Biomarker for diagnosing Alzheimer's disease as well as screening method and application of biomarker

Abstract

The invention discloses a biomarker for diagnosing Alzheimer's disease as well as a screening method and application thereof, relates to the technical field of biomarkers for Alzheimer's disease, and aims to solve the technical problems that the Alzheimer's disease lacks dynamic pathological monitoring indexes and stereoisomeric modification is difficult to systematically detect. According to the invention, the stereoisomeric modification level of peptide fragments KLDLSNVQSK and AKTDHGAEIVYK is used as the core biomarker for the first time, the modification level is remarkably related to the Braak staging of AD, and a detection method and a combined diagnosis model containing the marker are provided. A novel molecular target with high correlation and dynamic monitoring potential is provided for AD, the model diagnosis precision is remarkably improved, and a brand new solution is provided for disease mechanism research and drug development.

Description

Technical Field

[0001] This invention relates to the field of Alzheimer's disease biomarker technology. More specifically, this invention relates to a biomarker for diagnosing Alzheimer's disease, a screening method thereon, and its applications. Background Technology

[0002] Alzheimer's disease (AD) is a neurodegenerative disease characterized by progressive cognitive decline. Its pathophysiological process is complex, often including a preclinical period lasting decades. Typical neuropathological features of AD include senile plaques formed by extracellular β-amyloid (Aβ) deposition in the cerebral cortex and hippocampus, and neurofibrillary tangles (NFTs) formed by intracellular hyperphosphorylated Tau protein aggregation. However, the insidious period of decades between the appearance of pathological changes and the manifestation of clinical symptoms poses a significant challenge to early diagnosis and intervention. Currently, the gold standard for AD diagnosis still relies on postmortem neuropathological examination, specifically the detection of characteristic Aβ plaques and NFTs. Premortem diagnosis primarily relies on clinical symptoms, cognitive scale assessments, cerebrospinal fluid biomarkers (Aβ42, p-Tau, t-Tau), and positron emission tomography (PET) scans of amyloid protein. However, these methods all have limitations: clinical symptoms appear late; scale assessments are highly subjective; and PET scans are expensive and involve radiation exposure. Therefore, developing novel biomarkers that are objective and can dynamically reflect the pathological process has become an urgent need in the current field of AD research.

[0003] Post-translational modifications of proteins play a crucial role in the pathological mechanisms of Alzheimer's disease (AD). Among these, aberrant hyperphosphorylation of Tau protein is a core step in the formation of non-fibroblastic proteins (NFTs) and forms the basis for current mainstream fluid biomarkers (such as p-Tau181 and p-Tau217). However, phosphorylation is only one aspect of Tau pathological changes. Stereoisomerization is another type of post-translational modification that is increasingly gaining attention in aging and neurodegenerative diseases. It involves the racemic or isomerization of amino acid residues from their native L-configuration to their D-configuration. Unlike phosphorylation, stereosomerization does not alter the molecular weight of the protein but fundamentally changes the local chiral center, potentially significantly affecting the protein's secondary / tertiary structure, stability, interactions, and proteolytic properties.

[0004] Due to the limitations of low throughput and high cost of traditional detection technologies, systematic exploration of stereoisomerism in the proteome remains very limited, especially research on its dynamic correlation with disease pathological staging is almost non-existent. Whether it is possible to discover, with high throughput, specific stereoisomerism modification sites in complex biological samples that are strictly related to the pathological progression of Alzheimer's disease (AD) and possess diagnostic and staging potential remains an unsolved technical challenge in this field. Summary of the Invention

[0005] The purpose of this invention is to utilize advanced heterogeneous retrieval technology to deeply mine proteomics data of clinical samples covering different pathological stages of Alzheimer's disease (AD), identify specific stereoisomer-modified peptides that are highly correlated with AD Braak staging, and verify their efficacy in disease diagnosis and pathological staging by constructing statistical models, ultimately providing a new set of molecular markers with high clinical application potential.

