The yeast plasma membrane P4-ATPases are major transporters for lysophospholipids

Biochimica et Biophysica Acta 1791 (2009) 620–627

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a l i p

Review

The yeast plasma membrane P4-ATPases are major transporters for lysophospholipids☆ Wayne R. Riekhof, Dennis R. Voelker ⁎ Department of Medicine, National Jewish Health, 1400 Jackson St., Denver, CO 80206, USA

a r t i c l e

i n f o

Article history: Received 7 January 2009 Received in revised form 23 February 2009 Accepted 24 February 2009 Available online 5 March 2009 Keywords: Lysophospholipid P-type ATPase Edelfosine Miltefosine Membrane asymmetry Flippase Lipid transport Trafficking Membrane biogenesis Golgi Secretion

a b s t r a c t The transbilayer movement of phospholipids plays an essential role in establishing and maintaining the asymmetric distribution of lipids in biological membranes. The P4-ATPase family has been implicated as the major transporters of the aminoglycerophospholipids in both surface and endomembrane systems. Historically, fluorescent lipid analogs have been used to monitor the lipid transport activity of the P4ATPases. Recent evidence now demonstrates that lyso-phosphatidylethanolamine (lyso-PtdEtn) and lysophosphatidylcholine (lyso-PtdCho) are bona fide biological substrates transported by the yeast plasma membrane ATPases, Dnf1p and Dnf2p, in consort with a second protein Lem3p. Subsequent to transport, the lysophospholipids are acylated by the enzyme Ale1p to produce PtdEtn and PtdCho. The transport of the lysophospholipids occurs at rates sufficient to support all the PtdEtn and PtdCho synthesis required for rapid cell growth. The lysophospholipid transporters also utilize the anti-neoplastic and anti-parasitic ether lipid substrates related to edelfosine. The identification of biological substrates for the plasma membrane ATPases coupled with the power of yeast genetics now provides new tools to dissect the structure and function of the aminoglycerophospholipid transporters. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Membrane phospholipids are asymmetrically distributed across the bilayers of multiple cellular membranes, and this physical arrangement has important implications for membrane function. The asymmetric distribution of lipid species is largely the result of ATP-driven, intramembrane lipid translocation, catalyzed by ABC transporters and P-type ATPases in specific membrane compartments [1]. The first putative lipid-translocase to be identified at the primary sequence level, and the founding member of the P4-ATPase class, was the bovine adrenal gland chromaffin granule protein ATPase II, now named ATP8A1. ATP8A1 was shown to be a phosphatidylserine (PtdSer) stimulated ATPase, providing the first clue that this protein might act as a lipid translocase responsible for establishing or maintaining phospholipid asymmetry. Cloning of the yeast gene encoding the ATPase II homolog, Drs2p, and subsequent functional analysis of this and other family members, revealed these translocases are conserved across all eukaryotic phyla and probably act to directly

produce membrane phospholipid asymmetry. Additional analyses revealed that the Drs2p was principally localized in the Golgi apparatus, and played an important role in regulating constitutive protein secretion through the organelle [2]. Since the initial discoveries uncovering the Golgi-localized P4ATPases and their roles in protein secretion, the genes and cDNAs encoding multiple other members of this family have been cloned and their actions characterized. Members of this transporter family exhibit diverse functions and when mutated, disrupt multiple cellular and tissue-specific functions, including bile excretion [3], lysophospholipid uptake [4,5], endocytosis [6], drug metabolism [7– 10], and cold sensitivity [11]. This review will focus on the discovery, biochemical properties, and cellular functions of the P4-family of Ptype ATPases, and their requisite β-subunits of the Cdc50p family, with a particular emphasis on their role as lysophospholipid translocases and their functions as transporters of lysophospholipid-like drugs, the structures of which are given in Fig 1. 2. The Dnf1/2p-Lem3p complexes as plasma membrane phospholipid translocases

