Biomarkers for diagnosis of spontaneous isolated superior mesenteric artery dissection

Abstract

The invention discloses a biomarker for diagnosing a spontaneous isolated superior mesenteric artery dissection, wherein the biomarker is indole-3-methyl acetate and / or indole-3-pyruvic acid. The method can be used for early diagnosis of SISMAD, and a new strategy is provided for early diagnosis of spontaneous isolated superior mesenteric artery dissection.

Description

Technical Field

[0001] The present invention relates to the field of medical detection technology, and in particular to a biomarker for diagnosing spontaneous isolated superior mesenteric artery dissection. Background Art

[0002] Spontaneous isolated superior mesenteric artery dissection (SISMAD) refers to a dissection of the superior mesenteric artery (SMA). It is characterized by a tear in the SMA intima, resulting in the division of the vessel wall into two true and false lumens. This condition is often accompanied by acute intestinal ischemia and associated gastrointestinal complications. In severe cases, extensive intestinal necrosis can occur, which is life-threatening. Early symptoms of this disease are subtle, making misdiagnosis or delayed diagnosis a common problem. However, advances in modern imaging technologies, particularly computed tomography angiography (CTA) and magnetic resonance angiography (MRA), have significantly improved the early detection rate of SISMAD. However, due to the low incidence of the disease, clinical and basic research on this condition is limited. To date, research on SISMAD has focused on improving non-invasive imaging diagnostics and optimizing treatment options. The European and American Associations for Vascular Surgery have published guidelines on SISMAD, emphasizing CTA as the gold standard for diagnosis. Studies have shown that imaging classification helps to select conservative treatment (such as anticoagulation or antihypertensive therapy) or invasive treatment (such as intra-arterial stent implantation), and individualized treatment based on imaging and clinical symptoms can improve patient prognosis.

[0003] SISMAD is a rare vascular disease that typically presents with only symptoms such as abdominal pain. Although most patients can achieve a good prognosis with conservative treatment, in some cases, particularly when dissection leads to intestinal ischemia, interventional therapy or surgery may be necessary. Studies have suggested that the pathogenesis of SISMAD may be related to hemodynamic factors rather than simply hypertension or connective tissue disease. Therefore, timely diagnosis and appropriate treatment are crucial for preventing intestinal necrosis. Metabolomics, a rapidly developing "omics" technology, reveals changes in physiological or pathological states through comprehensive analysis of small molecule metabolites in biological samples. Metabolomics technology is widely used in disease diagnosis, prognosis, and treatment monitoring. In recent years, with advances in mass spectrometry (MS) and nuclear magnetic resonance (NMR), the application of metabolomics in cardiovascular disease, diabetes, cancer, and other fields has matured. Changes in metabolite profiles can serve as early biomarkers of disease, aiding early diagnosis. Summary of the Invention

[0004] The purpose of the present invention is to provide a biomarker for diagnosing spontaneous isolated superior mesenteric artery dissection, thereby providing a new strategy for the early diagnosis of spontaneous isolated superior mesenteric artery dissection.

[0005] The technical solution adopted by the present invention to solve its technical problem is: A biomarker for diagnosing spontaneous isolated superior mesenteric artery dissection, wherein the biomarker is indole-3-acetate methyl ester and / or indole-3-pyruvate.

[0006] Indole-3-acetic acid methyl ester, indole-3-pyruvic acid, or both can be used as markers.

[0007] The screening criteria for biomarkers of the present invention include: ① Statistical significance: the differential metabolites must meet P <0.05 and |FC|>1.5 to ensure that the expression differences between groups are credible; ② Model contribution: metabolites with VIP≥1 are screened through models such as OPLS-DA to reflect their key impact on the differences between groups; Diagnostic efficacy: AUC>0.8 (ROC curve) markers have high disease differentiation ability.

[0008] The cut-off value of the indole 3-acetic acid methyl ester is 10400000.

[0009] The cut-off value of indole-3-pyruvic acid is 13,200,000.

[0010] When the biomarker level is less than the cut-off value, it indicates abnormality. It is recommended to perform enhanced abdominal CT to diagnose or exclude spontaneous isolated superior mesenteric artery dissection.

