A new approach to untargeted integration of high resolution liquid chromatography–mass spectrometry data

A relative quantitative positive/negative ion switching method for untargeted lipidomics via high resolution LC-MS/MS from any biological source

INTRODUCTION

Advances in high-resolution mass spectrometry have created renewed interest for studying global lipid biochemistry in disease and biological systems.

OBJECTIVES

Here, we present an untargeted 30 min. LC-MS/MS platform that utilizes positive/negative polarity switching to perform unbiased data dependent acquisitions (DDA) via higher energy collisional dissociation (HCD) fragmentation to profile more than 1000-1500 lipid ions mainly from methyl-tert-butyl ether (MTBE) or chloroform:methanol extractions.

METHODS

The platform uses C18 reversed-phase chromatography coupled to a hybrid QExactive Plus/HF Orbitrap mass spectrometer and the entire procedure takes ~10 h from lipid extraction to identification/quantification for a data set containing 12 samples (~4 h for a single sample). Lipids are identified by both accurate precursor ion mass and fragmentation features and quantified using Lipid-Search and Elements software.

RESULTS

Using this approach, we are able to profile intact lipid ions from up to 18 different main lipid classes and 66 subclasses. We show several studies from different biological sources, including cultured cancer cells, resected tissues from mice such as lung and breast tumors and biological fluids such as plasma and urine.

CONCLUSIONS

Using mouse embryonic fibroblasts, we showed that TSC2-/- KD significantly abrogates lipid biosynthesis and that rapamycin can rescue triglyceride (TG) lipids and we show that SREBP-/- shuts down lipid biosynthesis significantly via mTORC1 signaling pathways. We show that in mouse EGFR driven lung tumors, a large number of TGs and phosphatidylmethanol (PMe) lipids are elevated while some phospholipids (PLs) show some of the largest decrease in lipid levels from ~ 2000 identified lipid ions. In addition, we identified more than 1500 unique lipid species from human blood plasma.

Post-acquisition spectral stitching. An alternative approach for data processing in untargeted metabolomics by UHPLC-ESI(-)-HRMS

INTRODUCTION

In the case of the MS-based metabolomics, the large number of false positives remains a fundamental issue.

OBJECTIVE

The aim of this study was to develop a new strategy, which highlights the number of the reliable features i.e. the detected features that correspond to a consistent peak according to chromatographic and mass spectrometric criteria.

METHOD

For the analysis blood samples from 20 chickens, which were administrated with naringin and 9 samples from control, were analyzed by UHPLC-HRMS (Orbitrap Velos). Two methodologies have been compared for data processing. In the first one (classical approach), all data in the 100-900 m/z mass-to charge range were included for the data processing procedure whereas for the newly developed methodology, the data were shred in 100Da slices generating 8 datasets, which have been then subjected to the downstream MS data processing. Each dataset was treated separately and the m/z_t_R features obtained by either VIP's or t-test values were merged and used as the input for the construction of the general model.

RESULTS

The new methodology resulted to a 4-fold increase of the peaks that could be considered chromatographically and mass spectrometrically valid.

CONCLUSION

A new strategy was reported on the detection of chromatographically reliable features during a metabolomic approach. The shredding of the LC-MS chromatograms into multiple m/z ranges increased the number of the identified chromatographically reliable features.

Non-targeted UHPLC-MS metabolomic data processing methods: a comparative investigation of normalisation, missing value imputation, transformation and scaling

INTRODUCTION

The generic metabolomics data processing workflow is constructed with a serial set of processes including peak picking, quality assurance, normalisation, missing value imputation, transformation and scaling. The combination of these processes should present the experimental data in an appropriate structure so to identify the biological changes in a valid and robust manner.

OBJECTIVES

Currently, different researchers apply different data processing methods and no assessment of the permutations applied to UHPLC-MS datasets has been published. Here we wish to define the most appropriate data processing workflow.

METHODS

We assess the influence of normalisation, missing value imputation, transformation and scaling methods on univariate and multivariate analysis of UHPLC-MS datasets acquired for different mammalian samples.

RESULTS

Our studies have shown that once data are filtered, missing values are not correlated with m/z, retention time or response. Following an exhaustive evaluation, we recommend PQN normalisation with no missing value imputation and no transformation or scaling for univariate analysis. For PCA we recommend applying PQN normalisation with Random Forest missing value imputation, glog transformation and no scaling method. For PLS-DA we recommend PQN normalisation, KNN as the missing value imputation method, generalised logarithm transformation and no scaling. These recommendations are based on searching for the biologically important metabolite features independent of their measured abundance.

CONCLUSION

The appropriate choice of normalisation, missing value imputation, transformation and scaling methods differs depending on the data analysis method and the choice of method is essential to maximise the biological derivations from UHPLC-MS datasets.

Identification strategies for flame retardants employing time-of-flight mass spectrometric detectors along with spectral and spectra-less databases

In the past, the preferred strategy for the identification of unknown compounds was to search in an appropriate mass spectral database for spectra obtained using either electron ionisation (GC-MS analyses) or collision-induced dissociation (LC-MS/MS analyses). Recently, an increase has been seen in the use of accurate mass instruments and spectra-less databases, based on monoisotopic accurate mass alone. In this article, we describe a systematic workflow for the screening and identification of new flame retardants. This approach utilises LC-quadrupole-time-of-flight MS and spectra-less databases based only on monoisotopic accurate mass for the identification of 'unknowns'. An in-house database was built, and the input parameters used in the data analysis process were optimised for flame retardant chemicals, so that it can be easily transferred to other laboratories. The procedure was successfully applied to dust, foam and textiles from car interiors and indoor consumer products. The developed method was demonstrated for the main new flame retardant present in Antiblaze V6 and for the three unreported reaction by-products/impurities present in the same technical mixture. Copyright © 2015 John Wiley & Sons, Ltd.