Metabolic biomarker combination for depression diagnosis

Abstract

The invention discloses a metabolic biomarker combination for diagnosing depression. The metabolic biomarker combination comprises proline, betaine, alanine, tryptophan, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine and 2-hydroxybutyric acid. Based on a metabolic biomarker combination, an optimal diagnosis model is screened through a machine learning model algorithm, and the kit for depression diagnosis is constructed.

Description

Technical Field

[0001] The present invention relates to the field of metabolomics, and more particularly, to a combination of metabolic biomarkers for the diagnosis of depression. Background Art

[0002] The low recognition rate of depression in the medical system is mainly because the clinical diagnosis of depression is mainly based on medical history, clinical symptoms, and disease course. Currently, the diagnosis of depression mainly relies on the symptoms described by the patient and the results of clinical interviews, and the diagnosis process is relatively subjective, which also causes problems such as low treatment rate and high recurrence rate of depression. The internationally common diagnostic criteria are ICD-10 and DSM-V. Moreover, there is a widespread phenomenon of low clinical visit rate among depression patients. Therefore, for the prevention and treatment of depression, an objective detection item is needed as a screening tool for depression management, which can not only advance the "defense line" of depression, actively detect a large number of potential depression patients, but also effectively supplement the overly subjective judgment means of the existing clinical diagnosis of depression.

[0003] Studies have shown that the medication compliance of depression patients is poor, and the recurrence rate within 1 year is about 33%, which seriously affects the treatment effect and treatment outcome of depression patients. After the patients return to society, the disease usually recurs irregularly, making it difficult to determine the treatment endpoint of depression. For depression patients at different stages, the required treatment methods and medications are not the same. Most patients cannot comply with the full-course treatment strategy, which adds difficulties to clinical work. Therefore, it is necessary to monitor the disease course of depression patients, including treatment outcome and recurrence prediction, which is particularly important for providing timely and effective treatment for depression patients.

[0004] In recent years, there has been an increasing number of studies on depression based on metabolomics. A study found that plasma hydroxylated sphingosine is an important predictor of the treatment outcome of depression after analyzing the metabolomics of 298 patients with depression (Joyce J B. Using Multi-Domain Data to Predict Remission of Unipolar Depressive Episodes Following Antidepressant Therapy[D]. College of Medicine-Mayo Clinic, 2020.). Another study showed that the levels of lactate, glucose, glutamine, creatinine, acetate, L-valine, L-alanine, and fatty acids were significantly decreased in serum, while the levels of acetone and choline were significantly increased (Song Z, Wang H, Yin X, et al. Application of NMR metabolomics to search for human disease biomarkers in blood[J]. Clinical Chemistry and Laboratory Medicine (CCLM), 2019, 57(4): 417-441.). Many studies have demonstrated that metabolites play important roles in the mechanism, medication, and prognosis of depression. Developing a kit for the selected molecular markers is of great significance for clinical application.

[0005] Although many research institutions have studied the diagnostic markers of depression, these existing studies do not provide a diagnostic method that can be accurately, simply, reliably, and widely applied clinically.

[0006] In Chinese patent applications with application numbers 201610105047.3 and 201080046087.6, it is disclosed that serum metabolic markers are used for the diagnosis of depression. However, it is difficult to collect serum samples, which requires professional medical staff and is difficult for patients to collect at home by themselves. In Chinese patent application with application number 201480019392.4, it is disclosed that biomarkers in serum or urine are used for diagnosis. A set of biomarkers used includes at least cGMP, cortisol, calprotectin, thromboxane, aldosterone, HVEM, and substance P. Since the markers used include both small molecule metabolites and large molecule proteins, multiple technical platforms need to be used for detection, such as mass spectrometry and immunological methods, which increases the detection difficulty and cost. Moreover, this invention uses serum and urine samples, and the storage and transportation costs are very high. Chinese patent application with application number 202010658243.X uses urine to make filter paper for sample collection. Although some problems existing in serum and plasma are effectively solved, its detection purpose is single, only predicting the diagnosis of depression. During the course of depression, graded diagnosis, differential targeting, treatment outcome, and recurrence also have a great impact on patients. Traditional analytical methods such as immunoassay have always been the most widely used technology in clinical diagnosis. In immunoassay methods, radioimmunoassay has been basically phased out due to radioactive contamination. ELISA methods are manually operated, and the results are easily affected by human factors. Due to the needs such as detection convenience and accuracy of detection results, new detection markers, detection methods, and detection tools need to be developed.

