A three-step assay for ceramide synthase activity using a fluorescent substrate and HPLC.

Ceramides are a family of signalling lipids with diverse physiological functions that include pro-differentiative and pro-apoptotic signalling. Ceramides and their derivatives are major constituents of myelin, maintaining neuronal conductivity. Ceramides are synthesized by ceramide synthases, of which there are six isoforms in mammals (CERS1-6). These enzymes catalyse the transfer of a variable length fatty acid to a sphingoid base, typically sphingosine or dihydrosphingosine. We previously reported a fluorescent thin-layer chromatography assay for ceramide synthase activity. In this paper we describe an improved fluorescent assay, using HPLC to achieve clear resolution of closely related ceramide species and to facilitate easy quantification of both product and substrate. Our HPLC assay protocol eliminates the need for a chloroform extraction step. Instead a simple three-step procedure is used: (1) reactions are run; (2) reactions are terminated with addition of methanol and centrifuged; (3) products are quantified with HPLC. HPLC resolution enables assays in which multiple fatty acid substrates are used in the same reaction. Using this approach, we show that CERS2 demonstrates a preference for the monounsaturated C24:1 fatty acid substrate compared to the saturated C24:0 substrate, potentially explaining why myelin is enriched in ceramides containing the monounsaturated form of very long chain fatty acids.

Cell-type-specific expression pattern of ceramide synthase 2 protein in mouse tissues.

Ceramide synthase 2 (CerS2) catalyzes the synthesis of dihydroceramides from dihydrosphingosine and very long fatty acyl (C22-C24)-CoAs. CerS2-deficient (gene trap) mice were reported to exhibit myelin and behavioral abnormalities, associated with the expression of CerS2 in oligodendrocytes and neurons based on expression of lacZ reporter cDNA instead of the cers2 gene in these mice. In order to clarify the cell-type-specific expression of CerS2 protein, we have raised antibodies that specifically recognize the glycosylated and non-glycosylated CerS2 protein in wild-type but not in CerS2-deficient mouse tissues. In early postnatal, juvenile and adult mouse brain, the new antibodies detect CerS2 protein only in oligodendrocytes but not in neurons, suggesting that the gene trap vector in CerS2-deficient mice led to ectopic expression of the lacZ reporter gene in neurons. In liver, the CerS2 protein is expressed in hepatocytes but not in Ito cells or Kupffer cells. We conclude that the behavioral abnormalities observed in CerS2-deficient mice originate primarily in oligodendrocytes and not in neurons. The identification of specific cell types in which CerS2 protein is expressed is prerequisite to further mechanistic characterization of phenotypic abnormalities exhibited by CerS2-deficient mice. The amount of CerS2 protein detected in different tissues by immunoblot analyses does not strictly correspond to the activity of the CerS2 enzyme. Disproportional results are likely due to post-translational regulation of the CerS2 protein.