Salivary metabolic marker combination for diagnosis of COPD and screening method thereof

Abstract

The present invention provides a combination of salivary metabolic markers for diagnosing COPD and a screening method thereof. The saliva of the subject can be analyzed to determine whether the subject suffers from COPD, and the sample is collected in the safest, non-invasive and rapid manner, thereby simplifying the COPD examination and analysis process. Furthermore, a mathematical analysis method is used to screen out reliable differential saliva metabolite combinations from multiple angles, and the influence of periodontal disease on COPD is excluded, thereby having a high degree of credibility.

Description

Technical Field

[0001] The invention relates to the field of biomedical technology, in particular to a combination of COPD-related metabolites in saliva and a screening method thereof. Background Art

[0002] Chronic obstructive pulmonary disease (COPD) is a progressive chronic disease characterized by persistent, incompletely reversible airflow limitation. Its main clinical manifestations are chronic cough, chest tightness, and dyspnea. COPD ranks fourth among the leading causes of death in the world.

[0003] Periodontitis (PD) is a disease characterized by inflammation of the periodontal tissue and alveolar bone resorption. In the late stage, it manifests as loose teeth and loss. It is considered the main cause of tooth loss in adults. Increasing evidence shows that periodontitis is associated with many systemic diseases, including diabetes, osteoporosis, cardiovascular disease and COPD.

[0004] Currently, the diagnosis of COPD still mainly relies on blood tests, imaging tests and lung function tests. However, the high cost and inconvenience of these tests, low sample stability and patient acceptance, and long detection time limit their application in some cases.

[0005] As a non-invasive, convenient and fast detection method, saliva testing has certain advantages, especially in primary care and large-scale screening. Saliva contains a variety of biomolecules, such as proteins, nucleic acids, metabolites, hormones, antibodies and cell fragments. As a non-invasive, convenient and economical detection method, saliva testing shows great potential in the diagnosis and monitoring of various diseases.

[0006] The levels of inflammatory markers and oxidative stress markers in the saliva of patients with periodontal disease may be similar to those of patients with COPD, which may lead to confusion in the test results; and the local inflammatory response caused by periodontal disease may produce metabolites related to COPD, thereby affecting the specificity and sensitivity of COPD-related biomarkers. In the diagnosis of COPD, there is still a lack of biomarkers with sufficient specificity and high sensitivity.

[0007] Saliva testing has certain potential in the diagnosis and monitoring of COPD, but it has not yet become a mainstream detection method. Therefore, it is crucial to find faster and more accurate metabolites to simplify the analytical process of COPD diagnosis, screen high-risk groups for COPD by performing oral examinations and saliva sample tests on patients with severe periodontal disease, so as to predict the risk of COPD, provide a research entry point for early detection and early diagnosis of COPD, and promote large-scale clinical application of the detection platform. Summary of the invention

[0008] The present invention collects saliva samples from subjects, performs metabolomics analysis on the samples using liquid chromatography, and further screens out saliva metabolic markers using mathematical analysis methods.

[0009] The purpose of the present invention is to screen significantly relevant salivary metabolic markers and establish a more convenient, rapid and accurate determination method.

[0010] In a first aspect, the present invention provides a salivary metabolite marker combination for diagnosing COPD, wherein the metabolites thereof are specifically: lysophospholipid LysoPC (18:2 (9Z, 12Z)), glutamine (N2-gamma-Glutamylglutamine), miglitol and 4-nitrophenol.

[0011] In a second aspect, the present invention provides a method for screening a combination of salivary metabolic markers for diagnosing COPD, comprising the following steps:

[0012] S1. Screening of subjects;

[0013] The subjects were divided into two groups: Group I, normal lung function and Group II, stable COPD;

[0014] S2, collect saliva samples from subjects;

[0015] S3. Perform metabolite mass spectrometry analysis on the subjects’ saliva samples;

[0016] S4. Performing data quality control on the raw data obtained from mass spectrometry analysis;

[0017] S5. Screening differential metabolites, comprising the following steps:

[0018] S5.1, primary screening;

[0019] S5.2, subgroup analysis was performed with periodontal disease;

[0020] S5.3, secondary screening;

[0021] S5.4, tertiary screening;

[0022] S6. Validation of the diagnostic efficacy of differential metabolites.