☆ Work in the authors' laboratory was supported by NIH grants F32-GM076798 (to W.R.R.), 5R37-GM32453, and GM081461 (to D.R.V). W.R.R. is a recipient of the American Cancer Society Great-West Division Post-Doctoral fellowship award (PF-06-288-01-CSM). ⁎ Corresponding author. Tel.: +1 303 398 1300. E-mail address: [email protected] (D.R. Voelker). 1388-1981/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.bbalip.2009.02.013

After the identification and initial characterization of the bovine ATPase II (ATP8A1) gene [12], it was recognized that the yeast genome contains five closely related homologs, with the Drs2p product showing the closest sequence identity to the ATP8A1 transporter

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

621

Fig. 1. Structures of substrates for P4-family ATPases. The structures of the P4-ATPase substrates discussed in the text are given, and abbreviations are: NBD-, 7-nitrobenz-2-oxa1,3-diazol-4-yl; PtdCho, phosphatidylcholine; PtdEtn, phosphatidylethanolamine; PtdSer, phosphatidylserine; R, saturated hydrocarbon chain of 15–17 carbons. The structure of the C6-NBD moiety is given in the structure for NBD-PtdSer, and abbreviated in the other lipids.

(Fig. 2). Drs2p was shown to exert its function in the Golgi and to be essential for proper vesicular trafficking [13]. Drs2p was also shown to directly interact with, and require for its proper function, the accessory protein Cdc50p [14]. Members of the Cdc50p family in yeast include the homologs Lem3p and Crf1p. These proteins are ∼450 amino acids, with two transmembrane helices at the N- and C-termini, a large, glycosylated, exoplasmic loop between the transmembrane domains, and short, cytoplasmic N- and C-terminal domains at the ends [2]. The functional Drs2p–Cdc50p transporter complex has been characterized as preferring PtdSer as a substrate, but this specific PtdSer transport function is not essential in vivo [15]. The body of work describing the initial characterization of Drs2p, Cdc50p, and the other yeast P4-ATPases and their β-subunits or accessory factors has been reviewed recently by several authors [1–3]. It is important to note at the outset that we refer to the P4-ATPases as lipid translocases throughout this review, but this activity has not been formally demonstrated in an unequivocal way. For example, unlike some of the mineral ion transporters of the P-type superfamily, the P4-lipid

translocases have not been purified and reconstituted in liposomes for activity assays measuring native substrates. Thus, it remains a formal possibility that the lipid translocase activity associated with the presence or absence of a particular P4-ATPase is due to a secondary effect resulting from loss of the transporter. We believe that biological parsimony coupled with the preponderance of evidence argues for the direct action of these ATPases as lipid translocases. In addition to Drs2, the proteins Neo1p and the Drs2–Neo1 family 1, 2, and 3 proteins (Dnf1–3p) are also P4-ATPases in yeast, with specific functions described below. The internalization of the fluorescent phospholipid analogs NBD-PtdCho and NBD-PtdEtn by yeast was initially described by the Nichols laboratory [16,17]. Elucidation of the activities and subcellular location of Drs2p and Dnf3p and their accessory proteins [13,18,19] clearly demonstrated these complexes do not function as the principal PtdCho and PtdEtn translocases of the plasma membrane, as originally postulated in yeast [20]. Two other members of the yeast P4-ATPase family encoded by the DNF1 and DNF2 genes have now been implicated in plasma

