Plasma metabolism marker combination for predicting human aging and application thereof

Abstract

The invention discloses a biomarker composition for predicting the biological age or aging state of an individual, and the biomarker composition comprises tetracosyl saturated fatty acid / tetracosyl monounsaturated fatty acid phosphatidylcholine (PC 24: 0 / 14: 1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), cysteine-glycine- The invention relates to a pharmaceutical composition for the preparation of a pharmaceutical composition, which comprises the following components in parts by weight: 20: 0 / 22: 5, 20 / docosapentaenoic acid monohexosyl sphingosine amide (HexCer 20: 0 / 22: 5), trigonelline, lithocholic acid glycine (Lithocholic acid glycine), and octadecanoic saturated fatty acid / octadecanoic monounsaturated fatty acid phosphatidylcholine (PC 18: 0 / 18: 1). The biomarker can be used for accurately characterizing the aging state of an organism, and provides a strategy for early screening, intervention evaluation, accurate health management and the like of related diseases of the elderly.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine, and specifically relates to the construction of a human plasma aging clock and the application of human plasma aging metabolic markers. Background Art

[0002] Individualized health management and early warning of geriatric diseases are becoming important research areas in life sciences and precision medicine. Traditional aging assessment methods rely primarily on physiological indicators, imaging examinations, or subjective questionnaires. These methods lack quantitative standards and cannot accurately reflect the degree of aging in individual physiological functions. To address this, researchers have proposed the concept of "biological age" and are attempting to establish a predictive model, the "Aging Clock," using a series of biomarkers.

[0003] Various aging clock models have been proposed, with the DNA methylation clock being a prominent example, capable of predicting individual biological age to a certain extent. However, these models suffer from high sample costs, limited tissue availability, and a lack of understanding at the level of metabolic function. In recent years, the rapid development of metabolomics and lipidomics technologies has made it possible to obtain large-scale metabolic profiles from body fluids such as plasma. Numerous studies have demonstrated that individuals develop stable patterns of metabolite changes during aging. These changes not only reflect their intrinsic physiological state but are also closely associated with various age-related diseases, such as cardiovascular disease and neurodegenerative diseases. In particular, certain lipid molecules, such as phospholipids, sphingolipids, and bile acids, have been found to show systemic upregulation or downregulation with age, suggesting potential as biomarkers of aging. Metabolomics and lipidomics technologies offer advantages such as low cost, clinical accessibility, and robust signal analysis, enabling high-throughput analysis of plasma samples to directly reflect individual physiological status at the level of functional metabolites.

[0004] While some studies have attempted to construct age prediction models using metabolites, most existing methods rely on a single modeling strategy or fail to fully consider the expression trends of markers across different age groups. For example, many models select variables based solely on model coefficients, ignoring whether they exhibit accumulation characteristics in the elderly population or are linearly correlated with age. This can result in limited stability and generalizability of the selected features across clinical populations. Despite the crucial role of lipid metabolism in the aging process, most existing models focus primarily on polar metabolites, with limited systematic analysis of lipid markers. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method for detecting the metabolite and lipid content in human plasma, screening out metabolite and lipid combinations that increase with age, are significantly enriched in the elderly population, and are sensitive to biological age prediction, and constructing a high-precision model that can be used for biological age assessment to achieve accurate assessment of the degree of aging of individuals.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] 1. A biomarker composition for predicting an individual's biological age or aging state, wherein the biomarkers include tetracosyl saturated fatty acid / tetracosyl monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosyl sphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and octadecyl saturated fatty acid / octadecyl monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1).

[0008] PC 24:0 / 14:1, twenty-four carbon saturated fatty acid / fourteen carbon monounsaturated fatty acid phosphatidylcholine, is a glycerophospholipid that contains a long-chain saturated fatty acid and a medium-chain unsaturated fatty acid. It is closely related to cell membrane stability, membrane fluidity changes, and aging-related lipid remodeling processes.

