Application of calnexin as marker for diagnosing mitral valve mucoma

Through joint transcriptome, proteome and omics analysis, multiple candidate biomarkers were identified and verified, solving the accuracy of early diagnosis of myxomatoma-like lesions in canine mitral valve, and improving treatment effect and quality of life.

CN119064583APending Publication Date: 2024-12-03HUAZHONG AGRI UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411260795.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The prior art is difficult to accurately diagnose dog mitral valve myxomatoid lesions (MMVD) in the early stage, resulting in delayed treatment, worsening of the condition, and nonspecificity of existing biomarkers, affecting diagnosis and treatment choices.

Method used

Candidate biomarkers such as MDM2, ROCK1, RIPK1, SNAP23, SERPINH1 and calnexin were identified and verified through transcriptome analysis, proteomic analysis, joint omics analysis and protein interaction network analysis, and were used in the early stages of canine MMVD Diagnosis and monitoring.

Benefits of technology

It improves the accuracy and therapeutic effect of early diagnosis of dog MMVD, provides reliable biomarkers for evaluating disease progression and treatment effects, delays the worsening of the disease, and improves the quality of life and lifespan of dogs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119064583A_ABST
    Figure CN119064583A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of tumor markers, in particular to application of calnin as a marker for diagnosing mitral valve mucoma. The invention relates to application of calnectin in serving as a marker for diagnosing mitral valve mucinous tumor. The calnectin is used as a marker of a dog mitral valve mucinous tumor lesion B2 stage. On the basis of transcriptome and proteome combined analysis and protein interaction network construction, a new canine MMVD biomarker-calnexin is excavated, the content of the calnexin in sick dogs in the MMVD B2 stage is remarkably increased, and the calnexin has the important potential of serving as the canine MMVD biomarker.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of tumor markers, and in particular to the application of calnexin as a marker for diagnosing mitral valve myxoma. Background Art

[0002] In recent years, with the improvement of living standards in modern society, animal health care and corresponding feeding levels have also improved, the life expectancy of domestic pets has increased, and the aging diseases of pets have gradually increased. Canine heart disease is one of the most common diseases in pets clinically, and acquired heart valve disease is the most common cause of clinical death. Among them, mitral valve myxomatous lesions (MMVD) are the most common heart disease and cause of cardiac death in dogs. For example, in North America, MMVD accounts for about 75% of heart disease cases found in veterinary clinics. As the most important heart disease in veterinary clinics, the early clinical diagnosis of canine MMVD is restricted by many factors. Clinically, most cases are already grade III severe when the disease is discovered, and the mortality rate is high. The surgical treatment of this disease has limited effect, and conservative treatment is currently the main treatment. Early diagnosis and intervention in its diagnosis and treatment can significantly affect the progression and prognosis of the disease. Early detection and diagnosis and appropriate treatment can help delay the deterioration of the disease and improve the quality of life and life expectancy of dogs.

[0003] Therefore, the study of biomarkers of MMVD is very important, which is of great significance for the early diagnosis, genetic counseling and treatment of the disease. It can also provide effective guidance for further understanding and research of the disease and dog health management in veterinary clinical practice. Summary of the invention

[0004] In order to solve the above problems, the present invention provides the use of calnexin as a marker for diagnosing mitral valve myxoma. The present invention uses transcriptome analysis, proteome analysis, omics joint analysis and protein interaction network analysis to identify and mine differentially expressed genes and proteins in canine mitral valve myxoma-like lesions. At the same time, the association analysis of known biomarkers and protein interaction networks combined with omics is used to further mine target candidate biomarkers, and the specificity and universality of the candidate markers are further verified through a large sample cohort, providing a theoretical basis for the development of canine MMVD biomarkers.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides the use of calnexin as a marker for diagnosing mitral valve myxoma.

[0007] Preferably, the mitral valve myxoma is a canine mitral valve myxoma-like lesion.

[0008] Preferably, the calnexin is used as a marker for canine mitral valve myxomatous lesions stage B2.

[0009] With the improvement of modern veterinary clinical diagnosis and treatment, canine heart disease is becoming more and more common in pet clinics. Among them, myxomatous mitral valve disease (MMVD) has a higher incidence in male, small and elderly dogs. It is the most common chronic acquired heart valve disease and cause of cardiac death in dogs in veterinary clinics. According to the American College of Veterinary Internal Medicine's consensus guidelines for the diagnosis and treatment of MMVD, 10% of dogs visit primary care veterinary hospitals for heart disease, of which about 75% (more than 4.5 million) suffer from MMVD (Keene et al 2019). Early diagnosis and appropriate treatment measures in the diagnosis and treatment of MMVD can help delay the deterioration of the disease and improve the quality of life and life expectancy of dogs. Biomarkers are of great value in diagnosis, prediction of disease progression and evaluation of treatment effects. Cardiac troponin-I (cTnI) and natriuretic peptide NT-ProBNP are commonly used canine MMVD markers and can be detected in serum and plasma. Because it is related to myocardial injury and ventricular wall stress, there is a certain degree of nonspecificity in diagnosis, which may lead to nonspecific diagnosis and treatment selection deviation in clinical practice. The early diagnosis of MMVD is particularly important in veterinary clinics. In view of its clinical research value, there are also literature reports on other biomarkers related to MMVD, such as cartilage intermediate layer protein 1 (CILP1) and matrix metalloproteinase 9 (MMP9). However, canine veterinary clinics still need reliable biomarkers with early diagnostic value to evaluate the occurrence, development, treatment and monitoring of MMVD.

[0010] The present invention systematically analyzes the expression differences at the transcriptome and proteome levels of peripheral blood of MMVDB2 stage dogs compared with healthy dogs by means of transcriptome analysis, proteome analysis, omics joint analysis and protein interaction network analysis, using real-time fluorescence quantitative PCR, enzyme-linked immunosorbent assay, Western blot and other technologies, screens, identifies and verifies key candidate genes and proteins involved in the early development of MMVD, and preliminarily verifies its reliability through cohort data, providing a theoretical research basis for the development of canine MMVD biomarkers that can be used for clinical diagnosis, and also providing new insights into the pathogenesis of canine MMVD. The specific research results of the present invention are as follows:

[0011] 1. Mining canine MMVD biomarkers based on transcriptome analysis

[0012] The present invention sequenced the peripheral blood transcriptome of 5 MMVDB2 stage poodles (MMVD group) and 3 healthy poodles (NC group), generating a total of 201.6Gb of raw transcriptome sequencing data. After filtering low-quality reads, 200.4Gb of high-quality data were obtained. After alignment with the canine reference genome, 884 new genes were predicted, and the expression quantification of 15,499 genes in 8 samples was obtained, including 880 new genes. 990 differentially expressed genes (DEGs) were identified between the MMVD group and the NC group, including 407 upregulated genes and 583 downregulated genes. Through GO and KEGG enrichment analysis, it was found that the downregulated expression DEGs involved platelet activation, NOD-like receptor signaling, ubiquitin protein transferase activity, phosphorylation and cytomegalovirus infection; the upregulated DEGs involved cofactor metabolic processes, pentose phosphate pathways and other functions. The weighted gene co-expression network analysis was carried out by integrating the basic parameters of experimental dogs and clinical cardiac structural parameters, and a key gene network module with strong and significant correlation with multiple clinical cardiac structural parameters was found. In this study, 100 key genes were screened from this module, of which 66 were differentially expressed genes, of which 64 were downregulated. Among the 64 downregulated genes, the most significant platelet activation pathway (cfa04611) pathway in the enrichment analysis and its related five downregulated DEGs (MDM2, ROCK1, RIPK1, SNAP23, ARHGAP35) were further selected as candidate biomarkers for further validation. In order to further validate the accuracy of the candidate biomarkers, a validation cohort of 108 multi-breed dogs was constructed in the study, including 56 MMVDB2 stage dogs (MMVD validation group) and 52 healthy dogs (NC validation group). The results of RT-qPCR experiments in the validation cohort showed that the expression of 4 genes (MDM2, ROCK1, RIPK1 and SNAP23) among the above 5 candidate markers was significantly downregulated in the MMVD dogs in the validation cohort, and can be used as potential biomarkers for the diagnosis of stage B2 of canine MMVD.

[0013] 2. Mining canine MMVD biomarkers based on proteomic analysis

[0014] Plasma TMT proteome analysis was carried out using the same samples as the transcriptome study, and 5,180 peptides and 574 proteins were identified. 90 differentially expressed proteins (DEPs) were identified between the MMVD group and the NC group, including 83 upregulated proteins and 7 downregulated proteins. Given that 35.17% of the proteins were classified as extracellular proteins, and the extracellular space and extracellular region were significantly enriched in GO, the extracellular matrix ECM was selected as the research target in the present invention. The present invention established a database containing 865 canine extracellular matrix (ECM)-related proteins. Comparison of the database found that 16 of the 90 DEPs were identified as ECM-related proteins, and further constructed a protein interaction network of 16 ECM-related DEPs, of which 13 ECM-related DEPs had protein interactions. Combined with network topology analysis, the present invention selected SERPINH1, SPTAN1, ACTA1 and HMGB1, the top four proteins with the highest protein expression change fold and related to ECM remodeling, as research targets, and then carried out subsequent experimental verification of the four candidate biomarkers. In the validation cohort, ELISA experiments verified that the abundance of SERPINH1 and HMGB1 in MMVD dogs was significantly higher than that in the NC group. Using echocardiography as the gold standard for the diagnosis of MMVD, ELISA results were used for ROC (Receiveroperating characteristic) curve analysis. The results showed that the AUC (Areaunder curve) value of SERPINH1 was as high as 0.885 (95% CI: 0.814 to 0.956), which can be used to accurately identify MMVD B2 dogs, while HMGB1 could not provide good diagnostic value. Western blot experiments further verified that the expression of SERPINH1 protein in MMVD B2 dogs was significantly higher than that in the NC group.

[0015] 3. Mining canine MMVD biomarkers based on combined omics analysis and protein interaction network analysis

[0016] By conducting a joint analysis of transcriptome and proteome, 9 differentially expressed genes / proteins identified by both omics data were obtained. Based on the literature analysis system, 34 known canine MMVD marker proteins reported in the literature were sorted out, and a canine MMVD biomarker protein interaction network was constructed based on the known proteins, and 723 possible canine MMVD biomarker-related proteins were mined. Based on the results of omics joint analysis and protein interaction network analysis, the present invention screened two candidate biomarkers: calnexin (CANX) and fibrillarin (FBL). CANX protein was upregulated to 1.83 times at the peripheral blood transcription level and 2.33 times at the plasma protein level; FBL protein was upregulated to 1.64 times at the peripheral blood transcription level and 1.90 times at the plasma protein level. Using echocardiography as the gold standard for MMVD diagnosis, an ELISA experiment was conducted on a validation cohort of 108 dogs, and it was found that the plasma CANX protein content in dogs with MMVD stage B2 increased by 41.3%, and its AUC for MMVD diagnosis reached 0.915 (95% CI: 0.861-0.969), indicating that CANX protein can be used as a biomarker to accurately identify MMVD stage B2 dogs. However, there was no significant difference in the plasma FBL protein content between MMVD dogs in the validation cohort and healthy control dogs.

[0017] in conclusion:

[0018] Four genes, MDM2, ROCK1, RIPK1 and SNAP23, can be used as potential biomarkers for canine MMVD B2, and SERPINH1 protein can be used to effectively diagnose and predict the phenotype of canine MMVD B2. Omics combined analysis has unearthed a new canine MMVD biomarker, calnexin CANX, which has important potential as a canine MMVD B2 biomarker. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required to be used in the embodiments are briefly introduced below.