[0007] The kit provided by the Chinese patent application with the application number 201610105047.3 can measure the contents of 14 metabolic markers screened in blood: methionine, nicotinic acid, phosphoserine, lactic acid, glutamic acid, hypoxanthine, choline, tryptophan, 5-hydroxytryptamine, propionyl carnitine, glutaric acid carnitine, estradiol, corticosterol and estriol. Using this kit, the specificity for the early diagnosis of depression reaches 86.8%, the sensitivity reaches 96.8%, and the accuracy rate reaches 93.0%. The Chinese patent application with the application number 201080046087.6 only relates to molecular markers for diagnosing depression and does not indicate the diagnostic efficacy of metabolite combinations. The literature (Doctoral dissertation of Chongqing Medical University, author: Zheng Peng, Screening of diagnostic markers for plasma and urine of depression based on metabolomics strategy, 2013 and Zheng P, Wang Y, Chen L, et al. Identification and validation of urinary metabolite biomarkers for major depressive disorder. Mol Cell Proteomics. 2013;12(1):207-214) discloses that 126 urine samples of depression patients and 134 normal control urine samples were measured by nuclear magnetic resonance, and a total of 23 differential metabolites were identified. Logistic regression analysis found that a biomarker group composed of five metabolites, namely malonate, formate, N-methylnicotinamide, M-hydroxyphenylacetate and alanine, has potential clinical diagnostic value. The AUC value of this biomarker group for distinguishing 82 depression patients and 82 normal controls in the training set is 0.812 (95% confidence interval: 0.829 - 0.961); the AUC value for distinguishing 44 depression patients and 52 normal controls in the test set is 0.895 (95% confidence interval: 0.829 - 0.961).

[0008] Based on the above research status, there is still a need to develop a better combination of metabolic biomarkers for diagnosing depression, so that it has excellent characteristics such as higher specificity, sensitivity, accuracy rate and is convenient for detection when used for diagnosing depression. Summary of the Invention

[0009] One of the purposes of the present invention is to provide a combination of metabolic biomarkers for diagnosing depression. Based on this combination of metabolic biomarkers, an optimal diagnostic model can be screened through a machine learning model algorithm, and a diagnostic kit for depression can be constructed.

[0010] One further object of the present invention is to provide the use of the metabolic biomarker combination in the preparation of a kit for diagnosing depression in a subject.

[0011] One further object of the present invention is to provide a detection method for testing the types and contents of metabolic biomarkers in a subject's dried blood spot sample.

[0012] Based on the above object of the invention, the present invention provides a metabolic biomarker combination for diagnosing depression, comprising: proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid. Preferably, the present invention provides a metabolic biomarker combination for diagnosing depression, which is composed of proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid.

[0013] The present invention further provides a metabolic biomarker combination for diagnosing depression, comprising: methionine, phenylalanine, guanine, hypoxanthine, γ-aminobutyric acid, allantoin, uric acid, histidine, aspartic acid, threonine, androstenedione, cytosine, xanthine, bilirubin, pyruvic acid, linoleic acid, glutamine, serine, adipic acid, pseudouridine, tyrosine, proline, valine, ornithine, betaine, alanine, tryptophan, leucine, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid, glutamic acid, glucose, lysine, arginine.

[0014] The present invention further provides a kit for diagnosing depression, which comprises the above-mentioned metabolic biomarker combination. Further, the present invention provides a kit for detecting the above-mentioned metabolic markers, including a standard product, a quality control product, an isotope internal standard, a diluent, formic acid, and an ammonium acetate solution of the metabolic marker. The extraction solvent of the metabolic marker is selected from one or a mixture of water, methanol, ethanol, acetonitrile, and isopropanol, and is required to be of chromatographic purity grade; the water, methanol, acetonitrile, isopropanol, ethanol, formic acid, and ammonium acetate used in the mobile phase all need to reach the chromatographic grade. The diluent is composed of one or more of methanol, water, and DMSO, and is required to be of chromatographic grade.

[0015] Further, the kit can be prepared according to the following method:

[0016] (1) Prepare a freeze-dried powder of a standard pure product of the metabolic marker;

[0017] (2) Prepare a freeze-dried powder of the isotope internal standard;

[0018] (3) Prepare an extraction solvent for the metabolic marker;

[0019] (4)Prepare the solvents for the standards and isotope internal standards;

[0020] (5)Assemble the above reagents into the kit.

[0021] Furthermore, the present invention provides a method for using the above kit, including:

[0022] (1)Collect a blood sample from the subject, prepare a dried blood spot paper, and use the dried blood spot sample as the test sample;

[0023] (2)Extract the dried blood spot sample with a metabolite extraction solvent and an isotope internal standard solution to obtain an extract;

[0024] (3)Use the targeted metabolomics analysis method of liquid chromatography-tandem mass spectrometry (LC-MS / MS) to determine the concentration of metabolites in the dried blood spot extract;

[0025] (4)Compare the measured metabolite concentration with the normal value of each metabolite concentration. If the concentrations of proline, betaine, alanine, tryptophan, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine, and 2-hydroxybutyric acid are significantly reduced (less than 0.8 times the normal value), it is determined as depression.

[0026] In addition, the test sample of the kit provided by the present invention can also be serum and / or plasma.