[0011] The cut-off value is obtained by relative quantification of the peak area after the sample is tested by HPLC-MS. The specific process of relative quantification is as follows: the offline data (.raw) file is imported into CD3.3 search software for processing. Each metabolite is simply screened for parameters such as retention time and mass-to-charge ratio. Then, peak area correction is performed with the first QC to ensure more accurate identification. Subsequently, peak extraction is performed with settings such as a mass deviation of 5 ppm, a signal intensity deviation of 30%, a minimum signal intensity, and adduct ions. Peak areas are quantified at the same time, and target ions are integrated. Molecular formulas are then predicted based on molecular ion peaks and fragment ions and compared with the mzCloud (https: / / www.mzcloud.org / ), mzVault, and Masslist databases. Background ions are removed using a blank sample. The raw quantitative results are normalized according to the formula: sample raw quantitative value / (sum of sample metabolite quantitative values / sum of QC1 sample metabolite quantitative values) to obtain relative peak areas. Compounds with relative peak area CVs greater than 30% in the QC samples are deleted. Finally, metabolite identification and relative quantification results are obtained.

[0012] The beneficial effects of the present invention are: it can be used for the early diagnosis of SISMAD, and provides a new strategy for the early diagnosis of spontaneous isolated superior mesenteric artery dissection. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The metabolic profiles of SISMAD patients and healthy controls (controls) were analyzed. A. PLS-DA analysis used partial least squares regression to establish a relationship model between metabolite expression and sample category to predict sample category. PLS-DA models were established for each comparison group, and the model evaluation parameters (R) were obtained through 7-fold cross-validation (seven cycles of cross-validation; when the number of biological replicates n <= 3, k cycles of cross-validation were used, with k = 2n). 2 , Q 2 ), if R 2 and Q 2 The closer it is to 1, the more stable and reliable the model is. B. In order to judge the quality of the model, the model will be sorted and verified to check whether the model is "overfitting". The grouping labels of each sample are randomly shuffled before modeling and prediction. Each modeling corresponds to a set of R 2 and Q 2 The value of Q after 200 scrambling and modeling 2 and R 2 The regression lines can be obtained from the values, and when R 2 Data is greater than Q 2 Data and Q2 When the intercept of the regression line and the Y-axis is less than 0, it indicates that the model is not "overfitted." C. The volcano plot can intuitively display the overall distribution of differential metabolites. The horizontal axis represents the fold change (log2(Fold Change)) of the metabolites in different groups, and the vertical axis represents the significance level (-log10(P-value)). Each point in the volcano plot represents a metabolite. Significantly upregulated metabolites are represented by red points, and significantly downregulated metabolites are represented by blue points. The size of the dot represents the VIP value.

[0014] Figure 2 This is a clustering and correlation analysis of differential metabolite patterns between the SISMAD and control groups. A. Cluster analysis is used to determine the metabolic patterns of metabolites under different experimental conditions. Metabolites with similar metabolic patterns may have similar functions or participate in the same metabolic process or cellular pathway. Therefore, by clustering metabolites with identical or similar metabolic patterns, the functions of certain metabolites can be inferred. B. Different metabolites may exhibit synergistic or mutually exclusive relationships. For example, if the trends of metabolites within a certain group are the same, a positive correlation is observed; if the trends of metabolites within a certain group are opposite to those of another group, a negative correlation is observed.

[0015] Figure 3 The expression differences and diagnostic values of M-IAA and IPA between SISMAD and Control groups. DETAILED DESCRIPTION

[0016] The technical solution of the present invention is further described in detail below through specific embodiments.

[0017] In the present invention, unless otherwise specified, the raw materials and equipment used can be purchased from the market or are commonly used in the art. The methods in the following examples, unless otherwise specified, are all conventional methods in the art.

[0018] Example 1: 1. Sample Collection and Processing 1. Study participants: To ensure sample representativeness, 30 patients diagnosed with SISMAD were screened from the First Affiliated Hospital of Zhejiang University School of Medicine (ethics approval number IIT20240766B-R1), and 30 healthy controls matched for age, sex, and body mass index (BMI) were recruited.

[0019] 2. Sample Processing: Within 72 hours of onset, 10 ml of peripheral venous blood was collected from the subjects and placed into EDTA-anticoagulant tubes. The tubes were centrifuged at 3000 rpm for 10 min at room temperature (Thermo ST 16R, USA). After centrifugation, the upper layer of pale yellow plasma was collected and aliquoted into 1 ml cryovials, with each tube containing ≥ 200 ml of plasma. The samples were then rapidly frozen in liquid nitrogen for 30 seconds, transferred to the biobank, and stored in a -80°C freezer to avoid repeated freeze-thaw cycles until sequencing.

