DNA methylation markers for assessing cognitive impairment and uses thereof

Abstract

The present application relates to the field of biotechnology, in particular to a DNA methylation marker for evaluating the risk of cognitive dysfunction and application thereof, and finds that 15 DNA methylation markers can be used for predicting the risk of cognitive dysfunction. The present application actively seeks DNA methylation sites affecting cognitive dysfunction in the elderly population from the perspective of reducing the risk of cognitive dysfunction, and constructs a DNA methylation score to evaluate the predictive power of cognitive dysfunction, so as to realize early identification of patients with cognitive dysfunction, early prevention and intervention of cognitive function of patients, and provide potential possibility for promoting the health of patients.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology, and specifically relates to a DNA methylation biomarker for assessing cognitive impairment and its application. Background Technology

[0002] With the accelerating aging of the global population, cognitive decline among the elderly has become a serious public health challenge, with a significant increase in the incidence of Alzheimer's Disease (AD) and its precursor stage—mild cognitive impairment (MCI). Studies show that approximately 80% of MCI patients will progress to AD within 6 years, and there is currently no effective cure for AD. Therefore, developing early detection indicators and predictive models for MCI is of significant clinical value for early intervention and slowing disease progression.

[0003] Current predictive models for MCI and AD primarily rely on three types of indicators: demographic characteristics (such as age, sex, and education level), lifestyle factors (such as sleep patterns and exercise), and clinical indicators (such as history of chronic diseases and imaging examinations). While imaging examinations are crucial for the diagnosis of MCI and AD, their high cost and reliance on specialized equipment limit their widespread application in rapid and simple screening. Furthermore, although some studies have attempted to integrate genetic factors (such as APOE genotype), the sensitivity and specificity of traditional predictive factors remain limited, failing to meet the demands for accurate prediction. In recent years, the application of multi-omics technologies (such as transcriptomics, proteomics, and brain imaging omics) has driven the optimization of predictive models; however, the role of epigenetic mechanisms, particularly DNA methylation, in the pathophysiology of MCI and AD remains largely unexplored.

[0004] DNA methylation, as a core epigenetic regulatory mechanism, participates in processes such as neuroplasticity, synaptic function, and neuroinflammation by dynamically modifying gene expression, and is closely related to the occurrence and development of neurodegenerative diseases. Current research has confirmed significant abnormal DNA methylation patterns in the peripheral blood and brain tissue of patients with microvascular Illness (MCI) and Advanced Neurological Disorder (AD), and these abnormalities are highly associated with the dysregulation of cognitive function-related genes (such as BDNF and SNAP25). Although DNA methylation analysis based on peripheral blood or cerebrospinal fluid provides a new direction for early prediction of MCI, current technologies still lack methods for systematically screening and validating specific methylation biomarkers, resulting in insufficient sensitivity and specificity of predictive models.

[0005] In summary, the following problems urgently need to be addressed in the existing technology: (1) traditional predictive factors have limited early warning capabilities for MCI and AD; (2) the key role of epigenetic mechanisms (especially DNA methylation) in the disease process has not been fully integrated into the prediction system; and (3) there is a lack of methods for screening and validating biomarkers based on multidimensional methylation data. To address these technological gaps, this invention focuses on the regulatory mechanism of DNA methylation in the occurrence and development of MCI, aiming to systematically screen and validate highly specific methylation biomarkers to provide technical support for the development of novel diagnostic tools and personalized prevention and treatment strategies. Summary of the Invention

[0006] To overcome the shortcomings of existing technologies, this invention provides a DNA methylation biomarker for assessing cognitive impairment and its application. The purpose of this invention is to construct an early prediction model for cognitive impairment (MCI) based on genomic DNA methylation characteristics. By screening and validating specific methylation biomarkers associated with MCI, accurate early prediction of MCI can be achieved, providing a reliable tool for early screening of MCI.