622

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

from chemical modification studies using trinitrobenzenesulfonate (TNBS), which derivatizes PtdEtn in the external leaflet of the plasma membrane from mutant cells. Although Drs2p localizes to the Golgi, there is evidence the protein transiently cycles through the plasma membrane and early endosomal compartments, thereby executing some translocation of NBD-aminoglycerophospholipids at the cell surface. The genetic elimination of Drs2p along with Dnf1p and Dnf2p further amplifies the detectable level of PtdEtn at the plasma membrane as assessed by Ro09-0198 sensitivity and TNBS reactivity [22–24]. In an intersecting line of experimentation Kato et al. [25] used a forward genetic screen to identify strains with enhanced Ro09-0198 sensitivity (ros-strains). The hypersensitivity of one ros strain was reversed by a gene designated ROS3, which was cloned by complementation. This gene was previously identified in an unrelated screen for yeast glucocorticoid responsiveness and assigned the name LEM3. The ros3/lem3 mutant shared many similarities with the dysfunctions in membrane asymmetry and NBD-lipid translocation associated with the dnf1Δ dnf2Δ mutants. In addition to hypersensitivity to Ro090198 peptide, the strains exhibited increased binding to biotinylated forms of the peptide as detected with FITC-streptavidin [26]. The import of NBD-PtdEtn and NBD-PtdCho, but not NBD-PtdSer was also severely impaired in the ros3 mutant. A Lem3p/Ros3p-EGFP fusion protein was expressed and shown to be localized in the plasma membrane, consistent with a role for the protein in phospholipid translocation. LEM3/ROS3 is a member of a three-gene family in yeast, which also includes CDC50 and YNR048W/CRF1 [2]. As described above, these proteins act as accessory β-subunits to the P4-ATPases, and function to specify their localization in the proper subcellular compartment, and perhaps play a role in substrate specificity (Fig. 3). A recent hypothesis put forward by Poulsen et al. [27] makes a compelling case for similarities in the structure and function of the P4-family β-subunits and the β- and γ-subunits of the mineral ion transporting P-type Fig. 2. Phylogenetic tree of P4-ATPases from major model organisms. The yeast P4ATPases Drs2p and Dnf1p were used as queries to search for protein sequences with E b 10− 50 encoded by following genomes: Caenorhabditis elegans (Ce); Mus musculus (Mm); Homo sapiens (Hs); Drosophila melanogaster (Dm); and Saccharomyces cerevisiae (Sc). When multiple transcriptional variants for a given protein were available, the longest version was used. Proteins were aligned and the rooted tree generated with the CLUSTALW program. The subfamily groupings are intended to highlight the evolutionary trends of the P4-ATPase family, i.e. mammalian genomes encode multiple individual ATPase paralogs, while invertebrate genomes (e.g. the fly and worm) encode only one or two members for each subfamily.

membrane lipid transport processes by multiple independent studies. Experiments by Pomorski et al. [6] showed that Dnf1p and Dnf2p localize to the plasma membrane. Yeast strains harboring inactivated alleles for DNF1 and DNF2 lack the ATP dependent translocase activity required for internalization of NBD-analogs of PtdSer, PtdEtn, and PtdCho. In dnf1Δ dnf2Δ strains, NBD-PtdCho and NBD-PtdEtn internalization was reduced by 75%, and NBD-PtdSer internalization was reduced 50%, with the remaining NBD-PtdSer internalization occurring via an ATP-independent mechanism. Dnf2p activity was responsible for the bulk of the phospholipid internalization activity since strains lacking DNF1 showed a b10% reduction in uptake for all three NBD-labeled aminoglycerophospholipids. The functional consequences of removing Dnf1p and Dnf2p from the plasma membrane, upon the steady state level of PtdEtn in the external leaflet of the plasma membrane bilayer, were probed using the cytolytic peptide Ro09-0198 (also known as cinnamycin), which binds to the lipid, thus initiating cell lysis [21–23]. These studies revealed that the mutant strains were hypersensitive to Ro09-0198, consistent with the conclusion that Dnf1p and Dnf2p play an important role in regulating the distribution of PtdEtn between the bilayer leaflets. Additional support for this conclusion was obtained

Fig. 3. Localization and substrate specificity of the P4-ATPases of yeast (Dnf1p, Dnf2p, Dnf3p, and Drs2p) and their β-subunits (Lem3p, Cdc50p, and Crf1p). A simplified model of the phospholipid translocase complexes, organized with regard to their location within the yeast cell and the substrates which they have been documented to transport. The metabolism of the transported lysophospholipids by the Ale1p acyltransferase, located in both the endoplasmic reticulum (ER) and mitochondriaassociated membrane (MAM) is also shown. ER, endoplasmic reticulum; lyso-, 1-acyl-2hydroxyl-; MAM, mitochondria-associated ER membrane, Mito, mitochondrion; NBD-, 7-nitrobenz-2-oxa-1,3-diazol-4-yl-; PtdCho, phosphatidylcholine; PtdEtn, phosphatidylethanolamine; PtdSer, phosphatidylserine.

