A biomarker combination based on mitochondrial oxidative stress for the auxiliary diagnosis of dilated cardiomyopathy and application thereof

Abstract

This application relates to the biomedical field and discloses a combination of biomarkers based on mitochondrial oxidative stress for the auxiliary diagnosis of dilated cardiomyopathy and its application. The biomarker combination includes the TARS2, NOX4, and SNCA genes. This biomarker combination can be used to prepare auxiliary diagnostic kits for dilated cardiomyopathy. By detecting the expression levels of each gene in the biomarker combination, the auxiliary diagnosis of dilated cardiomyopathy can be achieved efficiently and accurately. Compared with traditional diagnostic methods, this application, based on the characteristics of mitochondrial oxidative stress-related genes, has higher diagnostic accuracy and better applicability.

Description

Technical Field

[0001] This application relates to the field of biomedicine, and more specifically, it relates to a combination of mitochondrial oxidative stress-based biomarkers for the auxiliary diagnosis of dilated cardiomyopathy and their applications. Background Technology

[0002] Heart failure (HF) is one of the most prevalent cardiovascular diseases, severely impairing patients' quality of life and contributing to high mortality rates worldwide. HF affects more than 64 million people globally, imposing a significant socioeconomic burden. Dilated cardiomyopathy (DCM), the leading non-ischemic cause of HF, is characterized by progressive left ventricular dilation and systolic dysfunction. Despite advancements in treatment, the limited effectiveness of current pharmacological interventions often necessitates heart transplantation in end-stage DCM. The etiology of DCM is multifactorial, including gene mutations, cardiotoxic substances, and metabolic disorders, all contributing to its clinical heterogeneity. Therefore, elucidating novel molecular phenotypes beyond traditional clinical classifications is crucial for improving the diagnosis and treatment of DCM.

[0003] Approximately 30% of DCM cases have a clear genetic background, with most pathogenic variants primarily affecting structural proteins related to the cytoskeleton, sarcomeres, or nuclear membrane. However, the molecular mechanisms underlying a significant proportion of DCM cases remain poorly elucidated. Recent studies have shown that, in addition to structural protein abnormalities, mitochondrial dysfunction is also a key driver of DCM development. Impaired mitochondrial function not only weakens the energy supply to cardiomyocytes but also promotes apoptosis and exacerbates pathological ventricular remodeling.

[0004] Furthermore, mitochondria are highly sensitive to changes in redox homeostasis. Under pathological conditions, excessive accumulation of mitochondrial reactive oxygen species (mtROS) can directly induce oxidative damage in cardiomyocytes and amplify cellular stress responses. However, despite the increasing recognition of the role of mitochondrial oxidative stress in the development of diabetic cardiomyopathy (DCM), current clinical practice lacks specific diagnostic methods and effective intervention strategies for this process. Given that mitochondrial oxidative stress is one of the core pathological mechanisms shared by various types of DCM (including hereditary, idiopathic, and chemotherapy-induced cardiomyopathy), developing a biomarker system based on this biological foundation holds significant translational medical value and is expected to provide crucial support for early disease identification, risk stratification, and precision treatment. Summary of the Invention

[0005] The purpose of this invention is to provide a combination of biomarkers based on mitochondrial oxidative stress that is closely related to the occurrence and development of dilated cardiomyopathy, filling the current clinical gap of lacking specific diagnostic biomarkers for the mitochondrial oxidative stress process in dilated cardiomyopathy, enabling early auxiliary diagnosis and risk stratification of dilated cardiomyopathy, helping clinicians to develop more precise individualized treatment plans, improving the prognosis of patients with dilated cardiomyopathy, and also providing a target basis for the subsequent development of targeted drugs for mitochondrial oxidative stress in dilated cardiomyopathy.

[0006] To achieve the above-mentioned objectives, this invention constructs a combination of mitochondrial oxidative stress-based biomarkers for the auxiliary diagnosis of dilated cardiomyopathy, and provides the uses of the biomarker combination.

[0007] The technical solution proposed in this application is based on the following process:

[0008] a) Transcriptome sequencing data were obtained from heart tissue samples of 37 patients with dilated cardiomyopathy and 14 healthy controls. The data were stored in the Gene Expression Omnibus (GEO) database, dataset number GSE116250. Differential gene expression profiles between the dilated cardiomyopathy group and the healthy control group were analyzed using DESeq2 software to identify differentially expressed genes (DEGs) between the two groups.

