[page 6↓]

1  Introduction

All cells are surrounded by a plasma membrane consisting of two layers (leaflets) of amphipathic lipid molecules. This so-called lipid bilayer comprises a hydrophobic inner region formed by the hydrophobic tails of the lipid molecules and a polar outer region composed of the head groups of lipids. This lipid bilayer forms the physical barrier between the aqueous cytoplasm and the surrounding. Within the lipid bilayer proteins are embedded (“fluid-mosaic” model by Singer and Nicolson (Singer and Nicolson, 1972)). The proteins traverse the two leaflets (integral or intrinsic proteins) or are attached to membrane (peripheral proteins). In addition to the ubiquitous plasma membrane, eukaryotic cells contain subcellular membranes creating different intracellular compartments in which highly specific biochemical processes can be maintained and regulated. For the specific function of each compartment, distinct sets of lipids and proteins are essential. Moreover, lipids have to adopt the correct distribution over the two membrane leaflets. For example, in the plasma membrane of bacteria phospholipids are synthesized on the cytoplasmic leaflet of the plasma membrane. To ensure balanced growth and thus, stability of the biogenic membrane, half of the newly synthesized lipids must move to the opposing leaflet. A similar process must occur in the endoplasmic reticulum (ER) of eukaryotic cells. Newly synthesized lipids are initially located in the cytoplasmic leaflet of the ER but must flip across the ER to populate the exoplasmic leaflet to allow balanced membrane growth.

Furthermore, the plasma membrane of eukaryotic cells displays an asymmetric lipid distribution with the majority of aminophospholipids in the cytoplasmic leaflet and choline-containing phospholipids in the exoplasmic leaflet. Because this lipid asymmetry does not correspond to the asymmetry of lipid synthesis or hydrolysis, it must be formed and maintained by specific mechanisms that control lipid movement across the bilayer.

In protein-free model membranes, movement of most phospholipids from one leaflet to the other - the so-called flip-flop - is very slow, with half-times in the order of days (Eastman, et al., 1991;Kornberg and McConnell, 1971). The reason for the very slow flip-flop is the thermodynamically unfavorable transfer [page 7↓]of the hydrophilic head-group of a lipid molecule through the hydrophobic core of the lipid bilayer. Nevertheless, phospholipid transbilayer movement must occur at a considerable faster rate in membranes of living cells. This has led to the idea that lipid flip-flop is protein-mediated. The identification and characterization of the protein machinery involved in lipid flip-flop is a major challenge in current biology.

In the first chapter, an overview about the phospholipid flip-flop in eukaryotic cells is presented. Subsequently, the consequences of transport for function and structure of the originating and target membranes are discussed. Since this thesis focuses on the mechanisms of phospholipid flip-flop across the inner membrane in Escherichia coli (E.coli ), the present knowledge about the composition and functions of phospholipids in the E.coli envelope are summarized and the known phospholipid transport processes in bacteria will be discussed. In the last paragraph of this chapter, a number of techniques and methods used for investigations in transmembrane distribution and movement of (phospho)lipids will be described. Finally, the aims of the studies presented in this thesis are summarized.

[page 8↓]

1.1  Phospholipid transmembrane movement in eukaryotic cells

In eukaryotic cells, the phospholipid transbilayer distribution is specific for various subcellular membranes and seems to be regulated by specific membrane proteins ( for a review see e.g. (Pomorski, et al., 2001)).