[0009] Cys-Gly-Cys, a cysteine-glycine-cysteine tripeptide, is a small molecule peptide involved in glutathione metabolism and an important intracellular antioxidant peptide. It effectively scavenges reactive oxygen species (ROS), maintains the reduced state of cells, and prevents cell damage caused by oxidative stress. Cysteine residues are rich in thiols (–SH), which have metal ion chelation, antioxidant, and anti-toxic properties.

[0010] HexCer 20:0 / 22:5, eicosapentaenoic acid monohexosylsphingosine amide, is a complex sphingolipid that is widely present in the cell membranes of the nervous system and is closely related to processes such as inflammation, immune regulation and cell apoptosis.

[0011] Trigonelline, also known as trigonelline, is produced by niacin metabolism in the human body and has multiple biological functions, including neuroprotection, antioxidant, anti-inflammatory, and regulation of glucose and lipid metabolism.

[0012] Lithocholylglycine is a metabolite formed by the conjugation of lithocholic acid and glycine. It is one of the end products of bile acid metabolism and is associated with intestinal flora balance and hepatobiliary metabolism.

[0013] PC 18:0 / 18:1, also known as phosphatidylcholine, consists of an 18-carbon saturated fatty acid (stearic acid, 18:0) and an 18-carbon monounsaturated fatty acid (oleic acid, 18:1). It is one of the most common cell membrane lipids and is highly stable. It participates in membrane structure maintenance and the synthesis of signaling molecule precursors, and is closely related to energy metabolism and cell survival.

[0014] 2. A kit comprising reagents for detecting the content of biomarkers in a sample, and a label; wherein the biomarkers are 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1);

[0015] Preferably, the kit is used to predict an individual's biological age or aging state.

[0016] 3. The kit according to item 2, wherein the label records a formula for predicting an individual's biological age or aging state: Z = 5359.0277 + expression level of PC 24:0 / 14:1 × 928.22902 + expression level of Cys-Gly-Cys × 515.350141 + expression level of HexCer 20:0 / 22:5 × 466.9382 + expression level of Trigonelline × 203.360117 + expression level of Lithocholylglycine × (-112.135851) + expression level of PC 18:0 / 18:1 × (-452.99724);

[0017] Optionally, the expression level of the biomarker is detected by LC-MS, preferably the expression level is the value obtained by normalizing the relative peak value to the total peak value after LC-MS detection and performing log10;

[0018] Optionally, the label records that when the Z value is > 0, the individual is judged to be in an "aging" state; when the Z value is ≤ 0, the individual is judged to be in a "non-aging" state; the larger the Z value, the greater the degree to which the individual's metabolic characteristics deviate from the youthful state, and the higher the degree of physiological aging.

[0019] 4. The kit according to claim 2, wherein the sample is plasma, and preferably the actual age of the individual is 0-84.

[0020] 5. Use of a reagent for determining the content of a biomarker in preparing a kit for predicting the biological age or aging state of an individual; wherein the biomarkers are 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1).

[0021] 6. Use of a reagent for detecting the content of a biomarker in preparing a kit, wherein the biomarker is 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1), wherein the kit is used for:

[0022] (1) Identification of individual aging status and health risk assessment;

[0023] (2) Dynamic monitoring of the effects of aging interventions (e.g., medication, nutrition, lifestyle, etc.);

[0024] (3) Population screening and aging trend research; or

[0025] (4) Companion diagnostics in the development of clinical anti-aging therapies.

[0026] 7. A system for aging prediction, comprising:

[0027] (1) Biological plasma sample collection module, used for obtaining samples;

[0028] (2) A metabolite content determination module, which is used to determine the content of at least 6 aging metabolic markers in the obtained plasma sample, wherein the 6 aging metabolic markers are: 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), 20-carbon / 22-carbon pentaenoic acid monohexosyl sphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine (Lithocholylglycine), 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1);

[0029] (3) Metabolite aging assessment module, which is used to bring the measured metabolite expression levels into the assessment algorithm to assess the degree of human aging.