[0009] (b) A set of genes related to mitochondrial oxidative stress was obtained from previous studies and literature databases, and its intersection with the differentially expressed genes mentioned above was taken to obtain a set of differentially expressed genes related to mitochondrial oxidative stress. Further, this invention performed weighted gene co-expression network analysis (WGCNA) on the transcriptome data of dilated cardiomyopathy, screened co-expressed gene modules (CEGs) with the highest correlation to the disease phenotype, and took their intersection with the differentially expressed genes to obtain a set of co-expressed genes related to mitochondrial oxidative stress. Finally, the intersection of the above two gene sets was taken again to obtain 16 candidate core genes for dilated cardiomyopathy.

[0010] c) Multiple machine learning methods were used to further screen the 16 candidate core genes, including the least absolute shrinkage and selection operator (LASSO), support vector machine-recursive feature elimination (SVM-RFE), and random forest (RF). Through cross-screening using multiple algorithms, three stable candidate genes with high diagnostic value were obtained.

[0011] d) In the discovery cohort, receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficacy of single-gene and three-gene combination models for dilated cardiomyopathy, and a nomogram was constructed based on multivariate regression analysis to assist in clinical diagnostic decision-making. Furthermore, the diagnostic performance and stability of the single-gene and three-gene combination models were further validated in the validation cohort.

[0012] e) Construct a mouse model of dilated cardiomyopathy, and experimentally detect changes in mitochondrial oxidative stress levels in myocardial tissue, and verify the changes in expression levels of the three selected genes in myocardial tissue, thereby further confirming the reliability of the above biomarkers at the in vivo level.

[0013] The following technical solution is specifically adopted: In a first aspect, this application provides a combination of biomarkers for the auxiliary diagnosis of dilated cardiomyopathy, the combination of biomarkers including the TARS2, NOX4 and SNCA genes.

[0014] Secondly, this application provides the use of reagents for detecting the expression levels of TARS2, NOX4 and SNCA genes or their encoded proteins in the preparation of auxiliary diagnostic products for dilated cardiomyopathy.

[0015] Further, the reagent includes: (i) Primers or probes used to detect the mRNA expression levels of TARS2, NOX4, and SNCA genes; or (ii) Antibodies for detecting the levels of proteins encoded by TARS2, NOX4 and SNCA.

[0016] Thirdly, this application provides a kit for the auxiliary diagnosis of dilated cardiomyopathy, the kit comprising reagents for detecting the expression levels of TARS2, NOX4 and SNCA genes or the proteins they encode.

[0017] Further, the reagent includes: (i) Primers or probes used to detect the mRNA expression levels of TARS2, NOX4, and SNCA genes; or (ii) Antibodies for detecting the levels of proteins encoded by TARS2, NOX4 and SNCA.

[0018] Fourthly, this application provides the use of a combination of biomarkers, including the TARS2, NOX4, and SNCA genes, in screening candidate drugs for dilated cardiomyopathy.

[0019] Fifthly, this application provides a method for screening candidate drugs for dilated cardiomyopathy, the method comprising the following steps: Step 1: Contact the test substance with cells expressing TARS2, NOX4 and SNCA; Step 2: Detect the expression levels of TARS2, NOX4, and SNCA in the cells; Step 3: Evaluate whether the test substance can significantly reduce the expression levels of TARS2, NOX4, and SNCA. If it can significantly reduce these levels, it indicates that the test substance is a candidate drug for the treatment of dilated cardiomyopathy.

[0020] The technical solution of this application is based on the following principles: a) Pathological mechanism basis Mitochondrial oxidative stress is considered a key pathological mechanism involved in various types of dilated cardiomyopathy (including hereditary, idiopathic, and chemotherapy-induced cardiomyopathy). However, current clinical practice lacks effective diagnostic methods and intervention strategies targeting this mechanism. Based on this common pathological basis of mitochondrial oxidative stress, systematically screening related genes at the transcriptome level can help uncover biomarkers with potential diagnostic value, thus providing a theoretical basis for the early identification and auxiliary diagnosis of dilated cardiomyopathy.

[0021] b) Methodological basis With the development of high-throughput sequencing technology, transcriptome data has become an important resource for screening disease-related genes. Machine learning algorithms can extract key feature variables from large-scale biological data and effectively identify genes closely related to the occurrence and development of diseases. This invention improves the stability and reliability of candidate gene screening by introducing multiple machine learning methods for cross-screening and validation, thereby enhancing the robustness of the constructed biomarker system.