The plasma membranes of eukaryotic cells have a clearly defined transbilayer phospholipid distribution. The aminophospholipids phosphatidylethanolamine (PE) and phosphatidylserine (PS) are highly enriched in the cytoplasmic leaflet compared to the exoplasmic leaflet. This asymmetry is generated by a not yet identified aminophospholipid translocase (APLT) that uses hydrolyses of ATP to translocate aminophospholipids from the exoplasmic to the cytoplasmic leaflet (Seigneuret and Devaux, 1984;Tilley, et al., 1986). The APLT activity is diminished after modification of plasma membrane proteins with sulfhydryl reagents like NEM, in the presence of vanadate and after elevation of the intracellular calcium concentration, which leads to a randomization of phospholipids across the plasma membrane leaflets in each case (Bitbol, et al., 1987;Herrmann, et al., 1989;Williamson, et al., 1992). Furthermore, in human erythrocyte ghosts, it has been shown that PS translocase activity and a Mg2 +-ATPase activity are properties of the same protein (Beleznay, et al., 1997). In a variety of subcellular membranes like chromaffin granules, clathrin-coated vesicles and cholinergic vesicles from Torpedo electric organ Mg2+-ATPases of unknown function have been discovered. They share several properties with APLT, e.g. stimulation by PS and inhibition by vanadate and NEM. (Xie, et al., 1989;Yamagata and Parsons, 1989;Zachowski, et al., 1989). Although many attempts have been made to identify the APLT (Auland, et al., 1994;Morrot, et al., 1990;Zimmerman and Daleke, 1993), its molecular identity remains unclear. The asymmetric distribution of PS and PE across the plasma membrane is dissipated by the action of a putative, Ca2+-activated scramblase (Bevers, et al., 1999;Comfurius, et al., 1990). The resulting exposure of PS and PE in the outer leaflet leads to various cellular responses, e.g. recognition by macrophages (Chang, et al., 2000;Fadok, et al., 1992;Verhoven, et al., 1995) or blood coagulation (Solum, 1999).

[page 9↓]

A second member of ATP-dependent lipid transporter, the ATP-binding cassette (ABC) transporter family, was identified in studies, originally related to multidrug resistance (MDR) in cancer cells. ABC proteins transport a broad spectrum of structural unrelated substrates (Bosch and Croop, 1996). Studies on MDR1 P-glycoprotein (MDR1-Pgp) showed that MDR1-Pgp affects the transverse distribution of endogenous PS and PE (Bosch, et al., 1997;Pohl, et al., 2002). Unlike MDR1-Pgp, MDR3-Pgp is thought to regulate the secretion of PC into the bile (Elferink, et al., 1997;Smit, et al., 1993).

The ER membrane, in contrast to the plasma membrane, as a biogenic membrane, is assumed to have a symmetric lipid distribution. This transmembrane distribution and movement is thought to be mediated by a non-specific protein. It has been proposed that as a result of the activity of this protein, all phospholipids abundant in the microsomal membrane are continuously randomized and therefore, are equally distributed across the bilayer (Herrmann, et al., 1990;Williamson and Schlegel, 1994). It has been shown that phospholipids rapidly equilibrate over rat liver microsomal membranes by facilitated diffusion (Bishop and Bell, 1985;Buton, et al., 1996;Herrmann, et al., 1990). The measured translocation rates depended on the type of phospholipid analogue, temperature and differ slightly between methods used to assay this process (reviewed in (Menon, 1995)). Buton et al. reported a characteristic translocation half-time of fluorescent short-chain phospholipid analogues (see chapter 1.3.2) across the microsomal membrane of ~25 s (Buton, et al., 1996). This rapid movement was determined to be bi-directional, partially protease- and NEM-sensitive (Buton, et al., 1996). Buton and co-workers used a combined BSA extraction and filtration assay, which was limited in time resolution to properly monitor the very fast phospholipid flip-flop (Buton, et al., 1996). In a more recent study, Herrmann and co-workers improved the measurements of ultra-fast translocation processes (Marx, et al., 2000). They significantly enhanced the time resolution due to a combination of the classical BSA back-exchange assay with the stopped-flow method, which was intensively used and further optimized in this thesis. Marx et al. found that fluorescent and spin labeled, short-chain phospholipid analogues rapidly redistributed between the leaflets of microsomes with half-times of [page 10↓]62-148 s and 8-16 s, respectively (Marx, et al., 2000). Collectively, these studies revealed that the fast transbilayer movement in the ER is a protein-mediated, bi-directional process without phospholipid specificity. However, whether specific proteins are required for this phospholipid flip-flop remains to be elucidated. Recent studies discussed the possibility that transmembrane stretches of membrane proteins facilitate the phospholipid translocation (Kol, et al., 2001;Kol, et al., 2003;Kol, et al., 2003). In an attempt to identify the proteins involved in the rapid transmembrane movement of phospholipids across the ER, Menon and colleagues reconstituted detergent extracts and fractions of microsomal membranes separated by glycerol gradient centrifugation into proteoliposomes (Menon, et al., 2000). This approach yielded a chromatographic fraction of enhanced transport activity, which sedimented at 3.8 S in the glycerol gradient. However, they were not able to isolate a specific protein that was responsible for the translocation.