[0020] Figure 1Radiological imaging of experimental dogs, A. Echocardiographic LA / AO ratio measured using the early diastolic right short-axis view, the red arrow indicates the aortic size at the level of the aortic valve, the blue arrow indicates the left atrial size, B X-ray vertebral heart score VHS, where S and L indicate the short axis and long axis of the heart, respectively, C. Echocardiographic LVIDd measured using the late diastolic right long-axis view, the red dashed circle indicates the normal and thickened mitral valves in dogs at NC and MMVD B2 stage, the orange arrow indicates the measurement of the left ventricular internal diameter at diastole, D. Comparison of radiological data between the NC and MMVD B2 groups, LVIDd was standardized by body weight as LVIDDN = LVIDd (cm) / body weight (kg). Data are expressed as mean ± SEM, * indicates p < 0.05;

[0021] Figure 2 The difference in expression profiles between the MMVD group and the NC group. A. Density distribution of expression profiles. B. Principal component analysis. The NC group and the MMVD group were marked with different shapes and colors, and a 95% confidence ellipse was drawn for each group of samples. The percentages of the horizontal and vertical axes are the principal component contribution rates.

[0022] Figure 3 For differentially expressed gene identification and expression pattern analysis, A. Volcano plot of DEGs between NC group and MMVD group, the X-axis represents the expression fold change of DEG, the Y-axis represents the statistical significance of the expression fold change, red dots represent significantly up-regulated DEGs, green dots represent significantly down-regulated DEGs, and blue dots represent genes with no significant expression difference, B. Expression clustering heat map of DEGs, rows and columns represent genes and samples, respectively, and red and blue represent high and low expression of DEGs, respectively;

[0023] Figure 4 Key gene modules associated with clinical parameters identified by the WGCNA method. A. Gene clustering dendrogram and corresponding module colors. B. Correlation heatmap between co-expression modules and MMVD clinical parameters. Each cell contains the correlation coefficient and P value.

[0024] Figure 5 To validate the potential biomarkers of MMVD using the validation cohort, A. Real-time fluorescence quantitative PCR validation of candidate genes in the validation cohort (**P<0.01), B. Correlation analysis between clinical cardiac structural parameters and candidate genes in the validation cohort (**P<0.01);

[0025] Figure 6 is the peptide length range distribution;

[0026] Figure 7 is the parent ion mass tolerance distribution;

[0027] Figure 8 is the protein molecular weight distribution;

[0028] Fig. 9 Proteomic analysis revealed differential protein expression profiles between NC and MMVD dogs. A. PCA analysis of MMVD and NC samples based on identified proteins. Healthy control NC (green squares) and MMVD (red squares) dogs were marked with different shapes and colors, and 95% confidence ellipses were drawn for each group of samples. B. Volcano plot of differentially expressed proteins between MMVD and NC dogs. The fold change (Log2FoldChange) between the two groups of samples and the P value (-Log10) obtained by the t-test were used to draw a volcano plot to show the significantly different proteins between the two groups of samples. The red and green dots in the figure are significantly up-regulated and down-regulated proteins, respectively; the black dots are proteins with no significant differences. C. Hierarchical clustering analysis of differentially expressed proteins between MMVD and NC dogs. Each row in the figure represents a protein and each column represents a sample. Red, up-regulated expression; dark blue, down-regulated expression. In dogs with MMVD, two major protein clusters were observed, one of which was down-regulated and the other up-regulated.

[0029] Fig.10 GO and KEGG enrichment analysis of differentially expressed proteins, A. GO enrichment analysis of DEP. BP (orange) represents biological process; CC (red) represents cellular component; MF (blue) represents molecular function, B. The top 10 binding-related GOs significantly enriched in MMVD B2 dogs, C. KEGG classification of 90 DEPs, D. Significantly enriched KEGG pathways in MMVD B2 dogs;

[0030] Fig.11 Enriched expression of extracellular matrix-related proteins, A. Subcellular localization analysis of identified proteins, B. GO analysis of identified proteins, C. Overlap analysis of canine extracellular matrix-related proteins (865) and differentially expressed proteins between MMVD vs NC, D. Expression heat map of 16 extracellular matrix-related DEPs, red, high abundance; blue, low abundance;

[0031] Fig.12 GO chord diagram of ECM-related DEPs, where ECM-related DEPs are linked to their GO annotations through ribbons;

[0032] Fig.13Figure 2 shows the protein interaction network and betweenness centrality score distribution of ECM-related DEPs. A. 13 ECM-related proteins are predicted to participate in protein-protein interactions. Nodes represent proteins, and connections represent protein-protein interactions. Betweenness centrality determines the size of nodes, where red represents upregulation and green represents downregulation. B. Histogram of betweenness centrality scores of ECM-related DEPs.

[0033] Fig.14 Experimental validation of ECM-related differential DEPs

[0034] Note: AD. ELISA validation of SERPINH1, HMGB1, SPTAN1 and ACTA1 in plasma samples of NC (n=52) and MMVD groups (n=56). E. Canine NT-proBNP concentration (pmol / L) in NC and MMVD groups. F. ROC curve for diagnosis of MMVD stage B2. G. Western blot validation of SERPINH1 protein in plasma samples of NC and MMVD groups, using total protein loaded after Coomassie R-250 staining as a normalization reference. The red box represents SERPINH1 abundance. "NS" indicates no significant difference;

[0035] Fig.15 Venn diagrams for transcriptome and proteome expression regulation;

[0036] Fig.16 This is the protein interaction network of canine MMVD biomarkers. Yellow circles represent known biomarkers, blue circles represent interacting proteins of known biomarkers, and the size of the circle represents the number of interacting proteins.

[0037] Fig.17 KEGG and GO enrichment analysis of candidate biomarkers: A. KEGG; B. Molecular Function GO; C. Cellular Component GO; D. Biological Process GO;

[0038] Fig.18Figure 4 ELISA detection and ROC curve of candidate biomarkers CANX and FBL. A. ELISA validation standard curve of CANX in plasma samples of NC (n=52) and MMVD (n=56) groups. B. ELISA validation standard curve of FBL in plasma samples of NC (n=52) and MMVD (n=56) groups. C. ELISA validation values ​​of CANX in plasma samples of NC and MMVD groups. D. ELISA validation values ​​of FBL in plasma samples of NC and MMVD groups. E. ROC curve of CANX for diagnosing MMVD stage B2. DETAILED DESCRIPTION

[0039] The present invention provides the use of calnexin as a marker for diagnosing mitral valve myxoma. In the present invention, the mitral valve myxoma is preferably a canine mitral valve myxoma-like lesion. In the present invention, the calnexin is preferably used as a marker for canine mitral valve myxoma-like lesion B2.

[0040] In order to further illustrate the present invention, the present invention is described in detail below in conjunction with embodiments, but they should not be construed as limiting the scope of protection of the present invention.

[0041] Table 1 Abbreviations

[0042]

[0043] Example 1

[0044] Biomarker mining based on transcriptome analysis:

[0045] 1 Introduction

[0046] The transcriptome is located downstream of the genetic code. By detecting gene transcription and regulation at a given time point and in a specific tissue, transcriptome analysis can capture gene states more sensitively than DNA sequencing. Transcriptome sequencing RNA-Seq technology can detect the expression of all mRNAs and non-coding RNAs, and is a powerful technical means to analyze various life phenomena and mine disease biomarkers. With the rapid development of sequencing technology and bioinformatics analysis technology, transcriptomics research has provided an effective method for revealing new biomarkers and their related functions, and has played an important role in canine disease research, such as coronary artery disease (McCaffrey et al 2021), oral melanoma (Rahman et al 2020), pancreatic cancer (Capodanno et al 2020), etc. Weighted gene co-expression network analysis (WGCNA), as one of the most commonly used bioinformatics methods, is often used to identify key gene modules related to clinical characteristics. In recent years, many studies have used transcriptomes and WGCNA methods to conduct biological analysis, providing new ideas for the pathophysiology of cardiovascular diseases and biomarker mining.

[0047] 2 Materials and methods

[0048] 2.1 Main instruments and equipment

[0049] Bioanalyzer2100 (Agilent Technologies, USA), Qubit2.0 Fluorometer (Invitrogen, USA), NanoDrop Micro-volume Spectrophotometer (Thermo Fisher Scientific, USA), NovaSeq 6000 Sequencing Platform (Illumina, USA), StepOne Plus Real-time PCR System (Thermo Fisher Scientific, USA).

[0050] 2.2 Experimental Materials

[0051] 2.2.1 Main reagents

[0052] TRIzol kit (Invitrogen, USA), HiScriptⅡ reverse transcriptase (Vazyme, China), RNA extraction kit (Tiangen, China), RNaseH kit (Thermo Fisher Scientific, USA), AMPure XP Beads nucleic acid purification reagent (Beckman, USA), M-MuLV reverse transcriptase (Thermo Fisher Scientific, USA), DNA polymeraseⅠ (Thermo Fisher Scientific, USA), Ultra TM RNA Library Prep Kit (Illumina, USA).

[0053] 2.2.2 Sample collection

[0054] All experimental dogs in this invention are from the Veterinary Hospital of Huazhong Agricultural University. The Animal Use and Management Committee of Huazhong Agricultural University approved all experiments related to experimental dogs (approval number: HZAUMO-2015-12), and the informed consent of the dog owners was obtained. According to the American College of Veterinary Internal Medicine (ACVIM) consensus guidelines for the diagnosis and treatment of canine MMVD (Keene et al 2019), the experimental dogs were divided into two groups: a healthy control group and a naturally occurring MMVDB2 group. Among them, the healthy control dogs were examined by medical history, physical examination, cardiac examination (phonocardiogram, electrocardiogram, chest X-ray and echocardiogram) and routine laboratory tests (blood tests), and dogs with other heart diseases and concurrent systemic diseases (such as chronic inflammation, renal failure and liver failure) were excluded. The inclusion criteria for MMVDB2 dogs are as follows: 1) heart murmur intensity ≥ 3 / 6; 2) LA / AO ratio ≥ 1.6 on the right short-axis echocardiogram in early diastole; 3) diastolic left ventricular internal diameter (LVIDDN normalized by body weight) ≥ 1.7; 4) X-ray vertebral heart score (VHS) > 10.5. MMVD dogs included in the study need to meet the above four criteria.

[0055] The present invention involves two cohorts: a discovery cohort consisting of three healthy control dogs (NC group) and five MMVDB2 stage dogs (MMVD group), all of which were poodles; a validation cohort consisting of 52 healthy control dogs (NC validation group) and 56 MMVDB2 stage dogs (MMVD validation group), including 24 Poodles, 15 Pomeranians, 28 Beagles, 4 Golden Retrievers, 19 Chinese pastoral dogs, and 18 other dogs. Specific information is shown in Tables 2 and 3. Whole blood samples were collected from the vein of each dog and thoroughly mixed with TRIzol reagent at a ratio of 1:3. The mixture was incubated at room temperature for 10 minutes until the flocs disappeared, and then stored in an ultra-low temperature refrigerator at -80°C for further analysis.

