Biomarkers, analytical methods, kits and applications for early diagnosis of DKD

Abstract

The present invention discloses a biomarker for early diagnosis of DKD, which can be effectively used for monitoring the course of diabetic nephropathy, has high sensitivity, and accurate and reliable results. It includes a combination of one or more of the following proteins: ARL3, C7of25, and PSMC3. It also provides an analysis method for the biomarker for early diagnosis of DKD, a kit using the biomarker for early diagnosis of DKD, and applications.

Description

Technical Field

[0001] The present invention relates to the technical field of biomedicine, and in particular to a biomarker for early diagnosis of DKD, an analysis method for the biomarker for early diagnosis of DKD, a kit using the biomarker for early diagnosis of DKD, and applications. Background Art

[0002] Diabetic kidney disease (DKD) is a serious microvascular complication characterized by persistent abnormal urine albumin and / or decreased estimated glomerular filtration rate (eGFR) due to diabetes, affecting approximately 30% of patients with type 1 and type 2 diabetes. According to a 2021 report by the International Diabetes Federation, the global prevalence of diabetes is expected to continue to rise by 2045, ultimately affecting 780 million people. With this trend, the burden of DKD and diabetes-related renal failure is also expected to gradually increase. The aging population and the decline in diabetes-related mortality have further exacerbated the prevalence of DKD.

[0003] Current methods for identifying DKD progression include measuring urinary albumin excretion rate, urine protein / creatinine ratio (UACR), eGFR, and detecting markers of inflammation and oxidative stress in serum and urine. Although these are the gold standards for DKD staging, they have some limitations, such as the large number of indicators required for evaluation and the high variability of urinary albumin content, which is easily affected by external factors. These limitations reduce the reliability of diagnosis, not to mention the biomarkers for detecting disease progression early in the diabetic stage. Therefore, more sensitive, reliable and simple biomarker identification methods are needed.

[0004] With the advancement of analytical technology, proteomics can identify more differential proteins, reveal potential mechanisms of disease progression, and provide new biomarkers and therapeutic targets for future diagnosis and treatment. At present, research on identifying DKD biomarkers is relatively limited. Some studies mixed and retested plasma samples from healthy controls, patients with diabetes, early diabetic nephropathy, and late diabetic nephropathy, which inevitably affected the accuracy of the results. Recent studies have also reported biomarkers in renal tissue of DKD patients, but since most DKD patients do not undergo renal biopsy, such studies are more suitable for identifying therapeutic targets rather than identifying biomarkers of disease progression. Summary of the invention

[0005] In order to overcome the defects of the prior art, the technical problem to be solved by the present invention is to provide a biomarker for early diagnosis of DKD, which can be effectively used for monitoring the course of diabetic nephropathy, has high sensitivity, and produces accurate and reliable results.

[0006] The technical solution of the present invention is: a biomarker for early diagnosis of DKD, which includes a combination of one or more of the following proteins: ARL3, C7orf25, and PSMC3.

[0007] The beneficial effects of the present invention are: it can be effectively used for monitoring the course of diabetic nephropathy, has high sensitivity, and the results are accurate and reliable. It has a high diagnostic value in distinguishing between health and diabetes, and diabetic nephropathy stages, and can be used as a potential biomarker for early diagnosis and staging evaluation of the disease.

[0008] Also provided is a method for analyzing biomarkers for early diagnosis of DKD, comprising the following steps:

[0009] (1) Data collection: Disease indicators of the subjects were collected, and the subjects were divided into a healthy control group (HC) and a disease group. The disease group was further divided into a diabetes group (DM) and a diabetic nephropathy group (DKD). Proteomic analysis was performed on the serum samples of the healthy control group (HC), the diabetes group (DM), and the diabetic nephropathy group (DKD);

[0010] (2) Serum separation: Blood is collected from the subject’s vein and processed within 12 hours to separate serum;

[0011] (3) Enrichment of low-abundance proteins in serum samples;

[0012] (4) Data independent acquisition DIA mass spectrometry analysis: Use Thermo ScientificTM OrbitrapTM AstralTM and Vanquish Neo UHPLC systems to perform gun-type proteomics and DIA experiments and construct spectral libraries;

[0013] (5) Database search: The default factory settings of Spectronaut software were used for searching. The DIA data were analyzed using Spectronaut software and the spectral library constructed in step (4) was searched. The filtering criteria for data processing and protein analysis were: Score Sequest HT > 0, unique peptide count ≥ 1, localization probability ≥ 0.75 or Delta score ≥ 8. Missing value filtering and interpolation were then performed to obtain reliable proteins / sites. These reliable proteins / sites were then median normalized and log2 transformed.