[0022] c) Advantages of combined markers Because dilated cardiomyopathy exhibits significant heterogeneity, single biomarkers are easily affected by individual differences and detection stability, making them difficult to meet clinical application needs. In contrast, combinations of biomarkers constructed based on multiple genes can comprehensively reflect disease-related molecular characteristics, improving the ability to distinguish dilated cardiomyopathy from healthy states, thus offering significant advantages in diagnostic accuracy and stability.

[0023] In summary, this application has the following beneficial effects: a) A high-reliability screening strategy based on pathological mechanisms and multi-algorithm fusion This invention, based on mitochondrial oxidative stress-related molecules, combines differential expression analysis, co-expression network analysis, and various machine learning algorithms to systematically screen and cross-validate candidate genes, thereby obtaining a set of key genes with high stability and close association with dilated cardiomyopathy. This multi-method fusion strategy effectively reduces the bias introduced by a single algorithm, improving the reliability and reproducibility of biomarker screening results.

[0024] b) Biomarker combinations significantly improve diagnostic performance This invention is the first to discover that TARS2, NOX4, and SNCA are significantly upregulated in patients with dilated cardiomyopathy. Receiver operating characteristic (ROC) curve analysis showed that the area under the curve (AUC) of each individual gene was greater than 0.8, indicating good auxiliary diagnostic capabilities. Furthermore, when these three genes were combined into a biomarker combination, the AUC was further increased to 0.929, demonstrating high sensitivity and specificity. An AUC of 0.928 was also observed in an external validation cohort, demonstrating the stability of this biomarker combination for diagnosis. These results indicate that the biomarker combination described in this invention has a significant advantage in improving the detection rate of dilated cardiomyopathy and contributes to the early identification of the disease.

[0025] c) Develop risk assessment tools with clinical translation potential This invention constructs a nomogram model based on the aforementioned biomarker combination, which can be used for individualized risk assessment and clinical decision support in dilated cardiomyopathy. Furthermore, combining this biomarker combination with baseline clinical parameters (including age, sex, and cardiac function indicators) is expected to further improve the accuracy of disease diagnosis and provide support for patient prognosis assessment, thereby enhancing its clinical application value. Attached Figure Description

[0026] Figure 1Screening for core genes associated with mitochondrial oxidative stress in dilated cardiomyopathy. (A) Volcano plot of differentially expressed genes; (B) Intersection plot of genes related to mitochondrial oxidative stress; (CD) Heatmap of differential expression; (E) WGCNA results; (FL) Candidate gene screening; (MO) Expression of TARS2, NOX4, and SNCA in DCM.

[0027] Figure 2 Diagnostic efficacy assessment. (AC) ROC curves of individual genes; (D) ROC curves of the three-gene combined model; (E) DCM prediction nomogram based on the combined model; (FH) Expression differences and ROC curves in the external validation cohort.

[0028] Figure 3 Animal model validation. (A) Flowchart of doxorubicin-induced DCM mouse model construction; (B) Changes in the expression of heart failure marker genes Nppa and Nppb; (CE) Echocardiographic detection results; (F) RT-qPCR detection results. Detailed Implementation

[0029] The technical solutions and effects of this application will be further described in detail below with reference to the embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining the invention and are not intended to limit the invention.

[0030] Example Download dataset GSE116250 from the gene expression database GEO as the discovery cohort. This dataset contains transcriptome sequencing data of left ventricular tissue from 37 patients with dilated cardiomyopathy and 14 healthy controls. Download dataset GSE9800 as the validation cohort. This dataset contains transcriptome sequencing data of left ventricular tissue from 12 patients with dilated cardiomyopathy and 11 healthy controls.

[0031] a) Research Methodology 1.1 Machine Learning Algorithms 1.1.1 Minimal Absolute Contraction and Selection Operator LASSO regression was performed using the R package glmnet (v4.1-8). The gene expression matrix was log2 transformed and standardized. Ten-fold cross-validation (cv.glmnet) was used to select the optimal penalty parameter λ (lambda.min), and genes with non-zero coefficients under this λ were selected as candidate predictors. Coefficient path plots and cross-validation error curves were plotted to visualize the feature selection process.