[0023] Preferably, in step S3, the mass spectrometry is performed in two different measurement modes: positive ion mode and negative ion mode, so as to improve the sensitivity and coverage of the detection;

[0024] Preferably, in step S3, 12 quality control samples are added to ensure the accuracy and precision of the experiment;

[0025] Furthermore, in step S4, missing values ​​and outliers are processed, and the coefficient of variation (Coefficient of Variation) data quality control is performed;

[0026] Furthermore, in step S4, the statistical significance of the data was tested using t-test and Wilcoxon rank sum test (Mann-Whitney U);

[0027] Furthermore, step S5.1 initially screened the differential metabolites, and a total of 261 differential metabolites were screened;

[0028] Preferably, step S5.1 uses principal component analysis, partial least squares discriminant analysis, or orthogonal least squares discriminant analysis to screen data;

[0029] Furthermore, in step S5.1, the significance level of the selected metabolites is set to VIP>1, and P<0.05;

[0030] Furthermore, S5.2 conducted a subgroup analysis of patients with periodontal disease, with the following specific groups:

[0031] The group I includes two subgroups: ① a group with normal lung function and no periodontal disease and ② a group with normal lung function but periodontal disease;

[0032] The group II includes two subgroups, ③ a group with stable COPD and periodontal disease and ④ a group with stable COPD but no periodontal disease;

[0033] Preferably, the stability of the differential metabolites of the two subgroups in the two groups is analyzed by variance analysis, and the differential metabolites with obvious differences between the two subgroups are deleted.

[0034] Periodontitis (PD) is a disease characterized by inflammation of the periodontal tissue and alveolar bone resorption. In the late stage, it manifests as loose teeth and loss. It is considered the main cause of tooth loss in adults. Increasing evidence shows that periodontitis is associated with many systemic diseases, including diabetes, osteoporosis, cardiovascular disease and COPD.

[0035] The levels of inflammatory markers and oxidative stress markers in the saliva of patients with periodontal disease may be similar to those of patients with COPD, which may lead to confusion in the test results; and the local inflammatory response caused by periodontal disease may produce metabolites related to COPD, thereby affecting the specificity and sensitivity of COPD-related biomarkers. In the diagnosis of COPD, there is still a lack of biomarkers with sufficient specificity and high sensitivity.

[0036] In the present invention, the stability of metabolites of the two subgroups, group I and group II, was analyzed by subdividing the subgroups with combined periodontitis, and the metabolites with obvious differences between the two subgroups were deleted, thereby eliminating the influence of periodontal disease on the screening of salivary metabolites for the diagnosis of COPD.

[0037] Furthermore, step S5.3 performs differential metabolite grade scoring on the above data based on the local database, performs secondary screening, and screens out 16 “credible differential metabolites” in total;

[0038] Preferably, based on the matching results of m / z values, level B (known metabolites) and level A (standard metabolites) are screened out as “credible differential metabolites”;

[0039] Furthermore, step S5.4 screened significantly different metabolites, and a total of 4 metabolic markers were screened out;

[0040] Furthermore, step S5.4 calculated the correlation coefficient between the "credible differential metabolites" and the COPD diagnostic threshold (FEV1 / FVC%), and used the multivariate linear regression equation to screen the differential metabolites with significant influence. A total of 4 statistically significant differential metabolites were screened out, namely, lysophospholipid LysoPC (18:2 (9Z, 12Z)), glutamine (N2-gamma-Glutamylglutamine), Miglitol (Miglitol) and 4-nitrophenol (4-Nitrophenol);

[0041] Furthermore, the P values ​​of the correlation coefficients and regression coefficients of the four “credible differential metabolites” were all less than 0.05;

[0042] Furthermore, in step S6, a ROC curve is fitted to the combination of the four metabolite markers, and the AUC is 0.978, that is, the combination has a good efficacy in diagnosing COPD;

[0043] Based on the differential metabolites obtained through screening, COPD patients can be diagnosed or identified.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention simplifies the process of COPD examination and analysis and collects samples in a safe, non-invasive and rapid manner.