[0014] (6) Differentially expressed proteome profiling and pathway analysis: Identify proteins expressed in the human proteome, perform principal component analysis (PCA) to cluster samples, calculate protein fold-values, set logFC >2 or <0.5 and false discovery rate adjusted P value <0.05 as criteria for filtering differentially expressed proteins, annotate all identified proteins using GO and KEGG pathways, further perform GO and KEGG enrichment analysis on differentially expressed proteins, and use String to perform protein-protein interaction analysis;

[0015] (7) Analysis and validation of biomarkers: The differentially expressed proteins between the healthy control group HC, the diabetic group DM, and the diabetic nephropathy group DKD were intersected to obtain a total of 40 intersection proteins. These intersection proteins were intersected with the kidney highly expressed protein data to obtain 3 proteins. The ROC curve was then used to perform sensitivity and specificity analysis to verify the results.

[0016] A kit for biomarkers for early diagnosis of DKD is also provided, wherein the biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, PSMC3, and the sample is serum.

[0017] Also provided is the use of biomarkers for early diagnosis of DKD, wherein the biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, PSMC3, and are used for dynamic monitoring during the development of diabetic nephropathy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Shown is a basic flow chart of the present invention.

[0019] Figure 2 This is the PCA analysis and volcano plot analysis between groups. Figure 2 a- Figure 2 c PCA analysis of differential proteins among DM-HC, DKD-HC, and DKD-DM; Figure 2 d- Figure 2 f is the PCA analysis volcano plot analysis of differential proteins among DM-HC, DKD-HC, and DKD-DM.

[0020] Figure 3 This is the differential protein status among the three groups of DM-HC, DKD-HC, and DKD-DM, and the differential protein status after taking the intersection of the highly expressed kidney proteins downloaded from The human protein atlas.

[0021] Figure 4 Figure 2 shows the ROC analysis of ARL3, C7orf25 and PSMC3 at different stages in the discovery set data. Figure 4 a- Figure 4 c is the ROC curve analysis results of the three proteins in DM-HC, DKD-HC, and DKD-DM.

[0022] Figure 5 This is the ROC analysis of ARL3, C7orf25 and PSMC3 at different stages in the validation set data. Figure 5 a- Figure 5 c is the ROC curve analysis results of the three proteins in DM-HC, DKD-HC, and DKD-DM. DETAILED DESCRIPTION

[0023] A biomarker for early diagnosis of DKD, comprising a combination of one or more of the following proteins: ARL3, C7orf25, and PSMC3.

[0024] The beneficial effects of the present invention are: it can be effectively used for monitoring the course of diabetic nephropathy, has high sensitivity, and the results are accurate and reliable. It has a high diagnostic value in distinguishing between health and diabetes, and diabetic nephropathy stages, and can be used as a potential biomarker for early diagnosis and staging evaluation of the disease.

[0025] like Figure 1 As shown, a method for analyzing biomarkers for early diagnosis of DKD is also provided, which comprises the following steps:

[0026] (1) Data collection: Disease indicators of the subjects were collected, and the subjects were divided into a healthy control group (HC) and a disease group. The disease group was further divided into a diabetes group (DM) and a diabetic nephropathy group (DKD). Proteomic analysis was performed on the serum samples of the healthy control group (HC), the diabetes group (DM), and the diabetic nephropathy group (DKD);

[0027] (2) Serum separation: Blood is collected from the subject’s vein and processed within 12 hours to separate serum;

[0028] (3) Enrichment of low-abundance proteins in serum samples;

