Research And Application Of Metabolomic Memodology Based On Different Analytical Methods

Metabolomics, a relatively novel methodology arising from the post-genomics era, has been increasingly recognized as a valuable complementary approach to other well-established ’omic’ sciences to aid in the assessment of disease and toxicity. Metabolic profiling is the comprehensive studying of large numbers of all endogenous low molecular weight metabolites and their roles in various disease states in a global view. In the present study, The proton nuclear magnetic resonance (1H NMR), rapid resolution liquid chromatography (RRLC) and high performance liquid chromatography-mass spectrometry (HPLC-MS) based metabolomic methodology were established and utilized for the research of pathophysiological mechanisms of Arthus reaction, tumor biomarkers discovery and nanotoxicity. The main contents and results are as follows:1. A RRLC method was developed for the simultaneous determination of23amino acids in rat and human serum after pre-column derivatization with2,4-dinitrofluorobenzene (DNFB). The amino acid derivatives were separated on an Agilent Zorbax Eclipse Plus Cis (4.6mm x50mm,1.8um) column at45℃. Ultraviolet (UV) detection was set at360nm. Good separation of23amino acids was achieved within10min with a ternary gradient elution of mobile phase at a flow rate of1.5mL·min-1. Calibration curves were linear over the range from1to500μmol·L-1with coefficients0.9962or better for each amino acid. The lower limits of quantification (LLOQ) of all23amino acids were1.0μmol·L-1with signal-to-noise (S/N) ratio≥14. Intra-and Inter-day precisions, expressed as relative standard deviation (RSD) percentages, were ranged from0.32%to3.09%and0.67%to5.82%, respectively. Finally, it was successfully applied to the determination of amino acids in rat and human serum with recoveries ranged from90.8%to106.0%and88.2%to106.4%, and RSD percentages ranged from1.78%to4.68%and1.4%to8.5%, respectively.2. Arthus reaction (AR), a type of unconventional immune complex-mediated inflammation, is likely accompanied by alterations in circulating metabolites. Here, a1H NMR spectroscopy method coupled with a RRLC method was developed to evaluate the systemic metabolic consequences of AR and characterize metabolic aberrations. Serum and urine samples from AR rats and normal controls were compared to determine whether there were significant alterations associated with AR. The partial least squares discriminant analysis (PLS-DA) models of metabolomic results demonstrated good intergroup separations between AR rats and normal controls. Multivariate statistical analysis revealed significant alterations in the levels of35metabolites, which were termed as the disease-associated biomarkers. Differential metabolites identified from the metabolomic analysis suggested that AR caused dysfunctions of kidney and liver accompanied with changes in widespread metabolic pathways including the tricarboxylic acid (TCA) cycle, gut microbiota metabolism, lipids and cell membranes metabolism, glucose metabolism, fatty acid β3-oxidation, amino acids metabolism and ketogenesis.3. Here, a1H NMR-based metabolomic approach coupled with a relative simple RRLC-based focused metabolomic approach was developed and compared to characterize the systemic metabolic disturbances underlying esophageal cancer (EC) and identify possible early biomarkers for clinical prognosis. Serum metabolic profiling of patients with EC (n=25) and healthy controls (n=25) was performed by using both1H NMR and RRLC, and metabolite identification was achieved by multivariate statistical analysis. Using orthogonal partial least-squares-discriminant analysis (OPLS-DA), we could distinguish EC patients from healthy controls. The predictive power of the model derived from the RRLC-based focused metabolomics performed better in both sensitivity and specificity than the results from the NMR-based metabolomics, suggesting that the focused metabolomic technique may be of advantage in the future for the determination of biomarkers. Moreover, focused metabolic profiling is highly simple, accurate and specific, and should prove equally valuable in metabolomic research applications. A total of nineteen significantly altered metabolites were identified as the potential disease associated biomarkers. Significant changes in lipid metabolism, amino acid metabolism, glycolysis, ketogenesis, tricarboxylic acid (TCA) cycle and energy metabolism were observed in EC patients compared with the healthy controls.4. In this study, we used a’H NMR-based metabolomic approach and a RRLC-based focused metabolomic approach and profiled small-molecule metabolites in serum of25lung cancer patients and25healthy controls, to determine whether there are significant alterations associated with carcinogenesis. The metabolomic results demonstrate clear intergroup separations between healthy control subjects and lung cancer patients in the OPLS-DA models.15differential metabolites identified from both the two metabolomic analysis suggest a significantly up-regulated glycolysis and the disrupted lipid metabolism, amino acid metabolism and ketogenesis in lung cancer patients.5. In the present study, the toxicity of ZnO quantum dots (QDs) was investigated using HPLC-MS based metabolic profiling. The serum and urine of rats treated with ZnO QDs-PEG at different doses (1and10mg/kg) were analyzed by integrated metabolomic analysis and in combination with multivariate statistical analysis. All the results indicated that both the doses of1 and10mg/kg ZnO QDs-PEG can induce changes in endogenous metabolic profiles, but no overt sign of toxicity was found in either the two group. In addition, all the data generated from the current study completely supports the fact that the ZnO QDs-PEG are little toxic and metabolic profiling is promising for the development of a rapid in vivo screening tool for nanotoxicity.