[0046] The present invention screens metabolite markers from multiple perspectives including statistics, metabolite database and clinical parameters; screens metabolite markers and fits ROC curves, with an AUC value of 0.978, and the combination of metabolite markers has a high credibility.

[0047] The present invention takes into account the interactive effects of periodontal disease on saliva metabolites of patients with normal lung function and COPD, and adds subgroup analysis experiments, so that the screened metabolic markers have high specificity and sensitivity. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0049] Figure 1 For credible differential metabolites;

[0050] Figure 2 ROC curve of the potential metabolic biomarkers for COPD. DETAILED DESCRIPTION

[0051] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0052] The present invention provides a salivary metabolic marker combination for diagnosing COPD and a screening method thereof, which comprises the following steps:

[0053] S1. Screening of subjects;

[0054] S1.1. A total of 130 subjects with stable COPD and no lung function diseases were screened out from 278 people, including 71 subjects with stable COPD and 59 subjects with normal lung function. COPD patients were selected in the stable stage to reduce the impact of disease fluctuations on the research results;

[0055] S1.2. In order to study the effect of saliva metabolism on COPD determination in patients with periodontal disease and to quickly screen out more accurate metabolites, 130 subjects were subdivided. Among the 130 subjects, 41 had severe periodontal disease and 89 had mild or no periodontal disease.

[0056] S1.3. According to the classification of each subject, the subjects were divided into two groups: Group I, normal lung function and Group II, stable COPD;

[0057] S2, collect saliva samples from subjects;

[0058] S3. Perform metabolite mass spectrometry analysis on the subjects’ saliva samples;

[0059] Mass spectrometry can select two different measurement modes: positive ion mode and negative ion mode, to improve the sensitivity and coverage of detection;

[0060] Twelve quality control samples were added to ensure the accuracy and precision of the experiment. The metabolites in the samples were detected by ultra-high performance liquid chromatography tandem quadrupole orbitrap mass spectrometry (UPLC-Q-Exactive Orbitrap-MS) to obtain the original mass spectrometry data of the metabolites in each saliva sample.

[0061] A total of 5212 compounds in positive ion mode and 2233 compounds in negative ion mode were detected from all samples;

[0062] S4. Performing data quality control on the raw data obtained from mass spectrometry analysis;

[0063] S4.1. Preprocess the mass spectrometry data of metabolites in each saliva sample, perform qualitative and quantitative tests on the data based on local databases (MoNA, HMDB), and then clean the data, i.e., process missing values ​​and outliers, and perform data quality control on the coefficient of variation to ensure the accuracy and reliability of the data.

[0064] S4.2. The data were normalized by median and tested for normality. The t-test and Mann-Whitney U (wilcox.test) nonparametric tests were used to test whether the data were statistically significant.

[0065] S5, data analysis, screening of differential metabolites;

[0066] S5.1. Initial screening of differential metabolites, a total of 261 differential metabolites were screened;

[0067] S5.1.1. Calculate the fold change and P value of the data and create a volcano plot for visual analysis;

[0068] S5.1.2. Perform multivariate statistical analysis and use principal component analysis (PCA) to perform preliminary dimensionality reduction on the data.

