Introduction: In this paper, an analytical pipeline designed for untargeted lipidomic profiling in human plasma is proposed. The analytical pipeline was developed for case-control studies nested in prospective cohorts.
Methods: The procedure consisted of isopropanol protein precipitation followed by reverse phase liquid chromatography coupled to high resolution mass spectrometry and software-assisted data processing. The compounds are putatively annotated by matching experimental mass spectrometry data with spectral library data using LipidSearch software. The lipid profile of a pool of plasma samples from 10 healthy volunteers was detected in both positive and negative polarity modes. The impact of the chosen polarity on the number and quality of the lipid identification has been evaluated.
Results: More than 1000 lipids from 12 different classes were detected, 1150 in positive mode and 273 in negative mode. Nearly half of them were unambiguously identified by the software in positive mode, and about one-third in negative mode. The method repeatability was assessed on the plasma pool samples by means of variance components analysis. The intra- and inter-assay precision was measured for 10 lipids chosen among the most abundant found within the different lipid classes. The intra-assay coefficients of variation ranged from 2.56% to 4.56% while intra- and inter-day coefficients of variance never exceeded the 15% benchmark adopted. The lipidomic profiles of the 10 healthy volunteers were also investigated.
Discussion: This method detects a wide range of lipids and reports their degree of identification. It is particularly fit and well-designed for large case-control epidemiologic studies.
Keywords: Clinical and molecular epidemiology; epidemiology; mass spectrometry methods; metabolomics methods.