[0007] To achieve the objectives of this invention, the technical solution is as follows:

[0008] A DNA methylation biomarker for assessing cognitive impairment, the biomarker comprising one or more of the following 15 methylation sites: cg13618433 located in the upstream regulatory region of the CXCL13 gene on chromosome 4, cg12436019 located in the PCCA gene on chromosome 13, cg07021405 located in the upstream regulatory region of the SLC45A1 gene on chromosome 1, cg08149581 located in the NCOR2 gene on chromosome 12, cg18101488 located in the upstream regulatory region of the YY1 gene on chromosome 14, cg17348924 located in the HAPLN1 gene on chromosome 5, and cg0 located in the FBXO39 gene on chromosome 17. 4115740, cg22962669 located in the downstream regulatory region of the PEBP4 gene on chromosome 8, cg01632381 located in the downstream regulatory region of the TFAP2C gene on chromosome 20, cg01514843 located in the VEPH1 gene on chromosome 3, cg05843557 located in the downstream regulatory region of the FAM110B gene on chromosome 8, cg25979148 located in the GAK gene on chromosome 4, cg26159368 located in the downstream regulatory region of the EBF3 gene on chromosome 10, cg24862627 located in the EPHB4 gene on chromosome 7, and cg22660420 located in the upstream regulatory region of the PAOX gene on chromosome 10.

[0009] To further explain, the cognitive impairment status of the target subjects is assessed based on the DNA methylation level of the aforementioned markers in the target subjects.

[0010] This invention also provides a method for constructing a DNA methylation score, using the following formula:

[0011]

[0012] DMS stands for DNA methylation score. The weight of the i-th methylation site; represents the methylation level (M value) of the i-th methylation site in the i-th sample; n is the number of selected methylation sites, n = 15;

[0013] The method for determining the weight of methylation sites is as follows:

[0014] (1) 26 differentially methylated sites were identified by screening for differentially expressed sites in cases of cognitive impairment in the elderly population using epigenome association analysis of methylation.

[0015] (2) These 26 methylation sites were included in the minimum absolute contraction and selection operator model for site screening and weight determination, and finally 15 methylation sites were selected.

[0016] The methylation sites and their weights are as follows:

[0017]

[0018] To further explain, the specific formula for calculating the DMS is as follows:

[0019] DMS = 2.92688778 × cg13618433 + (-0.51702940) × cg12436019 + (-0.51131591) × cg07021405 + 1.28697396 × cg08149581 + 0.40121157 ×cg18101488 + 0.24181143 × cg17348924 + (-0.49135004) × cg04115740 + (-0.37437583 ) × cg22962669 + (-0.45821403 ) × cg01632381 + 0.36287111 ×cg01514843 + 0.46490156 × cg05843557 + (-0.52422779) × cg25979148 + (-0.78450358) × cg26159368 + (-0.72544161 ) × cg24862627 + (-0.11972678)×cg22660420.

[0020] The present invention also provides a kit for assessing cognitive impairment in target subjects, comprising primers, probes or chips for detecting said methylation sites.

[0021] To further clarify, the sample tested was whole blood leukocyte DNA from the target individual.

[0022] The present invention also provides a method for assessing cognitive impairment in a target subject, comprising the following steps:

[0023] (1) Detect and assess the methylation level of the methylation sites in the DNA of the target object as described above;

[0024] (2) Calculate the DNA methylation score based on the above formula;

[0025] (3) Combine basic predictive factors to predict the risk of cognitive impairment through the minimum absolute contraction and selection operator model.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. This invention clarifies the association between DNA methylation sites cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420 and cognitive impairment. The further constructed DMS can integrate information from multiple methylation sites to provide a more comprehensive and accurate risk assessment. Compared with traditional imaging tests, DNA methylation testing has the advantages of being less invasive and relatively less expensive. It is suitable for the early assessment of cognitive function in patients with mild cognitive impairment, and its testing burden is far lower than the burden that the development of cognitive impairment causes to society and individuals.

[0028] 2. The cognitive impairment assessment model constructed in this invention, which combines DMS with traditional risk factors, can significantly improve the ability of traditional models to assess cognitive impairment. This is beneficial for more accurate and efficient identification of high-risk groups for cognitive impairment, enabling early intervention, reducing the occurrence and development of cognitive impairment or even dementia, and developing individualized prevention and treatment plans for the elderly. Attached Figure Description

[0029] Figure 1 Genome-wide DNA methylation association analysis in mild cognitive impairment. The left figure is a Manhattan plot, where the x-axis represents the chromosomal location of DNA methylation sites, and the y-axis represents the -log association. 10 (p-value), the red dashed line represents P = 1.0 × 10⁻⁶. −5 -log of time 10 (p-value); The right figure is a quantile plot, with the x-axis and y-axis representing the expected value minus log₂. 10 (p-value) and observation-log 10 (p-value).