[0069] S5.1.3. Use Partial Least Squares Discrimination Analysis (PLS-DA) for further feature extraction and classification;

[0070] S5.1.4. Use orthogonal partial least squares discriminant analysis (OPLS-DA) to separate prediction and orthogonal information, thereby improving the transparency and interpretability of the model, and calculate the VIP value (Variable Importance in Projection) to measure the relative importance of each variable in the model;

[0071] S5.1.5, the above data processing and screening methods screened out a total of 261 differential metabolites;

[0072] The general standard for screening differential metabolites is: VIP value>1, and P value<0.05; the P value (t-test significance value) comes from the univariate analysis method;

[0073] S5.2, subgroup analysis was performed with periodontal disease;

[0074] The specific subgroups were divided into: Group I included two subgroups, ① normal lung function and no periodontal disease group and ② normal lung function but periodontal disease group;

[0075] Group II included two subgroups: ③ COPD stable stage with periodontal disease group and ④ COPD stable stage without periodontal disease group;

[0076] Subgroup analysis was performed by merging periodontal disease and variance analysis was used to analyze the stability of the two subgroups in the two groups. The mean and variance of each metabolite in the periodontitis group and the non-periodontitis group were calculated, respectively. The mean difference threshold was set to 0.1 and the variance threshold was set to 1.1, that is, those metabolites with a mean difference of less than 0.1 and a variance threshold of less than 1.1 were retained, and metabolites with significant differences between the two subgroups were deleted;

[0077] S5.2.1. The metabolites of subjects in group ① and group ② were statistically analyzed by variance analysis to obtain the unstable metabolites in the two groups;

[0078] S5.2.2. The metabolites of the subjects in groups ③ and ④ were statistically analyzed by variance analysis to obtain the unstable metabolites in the two groups;

[0079] The results showed that the mean difference threshold of the two groups of 12 saliva metabolites was greater than 0.1, and the variance threshold was greater than 1.1. Therefore, there were significant differences in these 12 saliva metabolites between the subjects in group ① and group ② or between the subjects in group ③ and group ④.

[0080] Since these 12 differential metabolites were affected by periodontal disease, especially considering that these differential metabolites affected by periodontal disease may interfere with the analysis results of COPD-related metabolites, in order to ensure the reliability and accuracy of subsequent data analysis, these 12 metabolites that had significant effects of periodontal disease on patients with normal lung function and COPD were excluded, and only the remaining 204 differential metabolites were retained for further analysis to explore the differential expression patterns of the remaining metabolites between COPD patients and subjects with normal lung function;

[0081] S5.3, secondary screening, screening out 16 “credible differential metabolites”;

[0082] The above differential metabolites were graded based on the local database and screened for secondary screening, and a total of 16 "credible differential metabolites" were screened out;

[0083] S5.3.1. Extracting the charge-to-mass ratio information: extracting the charge-to-mass ratio (m / z value) of the above 204 differential metabolites from the mass spectrometry data;

[0084] S5.3.2. Database search: Use local databases (MoNA, HMDB) and online metabolite databases to match m / z values;

[0085] S5.3.3. Screening of candidate metabolites: Based on the matching results of m / z values, the differential metabolites in the mass spectrometry data are rated, and level B (known metabolites) and level A (standard metabolites) are screened out. Level A and level B metabolites are considered to be more credible and are called "credible differential metabolites";

[0086] In this step, a total of 16 “credible differential metabolites” were screened out; Figure 1As shown in the figure, group C is COPD subjects, and group N is subjects with normal lung function. It can be seen from the figure that compared with subjects with normal lung function, COPD patients have 16 different metabolites, including malonylcarnitine, caffeine, uracil, Ser Pro Pro, Arginyl-Proline, Pyrraline, Ala-Ile, N-Acetylglutamic acid, Glutamicacid, Orotic acid acid) was up-regulated, while 5-acetylamino-6-amino-3-methyluracil, LysoPC (18:2(9Z,12Z)), glutamine (N2-gamma-Glutamylglutamine), Miglitol (Miglitol, IU1), 4-nitrophenol (4-Nitrophenol), and IU1 (USP14 proteasome inhibitor) were down-regulated;

[0087] S5.4, three screenings, four metabolic markers were screened out;

[0088] S5.4.1. Calculate the correlation coefficient between the saliva "credible differential metabolites" and the COPD diagnostic threshold (FEV1 / FVC%). FEV1 / FVC% is the ratio of the forced expiratory volume in the first second (FEV1) to the forced vital capacity (FVC). A ratio less than 70% indicates the presence of obstructive ventilatory dysfunction;

[0089] S5.4.2. Use the multivariate linear regression equation to screen out the differential metabolites that have a significant effect on FEV1 / FVC%. Select metabolites with a P value of <0.05 as metabolic markers;

[0090] From the above 16 differential metabolites, a total of 4 statistically significant metabolic markers were screened out, namely: lysophospholipid LysoPC (18:2 (9Z, 12Z)), glutamine (N2-gamma-Glutamylglutamine), miglitol and 4-nitrophenol. The detailed information is shown in Table 1.