[0032] 1.1.2 Support Vector Machine Recursive Feature Elimination Support Vector Machine (SVM) recursive feature elimination was implemented using the R packages caret (v6.0-94) and e1071 (v1.7-14). Standardized gene expression data and their group labels were used as input. Features were iteratively ranked using five-fold cross-validation, with the least important features being progressively removed; when the number of features exceeded 100, the number of features was halved in each iteration. The feature ranking results from each compromise were summarized to generate a final feature list, and the optimal number of features was determined by combining the error curve and accuracy curve.

[0033] 1.1.3 Random Forest The random forest model was trained using the R package randomForest (v4.7-1.1). All genes were used as predictors, and the number of trees was set to 1000 to balance stability and computational efficiency. The number of variables randomly sampled at each split (mtry) was set to the default value. The optimal number of trees was determined based on the minimum out-of-bag classification error, and gene importance was evaluated using the MeanDecreaseGini index.

[0034] 1.2 Weighted Gene Co-expression Network Analysis Weighted gene co-expression network analysis was performed using the R package WGCNA (v1.72) to identify co-expressed gene modules associated with dilated cardiomyopathy. The processed expression data were first preprocessed: genes with low expression or excessive missing values ​​were removed, and outliers were eliminated using the goodSamplesGenes function. A soft threshold power β = 6 was chosen to optimize the scale-free topology fit exponent Rmax. 2 =0.85, a scale-free co-expression network was constructed. Hierarchical clustering was performed based on the topological overlap matrix, grouping genes with similar expression patterns into the same module. Initial modules were defined using dynamic tree slicing; modules with highly similar characteristic genes were merged. The correlation between modules and phenotypes was calculated, identifying the modules most strongly associated with disease phenotypes for subsequent functional analysis. Within selected modules, hub genes were identified based on module membership relationships (correlation between gene expression and module characteristic genes) and gene significance (correlation with disease phenotype).

[0035] 1.3 Animals and Modeling Wild-type male C57BL / 6 mice (6–8 weeks old, purchased from Vital River Pharmaceuticals, Beijing) were housed in the animal facility of Zhongshan Hospital affiliated with Fudan University. The environment was maintained with a 12-hour light / dark cycle, a temperature of 22–24°C, and a relative humidity of 40%–70%. Mice had free access to water and food. All experimental protocols were reviewed and approved by the hospital's Laboratory Animal Ethics Committee.

[0036] To establish an adriamycin-induced cardiomyopathy model, mice were randomly divided into two groups: the model group received intraperitoneal injection of doxorubicin (Dox, 5 mg / kg, catalog number HY-15142A, MedChemExpress, USA), while the control group received an equal volume of solvent (dimethyl sulfoxide, DMSO, 200 µL). Body weight was monitored weekly, and the dosage was adjusted accordingly. The treatment was continued for 4 weeks, with a cumulative dose of 20 mg / kg.

[0037] 1.4 Mouse cardiac function test Transthoracic echocardiography was performed using a Vevo 2100 imaging system (VisualSonics, Canada) equipped with an MS-400 probe. Mice were anesthetized in a supine position on a 37°C constant-temperature heated platform and nasally inhaled with 1.5–2.0% isoflurane, maintaining a heart rate of 450–550 bpm. M-mode ultrasound images were acquired at the papillary muscle level, and myocardial motion was assessed using speckle-tracking-based strain analysis (VevoStrain module, B-mode recording). All measurements were performed double-blind and analyzed using VevoLAB software (v2.1.0). Left ventricular fractional shortening (FS): (LVIDd – LVIDs) / LVIDd; and left ventricular ejection fraction (EF): (LV Vol;d – LV Vol;s) / LV Vol;d × 100% were calculated based on the M-mode measurements.

[0038] 1.5 Real-time quantitative polymerase chain reaction (RT-qPCR) was used to detect TARS2 transcription levels. Total RNA was extracted from mouse heart tissue using the UNIQ-10 Trizol Total RNA Extraction Kit (Sangon Biotech, China, catalog number: B511321-0100). cDNA was synthesized using the high-quality RNA as a template with the PrimeScript RT Kit (Takara, Japan, catalog number: RR036A). RT-qPCR was performed on a CFX96 real-time quantitative PCR system (Bio-Rad, USA) using qPCR SYBR premix (Vazyme, China, catalog number: SGD386). The thermal cycling program was: 95°C denaturation for 10 seconds, 60°C annealing / extension for 30 seconds, for a total of 39 cycles. Gene expression levels were measured using 2... - The ΔΔCt method was used for calculation. The primer sequences are as follows. The thermal cycling parameters are as follows: denaturation at 95°C for 10 seconds, followed by annealing / extension at 60°C for 30 seconds, for a total of 39 cycles. Two... The ΔΔCt method is used to quantify gene expression.

