A fecal biomarker panel for the early diagnosis of Sjogren's syndrome

Abstract

The present invention provides a fecal biomarker combination for the early diagnosis of Sjogren's syndrome. The method is as follows: collect fecal specimens of the research subjects; collect fecal metabolites; use UPLC / Q-TOFMS technology to collect non-targeted metabolomics data of fecal metabolites to obtain a metabolic map; use XCMS to preprocess the metabolic map data; perform multivariate analysis to screen for differential metabolites and identify them in combination with relevant databases, retain endogenous differential metabolites, and further screen the endogenous differential metabolites using R language and establish a prediction model for verification to determine the biomarker combination for differentiating SS patients from healthy controls. Based on the metabolomics analysis of the feces of SS patients and healthy controls by UPLC / Q-TOFMS, a fecal biomarker combination for the early diagnosis or prevention of Sjogren's syndrome is obtained, which has high sensitivity and specificity, good predictability, and can be used for the early diagnosis of SS.

Description

Technical Field

[0001] The present invention relates to the technical field of medical biomarkers, and particularly relates to a fecal biomarker combination for early diagnosis of Sjogren's syndrome. Background Art

[0002] Sjogren's syndrome (SS) is a highly heterogeneous systemic autoimmune disease mainly characterized by the involvement of exocrine gland organs. The clinical manifestations mainly include dryness of exocrine glands such as the mouth and eyes, and extra-glandular symptoms such as arthritis, rash, and muscle pain may also occur. In severe cases, it can even affect internal organs and other organ systems, endangering life. At present, there is a lack of early diagnosis methods in clinical practice, often resulting in missed diagnosis and misdiagnosis. According to epidemiological investigation and statistics, the misdiagnosis rate from the first visit to the time of diagnosis is as high as 57.5%, and the average misdiagnosis time is as long as 5 - 7 years.

[0003] Currently, in the prior art, the diagnosis of Sjogren's syndrome mainly relies on various methods such as the patient's clinical manifestations, serum immunological indexes, and histological examinations. The main classification criteria currently in use are the 2016 ACR / EULAR Sjogren's syndrome classification criteria. This standard mainly evaluates the secretion ability of exocrine glands, screens for anti-Ro or anti-SSA autoantibodies, and assesses whether there is lymphocyte infiltration in the lip biopsy. Among them, the evaluation of the secretion ability of exocrine glands is mainly based on the Schirmer test and the natural saliva flow rate. Some studies have shown that its diagnostic specificity is not high, ranging from 34% to 76%. Anti-SSA antibody is the most commonly used autoantibody for diagnosing primary Sjogren's syndrome, but the diagnostic specificity of detecting anti-SSA antibody is not high. A negative serological antibody test does not rule out the diagnosis of SS, which is prone to misdiagnosis or missed diagnosis. Lip biopsy, as a morphological pathological examination, is known as the "gold standard" for the pathological diagnosis of SS, with high diagnostic sensitivity and specificity. However, due to its invasive nature, it is difficult for patients to accept, and its clinical utilization rate is low and it is not conducive to prognosis evaluation. In addition, the clinical manifestations of Sjogren's syndrome are extremely diverse, often involving multiple systems and organs, and are easily misdiagnosed as other common rheumatic immune diseases such as rheumatoid arthritis and systemic lupus erythematosus. It is estimated that currently about half of the patients have not been diagnosed. Therefore, there is an urgent need for new biomarkers to assist in the early diagnosis of SS and to provide prospects for monitoring the patient's condition, disease development, and developing specific treatment methods.

[0004] In recent years, a large number of studies have shown that autoimmune diseases are closely related to the gut microbiota, and the immune response caused by dysbiosis is the key immunopathological mechanism inducing the development of autoimmune diseases. Research has shown that there are significant differences in the gut microbiota between SS patients and normal people. In particular, the number of butyrate-producing bacteria is significantly reduced, and the disease severity is highly correlated with the microbiota diversity. Although the results of different gut microbiota studies are not completely consistent, there are indeed significant differences in the gut microbiota characteristics between patients with Sjogren's syndrome and healthy people. Gut metabolites are the bridge connecting the interaction between the microbiota and the host, and their changes are a direct reflection of the functions of the microbiota and the host.

[0005] Currently, metabolomics is widely used in the study of gut microbiota, which can provide high-throughput quantitative information on metabolites and explore the differences in gut microbiota-derived metabolites closely related to host physiological and pathological processes. Nowadays, metabolomics has been widely used to discover biomarkers and key pathways related to many diseases and explain their pathological mechanisms due to its high throughput, high sensitivity, wide coverage, and relatively low cost. Metabolomics analysis based on feces or gut contents can clearly show the metabolic status of gut microbiota and the host by detecting changes in gut metabolites. Therefore, feces, as a non-invasive and easily obtainable sample, studying the changing metabolites in the feces of SS patients and looking for potential biomarkers will help achieve the early diagnosis of SS. Summary of the Invention

[0006] Aiming at the deficiencies of the above existing methods, the present invention provides a fecal biomarker combination for the early diagnosis of Sjogren's syndrome to meet the urgent current need for using novel biomarkers to assist in the early diagnosis of SS.

[0007] Based on ultra-high performance liquid chromatography quadrupole time-of-flight tandem mass spectrometry (UPLC / Q-TOF MS), the present invention performs metabolomics analysis on the feces of SS patients and healthy controls, and obtains a fecal biomarker combination for the early diagnosis or prevention of Sjogren's syndrome, which has high sensitivity and specificity, good predictability, and can be used for the early diagnosis of SS. The kit detection of the present invention, as a non-invasive and repeatable novel examination method, may play a role in reducing the missed diagnosis rate of early Sjogren's syndrome and become a new method for disease condition assessment, and has great potential in the early diagnosis, development, and prognosis assessment of the disease.