[0027] The present invention can prepare the dried blood spot sample by the following method. The steps include: (1) Collect the dried blood spot card: The card should be clean, pollution-free, and mold-free, and should be sealed before use. Each card should have independent identification information (name, age, gender, date code). The card is a special card with a dedicated blood collection area. (2) The dried blood spot card is divided into two parts. One part is the blood collection area, which has an inner circle with a radius of 3 mm and an outer circle with a radius of 5 mm. Contact with this area is prohibited. The other part is the holding area, and clean medical gloves should be worn when holding this area. (3) The dried blood spot card contains an antioxidant (VC) and an enzyme inactivator 1-aminobenzotriazole (ABT). (4) Capillary blood and venous blood can be used for blood sampling. When collecting capillary blood from the fingertip, etc., it should be cleaned in advance, and non-volatile disinfectants such as iodophor should not be used. Drop a drop of blood into each inner circle of the dried blood spot card, ensure that the entire circle is covered and the filter paper is completely penetrated, and at the same time avoid the blood from soaking into the outer circle. The collection process should be carried out in the dark. If conditions do not permit, strong light should be avoided to prevent the influence on photosensitive substances. (5) After the blood spot is collected, avoid long-term exposure to the air. The nitrogen blowing method can be used for rapid drying. Or a vacuum instrument can be used for rapid drying. The collected dried blood spot is quickly immersed in liquid nitrogen to inactivate the enzyme, and then dried. The cards cannot be stacked. After being individually packed in a sealed bag, they are placed in a -80°C refrigerator for freezing.Figure 7 A schematic example of a dried blood spot card is provided.

[0028] In the present invention, the processing method of the dried blood spot sample is as follows: Use a punch to cut a small round piece with a diameter of 10 mm along the outer circle line in the dried blood spot sample area, place it in a 1.5 mL centrifuge tube, and then add 500 mL of an eluent / precipitant containing an internal standard. Vortex for about 5 minutes. Centrifuge at 15000 g for 10 minutes, take the supernatant, perform secondary centrifugation (or filter with a membrane), and inject the supernatant for analysis. During the punching process with the punch, punching must be strictly carried out according to the outer circle of the card paper to prevent cross-contamination between samples during the punching process and affect the test results.

[0029] The metabolic markers in the present invention can be used for the diagnosis of depression or mood disorders in a subject.

[0030] In the present invention, the depression is selected from mild depression, moderate to severe depression, or bipolar depression. For the sake of concise expression, in other parts of the present invention, it is expressed as "depression". Those skilled in the art can understand that the principles and purposes of the technical solutions thereof are equally applicable to other types of mood disorders.

[0031] In the present invention, a "metabolic marker" refers to a unique biological or bio-derived indicator of a process, event, or condition. The biomarker can be used for the diagnosis of depression. Using the biomarker provided by the present invention for the diagnosis of depression has high specificity, sensitivity, and accuracy. The present invention enables the diagnosis of depression to no longer rely entirely on the subjective judgment of clinicians and questionnaire scales, greatly improving the diagnosis and treatment accuracy.

[0032] In the present invention, the "subject" is a human.

[0033] Based on the targeted metabolomics analysis technology of liquid chromatography-tandem mass spectrometry (LC-MS / MS), the inventors of the present invention detected healthy people and depression patients to obtain various biomarkers that can be used for the diagnosis of depression, and can achieve high specificity, sensitivity, and accuracy.

[0034] Detecting metabolites in the dried blood spot paper of the present invention can not only effectively avoid relying on the subjective judgment of doctors, but also assist in the diagnosis of depression, which is of great clinical significance; the sample is stored on the dried blood spot paper, which has the advantages of simple production and collection, convenient transportation, and can be stored at room temperature. The present invention uses the liquid chromatography-tandem mass spectrometry (LC-MS / MS) method to target and quantify the biomarker to be detected, and the sample detection is accurate and convenient.

[0035] In the diagnosis of depression, the present invention adopts a generalized linear model (GLM), with an area under the curve (AUC) value of 1, an accuracy of 99.30%, a sensitivity of 100%, and a specificity of 98.96%. This indicates that the metabolic biomarker combination provided by the present invention and the constructed model have excellent diagnostic value, significantly superior to previous similar reports. The metabolic biomarker combination provided by the present invention is innovative. In particular, there is currently no study showing the correlation between 2-hydroxybutyric acid and the diagnosis of depression. 2-hydroxybutyric acid is related to insufficient energy metabolism and impaired glucose regulation. There is a study showing that 2-hydroxybutyric acid is significantly elevated in patients with bipolar disorder ( , , ,et al.MetabolomicProfiling of Bipolar Disorder by 1H-NMR in Serbian Patients[J].Metabolites,2023,13(5): 607.). However, there is no study showing that 2-hydroxybutyric acid can be used as a diagnostic molecular marker for depression. Adding 2-hydroxybutyric acid as a diagnostic molecular marker in the present invention is innovative. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Represents the chromatogram of metabolic markers.

[0037] Figure 2 Shows a PCA analysis of two groups of samples from healthy people and depression patients.

[0038] Figure 3 Shows another PCA analysis of two groups of samples from healthy people and depression patients.

[0039] Figure 4 Shows an OPLS-DA analysis of two groups of samples from healthy people and depression patients.

[0040] Figure 5 Shows another OPLS-DA analysis of two groups of samples from healthy people and depression patients.

[0041] Figure 6 Shows the ROC curve of the generalized linear model (GLM) for distinguishing healthy people from depression patients.