[0091] Table 1 Correlation analysis and multiple regression analysis between differential metabolites and FEV1 / FVC%

[0092]

[0093] Not all of these metabolites are directly related to the clinical manifestations or disease progression of COPD patients. The three-step screening process evaluates the strength of the relationship between credible differential metabolites and FEV1 / FVC% by constructing a multivariate regression model to select metabolites that have a significant effect on lung function. By associating credible differential metabolites with specific clinical parameters, it is possible to more accurately predict whether an individual has COPD and its severity.

[0094] S6. Diagnostic efficacy verification;

[0095] The ROC curve was constructed using the COPD metabolite marker combination composed of the above four metabolites, such as Figure 2 As shown, the ROC curve AUC is 0.978, the sensitivity is 0.924, the specificity is 0.927, the positive predictive value is 0.938, the negative predictive value is 0.911, and the accuracy is 0.918, that is, the combination has a good efficacy in diagnosing COPD.

[0096] The preferred specific embodiments of the present invention are described in detail above. It should be understood that ordinary technicians in the field can make many modifications and changes based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by technicians in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for screening a combination of salivary metabolic markers for diagnosing COPD, characterized in that: The steps include: S1. Screening of subjects; S2, collect saliva samples from subjects; S3. Perform metabolite mass spectrometry analysis on the subjects’ saliva samples; S4. Performing data quality control on the raw data obtained from mass spectrometry analysis; S5. Screen out differential metabolites; S6. Validation of diagnostic efficacy of differential metabolites; Step S5 includes the following steps: S5.1, primary screening; The principal component analysis, partial least squares discriminant analysis and orthogonal least squares discriminant analysis were used to screen the differential metabolites; S5.2, subgroup analysis was performed with periodontal disease; In step S5.2, the specific subgroups are divided into: Group I is a normal lung function group, including two subgroups, a normal lung function group without periodontal disease and a normal lung function group with periodontal disease; Group II is a COPD stable group, including two subgroups, a COPD stable group with periodontal disease and a COPD stable group without periodontal disease; In step S5.2, the stability of the two subgroups in the two groups was analyzed by variance analysis, and metabolites with a mean difference of greater than or equal to 0.1 and a variance threshold greater than or equal to 1.1 in the two subgroups were deleted; The salivary metabolic marker combination for diagnosing COPD consists of lysophospholipids, glutamine, miglitol and 4-nitrophenol.

2. The method for screening a combination of salivary metabolic markers for diagnosing COPD according to claim 1, characterized in that: The criteria for screening differential metabolites in step S5.1 are: VIP value>1, and P value<0.

05.

3. The method for screening a combination of salivary metabolic markers for diagnosing COPD according to claim 1, characterized in that: Step S5 includes the following steps: S5.1, primary screening; The principal component analysis, partial least squares discriminant analysis and orthogonal least squares discriminant analysis were used to screen for differential metabolites; S5.2, subgroup analysis was performed with periodontal disease; S5.3, secondary screening; The charge-to-mass ratios of differential metabolites were matched with the database and rated to screen out credible differential metabolites; S5.4, tertiary screening; The multivariate linear regression equation was used to screen out the differential metabolites that had a significant effect on FEV1 / FVC%, and the metabolites with a P value of < 0.05 were selected as metabolic markers; In step S5.3, based on the matching results of m / z values, metabolites of level B and level A are screened out as credible differential metabolites; the metabolites of level B are known metabolites, and the metabolites of level A are standard metabolites.

Images