc. At the same time, the reagent consumption and labor cost caused by the complex pretreatment steps, the long pretreatment time, and the large amount of plasma samples required limit the detection throughput of GC / LC-MS. Therefore, it is also difficult to achieve plasma metabolite peak extraction of metabolic syndrome using GC / LC-MS technology.

[0014] Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) is a new technology that has emerged in recent years. It forms co-crystals by mixing the analyte with the matrix material. Under laser irradiation, the analyte is ionized due to the laser energy transmitted by the matrix, so that subsequent mass spectrometry analysis can obtain the peak intensity. Compared with the above-mentioned traditional technology, MALDI-MS has the characteristics of high sensitivity, high detection flux, convenient sample pretreatment, fast analysis speed, and high result tolerance. However, there is currently no report on the detection of MetS and pre-MetS plasma metabolite peak intensity based on MALDI-MS. Based on the deficiencies of the above-mentioned prior art, the present invention is specially proposed. Summary of the invention

[0015] In response to the above-mentioned technical problems in the prior art, the present invention provides a plasma small molecule metabolic biomarker for the diagnosis of metabolic syndrome and its preclinical stage and its application. The plasma small molecule metabolic biomarker for the diagnosis of metabolic syndrome and its preclinical stage and its application is intended to solve the technical problem that the prior art is not effective in diagnosing metabolic syndrome and its preclinical stage.

[0016] The present invention provides a plasma small molecule metabolic biomarker for metabolic syndrome and preclinical diagnosis thereof, which is composed of the following substances: pyrimidine, (S)-β-aminoisobutyric acid ((S)-beta-Aminoisobutyric acid), 3-hydroxybutyric acid ((R)-3-Hydroxybutyric acid), L-serine (L-Serine), guanidineacetic acid (Guanidoacetic acid), 3-hydroxyisovaleric acid (3-Hydroxyisovaleric acid), L-homoserine (L-Homoserine), L-cysteine ​​(L-Cysteine), pipecolic acid (Pipecolic acid), hypotaurine (Hypotaurine), creatinine (Creatinine), phenylacetic acid (Phenylacetic acid), erythritol (Erythritol), picolinic acid (Picolinic acid), thiazolidinone (Thiazolidinone), pyridoxine (Pyridoxine ... acid), Taurine, Pyroglutamic acid, Citracanic acid, Malic acid, Tyramine, Isonicotinic acid, O-Phosphoethanolamine, L-Glutamine, Salicyluric acid, D-Glucose.

[0017] The present invention also provides a kit for detecting the above-mentioned plasma small molecule metabolic biomarkers, including a reagent for detecting the above-mentioned small molecule metabolic biomarkers, and using the reagent to detect the spectrum peak intensity of the plasma small molecule metabolic biomarkers in the subject using a matrix-assisted laser desorption ionization time-of-flight mass spectrometer, and calculating the risk score of whether each subject is MetS or pre-MetS by the following formula:

[0018] Log(P)=-2.69*creatinine peak intensity+1.04*phenylacetic acid peak intensity+1.72*erythritol peak intensity-1.76*picolinic acid peak intensity+2.55*taurine peak intensity+0.48*pyroglutamic acid peak intensity+0.22*citraconic acid peak intensity-0.61*malic acid peak intensity+1.01*tyramine peak intensity-0.67*isonicotinic acid peak intensity+0.23*o-phosphoethanolamine peak intensity+0.14*L-glutamine peak intensity+0.4*salicylic acid peak intensity+ 4.12*D-glucose peak intensity+0.38*pyrimidine peak intensity-0.35*(S)-β-aminoisobutyric acid peak intensity-1.08*3-hydroxybutyric acid peak intensity+2.59*L-serine peak intensity+0.44*guanidineacetic acid peak intensity-1.75*3-hydroxyisovaleric acid peak intensity-0.05*L-homoserine peak intensity-1.26*L-cysteine ​​peak intensity-0.17*piperidinic acid peak intensity+1.21*taurine peak intensity+0.038*age+0.066*sex;

[0019] Among them, gender is 1 for male and 0 for female;

[0020] The correction coefficient for age is 0.038, and the correction coefficient for gender is 0.066;

[0021] The predicted probability of whether each subject is MetS or pre-MetS:

[0022]

[0023] Among them, when Pro≤0.5, it indicates that the subjects are metabolically healthy;

[0024] When 0.5<Pro≤0.8, it indicates that the subjects are pre-MetS population;

[0025] When Pro>0.8, it indicates that the subject belongs to the MetS population.