[0030] 8. The system according to item 7, wherein the evaluation algorithm is: Z = 5359.0277 + expression level of PC 24:0 / 14:1 × 928.22902 + expression level of Cys-Gly-Cys × 515.350141 + expression level of HexCer 20:0 / 22:5 × 466.9382 + expression level of Trigonelline × 203.360117 + expression level of Lithocholylglycine × (-112.135851) + expression level of PC 18:0 / 18:1 × (-452.99724), wherein the expression levels are the values of the relative peak values after mass spectrometry detection after total peak normalization and log10, wherein the Z value is used to indicate the aging status of an individual. When the Z value > 0, the individual is judged to be in the "aging" state; when the Z value ≤ 0 When the Z value is greater, the individual is judged to be in a "non-aging" state; the larger the Z value, the greater the degree to which the individual's metabolic characteristics deviate from the youthful state, and the higher the degree of physiological aging.

[0031] In another aspect, the present invention provides a method for predicting aging based on plasma metabolites and lipids, comprising the following steps:

[0032] (1) Collect samples from people of different age groups (preferably covering 0 to 84 years old) and measure the content of metabolites and lipids in the samples.

[0033] (2) Candidate biomarker screening: Metabolites and lipids are screened through the following three dimensions:

[0034] a. Select metabolites / lipids that are significantly upregulated in the elderly group ≥65 years old

[0035] b. Select metabolites / lipids that increase linearly with age across all age groups;

[0036] c. Combined with the prediction model established by elastic net regression, the metabolites / lipids with significant weights in predicting age were screened.

[0037] (3) The intersection of the three types of markers above was taken to form a set of metabolites / lipids that were significantly correlated with age and enriched in the elderly. Multiple rounds of AUC scoring were then performed based on different numbers and combinations to screen out the combination with the strongest predictive ability in distinguishing between the non-elderly group (<40 years old) and the elderly group (>65 years old). This formed a set of aging prediction scoring systems that included six key metabolites and lipids.

[0038] The method is characterized in that the metabolites and lipids are obtained by performing non-targeted metabolomics and lipidomics detection on human plasma samples, and the detection method is liquid chromatography-mass spectrometry (LC-MS);

[0039] In step (2) a, the metabolites / lipids that were significantly upregulated in the elderly group aged ≥65 years were screened by statistical analysis (e.g., differential analysis, significance test). The difference in significance of the metabolites / lipids met the FDR-corrected p-value < 0.05, and the average abundance in the elderly group was significantly higher than that in other age groups.

[0040] The method described in step (2) b for selecting metabolites / lipids that linearly increase with age across all age groups uses a linear regression model, with an FDR-corrected p-value of < 0.05 as the significance criterion;

[0041] The method described in step (2) c for screening metabolites / lipids with significant weights for predicting age is based on the elastic net regression algorithm. The input is the relative abundance value of metabolites and / or lipids, and the output is the individual predicted age. The model performance is evaluated using 10-fold cross-validation, and the Pearson correlation coefficient R between the predicted age and the actual age is greater than 0.9.

[0042] The area under the ROC curve (AUC) of the combination in the above method in distinguishing the non-elderly group (<40 years old) from the elderly group (>65 years old) is not less than 0.90.

[0043] Each age-related metabolite / lipid has a commonly understood meaning. In certain embodiments, the exemplary PubChem numbers corresponding to each age-related metabolite / lipid are as follows:

[0044]

[0045] The six metabolites / lipids showed good predictive ability when the model was constructed with a single metabolite as the characteristic variable to predict the aging status. The corresponding areas under the receiver operating characteristic curves (AUCs) are shown below:

[0046]

[0047] The present invention has important value in clinical application:

[0048] (1) The metabolite and lipid markers screened by this invention have good age discrimination capabilities after rigorous statistical analysis and model screening. They have shown high predictive performance in multiple groups of people and are suitable for individual assessment at different age stages.

