Structure-specific, quantitative methods for analysis of sphingolipids by liquid chromatography-tandem mass...

Due to the large number of highly bioactive subspecies, elucidation of the roles of sphingolipids in cell structure, signaling, and function is beginning to require that one perform structure-specific and quantitative (i.e., "sphingolipidomic") analysis of all individual subspecies, or at least of those are relevant to the biologic system of interest. As part of the LIPID MAPS Consortium, methods have been developed and validated for the extraction, liquid chromatographic (LC) separation, and identification and quantitation by electrospray ionization (ESI), tandem mass spectrometry (MS/MS) using an internal standard cocktail that encompasses the signaling metabolites (e.g., ceramides, ceramide 1-phosphates, sphingoid bases, and sphingoid base 1-phosphates) as well as more complex species (sphingomyelins, mono- and di-hexosylceramides). The number of species that can be analyzed is growing rapidly with the addition of sulfatides and other complex sphingolipids as more internal standards become available. This review describes these methods as well as summarizes others from the published literature. Sphingolipids are an amazingly complex family of compounds that are found in all eukaryotes as well as some prokaryotes and viruses. The size of the sphingolipidome (i.e., all of the individual molecular species of sphingolipids) is not known, but must be immense considering mammals have over 400 headgroup variants (for a listing, see http://www.sphingomap.org), each of which is comprised of at least a few-and, in some cases, dozens-of lipid backbones. No methods have yet been developed that can encompass so many different compounds in a structurally specific and quantitative manner. Nonetheless, it is possible to analyze useful subsets of the sphingolipidome, such as the backbone sphingolipids involved in signaling (sphingoid bases, sphingoid base 1-phosphates, ceramides, and ceramide 1-phosphates) and metabolites at important branchpoints, such as the partitioning of ceramide into sphingomyelins, glucosylceramides, galactosylceramides, and ceramide 1-phosphate versus turnover to the backbone sphingoid base. This review describes methodology that has been developed as part of the LIPID MAPS Consortium (www.lipidmaps.org) as well as other methods that can be used for sphingolipidomic analysis to the extent that such is currently feasible. The focus of this review is primarily mammalian sphingolipids; hence, if readers are interested in methods to study other organisms, they should consult the excellent review by Stephen Levery in another volume of Methods in Enzymology (Levery, 2005), which covers additional species found in plants, fungi, and other organisms. It should be noted from the start that although many analytical challenges remain in the development of methods to analyze the full "sphingolipidome," the major impediment to progress is the limited availability of reliable internal standards for most of the compounds of interest. Because it is an intrinsic feature of mass spectrometry that ion yields tend to vary considerably among different compounds, sources, methods, and instruments, an analysis that purports to be quantitative will not be conclusive unless enough internal standards have been added to correct for these variables. Ideally, there should be some way of standardizing every compound in the unknown mixture; however, that is difficult, if not impossible, to do because the compounds are not available, and the inclusion of so many internal standards generates a spectrum that may be too complex to interpret. Therefore, a few representative internal standards are usually added, and any known differences in the ion yields of the analytes of interest versus the spiked standard are factored into the calculations. Identification of appropriate internal standards has been a major focus of the LIPID MAPS Consortium, and the methods described in this review are based on the development of a certified (i.e., compositionally and quantitatively defined by the supplier) internal standard cocktail that is now commercially available (Avanti Polar Lipids, Alabaster, AL). For practical and philosophical reasons, an internal standard cocktail was chosen over the process of an investigator adding individual standards for only the analytes of interest. On the practical level, addition of a single cocktail minimizes pipetting errors as well as keeping track of whether each internal standard is still usable (e.g., has it degraded while in solution?). Philosophically, the internal standard cocktail was chosen because an underlying premise of systems analysis asserts that, due to the high relevancy of unexpected interrelationships involving more distant components, one can only understand a biological system when factors outside the primary focus of the experiment have also been examined. Indeed, the first payoffs of "omics" and systems approaches involve the discoveries of interesting compounds in unexpected places when a "sphingolipidomic" analytical method was being used as routine practice instead of a simpler method that would have only measured the compound initially thought to be important (Zheng et al., 2006). Thus, routine addition of a broad internal standard cocktail at the outset of any analysis maximizes the opportunity for such discoveries, both at the time the original measurements are made and when one decides to return to the samples later, which can fortunately be done for many sphingolipids because they remain relatively stable in storage.