[0030] Figure 2 To illustrate the variation of prediction error with the penalty parameter logλ in the Least Absolute Shrinkage and Selection Operator (LASSO) model, the left vertical line represents the optimal λ (minimum mean square error) determined through cross-validation, and the right vertical line represents the simplest model obtained within one standard error range of the optimal λ value.

[0031] Figure 3 This describes the model selection path in the LASSO model based on the penalty parameter logλ. As the value of logλ increases, the penalty on the regression coefficients becomes more stringent, ultimately reducing more regression coefficients to zero.

[0032] Figure 4 The risk predictive efficacy of traditional risk factor reference models and methylation scores constructed by incorporating 15 differentially methylated sites was evaluated using receiver operating characteristic (ROC) curve analysis to assess the predictive efficacy of MCI in the healthy elderly cohort in the Hongshuihe area of ​​Guangxi.

[0033] Figure 5 The risk prediction efficacy of the public database E-MTAB-10600 for MCI was evaluated by assessing the traditional risk factor reference model and the methylation score constructed by adding 15 differentially methylated sites using receiver operating characteristic (ROC) curve analysis. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention. Unless otherwise specified, the experimental methods used in the following embodiments are conventional methods. Unless otherwise specified, the materials and reagents used in the following embodiments are commercially available.

[0035] Example 1: Screening of methylation sites

[0036] 1. Study Population: Based on the Guangxi Hongshuihe Healthy Elderly Cohort, this cohort recruited participants in 2016 from three towns (Donglan Town, Wuzhuan Town, and Sanshi Town) in the Hongshuihe River Basin of Hechi City, Guangxi Province, to establish a baseline. A total of 4,650 participants aged 60-118 years signed informed consent forms and completed the baseline survey, including a questionnaire [including the Mini-Mental State Examination (MMSE)], physical measurements, and blood sample collection. Based on the baseline data, 280 participants were randomly selected according to different ages for DNA methylation testing. Participants with missing MMSE scores, educational background, and APOE genotypes were further excluded, ultimately including 163 participants aged 60 years and older for analysis. Peripheral blood DNA was collected for Illumina 850K methylation chip detection.

[0037] 2. Assessment of Cognitive Function and Definition Criteria for Mild Cognitive Impairment (MCI): The MMSE scale is widely used for cognitive function assessment and is a tool for screening cognitive impairment in clinical, academic research, and community settings. The scale is divided into orientation, immediate memory, attention and calculation, recall, and language and practical skills (scoring 10, 3, 5, 3, and 9 points respectively). MMSE scores range from 0 to 30, with higher scores indicating better cognitive ability. MCI is a stage of cognitive decline and a precursor to Alzheimer's disease. This invention uses the MMSE scale to screen for MCI. Following the scale's usage principles, MCI is assessed based on educational attainment: ① Illiterate: MMSE score ≤ 17; ② Primary school education: MMSE score ≤ 20; ③ Secondary school education or above: MMSE score ≤ 24. Those meeting these criteria are defined as having MCI; otherwise, they are defined as having normal cognitive function.

[0038] 3. Main instruments and reagents for DNA methylation detection

[0039] Main instruments: pipettes (1mL, 200μL, 100μL, 10μL, Eppendorf, Germany), high / low speed centrifuges (Eppendorf, Germany), gel imaging system (Shanghai GeneScience & Technology Co., Ltd., China), bistable electrophoresis apparatus (Beijing Liuyi Biotechnology Co., Ltd., China), PCR gene amplification instrument (Stepone Plus, USA), PCR eight-tube strips (Hangzhou Aisijin Biotechnology Co., Ltd., China), micro-volume UV spectrophotometer (Thermo Fisher Scientific, USA), high-precision tube sheet heating system (Hybex, SciGene, USA), high-throughput genotyping system (iScan, Illumina, USA), etc.