[0049] (2) The detection method is simple and has strong applicability: The present invention relies on plasma metabolomics and lipidomics data, and can achieve high-throughput detection through non-invasive sampling (such as venous blood), which can quickly obtain a large amount of data. It has the advantages of simple sample collection, strong repeatability, low cost, and fast speed.

[0050] (3) Through quantitative assessment of individual aging status, this invention can be used to guide individualized intervention, anti-aging strategy formulation and long-term health management, and has clinical guidance value.

[0051] The term "biological age", also known as physiological age, refers to the measurement of the degree of aging of an individual's physiological functions by evaluating the functional status and metabolic levels of the body's cells, tissues and organs.

[0052] The term "chronological age", also known as calendar age, refers to the age calculated from the date of an individual's birth based on the passage of time, depending only on the age span since birth. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The metabolites and lipid quantities that are significantly enriched in the elderly population (≥65 years old).

[0054] Figure 2 Figure 2 shows the metabolite / lipid expression pattern that increases linearly with actual age in the linear model.

[0055] Figure 3 Graph for selecting the optimal clock in the elastic net model.

[0056] Figure 4 Figure 2 is the predictive ability graph of the optimal clock, showing the linear relationship between predicted age and actual age. R is the Pearson correlation coefficient, and MAE is the mean absolute error.

[0057] Figure 5 Venn diagram of the metabolite / lipid sets obtained for the three categories of screening criteria.

[0058] Figure 6 This is the receiver operating characteristic curve of the final combined model.

[0059] Figure 7 are the components of the final combined model. DETAILED DESCRIPTION

[0060] The embodiments of the present invention will be described in detail below with reference to examples. It should be understood that the following examples are only used to illustrate the present invention, rather than to limit the scope of protection of the present invention.

[0061] The human plasma samples used in this study were obtained from individuals undergoing routine health checkups at the Health Management Center of the First Affiliated Hospital of the University of Science and Technology of China. All subjects underwent a systematic physical examination to ensure they were healthy and free of major medical conditions, including but not limited to malignant tumors, autoimmune diseases, cardiovascular and cerebrovascular diseases, diabetes, and other metabolic diseases.

[0062] Example 1: Collection and analysis of plasma metabolites and lipids

[0063] This example aims to obtain comprehensive metabolite and lipid profile data by systematically performing non-targeted metabolomics and lipidomics testing on plasma samples of healthy people of different age groups, providing basic data support for subsequent metabolite screening and model building.

[0064] (1) Subject grouping and sample collection

[0065] A total of 136 healthy volunteers without major underlying medical conditions were recruited, ranging in age from birth to old age (0-84 years). They were categorized by chronological age as follows: neonatal period (0 years): n = 33; early childhood (1–6 years): n = 17; middle childhood (7–12 years): n = 14; adolescence (13–17 years): n = 11; early adulthood (18–40 years): n = 18; middle adulthood (41–64 years): n = 16; and old age (≥65 years): n = 27. Because all participants were healthy volunteers without major underlying medical conditions, chronological age was equated with biological age. Peripheral venous blood samples were collected from each participant in the early morning fasting state using plasma tubes containing EDTA anticoagulant. Within 30 minutes of collection, the samples were centrifuged (3000 rpm, 10 minutes, 4°C) to separate plasma, which was immediately stored at −80°C until metabolite and lipid analysis. To ensure the homogeneity of the research subjects and the reliability of the data, all included individuals met the following inclusion criteria:

[0066] 1. The actual age must be between 0 and 84 years old, with a clear date of birth;

[0067] 2. Physical examination results show that all blood tests, liver and kidney function, blood lipids, fasting blood sugar and other indicators are within normal range;

[0068] 3. No history of major chronic diseases;

[0069] 4. No recent acute infection;

[0070] 5. Sign the informed consent form and agree that their blood samples will be used for scientific research purposes.

[0071] All samples were collected under the guidance of professional medical staff. Blood was collected in the early morning on an empty stomach. Samples were processed immediately and stored at -80°C to ensure sample quality stability and consistency, providing a reliable basis for subsequent metabolomics analysis.

