Fig. 1: 1-deoxyDHCeramides are the primary mediators of 1-deoxySphingolipids induced toxicity.
From: Very long-chain fatty acids drive 1-deoxySphingolipid toxicity

a Schematic representation of 1-deoxySL metabolism. The horizontal arrow indicates desaturation of 1-deoxysphinganine (1-deoxySa, m18:0) at the Δ14(Z) position catalyzed by fatty acid desaturase 3 (FADS3). Vertical arrows represent N-acylation of 1-deoxySa by ceramide synthase isoforms (CERS1–6). The blue box highlights the acyl-CoA substrate specificities of individual CERS isoforms. b 1-deoxySa (m18:0) dose-response toxicity curves in HeLa cells in the presence or absence of the pan-CERS inhibitor fumonisin B1 (FB1, 35 µM). Toxicity was determined by CellTiter-GLO assay as described in the methods section. IC50 values are indicated below the individual treatments. c, d Formation of d3-1-deoxyDHCer (m18:0/xx:y+3) species in HeLa cells supplemented with d3-1-deoxySa and treated with pan-Ceramide synthase inhibitor Fuminosin B1 (FB1, 35 µM) or Vehicle. e Comparison of cytotoxicity of 1-deoxySa or (m18:0) 1-deoxysphingosine (1-deoxySo, m18:1, Δ14(Z)) on HeLa cells. f, g Intracellular accumulation of 1-deoxyDHCer and 1-deoxyCer species formed in HeLa cells supplemented with 1-deoxySa (m18:0) or 1-deoxySo (m18:1, Δ14(Z)). All lipid data experiments were performed with n = 3 biologically independent replicates (separately cultured and treated HeLa cell populations) and measured by an untargeted high-resolution LC-MS/MS lipidomics workflow and normalized to an internal standard and cell count. Data are shown as mean ± SD. Significance was determined using a two-tailed unpaired Student’s t-test with multiple testing correction (two-stage step-up method of Benjamini, Krieger, and Yekutieli). Reported p.adjusted values: p < 0.05 (*), p < 0.01 (**), p < 0.001(***). Exact p-values and statistical source data are provided in Source Data 1. Toxicity data were normalized to vehicle treated cells and represented as mean ± SD (n = 4, separately cultured and treated HeLa cell populations). Non-linear regression analysis was used to calculate toxicity curves and IC50 values (95% Confidence Interval). Illustrations were created in BioRender. Hornemann, T. (2025) https://BioRender.com/ifocfkk.