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

ATPases. Cdc50p pairs with Drs2p, and the complex of these proteins functions in the Golgi. In the absence of Cdc50p, Drs2p fails to exit from the ER, consistent with the current idea that Cdc50p acts as a molecular chaperone required for correct organelle targeting of the ATPase subunit. The similar phenotypes of dnf1Δ dnf2Δ and lem3/ ros3Δ mutations led Saito et al. [14] to test for the presence of a Dnf1p and Lem3p/Ros3p complex. Immunoprecipitation studies performed with membrane preparations from strains expressing epitope tagged versions of the proteins demonstrated specific interactions between Dnf1p and Lem3p/Ros3p. The similarities in the lipid uptake and plasma membrane PtdEtn asymmetry phenotypes of the lemsΔ/ros3Δ and dnf1Δ dnf2Δ mutant strains, provides strong evidence that Lem3p/Ros3p provides the same chaperone and/or localization functions to Dnf1p and Dnf2p [28], that Cdc50p provides for Drs2p [14]. Similar studies have implicated Crf1 as the accessory β-subunit for Dnf3 [28]. In addition to their roles in regulating the surface exposure of PtdEtn and translocation of NBDphospholipids, Dnf1p, Dnf2p, and Lem3p also function in the uptake of lysophospholipids and lysophospholipid-like drugs, described below. 3. Lysophospholipids and structurally related drugs are substrates for P-type ATPases 3.1. Lyso-PtdEtn uptake and acylation The structures of the relevant lysophospholipids, fluorescentphospholipids, and alkyl-lysophospholipid-like drugs are given in Fig 1. Recent work demonstrates specific, and high rate transport of lysophosphatidylethanolamine (lyso-PtdEtn) and lyso-phosphatidylcholine (lyso-PtdCho) by yeast cells. This line of investigation developed from genetic studies demonstrating that several yeast strains behaved as lysophospholipid auxotrophs [4]. In these screens, the ethanolamine auxotrophic psd1Δ psd2Δ yeast strain, devoid of PtdSer decarboxylase activity, was shown to use lyso-PtdEtn as a source for PtdEtn and PtdCho synthesis by the direct uptake and acylation of the lysophsopholipid from the growth medium. Surprisingly, lyso-PtdEtn is approximately 10-fold more efficient than free ethanolamine at supporting the growth of the psd1Δ psd2Δ mutant. The psd1Δ psd2Δ strains are deficient in mitochondrial PtdEtn content [29], and this membrane defect is not reversed when the strains are supplemented with free ethanolamine, which is used for PtdEtn synthesis in the endoplasmic reticulum (ER). This data indicates that PtdEtn derived from the Kennedy pathway in the ER in unable to effectively support the lipid content and function of mitochondria. However, the mitochondrial pool of PtdEtn in psd1Δ psd2Δ strains was fully restored to wild-type levels upon lyso-PtdEtn supplementation, demonstrating that the lysophospholipid enters novel transport pathways that support mitochondrial membrane biogenesis. The capacity for lyso-PtdEtn uptake by yeast is remarkable, and far exceeds the rate necessary for PtdEtn and PtdCho synthesis to support membrane biogenesis and cell division. The uptake of 1-palmitoyllyso-PtdEtn by yeast strains, as measured by the transport of radiolabeled lipid, was 11 nmol 107 cells− 1 h− 1, which is approximately three-times the rate required for phospholipid synthesis necessary for the maximum rate of cell growth The possibility that the lysophospholipid was first degraded and its constituents subsequently reassembled into phospholipid, was examined by radiolabeling experiments, using lyso-PtdEtn containing a fixed ratio of 3H-labeled fatty acid and 14C-labeled ethanolamine headgroup. The 3H/14C ratio of the PtdEtn, derived from the uptake and acylation of the radiolabeled lyso-PtdEtn, was nearly identical to that of the starting lysophospholipid. This latter result demonstrates the relatively high activity of the synthetic pathway involving the direct uptake and subsequent acylation of lyso-PtdEtn. The acylation of lyso-PtdEtn is now known to occur through the action of an acyltransferase named Ale1p, which uses multiple lysophospholipid substrates [30]. More-