[0008] The present invention provides a fecal biomarker combination for the early diagnosis of Sjogren's syndrome, including,

[0009] S1: Collect fecal specimens of the research subjects, where the research subjects include the SS patient group and the healthy control group HC;

[0010] S2: Collect fecal metabolites from the fecal specimens, where the fecal metabolites include polar-phase samples and non-polar-phase samples;

[0011] S3: Use UPLC / Q-TOF MS technology to collect non-targeted metabolomics data of the fecal metabolites and obtain a metabolic map;

[0012] S4: Use XCMS to preprocess the metabolic map data. The preprocessing includes peak detection and peak matching of the original chromatographic peaks of the metabolic map, and extracting a data matrix composed of the retention time, mass number, and corresponding peak intensity / peak area of metabolite fragments;

[0013] S5: Perform multivariate analysis on the data matrix to screen for differential metabolites and identify them in combination with relevant databases. Manually exclude exogenous differential metabolites and retain endogenous differential metabolites. Further screen the endogenous differential metabolites using R language and establish a prediction model for verification to determine a biomarker combination for differentiating the SS patients and the healthy controls. The biomarker combination is any one, any two, any three, or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide.

[0014] Preferably, antibiotics and probiotics have not been used within 1 month before sampling of the research subjects; the fecal specimens of the SS patient group are obtained within 1 to 2 days of the hospitalization of the SS patients and stored in a -80°C refrigerator within 4 hours; the HC fecal specimens are from a health examination center, obtained on the day of the examination, and stored in a -80°C refrigerator within 4 hours.

[0015] Preferably, the specific collection method in step S2 includes:

[0016] S21: Take an appropriate amount of the fecal specimen for lyophilization, and after grinding, accurately weigh an appropriate amount of the sample into a 1.5 mL centrifuge tube;

[0017] S22: Take an appropriate amount of ultrapure water and vortex it with the sample. The solvent ratio of the sample to the ultrapure water is 1:50. Ultrasonically extract for 20 min at 0°C, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation and place it in EP tube A;

[0018] S23: Add 1 mL of methanol solvent to the precipitate remaining in step S22, ultrasonically extract for 20 min at 0°C, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation again and place it in EP tube B;

[0019] S24: Take 500 μL of the supernatant from each of EP tube A and EP tube B, combine and mix them, add 1 mL of methanol, vortex to mix evenly, then concentrate and dry the supernatant using a nitrogen evaporator. After drying, take 100 μL of 80% methanol to redissolve, vortex to mix evenly, centrifuge at 13000 rpm for 15 min, and take the supernatant for injection to obtain the polar phase sample;

[0020] S25: Add 1 mL of methyl tert-butyl ether to the precipitate obtained in step S23, vortex to mix evenly, then perform ultrasonic extraction for 20 min. Centrifuge the solution after ultrasonic extraction at 3500 rpm for 20 min, take the supernatant after centrifugation and dry it. After drying, take 100 μL of 70% isopropanol to redissolve, vortex to mix evenly, centrifuge at 13000 rpm for 15 min, and take the supernatant for injection to obtain the non-polar phase sample.

[0021] Preferably, when using UPLC / Q-TOF MS technology to collect the untargeted metabolomics data of the fecal metabolites, the chromatographic conditions and mass spectrometry conditions of the polar phase sample are as follows:

[0022] Chromatographic conditions: Use an ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm. The mobile phase is divided into phase A and phase B. Phase A is 0.1% formic acid in water, and phase B is acetonitrile. The total running time is 39 min, and the elution gradient is as follows: 0 - 2 min, 2% B; 2 - 3.5 min, 15% B; 3.5 - 5.0 min, 15% B; 5.0 - 18 min, 60% B; 18 - 27 min, 60% B; 27 - 29 min, 95% B; 29 - 36 min, 95% B; 36 - 36.5 min, 2% B; 36.5 - 39 min, 2% B; the flow rate is 0.3 mL / min; the column temperature is 40 °C; the injection volume is 5 μL;

[0023] Mass spectrometry conditions: The ion source voltage is 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature is 450 °C, the declustering voltage is 60 eV or -60 eV, the collision energy is 35 eV or -35 eV, the collision energy spread voltage is 15 eV or -15 eV, the first auxiliary gas is 55 psi, the second auxiliary gas is 55 psi, the curtain gas is 30 psi, the parent ion scan range is 50 - 1500 m / z, and the daughter ion scan range is 50 - 1500 m / z;

[0024] The chromatographic conditions and mass spectrometry conditions of the non-polar phase sample are as follows:

[0025] Chromatographic conditions: An ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm was used; the mobile phase was divided into phase A' and phase B'. Phase A' was 60% acetonitrile water containing 0.1% formic acid, and phase B' was a solution containing 10 mM ammonium acetate. The volume ratio of isopropanol to acetonitrile in the solution was 9:1. The total running time was 34 min, and the elution gradient was as follows: 0 - 3 min, 32% B; 3 - 6 min, 45% B; 6 - 8 min, 52% B; 8 - 12 min, 58% B; 12 - 14 min, 66% B; 14 - 20 min, 70% B; 20 - 25 min, 75% B; 25 - 28 min, 99% B; 28 - 31 min, 99% B; 31 - 31.5 min, 32% B; 31.5 - 34 min, 32% B; the flow rate was 0.25 mL / min; the column temperature was 40°C; the injection volume was 5 μL;

[0026] Mass spectrometry conditions: The ion source voltage was 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature was 550°C, the declustering voltage was 60 eV or -60 eV, the collision energy was 40 eV or -40 eV, the collision energy spread voltage was 20 eV or -20 eV, the first auxiliary gas was 55 psi, the second auxiliary gas was 55 psi, the curtain gas was 30 psi, the parent ion scan range was 50 - 1500 m / z, and the daughter ion scan range was 50 - 1500 m / z.