[0042] Figure 7 Is an exemplary schematic diagram of a dried blood spot card. DETAILED DESCRIPTION OF THE INVENTION

[0043] To make the objectives, technical solutions and effects of the present invention clearer and more definite, the following further describes the present invention in detail with reference to the accompanying drawings and by way of examples. It should be understood that the specific examples described herein are only used to explain the present invention and are not used to limit the present invention.

[0044] The present invention provides a group of metabolic markers that can be used for the diagnosis of depression. The metabolic markers include: proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid. Alternatively, the metabolic markers include: methionine, phenylalanine, guanine, hypoxanthine, γ-aminobutyric acid, allantoin, uric acid, histidine, aspartic acid, threonine, androstenedione, cytosine, xanthine, bilirubin, pyruvic acid, linoleic acid, glutamine, serine, adipic acid, pseudouridine, tyrosine, proline, valine, ornithine, betaine, alanine, tryptophan, leucine, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid, glutamic acid, glucose, lysine, arginine. The present invention preferably uses targeted mass spectrometry technology, and other determination technologies such as immunoassay technology and chemiluminescence method can also be used for the determination of the metabolic markers of the present invention.

[0045] The metabolic markers of the present invention are screened by the following method:

[0046] (1) Select patients with depression and healthy control groups respectively, and take plasma samples respectively to make dried blood spots, and extract metabolites therefrom;

[0047] (2) By means of metabolomics, use liquid chromatography-tandem mass spectrometry (LC-MS / MS) to measure the contents of the main metabolites in the dried blood spots;

[0048] (3) Analyze the obtained data by orthogonal partial least squares discriminant analysis (OPLS-DA) and variable importance in projection (VIP) analysis to find out the metabolites with significant differences between the two groups as alternative metabolic markers;

[0049] (4) Through the receiver operator characteristic curve (ROC curve), further screen and determine a combination of metabolic markers with high accuracy and specificity.

[0050] These metabolic markers can be detected in the blood samples of the subjects. The present invention preferably uses the plasma samples of the subjects and makes dried blood spots as the detection samples.

[0051] During the determination, metabolite extraction from dried blood spots is carried out first. The present invention uses, but is not limited to, water, methanol, ethanol, acetonitrile, isopropanol, and mixed solutions of the above solvents to extract metabolites from dried blood spots, which may or may not contain isotope internal standards corresponding to each biomarker. Preferably, a 5-15% (volume ratio) aqueous methanol solution is used to extract metabolites from dried blood spots. Preferably, the extraction is carried out at room temperature (20-35°C) with shaking for 20-40 minutes, and the dried blood spot extract is subjected to subsequent determination.

[0052] During the determination, the present invention preferably uses a derivatization method to detect metabolite markers. The derivatization methods include, but are not limited to, dansyl chloride, phenylisothiocyanate, 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate, etc. Preferably, dansyl chloride is used for derivatization. Preferably, the supernatant of the dried blood spot extract is taken, and a sodium bicarbonate solution and a derivatization reagent are added. Shake for a period of time to allow sufficient reaction. Preferably, ethyl acetate extraction is used for the metabolite markers derivatized with dansyl chloride. The upper organic phase of the extraction solution is transferred to an appropriate container and dried under nitrogen. After reconstitution with an aqueous methanol solution, it is detected.

[0053] The present invention preferably uses liquid chromatography-tandem mass spectrometry (LC-MS / MS) to detect and analyze the above diagnostic marker composition.

[0054] Preferably, the liquid chromatography conditions are as follows: chromatographic column (which can be a C8 or C18 silica gel packing column), mobile phase (which can be an acetonitrile-aqueous solution, formic acid aqueous solution, methanol-aqueous solution, acetonitrile-isopropanol, methanol-isopropanol solution, acetic acid aqueous solution, ammonium acetate aqueous solution), and the elution program is: mobile phase A and mobile phase B are subjected to gradient elution in a certain proportion.

[0055] The liquid chromatography conditions of a specific embodiment of the present invention are as follows: the liquid chromatography column uses ZORBAX RRExtend-C18, 80A, 4.6×150 mm, 3.5 μm. The liquid chromatography conditions are: injection volume: 5 μL; flow rate: 0.6 mL / min; column temperature: 40°C; autoinjector temperature: 4°C; mobile phase A phase: an aqueous solution of 0.1% (volume ratio) formic acid; mobile phase B phase: methanol; the elution program is 0-1 min, linearly changing from 10% B phase to 40% B; 1-1.5 min, linearly changing from 40% B phase to 50% B phase; 1.5 min-2.0 min, linearly changing from 50% B to 60% B phase; 2.0 min-3.2 min, equilibrating at 80% B phase; 3.2 min-3.6 min, from 80% B phase to 10% B phase; 3.7-4.5 min, equilibrating at 10% B phase.

[0056] Preferably, triple quadrupole mass spectrometry is used for detection. An electrospray ionization source (ESI) is adopted, in the positive and negative ion transient modes, with a scanning time of 0 min to 4.5 min; CUR: 20; CAD: Medium; IS: 5000; TEM: 500; GS1: 30; GS2: 50; Interface heater (ihe): on; DP: ±36; CE: ±26; EP: ±15; CXP: ±10. Dwell Time: 20 ms; Resolution Q1: Unit; Resolution Q3: Unit; Pause between mass: 5.007 ms.

