Hair metabolic marker combination for diagnosing Alzheimer's disease and its application

Abstract

The present invention discloses a hair metabolite marker combination for diagnosing Alzheimer's disease and its application. The hair metabolite marker combination includes triacontanes, erucic acid, heneicosane, behenic acid, tetracosane, 3-hydroxybutyric acid, 15-methylhexadecanoic acid, arachidic acid, heptadecanoic acid, melatonin, stearic acid, tetracosanoic acid, heneicosanoic acid, and docosane. The present invention uses metabolomics methods to study, compare, and analyze the hair metabolites of AD patients, MCI patients, and cognitively normal subjects, explores potential biomarkers that may be contained therein, obtains a group of hair metabolite differences among AD patients, and evaluates the sensitivity and specificity of this group of substances for diagnosing AD. The present invention provides a new marker for the accurate diagnosis of AD and can be used in clinical auxiliary diagnosis of AD to improve diagnostic accuracy.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a hair metabolic marker combination for diagnosing Alzheimer's disease and an application thereof. Background Art

[0002] Alzheimer's disease (AD) is the most common type of dementia, accounting for approximately 60%-80% of all dementias, and is characterized by a decline in memory, language, orientation, visual-spatial, executive function, and daily activities. AD is an irreversible and gradually progressive disease with a long latency period before symptoms appear, diverse clinical symptoms, and insidious onset. It is easily overlooked by patients and clinicians in the early stages of the disease. Mild cognitive impairment (MCI), as an intermediate stage in the AD disease continuum, is often difficult to distinguish from cognitive decline caused by normal aging, and the annual conversion rate of AD is 5-17% [1]. Studies have shown that pathological changes have already begun to appear in the bodies of AD patients 20 years before the onset of clinical manifestations. Therefore, early diagnosis of AD is of great significance in delaying disease progression, reducing morbidity, and maintaining the early stage of the disease.

[0003] Currently, many diagnostic methods for AD are widely used, such as magnetic resonance imaging (MRI), functional magnetic resonance imaging (fMRI), positron emission tomography (PET), cerebrospinal fluid markers, and blood genetic analysis. However, these diagnostic methods each have drawbacks, such as being time-consuming, expensive, invasive, and unable to diagnose early. In recent years, there has been a surge in research exploring new diagnostic tools for AD, aiming to develop early, sensitive, and convenient methods for diagnosing and assessing prognosis, thereby helping physicians diagnose and treat patients promptly and providing clinical decision support and advice.

[0004] Metabolomics, as a new research technology, has become a hot topic in recent years and is widely used in the search for biomarkers for various diseases. It helps to study the pathogenesis, metabolic pathways, and new treatments of various diseases. Metabolomics focuses on endogenous metabolites with a molecular weight below 1000KDa in organisms. By revealing the changing trajectories of metabolite types and quantities under the influence of internal and external factors, it reflects the series of biological events occurring in a certain pathophysiological process and provides biological terminal information. These metabolites are derived from metabolic pathways such as sugars, fats, proteins, nucleic acids, bile acids, and commensal bacteria. In recent years, many scholars have used metabolomics methods to conduct metabolomics analysis of blood, saliva, brain tissue, cerebrospinal fluid, etc. to diagnose AD early and predict its progression, thereby gaining a deeper understanding of the disease.

[0005] Currently, the most commonly used separation methods for metabolism include gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), and nuclear magnetic resonance spectroscopy (NMR). NMR is more accurate than MS in identifying compound structures (e.g., isomers) and is more suitable for qualitative studies of objects. GC-MS is used to separate and analyze volatile components, which presupposes that they are volatile at high temperatures and have good thermal stability. Compared to LC-MS, GC-MS offers advantages such as high separation efficiency, high sensitivity, reduced sample usage, rapid analysis, and a large amount of information. Summary of the Invention

[0006] The purpose of the present invention is to address the above-mentioned problems and provide a hair metabolic marker combination for diagnosing Alzheimer's disease and its application.

