以下為提醒 CELLULAR & MOLECULAR BIOLOGY LETTERS Volume 10, (2005) pp 101 – 121 http://www.cmbl.org.pl Received 6 July 2004 Accepted 2 December 2004 *Corresponding author, e-mail: gummadi@iitm.ac.in Abbreviations used: PL – phospholipid; PC – phosphatidylcholine; PE – phosphatidylethanolamine; PS – phosphatidylserine; ER – endoplasmic reticulum; APT – aminophospholipid translocase; GPI – glycosylphosphatidylinositol; diC 4 PC – dibutyrylphosphatidylcholine; BSA – bovine serum albumin; NEM – N-ethylmaleimide; BMV – Bacillus megaterium vesicles; IIMV – inverted inner membrane vesicles; TNBS – trinitrobenzene sulfonic acid; PLA 2 – phospholipase A 2 ; DPPC – dipalmitoylphosphatidylcholine; DEPC – diethylpyrocarbonate; SL – spin labeled; PI – phosphatidylinositol; PG – phosphatidylglycerol. THE MYSTERY OF PHOSPHOLIPID FLIP-FLOP IN BIOGENIC MEMBRANES SATHYANARAYANA N. GUMMADI* and KRISHNA S. KUMAR Department of Biotechnology, Indian Institute of Technology-Madras, Chennai 600 036, India Abstract: Phospholipid flip-flop is required for bilayer assembly and the maintenance of biogenic (self-synthesizing) membranes such as the eukaryotic endoplasmic reticulum and the bacterial cytoplasmic membrane. Due to the membrane topology of phospholipid biosynthesis, newly synthesized phospholipids are initially located in the cytoplasmic leaflet of biogenic membranes and must be translocated to the exoplasmic leaflet to give uniform bilayer growth. It is clear from many studies that phospholipid flip-flop in biogenic membranes occurs very rapidly, within a period of a few minutes. These studies also reveal that phospholipid translocation in biogenic membranes occurs bi-directionally, independently of the phospholipid head group, via a facilitated diffusion process in the absence of metabolic energy input, and that this type of transport requires specific membrane proteins. These translocators have been termed biogenic membrane flippases, and they differ from metabolic energy-dependent transporters (ABC transporters and MDR proteins). No biogenic membrane flippases have been characterized. This review briefly discusses the importance of biogenic membrane flippases, the various assay methods used for measuring the rate of phospholipid flip-flop, and the progress that has been made towards identifying these proteins. Key Words: Phospholipids, Biogenic Membranes, Flippases, Bilayer Assembly, Endoplasmic Reticulum, Bacterial Cytoplasmic Membranes Page 2 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 102 INTRODUCTION Cellular membranes, which are involved in various biochemical functions including the transport and signalling mechanisms, are made of lipid bilayers. Seventy-eight years ago, Gorter and Grendel put forward the idea that biological membranes are based on a lipid bilayer [1]. Since then, important advances have been made in the understanding of membrane structure, phospholipid (PL) asymmetry, the organization of membrane embedded molecules and the biosynthesis of membrane components. Despite the technological advancement in molecular and cellular biology, the mechanism by which the assembly of the PL bilayer of cellular membranes occurs is not clearly known, and understanding the processes of membrane assembly remains a major challenge in molecular cell biological research. Fig. 1. Phospholipid flip-flop in biogenic membranes. (a) Phosphatidylcholine synthesis on the cytoplasmic leaflet of the eukaryotic ER and the translocation of PC to the opposite leaflet for membrane growth. (b) The synthesis of phosphatidylethanolamine and phosphatidylserine on the cytoplasmic side of the bacterial inner membrane and its flip-flop to the opposite leaflet. The double-headed arrows represent the rapid translocation of phospholipids between the two leaflets of the membrane by flippase (a membrane protein) without the requirement of metabolic energy. The question marks are by way of illustrating that the biogenic membrane flippase has not yet been identified. ER: endoplasmic reticulum; BCM: bacterial cytoplasmic membrane. Page 3 CELLULAR & MOLECULAR BIOLOGY LETTERS 103 It is a known fact that most of the glycerophospholipid-synthesizing enzymes in eukaryotic cells are membrane-bound proteins located in the microsomal fraction. These enzymes have their active sites oriented towards the cytoplasmic leaflet. For example, cholinephosphotransferase, a membrane-bound enzyme involved in the synthesis of phosphatidylcholine (PC), is present on the cytoplasmic face of the endoplasmic reticulum (ER) (Fig. 1A). In the prokaryotes, phosphatidylethanolamine (PE) is synthesized by two cytoplasmically disposed enzymes, namely phosphatidylserine synthase and phosphatidylserine decarboxylase (Fig. 1B). Due to the membrane topology of PL biosynthesis, newly synthesized phospholipids are initially located in the cytoplasmic leaflet of biogenic membranes and must be translocated to the exoplasmic leaflet to permit uniform bilayer growth [2]. Thermodynamic analysis revealed that PL translocation is energetically unfavorable because of the energy barrier to be overcome in translocating the polar head group of the phospholipids through the hydrophobic interior of the bilayer [2-4]. Consequently, PL translocation is slow in artificial liposomal systems and in certain biomembranes, lasting a few hours to days [5-7]. However, it is clear from many studies that phospholipid flip-flop in the ER and bacterial cytoplasmic membranes occurs very rapidly, taking a few seconds to minutes [8-13]. These studies revealed that PL translocation in the ER and in other biogenic membranes occurs bi-directionally independently of the PL head group. It was also demonstrated that this transport requires specific membrane proteins and does not require any metabolic energy. Based on these results, it was hypothesized that was another class of lipid translocators that facilitate the transbilayer movement of phospholipids in a metabolic energy-independent fashion [2, 9, 10, 13]. These translocators are named biogenic membrane flippases and they differ from metabolic energy- dependent transporters. For example, eukaryotic plasma membranes are equipped with special kinds of transporters that translocate phospholipids from one leaflet to other at the expense of ATP hydrolysis (Fig. 2). In this class of PL translocators are aminophospholipid translocase (APT), an inward-directed pump specific for phosphatidylserine (PS) and PE [14-16], and the ABC family of transporters, including the multi-drug resistance gene family of products, which are ATP-driven, outward directed pumps [17-19]. Apart from these, a special class of PL translocators exists, namely “scramblases”, which catalyze the bi-directional, ATP-independent, Ca 2+ -dependent transport of phospholipids resulting in their redistribution [20-22]. Even though ABC transporters, scramblases and APT are attractive candidates for lipid flippases, they are unlikely candidates for biogenic membrane flippases, as they require metabolic energy or calcium for the rapid flip-flop