Nature Methods
2, 14 - 15 (2005)
doi:10.1038/nmeth0105-14
Brilliant lipidsGerrit van Meer
& Rob M J LiskampGerrit van Meer is in the Department of Membrane Enzymology, Institute of Biomembranes, and Rob M. J. Liskamp is in the Department of Medicinal Chemistry, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, The Netherlands. g.vanmeer@chem.uu.nl A remarkably suitable fluorescent tag incorporated into lipids through some clever chemistry produces fluorescent lipids that are excellent mimics of their native counterparts.Wouldn't it be nice if the organization of cellular lipids in, for example, plasma membrane lipid rafts could be directly visualized by fluorescence? Lipids with a fluorescent reporter group have provided a wealth of information on the physical properties of membranes. Unfortunately, extrapolation to the cellular organization, fate and function of natural lipids has been limited by the effects of the bulky fluorophore. Kuerschner and colleagues1 have now devised a method of labeling lipids for fluorescence microscopy while closely mimicking their natural conformation. The striking similarity of the labeled lipids to their natural counterparts is illustrated in their report in this issue.
Ideally, a molecule containing a reporter group to be used in biological studies should be indistinguishable from the parent ligand. That this situation is never achieved is apparent from the large variety of fluorescent groups available for incorporation. Especially when fluorescent labels are incorporated into relatively small molecules like membrane lipids, the labels' effect on structure and behavior can be dominant and therefore mask, or even surpass, the contribution of the parent compound. Clever scientists realized this early and identified a naturally occurring fluorescent fatty acid with a conjugated system of four double bonds, parinaric acid2, but its emission wavelength of 416 nm was inconvenient for microscopy. Remarkably, only now, 30 years later, have Kuerschner et al.1 made use of this concept to develop a general tag to label lipids. The new tag was inspired by parinaric acid but contains five instead of four conjugated double bonds in the long chain of the fatty acid. This gives it excitation and emission wavelengths of 350 nm and 470 nm, respectively. With the advent of the two-photon excitation microscope used by Kuerschner et al., it is much easier to image compounds with excitation optima in the UV range of the spectrum without UV−mediated damage to the cells.
Why did it take 30 years? After the use of lipophilic probes in initial studies on membrane physics, the interest in fluorescent analogs of natural lipids was boosted by a series of elegant studies by Dick Pagano and colleagues with living cells in the 1980s (ref. 3). They used variants of natural lipids in which one fatty acid had been replaced by a short chain of six carbons (C6) carrying a fluorescent NBD moiety (C6-NBD). These were more water-soluble than the natural lipids and could be easily inserted into the plasma membrane, which allowed a detailed characterization of their intracellular transport and metabolism using a combination of biochemistry and fluorescence microscopy. The metabolic precursor C6-NBD−ceramide was instrumental in assigning sphingomyelin synthesis to the Golgi apparatus3 and revealed sorting between glucosylceramide and sphingomyelin in epithelial cells. Thought to reflect the occurrence of glycosphingolipid domains in the Golgi membrane, this observation formed the basis for the lipid 'raft' hypothesis4 but later led to the identification of the multidrug transporter, MDR1, as an outward lipid translocase at the plasma membrane5. C6-NBD−lipids were also crucial for identifying the inward lipid translocases5 and for delineating the lipid endocytotic pathways6. A new dimension was added in the 1990s when Pagano and colleagues used the density-dependent emission characteristics of the similar C5-BODIPY−lipids to visualize glycosphingolipid sorting by domain formation in the endosomes7. However, by then it had been determined that neither C6-NBD− nor C5-BODIPY−lipids accurately reflect the preference of their natural counterparts for liquid-ordered (raft) and liquid-disordered (non-raft) lipid phases8. Consequently, the need arose for more natural, and especially less polar, fluorescent groups such as the hydrocarbon pyrene9. However, the four-ring pyrene also strongly disfigures the slender fatty acid (Fig. 1).