[0007] In order to achieve its purpose, the present invention adopts the following technical solutions:

[0008] A first aspect of the present invention provides a hair metabolism marker combination for diagnosing Alzheimer's disease, wherein the hair metabolism marker combination includes triacontan, erucic acid, heneicosane, behenic acid, tetracosane, 3-hydroxybutyric acid, 15-methylhexadecanoic acid, arachidic acid, heptadecanoic acid, melatonin, stearic acid, tetracosanoic acid, heneicosanoic acid, and docosane.

[0009] The hair metabolic marker combination is used to diagnose and differentiate Alzheimer's disease patients, mild cognitive impairment patients and people with normal cognition.

[0010] Compared with cognitively normal subjects, in patients with Alzheimer's disease, melatonin, triacontanestheate, tetracosanoic acid, behenic acid, docosane, erucic acid, heneicosane, stearic acid, tetracosane, heptadecanoic acid, arachidic acid, and heneicosanoic acid were significantly upregulated, while 3-hydroxybutyrate was significantly downregulated.

[0011] A second aspect of the present invention provides use of the above-mentioned hair metabolic marker combination in the preparation of a kit for diagnosing Alzheimer's disease.

[0012] A third aspect of the present invention provides a kit for diagnosing Alzheimer's disease, comprising the above-mentioned standard product of the hair metabolic marker combination for diagnosing Alzheimer's disease.

[0013] The kit further comprises an extraction reagent for extracting the hair metabolism markers.

[0014] The extraction reagents include acetone, sodium hydroxide and methanol.

[0015] The kit also includes an internal standard D4-alanine.

[0016] The kit also includes reagents for gas chromatography-mass spectrometry analysis.

[0017] The present invention has the beneficial effect of using cutting-edge metabolomics methods to study, compare, and analyze the hair metabolites of AD patients, MCI patients (mild cognitive impairment, a pre-stage of AD), and normal controls (NC). This research explores potential biomarkers, resulting in a panel of differentially expressed hair metabolites in AD patients and assessing the sensitivity and specificity of this panel for AD diagnosis. This invention provides a new marker for the accurate diagnosis of AD, which can be used in clinical practice to aid in the diagnosis of AD and improve diagnostic accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Shown are: (A) PCA results of AD, MCI and NC groups; (B) PCA results grouped according to Clinical Dementia Rating (CDR) results; (C) OPLS-DA score graphs of MCI and NC groups; (D) OPLS-DA score graphs of AD and NC groups; (E) Volcano plot of metabolites in cognitive impairment group (MCI+AD) and NC group; (F) Line graph of fold changes of key metabolites between NC group and cognitive impairment group.

[0019] Figure 2 Volcano plot (AB) and line graph (CD) of metabolic changes among Alzheimer's disease (AD), mild cognitive impairment (MCI), and normal control (NC).

[0020] Figure 3 This is the Venn diagram of the common and unique differential metabolites found in the comparison between the AD group and the NC group, the MCI group and the NC group, and the CI group and the NC group.

[0021] Figure 4 LASSO logistic regression was used to construct the classification model. The differential metabolite set was used to construct the classification model results for NC vs AD and NC vs MCI respectively: (A, B) Lasso coefficient curves of differential metabolites with L1 regularization; (C, D) 10-fold cross-validation was used to select the optimal lambda (λ) value based on binomial deviation; (E, F) ROC curves of the classification results. DETAILED DESCRIPTION

[0022] The present invention will be further described below with reference to the embodiments, but the present invention is not limited thereto.

[0023] The experimental methods in the following examples are conventional methods unless otherwise specified.

[0024] Example 1

[0025] 1 Experimental methods

[0026] 1.1 Neuropsychological Scale Assessment

[0027] Participants were recruited from the Memory Disorders Clinic, Department of Geriatrics, Chongqing Medical University. All 213 participants and one of their reliable caregivers underwent standardized clinical interviews and neuropsychological assessments by trained nurses or experienced physicians for grouping. Complete clinical memory scale assessment data (MMSE, ADAS-Cog, Clock, AVLT, BNT, FCSRT, NPI, CDR, etc.) were obtained. The participants were divided into groups based on the Dementia Rating Scale-Global Score (CDR-GS): 35 with a score of 0, 72 with a score of 0.5, 40 with a score of 1, 35 with a score of 2, and 29 with a score of 3. The participants were also divided into groups based on diagnosis: 47 normal controls (NC), 40 with mild cognitive impairment (MCI), and 126 with Alzheimer's disease (AD).