[0027] Preferably, step S5 is specifically as follows:

[0028] S51: Divide the SS patient group and the HC into a training set and a validation set. Both the training set and the validation set include the SS patient group and the HC research objects. The ratio of the total number of people in the training set to the validation set is close to 3:1;

[0029] S52: Use SIMCA 14.0 software to perform multivariate analysis on the untargeted metabolomics data of the training set. The multivariate analysis includes principal component analysis PCA, partial least squares discriminant analysis PLS-DA, and orthogonal partial least squares discriminant analysis OPLS-DA. Screen for differential metabolites by combining the variable importance in projection VIP, fold change FC, and false discovery rate value FDR obtained from the OPLS-DA model. Consider metabolites with VIP > 1, FC > 1.5 or FC < 0.67, and FDR < 0.05 as differential metabolites. Finally, combine with relevant databases for identification, and manually exclude exogenous differential metabolites and retain endogenous differential metabolites;

[0030] S53: Further screen the endogenous differential metabolites based on the Lasso regression model, and use the ten-fold cross-validation method to screen out the optimal endogenous differential metabolites related to the early diagnosis of SS. Use R language to build a binary Logistics regression model to fit the optimal endogenous differential metabolites, and screen out the optimal endogenous differential metabolites with a P-value < 0.05, which are the biomarkers. The biomarkers include MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide;

[0031] S54: Establish prediction models according to two machine learning models, support vector machine SVM and random forest RF, and analyze and verify the biomarkers obtained in step S53 in the training set and the validation set to determine the biomarker combination for differentiating SS patients and healthy controls. The biomarker combination is any one, any two, any three, or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide.

[0032] Preferably, in step S52, KEGG pathway analysis was performed on the screened differential metabolites, and 13 metabolic pathways related to SS were screened out, including sphingolipid metabolism, biosynthesis of phenylalanine, tyrosine and tryptophan, phenylalanine metabolism, α-linolenic acid metabolism, arachidonic acid metabolism, glycerophospholipid metabolism, porphyrin and chlorophyll metabolism, pentose and glucuronate interconversion, cysteine and methionine metabolism, steroid biosynthesis, arginine and proline metabolism, pyruvate metabolism, and primary bile acid biosynthesis.

[0033] Preferably, the relevant databases include OSI / SMMS software and other online databases, and the other online databases include the Human Metabolome Database and the Lipidmaps Database.

[0034] Preferably, ROC curve analysis was performed on each biomarker and 4 biomarker combinations screened in step S53. The 4 biomarker combinations are MG(0:0 / 14:0 / 0:0)+LysoPE(14:0 / 0:0)+Sphinganine+Octadecanamide, and the 4 biomarker combinations are more effective in the early diagnosis of SS than single biomarkers.

[0035] The present invention also provides a detection kit for the early diagnosis of Sjogren's syndrome, which includes a biomarker combination for the early diagnosis of Sjogren's syndrome, and the biomarker combination is any one, any two, any three or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide.

[0036] Preferably, the sample detected by the kit is derived from a fecal sample.

[0037] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0038] 1. Based on UPLC / Q-TOF MS, the present invention performs metabolomics analysis on the feces of SS patients and healthy controls, and obtains a fecal biomarker combination for the early diagnosis of Sjogren's syndrome, which has high sensitivity and specificity, good predictability, and can be used for the early diagnosis of SS.

[0039] 2. The diagnostic index provided by the present invention is a fecal biomarker, which is a non-invasive and repeatable new detection method. It can play a role in reducing the missed diagnosis rate of early Sjogren's syndrome and become a new method for disease condition assessment. It has great potential in the early diagnosis, development and prognosis assessment of the disease. Therefore, the detection kit for the early diagnosis of Sjogren's syndrome has a good market prospect. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0041] Figure 1 It is the non-targeted metabolic profile map of one of the fecal samples participating in this experiment in the embodiment of the present invention, including the TIC total ion current maps of the positive and negative ion modes of the polar and non-polar phases.

[0042] Figure 2 It is the PCA model analysis diagram of the fecal metabolic profiles (positive and negative ion modes of the polar and non-polar phases) of the SS patient group and HC in the training set in the embodiment of the present invention.

[0043] Figure 3 It is the PLS-DA model analysis diagram of the fecal metabolic profiles (positive and negative ion modes of the polar and non-polar phases) of the SS patient group and HC in the training set in the embodiment of the present invention.

[0044] Figure 4 This is the OPLS-DA model analysis chart of the fecal metabolic profiles (positive and negative ion modes of polar and non-polar phases) of the SS patient group and HC in the training set of the embodiments of the present invention.

[0045] Figure 5 This is the volcano plot of differential metabolites in the embodiments of the present invention.

[0046] Figure 6 This is the bubble plot of differential metabolic pathways in the feces of the SS patient group and HC obtained through metabolic pathway enrichment analysis in the embodiments of the present invention.

[0047] Figure 7 This is the ROC analysis chart of 4 biomarker combinations in the embodiments of the present invention.

[0048] Figure 8 This is the accuracy chart of discriminating SS by 4 biomarkers in the support vector and random forest model verification training set in the embodiments of the present invention.

[0049] Figure 9 This is the accuracy chart of discriminating SS by 4 biomarkers in the support vector and random forest model verification validation set in the embodiments of the present invention.