[0006] To achieve these objectives and other advantages according to the present invention, a biomarker for diagnosing Alzheimer's disease is provided, said biomarker being a stereoisomerically modified peptide of the following: Peptide sequence 1: KLDLSNVQSK; Peptide sequence 2: AKTDHGAEIVYK; The stereoisomerization modification refers to the D-isomerization of at least one amino acid residue in peptide sequence 1 and / or peptide sequence 2, and the level of stereoisomerization modification of the biomarker is correlated with the Braak stage of neuropathology of Alzheimer's disease.

[0007] Preferably, the stereoisomerization level of the biomarker is correlated with the Braak neuropathological stage of Alzheimer's disease, as follows: as the severity of the pathology increases, the stereoisomerization level of peptide sequence 1 decreases, while the stereoisomerization level of peptide sequence 2 increases.

[0008] The present invention also provides the application of the above-mentioned biomarkers in the preparation of products for diagnosing or assisting in the diagnosis of Alzheimer's disease.

[0009] The present invention also provides the application of the above-mentioned biomarkers as targets in the preparation of products for treating Alzheimer's disease.

[0010] This invention also provides a method for screening the above-mentioned biomarkers, comprising the following steps: Obtain proteomics mass spectrometry data from samples of different pathological stages of Alzheimer's disease and healthy control samples; The proteomics mass spectrometry data were searched using a heterogeneity search algorithm to identify peptides in the proteomics mass spectrometry data that have undergone amino acid isomerization modification. Stereoisomer-modified peptides derived from Tau protein were screened from the identification results; Calculate the modification level of stereoisomerically modified peptides derived from Tau protein in samples from different pathological stages of Alzheimer's disease; The correlation between the modification level of each stereoisomer modified peptide and the pathological stage of Alzheimer's disease was analyzed. Stereoisomer-modified peptides whose modification levels were significantly correlated with the pathological stage of Alzheimer's disease were identified as candidate biomarkers.

[0011] Preferably, the screening method uses Spearman rank correlation analysis to assess the correlation, and peptides that satisfy the correlation coefficient |ρ|>0.6 and have a p-value <0.05 after multiple test correction are identified as candidate biomarkers.

[0012] The present invention also provides a method for constructing a model for diagnosing Alzheimer's disease, comprising the following steps: Obtain proteomics mass spectrometry data from samples of different pathological stages of Alzheimer's disease and healthy control samples; The stereoisomerization modification level of peptide sequence 1, the stereoisomerization modification level of peptide sequence 2, and the phosphorylation modification level of Tau217 protein were calculated and statistically analyzed in each sample. Univariate analysis was performed on the stereoisomerization modification levels of peptide sequence 1, the stereoisomerization modification levels of peptide sequence 2, and the phosphorylation modification levels of Tau217 protein to assess their respective associations with Alzheimer's disease. The stereoisomerization levels of peptide sequence 1, peptide sequence 2, and phosphorylation levels of Tau217 protein were tested for collinearity to confirm that they met the conditions for constructing a multivariate statistical model. Using Alzheimer's disease stage as the dependent variable, and the stereoisomerization modification level of peptide sequence 1, the stereoisomerization modification level of peptide sequence 2, the phosphorylation modification level of Tau217 protein, and the interaction term as independent variables, a stepwise regression method was used for feature selection to construct a multivariate logistic regression model, and the multivariate logistic regression coefficients were determined by maximum likelihood estimation.

[0013] The present invention has at least the following beneficial effects: Originality and Strong Correlation: This study is the first to systematically discover and validate the close correlation between stereoisomerism modifications of two specific peptide sequences (SEQ ID NO: 1 and SEQ ID NO: 2) in Tau protein and the pathological progression of Alzheimer's disease (AD). The correlation between the modification level and Braak stage is significantly better than that of many traditional phosphorylation markers, providing a novel perspective for understanding the molecular pathological mechanisms of AD.