[0026] In one embodiment, the present invention contains metabolite components present in a biological sample of a subject of a plasma metabolic fingerprint. The subject is a human, the biological sample is derived from the subject's plasma, and the reagents include deionized water and 1 ng / nL iron nanomatrix solution.

[0027] The present invention also provides an application of the spectrum peak intensity of plasma small molecule metabolic biomarkers in subjects obtained by matrix-assisted laser desorption ionization time-of-flight mass spectrometry in the preparation of products for predicting or detecting metabolic syndrome and its preclinical stage, wherein the plasma small molecule metabolic biomarkers are: pyrimidine, (S)-β-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidineacetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, pipecolic acid, hypotaurine, creatinine, phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, citraconic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid, D-glucose, and the following formula is used to calculate the risk score of whether each subject is MetS or pre-MetS:

[0028] Log(P)=-2.69*creatinine peak intensity+1.04*phenylacetic acid peak intensity+1.72*erythritol peak intensity-1.76*picolinic acid peak intensity+2.55*taurine peak intensity+0.48*pyroglutamic acid peak intensity+0.22*citraconic acid peak intensity-0.61*malic acid peak intensity+1.01*tyramine peak intensity-0.67*isonicotinic acid peak intensity+0.23*o-phosphoethanolamine peak intensity+0.14*L-glutamine peak intensity+0.4*salicylic acid peak intensity+ 4.12*D-glucose peak intensity+0.38*pyrimidine peak intensity-0.35*(S)-β-aminoisobutyric acid peak intensity-1.08*3-hydroxybutyric acid peak intensity+2.59*L-serine peak intensity+0.44*guanidineacetic acid peak intensity-1.75*3-hydroxyisovaleric acid peak intensity-0.05*L-homoserine peak intensity-1.26*L-cysteine ​​peak intensity-0.17*piperidinic acid peak intensity+1.21*taurine peak intensity+0.038*age+0.066*sex;

[0029] Among them, gender is 1 for male and 0 for female;

[0030] The correction coefficient for age is 0.038, and the correction coefficient for gender is 0.066;

[0031] The predicted probability of whether each subject is MetS or pre-MetS:

[0033] ;

[0034] Among them, when Pro≤0.5, it indicates that the subjects are metabolically healthy;

[0035] When 0.5<Pro≤0.8, it indicates that the subjects are pre-MetS population;

[0036] When Pro>0.8, it indicates that the subject belongs to the MetS population.

[0037] The present invention also provides the use of the above-mentioned plasma small molecule metabolic biomarkers in the preparation of a kit for diagnosing or detecting metabolic syndrome and its preclinical stage. The method of using the kit includes diagnosing or detecting whether the subject suffers from metabolic syndrome or has a preclinical stage of metabolic syndrome by testing the spectral peak intensity of the plasma small molecule metabolic biomarkers in the subject's plasma and correcting clinical indicators such as age and gender.

[0038] In one embodiment, the method for using the kit is to detect characteristic plasma small molecule metabolic biomarkers in the subject's plasma using the MALDI-MS method.

[0039] The present invention uses the MALDI-MS method to detect a group of plasma small molecule metabolic biomarkers that can specifically characterize metabolic syndrome and its preclinical stage in a large natural population cohort, and constructs a diagnostic tool based on the least absolute convergence and selection operator (LASSO) model of machine learning to facilitate clinical promotion and application.

[0040] The present invention provides a group of plasma small molecule metabolic biomarkers that can be used for the diagnosis of metabolic syndrome and its preclinical stage, and further provides a kit for the diagnosis of metabolic syndrome and its preclinical stage comprising the group of markers and an application method thereof.

