Targeted metabolomic profiling of acute ST-segment elevation myocardial infarction

Abstract

Myocardial infarction is a major cause of morbidity and mortality worldwide. Metabolomic investigations may be useful for understanding the pathogenesis of ST-segment elevation myocardial infarction (STEMI). STEMI patients were comprehensively examined via targeted metabolomic profiling, machine learning and weighted correlation network analysis. A total of 195 subjects, including 68 STEMI patients, 84 patients with stable angina pectoris (SAP) and 43 non-CVD patients, were enrolled in the study. Metabolomic profiling involving the quantitative analysis of 87 endogenous metabolites in plasma was conducted. This study is the first to perform targeted metabolomic profiling in patients with STEMI. We identified 36 significantly altered metabolites in STEMI patients. Increased levels of four amino acids, eight acylcarnitines, six metabolites of the NO-urea cycle and neurotransmitters, and three intermediates of tryptophan metabolism were detected. The following metabolites exhibited decreased levels: six amino acids, three acylcarnitines, three components of the NO-urea cycle and neurotransmitters, and three intermediates of tryptophan metabolism. We found that the significant changes in tryptophan metabolism observed in STEMI patients-the increase in anthranilic acid and tryptophol and decrease in xanthurenic acid and 3-OH-kynurenine-may play important roles in STEMI pathogenesis. On the basis of the differences in the constructed weighted correlation networks, new significant metabolite ratios were identified. Among the 22 significantly altered metabolite ratios identified, 13 were between STEMI patients and non-CVD patients, and 17 were between STEMI patients and SAP patients. Seven of these ratios were common to both comparisons (STEMI patients vs. non-CVD patients and STEMI patients vs. SAP patients). Additionally, two ratios were consistently observed among the STEMI, SAP and non-CVD groups (anthranilic acid: aspartic acid and GSG (glutamine: serine + glycine)). These findings provide new insight into the diagnosis and pathogenesis of STEMI.

Keywords: Metabolic pathways; Metabolomic profiling; STEMI pathogenesis.

Fig. 1

Trial profile.

Fig. 2

Heatmap correlation matrices of plasma metabolites and the biochemical analysis results.

Fig. 3

AUC ROC.

Fig. 4

Horizontal bar charts of the significantly altered metabolites among the groups of patients.

Fig. 5

Histogram representing metabolites with the greatest impact on the STEMI diagnostics established using the random forest model.

Fig. 6

Pathway analysis results. The pathways are shown in accordance with p values (y-axis) obtained from the pathway enrichment analysis, whereas pathway impact values (x-axis) are responsible for topology analysis. The colours of the pathways reflect the p value. Red corresponds to low statistical significance, and yellow indicates the highest statistical significance.

Fig. 7

Weighted correlation networks: A – STEMI group; B – SAP group; C – non-CVD group. I, II, III – metabolite clusters.

Fig. 8

Horizontal bar charts of the significantly altered ratios of the metabolites based on the weighted correlation network analysis results.

Fig. 9

Significantly altered metabolites and metabolic pathways in STEMI pathogenesis.