[0056] Table 2 Statistical and clinical data of experimental dogs used in this study

[0057]

[0058] Table 3 Serum biochemical parameters of MMVD B2 dogs and healthy control dogs

[0059]

[0060]

[0061] 2.3 Experimental methods

[0062] 2.3.1 RNA extraction and transcriptome sequencing

[0063] Total RNA was extracted using the TRIzol kit according to the manufacturer's instructions, and the total amount and integrity of the extracted RNA samples were assessed using the Bioanalyzer2100. Oligo(dT) magnetic beads were used to enrich mRNA with polyA tails, and then the obtained mRNA was randomly sheared using divalent cations in the Fragmentation Buffer. The fragmented mRNA was used as a template and random oligonucleotides as primers to synthesize the first-chain cDNA in the M-MuLV reverse transcriptase system, and then the RNA chain was degraded with RNaseH, and the second-chain cDNA was synthesized with dNTPs as raw materials in the DNA polymerase I system. After the purified double-stranded cDNA was end-repaired, A-tailed, and ligated to the sequencing adapter, AMPure XP Beads were used to screen cDNA with a length of about 370 to 420 bp, PCR amplification was performed, and the PCR product was purified again using AMPureXP Beads to obtain the sequencing library. After the library construction was completed, the Qubit2.0 fluorometer was used for preliminary quantification, the library was diluted to 1.5 ng / μl, and then the insert size of the library was detected using the Bioanalyzer2100. After the insert size met expectations, RT-qPCR was used to accurately quantify the effective concentration of the library (the effective concentration of the library was higher than 2 nM) to ensure the quality of the library. After the library inspection was qualified, 2*150bp double-end sequencing was performed using the NovaSeq6000 sequencing platform.

[0064] 2.3.2 Real-time fluorescence quantitative PCR verification

[0065] Total RNA was extracted using the TRIzol kit according to the instructions, and mRNA was reverse transcribed into cDNA using HiScriptⅡ reverse transcriptase. RT-qPCR was then performed on the StepOnePlus real-time PCR system using SYBR Green Master Mix. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene was used as an internal reference gene based on 2 -ΔΔCt The relative expression levels of mRNA were normalized by comparison method. Primers Premiers 5.0 software was used for primer sequence design, and GraphPad Prism software (v.8.0.2.236) was used to analyze the data, and P < 0.05 was considered statistically significant.

[0066] Table 4 Primers used for RT-qPCR

[0067]

[0068] 2.4 Data Analysis

[0069] 2.4.1 Transcriptome data analysis

[0070] Fastp software (v.0.23.2) was used to filter the raw reads obtained from transcriptome sequencing, mainly including removing reads with adapters, removing reads containing N (N means that the base information cannot be determined), and removing low-quality reads (reads with Qphred <= 20 bases accounting for more than 50% of the total read length) (Chen et al 2018). After filtering to obtain high-quality clean reads, Hisat2 software (v2.0.5) was used to align the clean reads to the canine reference genome CanFam3.1 ( http: / / asia.ensembl.org / Canis lupus familiaris / Info / Index )(Kim et al 2019). The reference genome-based approach in StringTie software (v1.3.3b) (Pertea et al 2015) was used to assemble and predict novel transcripts based on the reads aligned to each sample. The predicted novel transcripts were functionally annotated using the BLAST tool (Altschul et al 1990) and known databases, including Pfam (Finn et al 2016), GO (The Gene Ontology 2017), and KEGG (Kanehisa et al 2014).

[0071] 2.4.2 Gene expression analysis

[0072] FeatureCounts software (v1.5.0-p3) (Liao et al 2014) was used to determine the number of reads mapped to each gene, and then the FPKM (Fragments Per Kilobase of exon model per Million mapped fragments) value of each gene was calculated. Differentially Expressed Genes (DEGs) were identified using the negative binomial distribution model of the DESeq2R package (1.20.0) (Anders et al 2010). The P value of gene expression difference was adjusted using the Benjamini and Hochberg method, and genes with adjusted P values ​​< 0.05 and expression fold change > 1 were identified as DEGs. The identified DEGs were subjected to GO and KEGG enrichment analysis using the clusterProfilerR package (3.8.1) (Yu et al 2012). If BenjaminiadjustedPvalues≤0.05, the GO or KEGG was judged to be significantly enriched.

[0073] 2.4.3WGCNA Analysis

[0074] Genes expressed in at least one sample (FPKM>0) were screened, and an expression gene dataset was constructed. The gene co-expression network was constructed using the WGCNA package in R (Langfelder et al. 2008). First, WGCNA was performed by setting a soft threshold (R 2 =0.7) and a minimum module size of 10 to distinguish gene modules with different expression patterns. Secondly, the Pearson correlation coefficient is used to evaluate the weighted co-expression relationship of the adjacency matrix. Third, the matrix is ​​converted into a topological overlap matrix to estimate the co-expression relationship between genes. Finally, the network is established by merging genes with highly similar co-expression patterns into modules.

[0075] 3 Results

[0076] 3.1 Clinical characteristics of experimental dogs

[0077] The clinical characteristics of the 116 dogs included in the study are shown in Table 2. No significant differences in sex and weight were found between the healthy controls and MMVDB2 dogs in the cohort (P>0.05). However, there was a significant difference in age between the healthy controls and the MMVDB2 group in the validation cohort (P<0.05). MMVDB2 dogs had no obvious clinical symptoms, but had more severe mitral regurgitation, which led to accelerated mitral end-diastolic blood flow (Table 5), and this long-term hemodynamic disorder could lead to left atrial and ventricular enlargement ( Figure 1 A and B) (Boswood et al 2016a; Vezzos et al 2021) and leaflet thickening ( Figure 1 In both the discovery and validation cohorts, significant differences were found in the cardiac function parameters LA / AO, LVIDDN, and VHS between healthy controls and MMVDB2 dogs ( Figure 1 The blood biochemical data of healthy controls and MMVDB2 dogs are shown in Table 3. The study found significant differences in transaminase (ALT), alkaline phosphatase (AP), lactate dehydrogenase (LDH), Ca and phosphate levels between the two groups.

[0078] Table 5 Hemodynamic data of the discovery cohort

[0079] Variable NC(n=3) MMVDstageB2(n=5) P E 71.3±14.0 78.8±9.8 0.473 A 55±7.1 73.4±4.0 0.038 E / A 1.3±0.1 1.2±0.1 0.306 FS(%) 30.6±2.2 50.9±5.3 0.002 MR no mild -

[0080] Note: E, peak velocity of blood flow in early diastole of mitral valve; A, peak velocity of blood flow in late diastole of mitral valve; FS (%), fractional shortening; MR, mitral regurgitation. No = 0%, mild < 20%.

[0081] 3.2 Gene expression analysis results

[0082] The present invention performed peripheral blood transcriptome sequencing on a discovery cohort consisting of 5 MMVDB2 stage poodles and 3 healthy poodles, and a total of 201.6Gb of raw transcriptome sequencing data (rawreads) were obtained. The raw sequence data of this study have been uploaded to the genome sequence archive of the Beijing Institute of Genomics (BIG) Data Center of the Chinese Academy of Sciences (accession number: CRA008836). After filtering out low-quality reads, 200.4Gb (99.4%) of high-quality data (cleanreads) were obtained, and each sample included 23.1–27.7Gb of data. In the cleanreads, 98.3% and 95.2% of the bases had a phred quality score of more than 20 (99.0% base accuracy) and 30 (99.9% base accuracy) (Table 6). All cleanreads were aligned to the canine reference genome (CanFam3.1) using Hisat2 software (Kimetal 2019), and an average of 57.25 Mb of unique aligned reads were obtained for each sample (Table 6). Subsequently, 884 new genes were predicted in the transcriptome data using StringTie software (Pertea et al 2015) and a reference-based approach. The FPKM value was calculated to quantify the expression levels of new genes and 24,580 known genes in each sample. After filtering out genes that were not expressed in all samples, the expression quantification of 15,499 (60.87%) genes in 8 samples was finally obtained, including 880 new genes. The difference in the density distribution of FPKM values ​​between the MMVD and NC groups indicated that there were differences in gene expression between the two groups of dogs ( Figure 2 A). Principal component analysis (PCA) was performed based on the gene FPKM values ​​in all samples, and it was found that the first two principal components (PC1 and PC2) could separate MMVD and NC samples, explaining 58.7% and 8.7% of the variance, respectively ( Figure 2 (B) These results indicate that asymptomatic MMVD has a significant impact on the gene expression profile of canine blood.

[0083] Table 6 Summary of transcriptome sequencing data quality

[0084]

[0085]

[0086] 3.3 Identification of differentially expressed genes

[0087] To explore the specific expression profile induced by MMVD, 990 differentially expressed genes (DEGs) were identified between the MMVD group and the NC group based on the criteria of |log2(foldchange)|>0 ​​and P<0.05. Among the identified differentially expressed genes, 407 differentially expressed genes were up-regulated and 583 differentially expressed genes were down-regulated ( Figure 3 A), hierarchical clustering analysis revealed that the expression profiles of these DEGs were significantly different between the MMVD group and the NC group ( Figure 3 GO enrichment analysis showed that 10 DEGs were significantly involved in the molecular function of ubiquitin protein transferase activity (GO: 0004882, P = 7.68 × 10 -5 )(Table 7). KEGG enrichment analysis found that 54 DEGs were significantly involved in the platelet activation pathway (cfa04611, P = 3.47 × 10 -4 , 19 DEGs), NOD-like receptor signaling pathway (cfa04621, P = 7.68 × 10 -4 , 22 DEGs) and cytomegalovirus infection pathway (cfa05163, P = 2.01 × 10 -3 , 25 DEGs) (Table 7).

[0088] Table 7 The 10 most significantly enriched GO and KEGG

[0089]

[0090] 3.4 Identification of gene co-expression modules associated with MMVD clinical parameters

[0091] Weighted gene co-expression network analysis (WGCNA) can identify gene sets that are highly co-expressed and co-associated with specific phenotypic changes based on gene expression levels. In this study, the clinical information of the experimental dogs was recorded and analyzed (Table 2), including three basic parameters: sex, age, and weight, as well as five clinical cardiac structural parameters obtained by combined echocardiography and X-ray examination: LA (left atrial diameter in the right short-axis view in early diastole), AO (aortic diameter in the right short-axis view in early diastole), LA / AO ratio, LVIDDN, and VHS. Except for LA parameters (P=0.05), other clinical cardiac structural parameters showed significant differences between the MMVD group and the NC group (P<0.05) (Table 2). By performing WGCNA analysis on the gene expression profiles, a total of 41 co-regulated gene modules were identified ( Figure 4 By analyzing the correlation between the expression levels of gene modules and clinical cardiac structural parameters in MMVD dogs, it was found that the turquoise module (Turquoise) (according to Figure 4The color names in A) were significantly associated with multiple clinical cardiac structural parameters. However, no gene module was detected to be significantly associated with indicators such as gender, age, and weight ( Figure 4 The turquoise module consists of 1,324 genes and is significantly positively correlated with AO parameters (r 2 =0.80, P = 0.02), and was significantly negatively correlated with the other three clinical cardiac structural parameters: LA (r 2 =-0.74, P=0.04), LA / AO(r 2 =-0.92, P=0.001), LVIDDN(r 2 =-0.90, P = 0.002)( Figure 4 Middle B). The turquoise module is negatively correlated with VHS parameters, but not significantly (r 2 =-0.60, P = 0.11)( Figure 4 (middle B).