[0057] After data acquisition, the metabolic markers of each diagnostic marker are quantified using specific ion pairs. The peak area of the acquired signal is compared with the working curve of the corresponding standard, and is corrected by adding an internal standard solution to obtain the concentration value of the metabolic marker. The content of each metabolic marker in the original blood.

[0058] A specific embodiment of the present invention is a depression diagnosis kit, which includes: freeze-dried powder of 1 mg standard of metabolic marker, 1 mg quality control standard of marker, 13 C or 15 N or 2 Freeze-dried powder of a stable isotope internal standard mixture with at least three labels of H (deuterium), 5 mL of methanol-water-DMSO solution (diluent) for dissolving the standard powder and the isotope internal standard powder, 1 mL of formic acid solution, ammonium acetate solution with a concentration of 5 mol / L for adjusting the pH of the mobile phase, methanol solution as the metabolite extraction solvent, and 1 analytical chromatography column.

[0059] Preferably, in the above-mentioned depression detection kit, the metabolic markers include: proline, betaine, alanine, tryptophan, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine, 2-hydroxybutyric acid. Alternatively, the metabolic markers include: methionine, phenylalanine, guanine, hypoxanthine, γ-aminobutyric acid, allantoin, uric acid, histidine, aspartic acid, threonine, androstenedione, cytosine, xanthine, bilirubin, pyruvic acid, linoleic acid, glutamine, serine, adipic acid, pseudouridine, tyrosine, proline, valine, ornithine, betaine, alanine, tryptophan, leucine, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine, 2-hydroxybutyric acid, glutamic acid, glucose, lysine, arginine.

[0060] The preparation method of the above-mentioned depression detection kit is as follows:

[0061] (1) Prepare the sample: Take the sample to be analyzed and ensure that the sample is within the required concentration range. Dilute the sample as needed (usually diluted with the mobile phase). Filter the sample to remove possible particulate matter, usually using a 0.45 μm or 0.2 μm filter membrane.

[0062] (2) Select the mobile phase: Inject the metabolic extract into the Extend C18 chromatographic column (ZORBAX RR Extend-C18, 80A, 4.6×150 mm, 3.5 um, USA) through an autosampler to separate the metabolites in the blood sample. The specific liquid chromatography conditions are as follows: injection volume: 5 μL; flow rate: 0.6 mL / min; column temperature: 40 °C; autosampler temperature: 4 °C; mobile phase A: aqueous solution of 0.1% (volume ratio) formic acid; mobile phase B: methanol; elution program: 0 - 1 min, linearly change from 10% B to 40% B; 1 - 1.5 min, linearly change from 40% B to 50% B; 1.5 min - 2.0 min, linearly change from 50% B to 60% B; 2.0 min - 3.2 min, equilibrate at 80% B; 3.2 min - 3.6 min, from 80% B to 10% B; 3.7 - 4.5 min, equilibrate at 10% B.

[0063] (3) Set up the equipment: Check each part of the HPLC system, including the pump, injector, chromatographic column, and detector. Ensure that they are all in normal working condition. Introduce the metabolites separated by chromatography into a triple quadrupole mass spectrometer, and scan and detect the metabolites using the multiple reaction monitoring mode. Use an electrospray ionization source (ESI), positive and negative ion instantaneous switching mode, and the scanning time is 0 min - 4.5 min. CUR: 20; CAD: Medium; IS: 5000; TEM: 500; GS1: 30; GS2: 50; Interface heater (ihe): on; DP: ±36; CE: ±26; EP: ±15; CXP: ±10. Dwell Time: 20 ms; Resolution Q1: Unit; Resolution Q3: Unit; Pause between mass: 5.007 ms.

[0064] (4) Inject the sample: Use an injector to introduce the prepared sample into the system. The typical injection volume is 10 - 100 μL. Ensure that the liquid flow of the sample is smooth and record the injection time.

[0065] (5) After data acquisition, each diagnostic marker's metabolite is quantified using a specific ion pair. Compare the peak area of the acquired signal with the working curve of the corresponding standard product, and add an internal standard solution for calibration to obtain the concentration value of the metabolite.

[0066] The quantitative values of the metabolic markers obtained from the above tests are used to perform corresponding depression diagnosis risk assessments by comparing them with the normal values of each metabolite concentration. The measured metabolite concentrations are compared with the normal values of each metabolite concentration. If the concentrations of proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, and 2-hydroxybutyric acid are significantly reduced (less than 0.8 times the normal value), it is determined to be depression.

[0067] Example

[0068] The technical solution of the present invention will be further described below in conjunction with specific examples.

[0069] Example 1 Screening and model establishment of metabolic markers for depression diagnosis

[0070] 1. Inclusion of samples

[0071] In this example, the following inclusion criteria are adopted for the depression sample group: (1) All included depression patients meet the ICD-10 diagnostic criteria; (2) The score of the 17-item Hamilton Depression (HDRS) scale is greater than or equal to 18 points; (3) Patients with first-episode depression who have not taken any antidepressant drugs; (4) Aged over 18 years and less than 60 years.