[0041] The present invention realizes accurate spectral peak intensity measurement of peripheral blood metabolic small molecules through matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS), and combines it with machine learning algorithm model construction to achieve efficient and rapid diagnosis of metabolic syndrome and its preclinical stage, providing a new way to evaluate the real changes of human metabolic disorders and contributing to the health management of community populations. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Characteristic plasma small molecule metabolite peaks of MetS patients, pre-MetS patients and healthy controls detected by MALDI-MS.

[0043] Figure 2 This is the consistent signal spectrum of 13,554 plasma samples detected by MALDI-MS.

[0044] Figure 3 This is the small molecule screening analysis of differential metabolites in this example.

[0045] Figure 4 This is the diagnostic efficacy curve analysis of this embodiment.

[0046] Figure 5ROC curve analysis for the diagnosis of metabolic syndrome and its pre-stage using plasma small molecule metabolites.

[0047] Figure 6 To analyze the diagnostic indicators for metabolic syndrome and its early stages based on plasma small molecule metabolites. DETAILED DESCRIPTION

[0048] The following is an explanation of the preferred embodiments of the present invention in conjunction with the accompanying drawings. The preferred embodiments described herein are used to describe and explain the present invention rather than to limit the present invention.

[0049] Experimental methods without specific conditions specified in the examples are generally carried out under conventional conditions, such as those described in textbooks and experimental manuals, or under conditions recommended by manufacturers.

[0050] Example 1: Extraction of characteristic plasma small molecule metabolites of metabolic syndrome

[0051] 1. Research subjects and sample collection

[0052] The study was approved by the Ethics Committee of Renji Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, and subjects enrolled in the Shanghai Pudong natural population cohort from January 2019 to October 2020 were recruited.

[0053] 1.1 Inclusion and Exclusion Criteria:

[0054] Inclusion criteria: According to the diagnostic criteria recommended by the Chinese Medical Association Diabetes Society in 2004 (CDS 2004), patients who meet three or more of the following four criteria can be diagnosed as MeS: ① Overweight and / or obesity: body mass index (BMI) ≥ 25.0 kg / m 2 ;② Hyperglycemia: fasting blood glucose (FPG) ≥ 6.1mmol / L (110mg / dl) and (or) two-hour postprandial blood glucose (2hPG) ≥ 7.8mmol / L (140mg / dl), and (or) diagnosed with diabetes and treated;③ Hypertension: systolic blood pressure (SBP) / diastolic blood pressure (DBP) ≥ 140 / 90mmHg, and (or) confirmed with hypertension and treated;④ Dyslipidemia: fasting serum triglycerides (TG) ≥ 1.7mmol / L (150mg / dl), and (or) fasting blood high-density lipoprotein (HDL-C) < 0.9mmol / L (35mg / dl) (male) or < 1.0mmol / L (39mg / dl) (female). If one or two of the above four items are met, pre-MetS can be diagnosed.

[0055] Exclusion criteria: patients with acute and infectious clinical symptoms within three weeks before sampling, including but not limited to fever, headache, cough, malaise, sore throat, loss of smell, runny nose, abdominal pain and diarrhea.

[0056] According to the inclusion and exclusion criteria, 3504 MetS patients, 7776 pre-MetS patients, and 2274 healthy control subjects were finally included.

[0057] 1.2 Patient testing indicators:

[0058] Basic indicators: age, gender, height, weight, body mass index (BMI), blood pressure (systolic pressure, diastolic pressure);

[0059] Biochemical indicators: fasting blood sugar, two-hour postprandial blood sugar, fasting serum triglycerides, and fasting blood high-density lipoprotein.

[0060] 1.3 Plasma sample collection:

[0061] 1 ml of peripheral venous blood was collected from the patient in a fasting state, centrifuged at 4000 rpm for 10 min, and the supernatant was stored at -80°C for subsequent MALDI-MS determination. MALDI-MS detection only requires 100 nL of plasma sample.