[page 11↓]

1.2  The architecture and dynamic of the bacterial envelope

1.2.1 The membrane organization of E.coli

With regard to the cell membrane structure, bacteria can be divided in two classes: Gram-positive and Gram-negative bacteria. Gram-positive bacteria like Bacilli contain only an inner membrane and a peptidoglycan layer. Gram-negative bacteria such as E.coli have an inner and outer membrane, with an aqueous compartment between the two membranes - the periplasm. This compartment harbors the peptidoglycan layer.

Figure 1: Schematic representation of the E.coli envelope adapted from Raetz and Withfield with minor modifications (Raetz and Whitfield, 2002). Abbreviations: LPS, lipopolysaccharide; MDO, membrane-derived oligosaccharides; Kdo, 3-deoxy-d-manno-oct-2-ulosonic acid; PPEtn, phosphoethanolamine

The outer membrane forms a semipermeable border of the E.coli cell to its environment (Figure 1). It consists of lipopolysaccharide (LPS), glycerophospholipids and proteins. The outer membrane organization is highly asymmetric. The inner leaflet of the outer membrane is exclusively populated by phospholipid molecules (predominantly PE), while the outer leaflet exclusively contains LPS (Osborn, et al., 1972). LPS is a complex phospholipid with a non-repeating “core” oligosaccharide and a distal polysaccharide. The hydrophobic [page 12↓]anchor, lipid A or endotoxin, is a glucosamine dimer with six acyl chains attached. The peripheral polysaccharide chain contains distinct types of sugars. E.coli mutants lacking several sugar residues in the inner core of LPS (deep rough mutants, (Nikaido and Vaara, 1985)) are more sensitive for penetration of hydrophobic macromolecules (Nikaido and Vaara, 1985). Furthermore, deep rough mutants contain additional phospholipids in the outer leaflet, which is thought to contribute to enhanced permeability. LPS, in particular lipid A, plays an important role in activation of innate immune responses e.g. by. macrophages (Raetz and Whitfield, 2002).

The most abundant outer membrane proteins are integral pore forming proteins with a β-barrel structure, so-called porins (Nikaido and Vaara, 1985). These proteins allow non-specific passage of small hydrophilic molecules with a molecular mass of up to 600 Da (Decad and Nikaido, 1976). Several of the outer membrane β-barrel proteins transport water-soluble molecules like sugars and nucleotides in a more regulated manner. The outer membrane also contains a variety of peripheral and integral non-porin proteins with different functions e.g. receptors for vitamins (Nikaido and Vaara, 1985) and proteins involved in membrane and cell shape stability, like protein A (OmpA) (Pautsch and Schulz, 2000). The outer membrane also contains proteins with enzymatic activities, like OmpT and OMPLA (Dekker, et al., 1999;Luirink, et al., 1986;Pugsley and Schwartz, 1984;Stathopoulos, 1998;Vandeputte-Rutten, et al., 2001).

The periplasm (Figure 1) harbors the murein sacculus (peptidoglycan layer), which is a polymer of muropeptides (Weidel and Pelzer, 1964). The muropeptide polymer protects the cells from lysis in a hypotonic environment and is an important factor in maintaining the cell shape (Rogers, et al., 1980). The peptidoglycan layer is bound to the outer membrane by the major outer membrane protein (LPP) (Braun and Rehn, 1969;Braun and Wolff, 1970) via peptide bonds. OmpA is also involved in attaching the peptidoglycan wall to the outer membrane.