[0040] Main reagents: Genomic DNA extraction kit (Beijing Adley Biotechnology Co., Ltd.), red blood cell lysis buffer (Beijing Adley Biotechnology Co., Ltd.), nuclear lysis buffer (Beijing Adley Biotechnology Co., Ltd.), protein precipitation (Beijing Adley Biotechnology Co., Ltd.), DNA dissolving buffer (Beijing Adley Biotechnology Co., Ltd.), anhydrous ethanol (analytical grade, China National Pharmaceutical Group Chemical Reagent Co., Ltd.), EZ DNA Methylation–Gold conversion kit (Zymo Biotech, Inc., USA), Infinium Human Methylation EPIC methylation chip matching reagents (Illumina, Inc., USA), etc.

[0041] 4. Experimental Methods

[0042] 4.1 Whole Blood Leukocyte DNA Extraction: The frozen blood sample was placed at 4°C overnight, and brought to room temperature 30 minutes before the experiment. Red blood cell lysis buffer, nuclear lysis buffer, and protein precipitation buffer were added sequentially to lyse the nuclei of red blood cells and white blood cells, and to remove protein impurities, respectively. Subsequently, DNA was precipitated with isopropanol, washed with ethanol, and then incubated with DNA dissolving solution at 55°C overnight. After the DNA was fully dissolved, it was stored at -20°C.

[0043] 4.2 DNA Sample Quality Control: Take the above-mentioned DNA samples, mix them thoroughly, and then use a micro-volume UV spectrophotometer to detect the concentration and purity of the DNA. A DNA concentration between 50 and 60 ng / μL and a purity A260 / 280 between 1.8 and 2.0 are considered to meet the requirements for microarray experiments. Otherwise, concentration, dilution, or purification should be performed until the requirements are met. In addition, DNA agarose gel electrophoresis is required. If obvious tailing of the DNA electrophoresis bands occurs, it indicates that the DNA has degraded and is classified as an unqualified sample, which should be extracted again for testing.

[0044] 4.3 DNA bisulfite conversion: According to the instructions of the EZ DNA bisulfite conversion kit (Zymo), the quality-controlled DNA samples were subjected to bisulfite conversion and purification.

[0045] 4.4 Detection of whole-genome DNA methylation using an array: Experiments were conducted according to the Infinium HumanMethylationEPIC array (Illumina, USA) instruction manual. The scanned files from the iScan scanner were imported into GenomeStudio software, and the data was converted to IDAT format.

[0046] 5. Statistical Analysis

[0047] 5.1 Quality control and standardization of methylation chip data

[0048] The study used the CHAMPR software package to read the raw data of the IDAT format methylation microarray and further performed quality control and standardization on the read data. The following criteria were used to perform quality control on the samples and CpG probes: (1) Quality control of samples: samples with a probe deletion rate >10% were excluded (n=0). (2) Quality control of CpG probes: the following probes were excluded: ① probes with a low signal detection rate (detection P value >0.01) in all samples (n=17231); ② probes with a bead count <3 in more than 5% of the samples (n=4211); ③ non-CpG probes (n = 2586); ④ probes located on sex chromosomes (n=18280); ⑤ CpG probes located in the target region of SNPs (based on the Asian population 1000 Genomes Database) with a minimum allele frequency (MAF) >0.05 (n=7639); ⑥ CpG probes that hybridize to other genomic locations (n=39970).

[0049] Based on the above quality control methods, 776,001 CpG sites from 163 study subjects passed the quality control standards. Next, the probe distribution was standardized using Beta Mixture Quantile (BMIQ) to correct for type I and II probe bias. The β value of each CpG site, representing its methylation level, was calculated based on the ratio of unmethylated to methylated sites. Considering that the methylation M value is suitable for statistical analysis, the logarithm of the methylation β value was converted to an M value for subsequent analysis, M = log2[β / (1-β)].

[0050] 5.2 Genome-wide DNA methylation association analysis of MCI

[0051] First, this study used the ComBat method of the "sva" R software package to correct for batch effects between different chip sections. Second, the whole-genome DNA methylation association analysis of MCI employed a robust linear regression model based on empirical Bayesian methods in the "limma" software package, using the methylation M value as the dependent variable, while correcting for age, sex, body mass index (BMI), smoking status, education level, APOE genotype, and major leukocyte proportion. Since there is an intrinsic association between leukocyte proportion and methylation level, differences in its composition may lead to observed methylation changes not being entirely caused by the studied factors, thus introducing confounding bias. Therefore, it is necessary to correct for the proportion of major leukocytes (CD8T, CD4T, NK, B cells, monocytes, and granulocytes) in the EWAS analysis, which was assessed using the Houseman method. This study considers P < 1.0 × 10⁻⁶ as a negative criterion. −5 CpGs are defined as reaching suggestive genome-wide significance levels.