[0062] Example 2: MALDI-MS detection of plasma small molecule metabolite peaks

[0063] 2.1 Main instrument: Autoflex Speed ​​MALDI TOF / TOF matrix-assisted laser desorption ionization time-of-flight mass spectrometer (Bruker, Germany)

[0064] 2.2 Molecular biochemical reagents: deionized water, iron matrix materials

[0065] 2.3 Plasma sample processing:

[0066] 100 nL of the plasma sample to be tested was diluted with deionized water at a ratio of 10:1, and 500 nL of the diluted plasma sample was placed on the MALDI-MS target plate and dried at room temperature.

[0067] 2.4 Preparation of iron nanoparticle matrix:

[0068] A 1 ng / nL iron nanomatrix solution was prepared, and 500 nL of the iron nanomatrix solution was placed on the plasma sample of the MALDI-MS target plate and dried at room temperature.

[0069] 2.5MALDI-MS detection:

[0070] In the positive ion mode, MALDI-MS was used to perform mass spectrometry detection on the iron nanoparticle matrix on the target plate; and the results were preprocessed using Malab software, including data resampling, spectral line smoothing, baseline correction, spectral peak matching, and missing value filling, to finally obtain the spectral peak intensity (intensity) of the full spectrum of small molecule metabolites.

[0071] After comprehensive comparison with the secondary spectral library and the HMDB database (Human Metabolome Database), a total of 303 full-spectrum plasma small molecule metabolites with a mass-to-charge ratio (m / z) less than 300 were obtained, and the characteristic full-spectrum plasma small molecule metabolite peak intensity diagrams of MetS patients, pre-MetS patients and healthy controls were drawn ( Figure 1 ), and obtained characteristic signal spectra of 13554 plasma samples ( Figure 2 ).

[0072] 2.6 Screening of characteristic plasma small molecule metabolites for diagnosis of MetS and pre-MetS

[0073] To screen characteristic plasma small molecule metabolites for the diagnosis of MetS and pre-MetS, R 4.0.5 language was used for data processing and analysis. After testing the normal distribution of the data, the Kruskal-Wallis rank sum test function in R 4.0.5 language was used to evaluate the differentially expressed metabolites between healthy controls (HCs), pre-MetS, and MetS.

[0074] The significance level was set at p < 0.05. At the same time, the Bonferroni correction in R 4.0.5 was used to perform multiple comparisons of the three groups of data, and 24 plasma small molecule metabolites with significant expression differences were filtered out ( Figure 3 ):

[0075] Pyrimidine, (S)-beta-Aminoisobutyric acid, (R)-3-Hydroxybutyric acid, L-Serine, Guanidoacetic acid, 3-Hydroxyisovaleric acid, L-Homoserine, L-Cysteine, Pipecolic acid, Hypotaurine, Creatinine, Phenylacetic acid, Erythritol, Picolinic acid, Taurine, Pyroglutamic acid, Citraconic acid, Malic acid acid), tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid, D-glucose (Table 1).

[0076] Table 1. Characteristic plasma small molecule metabolites differentially expressed among the MetS group, pre-MetS group, and healthy controls

[0077]

[0078] 2.7 Establishment of a MetS diagnostic model based on characteristic plasma small molecule metabolites

[0079] MetS patients, pre-MetS patients and healthy controls who met the inclusion criteria and had complete indicators were screened and matched 1:1 for modeling.

[0080] Among them, in the MetS and healthy control distinction model (a total of 4548 cases were enrolled), 2274 healthy controls and 2274 MetS patients were enrolled; in the pre MetS and healthy control distinction model (a total of 4548 cases were enrolled), 2274 healthy controls and 2274 preMetS patients were enrolled; in the pre MetS and MetS distinction model (a total of 7008 cases were enrolled), 3504 MetS patients and 3504 preMetS patients were enrolled.

[0081] The area under the receiver operating characteristic (ROC) curve (AUC) statistical power analysis was performed using a two-sided z test on the sample size, with AUC0 = 0.75 as the reference and AUC1 = 0.80 as the estimate. The results showed that the diagnostic efficacy reached 100% ( Figure 4 ).

