Accurate quantification of lipid species affected by isobaric overlap in Fourier-transform mass spectrometry

Marcus Höring,Christer S Ejsing,Sabrina Krautbauer,Verena M Ertl,Ralph Burkhardt,Gerhard Liebisch

doi:10.1016/j.jlr.2021.100050

Abstract

AbstractLipidomics data require consideration of ions with near-identical masses, which comprises among others the Type-II isotopic overlap. This overlap occurs in series of lipid species differing only by number of double bonds (DBs) mainly because of the natural abundance of 13C-atoms. High-resolution mass spectrometry, such as Fourier-transform mass spectrometry (FTMS), is capable of resolving Type-II overlap depending on mass resolving power. In this work, we evaluated FTMS quantification accuracy of lipid species affected by Type-II overlap. Spike experiments with lipid species pairs of various lipid classes were analyzed by flow injection analysis-FTMS. Accuracy of quantification was evaluated without and with Type-II correction (using relative isotope abundance) as well as utilizing the first isotopic peak (M+1). Isobaric peaks, which were sufficiently resolved, were most accurate without Type-II correction. In cases of partially resolved peaks, we observed peak interference causing distortions in mass and intensity, which is a well-described phenomenon in FTMS. Concentrations of respective species were more accurate when calculated from M+1. Moreover, some minor species, affected by considerable Type-II overlap, could only be quantified by M+1. Unexpectedly, even completely unresolved peaks were substantially overcorrected by Type-II correction because of peak interference. The described method was validated including intraday and interday precisions for human serum and fibroblast samples. Taken together, our results show that accurate quantification of lipid species by FTMS requires resolution-depended data analysis.

Highlights

Supplementary key words lipidomics mass spectrometry lipids phospholipids sphingolipids triglycerides isotope correction data processing Fourier-transform mass spectrometry peak interferenc
Lipid standard pairs were analyzed with a constant concentration of species Mi and increasing concentrations of species Mi+1 on a QExactive Orbitrap using Fourier-transform mass spectrometry (FTMS) at a mass resolution setting of 140,000 [full-width at half maximum (FWHM) at m/z 200]
We evaluated the quantification of lipid species by flow injection analysis (FIA)-FTMS with particular consideration of the isobaric overlap occurring in double bond series

Summary

MATERIALS AND METHODS

Chloroform and 2-propanol were purchased from Roth (Karlsruhe, Germany) and methanol from Merck (Darmstadt, Germany). The mixture “M0” contained the following lipid species: LPE 18:0 (0.018 pmol/μl, concentration in the infusate), LPC 18:0 (0.18 pmol/μl), DG 36:0 (0.18 pmol/μl), CE 18:0 (0.18 pmol/μl), SM 36:1;2 (0.18 pmol/μl), Cer 42:1;2 (0.018 pmol/μl), and TG 54:0 (0.18 pmol/μl). Negative ion mode FTMS data were recorded in m/z range 400–650 for LPE ([M–H]−) and LPC ([M+HCOO]−) and m/z range 520–960 for Cer, SM, and PC quantification (as [M+HCOO]−). Full scan FTMS data were acquired in positive ion mode for 2 min in scan range m/z 470–1,030 using profile mode, a max injection time of 100 ms, AGC of 1 × 105, three microscans, and a target resolution setting of 480,000 and 120,000. The quantification was performed with nonendogenous internal standards by multiplication of the spiked IS amount with the analyte-to-IS ratio of the intensities after data processing

RESULTS

Evaluation of peak interference

DISCUSSION