[0050] Figure 10 This is the distribution chart of 4 biomarkers in the training set and validation set in the embodiments of the present invention, where A is the distribution chart in the training set and B is the distribution chart in the validation set. Detailed implementation manners

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0052] In the description of the present invention, it should be noted that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0053] The present invention provides a fecal biomarker combination for the early diagnosis of Sjogren's syndrome, including:

[0054] S1: Collect fecal specimens of research subjects, and the research subjects include the SS patient group and the healthy control group HC;

[0055] Preferably, antibiotics and probiotics were not used within 1 month before sampling the research subjects; the fecal specimens of the SS patient group were obtained within 1 - 2 days of the hospitalization of SS patients and stored in a -80°C refrigerator within 4 hours; the fecal specimens of the HC group were obtained from a health examination center on the day of the examination and stored in a -80°C refrigerator within 4 hours.

[0056] A total of 123 fecal specimens from SS patients and 52 fecal specimens from healthy controls were collected in this experiment. The 123 research subjects in the SS patient group were from the Department of Rheumatology and Immunology, the Second Hospital of Shanxi Medical University, and all met the 2016 ACR / EULAR diagnostic criteria for Sjogren's syndrome, without concurrent other autoimmune diseases, excluding other metabolic diseases such as diabetes and hyperlipidemia. Their demographic data, risk factors, and laboratory parameters were obtained from clinical records. The 52 research subjects in the healthy control group were from a health examination center, without autoimmune diseases, excluding diabetes, hypertension, digestive system-related diseases, etc.

[0057] S2: Collect fecal metabolites in the fecal specimens, and the fecal metabolites include polar phase samples and non-polar phase samples;

[0058] Preferably, the specific collection method in step S2 includes:

[0059] S21: Take an appropriate amount of the fecal specimen for lyophilization, and accurately weigh 20 mg of the sample in a 1.5 mL centrifuge tube after grinding;

[0060] S22: Take 1 mL of ultrapure water and vortex it with the sample. The solvent ratio of the sample to the ultrapure water is 1:50. Ultrasonically extract for 20 min at 0°C, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation and place it in EP tube A;

[0061] S23: Add 1 mL of methanol solvent to the precipitate remaining in step S22, ultrasonically extract for 20 min at 0°C, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation again and place it in EP tube B;

[0062] S24: Take 500 μL of supernatant from each of EP tube A and EP tube B and combine and mix them. Add 1 mL of methanol, vortex and mix well, then concentrate and dry the supernatant with a nitrogen blower. After drying, take 100 μL of 80% methanol for reconstitution, vortex and mix well, centrifuge at 13000 rpm for 15 min, and take the supernatant for injection to obtain the polar phase sample;

[0063] S25: Add 1 mL of methyl tert-butyl ether (MTBE) to the precipitate obtained in the step S23, vortex to mix evenly, then perform ultrasonic extraction for 20 min. Centrifuge the solution after ultrasonic extraction at 3500 rpm for 20 min. After centrifugation, take the supernatant and dry it. After drying, take 100 μL of 70% isopropanol to redissolve it, vortex to mix evenly, then centrifuge at 13000 rpm for 15 min, and take the supernatant for injection to obtain the non-polar phase sample.

[0064] S3: Use UPLC / Q-TOF MS technology to collect the untargeted metabolomics data of the fecal metabolites to obtain a metabolic map;

[0065] Preferably, when using UPLC / Q-TOF MS technology to collect the untargeted metabolomics data of the fecal metabolites, the chromatographic conditions and mass spectrometry conditions for the polar phase sample are respectively:

[0066] Chromatographic conditions: Use an ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm; the mobile phase is divided into phase A and phase B. Phase A is 0.1% formic acid in water, and phase B is acetonitrile. The total running time is 39 min, and the elution gradient is as follows: 0 - 2 min, 2% B; 2 - 3.5 min, 15% B; 3.5 - 5.0 min, 15% B; 5.0 - 18 min, 60% B; 18 - 27 min, 60% B; 27 - 29 min, 95% B; 29 - 36 min, 95% B; 36 - 36.5 min, 2% B; 36.5 - 39 min, 2% B; the flow rate is 0.3 mL / min; the column temperature is 40 °C; the injection volume is 5 μL;

[0067] Mass spectrometry conditions (positive and negative ion modes): The ion source voltage is 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature is 450 °C, the declustering voltage is 60 eV or -60 eV, the collision energy is 35 eV or -35 eV, the collision energy spread voltage is 15 eV or -15 eV, the first auxiliary gas is 55 psi, the second auxiliary gas is 55 psi, the curtain gas is 30 psi, the parent ion scan range is 50 - 1500 m / z, and the daughter ion scan range is 50 - 1500 m / z;

[0068] The chromatographic conditions and mass spectrometry conditions for the non-polar phase sample are respectively:

[0069] Chromatographic conditions: An ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm was used; the mobile phase was divided into phase A' and phase B'. Phase A' was 60% acetonitrile water containing 0.1% formic acid, and phase B' was a solution containing 10 mM ammonium acetate. The volume ratio of isopropanol to acetonitrile in the solution was 9:1. The total running time was 34 min, and the elution gradient was as follows: 0 - 3 min, 32% B; 3 - 6 min, 45% B; 6 - 8 min, 52% B; 8 - 12 min, 58% B; 12 - 14 min, 66% B; 14 - 20 min, 70% B; 20 - 25 min, 75% B; 25 - 28 min, 99% B; 28 - 31 min, 99% B; 31 - 31.5 min, 32% B; 31.5 - 34 min, 32% B; the flow rate was 0.25 mL / min; the column temperature was 40 °C; the injection volume was 5 μL;

[0070] Mass spectrometry conditions (positive and negative ion modes): The ion source voltage was 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature was 550 °C, the declustering voltage was 60 eV or -60 eV, the collision energy was 40 eV or -40 eV, the collision energy spread voltage was 20 eV or -20 eV, the first auxiliary gas was 55 psi, the second auxiliary gas was 55 psi, the curtain gas was 30 psi, the parent ion scan range was 50 - 1500 m / z, and the daughter ion scan range was 50 - 1500 m / z.

