Alzheimer's disease biomarker and novel method for searching Alzheimer's disease biomarker based on erythrocyte membrane lipidomics

Abstract

The invention provides application of lipid as a marker in preparation of a product for diagnosis or auxiliary diagnosis of Alzheimer's disease and / or mild cognitive impairment. The invention also provides an application of a substance for extracting lipid or a substance for detecting the expression level of lipid in preparation of a product for diagnosis or auxiliary diagnosis of Alzheimer's disease and / or mild cognitive impairment. The lipid is one or a combination of more of DAG 38: 2, TAG (14: 0 / 24: 1 (15Z) / 20: 0), TAG 58: 2, TAG 55: 5, TAG 53: 1, TAG 49: 3, TAG 45: 2, TAG 43: 1, PS 42: 8, PI (22: 1 (11Z) / 14: 0), PI (18: 2 (9Z, 12Z) / 18: 1 (9Z)), PC (22: 5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18: 0), LPC (24: 1 (15Z The substance for extracting the lipid is a combination of isopropanol and a mixture of isopropanol, acetonitrile and water in a volume ratio of 4: 3: 1. In the extraction of the lipid difference substances, the red cell membrane (RCM) is creatively used as an extraction object, an extraction method is provided for the lipid difference substances of AD in the red cell membrane, 15 potential AD biological diagnosis markers are finally determined, and lipids prompt chronic physiological changes and are closely related to the pathogenesis of AD.

Description

Technical Field

[0001] The present invention belongs to the field of biodiagnostic medicine, and specifically relates to Alzheimer's disease biomarkers and a method for searching for Alzheimer's disease biomarkers based on red blood cell membrane lipidomics. Background Art

[0002] The incidence of Alzheimer's disease (AD) is on the rise worldwide, and the diagnosis of AD usually relies on cerebrospinal fluid (CSF) and positron emission tomography (PET). However, their invasiveness and high cost often lead to late diagnosis, which limits their use in clinical applications, especially in early diagnosis. Studies have shown that timely medical intervention may slow the progression of the disease. The discovery of Alzheimer's disease biomarkers has greatly promoted early diagnosis, enabling therapeutic intervention to be initiated quickly and continuously monitored. The identification of biomarkers in blood samples, including specific proteins and metabolites, has been shown to not only distinguish between healthy individuals (NS), patients with mild cognitive impairment (aMCI) and patients with Alzheimer's disease.

[0003] The relationship between lipid metabolism and Alzheimer's disease has become a key area of ​​research. Lipidomics provides a comprehensive analytical framework for extensive lipid profile analysis, revealing potential biomarkers and facilitating disease diagnosis and monitoring. Studies have shown that changes in lipid profiles are associated with the pathogenesis of AD. The discovery of lipid biomarkers has the potential to revolutionize our diagnostic and therapeutic strategies for Alzheimer's disease. Therefore, continued exploration of lipidomics is crucial for identifying new lipid biomarkers. Currently, conventional methods use plasma or serum as the lipidome target for AD.

[0004] The importance of Alzheimer's disease (AD) biomarkers lies in their potential to enable early diagnosis, facilitate treatment monitoring, and improve understanding of disease progression. Uncovering distinct lipid profiles associated with the disease can aid diagnosis and understanding of its pathogenesis.

[0005] Therefore, it is necessary to provide more lipid biomarkers and marker screening methods to further advance AD ​​diagnosis. Summary of the Invention

[0006] Unlike prior art methods that directly target clinically differential substances, this invention instead further investigates correlations with clinical blood test parameters to ultimately identify targets. Furthermore, in the extraction of differential lipids, the invention creatively targets red blood cell membranes (RCMs), pioneering the inclusion of RCM lipids and providing extraction methods for differential lipids in RCMs associated with AD. This ultimately led to the identification of 15 potential biomarkers for AD. These lipids indicate chronic physiological changes and are closely related to the pathogenesis of AD.

[0007] The DAG 38:2 mentioned in the present invention means that DAG represents diacylglycerol, 38 represents the number of C atoms, and 2 represents the number of double bonds.

[0008] TAG (14:0 / 24:1(15Z) / 20:0) means: TAG stands for triacylglycerol, different fatty acid chains are separated by “ / ” in the brackets, 14, 24, and 20 represent the number of C atoms in each fatty acid chain, respectively, and 0, 1, and 0 represent the number of double bonds in each fatty acid chain, respectively. In 24:1(15Z), “(15Z)” means that the double bond is at the 15th position and is in a cis configuration.

[0009] TAG 58:2 means: TAG stands for triacylglycerol, 58 represents the number of C atoms, and 2 represents the number of double bonds.

[0010] TAG 55:5 means: TAG stands for triacylglycerol, 55 represents the number of C atoms, and 5 represents the number of double bonds.

[0011] TAG 53:1 means: TAG stands for triacylglycerol, 53 represents the number of C atoms, and 1 represents the number of double bonds.

[0012] TAG 49:3 means: TAG stands for triacylglycerol, 49 represents the number of C atoms, and 3 represents the number of double bonds.

[0013] TAG 45:2 means: TAG stands for triacylglycerol, 45 represents the number of C atoms, and 2 represents the number of double bonds.

[0014] TAG 43:1 means: TAG stands for triacylglycerol, 43 represents the number of C atoms, and 1 represents the number of double bonds.

[0015] PS 42:8 means: PS stands for phosphatidylserine, 42 represents the number of carbon atoms, and 8 represents the number of double bonds.

[0016] PI(22:1(11Z) / 14:0) means: PI stands for phosphatidylinositol, the “ / ” in the brackets separates different fatty acid chains, 22 and 14 represent the number of C atoms in the fatty acid chain, 1 and 0 represent the number of double bonds in the fatty acid chain, and “(11Z)” in 22:1(11Z) means that the double bond is at the 11th position and is in a cis configuration.

