Application of UHPLC-MS/MS-based nontargeted metabolomics in plasma metabolism in altitude-related hypertension among high-altitude migrants

Abstract

Objective To investigate the differences in plasma metabolites between patients with altitude-related hypertension (ARH) and healthy individuals, and analyze the potential pathogenesis of ARH. Methods Convenient sampling was conducted on a unit of male healthy officers and soldiers who resident at altitude of < 500 m and migrated to an altitude of 4 200 m in July 2020. Twenty of them diagnosed with ARH were assigned into the ARH group, and another 30 non-ARH individuals served as the control group. Their blood pressure, body mass index (BMI), blood oxygen saturation, and heart rate were measured and recorded, and fasting venous blood samples were harvested to screen and identify plasma metabolites with ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Metabolite fingerprinting was performed using unsupervised Principal Component Analysis (PCA) and supervised Orthogonal Partial Least Squares Discrimination Analysis (OPLS-DA) models in order to assist in biomarker screening. The quality of the OPLS-DA model was assessed and validated to guarantee the stability and reliability of the model. Differential plasma metabolites were screened using independent sample t test and fold change (FC) analysis, and volcano plots were drawn. Finally, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathway analysis was used to perform functional pathway enrichment and topological analysis on the screened differential metabolites. Results Compared to the control group, the ARH group showed significantly higher systolic and diastolic blood pressure and heart rate, and lower arterial oxygen saturation (P < 0.05). PCA analysis showed that 81.96% of the variance was explained in the positive ion mode and 79.25% in the negative ion mode, indicating significant metabolic differences between the 2 groups. OPLS-DA model analysis indicated that in the positive ion mode, PC1 explained 77.36% of the variance and PC2 explained 12.25% of the variance, with R2Y=0.96 and Q2Y=0.91; in the negative ion mode, PC1 explained 84.15% of the variance and PC2 explained 17.24% of the variance, with R2Y=0.99 and Q2Y=0.86. Inter-group difference exceeded 75%, and intra-group difference was less than 20%. The 7-fold cross-validation and 200 permutation test confirmed that the model was stable and reliable. In the positive ion mode, the Y-axis intercepts of the R2 and Q2 fitted lines were 0.58 and -0.48, respectively; in the negative ion mode, the Y-axis intercepts were 0.93 and -0.41, respectively. A total of 32 significantly different metabolites were screened out, including amino acids, nucleosides, fatty acids, and organic alkaloids. KEGG analysis revealed that among the 10 metabolic pathways, 4 were amino acid metabolic pathways, with the aminoacyl-tRNA biosynthesis pathway having the most enriched metabolites. Conclusion Based on UHPLC-MS/MS technology, untargeted metabolomics analysis identifies 32 significantly different metabolites, which may serve as characteristic biomarkers for ARH, and the aminoacyl-tRNA biosynthesis pathway may be associated with the pathogenesis of ARH.

Keywords

• nontargeted metabolomics
• altitude-related hypertension
• metabolic pathways