[0029] (4) Data independent acquisition DIA mass spectrometry analysis: Use Thermo ScientificTM OrbitrapTM AstralTM and Vanquish Neo UHPLC systems to perform gun-type proteomics and DIA experiments and construct spectral libraries;

[0030] (5) Database search: The default factory settings of Spectronaut software were used for searching. The DIA data were analyzed using Spectronaut software and the spectral library constructed in step (4) was searched. The filtering criteria for data processing and protein analysis were: Score Sequest HT > 0, unique peptide count ≥ 1, localization probability ≥ 0.75 or Delta score ≥ 8. Missing value filtering and interpolation were then performed to obtain reliable proteins / sites. These reliable proteins / sites were then median normalized and log2 transformed.

[0031] (6) Differentially expressed proteome profiling and pathway analysis: Identify proteins expressed in the human proteome, perform principal component analysis (PCA) to cluster samples, calculate protein fold-values, set logFC >2 or <0.5 and false discovery rate adjusted P value <0.05 as criteria for filtering differentially expressed proteins, annotate all identified proteins using GO and KEGG pathways, further perform GO and KEGG enrichment analysis on differentially expressed proteins, and use String to perform protein-protein interaction analysis;

[0032] (7) Analysis and validation of biomarkers: The differentially expressed proteins between the healthy control group HC, the diabetic group DM, and the diabetic nephropathy group DKD were intersected to obtain a total of 40 intersection proteins. These intersection proteins were intersected with the kidney highly expressed protein data to obtain 3 proteins. The ROC curve was then used to perform sensitivity and specificity analysis to verify the results.

[0033] Preferably, in step (1), the disease indicators include: age, weight, glycosylated hemoglobin, creatinine, urea, uric acid, 24-hour urine albumin, eGFR, urine red blood cells, urine white blood cells, cholesterol, triglycerides, urine ACR, total protein, ALT, AST, evaluation of serious diseases and commonly used drugs; the standard for DM is 24-hour urine protein < 0.15 g, and the standard for DKD is 24-hour urine protein > 0.5 g.

[0034] Preferably, in step (2), the serum is separated by centrifugation at 3000 g for 10 minutes and stored at −80°C until testing.

[0035] Preferably, in step (3), the low-abundance proteins in the serum sample are enriched using the EasyPept DeepP low-abundance protein enrichment and pretreatment kit, 1 mg of magnetic nanoparticle suspension is taken, the supernatant is removed by magnetic separation, after multiple washings, the magnetic beads are resuspended, 100 µL of serum / plasma is added, and the mixture is mixed at 37°C for 1 hour using a flip mixer, the supernatant is removed by magnetic separation, 300 µL of washing solution is added, and the mixture is washed by shaking for 5 minutes, the washing is repeated three times, the protein is converted into peptides by enzymatic hydrolysis, reduction alkylation and desalting are performed, and the total peptide concentration is determined by the nanodroplet method.

[0036] Preferably, in step (4), the sample is loaded on the EASY-nLCTM 1200 system and separated by a C18 column;

[0037] The linear gradient was set as follows: 0-0.1 min, 4-8% B; 0.1-1 min, 8-10% B; 1.1-1.199 min, 10% B; 1.199-11.199 min, 10-28% B; 11.199-13.499 min, 28-45% B; 13.499-13.899 min, 45-99% B; 13.899-14.999 min, 99% B;

[0038] The DDA MS data of the library were acquired by the PASEF method: MS / MS data were collected in the m / z range of 150–2000, with a TIMS cycle time of 0.6 s during each MS / MS data acquisition; each cycle contained 1 MS and 10 MS / MS scans of 100 ms TIMS; an average of 12 precursors were selected in the 10 PASEF MS / MS scans, resulting in an MS / MS data acquisition rate of 109 Hz;

[0039] For DIA, 56 DIA windows were acquired with an automatic gain control target of 3e6, injection time automatic, and collision energy adjusted linearly as a function of the shift from 59 eV at 1 / K0 = 1.6 Vs cm-2 to 20 eV at 1 / K0 = 0.6 Vs cm-2; MS / MS spectra were recorded from 150 to 2000 m / z.