623

over, only trace amounts of lyso-PtdEtn were detected within the cell, indicating that the acylation of this lipid proceeded at an extremely rapid rate. With longer incubations of high concentrations of lysoPtdEtn, the surplus lipid imported by the plasma membrane transporters is metabolized into triacylglycerol (Riekhof and Voelker, unpublished observations). The rates of lyso-PtdEtn transport are linear for at least 2 h , leading us to conclude that the transport process is not down-regulated upon continued exposure to lysophospholipid. The very high rate and capacity for lyso-PtdEtn influx was unexpected, and raised questions about the role of lyso-PtdEtn uptake in the biology and ecology of yeast. In a natural setting, lysophospholipid uptake may represent a previously unidentified nutrient scavenging mechanism. The main ecological niche of S. cerevisiae and other saccharophilic yeasts is in decaying organic matter, such as decomposing fruit. Thus, it is not surprising that yeast are proficient at extracting organic nutrients from their environment, and bypassing the requirement for de novo amino acid, fatty acid, carbohydrate and vitamin synthesis. The obvious advantage to importing and acylating lyso-PtdEtn, relative to de novo PtdEtn synthesis is in the savings of approximately 55 mole of ATP/mole of lysophospholipid, which would otherwise be required for the synthesis of fatty acids, the synthesis and esterification of glcyerol phosphate, and formation and transfer of the phosphoethanolamine headgroup. From the experimental observation of high rates of lyso-PtdEtn uptake by yeast cells, we reasoned that a dedicated, high-capacity active transport system would be required. Taking a candidate-gene approach, we targeted the plasma membrane localized P-type ATPase proteins, Dnf1p and Dnf2p, and their β-subunit, Lem3p, as potential lyso-PtdEtn transporters. Indeed, deletion of the LEM3 gene resulted in a N85% reduction in lyso-PtdEtn transport. Likewise, single deletion mutants lacking either DNF1 or DNF2 genes, resulted in a ∼ 15% or ∼40% reduction in lyso-PtdEtn transport rates, respectively. A tandem deletion mutant strain lacking DNF1 and DNF2 showed a further reduction, retaining only ∼ 25% of the wild-type capacity for lysoPtdEtn uptake. These results closely mirror those of previous work on the role of Dnf1p and Dnf2p on the uptake of NBD-labeled PtdEtn [6], but provides for more quantitative measurements of lipid transport under conditions of normal growth. These measurements also demonstrated that the rate of lyso-PtdEtn uptake in the dnf1Δ dnf2Δ mutant, while greatly reduced relative to the wild-type strain, was still significantly higher than that of the lem3Δ mutant. This result implies that, along with Dnf1p and Dnf2p, there is an additional, unidentified transport system for lyso-PtdEtn that is dependent on Lem3p function. The dnf1Δ dnf2Δ mutant exhibits defective endocytosis [6] and it is likely that Drs2p or Dnf3p become trapped at the plasma membrane and function as the unidentified transporters. Other groups [31,32] have previously reported that a minor population of Drs2p cycles between the plasma membrane and the Golgi via endocytosis, and disruption of endocytosis is expected to influence the population of Drs2p, Dnf3p, and/or Neo1p molecules found at the cell surface. Mutations arresting endocytic processes, such as end4 or vrp1 [33], result in the trapping of Drs2p at the plasma membrane, but the consequences of the tandem dnf1Δ dnf2Δ mutation upon the localization of Drs2p have not been investigated in detail. Further studies will be needed for elucidation of the residual lyso-PtdEtn transport activity in the dnf1Δ dnf2Δ strain. 3.2. Uptake and metabolism of Lyso-PtdCho and other lysophospholipids Lyso-PtdCho is also imported and rapidly acylated by yeast [5]. LysoPtdCho transport is primarily attributable to Dnf2p, since the dnf1Δ mutant showed only a 10% reduction in lyso-PtdCho transport rate, whereas the dnf2Δ mutant showed a 60% reduction in the lipid transport. Lyso-PtdCho is acylated by the Ale1p enzyme, described above as the lyso-PtdEtn acyltransferase. These results and relative contributions of Dnf1p and Dnf2p to the overall lyso-PtdCho transport