[0072] At the same time, the following exclusion criteria are adopted for the depression sample group: (1) Those with current or previous history of other mental diseases; (2) Those with combined organic brain diseases and severe brain trauma history, or those with combined heart, liver, kidney diseases, diabetes and other severe somatic diseases, etc.; (3) Those with abnormal routine laboratory tests (blood routine, liver function, urine routine); (4) Female research subjects during pregnancy, lactation, and menstrual periods; (5) History of drug and substance abuse.

[0073] The inclusion criteria for the healthy control group are: no history of neuropsychiatric diseases, no history of drug abuse or dependence, no systemic somatic diseases; no obvious abnormalities in routine laboratory tests.

[0074] 2. Preparation of dried blood spots

[0075] Prepare dried blood spot cards: The cards should be clean, free of contamination and mold, and sealed before use. Each card should have independent identification information (name, age, gender, code). The card has a blood collection area. The dried blood spot card is divided into two parts: The first part is the blood collection area, which has an inner circle with a radius of 3 mm and an outer circle with a radius of 5 mm, and contact with this area is prohibited; the second part is the holding area, and clean medical gloves should be worn when holding this area. The dried blood spot card contains an antioxidant (VC) and an enzyme inactivator 1-aminobenzotriazole (ABT).

[0076] Collecting blood samples using capillary blood and venous blood: Collect blood samples from the fingertip capillaries. Wash the fingertips in advance and do not use non-volatile disinfectants such as iodine. Then, drop one drop of blood into each inner circle of the dried blood spot card, ensuring that the entire circle is covered and the filter paper is completely penetrated, while avoiding blood staining on the outer circle. Avoid strong light during the collection process to prevent affecting photosensitive substances. After collecting the blood spots, avoid long-term exposure to air and quickly dry them using nitrogen blowing or a vacuum instrument. Quickly soak the collected dried blood spots in liquid nitrogen to inactivate the enzymes, and then air-dry. The cards cannot be stacked. After individually packaging them in sealed bags, store them in a -80°C freezer.

[0077] 3. Extraction of biomarkers from dried blood spot paper

[0078] (1) Use a punch to cut a small round piece with a diameter of 10 mm along the outer circle line in the dried blood spot sample area, place it in a 1.5 mL centrifuge tube, and then add 500 μL of the eluent / precipitant containing the internal standard. Vortex for about 5 minutes. Centrifuge at 15000 g for 10 minutes, take the supernatant, perform secondary centrifugation (or filter with a membrane), and inject the supernatant for analysis.

[0079] (2) Add 500 μL of the methanol-aqueous solution (metabolite extraction solvent) in the depression detection kit, and shake at 20°C and 1450 rpm for 45 minutes;

[0080] (3) Take 10 μL of the extract and add 90 μL of water. Then, sequentially add 200 μL of the internal standard working solution, 50 μL of 1M NaHCO 3 solution, 100 μL of 1 mg / mL derivatization reagent, and shake at 30°C and 1450 rpm for 30 minutes.

[0081] 4. Detection of biomarkers from dried blood spot paper

[0082] Based on the targeted metabolomics analysis method of liquid chromatography-tandem mass spectrometry (LC-MS / MS), the detected metabolites include: methionine, phenylalanine, guanine, hypoxanthine, γ-aminobutyric acid, allantoin, uric acid, histidine, aspartic acid, threonine, androstenedione, cytosine, xanthine, bilirubin, pyruvic acid, linoleic acid, glutamine, serine, adipic acid, pseudouridine, tyrosine, proline, valine, ornithine, betaine, alanine, tryptophan, leucine, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine, 2-hydroxybutyric acid, glutamic acid, glucose, lysine, arginine.

[0083] The specific steps are as follows:

[0084] (1) Prepare the sample: Take the sample to be analyzed and ensure that the sample is within the required concentration range. Dilute the sample as needed (usually diluted with the mobile phase). Filter the sample to remove possible particulate matter, usually using a 0.45 μm or 0.2 μm filter membrane.

[0085] (2) Select the mobile phase: Inject the metabolic extract into the Extend C18 chromatographic column (ZORBAX RR Extend-C18, 80A, 4.6 x 150 mm, 3.5 um, USA) through an autosampler to separate the metabolites in the blood sample. The specific liquid chromatography conditions are as follows: injection volume: 5 μL; flow rate: 0.6 mL / min; column temperature: 40 °C; autosampler temperature: 4 °C; mobile phase A: aqueous solution of 0.1% (volume ratio) formic acid; mobile phase B: methanol; elution program: 0 - 1 min, linearly changing from 10% B to 40% B; 1 - 1.5 min, linearly changing from 40% B to 50% B; 1.5 min - 2.0 min, linearly changing from 50% B to 60% B; 2.0 min - 3.2 min, equilibrating at 80% B; 3.2 min - 3.6 min, from 80% B to 10% B; 3.7 - 4.5 min, equilibrating at 10% B.