[0040] Preferably, in the step (5), Trypsin / P is used as the enzyme, a maximum of two missing cleavages are allowed, oxidation of Me is used as a variable modification, ethyl carbamate is used as a fixed modification, and the FDR for PSM, peptide and protein identification is 1%; the parameters of the Spectronaut software are set as follows: the precursor Q value cutoff and the protein Q value cutoff are both set to 0.01, the normalization strategy is set to local normalization, and MS2 is used as the quantitative MS level.

[0041] Preferably, in step (6), the muma v1.4 software package is used to perform principal component analysis (PCA) to perform cluster analysis of the samples, and the Genefilter R software package is used to calculate the protein fold increase.

[0042] A kit for biomarkers for early diagnosis of DKD is also provided, wherein the biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, PSMC3, and the sample is serum.

[0043] Also provided is the use of biomarkers for early diagnosis of DKD, wherein the biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, PSMC3, and are used for dynamic monitoring during the development of diabetic nephropathy.

[0044] The embodiments of the present invention are described in detail below.

[0045] The specific analysis methods of biomarkers for early diagnosis of DKD include:

[0046] Data Collection

[0047] The volunteers were evaluated for age, weight, glycated hemoglobin, creatinine, urea, uric acid, 24-hour urine albumin, eGFR, urine red blood cells, urine white blood cells, cholesterol, triglycerides, urine ACR, total protein, ALT, AST, as well as serious diseases (such as hyperlipidemia, kidney disease) and common medications. According to these disease indicators, the participants were divided into a healthy control group and a disease group. The disease group was further subdivided into three subgroups: DM (24-hour urine protein < 0.15 g) and DKD (24-hour urine protein > 0.5 g), each containing 10 volunteers. Serum samples from each subgroup were then subjected to proteomic analysis.

[0048] Serum separation

[0049] Blood was collected from participants’ veins and processed within 12 h to separate serum. Serum was separated by centrifugation at 3000 g for 10 min and stored at −80°C until testing.

[0050] Nanoparticle magnetic bead enrichment of low-abundance plasma proteins

[0051] The low-abundance proteins in serum samples were enriched using the EasyPept DeepP Low-Abundance Protein Enrichment and Pretreatment Kit (Omicsolution Co., Ltd., Shanghai, China). According to the manufacturer's instructions, 1 mg (40 µL) of magnetic nanoparticle suspension was taken and the supernatant was removed by magnetic separation. After multiple washes, the magnetic beads were resuspended and 100 µL of serum / plasma was added and mixed for 1 hour at 37 °C using a tumbling mixer. The supernatant was removed by magnetic separation, and 300 µL of washing solution was added and washed by shaking for 5 minutes, and the washing was repeated three times. The proteins were converted into peptides by enzymatic hydrolysis, reductive alkylation, and desalting. The total peptide concentration was determined by the nanodroplet method.

[0052] Data Independent Acquisition (DIA) Mass Spectrometry Analysis

[0053] Gun proteomics and DIA experiments were performed using Thermo ScientificTM OrbitrapTM AstralTM and Vanquish Neo UHPLC systems (ThermoFisher, USA). Samples were loaded on an EASY-nLCTM 1200 system (ThermoFisher, USA) and separated by a C18 column (25 cm × 75 µm). The linear gradient was set as follows: 0-0.1 min, 4-8% B; 0.1-1 min, 8-10% B; 1.1-1.199 min, 10% B; 1.199-11.199 min, 10-28% B; 11.199-13.499 min, 28-45% B; 13.499-13.899 min, 45-99% B; 13.899-14.999 min, 99% B. DDA MS data for the library were acquired by the PASEF method: MS / MS data were collected in the m / z range 150-2000, with a TIMS cycle time of 0.6 s during each MS / MS data acquisition; each cycle contained 100 ms TIMS scans for 1 MS and 10 MS / MS; an average of 12 precursors were selected in the 10 PASEF MS / MS scans, resulting in an MS / MS data acquisition rate of 109 Hz. For DIA, 56 DIA windows were acquired (auto gain control target 3e6, injection time automatic), and the collision energy was adjusted linearly as a function of the shift from 59 eV at 1 / K0 = 1.6 Vs cm-2 to 20 eV at 1 / K0 = 0.6 Vs cm-2. MS / MS spectra were recorded from 150 to 2000 m / z.