[0017] PI (18:2 (9Z, 12Z) / 18:1 (9Z)) means: PI stands for phosphatidylinositol, “ / ” in the brackets separates different fatty acid chains, 18 and 18 represent the number of C atoms in the fatty acid chain, 2 and 1 represent the number of double bonds in the fatty acid chain, respectively. In 18:2 (9Z, 12Z), “(9Z, 12Z)” means that the double bonds are at the 9th and 12th positions and are in a cis configuration. In 18:1 (9Z), “(9Z)” means that the double bond is at the 9th position and is in a cis configuration.

[0018] PC(22:5(7Z,10Z,13Z,16Z,19Z) / 18:0) means: PC stands for phosphatidylcholine, “ / ” in the brackets separates different fatty acid chains, 22 and 18 represent the number of C atoms in the fatty acid chain, 5 and 0 represent the number of double bonds in the fatty acid chain, and “(7Z,10Z,13Z,16Z,19Z)” in 22:5(7Z,10Z,13Z,16Z,19Z) means that the double bonds are at positions 7, 10, 13, 16, and 19 and are in cis configuration.

[0019] LPC(24:1(15Z)) means: LPC stands for lysophosphatidylcholine, 24 represents the number of carbon atoms, 1 represents the number of double bonds, and “(15Z)” indicates that the double bond is at the 15th position and is in the cis configuration.

[0020] LPC 22:0 means: LPC stands for lysophosphatidylcholine, 22 represents the number of carbon atoms, and 0 represents the number of double bonds.

[0021] FAHFA (16:0 / 7-O-18:0) means: FAHFA stands for fatty acid hydroxy fatty acid ester, the “ / ” in the brackets separates different parts, 16 and 18 represent the number of C atoms in different parts, 0 and 0 represent the number of double bonds, and “7-O-” indicates the presence of an oxygen connection structure at the 7th position.

[0022] The technical solutions of the present invention are as follows:

[0023] The first object of the present invention is to provide the use of lipids as markers in the preparation of products for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, wherein the lipids are a combination of one or more of DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0).

[0024] A second object of the present invention is to provide a substance for extracting lipids or a substance for detecting lipid expression levels in the preparation of a product for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, wherein the lipids are a combination of one or more of DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0).

[0025] Furthermore, the substance used for extracting lipids is a combination of isopropanol and a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:1.

[0026] Furthermore, the lipids are derived from the red blood cell membrane of the patient to be tested.

[0027] Furthermore, the decreased lipid expression level is closely related to the occurrence of Alzheimer's disease and / or mild cognitive impairment.

[0028] Furthermore, the application includes the following steps:

[0029] S1: Extract the red blood cell membrane of the patient to be tested;

[0030] S2: The erythrocyte membrane obtained in S1 was extracted with isopropanol and then redissolved in a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:1 to obtain the lipid sample to be tested;

[0031] S3: Detecting the expression level of the lipid in the lipid sample to be tested.

[0032] Furthermore, S2 specifically comprises mixing isopropanol with the red blood cell membrane obtained in S1, standing at 4°C for 20 minutes, shaking for 2 minutes, and then sonicating on ice for 10 minutes. The obtained sample mixture is centrifuged at 4°C and 12,000 rpm for 10 minutes, the supernatant is evaporated, and redissolved in a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:1.

[0033] Furthermore, the S3 is detected by ultra-high performance liquid chromatography tandem mass spectrometry, and the liquid chromatography conditions are:

[0034] A Waters ACQUITY UPLC CSH C18 column (1.7 μm, 100 mm × 2.1 mm) was used;

[0035] Column temperature 40°C;

[0036] Autosampler 8°C;

[0037] Flow rate 0.35 mL / min;

[0038] Mobile phase A was acetonitrile:water (60:40) containing 10 mM ammonium formate and 0.1% formic acid, and mobile phase B was isopropanol:acetonitrile (90:10) containing 10 mM ammonium formate and 0.1% formic acid;

[0039] Gradient elution conditions: 0-2.0 min, 40% B; 2.0-2.1 min, 45%-50% B; 2.1-11.0 min, 50-54% B; 11.0-11.1 min, 54%-70% B; 11.1-15.0 min, 70%-99% B; 15.0-18.0 min, return to 40% B;

[0040] Mass spectrometry conditions are:

[0041] DDA detection and data acquisition were performed in positive and negative ion modes with the following parameters: m / z range 50–1000 Da; capillary voltage +3.0 kV / −2.5 kV for positive / negative ion modes, respectively; sampling cone voltage 40 V; ion source temperature 100 °C; desolvation temperature 450 °C; cone gas flow rate 50 L / h; and desolvation gas flow rate 600 L / h.

[0042] The third object of the present invention is to provide a kit for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, the kit comprising a substance for extracting lipids or a substance for detecting the expression level of lipids, wherein the lipids are a combination of one or more of DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0).

[0043] The present invention uses a comprehensive blood lipid detection platform with optimized sample pretreatment and detection methods to analyze the blood lipid profiles of hundreds of subjects, including normal people, patients with mild cognitive impairment and AD patients. The results identified 101 differentially expressed lipids, and the 101 differentially expressed lipids were correlated with blood test parameters that have been reported to be closely related to AD (triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein cholesterol (HDL), lipoprotein (a) [LP(a)], low-cut whole blood viscosity (BVL), medium-cut whole blood viscosity (BVM), high-cut whole blood viscosity (BVH)). It was finally found that 68 of the 101 differentially expressed lipids were closely correlated with blood test parameters. In order to further accurately locate potential targets for MCI and AD, the above 68 differentially expressed lipids were analyzed for data between the normal group and the MCI group, and the normal group and the AD group. The ROC curve was used to accurately locate 15 potential biomarkers of mild cognitive impairment and AD. RCM lipids not only enhance the diagnostic landscape for mild cognitive impairment and AD but also highlight the innovative potential of RCM as a novel and underexplored source for uncovering novel biomarkers. The significance of lipid-derived biomarker discovery lies in their ability to provide insights into the metabolic dysregulation associated with mild cognitive impairment and AD, providing a window into the metabolic fingerprint of the disease and potentially enabling early diagnosis and treatment monitoring. This discovery underscores the importance of lipid-based biomarkers in advancing the field of AD diagnostics and opens new avenues for therapeutic intervention strategies.