[0039] The primer sequences used are shown below: Mouse TARS2, upstream primer TTGGCAGAACGATTTGGCCTT (SEQ ID NO:1), downstream primer GTTGTGTTCCATGCAACAGCA (SEQ ID NO:2); Mouse NOX4, upstream primer TGCCTGCTCATTTGGCTGT (SEQ ID NO:3), downstream primer CCGGCACATAGGTAAAAGGATG (SEQ ID NO:4). Mouse SNCA, upstream primer GGGAGTCCTCTATGTAGGTTCC (SEQ ID NO:5), downstream primer TCCAACATTTGTCACTTGCTCT (SEQ ID NO:6).

[0040] 1.6 Detection of mitochondrial oxidative stress levels Mitosolic reactive oxygen species (ROS) levels were detected using MitoSOX Red fluorescent indicator (Invitrogen, USA, catalog number: M36008). The procedure was as follows: the stock solution was diluted with calcium-magnesium HBSS buffer to prepare the working solution; cultured human AC16 cardiomyocytes or adult mouse ventricular myocytes (AMVMs) were incubated with the dye at 37°C for 10 minutes. Fluorescence was captured using a confocal laser scanning microscope (FV3000, Olympus, Japan), with excitation / emission wavelengths of approximately 396 / 610 nm.

[0041] b) Research Results 1. Screening for mitochondrial oxidative stress-related genes in dilated cardiomyopathy using multiple machine learning algorithms. We analyzed transcriptome sequencing (RNA-seq) data from cardiac tissues of 37 patients with dilated cardiomyopathy (DCM) and 14 healthy controls. A total of 4,836 differentially expressed genes (DEGs) were identified between DCM and normal samples, of which 4,606 were upregulated and 230 were downregulated. Figure 1 A). By integrating 2,030 mitochondrial-related genes and 1,387 oxidative stress-related genes from previous studies, we obtained 504 overlapping genes, which we defined as mitochondrial oxidative stress-related genes (MitOS genes). Figure 1 B). Among them, 69 were classified as differentially expressed genes due to mitochondrial oxidative stress (MitOS DEGs), most of which were significantly upregulated in DCM. Figure 1 CD).

[0042] We further performed weighted gene co-expression network analysis (WGCNA). Hierarchical clustering identified 13 distinct co-expression gene modules, among which the turquoise module showed the most significant positive correlation with the DCM phenotype. r = 0.58, P = 9e-06)( Figure 1 E). Combining this module with mitochondrial oxidative stress genes, we screened 35 mitochondrial oxidative stress genes (MitOS CEGs) co-expressed with the DCM phenotype. Figure 1 F). Protein-protein interaction network (PPI) analysis suggests that FASN, TARS2, UCP2, and SNCA may be key regulatory factors in the pathogenesis of DCM. Figure 1 G). Further screening at the intersection of MitOS DEGs and MitOS CEGs identified 16 core genes (G). Figure 1 H).

[0043] Subsequently, we employed various machine learning algorithms, including LASSO Minimum Absolute Shrinkage and Selection Operator (LASSO), SVM-RFE Recursive Feature Elimination (SVM-RFE), and Random Forest (RF), to prioritize 16 key candidate genes. The results showed that LASSO regression identified 8 genes with strong predictive value, while SVM-RFE identified 5 characteristic genes (…). Figure 1 IJ); meanwhile, RF analysis ranked the top 10 genes based on feature importance ( Figure 1 Based on the results of the above algorithms, three overlapping genes were finally identified: TARS2, NOX4, and SNCA (K). Figure 1 L). All three genes were significantly upregulated in DCM compared to healthy controls. Figure 1 MO).

[0044] 2. Diagnostic efficacy and prognostic model based on mitochondrial oxidative stress-related biomarkers We further analyzed the diagnostic performance of TARS2, NOX4, and SNCA, evaluating them by calculating the area under the receiver operating characteristic (ROC) curve (AUC) and its 95% confidence interval (CI). The results showed that TARS2 had the highest diagnostic accuracy (AUC = 0.894, 95% CI = 0.789–0.998), followed by NOX4 (AUC = 0.838, 95% CI = 0.719–0.957) and SNCA (AUC = 0.859, 95% CI = 0.738–0.980). Figure 2AC). The three-gene combined diagnostic model showed better accuracy (AUC = 0.929, 95% CI = 0.859–0.998). Figure 2 D). To improve its clinical application value, we further constructed a nomogram model for predicting DCM ( Figure 2 E). Differential expression levels of TARS2, NOX4, and SNCA were validated in an external validation cohort. Figure 2 F); ROC analysis showed that the AUC of each gene was greater than 0.800, while the AUC of the biomarker combination model reached 0.928 (F). Figure 2 The above results indicate that the combination of the three genes has robust diagnostic value in distinguishing DCM from controls, supporting its application prospects as a combination of biomarkers.