[0082] The diagnostic models of MetS and pre MetS were further established. In each model, the subjects were divided into training and test sets in a ratio of 7:3. Among them, in the differentiation model of MetS and healthy controls, 3184 subjects were included in the training set, and 1364 subjects were included in the test set. In the differentiation model of pre MetS and healthy controls, 3184 subjects were included in the training set, and 1364 subjects were included in the test set. In the differentiation model of pre MetS and MetS, 4906 subjects were included in the training set, and 2102 subjects were included in the test set.

[0083] The minimum absolute value convergence and selection operator and the generalized linear model via Lasso and Elastic-Net Regularization (glmnet) algorithm provided by the glmnet package of R 4.0.5 language were used in combination with 8-fold cross validation to perform machine learning on the peak intensity of the above-mentioned characteristic plasma small molecule metabolites, and the predicted probability of MetS and pre-MetS for each subject was determined, and the correction for age and gender was performed. Table 2 shows the metabolites and their parameters included in the model after correction for gender and age, among which the correction coefficient for age was 0.038 and the correction coefficient for gender was 0.066.

[0084] Table 2. Glmnet analysis of the MetS diagnostic model based on the peak intensity of 24 plasma metabolic small molecules (model parameter Lambda ≈ 0.0003)

[0085] Metabolites English name HMDB library number Model coefficient (β) OR Value Creatinine Creatinine HMDB0000562 -2.69 0.07 Phenylacetic acid Phenylacetic acid HMDB0000209 1.04 2.82 Erythritol Erythritol HMDB0002994 1.72 5.60 Picolinic acid Picolinic acid HMDB0002243 -1.76 0.17 Taurine Taurine HMDB0000251 -2.55 0.08 Pyroglutamate Pyroglutamic acid HMDB0000267 0.48 1.62 Citric Acid Citraconic acid HMDB0000634 0.22 1.25 Malic Acid Ma1ic acid HMDB0000156 -0.61 0.54 Tyramine Tyramine HMDB0000306 1.01 2.73 Isonicotinic acid Isonicotinic acid HMDB0060665 -0.67 0.51 o-Phosphoethanolamine 0-Phosphoethanolamine HMDB0000224 0.23 1.26 L-Glutamine L-Glutamine HMDB0000641 0.14 1.14 Salicylic acid Salicyluric acid HMDB0000840 0.40 1.49 D-Glucose D-Glucose HMDB0000122 4.12 61.71 Pyrimidine Pyrimidine HMDB0003361 0.38 1.46 (S)-β-Aminoisobutyric acid (S)-beta-Aminoisobutyric acid HMDB0002166 -0.35 0.71 3-Hydroxybutyric Acid (R)-3-Hydroxybutyric acid HMDB0000011 -1.08 0.34 L-Serine L-Serine HMDB0000187 2.59 13.37 Guanidineacetic acid Guanidoacetic acid HMDB0000128 0.44 1.56 3-Hydroxyisovaleric acid 3-Hydroxyisovaleric acid HMDB0000754 -1.75 0.17 L-Homoserine L-Homoserine HMDB0000719 -0.05 0.95 L-Cysteine L-Cysteine HMDB0000574 -1.26 0.28 Pipecolic acid Pipecolic acid HMDB0000070 -0.17 0.85 Hypotaurine Hypotaurine HMDB0000965 1.21 3.34

[0086] The glmnet() function built into the glmnet package of R 4.0.5 language was used for machine learning iteration. Its mathematical model is as follows, which can calculate the risk score of whether each subject is MetS or pre-MetS through the peak intensity of 24 plasma metabolic small molecules in the human body and gender and age:

[0087] Log(P)=-2.69*creatinine peak intensity+1.04*phenylacetic acid peak intensity+1.72*erythritol peak intensity-1.76*picolinic acid peak intensity+2.55*taurine peak intensity+0.48*pyroglutamic acid peak intensity+0.22*citraconic acid peak intensity-0.61*malic acid peak intensity+1.01*tyramine peak intensity-0.67*isonicotinic acid peak intensity+0.23*o-phosphoethanolamine peak intensity+0.14*L-glutamine peak intensity+0.4*salicylic acid peak intensity+ 4.12*D-glucose peak intensity+0.38*pyrimidine peak intensity-0.35*(S)-β-aminoisobutyric acid peak intensity-1.08*3-hydroxybutyric acid peak intensity+2.59*L-serine peak intensity+0.44*guanidineacetic acid peak intensity-1.75*3-hydroxyisovaleric acid peak intensity-0.05*L-homoserine peak intensity-1.26*L-cysteine ​​peak intensity-0.17*piperidinic acid peak intensity+1.21*taurine peak intensity+0.038*age+0.066*sex;