624

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

rate are similar to the general findings of Pomorski et al. for NBDPtdCho [6], but are much more quantitative in nature with regard to absolute rates of lipid transport. The identification of lyso-PtdEtn and lyso-PtdCho transport [4,5] also identified a direct, unambiguous growth assay for the measurement of Dnf1p/Dnf2p and Lem3p function. Specifically, in a strain with genetic deletions for the PtdEtn methyltransferases, Pem1p/Cho2p and Pem2p/Opi3p, growth is strictly dependent on the uptake and metabolism of either choline or lyso-PtdCho as a source of PtdCho for membrane biogenesis. When lyso-PtdCho is supplied as the sole precursor for PtdCho biosynthesis, the function of Lem3p becomes essential for the growth of this strain. Thus the pem1Δ pem2Δ lem3Δ strain shows an absolute growth requirement for free Cho, and cannot utilize lyso-PtdCho as a source of PtdCho. This finding will make the pem1Δ pem2Δ dnf1Δ dnf2Δ lem3Δ strain an ideal host for the expression and structure/function studies of both the ATPase and accessory β-subunits of the yeast P4-ATPases acting at the plasma membrane. Additionally, this strain will serve as an excellent host for the expression of heterologous, candidate transporter proteins from other species, such as the many uncharacterized “orphan” transporters detailed in the phylogenetic tree shown in Fig. 2. The identification of lyso-PtdEtn and lyso-PtdCho as substrates of the plasma membrane P4-ATPases raised the question of whether lyso-PtdSer or lyso-PtdIns are substrates for Dnf1p and Dnf2p [5]; and we tested the uptake and metabolism of these compounds by both biochemical and genetic analyses. First, we assessed the ability of these lysophospholipids to support the growth of the mutant strains defective in their respective biosynthetic pathways. For lyso-PtdIns, we tested whether the compound could support either the ino1Δ mutant (an inositol auxotroph), or a Pis1p temperature sensitive mutant, deficient in PtdIns synthesis at high temperature. We also tested whether lyso-PtdSer could support the growth of the cho1Δ mutant, which is deficient in PtdSer synthase activity, resulting in choline or ethanolamine auxotrophy. In neither case did the supplied lysophospholipid rescue the growth of the appropriate mutant. Since this result could either be due to failure to import the lipid, failure to transport the lipid to the site of the acyltransferase after uptake, or failure to acylate the lipid, we synthesized radiolabeled lyso-PtdSer and lyso-PtdIns and assessed their uptake and metabolism when incubated with yeast. Despite highly active lyso-PtdSer and lysoPtdIns acyltransferase activities observed in cell free extracts of yeast, which are attributable to the Ale1p acyltransferase that we recently identified [5,30], neither lyso-PtdSer nor lyso-PtdIns were taken up by yeast, consistent with the results of the genetic experiments. The result obtained with lyso-PtdSer was somewhat surprising, given that NBD-PtdSer had previously been observed to be internalized via the Dnf1p and Dnf2p transporters [6]. However our data, coupled with more recent work on NBD-PtdSer uptake [34], is in conflict with the original conclusions by others regarding transport of the lipid. Recent experiments by Stevens and Nichols, reported an increase in the ATP dependent import of NBD-PtdSer in the dnf1Δ dnf2Δ mutant, which was not suppressed by the removal of either Drs2p of Dnf3p. The authors raise the possibility that the essential P4-ATPase, Neo1p, might be a candidate for the translocase that is upregulated and/or mislocalized in the plasma membrane of the dnf1Δ dnf2Δ mutant. The discrepancies between the uptake and metabolism of lyso-PtdSer and NBD-PtdSer remain unresolved at this time, due to these conflicting reports [4–6,34]. Perhaps it is an oversimplification to assume that the general correspondence in the transport properties between NBDPtdEtn and lyso-PtdEtn, and between NBD-PtdCho and lyso-PtdEtn, will also extend to other pairs of NBD-PtdX and lyso-PtdX molecules. 3.3. Alkylphosphocholine drug transport by P4-ATPases Over the past 25 years, a class of anti-neoplastic and anti-parasitic agents named the alkylphosphocholines has been studied as candidates for the treatment of cancer [35] and certain infections caused by