[0092] Further, 100 hub genes were screened from the turquoise module based on the highest k-within coefficient, of which 66 hub genes (66%) were differentially expressed genes, and almost all differentially expressed hub genes (64 out of 66) were downregulated DEGs (Table 8). Platelets play a central role in the development of cardiovascular diseases, and the platelet activation pathway (cfa04611) is the most significantly enriched KEGG pathway for differentially expressed genes between MMVD and healthy control dogs (Table 7). It is worth noting that five downregulated differentially expressed hub genes (MDM2, ROCK1, RIPK1, SNAP23, ARHGAP35) in the turquoise module are involved in the platelet activation pathway (Table 8). The gene MDM2 encodes the MDM2 proto-oncogene (log2FC=-1.94, P=2.23×10 -2 ), ROCK1 encodes Rho-related protein kinase containing coiled-coli 1 (log2FC = -1.92, P = 1.34 × 10 -2 ), RIPK1 encoding receptor interacting serine / threonine kinase 1 (log2FC = -1.79, P = 1.14 × 10 -3 ), SNAP23 encodes synaptosome-associated protein 23 (log2FC=-1.69, P=3.14×10 -2 ), ARHGAP35 encoding RhoGTPase activating protein 35 (log2FC=-1.58, P=9.26×10 -3 ). The five hub genes can be used as candidate biomarkers for canine MMVD for further verification.

[0093] Table 8 66 hub genes belonging to DEGs in the turquoise module

[0094]

[0095]

[0096]

[0097] 3.5 Validation of candidate biomarkers for MMVD

[0098] To further validate the previously discovered candidate biomarkers for MMVD associated with canine clinical cardiac structural parameters, the present invention constructed a validation cohort of 108 dogs (multi-breeds), including 56 MMVD stage B2 dogs (MMVD validation group) and 52 healthy dogs (NC validation group). By conducting RT-qPCR experiments, the gene expression levels of the above five hubDEGs in the 108 dogs in the validation cohort were analyzed. The results showed that the expression of four DEGs (MDM2, ROCK1, RIPK1, and SNAP23) was significantly downregulated in MMVD dogs (P<0.01) ( Figure 5 A), while ARHGAP35 gene expression was not significantly downregulated. The clinical cardiac structural parameters of the validation cohort were further analyzed, and it was found that there were significant differences in four parameters (LA, LA / AO, LVIDDN and VHS) between the MMVD and NC validation groups (P<0.0001), while there was no significant difference in AO parameters (Table 2). The correlation analysis of gene expression levels and clinical cardiac structural parameters in the validation cohort showed that the expression levels of the four significantly downregulated genes MDM2, ROCK1, RIPK1 and SNAP23 were significantly negatively correlated with the four clinical cardiac structural parameters (LA, LA / AO, LVIDDN and VHS) ( Figure 5 B), but had almost no correlation with AO parameters. The highest significant correlation was found between LA / AO parameters and gene expression levels, while the correlation of LA parameters was relatively low. This result was slightly different from the transcriptome sequencing analysis results of the discovery cohort (same breed), which may be due to the specificity of LA and AO parameters in different breeds. In summary, the results of the large-scale validation cohort of multi-breed dogs showed that the four genes MDM2, ROCK1, RIPK and SNAP23 were significantly downregulated in MMVD dogs and strongly correlated with the clinical cardiac structural parameters of MMVD, and can be used as potential biomarkers for canine MMVD.

[0099] 4 Discussion

[0100] Canine MMVD can lead to long-term hemodynamic disturbances, manifested by significant changes in several key clinical cardiac structural parameters (such as LA / AO ratio, LVIDDN, and VHS). Peripheral blood carries mRNA that can regulate cardiac structure and function. By performing peripheral blood transcriptome sequencing, we can comprehensively analyze the specific gene expression profiles of asymptomatic MMVD stage B2 dogs, providing new insights into gene transcription changes caused by long-term hemodynamic disturbances in MMVD. WGCNA is an efficient systems biology analysis method that can be used to identify highly correlated gene modules and provide a basis for correlation analysis of important phenotypic traits (Langfelder et al 2008; Wang et al 2021b). The WGCNA method has been widely used in the prediction of some cardiovascular disease-related biomarkers, such as acute myocardial infarction (Chen et al 2022), post-infarction heart failure (Niu et al 2019), coronary artery disease (Zheng et al 2021), and dilated cardiomyopathy (Yuan et al 2022). Therefore, transcriptome sequencing and WGCNA analysis can be used to identify potential specific biomarkers for the clinical diagnosis of canine MMVD.

[0101] It is generally believed that omics studies of lesion tissue are the first choice for biomarker discovery, however, collecting valve tissue samples from MMVD dogs violates animal welfare regulations. The goal of this invention is to explore clinically applicable biomarkers. Minimally invasive or non-invasive testing of highly specific biomarkers is an important need for clinical diagnosis of canine MMVD. Body fluids such as peripheral blood, plasma, urine or saliva are the source of specific biomarkers (Necula et al 2019). Peripheral blood is an important part of the cardiovascular system, and hematomic studies are expected to provide important data references for screening of MMVD stage B2 dogs.

[0102] Many studies have shown that MMVD is an age-related disease (Cremer et al 2015; Keene et al2019; Meurs et al 2019). This was further confirmed in the present invention, where the mean age of dogs in MMVD stage B2 (9.1±1.3 years) was observed to be significantly higher than that of healthy controls (3.7±1.5 years) (P<0.01) (Table 2-1). A retrospective study of MMVD in miniature poodles found that the mean age of MMVD dogs was 11±3 years (Meurs et al 2019). The difference in the mean age of MMVD dogs between this study and the present invention may be due to differences in the stages of MMVD, which only include the MMVDB2 stage. In order to explore potential biomarkers applicable to multiple breeds, the present invention included six common breeds in the validation cohort, not just poodles, which may also be another reason for the difference in the mean age of MMVD between the studies.

[0103] The present invention carried out peripheral blood transcriptome sequencing of MMVD B2 stage dogs and found that the expression patterns of 990 genes changed significantly, among which the number of downregulated DEGs (583) exceeded the number of upregulated DEGs (407). GO and KEGG enrichment analysis found that the downregulated DEGs may be involved in regulating several important biological functions related to platelets, such as ubiquitin protein transferase activity (GO: 0004882), platelet activation (cfa04611), NOD-like receptor signaling (cfa04621) and cytomegalovirus infection (cfa05163). Platelets are disc-shaped anucleated cells released from megakaryocytes into the bloodstream (Montenont et al 2016) and play a crucial role in the development of various heart diseases such as atherosclerosis, myocardial infarction, and diabetic coronary heart disease (Bakogiannis et al 2019; Levent et al 2020a; Mauler et al 2019; Qi et al 2021; Sun et al 2021). High-speed turbulent blood flow and changes in shear stress around the mitral valve leaflets caused by MMVD may lead to platelet activation in dogs (Levent et al 2020a; Moesgaard et al 2009). NOD-like receptor signaling can also induce platelet activation, as has been reported in studies of heat stroke (Yin et al 2022) and sepsis (Zhong et al 2021). In addition, protein ubiquitination occurs extensively during platelet activation (Rorsman et al 2016; Unsworth et al 2019; Zhou et al 2016). Previous studies have also reported that human cytomegalovirus-platelet interactions lead to pro-inflammatory and pro-angiogenic responses, thereby exacerbating tissue damage and leading to atherosclerosis formation (Assinger et al 2014; Reinhardt et al 2005). In summary, the results of the present invention indicate that the pathogenesis of canine MMVD involves the suppression of the expression of certain key genes in canine peripheral blood.

[0104] Combining WGCNA with module-trait association analysis, the present invention discovered the gene co-expression module (turquoise module) that is most correlated with multiple clinical cardiac structural parameters of MMVD (i.e., LA / AO ratio, LVIDDN, and VHS). The significant negative correlation between them also supports the suppression of gene expression in canine peripheral blood during the pathogenesis of MMVD. The study found that the LA / AO ratio had the highest negative correlation with the turquoise module. The LA / AO ratio is a convenient and size-independent measurement method and a typical indicator for assessing the left atrial diameter of MMVD dogs (Hsue et al 2020; Keene et al 2019). Therefore, the turquoise module may be closely related to the progression of heart disease in dogs with MMVD. It is worth noting that the five hub genes in the turquoise module (MDM2, ROCK1, RIPK1, SNAP23, and ARHGAP35) are involved in platelet-related biological functions. After validation and analysis in an expanded cohort consisting of different dog breeds, four differentially expressed genes (MDM2, ROCK1, RIPK1, and SNAP23) were found to serve as potential biomarkers with broad applicability for the diagnosis and treatment of stage B2 MMVD in different dog breeds.

[0105] Among the four biomarkers, the MDM2 gene encodes a proto-oncogene, which is an E3 ubiquitin ligase containing a RING domain (Jean-Charles et al 2017). MDM2 can participate in cancer cell proliferation by activating platelet-derived growth factor and deubiquitination, and is upregulated in many soft tissue sarcomas, such as osteosarcoma (Bouaoud et al 2021), retroperitoneal liposarcoma (Casadei et al 2022), and liposarcoma (Thirasastr et al 2022). When patients with localized or metastatic solid tumors are treated with MDM2 antagonists, a decrease in platelet count is the most common abnormality (deWeger et al 2019; Patnaik et al 2015; Takahashi et al 2021). Of note, some MMVD-susceptible dog breeds (such as Cavalier King Charles Spaniels and Maltese dogs) have higher circulating platelet counts, which may represent a familial trait of premature onset of MMVD in susceptible breeds. et al 2018; Reimann et al 2021). The present invention shows that the inhibition of MDM2 gene expression may have a positive regulatory function, which will increase the count of circulating platelets in the peripheral blood of dogs in the MMVD B2 stage, leading to long-term hemodynamic disorders. In subsequent studies, the function of the MDM2 biomarker can be further verified by collecting platelet count data in the peripheral blood of dogs.

[0106] The ROCK1 (Rho-associated protein kinase 1) gene encodes a serine / threonine kinase and is an important effector of RhoAGTPase. RhoAGTPase can promote hemostasis by participating in the various vascular effects of vasoactive factors and plays a vital role in the occurrence and development of cardiovascular diseases (Aslan 2019; Shimokawa et al 2016). Previous studies have shown that inhibition of ROCK1 gene expression can lead to increased megakaryocyte polyploidy, abnormal proplatelet release, and giant thrombocytopenia (Donada et al 2019; Ito et al 2018; O'Sullivan et al 2021). It is worth noting that another study analyzed the effects of giant thrombocytopenia on the hemostatic function of dogs with MMVD and found that both giant thrombocytopenia and increased left ventricular shortening fraction are related to the hypercoagulable hemostatic system. The study also pointed out that dogs with giant thrombocytopenia have fewer activated platelets. et al 2019). The present results suggest that ROCK1 may promote the development of MMVD B2 stage by inhibiting platelet activation through downregulating its expression.

[0107] Regarding the other two biomarkers, RIPK1 encodes receptor interacting serine / threonine protein kinase 1, which is a key regulator in apoptosis, necrosis and inflammatory pathways. It can inhibit RIPK1 kinase-mediated cell death in inflammation, infection and embryogenesis, and is directly related to neurological and inflammatory diseases (Ashida et al 2020; Lalaoui et al 2020; Tao et al 2020). SNAP23 encodes synaptosome-associated protein 23, which is closely related to the induction of atrial natriuretic factor secretion (Oczak 2011). Although no studies have reported that these two genes are related to canine MMVD, the present invention confirms their feasibility as biomarkers for canine MMVD.