[0088] Among them, gender is 1 for male and 0 for female;

[0089] The predicted probability of whether each subject is MetS or pre-MetS is:

[0090]

[0091] The built-in predict() function of the glmnet package in R 4.0.5 is used to iterate and sort the predicted probability (Pro) of each subject, and the threshold range division of the predicted probability can be automatically output:

[0092] When Pro≤0.5, it indicates that the subjects are metabolically healthy;

[0093] When 0.5<Pro≤0.8, it indicates that the subjects are pre-MetS population;

[0094] When Pro>0.8, it indicates that the subject belongs to the MetS population.

[0095] On this basis, AUC was used to evaluate the predictive ability of the occurrence of MetS and pre-MetS in subjects based on the peak intensity of the characteristic plasma small molecule metabolite spectrum ( Figure 5 ).

[0096] The evaluation results based on AUC showed that an effective diagnostic model for MetS and pre-MetS could be established using the spectral peak intensity of this group of characteristic plasma small molecule metabolites.

[0097] Example 3: Model Evaluation

[0098] The AUC for distinguishing healthy controls from pre-MetS was 0.867 in the training set (sample size = 3184 people), with a 95% confidence interval of (0.854–0.881), and 0.864 in the test set (sample size = 1364 people), with a 95% confidence interval of (0.843–0.885). The result of DeLong's test was p = 0.813, indicating that there was no significant difference in the AUC of the model in the training set and the test set.

[0099] The AUC for distinguishing pre-MetS from MetS was 0.849 in the training set (sample size = 4906 people), with a 95% confidence interval of (0.838–0.860), and 0.835 in the test set (sample size = 2102 people), with a 95% confidence interval of (0.817–0.853). The result of DeLong's test was p = 0.195, indicating that there was no significant difference in the AUC of the model in the training set and the test set.

[0100] The AUC for distinguishing healthy controls from MetS was 0.891 in the training set, with a 95% confidence interval of (0.880–0.902), and 0.886 in the test set, with a 95% confidence interval of (0.868–0.903). The result of DeLong's test was p=0.605, indicating that there was no significant difference in the AUC of the model between the training set and the test set.

[0101] The above results show that the MetS and preMetS diagnostic model constructed using the peak intensity of this group of characteristic plasma small molecule metabolites has good diagnostic efficacy and strong generalization ability, and can be further promoted and applied. At the same time, the model can take into account good sensitivity, specificity and accuracy, and shows good positive prediction rate, negative prediction rate and F1 value ( Figure 6 ).

[0102] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A use of a plasma small molecule metabolic biomarker or a detection reagent thereof, characterized in that: Used for preparing a diagnostic reagent or a diagnostic kit, wherein the diagnostic reagent or the diagnostic kit is used for diagnosing metabolic syndrome or preclinical stage of metabolic syndrome; Wherein, the plasma small molecule metabolic biomarker is selected from the following group: pyrimidine, (S)-β-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidineacetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, pipecolic acid, hypotaurine, creatinine, phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, citraconic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid, D-glucose, or a combination thereof.

2. The use according to claim 1, characterized in that The plasma small molecule metabolic biomarkers are: pyrimidine, (S)-β-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidineacetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, pipecolic acid, hypotaurine, creatinine, phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, citraconic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid and D-glucose.

3. The use according to claim 1, characterized in that The kit has the following record: Based on the detection reagent, the plasma small molecule metabolic biomarker is detected using a matrix-assisted laser desorption ionization time-of-flight mass spectrometer.

4. The use according to claim 3, characterized in that The detection includes using the matrix-assisted laser desorption ionization time-of-flight mass spectrometer to detect and obtain the spectral peak intensity of the plasma small molecule metabolic biomarkers in the subject.