[0020] 2. Non-targeted metabolomics analysis 1. Experimental Methods: Plasma metabolite extraction 1) Use a pipette to draw 100 μL of plasma sample into a 1.5 ml sterile EP tube and add 400 μL of extraction solution (methanol:acetonitrile = 1:1); 2) Vortex for 30 seconds until the mixture is uniformly mixed, and then ultrasonicate in an ice-water bath for 10 minutes; 3) All samples were placed at -40°C for 60 minutes; 4) Centrifuge at 12,000 rpm for 15 min at 4°C, pipette 400 μL of the supernatant into a new 1.5 ml sterile EP tube, and vacuum dry for 15 min. 5) Add 200 μL of 50% acetonitrile to reconstitute, vortex mix for 30 seconds, and sonicate in an ice-water bath for 10 minutes; 6) Centrifuge all samples at 13,000 rpm for 15 min at 4°C; 7) Take 75 μL of supernatant and place it in a sample bottle for detection; 8) Take 10 μL of supernatant from each of the 60 samples and mix them into a quality control (QC) for testing.

[0021] 2. On-machine testing The present invention uses an ultra-performance liquid chromatography (UPLC) instrument (Agilent 1290) to separate the target compounds using a 2.1×100 mm, 1.7 μm liquid chromatography column. Phase A is an aqueous phase (i.e., the stationary phase) containing 25 mmol / L ammonium acetate and 25 mmol / L aqueous ammonia, while phase B is an organic phase (i.e., the mobile phase) containing acetonitrile. A gradient elution method is employed: 0–0.5 min, 95% B; 0.5–7 min, 95%–65% B; 7–8 min, 65%–40% B; 8–9 min, 40% B; 9–9.1 min, 40%–95% B; and 9.1–12 min, 95% B. The mobile phase flow rate was 0.5 ml / min, the column temperature was 25°C, the sample plate temperature was 4°C, and the injection volume was 2 μL in positive ion mode (POS) and 2 μL in negative ion mode (NEG).

[0022] This study used a high-resolution triple quadrupole time-of-flight mass spectrometry (Triple Q-TOF-MS) 6600 instrument in information-dependent acquisition (IDA) mode for high-resolution mass spectrometry data acquisition. In IDA mode, the data acquisition software (Analyst TF version 1.7, AB Sciex) automatically selected ions and acquired their secondary mass spectrometric data based on the primary mass spectrometric data and pre-defined criteria. During each cycle, the 12 ions with the highest intensities >100 were selected for secondary mass spectrometry scanning. The collision-induced dissociation energy was 30 eV, and the cycle time was 0.56 s. The ion source parameters were as follows: ① nebulizer gas pressure, 60 psi; ② auxiliary gas pressure, 60 psi; ③ curtain gas pressure, 35 psi; ④ ion source temperature, 600°C; ⑤ declustering voltage, 60 V; and ⑥ electrospray voltage, 5000 V or -4000 V, in POS and NEG modes, respectively.

[0023] 3. Data Processing Raw mass spectrometry data were converted to mzXML format using ProteoWizard software version 3.0 (http: / / proteowizard.sourceforge.net / ). Endogenous metabolite mass spectrometry data analysis software, XCMS version 3.2, was then used to perform baseline filtering and calibration, raw peak extraction, peak identification, peak alignment, and peak integration. The minfrac standard was set to 0.5, and the cutoff value was set to 0.3. Peak identification was performed using R language packages and secondary mass spectrometry databases, including the Human Metabolome Database (HMDB), the Mass Spectral Library of North America (MoNA), and METLIN.

[0024] Relative quantification: The off-line data (.raw) files were imported into CD3.3 database search software for processing. Each metabolite was briefly screened for parameters such as retention time and mass-to-charge ratio. Peak area correction was then performed using the first QC to ensure more accurate identification. Peak extraction was then performed using parameters such as a mass deviation of 5 ppm, a signal intensity deviation of 30%, a minimum signal intensity, and adduct ions. Peak areas were quantified and the target ions were integrated. Molecular formulas were then predicted based on molecular ion peaks and fragment ions and compared with the mzCloud (https: / / www.mzcloud.org / ), mzVault, and Masslist databases. Background ions were removed using a blank sample. The raw quantitative results were normalized using the formula: raw sample quantitative value / (sum of sample metabolite quantitative values / sum of QC1 sample metabolite quantitative values) to obtain relative peak areas. Compounds with relative peak area CVs greater than 30% in the QC samples were removed to obtain relative quantitative results for the metabolites.