of phospholipids. The importance of these metabolic energy-dependent flippases was discussed in many reviews [2, 16, 22]. Page 4 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 104 Fig. 2. The transbilayer movement of phospholipids in cellular membranes. Three classes of lipid translocators have been characterized so far. The gray double-headed arrows represent bi-directional energy-independent Ca 2+ -dependent PL translocation (scramblase), the black single-headed arrows represent vectorial, energy-dependent PL translocation (aminophospholipid and the ABC family of transporters), and the black double-headed arrows represent bi-directional energy-independent PL flip-flop (biogenic membrane flippases). Metabolic energy-dependent (ATP-dependent) lipid translocators are typically located in the plasma membranes and secretory granule membranes of eukaryotic cells, whereas biogenic membranes possess the metabolic energy- independent translocators (Fig. 2). In addition, there is a possibility that metabolic energy-dependent flippases might be present in biogenic membranes as a consequence of trafficking through the ER to the plasma membrane. However, no reports are available on the effect of the presence of these energy dependent flippases on biogenic membranes during trafficking through the ER to the plasma membrane. No biogenic membrane flippase have yet been identified. In this review, the state of knowledge on biogenic membrane flippases is discussed, including different assays used to measure flippase activity and the latest progress towards identifying them. IMPORTANCE OF BIOGENIC MEMBRANE FLIPPASES PL and glycolipid translocation across bilayers was found to be important for a number of biological processes [2]. For example, as discussed earlier, newly synthesized phospholipids on the cytoplasmic leaflet of eukaryotic ER membranes need to flip to the opposite leaflet for rapid bilayer growth [8, 11, Page 5 CELLULAR & MOLECULAR BIOLOGY LETTERS 105 13]. Also, during lipoprotein assembly, a continuous movement of lipids across the ER and Golgi membranes is required for packaging PC with triglycerides [22, 23]. Lipoproteins are critical components of the circulatory lipid-handling apparatus, and their proper assembly is therefore important in reducing the deposition of nutritionally acquired and endogeneously synthesized lipids into plaques that block circulation. Hence, flipping PL across the ER and other biogenic membranes is an important cellular process. The synthesis of glycosylphosphatidylinositol (GPI) initially starts on the cytoplasmic side of the ER, but the lipids must reach the lumenal side for GPI anchoring (Fig. 3) [2, 24- 26]. As indicated in Fig. 3, the synthesis of the GPI anchor has three flipping steps in the ER; they involve glycolipids and phospholipids. The last step in the synthesis of the GPI anchor is the flipping of PE from the cytoplasmic side to the lumenal side, without which the GPI anchor is not synthesized. The absence of GPI proteins leads to a chronic blood disorder (paroxysmal nocturnal hemoglobinuria) characterized by fragile red blood cells [27]. GPI proteins form the protective surface coats of parasitic protozoa such as trypanosomes. These parasites cause deadly diseases like sleeping sickness and Chagas disease, which Fig. 3. Different steps involved in the synthesis of the GPI anchor. It has been reported that during the whole process of GPI anchor synthesis, at least 3 different flipping steps are involved (indicated by the double-headed arrow): the flipping of (i) GluA-PI-PL, (ii) dolicholphosphate mannose and (iii) phospahtidylethanolamine from the cytoplasmic leaflet to the lumen of the endoplasmic reticulum. The double-headed arrows represent rapid translocation of phospholipids between the two leaflets of the membrane by flippase (membrane protein) without the requirement of metabolic energy. The question marks are by way of illustrating that experimental evidence is required for the involvement of biogenic membrane flippase, whereas experimental evidence is available for the third flipping step (the flipping of PE) [10]. Page 6 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 106 in turn lead to heart diseases [28]. These studies showed that PL and glycolipid flip-flop is fundamental to the processes of intracellular lipid traffic and metabolism in cell. Apart from this, energy-independent translocation of glycosphingolipids was observed in the Golgi [26]. These studies clearly suggested the importance of PL translocation across the ER and other biogenic membranes. In addition, energy-independent translocation was also observed during the synthesis of glycolipids in the ER. These translocators will not be discussed here in detail, since this review is focused on metabolic energy- independent phospholipid flippases in biogenic membranes. The asymmetry of PL biosynthesis in biogenic membranes is due to the fact that PL-synthesizing enzymes are asymmetrically embedded in the membrane bilayer. The concentration gradient generated by these newly synthesized phospholipids is degenerated by rapid flip-flop. Flip-flop creates a dynamic equilibrium PL distribution among the leaflets, thereby creating a different composition of lipids in them. It was found that PC and sphingomyelin are primarily located on the lumenal side, whereas aminophospholipids (PS & PE) are located on the cytoplasmic side of the plasma membrane. However, transbilayer asymmetry in biogenic membranes such as in the eukaryotic ER is controversial since no clear picture of asymmetry is available [29]. PL asymmetry has significant physiological consequences in processes such as membrane budding and membrane protein function. BIOCHEMICAL ASSAYS TO MONITOR PHOSPHOLIPID FLIP-FLOP The development of suitable assays to measure PL translocation is crucial in identifying biogenic membrane flippases. In the simplest assay format, a labeled phospholipid or phospholipid analogue is inserted into one of the leaflets of the PL bilayer, and its transbilayer orientation is evaluated after a suitable time interval [30]. This transbilayer orientation of the phospholipid can be evaluated in several ways, including via the derivation of amine-containing headgroups with membrane-impermeant reagents and accessibility to exogenously applied phospholipases or PL transfer proteins or PL exchange proteins [30, 31]. Most of the assays developed so far to measure PL translocation involve the use of PL analogues tagged with radioactive or fluorescent or spin label probes with a certain degree of water solubility [2]. However, these PL analogues may not perfectly mimic the behavior of natural long chain lipids. The different types of PL-translocation assay currently available and described in the literature are discussed in the following section. ASSAYS BASED ON PHOSPHOLIPID ANALOGUES Phospholipid flip-flop in the ER Bishop and Bell (1985) developed an assay using a water-soluble phospholipid analogue, dibutyrylphosphatidylcholine (diC 4 PC), to study the flippase