[0052] 5.3 The Least Absolute Contraction and Selection Operator (LASSO) model was used for further screening. MCI was used as the dependent variable, and the 26 differentially expressed CpG sites obtained from the genome-wide DNA methylation association analysis of MCI in step 5.2 were used as independent variables. Gender, age, education level, smoking status, BMI, and APOE genotype were included as covariates in the LASSO model. Feature selection was performed on the CpG sites in the prediction model. The LASSO model was performed using the R language package "glmnet", with parameters set as follows: alpha = 1.0, the link function "Binomial", and the penalty parameter lambda determined using 10-fold cross-validation. The lambda value that minimized the root mean square error was selected as the optimal model, thus performing variable screening. The L1 penalty term of the model compressed the regression coefficients of some sites to zero (these sites had weak independent contributions to MCI prediction), achieving variable selection. CpG sites with non-zero regression coefficients at the given λ value were considered to make significant contributions to MCI prediction. The regression coefficients of these sites were used as their relative weights for MCI prediction and were subsequently used to construct the DMS methylation score.

[0053] 6. Results

[0054] 6.1 MCI-related differential methylation sites

[0055] After adjusting for age, sex, BMI, smoking status, education level, APOE genotype, and major leukocyte percentage as covariates to identify candidate assessment factors, 26 differentially methylated sites were screened (P < 1.0 × 10⁻⁶). −5 (See Table 1 and Appendix) Figure 1 .

[0056] Table 1

[0057]

[0058] 6.2 Results of the LASSO model

[0059] The penalty parameter lambda of the model was determined using 10-fold cross-validation. The lambda value that minimizes the root mean square error (λ=0.01842928) was selected as the optimal model (see appendix). Figure 2 ), and variable selection is performed. The L1 penalty term of the model compresses the regression coefficients of some loci to zero (these loci have weak independent contributions to MCI prediction), thereby achieving variable selection (see appendix). Figure 3 Ultimately, the 15 CpG sites whose regression coefficients were not zero at this λ value were considered to make significant contributions to MCI prediction. The weights of these 15 sites are shown in Table 2.

[0060] Table 2

[0061]

[0062] Example 2: DMS Construction and Verification

[0063] 1. DMS Construction:

[0064] Scoring formula: The individual DMS value is calculated based on the methylation sites and their weight parameters determined in Example 1.

[0065] The weighted formula is:

[0066] DMS stands for DNA methylation score.

[0067] wi represents the weight of the i-th methylation site. Sites selected based on DNA methylation site differential analysis are further incorporated into the LASSO model, and the site weights are determined according to the LASSO parameters.

[0068] Mi represents the methylation level (M value) of the i-th methylation site in the i-th sample.

[0069] n represents the number of selected methylation sites, n=15.

[0070] The DMS of this invention is calculated as follows:

[0071] DMS = 2.92688778 × cg13618433 + (-0.51702940) × cg12436019 + (-0.51131591) × cg07021405 + 1.28697396 × cg08149581 + 0.40121157 ×cg18101488 + 0.24181143 × cg17348924 + (-0.49135004) × cg04115740 + (-0.37437583 ) × cg22962669 + (-0.45821403 ) × cg01632381 + 0.36287111 ×cg01514843 + 0.46490156 × cg05843557 + (-0.52422779) × cg25979148 + (-0.78450358) × cg26159368 + (-0.72544161 ) × cg24862627 + (-0.11972678)×cg22660420.

[0072] 2. Model Validation: The performance of the model is evaluated based on the LASSO model for MCI (binary variable).

[0073] The formula for evaluating the model is as follows:

[0074] Risk =

[0075] Here, Risk represents the probability that an individual is assessed as MCI (MCI=1). The regression coefficient represents the contribution of each predictor variable to the outcome. These are the predictor variables in the model.

[0076] Based on the assessment model constructed above, the diagnostic value of DMS for MCI was determined. Compared to the baseline model (age, sex, APOE genotype), the assessment model integrating DMS and baseline risk factors showed a significantly increased diagnostic value for MCI. (Receiver operating curves are shown in [link to receiving curve]). Figure 4 The results showed that the AUC value of DMS combined with baseline factors in diagnosing MCI was 0.956, which was significantly improved compared with the AUC of 0.721 of the baseline model, indicating that DMS has high diagnostic value as a methylation marker for MCI.