[0071] Refer to Figure 1 , when collecting the untargeted metabolomics data of the fecal metabolites using UPLC / Q-TOF MS technology, each sample would obtain metabolic profile maps for positive and negative ion mode injections of the polar and non-polar phases, a total of four Figure 1 which were the metabolic profile maps for positive and negative ion mode injections of the polar and non-polar phases of one of the samples participating in this experiment.

[0072] S4: Use XCMS (version 3.6.3) to preprocess the metabolic map data. The preprocessing includes peak detection and peak matching of the original chromatographic peaks in the metabolic map, and extracting a data matrix composed of the retention time, mass number, and corresponding peak intensity / peak area of the metabolite fragments;

[0073] S5: Perform multivariate analysis on the data matrix to screen for differential metabolites, identify them in combination with relevant databases, manually remove exogenous differential metabolites, retain endogenous differential metabolites, further screen the endogenous differential metabolites using R language, and establish and validate a prediction model to determine a biomarker combination for differentiating the SS patients from the healthy controls. The biomarker combination is any one, any two, any three, or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide.

[0074] Preferably, step S5 is specifically as follows:

[0075] S51: Divide the SS patient group and the HC into a training set and a validation set. Both the training set and the validation set include the SS patient group and the HC study subjects. The ratio of the total number of people in the training set to the validation set is close to 3:1.

[0076] S52: Perform multivariate analysis on the untargeted metabolome data of the training set using SIMCA 14.0 software. The multivariate analysis includes principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA). Screen for differential metabolites by combining the variable importance in projection (VIP), fold change (FC), and false discovery rate (FDR) values obtained from the OPLS-DA model. Consider metabolites with VIP > 1, FC > 1.5 or FC < 0.67, and FDR < 0.05 as differential metabolites. Finally, identify them in combination with relevant databases, manually remove exogenous differential metabolites, and retain endogenous differential metabolites.

[0077] PCA: It is an unsupervised data analysis method. It uses linear projection to transform the original multi-variable space into a set of new orthogonal variables, and describes the characteristics of the data with several main components (PCs). These components are linear combinations of the original variables, and the perpendicular intersection between these principal components ensures that when projecting from a high-dimensional space to a low-dimensional space, as much useful information as possible is retained. PLS-DA: It is a supervised method for multi-dimensional data compression. Before compressing the multi-dimensional data, it groups the data according to the differential factors to be found as needed, so that the variables most relevant to the factors used for grouping can be found, while reducing the influence of some other factors. OPLS-DA can filter out the noise irrelevant to the research object and is an analytical method that modifies PLS-DA. OPLS-DA divides the differences in data table X (variables) into two parts according to the differences in data table Y (grouping). The first part represents the differences related to Y, and the second part represents the differences unrelated (orthogonal) to Y. OPLS-DA can distinguish these two parts of differences. At the same time, through the OPLS-DA score plot, the inter-group differences can be directly distinguished from the predicted principal component (t1), improving the analytical ability and effectiveness of the model.

[0078] Refer to Figures 2-3 , in this experiment, 123 SS patients and 52 healthy controls included in the study were divided into a training set and a validation set. The training set included 93 SS patients and 42 healthy controls, and the validation set included 30 SS patients and 10 healthy controls. The clinical characteristics of SS patients and healthy controls in the training set and the validation set are shown in Table 1. In this experiment, the SIMCA 14.0 software was used to perform multivariate analysis on the untargeted metabolomic data of the training set. The results of the unsupervised PCA analysis (see Figure 2 ) showed that the fecal polar metabolite profiles of SS patients and healthy controls were significantly separated. Further, a supervised PLS-DA model was constructed (see Figure 3 ). The R 2 X, R 2 Y, and Q 2 of the model in the positive ion mode of the polar phase were 0.385, 0.945, and 0.785 respectively. The R 2 X, R 2 Y, and Q 2 of the model in the negative ion mode of the polar phase were 0.297, 0.888, and 0.627 respectively. The R 2 X, R 2 Y, and Q 2 of the model in the positive ion mode of the non-polar phase were 0.377, 0.839, and 0.629 respectively. The R 2 X, R 2 Y, and Q 2 of the model in the negative ion mode of the non-polar phase were 0.325, 0.831, and 0.454 respectively, indicating that the prediction ability of the model was good.

[0079] Referring to Figures 4-5 , to screen for differential metabolites, a supervised OPLS-DA model was constructed (see Figure 4 ) to obtain the VIP value, and all peaks were visualized as a volcano plot (see Figure 5 ). Figure 5 Figure 10 is the volcano plot of differential metabolites, where the Y coordinate is log2(FC) and the X coordinate is -log 10 q, the q value is the FDR value, each point represents a metabolite, the black points represent metabolites with significant differences and up-regulated in SS, the dark gray points represent metabolites with significant differences and down-regulated in SS, and the light gray points are metabolites without significant differences. The larger the absolute value of the abscissa, the greater the difference between the metabolites in the two groups. The larger the ordinate value, the more significant the expression difference of the metabolites between the groups and the more reliable the results.

[0080] Metabolites with VIP > 1, FC > 1.5 or FC < 0.67, and FDR < 0.05 were regarded as differential metabolites. Combining with relevant databases for identification, a total of 94 differential metabolites were identified in the polar part, of which 72 metabolites were enriched in SS and 22 metabolites were enriched in the healthy control; a total of 74 differential metabolites were identified in the non-polar part, of which 26 metabolites were enriched in SS and 48 metabolites were enriched in the healthy control. Therefore, a total of 168 differential metabolites were identified in this experiment, as shown in Table 2.