activity in rat liver microsomes [8]. This report was the first evidence for the existence of Page 7 CELLULAR & MOLECULAR BIOLOGY LETTERS 107 protein-facilitated PL translocation across the ER. A predetermined amount of [ 32 P]diC 4 PC was added to rat liver microsomes, and PL translocation was measured as the redistribution of radioactivity between the extravesicular and intravesicular space. The results showed that the transport was saturable, bi- directional and time-dependent, and also dependent on the amount of microsomes. The authors also reported that this transport led to an apparent equilibration between the extravesicular space and vesicular lumen. Later, Kawashima and Bell showed that a specific protein was involved in the transmembrane movement of lysophosphatidylcholine in liver microsomes [32]. The results clearly indicated that the transport of monoC 4 PC was saturable and dependent on time and the amount of microsomes. It was found that the monoC 4 PC transported into the microsomes was degraded to glycerophosphorylcholine. In addition, those authors showed that the transport of glycerophosphorylcholine was also saturable and dependent on time and the amount of microsomes. Backer and Dawidowicz reported on the reconstitution of PL flippase activity from rat liver microsomes into lipid vesicles [9]. Biochemical reconstitution is one of the powerful techniques available for the purification of integral membrane proteins. In this method, the membranes are solubilized in a suitable detergent and supplemented with exogenous phospholipids. This reconstitution mixture is subjected to dialysis or treated with detergent-adsorbing beads to form proteoliposomes. Using this technique, the authors reconstituted PC translocase activity from the ER. It was found that flippase activity was not detected in the reconstituted vesicles containing a microsomal lipid extract, suggesting that the translocase activity is protein mediated. The abundance of flippase among the total microsomal protein was estimated as 0.6% by weight. It was also reported separately that most glycerophospholipid classes exhibit flip-flop in microsomes, and that translocation rates differ depending on the type of phospholipid (analogue), the temperature, and the method used to study the translocation process [29]. Menon and co-workers used the assay developed by Bishop and Bell using diC 4 PC as the transporter reporter to further characterize diC 4 PC transport in rat liver microsomes [12]. The authors tested the PL translocation activity in the ER by dissolving ER membranes in Triton X-100 and reconstituting them into egg PC vesicles. Reconstituted proteoliposomes from detergent-solubilized ER vesicles were reported to be active, whereas protein-free liposomes containing ER lipids were found to be inactive. Furthermore, they reported that the slow step in the equilibration of diC 4 PC between the extra- and intravesicular space was due to the association of PL at the vesicle surface rather than the transbilayer movement itself. The amplitude of the transport assay (the extent of diC 4 PC uptake as a percentage of marker space) was found to be strongly dependent on the protein/PL ratio in proteoliposomes up to 100 mg/mmol. Beyond this ratio, the diC 4 PC uptake remained unchanged, indicating that at this ratio, all the vesicles contained at least one flippase. Page 8 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 108 The main disadvantage of an assay using a radiolabeled probe to measure the rate of transport is the time resolution. For this purpose, many researchers used fluorescent and spin-labeled PL analogues as transport reporters for their assays. There are several published reports on fluorescence-based assays to measure PL flip-flop [e.g. 33-36]. A fast bovine serum albumin (BSA) extraction method combined with fast filtration was used to assay the translocation of short-chain radio-labeled PE, PS, and PC, and their spin-labeled analogues across rat liver microsomes [10]. Briefly, a concentrated solution of phospholipids was added to microsomes, resulting in the incorporation of the lipids into the microsomes. The labeled lipids from the outer leaflet were extracted with BSA, following which the microsomes were isolated by manifold fast filtration. The amount of labeled phospholipid still associated with the microsomes was quantified and considered as the amount of flip-flop. Such results have been reported as identical for both spin and radiolabeled probes. The half-time of PL translocation was reported as 25 s for all the glycerophospholipids tested; this is much faster than what was reported earlier. The process of PL translocation was found to be bi-directional, and partially protease and N-ethylmaleimide (NEM) sensitive. Marx and co-workers developed a stopped flow assay with a high resolution to characterize the rapid transbilayer movement of phospholipids in the ER [37]. The authors used both spin and fluorescent-labeled PL analogues. In this assay, the extraction of analogues from the outer leaflet of the microsomes to the BSA was monitored by measuring the change in paramagnetic resonance or fluorescence. The time resolution of the assay revealed that the respective half- time for the transbilayer movement of spin and fluorescent-labeled phospholipid analogues was 3.5 and 9.5 s. Phospholipid flip-flop in the bacterial cytoplasmic membrane The bacterial cytoplasmic membrane is a site for the synthesis of phospholipids, some of which need to be transported to the other leaflet to maintain uniform bilayer growth. This transport of phospholipids is also mostly protein mediated, albeit insensitive to NEM. Based on pulse-labeling studies, it was demonstrated that the translocation of PE in the Gram-positive bacterium Bacillus megaterium has a half-time of 3 min [38], and that the PL translocation is independent of metabolic energy [39]. Hrafnsdóttir and co-workers showed the existence of PL flippase activity in the cytoplasmic membranes of Bacillus megaterium vesicles (BMV) [34]. The transport was monitored by measuring the increase in fluorescence as the transport substrate departed the quenching environment of the donor vesicles. The increase in fluorescence is due to two transport steps: (a) transfer from the outer leaflet of the donor vesicles to the outer leaflet of the acceptor vesicles; and (b) translocation across the membrane. In the case of PL translocation, biphasic fluorescence traces were obtained. The results showed that PL translocation in BMV occurs rapidly, with a half-time of 30 s. As for the ER, diC 4 PC translocase activity has also been demonstrated for Bacillus subtilis membranes [40]. The bacterial cytoplasmic membranes were Page 9 CELLULAR & MOLECULAR BIOLOGY LETTERS 109 solubilized in detergent and the extracts were reconstituted into proteoliposomes. The flippase activity was measured using diC 4 PC transport across egg PC vesicles. The half-time for PL translocation was found to be less than 1 min, and PL translocation was linearly