[0054] Database Search

[0055] Searches and library generation were performed using the default factory settings of Spectronaut Pulsar 18.4 (Biognosys, Switzerland) (including trypsin / P as enzyme, allowing up to two missed cleavages, oxidation of Me as variable modification, ethyl carbamate as fixed modification, and an FDR of 1% for PSM, peptide, and protein identification). DIA data were analyzed using Spectronaut, searching the spectral library constructed as described above. The main parameters of the software were set as follows: both the precursor Q-value cutoff and the protein Q-value cutoff were set to 0.01, the normalization strategy was set to local normalization, and MS2 was used as the quantitative MS level.

[0056] Data processing and filtering criteria for protein analysis: After extracting raw data from the database, the following filtering criteria were applied: Score Sequest HT > 0, unique peptide count ≥ 1, localization probability ≥ 0.75, or Delta score ≥ 8 (if available in the dataset). Missing value filtering and interpolation were then performed to obtain reliable proteins / sites. These reliable proteins / sites were then median-normalized and log2-transformed for further analysis.

[0057] Differentially expressed proteomic profiling and pathway analysis

[0058] In this study, a total of 2787 proteins expressed in the human proteome were identified. Principal component analysis (PCA) was performed using muma v1.4 software package for cluster analysis of samples. Fold values ​​of proteins were calculated using Genefilter R package. In the filtering of differentially expressed proteins, logFC >2 or <0.5 and false discovery rate adjusted P value (t test) <0.05 were set as criteria. All identified proteins were annotated using GO (http: / / www.blast2go.com / b2ghome; http: / / geneontology.org / ) and KEGG pathways (http: / / www.genome.jp / kegg / ). GO and KEGG enrichment analysis was further performed for differentially expressed proteins (DEPs). Protein-protein interaction analysis was performed using String (https: / / string-db.org / ).

[0059] Biomarker analysis and validation

[0060] Serum from 20 healthy volunteers, 20 patients with diabetes, and 20 patients with diabetic nephropathy was collected and subjected to proteomic analysis after the above pretreatment (the method is shown above), and the biomarkers screened in the discovery set were verified. Then, the sensitivity and specificity analysis was performed using the ROC curve to verify the results.

[0061] Study Results

[0062] The present invention firstly analyzes the differential proteins in the proteomics data, and the PCA results show that the three groups of HC, DM, and DKD can be effectively separated, indicating that the data has good classification potential ( Figure 2 a- Figure 2 c). This indicates that the identified differentially expressed proteins may play an important biological role in these three groups, laying the foundation for further research. The volcano plot results show the number of significantly upregulated and downregulated proteins between the groups. The number of differentially expressed proteins between DM and HC increased compared with that between DKD and HC, indicating that the disease is in a state of continuous deterioration ( Figure 2 d- Figure 2 f).

[0063] The differentially expressed proteins compared between the three groups were intersected, and a total of 40 intersection proteins were obtained. These intersection proteins were intersected with the kidney highly expressed protein data on The human protein atlas, and 3 proteins were obtained, namely: ARL3, C7orf25, and PSMC3 proteins. ARL3, C7orf25, and PSMC3 play an important role in diabetic nephropathy (DKD). As a small GTPase, ARL3 affects cell signaling and membrane transport, which may affect the adaptability of the kidney; C7orf25 is involved in immune response and may aggravate kidney damage through inflammatory mechanisms; and PSMC3, as part of the proteasome, regulates protein degradation and cell cycle, which may affect tubular and glomerular function. The study of these proteins will help to reveal the pathogenesis of DKD and find therapeutic targets ( Figure 3 ).