[0108] Example 2

[0109] Biomarker mining based on proteomic analysis:

[0110] 1 Introduction

[0111] Proteins are the end points of transcription and translation of genetic codes and are responsible for performing various biochemical and metabolic functions required for the survival and homeostasis of living organisms, including responses to diseases and various physiological states. Certain disease-specific protein molecules are discovered during comparative proteome analysis between normal and diseased individuals, which provides biomarker candidates for early diagnosis of the disease or provides candidates for the design of new drug molecular targets. Clinically, proteins can provide important information such as the physiological state, metabolic processes, and disease risks of living organisms, and compared to other tissue samples, peripheral blood collection is convenient and non-invasive. By analyzing changes in proteins in peripheral blood, potential biomarkers can be identified for early diagnosis of the disease, development monitoring, and prognosis tracking. Proteomics research has been widely used in canine disease research and biomarker mining, such as babesiosis. et al 2014), leishmaniasis (Martinez-Subiela et al 2017), oral tumors (Ploypetch et al 2019), and congestive heart failure caused by MMVD (Levent et al 2020b; Sarile et al 2022). TMT proteome sequencing technology (Tandem Mass Tags) is the most commonly used in vitro isobaric isotope labeling quantitative technology in protein research, which can perform relative and absolute quantification of proteins. It is a high-throughput technology commonly used in recent years for the study of quantitative proteomics. It labels the N-terminus or lysine side chain groups of protein polypeptides with a variety of isotope reagents, and then loads them onto a high-precision mass spectrometer tandem analyzer to compare the protein expression levels between different samples.

[0112] 2 Materials and methods

[0113] 2.1 Main instruments and equipment

[0114] EASY-nLC TM 1200 nanoliter UHPLC (Thermo Fisher Scientific, USA), QExactive TMHF-X mass spectrometer (Thermo Fisher Scientific, USA), low-temperature centrifuge (Scilogex, USA), freeze dryer Scan Speed ​​40 (Labogene, Denmark), electrophoresis instrument (Bio-Rad, USA), electrophoresis tank (Bio-Rad, USA), electronic balance (Sartorius, Germany), vortex mixer (Photosynthesis, China), microplate reader Multiskan FC (ThermoFisher Scientific, USA), ice machine (Xue Ke, China), tissue grinder (Shanghai Jingxin, China), ultrasonic cell disruptor (Ningbo Xinzhi, China), Image Quant LAS4000mini chemiluminescence imager (GE, USA), micropipette (eppendorf, Germany), centrifuge (TGL-16, domestic), microplate reader (Rayto-RT-6000, domestic), pure water system (RODI, domestic), constant temperature incubator (thermo, USA).

[0115] 2.2 Experimental Materials

[0116] 2.2.1 Main reagents

[0117] Mass Tagging Kits and Reagents (Thermo Fisher Scientific, USA), Bradford Protein Quantitation Kit (Beyotime, China), dithiothreitol (Sigma, USA), iodoacetamide (Sigma, USA), sodium dodecyl sulfate (Sinopharm, China), urea (Sinopharm, China), mass spectrometry grade trypsin (Promega, USA), ammonium bicarbonate (Sigma, USA), LC-MS grade ultrapure water (Thermo Fisher Scientific, USA), triethylammonium bicarbonate buffer (Sigma, USA), LC-MS grade acetonitrile (Thermo Fisher Scientific, USA), LC-MS grade formic acid (Thermo Fisher Scientific, USA), acetone (Beijing Chemical Plant, China), ammonia (Sigma, USA), ProteoMiner low abundance protein enrichment kit (Bio-Rad, USA), trifluoroacetic acid (Sigma, USA).

[0118] 2.2.2 Sample collection

[0119] The discovery cohort and validation cohort used the same cohort as in “2.1 Experimental Animals and Sample Collection” of Example 1.

[0120] 2.3 Experimental methods

[0121] 2.3.1 Total protein extraction

[0122] Add BioRAD Proteominer Bead to a 1.5ml centrifuge tube, add plasma sample, and mix in a vertical rotating mixer for 2h. Centrifuge at 4℃, 10000r, for 5min, discard the supernatant, add wash buffer to wash the Bead repeatedly. Add 0.4ml 1% TFA to the washed Bead, mix by inversion for 10min, remove the supernatant, repeat twice for elution. Collect the supernatant and freeze into dry powder in a freeze concentrator. Add an appropriate amount of protein dissolving solution (8M urea, 100mM TEAB, pH=8.5) to dissolve the protein precipitate, add 10mM DTT to the protein solution and react at 56℃ for 1h, then add sufficient IAM and react at room temperature in the dark for 1h.

[0123] 2.3.2 Protein detection

[0124] Use Bradford protein quantification kit to prepare BSA standard protein solution according to the instructions, with a concentration gradient range of 0-0.5μg / μL. Take BSA standard protein solution with different concentration gradients and sample solution with different dilution multiples and add them to 96-well plates, make up the volume to 20μL, and repeat each gradient 3 times. Quickly add 180μL G250 staining solution, leave at room temperature for 5min, and measure the absorbance at 595nm. Use the absorbance of standard protein solution to draw a standard curve and calculate the protein concentration of the sample to be tested. Take 20μg of protein sample to be tested for 12% SDS-PAGE gel electrophoresis, where the conditions of concentrated gel electrophoresis are 80V, 20min, and the conditions of separation gel electrophoresis are 120V, 90min. After the electrophoresis, Coomassie Brilliant Blue R-250 staining was performed and decolorization was performed until the bands were clear.

[0125] 2.3.3 TMT Marking

[0126] Take a protein sample, add DB protein dissolving solution (8M urea, 100mM TEAB, pH = 8.5) to make up the volume to 100μL, add trypsin and 100mM TEAB buffer, mix well and digest at 37℃ for 4h, then add trypsin and CaCl2 for digestion overnight. Add formic acid to adjust the pH to less than 3, mix well and centrifuge at room temperature and 12000g for 5min, take the supernatant and slowly pass it through the C18 desalting column, then use the cleaning solution (0.1% formic acid, 3% acetonitrile) to wash continuously for 3 times, then add an appropriate amount of eluent (0.1% formic acid, 70% acetonitrile), collect the filtrate, and freeze-dry. Add 100μL 0.1M TEAB buffer to re-dissolve, and add 41μL TMT labeling reagent dissolved in acetonitrile, invert and mix at room temperature for 2h. Then add ammonia water with a final concentration of 8% to terminate the reaction, take an equal volume of labeled samples, mix, desalt and freeze-dry.

[0127] 2.3.4 Fraction separation

[0128] Prepare mobile phase A (2% acetonitrile, 98% water, ammonia adjusted to pH = 10) and B (98% acetonitrile, 2% water). Dissolve the mixed freeze-dried powder with A solution and centrifuge at 12000g for 10 minutes at room temperature. Use L-3000 HPLC system, the chromatographic column is Waters BEHC18 (4.6×250mm, 5μm), the column temperature is set to 45°C, and the specific elution gradient is shown in Table 9. Collect 1 tube per minute, merge into 10 fractions, and add 0.1% formic acid to dissolve each after freeze-drying.

[0129] Table 9 Liquid chromatography elution gradient for peptide fraction separation

[0130] Time(min) Flow Rate (mL / min) MobilePhaseA(%) MobilePhaseB(%) 0 1 97 3 10 1 95 5 30 1 80 20 48 1 60 40 50 1 50 50 53 1 30 70 54 1 0 100

[0131] 2.3.5 Liquid quality testing

[0132] Prepare mobile phase A (100% water, 0.1% formic acid) and B (80% acetonitrile, 0.1% formic acid). Take 1 μg of each fraction supernatant for injection and perform LC-MS / MS detection. Use EASY-nLC TM 1200 nanoliter UHPLC system, pre-column is homemade pre-column (4.5cm×75μm, 3μm), analytical column is homemade analytical column (15cm×150μm, 1.9μm), HPLC elution conditions are shown in Table 10. QExactive TM HF-X Mass Spectrometer, NanosprayFlex TM The ESI ion source was set at 2.3 kV for ion spray voltage, 320 °C for ion transfer tube temperature, data-dependent acquisition mode for mass spectrometry, full scan range of m / z 350-1500, primary mass spectrometry resolution of 60000 (200 m / z), and C-trap maximum capacity of 3 × 10 6 The maximum injection time of C-trap was 20 ms. The parent ions with the top 40 ion intensities in the full scan were fragmented using the high-energy collision fragmentation (HCD) method for secondary mass spectrometry detection. The secondary mass spectrometry resolution was set to 45000 (200 m / z) and the maximum capacity of C-trap was 5×10 4 The maximum injection time of C-trap was 86 ms, the peptide fragmentation collision energy was set to 32%, and the threshold intensity was set to 1.2×10 5 , the dynamic exclusion range was set to 20 s, and the mass spectrometry detection raw data (.raw) were generated.

[0133] Table 10 Liquid chromatography elution gradient table

[0134] Time(min) Flow Rate (mL / min) MobilePhaseA(%) MobilePhaseB(%) 0 600 94 6 2 600 85 15 78.5 600 60 40 80.5 600 50 50 81.5 600 45 55 90 600 0 100

[0135] 2.3.6 ELISA analysis

[0136] After the plasma was thawed, the concentrations of four proteins, SERPINH1, HMGB1, SPTAN1, and ACTA1, were determined using a canine-specific enzyme-linked immunosorbent assay (ELISA) kit. Canine SERPINH1, HMGB1, SPTAN1, and ACTA1 protein ELISA detection kits were purchased from Shanghai Runyu Company and stored at 2-8°C. They were equilibrated at room temperature for 20 minutes before use, and then the ELISA experiment was performed according to the instructions, and each sample was measured in triplicate.

[0137] 2.3.7 Western blot analysis

[0138] Proteins were extracted from plasma and their concentrations were measured using the Pierce BCA protein assay kit. Equal amounts of protein samples were separated by SDS-PAGE, stained with Coomassie reagent for 40 minutes, and then destained with a destaining solution. Subsequently, the proteins were transferred to polyvinylidene fluoride (PVDF) membranes and hybridized with specific primary antibodies (1:1000) overnight. The membranes were washed and incubated with horseradish peroxidase-conjugated secondary antibodies, and then visualized and analyzed using an Image QuantLAS4000mini chemiluminescent imager. The grayscale values ​​of the protein bands were measured by Image-Pro Plus 6.0 software, and Coomassie staining of total protein in plasma samples was used as a loading control.

[0139] 2.4 Data Analysis

[0140] 2.4.1 Protein identification and quantification

[0141] Based on the canine reference genome CanFam3.1 ( http: / / asia.ensembl.org / Canis_lupus_ familiaris / Info / Index )(Kim et al 2019) protein sequences (58774 entries) were used to search the result spectra of each run using the library search software Proteome Discoverer 2.4 (PD2.4, Thermo Fisher). The library search parameters were set as follows: the mass tolerance of the precursor ion was 10ppm, and the mass tolerance of the fragment ion was 0.02Da. The fixed modification was alkylation of cysteine, the variable modification was oxidation of the side chain methionine, TMT tag modification of the peptide side chain, TMT tag modification, acetylation modification, methionine loss, and methionine loss + acetylation at the N-terminus, and up to 2 missed cleavage sites were allowed.

[0142] In order to improve the quality of the analysis results, the search results were further filtered based on the PD2.4 software: Peptide Spectrum Matches (PSMs) with a credibility of more than 99% were credible PSMs, and proteins containing at least one unique peptide were credible proteins. Only credible peptides and proteins were retained, and FDR verification was performed, and peptides and proteins with FDR greater than 1% were removed. The protein quantification results were statistically analyzed using the T-test, and proteins with significant quantitative differences between the experimental group and the control group (P≤0.05, FC≥1.2 or FC≤0.8) were defined as differentially expressed proteins (DEP).

[0143] 2.4.2 Functional analysis of proteins and DEPs

[0144] InterProScan software was used for GO and IPR functional annotation (including Pfam, PRINTS, ProDom, SMART, ProSite, PANTHER and other databases), and COG and KEGG were used to analyze the functional protein family and pathway of the identified proteins (Jones et al 2014). Volcano plot analysis, cluster heat map analysis, and GO, IPR and KEGG pathway enrichment analysis were performed for DEP (Huang da et al 2009).