[0044] By optimizing lipid extraction and data collection, this paper develops a comprehensive lipidomics method to identify potential biomarkers of mild cognitive impairment and Alzheimer's disease by analyzing the lipid profiles in red blood cell membrane samples from normal subjects, patients with mild cognitive impairment, and patients with Alzheimer's disease throughout the entire disease progression process, thereby facilitating the early diagnosis and treatment monitoring of Alzheimer's disease.

[0045] The present invention also provides an optimized method for discovering red blood cell membrane lipids as biomarkers for Alzheimer's disease, which optimizes the extraction of lipids from red blood cell membranes and selects four different solvent combinations to extract lipids from red blood cell membranes, including the following steps:

[0046] (1) Methanol extraction and redissolution in methanol (denoted as M1);

[0047] (2) after isopropanol extraction, the product was redissolved in a mixture of isopropanol:acetonitrile:water (4:3:1, volume ratio) (denoted as M2);

[0048] (3) extraction with chloroform:water (2:1, volume ratio) and then redissolved in a mixture of isopropanol:acetonitrile:water (4:3:1, volume ratio) (denoted as M3);

[0049] (4) After extraction with methanol:acetonitrile (99:1, volume ratio), the product was redissolved in a mixture of chloroform:methanol (1:1, volume ratio) (denoted as M4).

[0050] Ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS) was used to analyze lipids extracted from erythrocyte membranes and screen for the most efficient lipid extraction method. The results showed that the lipid extraction method using isopropanol extraction followed by resolubilization in a mixture of isopropanol:acetonitrile:water (4:3:1, volume ratio) (denoted as M2) provided a more comprehensive analysis of lipid metabolism and yielded the highest lipid counts.

[0051] The present invention selects the M2 method to extract erythrocyte membrane lipids from normal subjects (NS), patients with mild cognitive impairment (aMCI) and patients with Alzheimer's disease (AD), adopts LC-MS method to detect the lipid metabolites, and imports the LC-MS results into Progenesis QI for lipid metabolite identification.

[0052] At the same time, lipidomics analysis was performed on red blood cell membrane lipid metabolites. The overall distribution of each sample and the stability of the analysis process were observed by unsupervised principal component analysis (PCA), and the obvious clustering observed in the red blood cell membrane samples was determined. The results showed that significant changes occurred in lipids in normal subjects, mild cognitive impairment and Alzheimer's disease groups; supervised partial least squares discriminant analysis (PLS-DA) was used to evaluate the overall lipidomics differences between the three groups. The results showed that the three groups were clearly separated in the red blood cell membrane samples, and the lipids had significant differences.

[0053] This study statistically categorized red blood cell membrane lipid metabolites and used big data to analyze their correlation with Alzheimer's disease. The results showed that the detected lipids covered a wide range of lipid classes, all of which had varying degrees of correlation with Alzheimer's disease.

[0054] The present invention performed RaMP-DB database pathway enrichment analysis on differential lipids. The results showed that the top five enriched metabolic pathways included sphingolipid metabolism, G protein-coupled receptor 40 (GPR40) pathway, BDNF-TrkB signaling pathway and G protein signaling pathway. Studies have shown that most pathways have a significant correlation with neurological diseases.

[0055] The present invention conducts correlation analysis on key clinical blood test parameters closely related to differential lipids and Alzheimer's disease. The results show that 68 lipids are closely related to clinical practice. After further screening, DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) are closely related to clinical parameters closely related to AD. At the same time, we also further performed ROC curve analysis on the differential lipid data of normal subjects and Alzheimer's patients to further confirm the screened differential lipids DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS The feasibility of 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC (22:0) and FAHFA (16:0 / 7-O-18:0) as biological target combinations for AD.

[0056] Beneficial effects:

[0057] (1) The present invention selects four different solvent combinations to extract lipids from erythrocyte membranes. It is found that the lipid extraction method of extracting with isopropanol and then dissolving in a mixture of isopropanol:acetonitrile:water (4:3:1, volume ratio) (denoted as M2) can more comprehensively detect lipid metabolites, and based on this extraction method, the 15 markers described in the present invention can be completely extracted. On this basis, the present invention uses the M2 method to extract erythrocyte membrane lipids from clinical samples of normal subjects, patients with mild cognitive impairment, and patients with Alzheimer's disease, and detects lipids by ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS). The results of UPLC-MS are imported into Progenesis QI for lipid metabolite identification.

[0058] (2) The present invention proposes a comprehensive lipidomics method to perform lipidomics analysis on erythrocyte membrane lipid metabolites. By analyzing the lipid profiles in erythrocyte membrane samples of subjects throughout the entire disease progression, including normal subjects, patients with mild cognitive impairment, and patients with Alzheimer's disease, potential biomarkers of Alzheimer's disease are identified, which makes it possible to diagnose, monitor, and intervene in Alzheimer's disease in the early stage.

