Trace lipid detection with laser-induced fluorescence. The sensitivity of mass spectrometry, especially ion traps and triple quad SRM mode MS’s, has increased greatly in recent years, with detection limits in some cases in the low attomole range. Laser-induced fluorescence is still the king of detection limits, allowing single molecule detection and even spectroscopy in favorable cases. LIF can access the lower 6 orders of magnitude in mass between 1 attomole (1 amol) and 1 molecule (about 2 ymol; ymol=yoctomole, 10-24 mol). Because very active compounds involved in cell signaling, neurotransmission, and other tightly controlled biochemical cascades are generally present only for short times in very confined spaces, detection can be a challenge. Sampling with normal sampling methods, such as extraction or microdialysis, dilute the sample considerably, while direct sampling methods tend to isolate very tiny masses of compound. Either way, the capabilities of LIF would appear to be perfect for such analyses, and indeed it is widely used for that very purpose-for easily tagged functional groups like amines (which includes proteins and peptides). Quantitation can be a considerable challenge, but when the concentration drops below roughly 100 nM, just getting a tag onto a bioactive molecule is itself a challenge. We have used several approaches over time, including optimization of kinetics for maximum product and adsorption (J. Williams PhD dissertation) onto a solid phase sorbent for preconcentration and in-situ derivatization (see publication #21). Our current approach, which is by far the most versatile, is to confine the reaction to a rapidly mixed microdroplet. This approach allows the most efficient kinetic approach and provides for the derivatization of amol amounts of analyte or less. The focus is currently on fatty amine compounds, fatty acids, and fatty acid amides.

Multidimensional HPLC of lipids. The primary tool for the analysis of lipids is mass spectrometry. We use an ion trap mass spectrometer with gas chromatographic sample introduction and electron impact and chemical ionization interfaces for the analysis of simple acyls, but GC/MS(/MS) has its limitations with respect to the breadth of samples it can easily accomodate. Tandem MS with electrospray (ESI) and atmospheric pressure chemical ionization (APCI) is much more versatile, especially when coupled with high resolution mass analysis. We use the following mass spectrometers: QTOF, TOF, QqQ, and for simple analysis, a single quadrupole MS. However versatile MS is, there are some issues that require the introduction of the sample following chromatographic separation rather than by direct infusion-ion suppression by large fluxes of overabundant lipids (e.g. membrane lipids) when quantifying trace lipids and isobaric interferences (e.g. isomers). However, lipid samples are very complex, with a wide range of polarity, structure, and molecular weight. Single column separations often lack the peak capacity to handle a given lipid sample, depending on how much pre-LC sample preparation was done. In addition, when performing trace analysis, it pays to limit the number of steps that involve samples handling. Therefore, we are developing methods to streamline the lipid extraction step (see next) and streamlining the separation process by using multidimensional liquid chromatography for the direct introduction of purified lipid bands into tandem MS. A simplistic way of looking at the process is to envision the extraction process isolating a relatively broad range of lipid classes (though no extraction procedure extracts all lipids), which can then be separated into individual classes by a normal phase LC column. Each resulting class can then be separated into individual components by a reversed phase separation. In fact, many single classes are complex enough that even within that single class, a 2DLC separation is required for full separation (e.g. triglycerides and carotenoids). The approach we are currently working with is a liquid-liquid extraction (until the solid phase method is ready) that isolates mostly polar neutral lipids (monoacylglycerols, diacylglycerols, free fatty acids, primary fatty acid amides, and other fatty amides such as N-acyl glycines, N-acyl ethanolamines, and other N-acyl amino compounds like N-acyl taurine. The extract is then separated by normal phase followed by reverse phase. This approach seems to work pretty well. It also allows us to use a semi-automated method to analyze specific, single classes, such as primary fatty acid amides (PFAMs). Watch for the paper coming out on this application. The main instrument is a Dionex Ultimate 3000.

Solid phase lysis and extraction of lipids. As noted above, the initial extraction is an area that is a good candidate for streamlining to minimize sample loss and to enable the handling of small amounts of sample. By “small” we mean 103 cells or less, in contrast to current practices which usually require at least milligrams of tissue or greater than 106 cells. The strategy is to quickly lyse cells up against a stationary phase that is packed into either a capillary column or (currently) a microfluidic chip. The lipids adsorb onto the stationary phase and can be semi-selectively eluted. We expect to be able to integrate this step with the multidimensional LC step for fully automated lipidomics.

Lipidomics. The major instrumentation for this project is an Agilent QTOF with the ChipCube interface and an Agilent QqQ with UHPLC. The strategy is to incorporate the multidimensional LC and solid phase lysis and extraction to provide fully automated lipidomics with the ability to “dial in” the target range of compounds. The major applications are to profile the intracellular lipidome of (1) yeast cells and (2) N18TG2 cells. The former project (see below also) involves following the metabolism of phosphatidylinositols in yeast. Tandem MS with high mass accuracy is a critical tool for following isotopically lableled glycerophosphoinositol (GroPIns) and its phophate derivatives as it is transported into the cell and metabolized. The latter project involves establishing the metabolism of PFAMs by mapping the lipidome in cells that are known to produce PFAMs. When the cells are grown in labeled acids, such as oleic acid, the label gets incorporated into some of the PFAMs, but beyond that, little else is known about the fate of the label or how the label is incorporated into the PFAMs. This is a critical experiment, or question, regarding the metabolism of PFAMs and the effects of PFAM metabolism on affective disorders.

Metabolomics. With Dr. Jana Patton-Vogt of the Department of Biological Sciences at Duquesne. This could be called “Exometabolomics”. The goal in this project is to measure the exometabolome of yeast-the compounds excreted by living yeast cells. The idea is that the measurement of the exometabolome reflects the intracellular metabolic state as a kind of snapshot. Measuring it this way has a number of advantages compared to lysing the cell and measuring the intracellular metabolome (lipidomics project, above), mostly having to do with freezing the state of metabolic flux in the intracellular metabolome and in the number of compounds that need to be measured (the exometabolome is roughly 10% of the entire metabolome). The application, as above, is to understanding the deacylation pathways of phosphatidyl lipids. We are additionally fortunate that much of this metabolism seems to be carried out on or around the cell membrane, making EXOmetabolomics a good choice in any case. The primary tools are molecular biology methods that allow the Patton-Vogt group to manipulate the levels of the relevant deacylation enzymes (phospholipases) and LC/MS/MS for measuring the exometabolome. A secondary major project is the establishment of reliable databases for compounds in the exometabolome that are not currently in any of the public databases (see the Weblinks page).