[0045] 3. Expression of the validation biomarker combination in dilated cardiomyopathy To further validate the expression of the above biomarker combination in an animal model of dilated cardiomyopathy, we administered doxorubicin (Dox) (5 mg / kg) intraperitoneally to mice weekly for four weeks, accumulating a dose of 20 mg / kg. Figure 3 A). RT-qPCR analysis revealed that the heart failure marker genes Nppa and Nppb were significantly elevated in the Dox group. Figure 3 B). Echocardiographic results showed that Dox treatment reduced cardiac contractility and caused cardiac dilation in mice, specifically manifested as a decrease in left ventricular ejection fraction (LVEF) and fractional shortening (FS), and an increase in diastolic and systolic intraventricular diameter. Figure 3 CE indicates successful construction of dilated cardiomyopathy mice. RT-qPCR detection of the above biomarkers revealed significantly elevated expression of TASR2, NOX4, and SNCA in dilated cardiomyopathy. Figure 3 F).

[0046] The technical solution of this application can be implemented through the following process: a) In clinical practice, serum TARS2, NOX4 and SNCA protein levels can be detected by enzyme-linked immunosorbent assay (ELISA) to assist in the diagnosis of dilated cardiomyopathy. Combined with nomograms, prognostic risk assessment can be performed on patients undergoing treatment. When cardiac dysfunction is severe, myocardial biopsy can also be performed to further detect the expression changes of biomarkers in the heart through methods such as RT-qPCR, Western blotting and immunohistochemistry, which can more accurately assess mitochondrial oxidative stress levels and the progression of heart failure. At the same time, other diagnostic indicators need to be combined: (1) echocardiographic cardiac function parameters; (2) serum heart failure and myocardial injury-related biomarkers; (3) gene mutation detection.

[0047] b) This biomarker combination is suitable for large-scale community-based preliminary blood screening for various types of dilated cardiomyopathy.

[0048] c) All diagnostic, testing, and treatment procedures should be carried out in a medical center with cardiovascular diagnostic and treatment qualifications, and clinical trials based on this work should be conducted in collaboration with clinicians, basic research scholars, and biotechnology engineers.

[0049] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

Claims

1. A combination of biomarkers for the auxiliary diagnosis of dilated cardiomyopathy, characterized in that, The biomarker combination includes the TARS2, NOX4, and SNCA genes.

2. Use of reagents for detecting the expression levels of TARS2, NOX4 and SNCA genes or their encoded proteins in the preparation of auxiliary diagnostic products for dilated cardiomyopathy.

3. The use according to claim 2, characterized in that, The reagents include: (i) Primers or probes used to detect the mRNA expression levels of TARS2, NOX4, and SNCA genes; or (ii) Antibodies for detecting the levels of proteins encoded by TARS2, NOX4 and SNCA.

4. A reagent kit for the auxiliary diagnosis of dilated cardiomyopathy, characterized in that, The kit includes reagents for detecting the expression levels of the TARS2, NOX4, and SNCA genes or the proteins they encode.

5. The reagent kit according to claim 4, characterized in that, The reagents include: (i) Primers or probes used to detect the mRNA expression levels of TARS2, NOX4, and SNCA genes; or (ii) Antibodies for detecting the levels of proteins encoded by TARS2, NOX4 and SNCA.

6. Use of the biomarker combination of claim 1 in screening candidate drugs for dilated cardiomyopathy.

7. A method for screening candidate drugs for dilated cardiomyopathy, characterized in that, The method includes the following steps: Step 1: Contact the test substance with cells expressing TARS2, NOX4 and SNCA; Step 2: Detect the expression levels of TARS2, NOX4, and SNCA in the cells; Step 3: Evaluate whether the test substance can significantly reduce the expression levels of TARS2, NOX4, and SNCA. If it can significantly reduce these levels, it indicates that the test substance is a candidate drug for the treatment of dilated cardiomyopathy.

Images