[0077] 3. The above assessment model was validated using a public database. Based on the public database E-MTAB-10600 (https: / / www.ebi.ac.uk / biostudies / arrayexpress / studies / E-MTAB-10600), a sample size of 68 individuals (34 with MCI and 34 in the control group) was obtained, along with a DNA methylation database. Covariates included sex, age, and APOE genotype. First, based on the DMS parameter weights constructed in Example 1, the DMS in the public database was calculated. Then, the LASSO assessment model was used to validate the assessment power of the above model. The participant operating curve (NIPC) was obtained from the public database validation (see [link to relevant documentation]). Figure 5 The results showed that the AUC value of DMS combined with baseline factors in diagnosing MCI was 0.724, which was significantly improved compared with the AUC of 0.669 of the baseline model. This verifies the effectiveness of the above evaluation model and indicates that the DMS constructed in this invention has high diagnostic value as a methylation marker for MCI.

[0078] Example 3: The kit uses qPCR or methylation chip to detect target sites, and combines DMS scores and clinical data to output an MCI risk report.

[0079] This invention provides the application of the reagent for detecting the methylation level of the DNA methylation site in the preparation of reagents for assessing the risk of mild cognitive impairment (MCI), or in the preparation of kits for assessing the risk of mild cognitive impairment.

[0080] The present invention also provides the application of the reagent for detecting the methylation level of the DNA methylation site in the preparation of reagents for screening mild cognitive impairment, or in the preparation of kits for screening mild cognitive impairment.

[0081] Preferably, the kit can be any reagent known in the art for detecting site-specific DNA methylation levels, as long as it can detect the leukocyte DNA methylation levels at the following sites: cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420. This includes, but is not limited to, the embodiments listed below. The kit also includes, but is not limited to, specific primers for amplifying the cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420 sites. These primers can be designed using MethPrimer software. The reagents include a PCR kit and commonly used reagents required for the corresponding PCR techniques, such as dNTPs, MgCl2, double-distilled water, and Taq polymerase.

[0082] In a first embodiment, the kit includes reagents for detecting the leukocyte DNA methylation levels at the following sites in a sample using Targeted Bisulfite Sequencing (TBS): cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420. TBS (Transfer-Based Sequencing) can detect several to hundreds of gene / DNA methylation sites, offering advantages such as high accuracy, high throughput, low cost, and fast turnaround time. It is widely used for screening, validating, and translating methylation biomarkers at multiple sites in clinical samples. The process begins with library construction, involving the design and synthesis of BS-PCR primers for the target region or site. Simultaneously, sample DNA is extracted, and after passing initial testing, the DNA undergoes bisulfite conversion (EZ DNA Methylation Gold Kit, Zymo Research). A high-fidelity, U-resistant DNA polymerase is used to amplify the bisulfite-converted template using BSP (Bifermentation Sample Spectroscopy). BSP amplification products from the same sample are mixed and amplified with tagged primers, attaching Illumina sequencing adapters. This results in sequencing libraries with different tags for each sample. Each library undergoes purification, quantification, and multi-library mixing, followed by quality control before sequencing. After library approval, different libraries are pooled according to effective concentration and target data volume requirements before sequencing on the Illumina platform. The risk of MCI (Methylation-Induced Chronic Infection) is predicted based on the DNA methylation levels at these sites.

[0083] In a second embodiment, the kit includes reagents for detecting leukocyte DNA methylation levels at the following loci in a sample: cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420, using pyrosequencing. Pyrosequencing is a well-known technique in the art and is considered the "gold standard" method for detecting the methylation status of specific genes. It involves first converting the DNA with bisulfite, followed by pyrosequencing. Those skilled in the art can choose according to their needs, and will not be elaborated further here. Using this kit, the methylation level of the DNA methylation sites in the sample can be directly measured by pyrosequencing, and the risk of MCI can be predicted based on the DNA methylation level of these sites.

[0084] In the third embodiment, the kit includes reagents for detecting the leukocyte DNA methylation levels at sites cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420 in samples using DNA microarray detection methods (such as the Infinium Human Methylation EPIC methylation chip, for details please refer to the Illumina website's operation manual). The risk of MCI is predicted based on the DNA methylation levels at these sites.