Additionally, to a number of proteins involved in transport, detoxification, metabolic and catabolic processes (Beacham, 1979;Tam and Saier, 1993), the periplasm contains membrane-derived oligosaccharides (MDO). These sugar [page 13↓]chains are substituted with phosphoethanolamine, phosphoglycerol and O-succinyl ester (Kennedy, et al., 1976) and believed to play a role in osmotic regulation.

The inner membrane of E.coli separates the cytoplasm from the extracellular environment. It is the side of many essential processes, like nutrient uptake, oxidative phosphorylation, transport processes and export of toxic substances as well as metabolic products. Most of the proteins in E.coli that are involved in protein and lipid biosynthesis are associated with this membrane. Especially the catalytic active sites of proteins involved in phospholipid synthesis are faced to the cytosolic side of the membrane (Huijbregts, et al., 2000).

1.2.2 Glycerophospholipids in E.coli

The major phospholipids of E.coli are PE and phosphatidylglycerol (PG), which make up ~80% and 15% of the total phospholipid content, respectively. The third major phospholipid is cardiolipin (CL) with an abundance of about 5%.

Phospholipid synthesis starts in the cytosol with the generation of glycerol-3-phosphate (G3P) (Kito and Pizer, 1969). G3P is acylated in the sn-1 and sn-2 position and the resulting phosphatidic acid (PA) is immediately converted to CDP-glyceride (Sparrow and Raetz, 1985), as evident by the finding that PA is present only in trace amounts (~1%) in the plasma membrane of E.coli. CDP-glyceride is the precursor for the synthesis of PE, PG and CL. The first step in the synthesis of PE is the reaction of CDP-glyceride with serine, catalyzed by the phosphatidylserine synthetase (pss). PS is rapidly decarboxylated to PE by the PS decarboxylase. PE, the most abundant phospholipid in E.coli, is a zwitterionic molecule at physiological pH, due to the protonated amino group and the dissociated phosphate group.

The synthesis of the acidic phospholipid PG, the second major phospholipid in E.coli, starts with the formation of phosphatidylglycerolphosphate (PGP). The substrates of this reaction are CDP-glyceride and G3P. PGP is converted to PG by the enzyme PGP phosphatase.

[page 14↓]

CL is formed by the condensation of two molecules of PG (Hirschberg and Kennedy, 1972;Tunaitis and Cronan, 1973) by the enzyme CL synthetase.

1.2.3 Phospholipid movement across bacterial membranes

In Gram-negative bacteria, phospholipids have to move from the site of their synthesis, the cytoplasmic leaflet of the inner membrane, to the periplasmic leaflet. In addition, phospholipids have to be transported to the inner leaflet of the outer membrane. Both processes have been studied in in vivo and in vitro systems (for a recent review see (Huijbregts, et al., 2000)). The movement of phospholipids between the inner and the outer membrane was first shown by Osborn and co-workers (Osborn and Munson, 1974). They demonstrated with pulse labeling studies that PE was synthesized in the inner membrane of Salmonella typhimurium and finally transported to the outer membrane. Moreover, radiolabeled PS, introduced in the outer membrane, was shown to be rapidly transported to the inner membrane, where it becomes accessible to the enzyme PS decarboxylase and transformed to PE within five minutes (Jones and Osborn, 1977;Jones and Osborn, 1977). The resulting PE was transported back to the outer membrane. In the same study, it was shown that the transport of phospholipids in S. typhimurium was head-group independent, since not only the major phospholipids (PE, PG and CL) were transported, but in addition, the not naturally occurring phospholipid PC was also transported to the inner membrane of S. typhimurium.

In Gram-negative bacteria, membrane contact sides between inner and outer membrane, so called Bayer's bridges (Bayer, 1991), have been suggested to mediate intermembrane lipid transport. Additionally, in E.coli a rapid bi-directional transport of phospholipids between the inner and the outer membrane was observed on whole cells (Donohue-Rolfe and Schaechter, 1980;Langley, et al., 1982).