[0059] (3) This paper proposes a method for screening differential lipids as clinical targets by combining clinical blood test parameters, further integrating traditional target screening with clinical test results. To identify more clinically relevant lipid targets, we conducted further correlation analysis between differential lipids and lipid parameters associated with clinical AD characteristics. By integrating these analyses with big data, we narrowed down the candidate lipids and ultimately identified DAG 38:2, TAG (14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1(11Z) / 14:0), PI (18:2(9Z,12Z) / 18:1(9Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC (24:1(15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0) as a potential biomarker combination for tracking AD onset. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 The amount of lipids detected from erythrocyte membrane samples by four different sample preparation methods;

[0061] Figure 2 TIC showing lipids separated from erythrocyte membrane samples using the M2 sample preparation method;

[0062] Figure 3 PCA and PLS-DA score plots of erythrocyte membrane;

[0063] Figure 4 Histogram of lipid classification detected in red blood cell membranes (5A); Histogram of the number of lipid reports associated with Alzheimer's disease (5B);

[0064] Figure 5 Histogram of differential lipid enrichment in RaMP-DB;

[0065] Figure 6 Heat map of the correlation between differential lipids and clinical parameters;

[0066] Figure 7Normalized abundance of DAG 38:2 (7A); normalized abundance of TAG (14:0 / 24:1 (15Z) / 20:0) (7B); normalized abundance of TAG 58:2 (7C); normalized abundance of TAG 55:5 (7D); normalized abundance of TAG 53:1 (7E); normalized abundance of TAG 49:3 (7F); normalized abundance of TAG 45:2 (7G); Normalized abundance of p43:1 (7H); normalized abundance of PS42:8 (7I); normalized abundance of PI (22:1 (11Z) / 14:0) (7J); normalized abundance of PI (18:2 (9Z, 12Z) / 18:1 (9Z)) (7K); normalized abundance of PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0) (7L); normalized abundance of LPC (24:1 (15Z)) (7M); normalized abundance of LPC 22:0 (7N) and normalized abundance of FAHFA (16:0 / 7-O-18:0) (7O);

[0067] Figure 8 Performance of the final model when applied to the comparisons of NS versus AD (AUC, 1) and NS versus aMCI (AUC, 0.996), based on the full population for each group.

[0068] Figure 9 When applied to the comparison between NS and AD, ROC curves of DAG 38:2, TAG (14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1(11Z) / 14:0), PI (18:2(9Z,12Z) / 18:1(9Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC (24:1(15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) were presented based on the complete population of each group.

[0069] Figure 10When applied to the comparison between NS and aMCI, ROC curves of DAG 38:2, TAG (14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1(11Z) / 14:0), PI (18:2(9Z,12Z) / 18:1(9Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC (24:1(15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0) were presented based on the complete population of each group. DETAILED DESCRIPTION

[0070] The present invention is further explained below with reference to the following examples, but the examples do not limit the present invention in any form.

[0071] Experimental methods in the examples of this disclosure that do not specify specific conditions are generally based on conventional conditions or the conditions recommended by the raw material or product manufacturers. Reagents without specific sources are conventional reagents purchased from the market.

[0072] Example 1 Clinical sample collection and red blood cell membrane preparation

[0073] A total of 156 blood samples were collected from the Nanjing Medical University Geriatric Hospital and stored at -80°C prior to analysis. These samples included 63 normal subjects, 51 patients with mild cognitive impairment, and 42 patients with Alzheimer's disease (Table 1). The blood samples were mixed with an anticoagulant and centrifuged at 3000 rpm for 20 minutes. The lower layer contained erythrocytes (RBCs). The RBCs (200 μl) were mixed with 600 μl of pre-chilled isotonic phosphate buffer (pH 7.4), gently stirred, and centrifuged at 5000 rpm for 15 minutes at 4°C. The supernatant and surface sediment were removed, and this was repeated three times. Pre-chilled hypotonic Tris hydrochloride (10 mmol / L, pH 7.4) was added to the prepared RBC samples at a ratio of 1:40 (RBCs:Tris hydrochloride). The samples were gently stirred and incubated at 4°C for 2 hours until hemolysis was complete. Centrifuge at 4°C, 9000 rpm for 15 min to precipitate the erythrocyte membranes. Repeat this process three times to obtain the erythrocyte membrane sample.

[0074] Table 1

[0075]

[0076] Data are expressed as mean ± SEM values.

[0077] Example 2 Optimization of lipid extraction from erythrocyte membranes

[0078] In order to obtain a comprehensive and detailed lipid profile, four different sample pretreatment methods were selected to extract lipids from the erythrocyte membrane obtained in Example 1, including:

[0079] After methanol extraction, the sample was redissolved in methanol (denoted as M1); the specific operation was as follows: 100ul of methanol extract (pre-cooled on ice) was mixed with the red blood cell membrane sample, allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then ultrasonicated on ice for 10 minutes. The obtained sample mixture was centrifuged at 4°C, 12000rpm for 10 minutes, the supernatant was evaporated to complete the red blood cell membrane extraction, and the sample was redissolved in methanol for injection.

[0080] After isopropanol extraction, the sample was redissolved in a mixture of isopropanol:acetonitrile:water (4:3:1, volume ratio) (denoted as M2); the specific operation was as follows: 100ul of the extract isopropanol (pre-cooled on ice) was mixed with the red blood cell membrane sample, and the mixture was allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then sonicated on ice for 10 minutes. The obtained sample mixture was centrifuged at 4°C, 12000rpm for 10 minutes, the supernatant was evaporated to complete the red blood cell membrane extraction, and the sample was redissolved in isopropanol:acetonitrile:water (4:3:1) and injected.

[0081] After extraction with chloroform: water (2:1, volume ratio), the sample was redissolved in a mixture of isopropanol: acetonitrile: water (4:3:1, volume ratio) (denoted as M3); the specific operation was as follows: 100ul of the extract chloroform: water (pre-cooled on ice) was mixed with the red blood cell membrane sample, and the mixture was allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then sonicated on ice for 10 minutes. The obtained sample mixture was centrifuged at 4°C, 12000rpm for 10 minutes, the supernatant was evaporated to complete the red blood cell membrane extraction, and the sample was redissolved in isopropanol: acetonitrile: water (4:3:1) and injected.

[0082] After extraction with methanol:acetonitrile (99:1, volume ratio), the sample was redissolved in a mixture of chloroform:methanol (1:1, volume ratio) (denoted as M4). The specific operation was as follows: 100 μl of the extract (99:1, volume ratio) (pre-chilled on ice) was mixed with the red blood cell membrane sample, allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then sonicated on ice for 10 minutes. The resulting sample mixture was centrifuged at 4°C, 12,000 rpm for 10 minutes, and the supernatant was evaporated to dryness to complete the red blood cell membrane extraction. The sample was then redissolved in chloroform:methanol (1:1, volume ratio) and injected.