[0081] Table 1. Clinical characteristics of SS patients and HC in the training set and validation set

[0082]

[0083] Table 2. 168 differential metabolites in fecal samples of SS patients and HC

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093] Preferably, in step S52, KEGG pathway analysis was performed on the screened differential metabolites, and 13 metabolic pathways related to SS were screened out, including sphingolipid metabolism, phenylalanine, tyrosine and tryptophan biosynthesis, phenylalanine metabolism, alpha-linolenic acid metabolism, arachidonic acid metabolism, glycerophospholipid metabolism, porphyrin and chlorophyll metabolism, pentose and glucuronate interconversions, cysteine and methionine metabolism, steroid biosynthesis, arginine and proline metabolism, pyruvate metabolism, and primary bile acid biosynthesis.

[0094] Refer to Figure 6 , Figure 6 is an enrichment bubble chart of metabolic pathways, where the vertical axis is -log 10 p, the horizontal axis is pathway impact, which is a weight calculation based on topological analysis. The node color is based on the p-value (the p-value is the probability of obtaining the sample observation result or a more extreme result when the null hypothesis is true, and the p-value is directly calculated and generated in the software), from light to dark, and the p-value decreases from large to small; the node radius is based on its pathway impact value, the nodes increase from small to large, and the impact value increases from small to large. Therefore, the metabolic pathway in the upper right corner of the bubble chart is the most important.

[0095] KEGG (Kyoto Encyclopedia of Genes and Genomes) is a database for systematic analysis of gene functions and genomic information, and is a powerful tool for metabolomic analysis and metabolic network research in organisms. Performing KEGG enrichment analysis on differentially expressed genes can enrich significantly different pathways, which helps to identify biologically regulatory pathways with significant differential changes under experimental conditions. In this experiment, KEGG pathway analysis was performed using MetaboAnalyst 5.0.

[0096] Preferably, the relevant databases include OSI / SMMS software (a software for rapid identification of small molecule compounds in metabolomics, Dalian ChemData Solution Information Technology Co., Ltd, PR China) and other online databases. The other online databases include the Human Metabolome Database (http: / / www.hmdb.ca / ) and the Lipidmaps database (https: / / lipidmaps.org / ).

[0097] In this experiment, considering the complexity of metabolites in feces, 16 exogenous differential metabolites were manually removed before further screening and verification using R language, and only 152 endogenous differential metabolites were retained as data for further screening.

[0098] S53: Further screen the endogenous differential metabolites based on the Lasso regression model, and use the ten-fold cross-validation method to screen out the optimal endogenous differential metabolites related to the early diagnosis of SS. Use R language to build a binary Logistics regression model to fit the optimal endogenous differential metabolites, and screen out the optimal endogenous differential metabolites with a P value < 0.05, which are the biomarkers. The biomarkers include MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide;

[0099] In this experiment, the R package glmnet was used to further screen the endogenous differential metabolites based on Lasso. The regularization parameter λ of the regression coefficient was introduced by Lasso regression. The ten-fold cross-validation method was used to define the parameter λ. By adjusting the parameter λ, the coefficients of the metabolites not related to the early diagnosis of SS were reduced to zero, and the optimal endogenous differential metabolites related to the early diagnosis of SS were screened out. The optimal endogenous differential metabolites include: NAGly 17:0; O, Phenethylamineglucuronide, FA 24:1, Acetic acid, Sphinganine, Biotripyrrin-a, 5a-Dihydrotestosterone sulfate, LysoPE(14:0 / 0:0), MG(0:0 / 14:0 / 0:0), Cer(d18:0 / 14:0), Palmitic amide, FAHFA(18:1(9Z) / 8-O-18:0), Octadecanamide, Cer(d15:2 / 22:0), FA24:0, FA 16:0, FA 26:0; O, SPB 20:0; O2, SPB 16:0; O2, a total of 19 kinds. The binary Logistics regression model was built using R language to fit the 19 optimal endogenous differential metabolites, and the optimal endogenous differential metabolites with P value < 0.05 were screened out, which are the biomarkers, a total of 4 kinds, including MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide.

[0100] Preferably, the ROC curve analysis was performed on each biomarker screened in the step S53 and the combination of 4 biomarkers. The combination of 4 biomarkers is MG(0:0 / 14:0 / 0:0)+LysoPE(14:0 / 0:0)+Sphinganine+Octadecanamide. The diagnostic efficacy of the combination of 4 biomarkers in the early diagnosis of SS is better than that of a single biomarker.

[0101] Refer to Figure 7, in order to evaluate the accuracy of 4 biomarkers in the early diagnosis of SS, the receiver operating characteristic curve (ROC) was plotted to analyze the 4 biomarkers respectively. The area under the curve (AUC) of the curve was 0.901, 0.862, 0.825, and 0.836 respectively. The value range of AUC is between 0 and 1. The larger the AUC, the better the prediction effect. It can be seen that the accuracy of the 4 biomarkers in the early diagnosis of SS is very good. In order to improve the diagnostic efficiency of the biomarkers, 4 biomarkers were combined for multivariate Logistics model fitting. After fitting, the AUC value of the ROC curve was 0.990, which was greater than the AUC value of a single compound. Therefore, it can be seen that the combination of biomarkers is more superior to a single biomarker in the diagnostic efficiency of early diagnosis of SS.

[0102] S54: Prediction models were established according to two machine learning models, support vector machine (SVM) and random forest (RF), and the biomarkers obtained in step S53 were analyzed and verified in the training set and the validation set to determine the combination of biomarkers for differentiating the SS patients and the healthy controls. The combination of biomarkers is any one, any two, any three, or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide.