[0064] In the discovery data, ROC analysis was performed on the three proteins ARL3, C7orf25 and PSMC3, and the results showed that they all showed good sensitivity and specificity in different disease stages. In the DM-HC, DKD-HC and DKD-DM groups, the AUCs of PSMC3 protein were 0.85, 0.94 and 0.82, respectively; the AUCs of ARL3 were 0.88, 0.99 and 1, respectively; and the AUCs of C7orf25 were 0.83, 0.92 and 0.98, respectively. Figure 4 a- Figure 4 c). These results indicate that these three proteins have high diagnostic value in distinguishing between healthy and diabetic patients and diabetic nephropathy stages, suggesting that they may be used as potential biomarkers for early diagnosis and staging assessment of the disease.

[0065] The data of these three proteins in 60 serum samples were verified, and the results showed that they showed good sensitivity and specificity in different disease stages. In the DM-HC, DKD-HC and DKD-DM groups, the AUCs of PSMC3 protein were 0.69, 0.78 and 0.83, respectively; the AUCs of ARL3 were 0.89, 0.57 and 0.95, respectively; and the AUCs of C7orf25 were 0.54, 0.62 and 0.60, respectively. Figure 5 a- Figure 5 c). These results indicate that, although the sensitivity and specificity varied in some groups, overall the three proteins have a high potential in distinguishing different disease stages, suggesting that they may be used as biomarkers for the clinical diagnosis and monitoring of diabetes and its kidney disease.

[0066] The present invention firstly uses the serum of 30 volunteers (divided into 10 healthy volunteers, 10 diabetic patients, and 10 diabetic nephropathy patients) to conduct Astral-DIA proteomics research, and then uses bioinformatics to analyze and combine the data of the human protein atlas database to screen out 3 proteins that can represent the biomarkers of the disease pathogenesis. Finally, 60 volunteer serums (divided into 10 healthy volunteers, 10 diabetic patients, and 10 diabetic nephropathy patients) were collected for biomarker verification, and the results showed that the 3 protein biomarkers have good sensitivity and specificity and can be used as biomarkers for disease diagnosis.

[0067] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are still within the protection scope of the technical solution of the present invention.

Claims

1. A method for analyzing biomarkers for early diagnosis of DKD, characterized in that: The biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, PSMC3, and the method includes the following steps: (1) Data collection: Disease indicators of the subjects were collected, and the subjects were divided into a healthy control group (HC) and a disease group. The disease group was further divided into a diabetes group (DM) and a diabetic nephropathy group (DKD). Proteomic analysis was performed on the serum samples of the healthy control group (HC), the diabetes group (DM), and the diabetic nephropathy group (DKD); (2) Serum separation: Blood is collected from the subject’s vein and processed within 12 hours to separate serum; (3) Enrichment of low-abundance proteins in serum samples; (4) Data independent acquisition DIA mass spectrometry analysis: Use Thermo ScientificTM OrbitrapTM AstralTM and Vanquish Neo UHPLC systems to perform gun-type proteomics and DIA experiments and construct spectral libraries; (5) Database search: The default factory settings of Spectronaut software were used for searching. The DIA data were analyzed using Spectronaut software and the spectral library constructed in step (4) was searched. The filtering criteria for data processing and protein analysis were: Score Sequest HT > 0, unique peptide count ≥ 1, localization probability ≥ 0.75 or Delta score ≥ 8. Missing value filtering and interpolation were then performed to obtain reliable proteins / sites. These reliable proteins / sites were then median normalized and log2 transformed. (6) Differentially expressed proteome profiling and pathway analysis: Identify proteins expressed in the human proteome, perform principal component analysis (PCA) to cluster samples, calculate protein fold-values, set logFC >2 or <0.5 and false discovery rate adjusted P value <0.05 as criteria for filtering differentially expressed proteins, annotate all identified proteins using GO and KEGG pathways, further perform GO and KEGG enrichment analysis on differentially expressed proteins, and use String to perform protein-protein interaction analysis; (7) Analysis and validation of biomarkers: The differentially expressed proteins between the healthy control group HC, the diabetic group DM, and the diabetic nephropathy group DKD were intersected to obtain a total of 40 intersection proteins. These intersection proteins were intersected with the kidney highly expressed protein data to obtain three proteins, ARL3, C7orf25, and PSMC3. The ROC curve was then used to perform sensitivity and specificity analysis to verify the results.