[0083] A Waters ACQUITY I-Class UPLC liquid phase system (Milford, MA, USA) was used in conjunction with a Waters Xevo G2-XS Q-Tof mass spectrometer (Milford, MA, USA).

[0084] Chromatographic separation was performed on a Waters ACQUITY UPLC CSH C18 column (1.7 μm, 100 mm × 2.1 mm). The column temperature was 40°C, the autosampler was 8°C, and the flow rate was 0.35 mL / min. Mobile phase A consisted of acetonitrile:water (60:40) containing 10 mM ammonium formate and 0.1% formic acid, and mobile phase B consisted of isopropanol:acetonitrile (90:10) containing 10 mM ammonium formate and 0.1% formic acid.

[0085] Gradient elution conditions: 0-2.0 min, 40% B; 2.0-2.1 min, 45%-50% B; 2.1-11.0 min, 50-54% B; 11.0-11.1 min, 54%-70% B; 11.1-15.0 min, 70%-99% B; 15.0-18.0 min, returned to 40% B.

[0086] DDA detection and data acquisition were performed in positive and negative ion modes with the following parameters: m / z range 50–1000 Da; capillary voltage +3.0 kV / −2.5 kV for positive / negative ion modes, respectively; sampling cone voltage 40 V; ion source temperature 100°C; desolvation temperature 450°C; cone gas flow rate 50 L / h; and desolvation gas flow rate 600 L / h. MassLynx 4.1 was used for data acquisition.

[0087] The results showed that among M1 to M4, M2 had better lipid extraction efficiency, about 1.55 times that of M1 and M4, and about 2.25 times that of M3 ( Figure 1 A), the number of lipids that can be detected by the M2 method far exceeds that of other lipid extraction methods (M3 / M4). At the same time, M2 covers a variety of lipid categories with excellent chromatographic separation efficiency ( Figure 2 It is worth noting that the M2 method detected 265 lipids in the positive ion mode and 183 lipids in the negative ion mode. Only 33 lipids were detected in both positive and negative ion modes, suggesting the necessity of using different detection ion modes during sample analysis, thereby significantly broadening the scope of lipid detection ( Figure 1 B). Therefore, to obtain a comprehensive and detailed lipid profile, the M2 method was used to extract lipids from erythrocyte membranes and to detect them using mass spectrometry in both positive and negative modes.

[0088] Example 3 Lipidomics Analysis of Red Blood Cell Membrane Samples

[0089] Based on Example 2, we performed lipidomic analysis on erythrocyte membrane samples from normal subjects, patients with mild cognitive impairment, and patients with Alzheimer's disease. The LC-MS results were imported into QI for identification.

[0090] The erythrocyte membrane sample from Example 2 was extracted for lipids using the M2 method. 100 μl of the extract (pre-chilled on ice) in isopropanol was mixed with the erythrocyte membrane sample and allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then sonicated on ice for 10 minutes. The resulting sample mixture was centrifuged at 12,000 rpm for 10 minutes at 4°C. The supernatant was evaporated to dryness and reconstituted in isopropanol:acetonitrile:water (4:3:1) for injection.

[0091] Metaboanalyst software (https: / / www.metaboanalyst.ca / ) was used to perform principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). Unsupervised principal component analysis (PCA) was used to observe the overall distribution of each sample and the stability of the analysis process. The scores of the first two principal components, PC1 and PC2, accounted for 35.4% and 15.6% of the variance, respectively ( Figure 3 A), samples showed a certain degree of clustering, indicating that lipid profiles were significantly altered in the three groups: normal subjects, patients with mild cognitive impairment, and patients with Alzheimer's disease.

[0092] The overall lipidomic differences among the three groups were evaluated by supervised partial least squares discriminant analysis (PLS-DA). The results showed that the lipids detected from the red blood cell membrane samples could distinguish the three groups to a certain extent ( Figure 3 B).

[0093] Comparison of the PLS-DA results with the PCA results highlighted distinct lipid profiles among the three groups. The more pronounced separation in the erythrocyte membrane sample suggests it is a superior matrix for lipid analysis, particularly for Alzheimer's disease monitoring. Erythrocyte membranes are primarily composed of lipids, providing a purer, more concentrated lipid profile. Therefore, using erythrocyte membranes as a lipid source for biological samples for Alzheimer's disease detection has some reliability.

[0094] Example 4 Confirmation of differential lipids

[0095] Based on Examples 1-3, the extracted differential lipids were identified in QI using databases such as the Human Metabolome Database (HMDB) (http: / / www.hmdb.ca / ), MONA (http: / / mona.fiehnlab.ucdavis.edu / ), Lipid Atlas (https: / / www.lipidmaps.org / ), and METLIN (https: / / metlin.scripps.edu / ). Differential lipid screening criteria: VIP value (Variable importance in the projection, VIP) > 1; P value (pvalue) < 0.05; relative abundance ratio (Fold change, FC) between samples > 1.5. Identification parameters were: precursor ion matching amount of 10 ppm; fragment ion matching of 20 ppm, isotope similarity > 80%; precursor ion library score > 40%.

[0096] A total of 101 differential lipids in 11 categories were detected in the red blood cell membrane samples, including DG, TAGs, SMs, PE, PC, LPCs, FAHFAs, PI, PSs, Cers (GlcCer, Cer) ( Figure 4 A), big data analysis shows that they have a strong correlation with AD ( Figure 4 B) Therefore, lipid-based research in Alzheimer's disease is feasible.

[0097] Example 5 Pathway enrichment analysis of differential lipids

[0098] The enrichment analysis of differential lipids was performed using the RaMP-DB database (integrated with KEGG through HMDB, Reactome, and WikiPathways). The top five metabolic pathways enriched included sphingolipid metabolism, G protein-coupled receptor 40 (GPR40) pathway, BDNF-TrkB signaling pathway, and G protein signaling pathway ( Figure 5 ).