dependent on the protein/PL content in reconstituted vesicles up to a ratio of 50 mg/mmol. Kubelt and co-workers also reported on the transbilayer movement of fluorescent-labeled PL in the inverted inner membrane vesicles (IIMV) of Escherichia coli [41]. PC and PE translocation across the IIMV was measured using the fluorescence stopped- flow technique and BSA back extraction. The half-times for PC and PE translocation were found to be the same, at1 min. PL translocation in the IIMV was found to be sensitive to protease, indicating that the rapid transbilayer movement of PLs is protein mediated. ASSAYS BASED ON NATURALLY OCCURRING PHOSPHOLIPIDS The assays discussed so far dealt with the use of PL analogues, and in such studies, the half-times for PL translocation were quite low. However, PL analogues may not perfectly mimic the natural long-chain phospholipids. Few reports are available on assays using natural phospholipids as transport reporters. Nicolson and Mayinger reconstituted ER membrane proteins prepared from a yeast microsomal detergent extract in proteoliposomes made of endogenous aminophospholipids [42]. In this assay, the proteoliposomes or liposomes were labeled with [ 14 C]ethanolamine for 30 min at pH 5.7 and room temperature, and treated with trinitrobenzene sulfonic acid (TNBS) for 10 min at pH 9.0 and 20ºC to modify the labeled [ 14 C]PE on the outer leaflet. It was observed that [ 14 C]ethanolamine was successfully incorporated into liposomes as [ 14 C]PE at pH 5.7, and that the labeling reaction can be separated from translocation of [ 14 C]PE by changing the pH from 5.7 to 7.4. These results clearly indicated that the TNBS only reacted with PE on the outer leaflet, and the TNBS-resistant pool of PE was sequestered to the inner leaflet of the liposomes. In addition, it was found that a pool of [ 14 C]PE unavailable for TNBS modification appeared when experiments were performed with proteoliposomes derived from yeast microsomes. The amount of [ 14 C]PE unavailable for TNBS modification depended on the protein content in the proteoliposomes. The half-time for PE translocation was reported as 10 min. [ 14 C] PE translocation has also been reported as protein-mediated based on its sensitivity to protease. Recently, a report was published on an assay to measure the translocation of dipalmitoylphosphatidylcholine (DPPC) in proteoliposomes derived from detergent extracts of rat liver ER [13, 43]. PL translocation was measured depending on the ability of phospholipase A 2 (PLA 2 ) to hydrolyze phospholipids located in the outer leaflet of a vesicle membrane bilayer. It was found that PLA 2 rapidly hydrolyzed phospholipids on the outer leaflet of both proteoliposomes and liposomes, with a half-time of 0.1 min). In the case of proteoliposomes, the initial rapid hydrolysis was followed by a slower phase indicating the translocation of phospholipids from the inner to outer leaflet. The amount of Page 10 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 110 [ 3 H] DPPC hydrolysis depended on the protein content in the reconstituted vesicles up to a ratio of 60 mg/mmol of PL. The half-time for translocation of [ 3 H] DPPC was reported as 3.3 min. Tab. 1 Summary of different assays used to measure biogenic membrane flippase activity. S.No. Transport reporter Type of mem- brane used to measure activity Source of flippase Important results References 1 [ 32 P] diC 4 PC ER microsomes Rat liver ER diC 4 PC transport was inhibited by NEM and TNBS, and was protease sensitive. Transport rate is 2.7 nmol/min/mg 8 2 [ 32 P] monoC 4 PC and [ 32 P] glycerolphosphoryl choline ER microsomes Rat liver ER monoC 4 PC-inhibited by trypsin, NEM and TNBS glycerolphosphoryl choline not inhibited by trypsin, NEM or TNBS 32 3 Brominated PC (density labeled) egg PC:brominated PC:egg PE (1:2.5:1.2) Rat liver ER Insensitive to protein modifying reagents. 15 min for complete translocation of PC 9 4 [ 14 C]-PC; (SL)-PC; (SL)-PS ER microsomes Rat liver ER Partially protease and NEM sensitive. Half-time: 25 sec 10 5 C 6 NBD-PC; C 6 NBD-PE; (SL)-PC; (SL)-PE ER microsomes Rat liver ER Half-time: C 6 NBD-PC, C 6 NBD-PE: 9.5s (SL)-PC, (SL)-PE: 3.5s 37 6 [ 14 C]-PE PC:PI:PE:PS (40:28:12:7) Yeast ER Partially protease sensitive. Half-time: 10 min 42 7 [ 3 H]-DPPC egg PC Rat liver ER Trypsin sensitive. Half- time: 3.3 min 13 8 C 6 NBD-PC; C 6 NBD-PE and C 6 NBD-PG Bacillus megaterium lipid vesicles Bacillus megaterium cytoplasmic membranes Sensitive to protease, Insensitive to NEM. Half- time: 30s 34 9 [ 3 H]diC 4 PC egg PC Bacillus subtilis membranes Sensitive to protease. Half- time: 1 min 40 10 C 6 NBD-PC; C 6 NBD-PE egg PC Inverted inner membrane vesicles of Escherichia coli Sensitive to protease. Half- time: 1 min 41 11 [ 3 H]diC 4 PC egg PC Rat liver ER Sensitive to protease. Half- time: 1 min 12 12 [ 14 C]PE Inner membranes of Escherichia Inner membranes of Escherichia coli Insensitive to sulfhydryl reagents and protease. Half-time: < 1min. 44 Page 11 CELLULAR & MOLECULAR BIOLOGY LETTERS 111 Huijbregts and co-workers were the first to report on the transmembrane movement of an endogenously synthesized phospholipid across the inner membrane of E. coli [44]. The addition of a wild-type cell lysate containing phosphatidylserine synthetase, cytidine triphosphate and [ 14 C] serine to the inside-out vesicles of a PE-deficient strain resulted in the synthesis of [ 14 C]PS, which was further converted into [ 14 C]PE. The newly synthesized [ 14 C]PE was immediately equilibrated over both membrane leaflets, as determined by its accessibility to the amino-reactive reagent fluorescamine. The half-time of translocation was found to be less than 1 min. The transport process of PE was not influenced by the presence of ATP or the proton motive force in the inside- out vesicles. The rate of PE translocation was also found to be insensitive to protease and sulfhydryl reagents. The summary of the various assays and type of lipid and membrane is given in Tab. 1. CHARACTERIZATION OF BIOGENIC MEMBRANE FLIPPASES Although the flippases themselves have not yet been identified, reports are available on the characterization of flippase activity in eukaryotes and prokaryotes. Bishop and Bell showed that diC 4 PC transport was inhibited when microsomes were treated with NEM, TNBS and proteases, suggesting that the transbilayer movement of diC 4 PC across ER (biogenic) membranes was protein mediated [8]. Similarly, the translocation of lyso PC in rat liver microsomes was also inhibited by trypsin, NEM and TNBS, suggesting once again that the transport of monoC 4 PC was protein mediated [32]. Interestingly, glycerophosphorylcholine transport was not inhibited by treatment with trypsin and with the other aforementioned protein modification reagents. This clearly suggested that a separate and distinct transport system exists for phosphatidylcholine metabolites. However, in these reports, the flippase activity was not completely stopped in intact microsomes. In another study, it was found that the