The first investigations on transbilayer movement of phospholipids in bacteria has been carried out on Gram-positive bacteria (Rothman and Kennedy, 1977). Rothman and Kennedy observed that the translocation of newly synthesized PE from the inner to the outer leaflet in Bacilli occurred with a half-[page 15↓]times of 1.5-3 min at 37°C (Rothman and Kennedy, 1977;Rothman and Kennedy, 1977). A more recent study (Hrafnsdottir, et al., 1997) showed that short-chain, fluorescent labeled phospholipid analogues translocated rapidly across the Bacillus megaterium membrane with a half-time of ~30 s at 37°C. This transport was demonstrated to be protease sensitive but not head-group dependent. Transbilayer movement of phospholipids was studied with the use of reconstituted, transport-competent proteoliposomes derived from detergent-solubilized Bacillus subtilis plasma membranes (Hrafnsdottir and Menon, 2000). The resulting proteoliposomes were shown to be capable of transporting a short-chain, water soluble analogue of PC (half-time about one minute). To prove that the short-chain, water soluble PC analogue reflects the behavior of endogenous phospholipids, a more natural long-chain phospholipid dipalmitoyl-PC (DPPC) was reconstituted into proteoliposomes and the extent of hydrolysis by the phospholipase A2 (PLA2) was measured (Hrafnsdottir and Menon, 2000). Indeed, the extent of hydrolysis was shown to be a function of the protein/phospholipid ratio reconstituted into proteoliposomes derived from B. subtilis. This indicates that the short-chain as well as the long-chain phospholipid analogues were transported across the vesicle membrane by proteins.

Pulse labeling studies on separated inner and outer membrane fractions from E.coli demonstrated that newly synthesized PE reached the outer membrane within 2.8 min. The transport of anionic phospholipids had a half-time of less than 30 s (Donohue-Rolfe and Schaechter, 1980;Langley, et al., 1982). Investigations on inverted inner membrane vesicles (IIMV) from E.coli also demonstrated a rapid transbilayer movement of phospholipids across the vesicle membrane (Huijbregts, et al., 1996). Utilizing short-chain fluorescent analogues of phospholipids, Huijbregts et al. showed that exogenously added analogues rapidly flip across the inner membrane of E.coli with a half-time about seven minutes at 37°C. This transport was temperature dependent, bi-directional and not influenced by treatment with sulfhydryl reagents or proteinase K, nor by the presence of ATP or a pH gradient across the membrane of IIMV (Huijbregts, et al., 1996). Huijbregts, et al. also studied transmembrane movement of endogenously synthesized phospholipids across the inner membrane of E.coli (Huijbregts, et al., 1998). Radioactive labeled PE was biosynthetically introduced into IIMV from [page 16↓]PE-deficient E.coli strain AD93 by reconstitution with the enzyme pss and the addition of wild-type lysate, metabolic substrates and [14C]serine. Another approach utilized right-side out vesicles, in which the active site of pss is situated in the lumen of the vesicles. Under these circumstances, the PS conversion took place in the lumen of the vesicles by reconstitution and the appearance of PE on the outer leaflet was measured. Both approaches demonstrated that the redistribution of newly synthesized radiolabeled PE occurred with a half-time of less than one minute. However, these earlier studies did not demonstrate a strong requirement for protein in the translocation process or lacked the time resolution to measure an accurate translocation rate. Furthermore, the molecular basis of the putative protein dependent mechanism of phospholipid flip-flop is still unknown.

Only little is known about the transverse distribution of phospholipids in bacterial membranes. In the plasma membrane of the Gram-positive bacterium Micrococcus luteus the distribution of PG and CL was studied using photoreactive lipid analogues (de Bony, et al., 1989). A slight asymmetric distribution of PG with about 60% of the PG in the outer leaflet was found. CL was equally distributed between the two leaflets. However, in a later report it was suggested, that this distribution strongly depends on cell growth and division (Welby, et al., 1996). When Huijbregts and colleagues investigated the transbilayer distribution of phospholipids in IIMV and right-side out vesicles, they detected an asymmetric transbilayer distribution of radiolabeled, newly synthesized PE in the inner membrane of 35% in the cytoplasmic and 65% in the periplasmic leaflet (Huijbregts, et al., 1998).