 | | |
This is when the pentaene lipids came onto the scene. Ten years ago, Amat-Guerri et al.10 synthesized conjugated pentaenoic acids by a Wittig olefination reaction, which has proven to be very versatile for the synthesis of all kinds of important compounds containing many conjugated double bonds, including vitamin A. The aim was to obtain polyunsaturated molecules with absorption at a longer wavelength for use in fluorescence studies of lipid behavior in model membrane systems. Only very recently did Kuerschner et al. realize the potential of the pentaene−fatty acids for studies of living cells, and they have beautifully demonstrated in their present paper1 how biologists and chemists can team up to develop new tools for biological research. By using pentaene−fatty acids as fluorescent mimics of regular fatty acids, one very closely approaches the ideal structural resemblance. This is even more impressive after incorporation of the pentaene−fatty acid into the structure of membrane lipids such as sphingomyelin (Fig. 1). Pentaene-sphingomyelin nicely mimicked the preference of natural sphingomyelin for the liquid-ordered phase. The versatility of the probes was illustrated by allowing cells to incorporate them into membrane phospholipids with lipid backbones as different as ceramide, diacylglycerol and the ether lipid alkylacylglycerol. Unexpectedly, ceramide synthase, which normally utilizes saturated fatty acids, strongly favored pentaene−fatty acids, whereas these were incorporated like cis-unsaturated fatty acids in the SN2 position of the glycerol in glycerophospholipids. Thus, cells still recognize pentaene−fatty acids as special, which illustrates the exquisite selectivity with which enzymes perceive lipids.
The most popular lipid fluorophores, NBD and BODIPY, are readily visualized by fluorescence microscopy. However, they dramatically change the physical properties of the fatty acid to which they are attached, which complicates extrapolation of their behavior to that of natural lipids. Pyrene is a better mimic11, but pyrene fluorescence microscopy is complex because the monomer emits at 380 nm, and only high concentrations can be visualized through the formation of excimers that emit at 470 nm. The pentaene−fatty acids are the best mimic. They resemble fatty acids in shape and metabolism and do not have dramatic effects on the phase behavior of at least the sphingolipids. The images in the Kuerschner et al. paper show that pentaene-labeled lipids can be nicely visualized. Pentaene-phosphatidylcholine stained all membranes, especially the endoplasmic reticulum (ER). Newly assembled ether lipids, never before visualized, showed a more punctate distribution (although they were not found in peroxisomes, where they are normally synthesized), and sphingomyelin highlighted the plasma membrane with a far lower labeling of Golgi than was to be expected from NBD- and BODIPY-ceramide experiments. These discrete distributions of the various types of pentaene-labeled lipids hold great promise for their application in studies of lipid organization, traffic and metabolism, and this work shows how by clever design one can obtain excellent mimics of biologically meaningful compounds. But a critical attitude remains essential. In the end, only the natural lipids themselves know how to behave properly.
REFERENCES
-
Kuerschner et al. Nat. Methods 1, 39−45 (2004).
-
Sklar, L.A., Hudson, B.S. & Simoni, R.D. Proc. Natl. Acad. Sci. USA 72, 1649−1653 (1975). | PubMed | ChemPort |
-
Pagano, R.E. & Sleight, R.G. Science 229, 1051−1057 (1985).. | PubMed | ChemPort |
-
Simons, K. & van Meer, G. Biochemistry 27, 6197−6202 (1988). | PubMed | ISI | ChemPort |
-
Pomorski, T., Holthuis, J.C., Herrmann, A. & Van Meer, G. J. Cell Sci. 117, 805−813. (2004). | Article | PubMed | ChemPort |
-
Maier, O., Oberle, V. & Hoekstra, D. Chem. Phys. Lipids 116, 3−18 (2002). | PubMed | ChemPort |
-
Sharma, D.K. et al. J. Biol. Chem. 278, 7564−7572. (2003). | Article | PubMed | ISI | ChemPort |
-
Wang, T.Y. & Silvius, J.R. Biophys. J. 79, 1478−1489 (2000). | PubMed | ISI | ChemPort |
-
Koivusalo, M., Alvesalo, J., Virtanen, J.A. & Somerharju, P. Biophys. J. 86, 923−935 (2004). | PubMed | ChemPort |
-
Souto, A.A., Acuna, A.U. & Amat-Guerri, F. Tetrahedron Lett. 35, 5907−5910 (1994). | Article | ChemPort |
-
Kasurinen, J. & Somerharju, P. J. Biol. Chem. 267, 6563−6569 (1992). | PubMed | ChemPort |
-
Chattopadhyay, A. & London, E. Biochemistry 26, 39−45 (1987). | PubMed | ISI | ChemPort |
|