translocase activity in rat liver microsomes was insensitive to protein modifying reagents [9]. This result is in contrast to the report published by Bishop and Bell [8]. Furthermore, there are a number of reports available on the sensitivity of flippases to proteases and NEM [10, 13, 41, 42]. Glycerol gradient analysis of a Triton X-100 extract of ER microsomes showed that the specific activity of the diC 4 PC transporter was found to be at its maximum in the top fractions containing less vesicular protein [12]. A majority of the diC 4 PC transporter activity was recovered in the top fractions of the gradient, with a peak corresponding to a sedimentation coefficient of 3.8S. The results clearly indicated that the detergent extract of ER could resolve into pools with different specific activities of diC 4 PC transport. This supported the hypothesis that specific proteins in the ER are responsible for diC 4 PC transport. This transport activity was also found to be sensitive to protease treatment. Anion exchange chromatography of the detergent extract of ER microsomes exhibited flippase activity in both the bound and unbound fractions. Interestingly, each of these fractions showed a difference in their sensitivity to Page 12 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 112 protease treatments, suggesting that there may be at least two different ER membrane proteins capable of translocating diC 4 PC bi-directionally without using metabolic energy. However, diC 4 PC transport activity was not stopped completely. Similarly, a glycerol gradient analysis was performed for Bacillus subtilis membranes [40]. The majority of the activity was recovered in the top fractions suggesting that the proteins responsible for diC 4 PC sediment in the range of 3.6S to 4.3S. These results also clearly indicated that only specific proteins are capable of transporting diC 4 PC. The sedimentation coefficient of bacterial flippase concurs with that for eukaryotic ER flippase. In an another study, it was demonstrated that proteoliposomes generated from a flippase-enriched Triton X-100 extract of ER can flip analogues of PC, PE and PS [11]. The treatment of a detergent extract of ER with NEM could abolish 40% of functional flippases, whereas diethylpyrocarbonate (DEPC) treatment could abolish 50% of functional flippase activity. Interestingly, combined treatment with NEM and DEPC could abolish all flippase activity. It can be concluded that glycerophospholipid flippase activity in the ER can be functionally differentiated into two independently active components based on the pattern of sensitivity to the protein-modifying reagents NEM and DEPC. Further characterization work on the NEM-sensitive class of flippase revealed that the functionally critical sulfhydryl group in the flippase protein is buried in a hydrophobic environment in the membrane, but becomes reactive on extraction of the protein into Triton X-100. These advances in understanding the properties of flippases will be crucial in the efforts to purify them. PROPOSED HYPOTHESIS TO EXPLAIN THE PL FLIP-FLOP During the last three decades, several hypotheses have been proposed to explain how phospholipids can rapidly move across the bilayer. It was suggested that PL translocation across the bilayer is due to the excess accumulation of lipids/proteins on one side of the membrane, preferably on the side where the lipids are synthesized [45]. It was also hypothesized that bilayer perturbations involving the incorporation of integral membrane proteins could lead to PL flip- flop across the bilayer without the need for metabolic energy. For example, PL translocation was observed when the integral membrane protein glycophorin was incorporated into lipid vesicles [46]. However, this might not be the actual scenario in biogenic membranes. In addition, experimental results clearly showed that only specific membrane proteins in biogenic membranes are capable of translocating PL without metabolic energy [12, 40]. The PL translocation was found to be strongly influenced by lipid composition, and PL flip-flop was inhibited by the addition of cholesterol to the membrane [47]. In another report, the effect of the phase transition of phospholipid membranes on the phospholipid flip-flop was studied [48]. It was found that there is a drastic increase in the transbilayer movement of phospholipids at the lipid phase transition of DPPC and dimyristoylphosphatidylcholine. Again, it was observed in that study that rapid transbilayer movement could be stopped when the phase Page 13 CELLULAR & MOLECULAR BIOLOGY LETTERS 113 transition was suppressed by the addition of cholesterol. This could also be speculated as one of the reasons for the absence of metabolic energy- independent PL translocation at the plasma membrane, which is very much enriched with cholesterol, whereas metabolic energy-independent PL translocation is observed in biogenic membranes such as the ER, where the cholesterol content is negligible. These studies clearly indicated that membrane composition plays an important role in phospholipid flip-flop. Scramblases, ABC transporters and MDR proteins were also thought to be responsible for PL translocation across the ER and bacterial cytoplasmic membranes. ABC transporters are substrate specific and facilitate the unidirectional transport of substrates, additionally using ATP for this transport [17, 18]. However, flip-flop in biogenic membranes was found to be bi-directional, non-specific to the phospholipid headgroup, and metabolic energy independent [8-13, 34, 41]. The observation that PL flip-flop is rapid, bi-directional and independent of ATP does not exclude an additional role for ABC transporters and MDR proteins in PL translocation in the ER and bacterial plasma membranes. It was postulated that the mechanism by which phospholipids flop (inside to outside) across the bacterial cytoplasmic membrane is mediated by the presence of membrane-spanning segments of inner membrane proteins, rather than by dedicated flippases [49, 50]. It was demonstrated that transmembrane alpha- helical peptides, mimicking membrane-spanning segments, mediate the flop of 2-6-(7-nitro-2,1,3-benzoxadiazol-4-yl) aminocaproyl (C6-NBD)-phospholipids. Based on experimental results, it was also concluded that two integral proteins of the bacterial cytoplasmic membrane, leader peptidase and the potassium channel KcsA, induced PL translocation mediated by their transmembrane domains. The respective half-times for phospholipid translocation when leader peptidase and KcsA were incorporated into the liposomes were 4.3 and 3.3 h. These half-times were found to be very much higher than the half-time of PL translocation in biogenic membranes. PL flip-flop can also be artificially promoted by chemical agents, which do not disrupt the membrane [51, 52]. A series of low molecular weight synthetic translocases, and amide and sulfonamide (tris(aminoethyl)amine) derivatives appear to enhance flip-flop by hydrogen bonding to the phospholipid head group, thereby