[0028] Inclusion Criteria: 1. Normal (NC): No diagnosis of cognitive impairment confirmed by clinical evaluation and cognitive assessment (e.g., MMSE score >28). 2. Mild Cognitive Impairment (MCI): Patients diagnosed with mild cognitive impairment according to the Petersen criteria, including subjective memory decline, objective memory impairment, and normal general cognitive function with no significant impairment in daily activities. 3. Alzheimer's Disease (AD): Meets the NIA-AA diagnostic criteria published in 2011 by the National Institute on Aging and the Alzheimer's Association. 4. Age ≥60 years and ≤85 years, voluntary participation, and informed consent signed by the patient or by a direct relative.

[0029] Exclusion criteria: 1. Patients with other types of dementia, such as vascular dementia; 2. Patients with any existing acute physical illness; 3. Patients with a history of illegal drug use, substance abuse or alcoholism; 4. Patients or their families refused to join the study.

[0030] 1.2 Sampling and storage

[0031] Hair samples were cut 1.0 cm from the scalp, wrapped in aluminum foil, and stored at −80°C. Specimen IDs were labeled with a digital code to ensure that patient personal information was not associated with the samples. All hair samples underwent only one freeze-thaw cycle before analysis. After collection, 5.00 ± 0.50 mg of hair samples were weighed and placed in 2-ml tubes. The samples were washed with 2 ml of acetone, 1 ml of water, and 2 ml of acetone for 10 minutes, and then air-dried in a fume hood at room temperature. After drying, 400 μL of 1 M sodium hydroxide and 4 μL of internal control (10 mM D4-alanine) were added, and the tubes were incubated at 85°C for 25 minutes. The samples were then allowed to cool naturally to room temperature. After cooling, 400 μL of methanol was added, the tubes were vortexed for 30 seconds, and centrifuged at 4000 g for 5 minutes. 300 μL of the supernatant was collected for derivatization. Quality control (QC) samples were prepared by taking a certain amount of supernatant from each sample, mixing a total of 20 ml into a new tube, and then performing MCF derivatization as described above to obtain 15 QC samples. After every 15 samples, a QC sample was loaded to obtain a QC spectrum.

[0032] 1.3 Methyl chloroformate (MCF) derivatization and gas chromatography-mass spectrometry analysis

[0033] According to Nature Protocols [Smart KF, Aggio RB, Van Houtte JR, All prepared extracts were chemically modified using MCF derivatization to lower their boiling points according to the method described in SG. (2010) Analytical platform for metabolome analysis of microbial cells using methyl chloroformate derivatization followed by gas chromatography-mass spectrometry. Nat Protoc. 5: 1709-29.10.1038 / nprot.2010.108].

[0034] Derivatized metabolites were analyzed using an Agilent Intuvo 9000 gas chromatography system coupled with a 5977B series mass selective detector (MSD). The GC column was a BD-1701 (30 m × 250 μm ID × 0.25 μm, Agilent). Derivatized samples were injected into the splitless inlet at 290°C with a helium flow rate of 1 mL / min. The first stage of the GC-oven program was held at 45°C for 2 minutes. The second stage involved increasing the temperature from 45°C to 180°C at a rate of 9°C / min and holding for 5 minutes. The third stage involved increasing the temperature to 220°C at a rate of 40°C / min and holding for 5 minutes. The fourth stage involved increasing the temperature to 240°C at a rate of 40°C / min and holding for 11.5 minutes. Finally, the temperature was increased to 280°C at a rate of 80°C / min. The guard chip, auxiliary components, quadrupole mass spectrometer, and mass spectrometer source temperatures were 280°C, 250°C, 230°C, and 150°C, respectively. The mass range was 30 μm to 550 μm, the scanning speed was set to 1.563 μ / s, and the solvent delay time was 5.5 min.