[0014] Excellent potential for dynamic monitoring: The level of this modification changes regularly and gradually with disease progression, making these two biomarkers not only indicators for distinguishing disease states, but also potential "yardsticks" for disease progression. Therefore, they are very suitable for dynamically monitoring the natural history of disease, predicting clinical translational milestones, and objectively evaluating the efficacy of drug interventions.

[0015] High specificity and auxiliary diagnostic value: Abnormal aggregation of Tau protein is one of the core pathological features of Alzheimer's disease (AD). SEQ ID NO: 1 and SEQ ID NO: 2 are located in the microtubule-binding repeat region (MTBR) and C-terminal domain of Tau protein, respectively. These regions are highly correlated with the formation of AD-specific neurofibrillary tangles. Therefore, their modification patterns may have good AD specificity and help differentiate other types of Tau protein diseases or neurodegenerative diseases in pathological diagnosis.

[0016] Providing new targets for treatment development: This discovery suggests that regulating the stereoisomerization process of specific residues in the Tau protein (e.g., inhibiting the activity of enzymes that promote this modification or activating deisomerizing enzymes) may be a novel therapeutic strategy for intervening in the progression of AD pathology, opening up new targets for drug development.

[0017] The methodology is universal: the systematic discovery process based on heterogeneous retrieval adopted in this invention can not only be applied to AD and Tau protein, but also provide a reference technical route for finding stereoisomer modification biomarkers with staging value in other diseases.

[0018] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0019] Figure 1 This is a full-length stereoisomer modification map of human Tau protein (longest isoform, 2N4R) and a schematic diagram of the location of key modification sequences, drawn based on human brain tissue proteomics data according to the present invention.

[0020] Figure 2 This study demonstrates the correlation between the stereoisomerization modification levels of key Tau protein peptides and the Braak stage of Alzheimer's disease (AD). The scatter plot reflects the change in the proportion of D-isomers (Y-axis) of multiple peptides (including SEQ ID NO: 1, SEQ ID NO: 2, and other control peptides) with the sample Braak stage (X-axis, stages 0-VI). The trend lines clearly show that the modification level of SEQ ID NO: 1 significantly decreased with pathological progression (negative correlation), the modification level of SEQ ID NO: 2 significantly increased with pathological progression (positive correlation), while the trends of other peptides were relatively flat or showed no significant pattern.

[0021] Figure 3This is a performance evaluation graph (receiver operating characteristic curve, ROC curve) of the AD diagnostic model constructed in Example 2. The graph compares the ROC curves of four models: A (model using only the stereoisomer modification level of SEQ ID NO: 1), B (model using only the stereoisomer modification level of SEQ ID NO: 2), C (model using only the phosphorylation modification level of Tau217), and D (combined model). The legend shows that model D has the largest area under the curve (AUC), indicating that the combined biomarker model has the best diagnostic accuracy. Detailed Implementation

[0022] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0023] It should be noted that, unless otherwise specified, the experimental methods described in the following implementation plan are conventional methods, and the reagents and materials mentioned are commercially available unless otherwise specified. Example 1: Discovery of AD-related stereoheterogeneous modification biomarkers based on human brain tissue samples 1. Sample and data preparation: Sample Source: The data analyzed in this study were obtained from a published public proteomics dataset (PRIDE database, accession number PXD020517). This dataset contains deep proteomics data (analyzed using liquid chromatography-tandem mass spectrometry LC-MS / MS) of cerebral cortex tissue from 49 AD patients and 42 healthy human donors. All samples underwent systematic neuropathological evaluation and Braak staging was determined.

[0024] Data acquisition: Download the corresponding raw mass spectrometry file (.raw format) and related metadata (including Braak staging) from the database.