[0086] (3) Set up the equipment: Check each part of the HPLC system, including the pump, injector, chromatographic column, and detector. Ensure that they are all in normal working condition. Introduce the metabolites separated by chromatography into a triple quadrupole mass spectrometer, and scan and detect the metabolites using the multiple reaction monitoring mode. Use an electrospray ionization source (ESI), positive and negative ion instantaneous switching mode, and the scanning time is 0 min - 4.5 min. CUR: 20; CAD: Medium; IS: 5000; TEM: 500; GS1: 30; GS2: 50; Interface heater (ihe): on; DP: ±36; CE: ±26; EP: ±15; CXP: ±10. Dwell Time: 20 ms; Resolution Q1: Unit; Resolution Q3: Unit; Pause between mass: 5.007 ms.

[0087] (4) Inject the sample: Use an injector to introduce the prepared sample into the system. The typical injection volume is 10 - 100 μL. Ensure that the liquid flow of the sample is smooth and record the injection time.

[0088] (5) After data collection, the metabolic markers of each diagnostic marker were quantified using specific ion pairs. The peak area of the collected signal was compared with the working curve of the corresponding standard, and an internal standard solution was added for calibration to obtain the concentration value of the metabolic marker. Then, the metabolite concentrations of the measured depression sample group and the healthy control group were compared. 4) Stability experiment:

[0089] After the dried blood spot paper was air-dried, it was stored at -80 °C, -20 °C, and room temperature respectively. The content of substances in the dried blood spot was determined by liquid chromatography-mass spectrometry. The samples stored at -80 °C were measured once a month, the samples stored at -20 °C were measured once every half month, and the samples stored at room temperature were measured once every other day. The relative standard deviation (RSD) was calculated, with the initial measurement mean as the initial value, and the recovery rate was calculated. The results showed that the precision was within 10%, and the recovery rate was between 80% and 110%. It indicated that the dried blood spot paper could be stably stored for at least one year at -80 °C, at least 3 months at -20 °C, and 15 days at room temperature.

[0090] 5) Test results

[0091] 150 dry blood spot samples of normal people were collected, and the metabolites in the dry blood spot samples were quantified to determine the normal reference range of the markers. The chromatogram of the metabolic marker is as Figure 1 shown. During the analysis of dry blood spot samples, one QC sample (Quality Control samples) was inserted for every 10 research samples to evaluate the variability of the derivatization process and instrumental analysis. QC samples are crucial for ensuring the repeatability, reliability, accuracy, and robustness of metabolite quantitative analysis.

[0092] At the same time, 150 healthy people and 60 depression patients' samples were collected and detected. The sample test results showed that the differences in the following metabolite levels could well distinguish normal people from depression patients: proline, betaine, alanine, tryptophan, kynurenine, 5-hydroxytryptamine, creatine, succinic acid, taurine, 2-hydroxybutyric acid. That is, the concentrations of these metabolic biomarkers in the depression sample group were significantly lower than those in the healthy control group (Table 1).

[0093] Table 1. Differentially expressed metabolites in the depression group compared with the healthy group

[0094]

[0095] Figure 2 、 Figure 3It is the score plot of principal component analysis (PCA). As shown, the QC samples are closely clustered relative to other samples of dried blood spots, indicating that the method of this embodiment has good repeatability. Here, the QC sample is BSA (bovine serum albumin solution). Orthogonal partial least squares-discriminant analysis (OPLS-DA) was performed. As Figure 4 , Figure 5 shown, normal and depression samples can be well distinguished, indicating that there are significant differences in metabolism between depression patients and normal people.

[0096] Furthermore, different machine learning algorithms are used to build models. The machine learning algorithms used include generalized linear models, linear discriminant analysis, K-nearest neighbor algorithm, logistic regression, lasso regression, decision trees, artificial neural networks, support vector machines, extreme gradient boosting, random forests, principal component analysis, Bayesian networks, and linear regression. Among them, the operation method of the generalized linear model is as follows: select an appropriate link function and distribution, and use maximum likelihood estimation (MLE) to fit the model. It can be implemented through statistical software (such as the glm function in R). The operation method of linear discriminant analysis (LDA) is: calculate the means and covariance matrices of each class, and solve the eigenvalues and eigenvectors to obtain the optimal partition. It is usually implemented in R or sklearn in Python. The operation method of the K-nearest neighbor algorithm (KNN) is: select the value of K, calculate the distances between the point to be classified and all points in the training data, sort by distance, select the K nearest neighbors, and determine the class by voting. In Python, KNeighborsClassifier in sklearn can be used. The operation method of logistic regression is: model using the logistic function (sigmoid) and optimize the parameters using maximum likelihood estimation. It can be implemented through glm in R or sklearn in Python. The operation method of lasso regression is: add an L1 regularization term on the basis of linear regression and implement it using coordinate descent or the least squares method. There is a Lasso class in sklearn of Python. The operation method of decision trees is: recursively select the best feature for node partitioning until the stopping condition is met. It can be implemented using DecisionTreeClassifier or DecisionTreeRegressor in sklearn. The operation method of artificial neural networks is: design the network structure (number of layers, number of neurons), and use the backpropagation algorithm to optimize the parameters. Deep learning frameworks such as TensorFlow or PyTorch can be used. The operation method of support vector machines (SVM) is: select an appropriate kernel function and optimize the objective function to find the best hyperplane. It can be implemented through SVC in sklearn. The operation method of extreme gradient boosting is: build gradient boosting trees and iteratively optimize the loss function. The xgboost library can be used for implementation, and hyperparameters are set for tuning. The operation method of random forests is: build multiple decision trees and use the Bootstrap sampling method to generate different training sets. Finally, vote or average the prediction results of each tree. RandomForestClassifier or RandomForestRegressor in sklearn can be used. The operation method of principal component analysis (PCA) is: standardize the data, calculate the covariance matrix, find the eigenvalues and eigenvectors, and select the first K eigenvectors as the principal components. PCA in sklearn can be used.Operation methods of Bayesian network: Define the network structure, update probabilities using Bayes' theorem, and usually perform parameter learning using maximum likelihood estimation or Bayesian estimation. It can be implemented using the pgmpy library. Operation methods of linear regression: Establish a linear equation and fit parameters using the least squares method. It can be implemented using lm in R or LinearRegression in sklearn of Python.