[0035] 1.3 Metabolite identification and quantification

[0036] Chromatographic features were processed and identified using Automated Mass Spectral Deconvolution and Identification System (AMDIS) software. Metabolites were confirmed by matching spectra with an in-house MFC library, with a similarity score exceeding 85%, and ensuring that their GC retention times were within 30 seconds. Other potential compounds were identified using the commercially available NIST mass spectral library. Relative concentrations of metabolites were extracted using a Mass Omics-R-based script by determining the peak height of the most abundant fragment ion mass within a predefined retention time. Background contamination and potential carryover of identified metabolites were subtracted using blank samples. To enhance quantitative stability, relative concentrations of identified compounds were normalized using an internal reference (D4-alanine) based on their correlation with metabolites in quality control (QC) samples. The median of the QC samples was used to adjust for daily batch effects, and hair samples were corrected based on measured weight. Amino acids, fatty acids, and lactic acid were quantified using chemical reference standards. Metabolite levels were first normalized to the corresponding internal standards and then converted to absolute concentrations using calibration curves derived from the respective chemical standards (5 concentrations from 0 to 55.4 mM).

[0037] 1.4 Statistical analysis

[0038] Quantitative data were expressed as mean ± standard deviation (SD), and categorical variables were expressed as frequency and percentage. Categorical variables were compared using the χ2 test. 2Test. For continuous variables with normal distribution, one-way analysis of variance (ANOVA) was used; for non-normally distributed variables, the Kruskal-Wallis nonparametric test was used to compare the differences between the NC group, MCI group, and AD group. Spearman correlation analysis was used to test the correlation between two non-normally distributed continuous variables. Principal component analysis was used to analyze the differences in metabolites among the groups, and |Log2 FC|>0 was used as the standard to identify metabolite differences, with an adjusted p value of <0.05. The ggplot2 package of R was used to generate a volcano plot, and Venn analysis was performed to identify common differential metabolites. The metabolic characteristics were modeled using the LASSO logistic regression of the glmnet package. To evaluate the ability of the LASSO model to identify AD, ROC analysis was performed using the pROC package. All statistical analyses were performed using R software.

[0039] 2 Experimental results

[0040] First, PCA analysis was performed on the hair metabolites of different groups, and the results showed that there were differences in the metabolite levels among the three groups. Figure 1 As shown in A, clustering patterns were observed between AD, MCI, and NC groups, indicating that there were differences in metabolites between groups. PCA analysis based on CDR scores showed that there were some differences between group 0 and other groups ( Figure 1 B). PLS-DA further confirmed the significant metabolic separation ( Figure 1 CD), MCI and AD groups showed a clear trend of separation from the NC group. The volcano plot highlights the specific differential metabolites between the cognitive impairment groups (including MCI and AD) and the NC group. Red marks are upregulated metabolites such as linoleic acid, behenic acid and eicosanoid acid, and green marks are downregulated metabolites such as quinolinic acid, octanoic acid and myristic acid. Metabolites were filtered by the standard of |Log2FC|>0, and the adjusted p value was <0.05. The figure details the differential fold changes in metabolite levels and classifies them into chemical categories, including cholesterol derivatives, amino acids, tricarboxylic acid cycle intermediates, alkanes and fatty acids. Results ( Figure 1 ) showed that the metabolites linoleic acid, cholestenone, arachidonic acid, and docosanoic acid were significantly increased in the AD group, while quinolinic acid and myristic acid were decreased. These findings highlight the presence of significant increases or decreases in the levels of different metabolites in AD compared with normal controls.

[0041] Figure 2 Metabolic changes between Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal controls (NC) were analyzed and visualized using volcano plots and line graphs. Figure 2In A, the volcano plots of MCI and NC show the top five upregulated metabolites, including linoleic acid, phenylacetic acid, heptanoic acid, cinnamic acid, and sebacic acid (red), and the top five downregulated metabolites, including quinolinic acid, octanoic acid, methylmalonic acid, 4-hydroxyphenylacetic acid, and malic acid (green), with the selection criteria of |Log2FC|>0 and adjusted p-value<0.05. Figure 2 B shows the differential fold changes (Log2FC) of these differential metabolites when categorized by chemical class. Notably, upregulated fatty acids such as linoleic acid, capric acid, and cinnamic acid increased significantly with disease progression, while downregulated metabolites such as quinolinic acid, methylmalonic acid, and malic acid decreased significantly with disease progression. Figure 2 In C, the volcano plot of AD versus NC highlights the top five upregulated metabolites in the AD group, including cholestenone, arachidonic acid, dodecanoic acid, linoleic acid, and decanoic acid (marked in red), and the top five downregulated metabolites, such as quinolinic acid, decanoic acid, myristic acid, malic acid, and threonine (marked in green). Figure 2 D shows the fold change (Log2FC) of differential metabolites in the AD group, showing significant upregulation of 4,6-cholestadien-3-one, arachidonic acid, and behenic acid, and downregulation of quinolinic acid, malic acid, and decanoic acid. These findings indicate that compared with NC, both MCI and AD groups undergo significant metabolic reprogramming (a process by which cells, under specific physiological and pathological conditions, undergo systematic adjustments and shifts in their metabolic patterns to adapt to changes in the external environment and meet their own growth and differentiation needs), as evidenced by upregulation of fatty acid and cholesterol-related metabolites and downregulation of metabolites related to energy metabolism. These results provide valuable information on the metabolic characteristics and potential disease mechanisms of MCI and AD.