[0025] 2. Proteomics data reanalysis workflow: Standard Database Search: First, the raw data underwent a standard database search analysis using MaxQuant software. Key parameter settings were as follows: precursor quality tolerance was 20 ppm for the first search and 4.5 ppm for the second; enzyme digestion was Trypsin / P, allowing a maximum of two missed cleavage sites; fixed modification was cysteine ​​alkylation (Carbamidomethyl, +57.021 Da); variable modifications included methionine oxidation, N-terminal acetylation, serine / threonine / tyrosine phosphorylation, lysine acetylation, and ubiquitination (GlyGly). False discovery rates (FDR) for both peptide and protein levels were set to 1%, controlled using a reverse database. The reference database for the search was the human UniProt protein database, with other parameters remaining at default. This step aimed to obtain baseline information for protein identification and quantification.

[0026] Heterogeneous Search Analysis (Core Step): All peptides derived from Tau protein (UniProtID: P10636-8) were extracted from the conventional database search results. A heterogeneous search algorithm based on spectral feature matching was then executed. For stereoisomerically modified peptides derived from Tau protein (2N4R isomer) stably identified in multiple samples, the relative modification level of their D-isomers was calculated. For identical sequences identified at different retention times, quantification and comparison were performed using primary mass spectrometry peak intensities. Since natural standard proteins are primarily composed of L-type amino acids, the higher intensity is assumed to be the L-configuration, and the lower intensity the D-configuration. The calculation formula is: Modification Level (%) = [Mass spectral peak intensity of D-isomer / (Mass spectral peak intensity of L-isomer + Mass spectral peak intensity of D-isomer)] × 100%. Modification levels are calculated using primary mass spectrometry peak intensities (the calculation of phosphorylation modification levels is also performed in this way). The output results include sequence information and quantitative data for all stereoisomerically modified peptides identified in each sample.

[0027] 3. Data post-processing and biomarker selection: Quantitative statistical analysis of modification levels: For the stereoisomerically modified peptides derived from Tau protein that were stably identified, the differences in modification levels between the AD group and the healthy control group, as well as between different Braak stage groups, were statistically analyzed.

[0028] Correlation analysis: Spearman rank correlation analysis was performed on the modification level of each peptide in each sample and its corresponding Braak stage value (0=0, I=1, II=2, III-IV=3.5, IV-V=4.5, V=5, V-VI=5.5, VI=6) to calculate the correlation coefficient (ρ) and significance p value.

[0029] Results: See attached. Figure 1As shown, a total of 19 stereoisomerically modified peptides derived from Tau protein were identified. Correlation analysis showed that ( Figure 2 The modification levels of the peptides “KLDLSNVQSK” (SEQ ID NO: 1) located in the microtubule-binding repeat region (MTBR) and “AKTDHGAEIVYK” (SEQ ID NO: 2) located in the C-terminal domain were correlated with Braak staging. Specifically: SEQ ID NO: 1: ρ = -0.62, p = 0.002. The mean modification level was approximately 38% in the healthy control group (Braak stage 0) and decreased to approximately 5% in the advanced AD group (Braak stage VI).

[0030] SEQ ID NO: 2: p = 0.65, p = 0.005. The mean modification level was approximately 5% in the healthy control group (Braak stage 0) and increased to approximately 24% in the advanced AD group (Braak stage VI).

[0031] Other peptides showed weak correlations (|ρ|<0.6) or failed the significance test. This indicates that SEQ ID NO: 1 and SEQ ID NO: 2 are highly specific molecular markers closely associated with the progression of Alzheimer's disease.

[0032] Example 2: Construction and Validation of AD Diagnostic Model 1. Model building: 1) Univariate analysis First, in the data sample, the ability of candidate biomarkers to distinguish AD from healthy controls and their association with Braak staging were evaluated separately.

[0033] Analysis: The associations between the stereoisomerization modification levels of SEQ ID NO: 1 (X1), SEQ ID NO: 2 (X2), and Tau217 (X3) and disease status (AD / control) and Braak stage (ordination variable) were calculated. The phosphorylation modification level of Tau217 (X3) was calculated as follows: Phosphorylation modification level (%) = Mass spectrum peak intensity of the isophosphorylated sequence / (Mass spectrum peak intensity of the phosphorylated sequence + Mass spectrum peak intensity of the non-phosphorylated sequence) × 100%.