[page 17↓]

1.3  Methods for the characterization of transmembrane distribution and movement of phospholipids in biological membranes

A variety of techniques are used to investigate the phospholipid transbilayer distribution and movement in model membranes and biomembranes such as plasma membranes or cellular organelles. In the past, chemical reagents or enzymes for altering endogenous phospholipids or high affinity probes, which react with distinct phospholipid species, were utilized. In the last two decades, lipid analogues, which mimic the physicochemical properties of their endogenous counterparts, evolved to be the most important tools to study the behavior of lipids.

1.3.1 Assays for the determination of transmembrane movement and distribution of endogenous phospholipids

One of the first methods used to determine the transbilayer distribution of phospholipids was the modification of endogenous phospholipids by phospholipases A2 (PLA2) and C (PLC). Endogenous phospholipids located in the outer monolayer of membranes, such as the plasma membrane of erythrocytes (Dolis, et al., 1996;Roelofsen and Zwaal, 1976) or the plasma membrane of prokaryotes (Nanninga, et al., 1973), were treated with PLA2 or PLC. Subsequently, the products were analyzed by e.g. chromatographic techniques. This invasive method has a number of shortcomings. Phospholipase treatment can lead to the release of lyso lipids and free fatty acids from the membrane (Nanninga, et al., 1973), and therefore, may induce transmembrane movement of phospholipids and their derivatives or perturbations of the membrane organization. Furthermore, these assays suffer from the limited time resolution.

Other approaches are based on modifications of the head group or fatty acid of lipids by chemicals (reviewed in (Op den Kamp, 1979)), like trinitrobenzene sulfonic acid (TNBS), isothionyl acetimidate, fluorescamine or anthracene (Welby, et al., 1996). TNBS reacts specifically with PE and does not permeate the membrane. It has been widely used for investigations on the transverse distribution of this phospholipid in both prokaryotic (Rothman and Kennedy, 1977) and eukaryotic cells (Bonsall and Hunt, 1971;Cerbon and Calderon, 1991;Fontaine, et al., 1980;Marinetti and Love, 1976;Musters, et al., 1993;Sandra and Cai, 1991). Another chemical reagent, which has been utilized to determine the distribution and transmembrane movement of PG across vesicle membranes is the α-diol group oxidizing periodate (de Bony, et al., 1989;Hope, et al., 1989;Huijbregts, et al., 1997). But permeation of periodate through distinct membranes leads to inaccuracies in the quantification of the PG distribution across the membrane (Huijbregts, et al., 1997).

Non modifying techniques were also applied, such as the annexin V approach. Annexin V is a protein, which is able to non-covalently bind to various phospholipids in a calcium dependent manner but with a clear preference for PS (Swairjo and Seaton, 1994). Annexin V does not penetrate the membrane. However, annexin V is only appropriate for the determination of the exposure or presence of phospholipids rather than for the assessment of transmembrane movement and distribution of phospholipids.

Lipid transfer proteins, which can transfer lipids between two membranes (Wirtz, 1991) have also been used to study the distribution of phospholipids in membranes (van Meer, 1989;Wirtz, 1991).

1.3.2  Assays for the determination of transmembrane movement and distribution of phospholipid analogues

Substantial progress in the study of transmembrane distribution and transbilayer movement of phospholipids across membranes has been made with the introduction of phospholipid analogues (Kornberg and McConnell, 1971;McIntyre and Sleight, 1991). Fluorescent and spin labeled analogues have been extensively used in many studies. One of the naturally occurring long-chain fatty acids (usually in the sn-2 position) of the phospholipid is substituted by an acyl chain (commonly 4-12 carbon atoms long) carrying a fluorescent (e.g. N-(4-nitrobenzo-2-oxa-1,3-diazole) (NBD)) or paramagnetic (e.g. 4-doxyl) reporter group. This replacement results in a slight change of the physicochemical properties of the analogue. The shorter fatty acid in the sn-2 position makes the analogues slightly more water-soluble compared to their endogenous counterparts. This is advantageous for rapid incorporation of these analogues into

[page 19↓]

Figure 2: Structures of fluorescent short-chain phospholipid analogues used in the course of this thesis.