[0099] Sphingolipids are members of a broad family of lipids that play a crucial role in biological signal transduction. Numerous studies have shown a close association between sphingolipids and cognitive impairment and neurological diseases. GPR40 is being investigated for its potential role in the pathogenesis and treatment of diseases such as Alzheimer's disease, type 2 diabetes, and dementia. GPR40 agonists have been reported to reduce pathological neuroinflammation in Alzheimer's disease by modulating the gut microbiome and the immune system. Studies have shown that GPR40 is a potential drug target for Alzheimer's disease, type 2 diabetes, and dementia. Furthermore, G protein signaling has been shown to reduce β-amyloid levels, leading to a reduction in amyloid plaques. These results suggest that G proteins may represent a safer therapeutic strategy for the development of G protein-coupled receptors for the treatment of Alzheimer's disease. BDNF and its high-affinity receptor, TrkB, are abundantly expressed in the cortex and hippocampus and play an important role in learning and memory. Abnormal changes in the expression and function of the BDNF / TrkB signaling pathway may be associated with the pathogenesis of neurodegenerative diseases such as Alzheimer's disease. Most pathways associated with lipid targets in the samples have been shown to be significantly associated with neurological diseases, indicating that the lipid targets we studied are accurate and practical, which can help enhance our understanding of their role in Alzheimer's disease and potentially discover new avenues for therapeutic intervention.

[0100] Example 6 Discovery of potential biomarkers

[0101] Based on Example 5, in order to improve the accuracy of differential lipid confirmation related to clinical practice, we performed correlation analysis on lipids and key clinical blood test clinical parameters related to Alzheimer's disease, including triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), lipoprotein (a) [LP (a)], low shear whole blood viscosity (BVL), medium shear whole blood viscosity (BVM), high shear whole blood viscosity (BVH). The correlation between differential lipids and clinical parameters was analyzed, and it was found that among these 101 differential lipids, 68 lipids were closely related to clinical practice ( Figure 6 38 lipids were strongly correlated with TC and TG, 32 lipids were significantly correlated with LDL and HDL, and 27 lipids were closely correlated with BVL, BVM, and BVH.

[0102] On this basis, we further screened the differential metabolites by the AUC value of the ROC curve to obtain more representative biomarkers that can distinguish the NS and AD groups and the NC and aMCI groups. With AUC>90 as the screening cutoff value, the 68 differential lipids closely related to clinical blood test parameters were further screened to determine DAG 38:2 (NS-AD:0.908; NS-aMCI:0.924), TAG (14:0 / 24:1(15Z) / 20:0) (NS-AD:0.964; NS-aMCI:0.982), TAG 58:2 (NS-AD:0.963; NS-aMCI:0.967), TAG 55:5 (NS-AD:0.962; NS-aMCI:0.966), TAG 53:1 (NS-AD:0.942; NS-aMCI:0.932), TAG 49:3 (NS-AD:0.928; NS-aMCI:0.941), and TAG 45:2(NS-AD:0.955; NS-aMCI:1), TAG 43:1(NS-AD:0.942; NS-aMCI:1), PS 42:8(NS-AD:0.926; NS-aMCI:0.916), PI(22:1(11Z) / 14:0)(NS-AD:0.946; NS-aMCI:0.994), PI(18 :2(9Z,12Z) / 18:1(9Z))(NS-AD:0.904; NS-aMCI:0.949), PC(22:5(7Z,10Z,13Z,16Z,19Z) / 18:0)(NS -AD:0.903; NS-aMCI:0.949), LPC (24:1 (15Z)) (NS-AD:0.977; NS-aMCI:0.992), LPC22:0 (NS-AD:0.930; NS-aMCI:0.958) and FAHFA (16:0 / 7-O-18:0) (NS-AD:0.958; NS-aMCI:0.979) as a potential biomarker combination for tracking the onset of AD. The expression levels of these 15 lipids gradually decreased with the aggravation of AD ( Figure 7 These results also emphasize the importance of AD biomarker discovery and the innovation and necessity of using RCM as a detection matrix.

[0103] Example 7 Diagnostic Validation

[0104] At the same time, to further confirm whether the selected targets are representative, we performed ROC analysis and measured the overall performance of the model by the area under the ROC curve.

[0105] The experimental process is as follows:

[0106] (1)DAG 38:2, TAG(14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG49:3, TAG 45:2, TAG 43:1, PS 42:8, PI(22:1(11Z) / 14:0), PI(18:2(9Z,12Z) / 18:1(9Z)), PC(22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC(24:1(15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) separate validation group: 120 blood samples were obtained from Nanjing Medical University Geriatric Hospital, different from those in Example 1, and stored at -80°C before analysis. Among them, 40 were normal subjects, 40 were patients with mild cognitive impairment, and 40 were patients with Alzheimer's disease.

[0107] The blood sample was mixed with an anticoagulant and centrifuged at 3000 rpm for 20 minutes. The lower layer was the erythrocytes. The resulting erythrocytes (200 μl) were mixed with 600 μl of pre-chilled isotonic phosphate buffer (pH 7.4), gently stirred, and centrifuged at 5000 rpm for 15 minutes at 4°C. The supernatant and surface sediment were removed. This was repeated three times. Pre-chilled hypotonic Tris hydrochloride (10 mmol / L, pH 7.4) was added to the prepared erythrocyte sample at a ratio of 1:40 (erythrocytes:Tris hydrochloride). The mixture was gently stirred and incubated at 4°C for 2 hours until hemolysis was complete. The erythrocyte membranes were pelleted by centrifugation at 9000 rpm for 15 minutes at 4°C. This was repeated three times to obtain the erythrocyte membrane sample.

[0108] To obtain a comprehensive and detailed lipid profile, 100 μl of isopropanol extract (pre-cooled on ice) was mixed with the red blood cell membrane sample, allowed to stand at 4°C for 20 minutes, shaken for 2 minutes, and then sonicated on ice for 10 minutes. The resulting sample mixture was centrifuged at 4°C, 12,000 rpm for 10 minutes, the supernatant was evaporated, and reconstituted in isopropanol:acetonitrile:water (4:3:1) for injection.