[0145] 2.4.3 Identification of ECM-related proteins

[0146] In the Uniprot database ( https: / / www.uniprot.org / uniprot / ) was searched using the keywords "extracellular matrix" and "dog" and limiting "organism" to "Canis lupus familiaris (Dog) (Canis familiaris)"

[9615] , and a canine ECM-related protein database was established based on the retrieved proteins.

[0147] 2.4.4 Construction of protein interaction network of ECM-related proteins

[0148] The protein-protein interaction network of 16 ECM-related proteins was constructed using the STRING DB (https: / / string-db.org / , v.11.5) database (Szklarczyk et al 2023). The betweenness centrality (BC) score was calculated using the Cytoscape plugin tool cytoNCA (v2.1.6). The BC score is a general measure of centrality, and the ECM-related DEPs with higher BC scores were identified as candidate biomarkers.

[0149] 2.4.5 Statistical analysis

[0150] Data are presented as mean ± SEM. GraphPadPrism software (v8.0.2.236) and Student's t test were used to compare continuous variables between groups, and ROC curves and related AUCs for MMVD B2 stage prediction were calculated and constructed. Partial least squares discriminant analysis (PLSsingle bondDA) was performed using the prcomp package in R software (v3.5.0) to visualize high-dimensional data. In all statistical analyses, differences were considered statistically significant when the two-sided P value was ≤0.05.

[0151] 3 Results

[0152] 3.1 Protein identification and quantification

[0153] In order to identify protein features associated with MMVD, proteomic analysis was performed in the discovery cohort (3 NC dogs, 5 MMVD stage B2 dogs). The proteomic data have been deposited in the ProteomeXchange database with the dataset ID PXD037557. A total of 5180 peptides and 574 proteins were found through TMT quantitative proteomics identification. The measurement range of the mass spectrometer is fixed, and only peptides within its target range can be identified. Peptides beyond its target range cannot be detected in the mass spectrometer. Improper selection of proteases will result in peptides in the identification results that are generally too short or too long to be detected. The 5180 peptides discovered in the present invention are distributed in lengths between 7-24 amino acids ( Figure 6 ), which is consistent with the TMT results reported in the literature. The important indicator for measuring the quality of mass spectrometry performance and mass spectrometry identification results is the deviation between the molecular weight of the peptide ion measured by mass spectrometry and the theoretical molecular weight of the peptide. The mass deviation peak of the measured molecular weight and the theoretical molecular weight of the peptide parent ion identified by the present invention is concentrated near 0 ( Figure 7 ), the mass deviation is small. Protein molecular weight distribution is an important indicator for evaluating the size of identified proteins, and the molecular weight range is proportional to the range of identified proteins. The molecular weights of the 574 proteins identified in this study have no molecular weight deviation and are distributed in the range of 10-100 kDa (kilodaltons) and greater than 100 kDa ( Figure 8). Based on the identification of peptides and proteins, the proteins were quantitatively analyzed. The relative quantitative value of each PSM in each sample was first obtained, the relative quantitative value of the unique peptide was corrected, and finally the relative quantitative value of each protein was corrected. In order to further determine the reliability of the proteomics data, principal component analysis was performed on the proteins identified by proteomics in all samples. The results showed that the first and second principal components (PC1 and PC2) could separate all samples, contributing 34.24% and 17.30% of the variance explanation, respectively ( Fig. 9 The 574 proteins identified in the present invention are much higher than the 157 proteins identified in a canine serum proteome study at different stages of CHF caused by MMVD published in 2022 (Saril et al 2022).

[0154] 3.2 Identification of differentially expressed proteins

[0155] Differentially expressed proteins (DEPs) between MMVD and NC dogs were screened using FC ≥ 1.2 or FC ≤ 0.8 and P ≤ 0.05 as criteria. 90 DEPs were identified in the MMVD group compared with the NC group. Among these 90 DEPs, 83 proteins (92.2%) had increased expression and 7 proteins (7.8%) had decreased expression in dogs with MMVD. Transcriptomic analysis results showed that the gene expression levels of the 9 DEPs were also significantly different between MMVD and NC dogs, and the change trend was consistent with that of the proteins. Fig. 9 The volcano plot in middle B depicts the distribution of all proteins based on the MMVD / NC protein abundance ratio and P value. Fig. 9 Middle (C) shows the expression heat map of all 90 DEPs. Cluster analysis of DEP expression showed that the protein expression pattern of MMVD dogs was different from that of NC dogs, and the protein expression in each group was clustered together.

[0156] In order to analyze which biological functions DEPs are significantly associated with, GO enrichment analysis of DEPs was performed. At the biological process (BP), cellular component (CC), and molecular function (MF) levels, metabolism (46 DEPs), intracellular part (27 DEPs), and binding (47 DEPs) GO accounted for the largest proportions ( Fig.10 In A), binding-related GO contains the largest number of DEPs. Binding-related proteins play an important role in atrial fibrillation, congenital heart defects and cardiac development. Therefore, the present invention further analyzed the enriched binding-related GOs, and the top 10 binding-related GOs enriched by DEPs were as follows: Fig.10 As shown in B. In MF, cytoskeletal protein binding (GO: 0008092, P value is 9.29×10 -3, 7 DEPs) and actin binding (GO:0003779, P value 1.79 × 10 -2 , 5 DEPs) were the most significantly enriched GOs.

[0157] Using KEGG functional annotation, 90 DEPs were classified to determine the pathways they participated in. Among metabolism, cellular processes, organic systems, and human diseases, metabolic pathways (20 DEPs), phagosomes (4 DEPs), pancreatic secretion (3 DEPs), and Alzheimer's disease (5 DEPs) accounted for the largest proportions, respectively ( Fig.10 (C) The significantly enriched pathways of 90 DEPs are as follows Fig.10 As shown in D, the differentially expressed proteins were mainly involved in the fructose and mannose metabolic pathways (map00051, P value was 3.19×10 -3 , 4 DEPs), metabolic pathway (map01100, P value was 5.64×10 -3 , 20 DEPs) and Alzheimer's disease (map05010, P value was 4.28×10 -3 , 5 DEPs).

[0158] 3.3 Enrichment of ECM-related proteins

[0159] Among the 574 proteins identified in this study, subcellular localization analysis showed that 35.17% of the proteins were classified as extracellular proteins ( Figure 8 GO analysis of the identified proteins showed that the GO terms significantly enriched in the cellular component (CC) included extracellular space and extracellular region ( Fig.11 In B). Given that 35.17% of proteins were classified as extracellular proteins and that the extracellular space and extracellular region were significantly enriched in GO, this study selected extracellular matrix (ECM)-related proteins as the research target. In order to elucidate the potential changes in ECM in the B2 stage of MMVD, the present invention established a database containing 865 canine ECM-related proteins. Then, sequence alignment analysis was performed on the 865 ECM-related proteins and 90 DEPs, and 16 of the 90 DEPs were identified as ECM-related proteins ( Fig.11 C, Table 11).

[0160] Table 1116 ECM related DEPs

[0161]

[0162]

[0163] Cluster analysis of the abundance of these 16 ECM-related DEP proteins ( Fig.11Figure D) shows that the protein abundance patterns of healthy control dogs and MMVD dogs are different, and the proteins in each group are clustered together. To better understand the functions of the 16 ECM-related DEPs, GO analysis was performed. The GO chord diagram of the 10 ECM-related DEPs with GO annotations is shown in Figure 3. Fig.12 As shown in Figure 3, five ECM-related DEPs (SPTAN1, FKBP4, RACK1, ATRN, and F11) were significantly enriched in protein binding (GO:0005515), and three ECM-related DEPs (HMGB1, SERPINH1, and F11) were significantly enriched in the extracellular space (GO:0005576).

[0164] 3.4 Protein interaction network analysis of ECM-related DEPs

[0165] The protein interaction network of 16 ECM-related DEPs was constructed using the STRING protein interaction database and Cytoscape tools. The results showed that 13 ECM-related DEPs had protein interactions ( Fig.13 The nodes in the figure represent proteins, and up-regulated and down-regulated proteins are shown in red and green, respectively. The Cytoscape tool is used to set the size of the nodes to reflect their betweenness centrality ( Fig.13 In Figure B), the top 10 proteins in the network betweenness centrality score are FKBP4, SERPINH1, SPTAN1, ACTR1A, ADIPOQ, TLN1, ACTA2, ACTA1, S10012A, and HMGB1. Among them, the top four proteins with the highest expression changes in MMVD dogs are SERPINH1, SPTAN1, ACTA1, and HMGB1, with fold changes of 2.27, 1.75, 1.75, and 1.71, respectively (P<0.05). In addition, these four proteins have been reported to be associated with ECM remodeling (Gallo et al 2020; Li et al 2021; Lindert et al 2015; Lubbers et al 2019). Therefore, these four proteins were selected as candidate biomarkers for subsequent experimental validation.

[0166] 3.5 Experimental verification and diagnostic effect of ECM-related DEP

[0167] In the validation cohort, the abundance levels of HMGB1, SPTAN1, ACTA1, and SERPINH1 in plasma were verified by ELISA experiments. The abundance of SERPINH1 and HMGB1 in MMVD dogs was significantly higher than that in the NC group (SERPINH1, P<0.0001; HMGB1, P<0.05) ( Fig.14 The trend was consistent with the TMT proteomic data, and the SERPINH1 gene expression level was also increased by 5.45 times in the transcriptomic data (Section 3.2). The typical heart disease biomarker NT-proBNP was also significantly higher in MMVD dogs than in healthy control dogs ( Fig.14 Middle E).

[0168] In order to evaluate the diagnostic efficacy of HMGB1, SERPINH1 and NT-proBNP for MMVD stage B2, ROC curve analysis was performed using ELISA results, while echocardiography was used as the gold standard for MMVD diagnosis. Fig.14 In F), the AUC curve of HMGB1 protein was <0.800 (AUC=0.661, 95%CI:0.557 to 0.765), indicating that HMGB1 could not provide good diagnostic value. However, the AUC value of SERPINH1 protein was as high as 0.885 (95%CI:0.814 to 0.956), indicating that SERPINH1 can be used to accurately identify MMVD B2 stage dogs. The changes in SERPINH1 protein level were further verified by Western blot experiments, and the results showed that the expression of SERPINH1 protein in MMVD dogs was significantly higher than that in the NC group ( Fig.14 Middle G).

[0169] 4 Discussion

[0170] The present invention uses TMT labeling combined with LC-MS / MS and ECM protein network analysis to compare the plasma proteome of MMVD B2 dogs and healthy control dogs, and further verifies the abundance difference of ECM-related proteins by ELISA and Western blot. In addition, using the imaging diagnosis results as the gold standard, ROC analysis was used to study the role of candidate biomarker proteins in the diagnosis of canine MMVD B2.

[0171] In 2020, a report based on 10 healthy dogs and 10 dogs with congestive heart failure due to MMVD used an LC / MS-based label-free differential proteome expression analysis method to study changes in the platelet proteome in dogs (Levent et al 2020b). This study identified 104 proteins and 10 differentially expressed proteins, of which 4 proteins were upregulated in CHF dogs and 6 proteins were downregulated. In 2022, another study also used an LC / MS-based label-free differential proteome expression analysis method to analyze the changes in the serum proteome of dogs at different stages of CHF caused by MMVD and after conventional treatment in 32 dogs (Saril et al2022). This study identified 157 proteins and 32 differentially expressed proteins, of which 11 proteins were upregulated in CHF dogs and 21 proteins were downregulated. In this study, 574 proteins and 90 differentially expressed proteins were identified in canine plasma using TMT labeling proteomics technology, and the number of proteins and DEPs were much higher than those in the above two studies.

