Four-dimensional lipidomics profiling in X-linked adrenoleukodystrophy using trapped ion mobility mass spectrometry

Abstract

Lipids play pivotal roles in an extensive range of metabolic and physiological processes. In recent years, the convergence of trapped ion mobility spectrometry and MS has enabled 4D-lipidomics, a highly promising technology for comprehensive lipid analysis. 4D-lipidomics assesses lipid annotations across four distinct dimensions-retention time, collisional cross section, m/z (mass-to-charge ratio), and MS/MS spectra-providing a heightened level of confidence in lipid annotation. These advantages prove particularly valuable when investigating complex disorders involving lipid metabolism, such as adrenoleukodystrophy (ALD). ALD is characterized by the accumulation of very-long-chain fatty acids (VLCFAs) due to pathogenic variants in the ABCD1 gene. A comprehensive 4D-lipidomics strategy of ALD fibroblasts demonstrated significant elevations of various lipids from multiple classes. This indicates that the changes observed in ALD are not confined to a single lipid class and likely impacts a broad spectrum of lipid-mediated physiological processes. Our findings highlight the incorporation of mainly saturated and monounsaturated VLCFA variants into a range of lipid classes, encompassing phosphatidylcholines, triacylglycerols, and cholesterol esters. These include ultra-long-chain fatty acids with a length of up to thirty carbon atoms. Lipid species containing C26:0 and C26:1 were the most frequently detected VLCFA lipids in our study. Furthermore, we report a panel of 121 new candidate biomarkers in fibroblasts, exhibiting significant differentiation between controls and individuals with ALD. In summary, this study demonstrates the capabilities of a 4D-lipid profiling workflow in unraveling novel insights into the intricate lipid modifications associated with metabolic disorders like ALD.

Keywords: 4D-Lipidomics; adrenoleukodystrophy; parallel accumulation serial fragmentation; trapped ion mobility spectrometry; very-long-chain fatty acids.

Graphical abstract

Fig. 1

4D lipidomics profiling using PASEF (A) 4D lipidomics workflow. Samples are extracted using methanol:chloroform, followed by LC and TIMS mass spectrometry. Lipids are detected using PASEF which enables efficient precursor selection and MSMS acquisition. Features are identified using accurate mass, retention time, isotopic pattern, MSMS spectra, and CCS. B: Bar chart of detected lipid species per class. C: Three-dimensional representation (RT, m/z, CCS) of detected lipids. Different colors represent different lipid classes. D: Chromatogram of PE 34:1 (blue) and PC O-32:1 (red) show overlapping peaks in the LC dimension but are not overlapping in the ion mobility dimension. PASEF enables MSMS coverage for nearly every ion mobility resolved peak leading to noninterfering MSMS spectra of PE 34:1 and PC O-32:1. CCS, collisional cross section; PASEF, parallel accumulation serial fragmentation; PC, phosphatidylcholine; PE, phosphatidylethanolamine; RT, retention time; TIMS, trapped ion mobility spectrometry.

Fig. 2

ALD induces in significant changes in the lipidome. A: Volcano plot of lipid levels. The vertical axis contains the P-value (−log10) from Welch t-tests between ALD and controls, and the horizontal axis the fold change (log2) between ALD and controls. Colored dots are lipids with a P-value of <0.05. Each color represents a different lipid class. B: PCA analysis shows distinct group of ALD and control fibroblasts. C: Heatmap of Z-scores of top 50 lipids based on P-value sorted on lipid class. D: Box plot of the newborn screening marker LPC 26:0 in ALD and control fibroblasts (∗∗∗∗P < 0.0001). ALD, adrenoleukodystrophy; LPC, lysophosphatidylchloline; PCA, principal component analysis.

Fig. 3

Characterization of VLCFA in the ALD lipidome. A: Trend lines illustrating the log fold change (logFC) and the total chain length of LPC, PC, and TG lipids when ALD is compared to controls. B: Heatmap illustrating the log fold change, total chain unsaturation and chain length for LPC, PC, and TG when ALD is compared to controls. C: Bar plot of amount of detected VLCFA containing species for LPC, PC, and TG. D: Sum of lipid species containing specific VLCFA variants for LPC, PC, and TG. ALD, adrenoleukodystrophy; LPC, lysophosphatidylchloline; PC, phosphatidylcholine; TG, triacylglycerol; VLCFA, very-long-chain fatty acid.

Fig. 4

4D-lipidomics allows for the identification of new biomarker candidates in ALD. Examples of biomarker candidates in ALD are presented. Lipid abundances were calculated by dividing the analyte response by that of the corresponding internal standard (ratio to IS). (∗∗∗∗P < 0.0001). ALD, adrenoleukodystrophy.