[0025] 4. Data Analysis 4.1 Raw Data Preprocessing 1) Filter a single peak to remove noise and filter deviations based on relative standard deviation; 2) Filter individual peaks and retain only peak area data with null values ≤ 50% for the AAA and / or HC groups; 3) Simulate missing values and fill them with half of the minimum value in the existing data; 4) Normalize the data based on the signal changes during the QC sample simulation data acquisition process.

[0026] 4.2 PCA analysis Metabolomics data sets are diverse, and multivariate datasets require a high-dimensional spatial coordinate system for presentation. Therefore, using PCA to convert high-dimensional data to low dimensions is an essential analytical step. Using SIMCA software version 15.0.2, the raw data were first logarithmically transformed and centered, then formatted, and finally automatically modeled and analyzed.

[0027] 4.3 PLS-DA analysis First, partial least squares-discrimination analysis (PLS-DA) modeling was performed on the first principal component. A 7-fold cross-validation was used to test the model quality. The evaluation indicators of the model effectiveness after verification mainly included the following: ① R2Y value—the interpretability of the model for the categorical variable Y; ② Q value—the predictability of the model. Based on the above evaluation, a permutation test was used to randomly change the order of the categorical variable Y to obtain the corresponding random Q value, and the effectiveness of the model was tested again.

[0028] 4.4 Multivariate Statistical Analysis The present invention uses multivariate statistics to analyze data. Previous univariate statistical methods, such as t-tests and analysis of variance, focus solely on individual changes in metabolite expression. In contrast, multivariate statistics better align with the characteristics of untargeted metabolomics datasets. Their advantages can be summarized in two aspects: the ability to uncover potential connections between metabolites and the ability to explore the potential promoting and antagonistic effects of different metabolites on pathophysiological processes. The simultaneous use of these two statistical methods not only facilitates considering the data from different perspectives but also significantly reduces the likelihood of overfitting or false-positive results. The primary metabolite screening criteria used in the present invention are a t-test P < 0.05 and a variable importance in the projection (VIP) of the first principal component of the OPLS-DA model > 1. A secondary screening criterion is the logarithmic (base 2) absolute value of the fold change (greater than 2 or less than 0.5) of a metabolite expression between the AAA group and the HC group, i.e., |log2(fold change)| > 1.

[0029] 4.5 Hierarchical cluster analysis of differential metabolites The present invention calculated a Euclidean distance matrix for the quantitative values of differential metabolites and hierarchically clustered them using a complete linkage method. From the perspective of biological structure and function, the differential metabolites identified in this analysis exhibit certain similarities and / or complementarities, or are regulated by the same metabolic pathways, with similar or opposite expression patterns between the AAA and HC groups. By integrating these features through hierarchical cluster analysis, metabolites with similar characteristics can be grouped together, facilitating the investigation of the specific variations in plasma metabolites between the AAA and HC groups.

[0030] 4.6 Correlation Analysis of Differential Metabolites The present invention uses the Pearson test to calculate the correlation between the quantitative values of differential metabolites. The correlation between two metabolites is measured by the correlation coefficient R. For a positive correlation, 0 < R < 1; for a negative correlation, -1 < R < 0. The closer the absolute value of R is to 1, the stronger the correlation between the two variables; the closer the absolute value of R is to 0, the weaker the correlation between the two variables.

[0031] 4.7 Receiver Operating Characteristic (ROC) Analysis of Differential Metabolites Receiver operating characteristic curve (ROC) analysis was performed on the differential metabolites, and the area under the curve (AUC) was calculated.

[0032] 4.8 Cut-off value analysis 4.8.1 Data Preparation Import the table file (.csv or .txt) containing the peak areas and sample groups (SISMAD vs. Control) into the Statistical Analysis [one factor] module of MetaboAnalyst 6.0.

[0033] 4.8.2 Data Preprocessing ① Missing value processing: Filter metabolites with more than 50% missing values, and fill the remaining missing values with KNN or minimum value; ② Standardization: It is recommended to choose Auto-scaling or Log transformation to eliminate dimensional differences.

[0034] 4.8.3 ROC Analysis Operation ① Enter the Biomarker Analysis → ROC Analysis module.