[0109] A Waters ACQUITY I-Class UPLC liquid phase system (Milford, MA, USA) was used in conjunction with a Waters Xevo G2-XS Q-Tof mass spectrometer (Milford, MA, USA).

[0110] Chromatographic separation was performed on a Waters ACQUITY UPLC CSH C18 column (1.7 μm, 100 mm × 2.1 mm). The column temperature was 40°C, the autosampler was 8°C, and the flow rate was 0.35 mL / min. Mobile phase A consisted of acetonitrile:water (60:40) containing 10 mM ammonium formate and 0.1% formic acid, and mobile phase B consisted of isopropanol:acetonitrile (90:10) containing 10 mM ammonium formate and 0.1% formic acid. Gradient elution conditions were: 40% B (0-2.0 min); 45%-50% B (2.0-2.1 min); 50%-54% B (2.1-11.0 min); 54%-70% B (11.0-11.1 min); 70%-99% B (11.1-15.0 min); and 40% B (15.0-18.0 min).

[0111] DDA detection and data acquisition were performed in positive and negative ion modes with the following parameters: m / z range 50–1000 Da; capillary voltage +3.0 kV / −2.5 kV for positive / negative ion modes, respectively; sampling cone voltage 40 V; ion source temperature 100°C; desolvation temperature 450°C; cone gas flow rate 50 L / h; and desolvation gas flow rate 600 L / h. MassLynx 4.1 was used for data acquisition.

[0112] After data collection, data analysis was performed on the NS group and AD group, and the NS group and aMCI group, respectively, and the area under the ROC curve was used to measure whether the target was representative.

[0113] (2) DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) combination validation group: The experimental process is the same as the above experimental process (1).

[0114] The results show:

[0115] (1) ROC analysis of DAG 38:2: DAG 38:2 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.908, with a 95% confidence interval of 0.8310.966 ( Figure 9A); The AUC value of the diagnostic model NS-aMCI was 0.863, with a 95% confidence interval of 0.7620.942 ( Figure 10 A).

[0116] (2) ROC analysis of TAG (14:0 / 24:1 (15Z) / 20:0): TAG (14:0 / 24:1 (15Z) / 20:0) was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.983, with a 95% confidence interval of 0.945-0.999 ( Figure 9 B); The AUC value of the diagnostic model NS-aMCI was 0.95, with a 95% confidence interval of 0.889-0.99 ( Figure 10 B).

[0117] (3) ROC analysis of TAG 58:2: TAG 58:2 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.95, with a 95% confidence interval of 0.889-0.991 ( Figure 9 C); the AUC value of the diagnostic model NS-aMCI was 0.944, with a 95% confidence interval of 0.867-0.998 ( Figure 10 C).

[0118] (4) ROC analysis of TAG 55:5: TAG 55:5 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.959, with a 95% confidence interval of 0.886-1 ( Figure 9 D); the AUC value of the diagnostic model NS-aMCI was 0.943, 95% confidence interval: 0.878-0.986 ( Figure 10 D).

[0119] (5) ROC analysis of TAG 53:1: TAG 53:1 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.926, with a 95% confidence interval of 0.838-0.988 ( Figure 9 E); the AUC value of the diagnostic model NS-aMCI was 0.908, with a 95% confidence interval of 0.83-0.968 ( Figure 10 E).

[0120] (6) ROC analysis of TAG 49:3: TAG 49:3 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.94, with a 95% confidence interval of 0.86-1 ( Figure 9 F); The AUC value of the diagnostic model NS-aMCI was 0.917, with a 95% confidence interval of 0.844-0.974 ( Figure 10 F).

[0121] (7) ROC analysis of TAG 45:2: TAG 45:2 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 1, with a 95% confidence interval of 1-1 ( Figure 9 G); the AUC value of the diagnostic model NS-aMCI was 0.956, with a 95% confidence interval of 0.885-1 ( Figure 10 G).

[0122] (8) ROC analysis of TAG 43:1: TAG 43:1 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 1, with a 95% confidence interval of 1-1 ( Figure 9 H); the AUC value of the diagnostic model NS-aMCI was 0.953, with a 95% confidence interval of 0.89-0.993 ( Figure 10 H).

[0123] (9) ROC analysis of PS 42:8: PS 42:8 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.902, with a 95% confidence interval of 0.829-0.952 ( Figure 9 I); the AUC value of the diagnostic model NS-aMCI was 0.913, 95% confidence interval: 0.842-0.962 ( Figure 10 I).

[0124] (10) ROC analysis of PI (22:1 (11Z) / 14:0): PI (22:1 (11Z) / 14:0) was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.99, with a 95% confidence interval of 0.972-1 ( Figure 9 J); The AUC value of the diagnostic model NS-aMCI was 0.908, with a 95% confidence interval of 0.836-0.955 ( Figure 10 J).

[0125] (11) ROC analysis of PI (18:2 (9Z, 12Z) / 18:1 (9Z)): PI (18:2 (9Z, 12Z) / 18:1 (9Z)) was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.939, with a 95% confidence interval of 0.882-0.979 ( Figure 9 K); the AUC value of the diagnostic model NS-aMCI was 0.877, with a 95% confidence interval of 0.8-0.946 ( Figure 10 K).

[0126] (12) ROC analysis of PC(22:5(7Z,10Z,13Z,16Z,19Z) / 18:0):

[0127] PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0) as a biomarker for Alzheimer's disease, after binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.932, with a 95% confidence interval of 0.871-0.981 ( Figure 9 L); the AUC value of the diagnostic model NS-aMCI was 0.873, 95% confidence interval: 0.768-0.943 ( Figure 10 L).

[0128] (13) ROC analysis of LPC (24:1 (15Z)): LPC (24:1 (15Z)) was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.989, with a 95% confidence interval of 0.964-1 ( Figure 9 M); the AUC value of the diagnostic model NS-aMCI was 0.966, with a 95% confidence interval of 0.923-0.992 ( Figure 10 M).