[0103] In this experiment, the support vector machine (SVM) used the R package e1071 to construct the prediction model, and the random forest (RF) used the R package randomForest to construct the prediction model.

[0104] Refer to Figures 8-9 , in this experiment, two models, RF and SVM, were used to analyze the discrimination ability of 4 biomarkers in the training set. It can be seen from Figure 8 that the accuracy of differentiating the HC and SS patient groups by 4 biomarkers in the random forest model was 100%, and the classification accuracy of the support vector machine was 97.04%. Two models, RF and SVM, were used to verify the discrimination ability of 4 biomarkers in the validation set. It can be seen from Figure 9 that the accuracy of differentiating the HC and SS patient groups by the random forest model was 100%, and the classification accuracy of the support vector machine was 92.5%. Moreover, the changing trend of the 4 biomarkers in the validation set was consistent with that in the training set. In summary, the 4 screened biomarkers have good discrimination ability for differentiating SS.

[0105] Figure 10 It is the distribution diagram of 4 biomarkers in the SS patient group and HC in the training set and the validation set, where Figure 10 A is the distribution diagram of 4 biomarkers in the training set, Figure 10Figure B shows the distribution maps of 4 biomarkers in the validation set. It can be seen from Figure 10 that the distribution trends of the 4 selected biomarkers in the training set and the validation set are consistent, further verifying the efficacy of the biomarkers in discriminating SS.

[0106] The embodiment of the present invention also provides a detection kit for the early diagnosis of Sjogren's syndrome. The kit includes a biomarker combination for the early diagnosis of Sjogren's syndrome, and the biomarker combination is any one, any two, any three or any four of MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide.

[0107] Preferably, the sample detected by the kit is derived from a fecal sample.

[0108] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. Use of a biomarker combination in the preparation of a kit for early diagnosis of Sjogren's syndrome, characterized in that, The biomarker combination is MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide.

2. Use of the biomarker combination according to claim 1 in the preparation of a kit for early diagnosis of Sjogren's syndrome, characterized in that, The biomarker combination is derived from fecal samples.

3. Use of the biomarker combination according to claim 1 in the preparation of a kit for the early diagnosis of Sjogren's syndrome, characterized in that, The screening method of the biomarker combination specifically includes: S1: Collect fecal specimens of the research subjects, where the research subjects include the SS patient group and the healthy control group HC; S2: Collect fecal metabolites in the fecal specimens, where the fecal metabolites include polar phase samples and non-polar phase samples; S3: Use UPLC / Q-TOF MS technology to collect non-targeted metabolomics data of the fecal metabolites to obtain a metabolic map; S4: Use XCMS to preprocess the metabolic map data. The preprocessing includes peak detection and peak matching of the original chromatographic peaks of the metabolic map, and extracting a data matrix composed of the retention time, mass number and corresponding peak intensity / peak area of the metabolite fragments; S5: Perform multivariate analysis on the data matrix to screen for differential metabolites and identify them in combination with relevant databases. Manually exclude exogenous differential metabolites and retain endogenous differential metabolites. Further screen the endogenous differential metabolites using R language and establish a prediction model for verification to determine the biomarker combination for differentiating the SS patients and the healthy controls. The biomarker combination is MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine and Octadecanamide.

4. Use of the biomarker combination according to claim 3 in the preparation of a kit for the early diagnosis of Sjogren's syndrome, characterized in that, The research subjects did not use antibiotics and probiotics within 1 month before sampling; the fecal specimens of the SS patient group were obtained within 1 - 2 days of the SS patients' hospitalization and stored in a -80°C refrigerator within 4 hours; the fecal specimens of the HC were from a health examination center, obtained on the day of the examination and stored in a -80°C refrigerator within 4 hours.

5. Use of the biomarker combination according to claim 4 in the preparation of a kit for early diagnosis of Sjogren's syndrome, characterized in that, The specific collection method in step S2 includes: S21: Take an appropriate amount of the fecal specimen for freeze-drying, and accurately weigh an appropriate amount of the sample into a 1.5 mL centrifuge tube after grinding; S22: Take an appropriate amount of ultrapure water and vortex it with the sample, ultrasonically extract it at 0°C for 20 min, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation and place it in EP tube A; S23: Add 1 mL of methanol solvent to the precipitate remaining in step S22, ultrasonically extract it at 0°C for 20 min, centrifuge the ultrasonically extracted solution at 3500 rpm for 20 min, and take out the supernatant obtained after centrifugation again and place it in EP tube B; S24: Take 500 μL of supernatant from each of EP tube A and EP tube B, combine and mix them, add 1 mL of methanol, vortex and mix well, then concentrate and dry the supernatant with a nitrogen blower. After drying, take 100 μL of 80% methanol to redissolve it, vortex and mix well, centrifuge it at 13000 rpm for 15 min, and take the supernatant for injection to obtain the polar phase sample; S25: Add 1 mL of methyl tert-butyl ether to the precipitate obtained in the step S23, vortex to mix evenly, then perform ultrasonic extraction for 20 min. Centrifuge the solution after ultrasonic extraction at 3500 rpm for 20 min. After centrifugation, take the supernatant and dry it. After drying, take 100 μL of 70% isopropanol to redissolve it, vortex to mix evenly, then centrifuge at 13000 rpm for 15 min, and take the supernatant for injection to obtain the non-polar phase sample.

