Organellar lipidomics

Abstract

A wealth of information related to lipid metabolism and signaling has been revealed in recent years using mass spectrometric-based lipidomics methods. Although quantitatively sensitive, these compositional profiling methods rely on conventional tissue extractions of total lipids which results in a loss of original cellular context of lipid metabolites. We described the development of direct organelle mass spectrometry (DOMS), a high resolution MS profiling method providing the capability to directly visualize, extract and analyze the lipid compositions in single, individual lipid droplets (LDs) from plant tissues. DOMS of lipid droplets isolated from mature cotton embryos demonstrated a surprising lipid droplet-to-droplet variability in triacylglycerol (TAG) composition that would have been concealed through conventional profiling methods and might be important for the understanding of LD biogenesis in oilseeds. Additional applications directed toward the identification of lipid and protein compositions of other organelles could have a significant impact on our general understanding of metabolism and suggest new ideas about how cells coordinate the functions of their subcellular compartments.

Mass spectrometric-based lipidomics methods in recent years have revealed considerable complexity of lipid composition within organisms and tissues providing a wealth of information related to lipid metabolism and signaling.^1,2 Although quite sensitive, these profiling methods rely on conventional tissue extractions of total lipids which results in a loss of original cellular context of lipid metabolites. The spatial distribution of lipids in cells is certain to be as important to their functions as their changes in concentrations, and we have focused on developing approaches for high resolution sampling of lipids in cellular and subcellular compartments.

Lipid droplets (LDs) are dynamic organelles composed of a neutral lipid core surrounded by a phospholipid monolayer with integrated and associated proteins.^3,4 LDs are present in essentially all organisms and are involved in storage,^5,6 signaling^7,8 and cellular trafficking.⁹ In the manuscript by Horn et al. we describe the development of direct organelle mass spectrometry (DOMS), enabling the analysis of lipid compositions in single, individual LDs from plant tissues.¹⁰ In this approach, LDs are visualized with selective fluorescent dyes directly on an epifluorescence microscope stage. Individual LDs are selected and drawn into a nanospray capillary using a custom, user-controlled nanomanipulation apparatus.^11,12 Lipids are solvent-microextracted in the tip and directly applied into a nanospray ionization source for MS analysis. DOMS analysis provides the capability to directly visualize, extract and analyze the lipid content of individual LDs. The identification of lipid compositions at the organelle level could have a significant impact on our general understanding of cellular lipid metabolism.¹⁰

DOMS of lipid droplets isolated from mature cotton embryos demonstrated a surprising droplet-to-droplet variability in triacylglycerol (TAG) composition (e.g., over seven individual LDs sampled, TAG species containing 16:0/16:0/18:2 and 16:0/18:2/18:2 varied from ∼7–17% and ∼16–28% of total TAG in each droplet, respectively).¹⁰ If all the lipids were packaged at the same rate based on their relative abundance one would expect to see less variability among these LDs. This individual heterogeneity initially suggests that there may be substantial variation in the intracellular packaging process for neutral lipids in plant tissues. Indeed when individual LD compositions are averaged, the overall TAG composition resembles that of standard cottonseed oil. It is unclear if the differences are a result of tissue or cell-type specific differences in TAG machinery of embryos or if they arise through differential expression of TAG metabolic enzymes during embryo development. Perhaps DOMS applied to embryo sections at different stages of development can help resolve these questions. Nonetheless, this organelle-to-organelle variability would have been otherwise concealed with traditional lipid analysis that reports the average composition among the population of LD samples, and this new information will be important when constructing models of LD biogenesis in oilseeds.