(bio)membranes, since the short-chain analogues spontaneously incorporate into membranes. Their transmembrane distribution and movement can be monitored either by fluorescence spectroscopy, fluorescence microscopy for fluorescent labeled or by EPR spectroscopy for spin labeled analogues. Furthermore, the reporter groups can be chemically modified either by conversion of spin labeled analogues into diamagnetic species using ascorbate (Kornberg and McConnell, 1971) or by turning fluorescent analogues into non-fluorescent derivatives using dithionite (Huijbregts, et al., 1996;McIntyre and Sleight, 1991;Pomorski, et al., 1994).

[page 20↓]

Short-chain (up to six carbon atoms with the fluorophore attached) phospholipid analogues (Figure 2) are extractable by e.g. BSA (Haest, et al., 1981). That makes the so-called BSA back-exchange assay a powerful tool in the investigation of the transbilayer movement and distribution of phospholipids. The time resolution of this assay is higher than of assays that utilize enzymes for the determination of the transmembrane distribution of endogenous phospholipids. However, for some membrane systems it has been shown that BSA is not capable of quantitatively extracting short-chain phospholipid analogues. In such cases, time-consuming centrifugation steps are required (Pomorski, et al., 1996).

A general concern in these studies is to what extent these analogues mimic the behavior of endogenous phospholipids in membranes. Since their sn-2 acyl are shorter than the respective acyl chains of the native counterparts, and therefore, possess some aqueous solubility, analogues are capable of spontaneous monomeric exchange between (intracellular) membranes of nucleated cells (Bai and Pagano, 1997;Kean, et al., 1993;Martin and Pagano, 1987;Nichols and Pagano, 1982). Consequently, as far as nucleated cells are concerned, the transmembrane equilibrium distribution of analogues does not provide a quantitative measurement of the actual steady-state distribution of the corresponding endogenous phospholipids. However, it has been shown that the transmembrane equilibrium distribution of analogues qualitatively reflects the distribution of endogenous phospholipids (Bratton, et al., 1997;Verhoven, et al., 1995).

Another concern is the presence of a reporter group that might influence the kinetics and the extent of transmembrane movement of the analogue. Tilley et al.obtained kinetic data for phospholipid translocation of spin-labeled analogues very similar to data obtained with radioactive, long-chain phospholipids (Tilley, et al., 1986). As the doxyl group is a comparatively small reporter group, the behavior is likely to resemble endogenous phospholipids with respect to the transmembrane movement.

The NBD moiety, a fluorescent reporter group, is much bulkier and more polar than the doxyl group. It was found that this group diminishes the affinity of PE analogues to the APLT (Colleau, et al., 1991). Additionally, Chattopadhyay and London, who used fluorescence quenching by spin-labeled phospholipids, [page 21↓]concluded that presumably the polarity of the NBD group results in “loop back” of the reporter group to the membrane surface (Chattopadhyay and London, 1987). Indeed, as shown by dithionite fluorescence assay, long-chain analogues react more favorably with dithionite, indicating a better accessibility of the probe by dithionite present in the aqueous phase (Huster, et al., 2001;Huster, et al., 2003). It cannot be excluded that the bending of the sn-2 acyl chain might exert an influence on kinetic analysis.

However, short-chain, NBD-labeled analogues (Figure 2) have proven to be faithful analogues of their endogenous counterparts in a variety of membrane systems (Bratton, et al., 1997;Kean, et al., 1997;Marx, et al., 2000;Seigneuret and Devaux, 1984;Verhoven, et al., 1995). In spite of the concerns mentioned above, short-chain analogues are good tools that have proven to be useful for the identification of flippase proteins.

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 3.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated: