Metabolic profiling of human lung injury by 1H high-resolution nuclear magnetic resonance spectroscopy of blood serum

Abstract

Acute respiratory distress syndrome (ARDS) contributes substantially to mortality and morbidity in USA and worldwide. Due to limitations in early diagnostics of ARDS by classical methods, there has been need for discovery of novel methods and biomarkers for its characterization. We present here first high-resolution 1H nuclear magnetic resonance (NMR) metabolic profiling of serum from ARDS patients and non ARDS (NARDS) controls to search for novel biomarkers in blood serum for better diagnostics and prognostics. We have carried out study with serum samples from a total of 45 subjects, which included 26 ARDS patients and 19 NARDS controls. Principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA) were performed on 1H NMR spectra of serum for group discrimination between the two. PCA and PLS-DA on 1H NMR spectra of serum clearly discriminated between NARDS control and ARDS diseased samples. In our study, lipid resonances showed major contribution for this separation in loading plots. In order to highlight role of other small molecular weight metabolites, the analysis was also carried out after removing lipids resonances from NMR spectrum. There was significant increase in concentrations of N-acetylglycoproteins (NAC) (p = 0.001), acetoacetate (p = 0.001), lactate (p = 0.001), creatinine (p = 0.003), histidine (p = 0.03), formate (p = 0.04) and aromatic amino acids serum of ARDS patients. Along with small metabolites, lipids play a very important role in this discrimination and can differentiate between two classes. Our study has given new avenue in the metabolic profiling of lung injuries.