2. The method for analyzing biomarkers for early diagnosis of DKD according to claim 1, characterized in that: In step (1), the disease indicators include: age, weight, glycated hemoglobin, creatinine, urea, uric acid, 24-hour urine albumin, eGFR, urine red blood cells, urine white blood cells, cholesterol, triglycerides, urine ACR, total protein, ALT, AST, evaluation of serious diseases and commonly used drugs; the standard for DM is 24-hour urine protein < 0.15 g, and the standard for DKD is 24-hour urine protein > 0.5 g.

3. The method for analyzing biomarkers for early diagnosis of DKD according to claim 2, characterized in that: In step (2), serum was separated by centrifugation at 3000 g for 10 min and stored at −80°C until testing.

4. The method for analyzing biomarkers for early diagnosis of DKD according to claim 3, characterized in that: In the step (3), the low-abundance proteins in the serum sample are enriched using the EasyPept DeepP low-abundance protein enrichment and pretreatment kit, 1 mg of magnetic nanoparticle suspension is taken, the supernatant is removed by magnetic separation, after multiple washes, the magnetic beads are resuspended, 100 µL of serum / plasma is added, and the mixture is mixed at 37°C for 1 hour using a flip mixer, the supernatant is removed by magnetic separation, 300 µL of washing solution is added, and the mixture is washed by shaking for 5 minutes, the washing is repeated three times, the protein is converted into peptides by enzymatic hydrolysis, reduction alkylation and desalting are performed, and the total peptide concentration is determined by the nanodroplet method.

5. The method for analyzing biomarkers for early diagnosis of DKD according to claim 4, characterized in that: In step (4), the sample is loaded on the EASY-nLCTM 1200 system and separated by a C18 column; The linear gradient was set as follows: 0-0.1 min, 4-8% B; 0.1-1 min, 8-10% B; 1.1-1.199 min, 10% B; 1.199-11.199 min, 10-28% B; 11.199-13.499 min, 28-45% B; 13.499-13.899 min, 45-99% B; 13.899-14.999 min, 99% B; The DDA MS data of the library were acquired by the PASEF method: MS / MS data were collected in the m / z range of 150–2000, with a TIMS cycle time of 0.6 s during each MS / MS data acquisition; each cycle contained 1 MS and 10 MS / MS scans of 100 ms TIMS; an average of 12 precursors were selected in the 10 PASEF MS / MS scans, resulting in an MS / MS data acquisition rate of 109 Hz; For DIA, 56 DIA windows were acquired with an automatic gain control target of 3e6, injection time automatic, and collision energy adjusted linearly as a function of transfer from 59 eV at 1 / K0 = 1.6 Vs cm-2 to 20 eV at 1 / K0 = 0.6 Vs cm-2; MS / MS spectra were recorded from 150 to 2000 m / z.

6. The method for analyzing biomarkers for early diagnosis of DKD according to claim 5, characterized in that: In the step (5), Trypsin / P was used as the enzyme, a maximum of two missing cleavages were allowed, oxidation of Me was used as a variable modification, ethyl carbamate was used as a fixed modification, and the FDR for PSM, peptide, and protein identification was 1%; the parameters of the Spectronaut software were set as follows: the precursor Q value cutoff and the protein Q value cutoff were both set to 0.01, the normalization strategy was set to local normalization, and MS2 was used as the quantitative MS level.

7. The method for analyzing biomarkers for early diagnosis of DKD according to claim 6, characterized in that: In the step (6), the muma v1.4 software package was used to perform principal component analysis (PCA) to perform cluster analysis of the samples, and the Genefilter R software package was used to calculate the protein fold increase.

8. Use of a biomarker in the preparation of a kit for dynamic monitoring of the pathogenesis of diabetic nephropathy, characterized in that: The biomarkers include a combination of one or more of the following proteins: ARL3, C7orf25, and PSMC3.

Images