5. The use according to any one of claims 1 to 3, characterized in that The subjects are humans, and the biological samples are derived from the subjects’ plasma; And the detection reagent is used to detect the plasma small molecule metabolic biomarker, and the detection reagent includes deionized water and 1 ng / nL iron nanomatrix solution.

6. The use according to claim 1, characterized in that The kit also records: calculating the risk score of whether each subject has metabolic syndrome (MetS) or preclinical stage of metabolic syndrome (pre-MetS) based on the spectral peak intensity of the plasma small molecule metabolic biomarker, and obtaining the predicted probability of whether each subject has MetS or pre-MetS based on the risk score.

7. The use according to claim 6, characterized in that The risk score calculation formula is: Log(P)=-2.69*creatinine peak intensity+1.04*phenylacetic acid peak intensity+1.72*erythritol peak intensity-1.76*picolinic acid peak intensity+2.55*taurine peak intensity+0.48*pyroglutamic acid peak intensity+0.22*citraconic acid peak intensity-0.61*malic acid peak intensity+1.01*tyramine peak intensity-0.67*isonicotinic acid peak intensity+0.23*o-phosphoethanolamine peak intensity+0.14*L-glutamine peak intensity+0.4*salicylic acid peak intensity+ 4.12*D-glucose peak intensity+0.38*pyrimidine peak intensity-0.35*(S)-β-aminoisobutyric acid peak intensity-1.08*3-hydroxybutyric acid peak intensity+2.59*L-serine peak intensity+0.44*guanidineacetic acid peak intensity-1.75*3-hydroxyisovaleric acid peak intensity-0.05*L-homoserine peak intensity-1.26*L-cysteine ​​peak intensity-0.17*piperidinic acid peak intensity+1.21*taurine peak intensity+0.038*age+0.066*sex; Among them, gender is 1 for male and 0 for female; The correction coefficient for age is 0.038, and the correction coefficient for gender is 0.

066.

8. The use according to claim 6, characterized in that The predicted probability Pro of whether each subject is MetS or pre-MetS is: Among them, when Pro≤0.5, it indicates that the subjects are metabolically healthy; When 0.5<Pro≤0.8, it indicates that the subjects are pre-MetS population; When Pro>0.8, it indicates that the subject belongs to the MetS population.

9. A plasma small molecule metabolic biomarker set for diagnosis of metabolic syndrome or preclinical diagnosis of metabolic syndrome, characterized in that: The plasma small molecule metabolic biomarker set is: pyrimidine, (S)-β-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidineacetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, pipecolic acid, hypotaurine, creatinine, phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, citraconic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid and D-glucose.

10. A method for constructing a diagnostic model for metabolic syndrome or preclinical stage of metabolic syndrome, characterized in that: The following steps are involved: (s1) providing a data set for model training, the data set comprising characteristic full-spectrum plasma small molecule metabolite spectrum peak intensity data of the following subjects: patients with metabolic syndrome, patients with preclinical stage of metabolic syndrome and healthy control population; Providing a data set for model validation, the data set includes characteristic full-spectrum plasma small molecule metabolite spectrum peak intensity data of the following subjects: patients with metabolic syndrome, patients with preclinical stage of metabolic syndrome and healthy control population; (s2) using a machine learning method to select a machine learning model to train the data in the data set for model training, thereby obtaining characteristic plasma small molecule metabolites for the diagnosis of MetS and pre-MetS, thereby obtaining the metabolic syndrome and its preclinical diagnostic model; and (s3) validating the diagnostic model on the data set for model validation; Wherein, the characteristic plasma small molecule metabolites are selected from the following group: pyrimidine, (S)-β-aminoisobutyric acid, 3-hydroxybutyric acid, L-serine, guanidineacetic acid, 3-hydroxyisovaleric acid, L-homoserine, L-cysteine, pipecolic acid, hypotaurine, creatinine, phenylacetic acid, erythritol, picolinic acid, taurine, pyroglutamic acid, citraconic acid, malic acid, tyramine, isonicotinic acid, o-phosphoethanolamine, L-glutamine, salicylic acid, D-glucose, or a combination thereof.

Images