lowering its polarity and enabling translocation into the hydrophobic membrane interior. These derivatives do not cause any leakage of contents and do not change the membrane fluidity. Synthetic translocases that selectively facilitate the translocation of specific phospholipids have also been synthesized. However, such conditions are unlikely to be mimicked in vivo. It has also been hypothesized that Escherichia coli translocon mediates phospholipid flip-flop [53]. To test this hypothesis, those authors used Escherichia coli depleted of SecYE and YidC, and assayed for PL flip-flop in both intact cells and in proteoliposomes reconstituted from the inner membrane. The experimental results showed that flippase activity does not require translocon. Taking into account the hypotheses proposed so far, the theory that Page 14 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 114 specific flippases help in PL translocation across biogenic membranes is the one that is most popular and well suited to explain experimental findings, and to mimic the reality in vivo [8-13]. CONCLUSIONS AND PERSPECTIVES The main aims of this article are to highlight the importance of biogenic membrane flippase, to review the different assays used to measure the rate of phospholipid flip-flop and to describe the progress towards identifying biogenic membrane flippases. It is clear from many studies that, in biogenic membranes, phospholipid flip-flop is a facilitated diffusion process occurring in the absence of metabolic energy input and requiring specific membrane proteins, and that it occurs very rapidly (only taking up to a few minutes), bi-directionally, and independently of the PL head group. These specific membrane proteins are translocators that have been termed biogenic membrane flippases, and they are different from metabolic energy-dependent PL transporters. Newly synthesized phospholipids on the cytoplasmic leaflet of eukaryotic ER membranes and bacterial cytoplasmic membranes need to flip to facilitate rapid bilayer growth. PL and glycolipid flip-flop is fundamental to the processes of intracellular lipid traffic and to lipid metabolism in the cell. A number of assays were used to measure the phospholipid flippase activity in biogenic membranes; these are outlined in the sections above, and summarized in Tab. 1. Most of these assays used PL analogues, and there are few reports available on studies involving natural phospholipids. It is clear from Tab. 1 that the half-time for phospholipid translocation mainly depends on the lipid composition of the membranes and the type of phospholipid probe, either a short chain phospholipid analogue or a natural phospholipid. As the half-times for PL translocation using natural phospholipids are higher [42, 43] than those reported using short-chain phospholipid analogues [8, 10, 12, 26], it is reasonable to assume that the rates of translocation from assays using PL analogues may not perfectly mimic the rates for natural phospholipids. PL flip-flop assays can be performed directly on microsomes [8] as well as using reconstituted proteoliposomes [11-13, 32, 34, 44, 45]. However, the assays using biochemical reconstitution procedures are advantageous because they can be used to monitor the flippase activity during purification. All the reported assays are good enough to measure the phospholipid flip-flop activity in biogenic membranes. However, the important point to be considered is whether these assays can be extended to monitor the enrichment of flippase activity during biochemical purification. The major limitation of these assays is discussed in Fig. 4. In these assays, the amplitude of the assay is a linear function of the protein/PL ratio in the proteoliposomes up to an extent (dose-response curve, see Fig. 4), and amplitude reaches a constant value beyond a certain ratio of protein/PL [12, 13, 40]. The slope of the dose-response curve indicates the specific activity, and the flippase activity can be measured only in the linear region in Fig. 4 as a function of protein/PL [12, 13]. As purification proceeds, Page 15 CELLULAR & MOLECULAR BIOLOGY LETTERS 115 the slope of the line increases and reaches a value close to ordinate (Fig. 4), suggesting that the protein/PL ratio reaches a very low value. It was observed that the generation of proteoliposomes at such a low protein/PL ratio and the measurement of protein content become highly difficult (unpublished result). Thus purification of flippase to homogeneity within the working range of assays (Fig. 4) might pose a constraint to the identification of flippase using vesicle population-based assays. Fig. 4. Interpretation of transport assays used to measure flippase activity reconstituted from a detergent extract of biogenic membranes. As the protein/PL ratio in the proteoliposomes increases, the diC 4 PC uptake (% content marker) or rate of DPPC hydrolysis increases and then remains constant when the ratio of protein/PL reaches a critical value, indicating that every vesicle contains at least one flippase. Below this critical ratio, the amplitude of the assay should increase with increasing protein/PL ratio, whereas the rate of flipping remains constant (3.3 min in the case of the DPPC assay [13]). Beyond this ratio, the amplitude of the assay remains constant but the rate of flipping should increase. The slope of the line gives a measure of the specific activity for flippase 1 . The lowest line indicates the protein/PL dependency on the amplitude of the assay for crude protein, reaching a critical value at around 100 mg/mmol. As the flippase source purifies, the slope of the line should increase, as shown by the other lines in the figure. For a higher enrichment of flippase activity during purification, the slope of the line approaches infinity, indicating that the measurement of specific activity will not be possible, since the protein content in the proteoliposome will be negligible. The generation of proteoliposomes at a very low protein/PL ratio and the measurement of protein will become highly difficult as purification proceeds. 