Metabolomics studies generate increasingly complex data tables, which are hard to summarize and visualize without appropriate tools. The use of chemometrics tools, e.g., principal component analysis (PCA), partial least-squares to latent structures (PLS), and orthogonal PLS (OPLS), is therefore of great importance as these include efficient, validated, and robust methods for modeling information-rich chemical and biological data. Here the S-plot is proposed as a tool for visualization and interpretation of multivariate classification models, e.g., OPLS discriminate analysis, having two or more classes. The S-plot visualizes both the covariance and correlation between the metabolites and the modeled class designation. Thereby the S-plot helps identifying statistically significant and potentially biochemically significant metabolites, based both on contributions to the model and their reliability. An extension of the S-plot, the SUS-plot (shared and unique structure), is applied to compare the outcome of multiple classification models compared to a common reference, e.g., control. The used example is a gas chromatography coupled mass spectroscopy based metabolomics study in plant biology where two different transgenic poplar lines are compared to wild type. By using OPLS, an improved visualization and discrimination of interesting metabolites could be demonstrated.

The production of 'global' metabolite profiles involves measuring low molecular-weight metabolites (<1 kDa) in complex biofluids/tissues to study perturbations in response to physiological challenges, toxic insults or disease processes. Information-rich analytical platforms, such as mass spectrometry (MS), are needed. Here we describe the application of ultra-performance liquid chromatography-MS (UPLC-MS) to urinary metabolite profiling, including sample preparation, stability/storage and the selection of chromatographic conditions that balance metabolome coverage, chromatographic resolution and throughput. We discuss quality control and metabolite identification, as well as provide details of multivariate data analysis approaches for analyzing such MS data. Using this protocol, the analysis of a sample set in 96-well plate format, would take ca. 30 h, including 1 h for system setup, 1-2 h for sample preparation, 24 h for UPLC-MS analysis and 1-2 h for initial data processing. The use of UPLC-MS for metabolic profiling in this way is not faster than the conventional HPLC-based methods but, because of improved chromatographic performance, provides superior metabolome coverage.

No abstract

Obtaining comprehensive, untargeted metabolic profiles for complex solid samples, e.g., animal tissues, requires sample preparation and access to information-rich analytical methodologies such as mass spectrometry (MS). Here we describe a practical two-step process for tissue samples that is based on extraction into 'aqueous' and 'organic' phases for polar and nonpolar metabolites. Separation methods such as ultraperformance liquid chromatography (UPLC) in combination with MS are needed to obtain sufficient resolution to create diagnostic metabolic profiles and identify candidate biomarkers. We provide detailed protocols for sample preparation, chromatographic procedures, multivariate analysis and metabolite identification via tandem MS (MS/MS) techniques and high-resolution MS. By using these optimized approaches, analysis of a set of samples using a 96-well plate format would take ~48 h: 1 h for system setup, 8-10 h for sample preparation, 34 h for UPLC-MS analysis and 2-3 h for preliminary/exploratory data processing, representing a robust method for untargeted metabolic screening of tissue samples.

Metabolomics represents a paradigm shift in metabolic research, away from approaches that focus on a limited number of enzymatic reactions or single pathways, to approaches that attempt to capture the complexity of metabolic networks. Additionally, the high-throughput nature of metabolomics makes it ideal to perform biomarker screens for diseases or follow drug efficacy. In this Review, we explore the role of metabolomics in gaining mechanistic insight into cardiac disease processes, and in the search for novel biomarkers. High-resolution NMR spectroscopy and mass spectrometry are both highly discriminatory for a range of pathological processes affecting the heart, including cardiac ischemia, myocardial infarction, and heart failure. We also discuss the position of metabolomics in the range of functional-genomic approaches, being complementary to proteomic and transcriptomic studies, and having subdivisions such as lipidomics (the study of intact lipid species). In addition to techniques that monitor changes in the total sizes of pools of metabolites in the heart and biofluids, the role of stable-isotope methods for monitoring fluxes through pathways is examined. The use of these novel functional-genomic tools to study metabolism provides a unique insight into cardiac disease progression.

Metabolite identification studies involve the detection and structural characterization of the biotransformation products of drug candidates. These experiments are necessary throughout the drug discovery and development process. The use of high-resolution chromatography and high-resolution mass spectrometry together with data processing using mass defect filtering is described for in vitro and in vivo metabolite identification studies. Data collection was done using UPLC coupled with an orthogonal hybrid quadrupole time-of-flight mass spectrometer. This experimental approach enabled the use of MS(E) data collection (where E represents collision energy) which has previously been shown to be a powerful approach for metabolite identification studies. Post-acquisition processing with a prototype mass defect filtering program was used to eliminate endogenous interferences in the study samples, greatly enhancing the discovery of metabolites. The ease of this approach is illustrated by results showing the detection and structural characterization of metabolites in plasma from a preclinical rat pharmacokinetic study.

The systemic biochemical effects of oral hydrazine administration (dosed at 75, 90, and 120 mg/kg) have been investigated in male Han Wistar rats using metabonomic analysis of (1)H NMR spectra of urine and plasma, conventional clinical chemistry, and liver histopathology. Plasma samples were collected both pre- and 24 h postdose, while urine was collected predose and daily over a 7 day postdose period. (1)H NMR spectra of the biofluids were analyzed visually and via pattern recognition using principal component analysis. The latter showed that there was a dose-dependent biochemical effect of hydrazine treatment on the levels of a range of low molecular weight compounds in urine and plasma, which was correlated with the severity of the hydrazine induced liver lesions. In plasma, increases in the levels of free glycine, alanine, isoleucine, valine, lysine, arginine, tyrosine, citrulline, 3-D-hydroxybutyrate, creatine, histidine, and threonine were observed. Urinary excretion of hippurate, citrate, succinate, 2-oxoglutarate, trimethylamine-N-oxide, fumarate and creatinine were decreased following hydrazine dosing, whereas taurine, creatine, threonine, N-methylnicotinic acid, tyrosine, beta-alanine, citrulline, Nalpha-acetylcitrulline and argininosuccinate excretion was increased. Moreover, the most notable effect was the appearance in urine and plasma of 2-aminoadipate, which has previously been shown to lead to neurological effects in rats. High urinary levels of 2-aminoadipate may explain the hitherto poorly understood neurological effects of hydrazine. Metabonomic analysis of high-resolution (1)H NMR spectra of biofluids has provided a means of monitoring the progression of toxicity and recovery, while also allowing the identification of novel biomarkers of development and regression of the lesion.

A fast and robust method for lipid profiling utilizing liquid chromatography coupled with mass spectrometry has been demonstrated and validated for the analysis of human plasma. This method allowed quantification and identification of lipids in human plasma using parallel alternating low energy and high energy collision spectral acquisition modes. A total of 275 [corrected] lipids were identified and quantified (as relative concentrations) in both positive and negative ion electrospray ionization mode. The method was validated with five nonendogenous lipids, and the linearity (r(2) better than 0.994) and the intraday and interday repeatability (relative standard deviation, 4-6% and 5-8%, respectively) were satisfactory. The developed lipid profiling method was successfully applied for the analysis of plasma from osteoarthritis (OA) patients. The multivariate statistical analysis by partial least-squares-discrimination analysis suggested an altered lipid metabolism associated with osteoarthritis and the release of arachidonic acid from phospholipids.