[0172] The extracellular matrix is ​​a non-cellular component of the cardiovascular system and other organ tissues throughout the body that regulates cell function by providing a basic physical scaffold to preserve tissue structure and various biochemical signals (Del Monte-Nieto et al 2020; Zhu et al 2015a). In addition, Derrick et al. found that ECM is an important driver of cardiac development and repair (Derrick et al 2021). ECM remodeling and the accumulation of proteoglycans in the mitral valve can lead to increased leaflet thickness, leaflet prolapse, or flail, which are typical pathological features of MMVD (Maréchaux et al 2017; Oyama et al 2020). A proteomic study of 16 different heart regions showed that the valve had the highest proportion of protein abundance dedicated to ECM compared with other heart regions (Doll et al 2017). This is consistent with the results of the subcellular localization analysis of this study, that is, ECM-related proteins accounted for the highest proportion of proteins identified in plasma (35.17%) ( Fig.11 Similarly, among the 90 DEPs identified between healthy and MMVD B2 dogs, 16 were ECM-related proteins ( Fig.11 In conclusion, the present invention shows that the significant difference in the abundance of ECM-related proteins is a significant molecular feature of MMVD B2 stage, so this study focuses on ECM-related DEPs that are closely related to MMVD DB2 stage.

[0173] The SERPINH1 (also known as HSP47) gene encodes a collagen-binding protein and is located on chromosome 11q 13.5. HSP47 has been reported to be a node hub in a regulatory network that regulates ECM gene transcription (Zhu et al 2015b). Increased SERPINH1 abundance has been shown to be associated with ECM remodeling in fibrotic diseases affecting the breast (Zhu et al 2015b), lung (Staab-Weijnitz 2021), and liver (Ramani et al 2022). Therefore, inhibiting SERPINH1 binding is considered an effective strategy to prevent collagen deposition and ECM remodeling (Taguchi et al 2007; Zhu et al 2015a). SERPINH1 is a molecular chaperone required for the correct folding and secretion of collagen (Tasab et al 2002; Zhu et al 2015a). Myofibroblast-specific ablation of SERPINH1 prevents fibrosis and deposition of type I, III, and IV collagen after pressure overload and significantly reduces cardiac hypertrophy (Khalil et al 2019). In a study on the correlation between heart valve interstitial cell stiffness and transvalvular pressure, it was found that HSP47 secreted by heart valve interstitial cells was positively correlated with collagen biosynthesis, thereby affecting heart valve interstitial cell stiffness and transvalvular pressure (Merryman et al 2006). The above studies showed that SERPINH1 abundance and collagen biosynthesis were significantly parallel (Bellaye et al 2014). However, the proteomic analysis in the present invention did not find significant differences in collagen abundance between healthy control dogs and MMVD stage B2 dogs. The main reason may be that ECM remodeling in the mitral leaflet is a dynamic and complex physiological and pathological process, which is constantly modified by resident valvular interstitial cells to achieve a balance between ECM synthesis and degradation (Oyama et al 2020). Therefore, the main form of mitral valve ECM remodeling may vary at different stages of MMVD. In addition, ECM-related proteins or cleavage processing products are often released into the systemic circulation. However, most ECM remodeling pathways are shared between organs (de Boer et al 2019). Therefore, the circulating levels of ECM remodeling markers may be inconsistent with tissue levels. In summary, the present invention did not find a parallel between SERPINH1 abundance and collagen biosynthesis, which may be due to small sample size, different histological types, and different disease progression.

[0174] ROC curve analysis is a good tool to evaluate the accuracy of diagnostic tests or predictive models, especially in medical imaging techniques, which account for approximately 80% of published ROC studies (Obuchowski et al 2018). A reference standard or gold standard is a necessary prerequisite for ROC curve analysis. ACVIM published a gold standard for diagnosing MMVD based on echocardiography (Keene et al 2019). The AUC of SERPINH1 obtained from the ROC curve was 0.885 (95% CI = 0.814-0.956, P < 0.0001) ( Fig.14 This result was further confirmed by Western blotting ( Fig.14 G). At the same time, plasma NT-proBNP concentrations were significantly increased in MMVD stage B2 dogs compared with healthy control dogs, consistent with previous studies (Klein et al 2022; Saril et al 2022). Although NT-proBNP concentrations are increased in MMVD stage B2 dogs, their usefulness is limited because it is not specific to ECM remodeling caused by valvular tissue, may be at high levels in animals with dilated cardiomyopathy and other chronic heart failure, and is also significantly elevated in severe MMVD (stages C and D). et al 2018; Kleinet al 2022; Saril et al 2022). In this study, the AUC of NT-proBNP (AUC = 0.851) was slightly lower than that of SERPINH1, indicating that the diagnostic accuracy and sensitivity of SERPINH1 for MMVD stage B2 dogs were higher than NT-proBNP, highlighting the potential role of SERPINH1 in the diagnosis and grading of MMVD stage B2. However, due to the design of this study, there were no samples of MMVD dogs in stages B1, C, and D, so the potential effectiveness of targeting SERPINH1 in dogs in stages B1, C, and D remains to be evaluated.

[0175] HMGB1 (high mobility group protein 1) is a highly conserved and ubiquitously expressed nuclear protein. Generally speaking, HMGB1 is present in the nucleus and can be released into the ECM by necrotic cells or granulocytes, acting as a cytokine during inflammatory diseases (Andersson et al 2018), cancer (Ahmed et al 2020), and ischemic heart disease (Pellegrini et al 2019). HMGB1 induces osteoblastic differentiation and calcification of aortic valve interstitial cells through the TLR4-JNK-NF-κB and / or RAGE / ER stress / Sox9 pathways (Wang et al 2016; Wang et al 2017). It is generally believed that calcification is an important pathological process in degenerative valvular heart disease and a common complication (Wang et al 2021a). However, the relationship between HMGB1 and MMVD B2 stage remains unclear. The present invention found that HMGB1 was significantly upregulated in MMVD B2 stage dogs (P<0.05) ( Fig.14 However, whether HGBM1 induces mitral valve calcification in MMVD stage B2 dogs requires further study.

[0176] Example 3

[0177] Biomarker mining based on joint analysis of omics:

[0178] 1 Introduction

[0179] With the development of sequencing technology, omics technology has become the most commonly used means to study life activities. However, life activities are often the result of multi-dimensional and multi-omics co-regulation. A single omics cannot fully and correctly analyze the entire process of complex life activities. Multi-omics joint analysis and multi-dimensional analysis of life processes are becoming more and more popular in modern scientific research. In canine disease research, multi-omics joint analysis methods have also been gradually applied, such as metabolomics and transcriptomics analysis of degenerative mitral valve disease (DMVD) (Li et al 2015a), genome, transcriptome and epigenomic analysis of histiocytic sarcoma (Evans et al 2021), etc. Protein interaction networks (PPIs) include information about cellular biological processes and molecular functions, and have been widely used to analyze the potential mechanisms of action of complex disease-related genes (Vidal et al 2011; Wu et al 2010). Using protein interaction networks and using known biomarkers as "bait" (bait), new candidate biomarkers (prey) can be discovered. For example, in the study of human endometriosis, based on 100 known related genes reported in the literature, the atBioNet tool was used to construct an endometriosis protein interaction network containing 2429 genes, and 15 candidate biomarker genes were mined based on network analysis (Xiao et al 2017). With the completion of canine genome sequencing and the accumulation of protein interaction data, it is possible to integrate multi-omics analysis and protein interaction network analysis to mine new biomarkers for canine mitral valve myxomatous lesions.

[0180] 2 Materials and methods

[0181] 2.1 Literature search

[0182] Literature related to biomarkers of canine mitral valve myxomatous lesions were searched in the Web of Science core database. The search formula was: TS = (((myxomatous mitral valve disease) or MMVD) AND (biomarker) AND (dog or canine)). The search time was April 2023, and the search years were all years.

[0183] 2.2 Network Construction

[0184] For the known canine MMVD biomarker proteins reported in the literature, their Ensembl database IDs (https: / / www.ensembl.org, Canfam3.1) were retrieved. Based on the STRING database (https: / / string-db.org, Version 11.5) (Szklarczyk et al 2023), the "Multiple Proteins" search was performed using the known MMVD biomarker proteins to obtain interacting proteins, and the protein interaction network was drawn using Gephi (version 0.9.7) software (Bastian et al 2009).

[0185] 2.3 Functional enrichment analysis of candidate marker genes

[0186] For the canine MMVD candidate marker genes obtained by network analysis, KEGG and GO enrichment analysis was performed using the STRING database. The enriched KEGG and GO annotation results were screened based on FDR<0.05, and the enriched KEGG and GO were ranked and mapped based on the enrichment strength of the STRING database.

[0187] 2.4 ELISA analysis

[0188] ELISA validation was performed using the validation cohort in Example 1 “2.1 Experimental Animals and Sample Collection”. Canine CANX and FBL protein ELISA detection kits were purchased from Shanghai Runyu Company and stored at 2-8°C. They were equilibrated at room temperature for 20 minutes before use, and then the ELISA experiment was performed according to the instructions. Each sample was measured in triplicate.

[0189] 2.5 Main instruments

[0190] Micropipette (eppendorf, Germany), centrifuge (TGL-16, domestic), microplate reader (Rayto-RT-6000, domestic), pure water system (RODI, domestic), constant temperature incubator (thermo, United States).

[0191] 3 Results

[0192] 3.1 Joint analysis of transcriptome and proteome

[0193] In order to further explore potential candidate biomarkers for canine MMVD, a joint analysis of the transcriptome and proteome was performed. Among the 15,499 expressed genes identified by transcriptome analysis and 574 proteins identified by proteome, the two omics analyses jointly identified 445 genes / proteins, indicating that 77.53% of the proteome-identifiable proteins had gene expression in peripheral blood. The proteome identified 129 proteins that were expressed but not expressed at the transcriptome level, and 15,054 proteins that were expressed at the gene level but not identified by the proteome, indicating that the two omics technologies are complementary in terms of coverage. Among the 990 differentially expressed genes and 90 differentially expressed proteins identified, there were 9 genes ( Fig.15 ), and the trend of gene expression was consistent with that of protein expression (Table 12). Among these 9 proteins, SERPINH1 protein has been verified as a biomarker for canine MMVD by ELISA and Western blot in Section 3.5. For the other 8 proteins, in order to further screen candidate biomarkers for the next experimental verification, a protein interaction network analysis strategy based on known biomarkers reported in the literature is proposed.

[0194] Table 12 Differentially expressed genes / proteins identified by transcriptome and proteome analysis

[0195]

[0196] 3.2 Literature search

[0197] Through the search of Web of Science database, 47 articles related to biomarkers of canine mitral valve myxomatous lesions were obtained, including 2 reviews and 1 conference paper. By reading the retrieved articles one by one, excluding reviews, MMVD prognostic markers (such as ESM-1) (Hong et al 2022), treatment effects (such as omega-3) (Nasciutti et al 2021), biomarker detection methods (such as LC-MS-MRM detection method of podocin peptide fragments) (Szczepankiewicze et al 2019), non-protein biomarkers (such as red blood cell volume distribution width RDW (Guglielmini et al 2021), circulating microRNA (Ro et al 2021b)), etc., a total of 11 articles involved canine MMVD protein biomarkers, covering 34 proteins (Table 13). Most of these studies were conducted on a single biomarker, and only one article reported serum biomarker mining based on LC-MS / MS proteomic analysis, identifying 29 differentially expressed proteins as canine MMVD biomarkers. In addition to proteomics techniques, most other studies were based on ELISA, while ANP used chemiluminescent enzyme immunoassay and MMP-9 used gelatin zymography.