[0072] (2) Sample processing and mass spectrometry detection

[0073] A non-targeted combined metabolomics and lipidomics platform based on liquid chromatography-high-resolution mass spectrometry (LC-HRMS) was employed. Data quality was ensured through a standardized quality control process, including the insertion of a quality control sample (pool QC) for every three samples; correction for retention time drift and mass-to-charge ratio variation; batch calibration using QC support vector regression (QC-SVR); and monitoring the reproducibility of QC samples (RSD < 30%) to ensure assay stability.

[0074] (3) Data processing and standardization

[0075] Raw mass spectrometry data were converted to MzXML format using ProteoWizard MSConvert (v.3.0.6428) and processed in XCMS Online Software (v.3.7.1). Fragmentation patterns were compared with the spectral database, and all identified metabolites met at least MSI level 2 criteria. Lipids were matched and identified using LipidSearch (v.4.0). Normalized intensities were obtained by normalizing to the total peak (dividing the raw intensity by the sum of each sample). Metabolite / lipid values are log10 normalized to the total peak.

[0076] After the above processing, the metabolite / lipid expression profile of each sample under a unified standard is finally obtained, providing high-quality input data for subsequent metabolite screening and model building.

[0077] Statistical analysis was performed using the stats package (version 4.1.2) in R software. A two-sided Student's t-test was used to compare the elderly group aged ≥65 years with samples from all other age groups, yielding preliminary significant differences. The Benjamini–Hochberg (BH) method was used to correct the obtained P values. The screening criterion was a corrected P value < 0.05, resulting in a group of metabolites / lipids that were considered to have significant enrichment characteristics in the elderly population ( Figure 1 ).

[0078] Example 2: Screening of metabolites and lipids that increase with chronological age based on a linear model

[0079] The standardized metabolite and lipid data were used as independent variables (X), and the actual age was used as the dependent variable (Y). A linear regression model was established ( Figure 2 Because chronological age is a continuous variable, this step accurately captures trends in metabolite changes with chronological age. The model formula is: Expression = α + β1 × chronological age + β2 × sex + ε. α represents the intercept, β represents the regression coefficient, and ε represents the residual. Model fitting was performed using the lm function in R software, and regression analysis was performed using Type II analysis of variance (Anova) in the car package (v.3.1.2). The screening criterion was an adjusted P value < 0.05 for the chronological age variable in the model. Selected metabolites / lipids were considered to increase linearly with chronological age. This yielded a set of metabolites and lipids that increase linearly with chronological age over the lifespan.

[0080] Example 3: Identifying metabolites and lipids that contribute significantly to biological age prediction based on the elastic net regression model

[0081] Elastic Net Regression was used to construct a metabolic clock model for predicting biological age. The model was built using the subject's actual age as the dependent variable and all metabolite / lipid expression data as the independent variables. Multiple bootstrap sampling was performed using the Rsample() function, with 80% of the samples drawn from the original data with replacement. The samples were then subjected to variable selection (α) using the R glmnet package. A 10× cross-validation was performed using the glmnet function to control model complexity and prevent overfitting ( Figure 3 After obtaining the regression coefficients of all metabolites / lipids, the metabolites / lipids with a coefficient of 0 are removed, and the remaining metabolites and lipids are those that contribute significantly to the prediction of biological age ( Figure 4 ). A set of metabolites and lipids with non-zero regression coefficients in the model and significant contributions to the prediction results were obtained.

[0082] Example 4: Metabolite / lipid combination optimization and final model construction

[0083] The core metabolites obtained in Examples 1-3 were intersected with lipids ( Figure 5 ), yielding a set of characteristic metabolites that increase linearly with chronological age, ultimately accumulating in the elderly population and being sensitive to biological age prediction. Core metabolites and lipids were combined and classified, and binary classification models for different combinations (non-elderly population <40 years vs. elderly population ≥65 years) were constructed using logistic regression. The training set:test set ratio was 8:2. The AUC values were calculated for each model to evaluate predictive performance.