6. Use of the biomarker combination according to claim 5 in the preparation of a kit for the early diagnosis of Sjogren's syndrome, characterized in that, When collecting the untargeted metabolomics data of the fecal metabolites by using the UPLC / Q-TOF MS technology, the chromatographic conditions and mass spectrometry conditions of the polar phase sample are respectively: Chromatographic conditions: Use an ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm; the mobile phase is divided into phase A and phase B. Phase A is 0.1% formic acid water, and phase B is acetonitrile. The total running time is 39 min, and the elution gradient is as follows: 0 - 2 min, 2% B; 2 - 3.5 min, 15% B; 3.5 - 5.0 min, 15% B; 5.0 - 18 min, 60% B; 18 - 27 min, 60% B; 27 - 29 min, 95% B; 29 - 36 min, 95% B; 36 - 36.5 min, 2% B; 36.5 - 39 min, 2% B; the flow rate is 0.3 mL / min; the column temperature is 40 °C; the injection volume is 5 μL; Mass spectrometry conditions: The ion source voltage is 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature is 450 °C, the declustering voltage is 60 eV or -60 eV, the collision energy is 35 eV or -35 eV, the collision energy spread voltage is 15 eV or -15 eV, the first auxiliary gas is 55 psi, the second auxiliary gas is 55 psi, the curtain gas is 30 psi, the parent ion scanning range is 50 - 1500 m / z, and the daughter ion scanning range is 50 - 1500 m / z; The chromatographic conditions and mass spectrometry conditions of the non-polar phase sample are respectively: Chromatographic conditions: Use an ACQUITY UPLC HSS T3 chromatographic column with a specification of 2.1×100 mm and 1.8 μm; the mobile phase is divided into phase A' and phase B'. Phase A' is 60% acetonitrile water containing 0.1% formic acid, and phase B' is a solution containing 10 mM ammonium acetate, and the volume ratio of isopropanol to acetonitrile in the solution is 9:

1. The total running time is 34 min, and the elution gradient is as follows: 0 - 3 min, 32% B; 3 - 6 min, 45% B; 6 - 8 min, 52% B; 8 - 12 min, 58% B; 12 - 14 min, 66% B; 14 - 20 min, 70% B; 20 - 25 min, 75% B; 25 - 28 min, 99% B; 28 - 31 min, 99% B; 31 - 31.5 min, 32% B; 31.5 - 34 min, 32% B; the flow rate is 0.25 mL / min; the column temperature is 40 °C; the injection volume is 5 μL; Mass spectrometry conditions: The ion source voltage is 5500 eV for positive ions and -4500 eV for negative ions, the ion source temperature is 550 °C, the declustering voltage is 60 eV or -60 eV, the collision energy is 40 eV or -40 eV, the collision energy spread voltage is 20 eV or -20 eV, the first auxiliary gas is 55 psi, the second auxiliary gas is 55 psi, the curtain gas is 30 psi, the parent ion scan range is 50 - 1500 m / z, and the daughter ion scan range is 50 - 1500 m / z.

7. Use of the biomarker combination according to claim 6 in the preparation of a kit for early diagnosis of Sjogren's syndrome, characterized in that, Step S5 is specifically as follows: S51: Divide the SS patient group and the HC into a training set and a validation set. Both the training set and the validation set include the SS patient group and the HC research objects. The ratio of the total number of people in the training set to the validation set is close to 3:

1. S52: Use SIMCA 14.0 software to perform multivariate analysis on the untargeted metabolome data of the training set. The multivariate analysis includes principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA). Screen for differential metabolites by combining the variable importance in projection (VIP), fold change (FC), and false discovery rate (FDR) values obtained from the OPLS-DA model. Consider metabolites with VIP > 1, FC > 1.5 or FC < 0.67, and FDR < 0.05 as differential metabolites. Finally, combine with relevant databases for identification, and manually remove exogenous differential metabolites, retaining endogenous differential metabolites. S53: Further screen the endogenous differential metabolites based on the Lasso regression model. Use the ten-fold cross-validation method to screen out the optimal endogenous differential metabolites related to the early diagnosis of SS. Use R language to build a binary Logistics regression model to fit the optimal endogenous differential metabolites, and screen out the optimal endogenous differential metabolites with P value < 0.05, which are the biomarkers. The biomarkers include MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide. S54: Establish prediction models according to two machine learning models, support vector machine (SVM) and random forest (RF), and analyze and verify the biomarkers obtained in step S53 in the training set and the validation set to determine the biomarker combination for differentiating the SS patients and the healthy controls. The biomarker combination is MG(0:0 / 14:0 / 0:0), LysoPE(14:0 / 0:0), Sphinganine, and Octadecanamide.

8. Use of the biomarker combination according to claim 7 in the preparation of a kit for the early diagnosis of Sjogren's syndrome, characterized in that, In step S52, KEGG pathway analysis was performed on the screened differential metabolites, and 13 metabolic pathways related to SS were screened out, including sphingolipid metabolism, biosynthesis of phenylalanine, tyrosine and tryptophan, phenylalanine metabolism, α-linolenic acid metabolism, arachidonic acid metabolism, glycerophospholipid metabolism, porphyrin and chlorophyll metabolism, pentose and glucuronate interconversions, cysteine and methionine metabolism, steroid biosynthesis, arginine and proline metabolism, pyruvate metabolism, and primary bile acid biosynthesis.

9. Use of the biomarker combination according to claim 7 in the preparation of a kit for the early diagnosis of Sjogren's syndrome, characterized in that, The relevant databases include the OSI / SMMS software and other online databases, and the other online databases include the Human Metabolome Database and the Lipidmaps Database.

10. Use of the biomarker combination according to claim 7 in the preparation of a kit for early diagnosis of Sjogren's syndrome, characterized in that, ROC curve analysis was performed on each biomarker and 4 biomarker combinations screened in step S53. The 4 biomarker combination is MG(0:0 / 14:0 / 0:0)+LysoPE(14:0 / 0:0)+Sphinganine+Octadecanamide, and the diagnostic efficacy of the 4 biomarker combination in the early diagnosis of SS is better than that of single biomarkers.

Images