[0042] The following fourteen metabolites were found to be different among the three groups: triacontanes, erucic acid, heneicosane, behenic acid, tetracosane, 3-hydroxybutyric acid, 15-methylhexadecanoic acid, arachidic acid, heptadecanoic acid, melatonin, stearic acid, tetracosanoic acid, heneicosane, and docosane, which were distributed in fatty acids and their complexes, alkanes, and organic compounds.

[0043] Figure 3 Venn diagrams show the common and unique differential metabolites found in the comparisons between the AD and NC groups, the MCI and NC groups, and the CI and NC groups. There were 14 common differential metabolites across all three groups, while there were 15, 6, and 34 unique metabolites between the AD and NC groups, the CI and NC groups, and the MCI and NC groups, respectively. The levels of the 14 common differential metabolites are shown in Table 1.

[0044] Table 1 The results of the comparison of the different metabolite levels among the three groups and the differences between the groups are as follows

[0045]

[0046]

[0047] 1 Median [IQR]; Mean (SD)

[0048] 2 Kruskal-Wallis rank sum test; one-way ANOVA

[0049] LASSO logistic regression was used to construct a classification model. The 14 differential metabolite sets in Table 1 were used to construct classification models for NC vs AD and NC vs MCI. Figure 4 It can be seen that the AUC value for the NC vs AD binary classification is 0.839, and the AUC value for the NC vs MCI binary classification is 0.927, both of which achieved good classification performance (the closer the AUC is to 1.0, the better the classification performance of the model), suggesting that differential hair metabolites have potential clinical application value in the diagnosis of Alzheimer's disease.

[0050] In this study, we found significant differences in hair metabolites, including triacontanes, erucic acid, heneicosane, behenic acid, and tetracosane, between the AD and MCI groups compared with the NC group. The differential metabolites in the MCI group were primarily amino acids and fatty acids, while those in the AD group were primarily fatty acids and lipids, suggesting that hair metabolite levels fluctuate during the progression from NC to MCI to AD. Notably, we developed diagnostic biomarkers for MCI and AD based on hair metabolites. This method is minimally invasive and cost-effective, and represents a promising tool for large-scale screening of Alzheimer's disease in the general elderly population.

Claims

1. A combination of hair metabolic markers for diagnosing Alzheimer's disease, characterized by: The hair metabolism marker combination includes triacontan, erucic acid, heneicosane, behenic acid, tetracosane, 3-hydroxybutyric acid, 15-methylhexadecanoic acid, arachidic acid, heptadecanoic acid, melatonin, stearic acid, tetracosanoic acid, heneicosane, and docosane.

2. Use of the hair metabolic marker combination according to claim 1 in preparing a kit for diagnosing Alzheimer's disease.

3. A kit for diagnosing Alzheimer's disease, characterized in that: A standard product comprising the hair metabolic marker combination for diagnosing Alzheimer's disease according to claim 1.

4. The kit according to claim 3, wherein: Also included is an extraction reagent for extracting the hair metabolism marker.

5. The kit according to claim 4, wherein: The extraction reagents include acetone, sodium hydroxide and methanol.

6. The kit according to claim 3, wherein: An internal standard, D4-alanine, was also included.

7. The kit according to claim 3, wherein: Also included are reagents for gas chromatography-mass spectrometry analysis.

Images