1 Calculation of flippase activity using [ 3 H]DPPC hydrolysis [13]: Amplitude of assay for liposomes = a, amplitude of assay for proteoliposomes = b; Amplitude of assay due to flippase (x) = b-a ; protein/PL ratio in proteoliposme = r Specific activity of flippase = x/r (slope of the dose-response curve, Fig. 4) Activity of flippase = specific activity x total protein content in proteoliposomes. Page 16 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 116 The assays developed so far to measure PL translocation involve the use of PL analogues tagged with radioactive, fluorescent or spin labels. The radioactive- labeled diC 4 PC assay is a powerful technique for monitoring the translocation of short-chain phospholipid analogues. The termination of PL translocation and filtering of the reaction mixture on a manifold filter should be performed very rapidly, in less than one minute. Fluorescent labeling assays (with fluorescent quenching) are easy to perform with a higher time resolution than radioactive assays. The probes are generally quite stable; however, steric perturbations caused by the size of the fluorescent probe should be taken into consideration. Other than the usual fluorescent technique which uses dithionite quenching, BSA back extraction was also used. The flippase assay using PLA 2 can be used to monitor the flippase activity using long chain PL instead of analogues as a transport reporter. All these assays can be successfully used to monitor the enrichment of flippase activity during biochemical purification. The endogenous PL translocation assay [42] and in vitro assays to measure PL translocation [44] will become tedious and cumbersome when these assays are used to measure activity during purification for a larger number of samples. However, these assays are very useful in identifying biogenic membrane flippases via the genetic engineering approach. Based on experimental evidence, it has also been proposed that there may be more than one characteristic protein responsible for translocation in the ER; this was tested with an assay using a PL analogue [12]. This is also supported by experimental findings that flippase activity was higher in both unbound and bound fractions during purification experiments (unpublished results from ion exchange chromatography confirmed with an assay using natural phospholipids as the transport reporter). This will also pose a constraint on the identification of biogenic membrane flippase(s). There have been suggestions of using model systems like Mycoplasma where the content of flippase(s) is 1% by weight of reconstituted mycoplasma membrane proteins (Watkins et al. 2003, Personal Communication), and is thus higher than that for eukaryotic ER membranes (0.16-0.2%) [12, 13]. Alternative genetic approaches to identifying biogenic membrane flippases are required. The transmembrane movement of lipid-linked Man 5 GlcNAc 2 oligosaccharide is of fundamental importance in this biosynthetic pathway, and the process is predicted to be catalyzed by proteins termed flippases [54]. It was shown that the yeast Rft1 gene encodes an evolutionarily conserved protein required for the translocation of Man 5 GlcNAc 2 -PP-Dol from the cytoplasmic to the lumenal leaflet of the ER membrane. Under Rft1 repression conditions, it was demonstrated that N-linked glycosylation was reduced in the yeast strain, suggesting that Rft1 is required for the translocation of Man 5 GlcNAc 2 -PP-Dol across the ER membrane. Interestingly, it was reported that RFT1 does not reveal any ATP-binding domains, unlike plasma membrane transporters, e.g. scramblase, fatty acid transporter or bacterial Wzx protein. Similar genetic approaches are required for the identification of biogenic membrane flipases. For Page 17 CELLULAR & MOLECULAR BIOLOGY LETTERS 117 example, screening mutants of bacteria or yeast that are defective in transporting phospholipids across the membrane bilayer is required. In the literature, various hypotheses have been put forward to explain the phospholipid flip-flop across the membranes. It is clear from these studies that the lipid composition in the membrane also plays a crucial role in phospholipid flip-flop. For example, the presence of cholesterol in membranes stopped phospholipid translocation. Among the hypotheses, the theory that characteristic flippases help in PL translocation across biogenic membranes is the most popular and well suited to explain experimental findings as well as to mimic the situation found in vivo. Effective purification strategies, novel biochemical assays and molecular and genetic approaches are required to crack the mystery of phospholipid flip-flop. Acknowledgements. Sathyanarayana Gummadi would like to dedicate this paper to Prof. A. K. Menon, and to acknowledge GRS and GSS for their support. Dr. Manoj and Dr. Gopal are acknowledged for their critical reading of the manuscript. REFERENCES 1. Gorter, E. and Grendel F. On biomolecular layers of lipids on the chromocytes of the blood. J. Exp. Med. 41 (1925) 439-443. 2. Menon A.K. Flippases. Trends Cell Biol. 5 (1995) 355-360. 3. Bretscher, M.S. Some general principles of membrane structure. in: The cell surface and development, (Moscana, A., Ed.), Wiley, New York, 1973, 17- 27. 4. Raggers, R.J., Pomorski, T., Holthuis, J.C.M., Kalin, M. and van Meer, G. Lipid traffic: the ABC of transbilayer movement. Traffic 1 (2000) 226-234. 5. Kornberg, R.D. and McConnell, H.M. Inside-outside transitions of phospholipids in vesicle membranes. Biochemistry 10 (1971) 1111-1120. 6. Lenard, J. and Rothman J.E. Transbilayer distribution and movement of cholesterol and phospholipid in the membrane of influenza virus. Proc. Natl. Acad. Sci. U.S.A. 73 (1976) 391-395. 7. Eastman, S.J., Hope, M.J. and Cullis, P.R. Transbilayer transport of phosphatidic acid in response to transmembrane pH gradients. Biochemistry 30 (1991) 1740-1745. 8. Bishop, W.R. and Bell, R.M. Assembly of endoplasmic reticulum phospholipid bilayer: the phosphatidylcholine transporter. Cell 42 (1985) 51-60. 9. Backer, J.M. and Dawidowicz, E.A. Reconstitution of a phospholipid flippase from rat liver microsomes. Nature 327 (1987) 341-343. 10. Buton, X., Morrot, G., Fellmann, P. and Seigneuret, M. Ultrafast glycerophospholipid selective transbilayer motion mediated by a protein in the endoplasmic reticulum membrane. J. Biol. Chem. 271 (1996) 6651- 6657. Page 18 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 118 11. Chang, Q.L., Gummadi, S.N. and Menon, A.K. Chemical modification identifies two populations of glycerophospholipid flippase in rat liver ER. Biochemistry 43 (2004) 10710-10718. 12. Menon, A.K., Watkins, W.E. and Hrafnsdóttir, S. Specific proteins are required to translocate phosphatidylcholine bidirectionally across the endoplasmic reticulum. Curr. Biol. 10 (2000) 241-252. 13. Gummadi, S.N. and Menon, A.K. Transbilayer movement of dipalmitoylphosphatidylcholine in proteoliposomes reconstituted from detergent extracts of endoplasmic reticulum. Kinetics of transbilayer transport mediated by a single flippase and identification of protein fractions enriched in flippase activity. J. Biol. Chem. 277 (2002) 25337-25343. 14. Devaux, P.F. Protein involvement in transmembrane lipid asymmetry. Annu. Rev. Biophys. Biomol. Struct. 21 (1992) 417-419. 15. Tang, X., Halleck, M.S., Schlegel, R.A. and Williamson, P. A subfamily of P-type ATPases with aminophospholipid transporting activity. Science 272 (1996) 1495-1497. 16. Daleke, D.L. Regulation of transbilayer plasma membrane phospholipid asymmetry. J. Lipid Res. 44 (2003) 233-242. 17. Romsicki, Y. and Sharom, F.J. Phospholipid flippase activity of the reconstituted P-glycoprotein multidrug transporter. Biochemistry 40 (2001) 6937-6947. 18. van Helvoort, A., Smith, A.J., Sprong, H., Fritzsche, I., Schinkel, A.H. Borst, P. and van Meer, G. MDR1 P-glycoprotein is a lipid translocase of broad specificity, while MDR3 P-glycoprotein specifically translocates phosphtidylcholine. Cell 87 (1996) 507-517. 