[0097] It was found that, as shown in Table 2, the best model was the generalized linear model (GLM). The significantly changed metabolites were used to draw the receiver operating characteristic curve (ROC). The area under the curve (AUC) value was 1, the accuracy was 99.30%, the sensitivity was 100%, and the specificity was 98.96%. See Figure 6 . It indicates that the model has good diagnostic value. The combination of metabolites proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, and 2-hydroxybutyric acid showed excellent specificity and sensitivity.

[0098] Table 2. Comparative analysis of the results of different models for diagnosing depression

[0099]

[0100] Example 2 Evaluate the diagnostic effect of the metabolic biomarker combination and diagnostic model established in Example 1 on depression

[0101] Randomly select dried blood spot samples from outpatients or inpatients in a certain hospital and dried blood spot samples from recruited healthy controls (a total of 300 samples. All samples were collected with clear diagnostic data by professional doctors according to the ICD-10 standard and informed consent forms were signed). Then, directly detect the ion abundance ratios of the biomarkers (proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinic acid, taurine, 2-hydroxybutyric acid) and internal standards according to the method of Example 1. Input the data into the diagnostic model established in Example 1 to obtain the diagnostic results. Compare the diagnostic results with the results of diagnosing depression for these 300 samples by professional doctors using ICD-10. The results are shown in Table 3.

[0102] Table 3. Evaluation of the diagnostic effect of the metabolic marker kit for depression

[0103]

[0104] As can be seen from Table 3, using the biomarker combination in Example 1 and the diagnostic model established therefrom, the accuracy for diagnosing depression was 100%, the sensitivity was 100%, and the specificity was 100%.

[0105] The above has described the embodiments of the present invention in detail, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principle and spirit of the present invention, various changes, modifications, substitutions, and variations made to these embodiments still fall within the protection scope of the present invention.

Claims

1. A metabolic biomarker combination for the diagnosis of depression, characterized in that: The metabolic biomarker combination consists of proline, betaine, alanine, tryptophan, kynurenine, serotonin, creatine, succinate, taurine and 2-hydroxybutyrate, wherein the depression is selected from mild depression, moderate to severe depression or bipolar depression.

2. The metabolic biomarker combination according to claim 1, wherein: The detection sample for the metabolic biomarkers is human serum, plasma, or dried blood spots.

3. A kit for diagnosing depression, characterized in that: A reagent for measuring the metabolic biomarker combination according to any one of claims 1 to 2.

4. The kit according to claim 3, further comprising: a standard substance of the metabolic biomarker combination, a quality control substance, an isotope internal standard, a diluent, formic acid and an ammonium acetate solution.

5. The kit according to claim 4, comprising: 1 mg of lyophilized powder of a standard substance of a metabolic marker; 1 mg of a quality control standard substance of the marker; 13 C or 15 N or 2 H-labeled stable isotope internal standard mixed lyophilized powder with at least three ions; 5 mL of methanol-water-DMSO solution, used to dissolve the standard powder and isotope internal standard powder; 1 mL of formic acid solution and 5 mol / L ammonium acetate solution, used to adjust the pH of the mobile phase; Methanol solution was used as the metabolite extraction solvent; 1 analytical column.

6. The kit according to claim 5, wherein the method for using the kit is: (1) Collecting a blood sample from a subject, preparing a dried blood spot paper, and using the dried blood spot sample as a test sample; (2) extracting the dried blood spot sample using a metabolite extraction solvent and an isotope internal standard solution to obtain an extract; (3) Using a targeted metabolomics analysis method coupled with liquid chromatography-tandem mass spectrometry, the concentration of metabolites in the dried blood spot extract was measured; (4) The measured metabolite concentrations are compared with the normal values ​​of each metabolite concentration. If the concentrations of proline, betaine, alanine, tryptophan, kynurenine, 5-hydroxytryptamine, creatine, succinate, taurine and 2-hydroxybutyrate are less than 0.8 times the normal value, it is diagnosed as depression.

Images