The widely accepted model for the formation of LDs is through budding off from the endoplasmic reticulum (ER) membrane^13,14 although there are a few alternative model variations that still need to be experimentally tested.^15,16 Identifying the lipid and protein compositional heterogeneity of LDs within tissues and single cells will permit a more thorough understanding of how these LDs are formed from a biophysical and cellular perspective. Based on the size and morphology variability of LDs within certain tissues, cells and organisms^17,18 there is considerable flexibility in the ability of cells to package neutral lipids (and proteins) into a single LD. Recent biophysical studies have suggested that LDs are formed as single, small particles from the ER and coalesce in the cytosol to form the LD of final size in situ.¹⁴ From a statistical standpoint if there were small compositional differences in initial LD particles, it would be reasonable to assume that a majority of LDs formed at maturity within cells will represent the average LD lipid composition¹⁹ and that there could be individual LDs present that show significant compositional differences. But it remains to be examined whether LD heterogeneity arises within cells, or is determined at the tissue or developmental levels. This concept of lipid packaging and heterogeneity may be important to consider in the genetic modification of lipids in transgenic oilseeds²⁰ if the process of LD formation is influenced by composition or vice versa.

DOMS may be useful in helping to infer additional functions for LDs in different plant tissues. LDs in plants are mostly considered to be a storage repository for energy reserves that are mobilized during seed germination.²¹ But they also serve as the building blocks for recycling of biological membranes,⁶ and temporary reserve during metabolic fluxes of lipids.^6,22 However, LDs are present in essentially all plant cells and there may be additional functions for LDs independent from temporary lipid storage. In other organisms, LDs have been suggested to play a role in signaling and cellular trafficking^7–9 and even in the intracellular propagation and persistence of viruses.²³ Moreover, there are several lines of evidence in different organisms that LDs directly interact or are closely associated with the ER,²⁴ endosomes,²⁵ peroxisomes,²⁶ mitochondria²⁷ and chloroplasts.²⁸ Elucidating the lipid species located in LDs derived from various plant sources may help suggest additional functions for these dynamic organelles.

The potential capability to analyze integrated proteins within a single LD or even a small group of LDs by DOMS through a trypsin-mediated microphase extraction could provide additional information about LD ontogeny and functions. Several proteins that have been shown to localize to LDs in different tissues and some of these could potentially have a role in plant signaling^6,29,30 especially in nonseed tissues that are not encompassed by oleosin proteins. There is very little information about surface proteins in plant LDs beyond the well characterized oleosins, and the proteins that interact with the LD surface are important to overall function. For example, in mammalian (HeLa) cell cultures fatty acid amide hydrolase (FAAH-2) was found was to be localized to LDs were it inactivates known (plant) signaling lipids, namely the N-acylethanolamines.⁷ The potential for expanding DOMS to protein identification in individual LDs could markedly influence insights into the functions of plant LDs beyond that of storage depots.

The use of DOMS for the analysis of lipids other than those found in LDs could be profound. For example, it is likely that signal transduction mechanisms involving lipids relies on highly localized changes in specific lipid species, and these may be difficult to detect against the large backdrop of lipids in tissue extracts. If DOMS could be applied to regions of cell membranes or even to isolated membrane fractions, previously undetected changes in signaling lipids might be possible to detect. Certainly major considerations of tissue preparation and experimental design would be important in such studies, but the potential to uncover localized changes of signaling metabolites offered by DOMS may be worth exploring.

Still other applications for DOMS include examining organelles other that LDs. Questions about organelle-toorganelle heterogeneity in composition and function have not really been possible to address until now. Such information applied to plastids, mitochondria, nuclei, vacuoles, or other subcellular compartments could suggest new ideas about how cells coordinate the functions of their subcellular constituents for maximum overall efficiency. For example, do all chloroplasts in a cell have the same capacity for photosynthesis, or produce the same amounts of requisite metabolites? Or do all mitochondria contain the same complement of respiratory complexes, alternate or otherwise, or might these be partitioned in different organelles? The many questions of organelle heterogeneity in function go hand-in-hand with analysis of composition, and the high-resolution capabilities of DOMS provide opportunities to address these questions at a new level.

Addendum to: Horn PJ, Ledbetter NR, James CN, Hoffman WD, Case CR, Verbeck GF, et al. Visualization of lipid droplet composition by direct organelle mass spectrometry. J Biol Chem. 2011;286:3298–3306. doi: 10.1074/jbc.M110.186353.