[0035] ② Parameter setting: Select target metabolites (metabolites that need to be significantly different can be screened by P value, VIP value and / or Foldchange).

[0036] ③Set the grouping variable (binary classification, such as 0=Control, 1=SISMAD).

[0037] ④Result output: ⑤ROC curve: displays AUC value, sensitivity and specificity.

[0038] ⑥Cut-off value determination: The optimal threshold is determined based on the point where the Youden index (J = sensitivity + specificity - 1) is maximized.

[0039] 5. Results Analysis Figure 1 The metabolic profile test results of SISMAD patients and healthy controls (Control) are shown in Figure 2. Figure 1 A, B, and C indicate that there are significant differences in the metabolic profiles between the SISMAD and Control groups (overall), and the model is well constructed, and the data are suitable for further differential analysis (specific differential metabolites).

[0040] Table 1 shows the results of differential metabolite screening between the SISMAD and Control groups. The screening of differential metabolites mainly refers to three parameters: VIP, FC, and value. VIP refers to the variable importance in the projection of the first principal component of the PLS-DA model. The VIP value indicates the contribution of the metabolite to the grouping. FC refers to the fold change, which is the ratio of the mean of the quantitative values of all biological replicates of each metabolite in the comparison group. The P value is calculated by T-test and indicates the level of significance of the difference. The thresholds are set as VIP>1.0, FC>1.5 or FC<0.667 and P value<0.05. The differential metabolites screened are shown in the following table: Sample control Total number of metabolites Total number of differential metabolites Upregulated differential metabolites Down-regulated differential metabolites Positive ion mode 1055 123 27 96 Negative ion mode 1097 128 44 84

[0041] Figure 2 Clustering and correlation analysis of differential metabolites between SISMAD and Control groups. Figure 2 A shows the types and correlations of differential metabolites. Among them, there is a significant correlation between the two indole metabolites, suggesting the importance of indole metabolism in the diagnosis of SISMAD and its potential therapeutic effect. The purpose of differential metabolite correlation analysis is to check the consistency of metabolite and metabolite change trends. The correlation between each metabolite is analyzed by calculating the Pearson correlation coefficient between all metabolites. When the linear relationship between two metabolites is enhanced, it tends to 1 when positively correlated and tends to -1 when negatively correlated ( Figure 2B). At the same time, a statistical significance test was performed on the metabolite correlation analysis, and the significance level P-value < 0.05 was selected as the threshold for significant correlation. The correlation analysis found that the indole metabolites, methyl 2-(1H-indol-3-yl)acetate (M-IAA) and indole-3-pyruvic acid (IPA) were significantly correlated (correlation coefficient was 0.989, P <0.001).

[0042] Calculation and analysis revealed a cutoff value of 10,400,000 for indole-3-acetic acid methyl ester and 13,200,000 for indole-3-pyruvic acid. A biomarker level below the cutoff value indicates an abnormality. A contrast-enhanced abdominal CT scan is recommended to diagnose or exclude spontaneous isolated superior mesenteric artery dissection.

[0043] Figure 3 Figure 1: Differences in expression and diagnostic value of M-IAA and IPA between the SISMAD and control groups. A, B: Plasma expression of M-IAA and IPA was significantly lower in SISMAD patients compared with the control group (P < 0.001). C, D: To explore the diagnostic value of M-IAA and IPA, receiver operating characteristic (ROC) analysis was performed using MetaboAnalyst 6.0. The AUC values for M-IAA and IPA were 0.802 and 0.801, respectively, demonstrating reasonable diagnostic accuracy.

[0044] The embodiment described above is only a preferred solution of the present invention and does not limit the present invention in any form. Other variations and modifications are possible without exceeding the technical solution described in the claims.

Claims

1. A biomarker for diagnosing spontaneous isolated superior mesenteric artery dissection, characterized in that: The biomarkers are indole-3-acetic acid methyl ester and / or indole-3-pyruvic acid.

2. The biomarker according to claim 1, characterized in that The cut-off value of the indole 3-acetic acid methyl ester is 10400000.

3. The biomarker according to claim 1, characterized in that The cut-off value of indole-3-pyruvic acid is 13,200,000.

4. The biomarker according to claim 2 or 3, characterized in that When the biomarker content is less than the cut-off value, it represents an abnormality.

5. The biomarker according to claim 2 or 3, characterized in that The cut-off value is obtained by relative quantification of the peak area after the sample is tested by HPLC-MS.

Images