[0084] After multiple rounds of screening, cross-combination and AUC evaluation, the optimal prediction combination of 6 components was finally determined: 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine (Lithocholylglycine), 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1) ( Figure 7 ).

[0085] The final prediction formula was constructed as follows, and the Z value was calculated to determine whether an individual was in an aging state: Z = 5359.0277 + expression level of PC 24:0 / 14:1 × 928.22902 + expression level of Cys-Gly-Cys × 515.350141 + expression level of HexCer 20:0 / 22:5 × 466.9382 + expression level of Trigonelline × 203.360117 + expression level of Lithocholylglycine × (-112.135851) + expression level of PC 18:0 / 18:1 × (-452.99724), where the expression level is the value after the relative peak value after LC-MS detection is normalized to the total peak and log10.

[0086] Subsequently, 28 samples of subjects different from the training set were extracted as the test set to verify the above model. To ensure the homogeneity of the research subjects and the reliability of the data, all individuals included in the test set met the following inclusion criteria:

[0087] 1. The actual age must be between 0 and 84 years old, with a clear date of birth;

[0088] 2. Physical examination results show that all blood tests, liver and kidney function, blood lipids, fasting blood sugar and other indicators are within normal range;

[0089] 3. No history of major chronic diseases;

[0090] 4. No recent acute infection;

[0091] 5. Sign the informed consent form and agree that their blood samples will be used for scientific research purposes.

[0092] All samples were collected under the guidance of professional medical staff. Blood was collected in the early morning on an empty stomach. Samples were processed immediately and stored at -80°C to ensure sample quality stability and consistency, providing a reliable basis for subsequent metabolomics analysis.

[0093] Depend on Figure 6 and Figure 7 It can be seen that the prediction AUC of the combination of the six markers is 0.985 ( Figure 6 ), which is much higher than the existing technology of 0.8 and also higher than the AUC of a single marker, indicating that the prediction combination of the present invention can quickly and accurately determine the degree of aging in individuals. The test set Z value results are as follows:

[0094] Table 1 Test set Z value results

[0095]

[0096] The Z value determination rule is:

[0097] Z ≤ 0: judged as “non-aging” state;

[0098] Z > 0: Determined as "aging" state; the higher the Z value, the more the individual's metabolic characteristics deviate from the young reference state, and the higher the degree of biological aging.

[0099] Example 5: Practical application verification

[0100] Another batch of 9 plasma samples from an independent population, different from those in the above examples, were selected to further validate the method of the present invention. To ensure the homogeneity of the research subjects and the reliability of the data, all included individuals met the following inclusion criteria:

[0101] 1. The actual age must be between 0 and 84 years old, with a clear date of birth;

[0102] 2. Physical examination results show that all blood tests, liver and kidney function, blood lipids, fasting blood sugar and other indicators are within normal range;

[0103] 3. No history of major chronic diseases;

[0104] 4. No recent acute infection;

[0105] 5. Sign the informed consent form and agree that their blood samples will be used for scientific research purposes.

[0106] All samples were collected under the guidance of professional medical staff. Blood was collected in the early morning on an empty stomach. Samples were processed immediately and stored at -80°C to ensure sample quality stability and consistency, providing a reliable basis for subsequent metabolomics analysis.

[0107] The actual age of the independent population was equivalent to their biological age. The plasma samples were subjected to the same metabolomics and lipidomics mass spectrometry analysis and data preprocessing (total peak normalization and log10 transformation) and their values were entered into the formula constructed in Example 4. The validation data obtained were as follows:

[0108] Table 2 Verification results of independent population Z value

[0109]

[0110] Based on the calculated Z value, it was found that the model can effectively distinguish the age groups of the samples with an accuracy of 100%, which is highly consistent with their actual age and physical signs, verifying the application stability and promotion potential of the model in real scenarios.