[0034] Preliminary results: Analysis showed that the modification level of SEQ ID NO: 1 (X1) was significantly lower in the AD group than in the control group, the modification level of SEQ ID NO: 2 (X2) was significantly higher in the AD group than in the control group, and the phosphorylation modification level of Tau217 (X3) was significantly increased in the AD group. The first two were significantly correlated with Braak stage (p<0.001). However, there is still room for improvement in the diagnostic efficacy of single biomarkers (AUC between 0.75 and 0.88).

[0035] 2) Collinearity and complementarity tests Before constructing a multivariate model, it is necessary to ensure that there is no severe multicollinearity among the included independent variables in order to avoid model instability and distorted coefficient estimates.

[0036] Test method: Calculate the Pearson correlation coefficient between each pair of X1, X2 and X3, and calculate the variance inflation factor (VIF).

[0037] Results: The correlation coefficients between X1 and X2 were at a moderate level (|r|<0.6), and the correlation coefficient with X3 was also low (|r|<0.3). The VIF values ​​for all variables were less than 3, far below the commonly accepted collinearity threshold (VIF=5 or 10). This indicates that the pathological information carried by the three biomarkers is relatively independent, and that X1 and X2 exhibit opposite trends, providing complementary discriminative information in the model, thus meeting the prerequisite for constructing a stable multivariate model.

[0038] 3) Construction of a multivariate logistic regression model To comprehensively assess disease risk and capture potential interactions between biomarkers, a multivariate logistic regression model including main effects and interaction terms was constructed.

[0039] Model Specification: The disease stage of the samples (0=0, I=1, II=2, III-IV=3.5, IV-V=4.5, V=5, V-VI=5.5, VI=6) was used as the initial modeling objective. X1, X2, X3, and their pairwise interactions (X1X2, X1X3, X2X3) and even triple interactions (X1X2X3) were included in the candidate independent variable set. Considering the potential synergistic or antagonistic effects of different Tau protein modifications in AD pathology, the model formula structure is as follows: Logit(P(AD)) = β0 + β1*X1 + β2*X2 + β3*X3 + β4*X1*X2 + β5*X1*X3 + β6*X2*X3 + β7*X1*X2*X3 Where P(AD) is the probability that a sample is diagnosed with AD, β0 is the intercept, and β1 to β7 are the regression coefficients of each variable and the interaction term.

[0040] Model Training and Coefficient Determination: Both the training and test sets utilize the complete set of data from the aforementioned public proteomics dataset. On the training set, stepwise regression (combined with the AIC criterion) is used for feature selection to eliminate insignificant terms and prevent overfitting. The final simplified model is iteratively optimized using maximum likelihood estimation to obtain the coefficient (β) values ​​that maximize the probability of the observed data.

[0041] 2. Model Evaluation 1) Model performance evaluation Assessment method: Calculate the risk score for each sample in the test set according to the model formula, and use the score as the test variable to plot the receiver operating characteristic (ROC curve). Calculate the area under the curve (AUC) as the overall indicator of discrimination ability.

[0042] Evaluation results: The joint model of this invention (integrating X1, X2, and X3) performs exceptionally well in dataset validation, with an AUC of 0.99, indicating its ability to accurately distinguish between AD and healthy states.

[0043] 2) Compare model performance To objectively evaluate the superiority of the joint model of this invention, comparative diagnostic models based on only a single marker (X1, X2, or X3) were simultaneously evaluated on the same test set.

[0044] Comparative models: Using the same statistical framework (logistic regression), but using only the stereoisomerization level of SEQ ID NO: 1 (X1), the stereoisomerization level of SEQ ID NO: 2 (X2), and the phosphorylation level of Tau217 (X3) as independent variables to construct diagnostic models.