[0198] Table 13 Literature on canine MMVD biomarkers

[0199]

[0200]

[0201] 3.3 Network Construction

[0202] String database (Version 11.5) was searched using 34 known MMVD marker proteins, including cTnI, NT-proBNP, MMP9, proANP, CILP1, AGT, and CPN (excluding new candidate markers identified in this study). Interacting proteins were retrieved for 19 proteins, and a total of 723 interacting proteins were retrieved. A protein interaction network was constructed, including 1026 pairs of interacting proteins. Among the 19 known MMVD biomarkers with interacting proteins, 184 interacting proteins were identified for angiotensinogen (AGT), 163 interacting proteins were identified for matrix metalloproteinase 9 (MMP9), and 104 interacting proteins were identified for platelet basic protein precursor (PPBP). Among the 723 interacting proteins identified, PLG and CRP interacted with 6 known MMVD biomarkers, and CPB2, APOA1, and SERPINC1 interacted with 5 known MMVD biomarkers.

[0203] 3.4 Functional enrichment analysis of candidate markers

[0204] The 723 interacting proteins retrieved by protein interaction network analysis are new candidate markers for canine MMVD. KEGG pathway enrichment analysis was carried out for the 723 candidate markers, and a total of 180 KEGG pathways were enriched, among which the 20 KEGG pathways with the highest STRING enrichment intensity were shown in Figure 2. Fig.17 A. Among them, KEGG pathways such as Dilated cardiomyopathy (cfa05414), Hypertrophic cardiomyopathy (cfa05410), Adrenergic signaling in cardiomyocytes (cfa04261), Chemokine signaling pathway (cfa04062), and Complement and coagulation cascades (cfa04610) are related to heart disease or cardiovascular function. The platelet activation pathway (cfa04611) enriched in differentially expressed genes in transcriptome analysis was also enriched in 723 interacting proteins (21 proteins, P = 4.37 × 10 -7). GO enrichment analysis of 723 candidate markers showed that Molecular Function, Cellular Component, and Biological Process were enriched in 145, 67, and 1035 GOs, respectively. Among them, the 20 GOs with the highest STRING enrichment intensity were found in Fig.17 B, C, and D. Among them, GOs such as Angiotensin receptor activity (GO:0001595), Angiotensin type II receptor activity (GO:0004945), Troponin I binding (GO:0031013), Regulation of angiotensin levels in blood (GO:0002002), Negative regulation of endothelial cell chemotaxis (GO:2001027), and Troponin complex (GO:0005861) are related to heart disease or cardiovascular function. Although GO:0031672 (extracellular matrix) did not enter the 20 GOs with the highest enrichment intensity of STRING, it was also significantly enriched in 723 interacting proteins, including 15 interacting proteins with a P value of 9.93×10 -9 .

[0205] 3.5 Screening and validation of candidate biomarkers for MMVD

[0206] After the intersection of the 9 differentially expressed proteins jointly identified by transcriptome and proteome analysis and 723 canine MMVD candidate markers mined based on protein interaction network, two proteins were obtained for further experimental verification: calnexin (CANX) and fibrillarin (FBL). Analysis of the changes in the two proteins in the blood transcriptome and plasma proteome of canine MMVD found that the quantitative level of calnexin CANX increased to 2.33 times in the plasma of MMVD dogs, and its gene expression was upregulated to 1.83 times in the blood of MMVD dogs; the quantitative level of fibrillarin FBL increased to 1.90 times in the plasma of MMVD dogs, and its gene expression was upregulated to 1.64 times in the blood of MMVD dogs. In the validation cohort consisting of 52 healthy control dogs and 56 MMVD B2 dogs, the standard curve was drawn using an ELISA kit, and the plasma concentrations of CANX and FBL were measured ( Fig.18 Middle A, Fig.18The test results showed that the CANX protein content in the plasma of 52 healthy dogs was 47.21±4.42 ng / mL, while the CANX protein content in the plasma of 56 MMVD dogs reached 66.71±11.61 ng / mL, an increase of 41.3% (P<0.001, Fig.18 Middle C). ROC analysis ( Fig.18 Figure E) shows that the AUC of CANX for MMVD diagnosis reached 0.915 (95% CI: 0.861-0.969), which is higher than the SERPINH1 protein identified in Example 2, indicating that CANX protein can be used as a biomarker to accurately identify MMVD B2 stage dogs. However, the FBL protein content in the plasma of healthy dogs in the validation cohort was 6.25±0.87 ng / mL, and the FBL protein content in the plasma of MMVD-affected dogs was 6.57±1.29 ng / mL, and there was no significant difference between the two (P=0.061, Fig.18 (middle D).

[0207] 4 Discussion

[0208] Multi-omics analysis can integrate data from multiple omics levels to more comprehensively and systematically study the functions, regulatory mechanisms and pathogenesis of complex diseases of biological systems. Combined with protein interaction network analysis, it can help to further explore canine MMVD biomarkers.

[0209] In this study, transcriptomics (Example 1) and proteomics (Example 2) studies were carried out on a discovery cohort consisting of 5 MMVD B2 poodles and 3 healthy poodles, identifying 15,499 expressed genes and 574 proteins. Comparative analysis of MMVD dogs and healthy control dogs identified 990 differentially expressed genes and 90 differentially expressed proteins. Candidate biomarkers mined based on differentially expressed genes are mainly involved in the platelet activation pathway, and candidate biomarkers mined based on differentially expressed proteins are mainly related to the extracellular matrix. There are differences in the candidate biomarkers mined by the two omics analyses, which may be due to the preference of different omics technologies and the differences in gene transcription levels and protein translation levels. In order to more fully explore the information contained in transcriptome and proteome data, this study obtained 9 differentially expressed genes / proteins identified by both omics data through transcriptome and proteome joint analysis, among which SERPINH1 protein has been verified by ELISA and Western blot in Example 2 as a biomarker for canine MMVD.

[0210] Based on the literature analysis, 34 known canine MMVD marker proteins reported in the literature were systematically sorted out, and 723 new canine MMVD biomarkers were mined by constructing a canine MMVD biomarker protein interaction network. KEGG and GO enrichment analysis showed that the new candidate biomarkers were involved in multiple pathways related to heart disease or cardiovascular function, proving the reliability and application potential of the newly mined biomarkers. Chemokines, as a large class of small secreted proteins, perform signal transduction through transmembrane helical receptors coupled to cell surface G proteins, play a core role in the development and homeostasis of the immune system, and participate in all protective or destructive immune and inflammatory responses (Hughes et al2018). Among the chemokines discovered, 19 are considered to be related to heart disease and are involved in various stages of cardiovascular injury response (Dusi et al 2016). In a human MMVD chip study, by comparing the gene expression patterns of 11 healthy controls and 11 MMVD patients, it was found that the expression of the chemokine receptor 1 gene was increased in MMVD patients (Thalji et al 2015), indicating that the chemokine signaling pathway is associated with MMVD. The renin-angiotensin-aldosterone system (RAAS) plays an important role in regulating blood pressure and fluid homeostasis and is an important target for the treatment of cardiovascular and renal diseases (MirabitoColafella et al 2019). An important consequence of canine mitral valve myxomatous lesions is the activation of the RAAS system (Ames et al 2019). The GO enrichment analysis of the new candidate biomarkers mined in this study showed that multiple GOs enriched were related to angiotensin receptors. In addition, the platelet activation pathway (cfa04611) and GO:0031672 (extracellular matrix) were also enriched in 723 interacting proteins, which was consistent with the results of KEGG and GO enrichment analysis of transcriptome and proteome.

[0211] CANX and FBL proteins were screened for experimental validation by combined transcriptome and proteome analysis, combined with protein interaction network analysis based on known biomarkers reported in the literature. The ELISA results of the large-scale validation cohort showed that the content of CANX protein increased by 41.3% in the plasma of MMVD-stricken dogs, and the AUC for MMVD diagnosis reached 0.915, indicating that CANX protein can be used as a biomarker to accurately identify MMVD stage B2 dogs. Calnexin is a molecular chaperone transmembrane protein in the endoplasmic reticulum. It was first discovered in dogs and named pp90 (Wada et al 1991). Its sequence is conserved in multiple species, especially the lumenal domain responsible for the molecular chaperone function (Paskevicius et al 2023). The structure of calnexin is similar to that of another endoplasmic reticulum-localized molecular chaperone, calreticulin. Calnexin and calreticulin, as well as other endoplasmic reticulum auxiliary proteins including ERp57, participate in the calnexin / calreticulin cycle, which is responsible for the folding and maturation of glycosylated proteins synthesized in the endoplasmic reticulum. Studies in mice have shown that calreticulin is involved in heart development, and knocking out the calreticulin gene can lead to impaired calcium-dependent transcriptional regulation and thus cardiac developmental defects (Mesaeli et al 1999), but studies on the relationship between calreticulin and heart disease have not been reported. In a study based on 100 healthy controls and 195 lung cancer patients, it was found that the level of calreticulin in the serum of lung cancer patients was significantly higher than that of healthy controls, and calreticulin is expected to become a new serum diagnostic marker for lung cancer (Kobayashi et al 2015). A study in human cardiomyocytes showed that paraoxonase 2 (PON2) may inhibit Ang II-induced oxidation, inflammation, hypertrophy and damage of AC16 cells by mediating CANX / NOX4 signaling (Ye et al 2023). A genome-wide association analysis conducted in a population cohort of 1,737 Italians showed that calreticulin was associated with resting heart rate, and previous studies have shown that resting heart rate may be an independent predictor of cardiovascular mortality and morbidity (Mezzavilla et al 2014). Although the above studies were conducted in humans, they provide a reference for calnexin as a disease marker in dogs.

[0212] It can be concluded from the above examples that mitral valve myxomatous lesions (MMVD) are the most common heart disease and cause of cardiac death in domestic dogs, and biomarkers play an important role in the early diagnosis and treatment effect evaluation of canine MMVD. The present invention focuses on canine MMVD B2 stage, and comprehensively applies transcriptome analysis, proteome analysis, omics joint analysis and protein interaction network analysis to explore more candidate biomarkers. The main research conclusions are as follows:

[0213] 1) Through transcriptome analysis of the discovery cohort and real-time fluorescence quantitative PCR validation of the validation cohort, it was found that four genes, MDM2, ROCK1, RIPK1 and SNAP23, could serve as potential biomarkers for canine MMVD stage B2.

[0214] 2) Through proteomic analysis of the discovery cohort and ELISA and Western blot validation of the validation cohort, it was found that SERPINH1 protein can be used to effectively diagnose and predict the canine MMVD B2 phenotype.

[0215] 3) Based on the joint analysis of transcriptome and proteome and the construction of protein interaction network, a new canine MMVD biomarker, calnexin, was discovered. Its content was significantly increased in dogs with MMVD B2, and it has important potential as a canine MMVD biomarker.

[0216] Although the above embodiment describes the present invention in detail, it is only a part of the embodiments of the present invention, not all of the embodiments. People can also obtain other embodiments based on this embodiment without creativity, and these embodiments all fall within the protection scope of the present invention.

Claims

1. Application of calnexin as a biomarker for the diagnosis of mitral valve myxoma.

2. The use according to claim 1, characterized in that: The mitral valve myxoma is a canine mitral valve myxoma-like lesion.

3. The use according to claim 1, characterized in that: The calnexin is used as a marker for canine mitral valve myxomatous lesions stage B2.

Citation Information

Patent Citations

  • Withdrawing device for form panels

    EP0005515A1