[0085] Furthermore, the present invention also provides the application of the aforementioned DNA methylation sites for screening drug targets for the prevention or treatment of mild cognitive impairment.

[0086] This invention also provides the application of the aforementioned DNA methylation sites as targets for studying the causal mechanisms of mild cognitive impairment.

[0087] Preferably, the invention provides DNA methylation sites associated with MCI. DNA methylation is a reversible epigenetic modification that can influence the methylation level at specific sites through lifestyle changes, medication use, etc., thereby achieving the purpose of disease prevention or treatment. Therefore, this invention provides highly valuable targets for the prevention and treatment of MCI, which can be used for drug development.

[0088] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, any improvements and changes made without departing from the inventive concept of the present invention are within the protection scope of the present invention.

Claims

1. A DNA methylation biomarker for assessing cognitive impairment, characterized in that, The biomarker consists of the following 15 methylation sites: cg13618433 located in the upstream regulatory region of the CXCL13 gene on chromosome 4; cg12436019 located in the PCCA gene on chromosome 13; cg07021405 located in the upstream regulatory region of the SLC45A1 gene on chromosome 1; cg08149581 located in the NCOR2 gene on chromosome 12; cg18101488 located in the upstream regulatory region of the YY1 gene on chromosome 14; cg17348924 located in the HAPLN1 gene on chromosome 5; cg04115740 located in the FBXO39 gene on chromosome 17; and cg04115740 located in the downstream regulatory region of the PEBP4 gene on chromosome 8. The biomarkers cg22962669 in the control region, cg01632381 in the downstream regulatory region of the TFAP2C gene on chromosome 20, cg01514843 in the VEPH1 gene on chromosome 3, cg05843557 in the downstream regulatory region of the FAM110B gene on chromosome 8, cg25979148 in the GAK gene on chromosome 4, cg26159368 in the downstream regulatory region of the EBF3 gene on chromosome 10, cg24862627 in the EPHB4 gene on chromosome 7, and cg22660420 in the upstream regulatory region of the PAOX gene on chromosome 10 are derived from whole blood leukocyte DNA of the target subjects.

2. The DNA methylation marker according to claim 1, characterized in that, The cognitive impairment status of the target subjects is assessed based on the DNA methylation level of the aforementioned markers in the target subjects.

3. A method for constructing a DNA methylation score, characterized in that, The following formula is used: DMS stands for DNA methylation score. The weight of the i-th methylation site; represents the methylation level (M value) of the i-th methylation site in the i-th sample; n is the number of selected methylation sites, n = 15; The method for determining the weight of methylation sites is as follows: (1) 26 differentially methylated sites were identified by screening for differentially expressed sites in cases of cognitive impairment in the elderly population using epigenome association analysis of methylation; (2) These 26 methylation sites were included in the minimum absolute contraction and selection operator model for site screening and weight determination, and finally 15 methylation sites were selected. The methylation sites and their weights are as follows: 。 4. The method for DNA methylation scoring according to claim 3, characterized in that, The specific formula for calculating DMS is as follows: DMS = 2.92688778 × cg13618433 + (-0.51702940) × cg12436019 + (-0.51131591) × cg07021405 + 1.28697396 × cg08149581 + 0.40121157 ×cg18101488 + 0.24181143 × cg17348924 + (-0.49135004) × cg04115740 + (-0.37437583 ) × cg22962669 + (-0.45821403 ) × cg01632381 + 0.36287111 ×cg01514843 + 0.46490156 × cg05843557 + (-0.52422779) × cg25979148 + (-0.78450358) × cg26159368 + (-0.72544161 ) × cg24862627 + (-0.11972678)×cg22660420; Wherein, cg13618433, cg12436019, cg07021405, cg08149581, cg18101488, cg17348924, cg04115740, cg22962669, cg01632381, cg01514843, cg05843557, cg25979148, cg26159368, cg24862627, and cg22660420 represent the methylation level (M value) of the corresponding sites.

5. A kit for assessing cognitive impairment in a target subject, characterized in that, It includes primers, probes, or chips for detecting the methylation sites of claim 1.

6. The use of the DNA methylation marker of claim 1 in the preparation of a kit for assessing cognitive impairment.

Images