[0045] Comparison results: like Figure 3 The results show that the diagnostic efficacy of the combined model of the present invention (integrating X1, X2, and X3) is significantly better than any single biomarker model, verifying the superiority of the multi-biomarker combined strategy.

[0046] The present invention also provides the application of the above-mentioned biomarkers in the preparation of products for diagnosing or assisting in the diagnosis of Alzheimer's disease.

[0047] The present invention also provides the application of the above-mentioned biomarkers as targets in the preparation of products for treating Alzheimer's disease.

[0048] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A biomarker for diagnosing Alzheimer's disease, characterized in that, The biomarkers are the following peptides that have undergone stereoisomer modification: Peptide sequence 1: KLDLSNVQSK; Peptide sequence 2: AKTDHGAEIVYK; The stereoisomerization modification refers to the D-isomerization of at least one amino acid residue in peptide sequence 1 and / or peptide sequence 2, and the level of stereoisomerization modification of the biomarker is correlated with the Braak stage of neuropathology of Alzheimer's disease.

2. The biomarker as described in claim 1, characterized in that, The stereoisomerization level of the biomarkers was correlated with the Braak neuropathological stage of Alzheimer's disease, as follows: as the severity of the pathology increased, the stereoisomerization level of peptide sequence 1 decreased, while the stereoisomerization level of peptide sequence 2 increased.

3. The application of the biomarker as described in claim 1 in the preparation of products for diagnosing or assisting in the diagnosis of Alzheimer's disease.

4. The use of the biomarker as a target in the preparation of products for treating Alzheimer's disease, as described in claim 1.

5. The method for screening biomarkers as described in claim 1, characterized in that, Includes the following steps: Obtain proteomics mass spectrometry data from samples of different pathological stages of Alzheimer's disease and healthy control samples; The proteomics mass spectrometry data were searched using a heterogeneity search algorithm to identify peptides in the proteomics mass spectrometry data that have undergone amino acid isomerization modification. Stereoisomer-modified peptides derived from Tau protein were screened from the identification results; Calculate the modification level of stereoisomerically modified peptides derived from Tau protein in samples from different pathological stages of Alzheimer's disease; The correlation between the modification level of each stereoisomer modified peptide and the pathological stage of Alzheimer's disease was analyzed. Stereoisomer-modified peptides whose modification levels were significantly correlated with the pathological stage of Alzheimer's disease were identified as candidate biomarkers.

6. The method for screening biomarkers as described in claim 5, characterized in that, The correlation was assessed using Spearman rank correlation analysis, and peptides that met the correlation coefficient |ρ|>0.6 and p-value < 0.05 after multiple test correction were identified as candidate biomarkers.

7. A method for constructing a model for diagnosing Alzheimer's disease, characterized in that, Includes the following steps: Obtain proteomics mass spectrometry data from samples of different pathological stages of Alzheimer's disease and healthy control samples; The stereoisomerization modification level of peptide sequence 1 as described in claim 1, the stereoisomerization modification level of peptide sequence 2 as described in claim 1, and the phosphorylation modification level of Tau217 protein were calculated and statistically analyzed in each sample. Univariate analysis was performed on the stereoisomerization modification levels of peptide sequence 1, the stereoisomerization modification levels of peptide sequence 2, and the phosphorylation modification levels of Tau217 protein to assess their respective associations with Alzheimer's disease. The stereoisomerization levels of peptide sequence 1, peptide sequence 2, and phosphorylation levels of Tau217 protein were tested for collinearity to confirm that they met the conditions for constructing a multivariate statistical model. Using Alzheimer's disease stage as the dependent variable, and the stereoisomerization modification level of peptide sequence 1, the stereoisomerization modification level of peptide sequence 2, the phosphorylation modification level of Tau217 protein, and the interaction term as independent variables, a stepwise regression method was used for feature selection to construct a multivariate logistic regression model, and the multivariate logistic regression coefficients were determined by maximum likelihood estimation.

Images