[0129] (14) ROC analysis of LPC 22:0: LPC 22:0 was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 0.954, with a 95% confidence interval of 0.895-0.991 ( Figure 9 N); the AUC value of the diagnostic model NS-aMCI was 0.927, with a 95% confidence interval of 0.855-0.976 ( Figure 10 N).

[0130] (15) ROC analysis of FAHFA (16:0 / 7-O-18:0): FAHFA (16:0 / 7-O-18:0) was selected as a biomarker for Alzheimer's disease. After binary logistic regression and ROC analysis, the AUC value of the diagnostic model NS-AD was 1, with a 95% confidence interval of 1-1 ( Figure 9 O); the AUC value of the diagnostic model NS-aMCI was 0.999, with a 95% confidence interval of 0.994-1 ( Figure 10 O).

[0131] (16) ROC analysis of DAG 38:2, TAG (14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1(11Z) / 14:0), PI (18:2(9Z,12Z) / 18:1(9Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC (24:1(15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0):

[0132] The combined validation model of DAG 38:2, TAG (14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG45:2, TAG 43:1, PS 42:8, PI (22:1(11Z) / 14:0), PI (18:2(9Z,12Z) / 18:1(9Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z) / 18:0), LPC (24:1(15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) accurately distinguished AD and NS patients (AUC, 1).

[0133] Importantly, the model also discriminated between patients with aMCI and NS (AUC, 0.999), indicating its potential diagnostic utility in early disease stages ( Figure 8 ).

[0134] These findings describe a sensitive biomarker panel that may aid in the specific detection of early AD by analyzing plasma samples.

[0135] Therefore, the final selections are DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0) are potential indicators for tracking the progression of AD and biomarkers for early diagnosis of AD. These 15 lipids can be used alone or in combination as diagnostic markers or auxiliary diagnostic markers for AD. These results also emphasize the importance of AD biomarker discovery and the innovation and necessity of using red blood cell membranes as sample sources.

[0136] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. Use of lipids as markers in the preparation of products for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, characterized in that: The lipids are DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), A combination of one or more of PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0).

2. Use of a substance for extracting lipids or a substance for detecting lipid expression levels in the preparation of a product for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, characterized in that: The lipid is DAG 38:2, TAG(14:0 / 24:1(15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG43:1, PS 42:8, PI(22:1(11Z) / 14:0), PI(18:2(9Z,12Z) / 18:1(9Z)), A combination of one or more of PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0 and FAHFA (16:0 / 7-O-18:0).

3. The use according to claim 2, characterized in that The substance used for extracting lipids is a combination of isopropanol and a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:

1.

4. The use according to claim 1 or 2, characterized in that The lipids are derived from the red blood cell membranes of the patient to be tested.

5. The use according to claim 1 or 2, characterized in that: The decreased expression level of the lipids is closely related to the occurrence of Alzheimer's disease and / or mild cognitive impairment.

6. The use according to claim 1 or 2, characterized in that The application The following steps are involved: S1: Extract the red blood cell membrane of the patient to be tested; S2: The erythrocyte membrane obtained in S1 was extracted with isopropanol and then redissolved in a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:1 to obtain the lipid sample to be tested; S3: Detecting the expression level of the lipid in the lipid sample to be tested.

7. The use according to claim 6, characterized in that Specifically, S2 is prepared by mixing isopropanol with the red blood cell membrane obtained in S1, allowing the mixture to stand at 4°C for 20 minutes, shaking for 2 minutes, and then sonicating on ice for 10 minutes. The obtained sample mixture is centrifuged at 4°C and 12,000 rpm for 10 minutes, the supernatant is evaporated, and redissolved in a mixture of isopropanol:acetonitrile:water in a volume ratio of 4:3:

1.

8. The use according to claim 6, characterized in that The S3 is detected by ultra-high performance liquid chromatography tandem mass spectrometry, and the liquid chromatography conditions are: A Waters ACQUITY UPLC CSH C18 column (1.7 μm, 100 mm × 2.1 mm) was used; Column temperature 40°C; Autosampler 8°C; Flow rate 0.35 mL / min; Mobile phase A was acetonitrile:water (60:40) containing 10 mM ammonium formate and 0.1% formic acid, and mobile phase B was isopropanol:acetonitrile (90:10) containing 10 mM ammonium formate and 0.1% formic acid; Gradient elution conditions: 0-2.0 min, 40% B; 2.0-2.1min, 45%-50%B; 2.1-11.0min, 50-54%B; 11.0-11.1min, 54%-70%B; 11.1-15.0min, 70%-99%B; 15.0-18.0 min, recovered to 40% B; Mass spectrometry conditions are: DDA detection and data acquisition were performed in positive and negative ion modes with the following parameters: m / z range 50–1000 Da; capillary voltage +3.0 kV / −2.5 kV for positive / negative ion modes, respectively; sampling cone voltage 40 V; ion source temperature 100 °C; desolvation temperature 450 °C; cone gas flow rate 50 L / h; and desolvation gas flow rate 600 L / h.

9. A kit for diagnosing or assisting in the diagnosis of Alzheimer's disease and / or mild cognitive impairment, characterized in that: The kit comprises the substance for extracting lipids or the substance for detecting the expression level of lipids as described in claim 2 or 3, wherein the lipids are a combination of one or more of DAG 38:2, TAG (14:0 / 24:1 (15Z) / 20:0), TAG 58:2, TAG 55:5, TAG 53:1, TAG 49:3, TAG 45:2, TAG 43:1, PS 42:8, PI (22:1 (11Z) / 14:0), PI (18:2 (9Z, 12Z) / 18:1 (9Z)), PC (22:5 (7Z, 10Z, 13Z, 16Z, 19Z) / 18:0), LPC (24:1 (15Z)), LPC 22:0, and FAHFA (16:0 / 7-O-18:0).

Images