A High Throughput Lipidomics Method Using Scheduled Multiple Reaction Monitoring

Abstract

Lipid compositions of cells, tissues, and bio-fluids are complex, with varying concentrations and structural diversity making their identification challenging. Newer methods for comprehensive analysis of lipids are thus necessary. Herein, we propose a targeted-mass spectrometry based lipidomics screening method using a combination of variable retention time window and relative dwell time weightage. Using this method, we identified more than 1000 lipid species within 24-min. The limit of detection varied from the femtomolar to the nanomolar range. About 883 lipid species were detected with a coefficient of variance <30%. We used this method to identify plasma lipids altered due to vitamin B12 deficiency and found a total of 18 lipid species to be altered. Some of the lipid species with ω-6 fatty acid chains were found to be significantly increased while ω-3 decreased in vitamin B12 deficient samples. This method enables rapid screening of a large number of lipid species in a single experiment and would substantially advance our understanding of the role of lipids in biological processes.

Keywords: dwell time; isomers; lipidomics; mass spectrometry; plasma lipidome; scheduled MRM; variable RT window; vitamin B12.

Figure 1

Comparison of three different MS methods (Variable RTW-Relative DTW, sMRM and MRM) on three different days. (a) Total number lipid species obtained through Variable RTW-Relative DTW (829, 901 and 852), sMRM (728, 832, and 802) and MRM (141, 143, and 138) methods. Data shown as mean ± SD of experimental replicates on three different days. (b) Comparison of AUC for different lipid species detected through variable RTW- relative DTW method with sMRM and MRM method. The x-axis is the area under the curve (AUC) of XIC (extracted ion chromatogram) for detected lipids in sMRM and MRM and y-axis is AUC for variable RTW- relative DTW. Each dot represents lipid species. Increase in AUC for lipid species is associated with the distance between reference line (diagonal) and the trend line. For equal AUC obtained in different methods, the point should scatter around the reference line while we observed deviation from linearity (reference line) toward y-axis (variable-RTW and relative-DTW). Trend line shift toward y-axis indicate the improved data quality for variable RTW- relative DTW as compared to sMRM and MRM.

Figure 2

XIC (extracted ion chromatogram) of nine isomers of TAG (52:6). Parent m/z for all was 868.8 while the product m/z was derived from the remaining mass (R1 + R2 with glycerol backbone) after the loss of fatty acid released from the parent ion. R1 + R2 can be any composition of fatty acid which sum-up to give product ion. Different colored dots represent different isomers.

Figure 3

Abundance of different lipids. (a(i)) Abundance of different TAGs on the basis of total chain length (as a function of main-chain carbon atoms) and unsaturation. (a(ii)) Abundance of different phospholipids on the basis of total chain length and unsaturation. (b) A total of 339 TAG isomers were detected from 77 different species of TAG. (c) Abundance of 365 phospholipids belonging to 6 classes (PC, PE, PG, PI, PS, and PA), different dots of same color represent isomers. The abundance of the difference lipids/isomers is represented by the varying size of the bubble in all of the panels.

Figure 4

Representative graphs from positive and negative ion mode showing LoD, LoQ and coefficient of determination, x and y-axis was log transformed. (a) SM from positive ion mode and (b) PC from negative ion mode. The grey area represents the concentration below the linear range while the yellow region is indicative of linear range. The error bar represents the variance/standard deviation obtained in three replicates, reflecting inter day variability.

Figure 5

Validation of the method. (a) Spike and recovery of different lipid classes where blue bars represent the recovery of lipids when known concentration of lipid standards was spiked during extraction and green bars represents the reference (same concentration of lipid standard spiked after extraction). (b) Coefficient of variance on day 2 where 1043 lipid species from 16 lipid classes were detected (n = 5). The color scale of the bubble is based on the function of coefficient of variance in the increasing order. The right-hand panel represents the density function w.r.t. coefficient of variance.

Figure 6

Significantly altered lipid species in vitamin B₁₂ deficiency. The significance level for different lipid species are represented by p-value ( * = p < 0.05, ** = p < 0.01 and *** = p < 0.001).