Background: Ceramides and their role in various signaling pathways are believed to regulate distinct cellular effects. Accumulation of ceramides seems to be associated with an inflammatory phenotype negatively affecting cellular viability and mitochondrial metabolism.

Aim: The purpose of our study is to identify underlying ceramide signaling pathways and characterize downstream effects on cell survival and metabolism.

Methods: Ceramide Synthase 2 (CerS2) was overexpressed in human fibroblasts (hFSF) with the p3xFLAG-CMV7 vector (FLAG-CerS2). The empty vector was used as transfection control (FLAG-TC). In parallel cells were preincubated with MitoTempo (20 µM, 2h, MT) or Fumonisin B (100 µM, 4h, FuB) to mitigate CerS2 effects. Transfection success as well as expression of other proteins of interest was measured with Western Blot (WB) and immunofluorescence (IF). Ceramide levels were determined with mass spectrometry. Illustrations were done with laser scanning microscopy (LSM) or electron microscopy. Gene expression was assessed with qPCR. Mitochondrial viability and activity were measured with seahorse analysis.

Results: CerS2 overexpression (1.6 ± 0.18, p<0.001 for IF, 4.2 ± 0.44, p=0.04 for WB) led to increased long chain and very long chain ceramides (C14:0-C24:1) (2.12 ± 0.08, p=0.009), with the highest peak for C24:0 (2.9 ± 0.15, p=0.006). Coherently to the increase of proinflammatory ceramides, cytokines IL-6 (2.0 ± 0.40, p=0.03) and IL-1β (8.5 ± 0.57, p=0.05) as well as mitochondrial ROS (1.60 ± 0.12, p<0.001) were elevated. Preincubation with MT as well as FuB reduced cytokine and mitochondrial ROS level. Mitochondrial impairment was assessed with MitoTracker staining, electron microscopy, qPCR and seahorse analysis. qPCR revealed reduction of the mitochondria fusion genes MFN1 (0.86 ± 0.08, p=0.04) and MFN2 (0.68 ± 0.10, p=0.02), while seahorse analysis verified mitochondrial damage through reduced ATP production (0.82 ± 0.10, p=0.003). Preincubation with MT and FuB reversed decreased MFN1 and MFN2 level, but were not strong enough to reverse reduced ATP production. CerS2 overexpression increased fatty acid associated gene expression of CD36 (3.61 ± 0.26, p<0.001) and reduced CPT1B (0.42 ± 0.65, p=0.02) expression. CD36 was reversed to non- significant level after FuB preincubation but not with MT.

Conclusion: CerS2 overexpression led to increased ceramide production causing lipotoxicity accompanied by inflammation characterized and mediated by increased cytokines and mitochondrial ROS level. The subsequent impairment of mitochondrial structure and function was in part reversed by the reduction of ceramides with FuB or reduction or mitochondrial ROS with MT improving cell viability. Thus, regulation of ceramide levels through control of CerS2 activity might represent a novel therapeutic option for the control of cardiotoxicity.