19. Ruetz, S. and Gros, P. Phosphatidylcholine translocase: a physiological role for the mdr2 gene. Cell 77 (1994) 1071-1081. 20. Bevers, E.M., Comfurius, P., Dekkers, D.W.C. and Zwaal, R.F.A. Lipid translocation across the plasma membrane of mammalian cells. Biochim. Biophys. Acta 1439 (1999) 317-330. 21. Zhou, Q., Zhao, J., Stout, J.G., Luhm, R.A., Wiedmer, T. and Sims, P.J. Molecular cloning of human plasma membrane phospholipid scramblase. J. Biol. Chem. 272 (1997) 18240-18244. 22. Daleke, D.L. and Lyles, J.V. Identification and purification of aminophospholipid flippases. Biochim. Biophys. Acta 1486 (2000) 108- 127. 23. Rush, J.S. and Waechter, C.J. Transmembrane movement of a water-soluble analogue of mannosylphosphoryldolichol is mediated by an endoplasmic reticulum protein. J. Cell Biol. 130 (1995) 529-536. 24. Maeda, Y., Watanabe, R., Harris, C.L., Hong, Y., Ohishi, K., Kinoshita, K. and Kinoshita, T. PIG-M transfers the first mannose to glycosylphosphatidylinositol on the lumenal side of the ER. EMBO J. 20 (2001) 250-261. Page 19 CELLULAR & MOLECULAR BIOLOGY LETTERS 119 25. Vidugiriene, J. and Menon, A.K. The GPI anchor of cell-surface proteins is synthesized on the cytoplasmic face of the endoplasmic reticulum. J. Cell Biol. 127 (1994) 333-341. 26. Sprong, H., van der Sluijs, P. and van Meer, G. How proteins move lipids and lipids move proteins. Nat. Rev. Mol. Cell Biol. 2 (2001) 504-513. 27. Leidich, S.D., Drapp, D.A. and Orlean, P. A yeast mutant blocked at the first step in GPI anchor synthesis. J. Cell Biol. 269 10193-10196. 28. Anselmi, A. and Moleiro, F. Chagas disease and chagasic myocardiopathies. Arch. Inst. Cardiol. Mex. 42 (1972) 622-628. 29. Devaux, P.F. Phospholipid translocation in the endoplasmic reticulum. in: Subcellular Biochemistry (Borgese, N. and Harris, J.R., Eds.) Plenum Press, New York, 1993, 273-285. 30. Zachowski, A. Phospholipids in animal eukaryotic membranes: transverse asymmetry and movement. Biochem. J. 294 (1993) 1-14. 31. Pomorski, T., Hrafnsdóttir, S., Devaux, P.F. and van Meer, G. Lipid distribution and transport across cellular membranes. Sem. Cell Develop. Biol. 12 (2001) 139-148. 32. Kawashima, Y. and Bell, R.M. Assembly of the endoplasmic reticulum phospholipid bilayer. Transporters for phosphatidylcholine and metabolites. J. Biol. Chem. 262 (1987) 16495-16502. 33. Herrmann, A., Zachowski, A. and Devaux, P.F. Protein-mediated phospholipid translocation in the endoplasmic reticulum with a low lipid specificity. Biochemistry 29 (1990) 2023-2027. 34. Hrafnsdóttir, S., Nichols, J.W. and Menon, A.K. Transbilayer movement of fluorescent phospholipids in Bacillus megaterium membrane vesicles. Biochemistry 36 (1997) 4969-4978. 35. Pohl, A., Lage, H., Müller, P., Pomorski, T. and Herrmann, A. Transport of phosphatidylserine via MDR1 (multidrug resistance 1) P-glycoprotein in a human gastric carcinoma cell line. Biochem. J. 365 (2002) 259-268. 36. McIntyre, J.C. and Sleight R.G. Fluorescence assay for phospholipid asymmetry. Biochemistry 30 (1991) 11819-11827. 37. Marx, U., Lassmann, G., Holzhütter, H-G., Wüstner, D., Müller, P., Höhlig, A., Kubelt, J. and Herrmann, A. Rapid flip-flop of phospholipids in endoplasmic reticulum membranes studied by a stopped flow approach. Biophys. J. 78 (2000) 2628-2640. 38. Rothman, J.E. and Kennedy, E.P. Rapid transmembrane movement of newly synthesized phospholipids during membrane assembly. Proc. Natl. Acad. Sci. USA 74 (1977) 1821-1825. 39. Langley, K.E. and Kennedy, E.P. Energetics of rapid transmembrane movement and of compositional asymmetry of phosphatidylethanolamine in membranes of Bacillus megaterium. Proc. Natl. Acad. Sci. U.S.A. 76 (1979) 6245-6249. Page 20 CELL. MOL. BIOL. LETT. Vol. 10. No. 1. 2005 120 40. Hrafnsdóttir, S. and Menon, A.K. Reconstitution and partial characterization of phospholipid flippase activity from detergent extracts of Bacillus subtilis. J. Bacteriol. 182 (2000) 4198-4206. 41. Kubelt, J., Menon, A.K., Müller, P. and Hermann, A. Transbilayer movement of fluorescent phospholipid analogues in the cytoplasmic membrane of Escherichia coli. Biochemistry 41 (2002) 5605-5612. 42. Nicolson, T. and Mayinger, P. Reconstitution of yeast microsomal lipid flip-flop using endogenous aminophospholipids. FEBS Lett. 476 (2000) 277-281. 43. Gummadi, S.N., Hrafnsdottir, S., Walent, J.E., Watkins, W. and Menon, A.K. Reconstitution and assay of biogenic membrane-derived phospholipid flippase activity in proteoliposomes in: Methods in Molecular Biology (Selinsky, B.S. Ed.), Humana Press, New Jersey, 2004, 271-280. 44. Huijbregts, R.P.H., de Kroon, A.I.P.M. and de Kruijff, B. Rapid transmembrane movement of newly synthesized phosphatidylethanolamine across the inner membrane of Escherichia coli. J. Biol. Chem. 273 (1998) 18936-18942. 45. van der Steen, A.T.M., Taraschi, T.F., Voorhout, W.F. and de Kruijff, B. Barrier properties of glycophorin-phospholipid systems prepared by different methods. Biochim. Biophys. Acta 733 (1983) 51-64. 46. de Kruijff, B., van Zoelen, E.J. and van Deenen, L.L. Glycophorin facilitates the transbilayer movement of phosphatidylcholine in vesicles. Biochim. Biophys. Acta 509 1978 537-542. 47. Kol, M.A., van Laak, A.N., Rijkers, D.T., Killian, J.A., de Kroon, A.I.P.M. and de Kruijff, B. Phospholipid flop induced by transmembrane peptides in model membranes is modulated by lipid composition. Biochemistry 42 (2003) 231-237. 48. John, K., Schreiber, S., Kubelt, J., Herrmann, A. and Müller, P. Transbilayer movement of phospholipids at the main phase transition of lipid membranes: implications for rapid flip-flop in biological membranes. Biophys. J. 83 (2002) 3315-3323. 49. Kol, M.A., de Kroon, A.I.P.M., Rijkers, D.T., Killian, J.A. and de Kruijff, B. Membrane-spanning peptides induce phospholipid flop: a model for phospholipid translocation across the inner membrane of E. coli. Biochemistry 40 (2001) 10500-10506. 50. Kol, M.A., van Dalen, A., de Kroon, A.I.P.M., de Kruijff, B. Translocation of phospholipids is facilitated by a subset of membrane-spanning proteins of the bacterial cytoplasmic membrane. J. Biol. Chem. 278 (2003) 24586- 24593. 51. Boon, J.M. and Smith, B.D. Facilitated phospholipid translocation across vesicle membranes using low molecular weight synthetic flippases. J. Am. Chem. Soc. 121 (1999) 11924-11925. 52. Boon, J.M. and Smith, B.D. Facilitated phosphatidylcholine flip-flop across erythrocyte membranes using low-molecular weight synthetic translocases. J. Am. Chem. Soc. 123 (2001) 6221-6226. Page 21 CELLULAR & MOLECULAR BIOLOGY LETTERS 121 53. Watkins, W.E. and Menon, A.K. Reconstitution of a phospholipid flippase activity from E. coli inner membrane: a test of the protein translocon as a candidate flippase. Biol. Chem. 383 (2002) 1435-1440. 54. Helenius, J., Ng, D.T.W., Marolda, C.L., Walter, P., Valvano, M.A. and Aebi, M. Translocation of lipid-linked oligosaccharides across the ER membranes requires Rft1 protein. Nature 415 (2002) 447-450. 以上內容來自 "the work of transmembrane phosphatic acid my done by this"