Claims

1. A biomarker composition for predicting an individual's biological age or aging state, wherein: The biomarkers include twenty-tetradecanoic acid saturated fatty acid / fourteen-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer20:0 / 22:5), trigonelline, lithocholic acid glycine (Lithocholylglycine), and eighteen-carbon saturated fatty acid / eighteen-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1).

2. A kit comprising reagents for detecting the content of biomarkers in a sample, and a label; wherein the biomarkers are 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1); Preferably, the kit is used to predict an individual's biological age or aging state.

3. The kit according to claim 2, wherein The label records a formula for predicting an individual's biological age or aging state: Z = 5359.0277 + expression level of PC 24:0 / 14:1 × 928.22902 + expression level of Cys-Gly-Cys × 515.350141 + expression level of HexCer 20:0 / 22:5 × 466.9382 + expression level of Trigonelline × 203.360117 + expression level of Lithocholylglycine × (-112.135851) + expression level of PC 18:0 / 18:1 × (-452.99724); Optionally, the expression level of the biomarker is detected by LC-MS, preferably the expression level is the value obtained by normalizing the relative peak value to the total peak value after LC-MS detection and performing log10; Optionally, the label records that when the Z value is greater than 0, the individual is judged to be in an "aging" state; when the Z value is less than or equal to 0, the individual is judged to be in a "non-aging" state; the larger the Z value, the greater the degree to which the individual's metabolic characteristics deviate from the youthful state, and the higher the degree of physiological aging.

4. The kit according to claim 2, wherein The sample is plasma.

5. Use of a reagent for determining the content of a biomarker in preparing a kit, wherein the kit is used to predict the biological age or aging state of an individual; wherein the biomarkers are 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), eicosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1).

6. Use of a reagent for detecting the content of a biomarker in preparing a kit, wherein the biomarker is 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), 20-carbon / docosapentaenoic acid monohexosylsphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine, and 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1), wherein the kit is used for: (1) Identification of individual aging status and health risk assessment; (2) Dynamic monitoring of the effects of aging interventions (e.g., medication, nutrition, lifestyle, etc.); (3) Population screening and aging trend research; or (4) Companion diagnostics in the development of clinical anti-aging therapies.

7. A system for aging prediction, comprising: (1) Biological plasma sample collection module, used for obtaining samples; (2) A metabolite content determination module, which is used to determine the content of at least 6 aging metabolic markers in the obtained plasma sample, wherein the 6 aging metabolic markers are: 24-carbon saturated fatty acid / 14-carbon monounsaturated fatty acid phosphatidylcholine (PC 24:0 / 14:1), cysteine-glycine-cysteine tripeptide (Cys-Gly-Cys), 20-carbon / 22-carbon pentaenoic acid monohexosyl sphingosine amide (HexCer 20:0 / 22:5), trigonelline, lithocholic acid glycine (Lithocholylglycine), 18-carbon saturated fatty acid / 18-carbon monounsaturated fatty acid phosphatidylcholine (PC 18:0 / 18:1); (3) Metabolite aging assessment module, which is used to bring the measured metabolite expression levels into the assessment algorithm to assess the degree of human aging.

8. The system according to claim 7, wherein the evaluation algorithm is: Z = 5359.0277 + expression level of PC 24:0 / 14:1 × 928.22902 + expression level of Cys-Gly-Cys × 515.350141 + expression level of HexCer 20:0 / 22:5 × 466.9382 + expression level of Trigonelline × 203.360117 + expression level of Lithocholylglycine × (-112.135851) + expression level of PC 18:0 / 18:1 × (-452.99724), wherein the expression levels are the values of the relative peak values after mass spectrometry detection, normalized to the total peak value, and log10, and the Z value is used to indicate the aging status of an individual. When the Z value is > 0, the individual is judged to be in the "aging" state; when the Z value is ≤ 0, the individual is judged to be in the "aging" state. When the Z value is greater, the individual is judged to be in a "non-aging" state; the larger the Z value, the greater the degree to which the individual's metabolic characteristics deviate from the youthful state, and the higher the degree of physiological aging.

Images