apicomplexan and trypanosomatid parasites, most notably for the treatment of visceral leishmaniasis [36]. The canonical member of this group is edelfosine (also referred to as ET-18-OCH3), and its structure is given in Fig. 1, along with other related members of this drug class. The obvious structural similarity to lysophospholipids, as well as the limited metabolism of the ether and methoxy moieties, immediately suggests that these compounds might act by disrupting membrane structure and/or inhibiting membrane lipid biosynthesis. Several studies have provided support for this idea [37–40], but the exact mechanism of action of these drugs remains poorly defined, and may differ in diverse cell types. Treatment of yeast with edelfosine alters the distribution of a plasma membrane protein, Pma1p, between detergent-resistant and detergent-soluble membrane domains [41]. A recent study shows similar actions in mammalian cells [42], and implicates edelfosine in the disruption of ER function, inducing an ER stress response, leading to apoptosis in solid tumors. The structure of edelfosine is quite similar to that of lyso-platelet activating factor (lyso-PAF), with the only difference being the 2-Omethyl moiety in the glycerol backbone of edelfosine, relative to the 2hydroxyl moiety of lyso-PAF (Fig. 1). Lyso-PAF can be taken up by yeast and converted to the corresponding 1-alkyl-2-acyl-PtdCho species, but in the absence of the Ale1p lysophospholipid acyltransferase [30], lyso-PAF accumulates in the cell and is as toxic to yeast as edelfosine [43]. This fact, coupled with the aforementioned structural similarity between lyso-PAF and edelfosine, suggests that these compounds use the same transport system to gain entry into the yeast cell. The accumulation of toxic levels of lyso-PAF and edelfosine in the cell also provides further evidence that the lysophospholipid transport system lacks feedback regulation, In an initial study aimed at identifying genes that mediate edelfosine uptake by yeast, Hanson et al. [44] isolated a series of mutants that are resistant to the effects of this drug, and then further characterized these strains for their ability to take up NBD-PtdCho. One of the edelfosine resistant, NBD-PtdCho transport-deficient mutants that was characterized contained a mutation in the LEM3 gene, which identified a previously unknown role for this Cdc50p homolog. Later studies, as described above, implicated Lem3p as an essential component of the transport complex, in combination with either Dnf1p or Dnf2p [14]. The study by Hanson et al. [44] also showed that the alkylphosphocholine drug miltefosine (hexadecylphosphocholine) is also toxic to yeast at roughly the same concentration as edelfosine, and that the lem3 mutation renders the strain resistant to this compound, as well. The finding that miltefosine is apparently also a substrate for Dnf1p or Dnf2p provides important information about the substrate specificity of these transporters. Coupled with work from our lab characterizing the uptake of lyso-phospholipids by Dnf1p and Dnf2p [4,5], the structural requirements for competence as a substrate are the presence of a phosphocholine or phosphoethanolamine headgroup, and a long-chain hydrophobic alkyl or acyl-moiety. Alkylphosphocholine and alkyl-lysophospholipid-like compounds have found utility as anti-parasitic drugs, especially for treatment of visceral leishmaniasis, caused primarily by Leishmania donovanii or closely related Leishmania spp. [36]. As with any drug, there is a powerful selective pressure on this pathogen to develop resistance to miltefosine. Miltefosine-resistant leishmaniasis is known and is anticipated to become a major problem with widespread use of this and related drugs [36,45,46]. The possible mechanisms of miltefosine resistance have been investigated by Pérez-Victoria et al. [7,47]. These authors analyzed a naturally occurring miltefosine and edelfosine resistant mutant of L. donovanii, denoted as strain M-40 R, and determined that the basis of resistance is altered uptake of the drug. Additional experiments showed that uptake of NBD-PtdSer, NBDPtdEtn, and NBD-PtdCho was also blocked in the M-40 R strain relative to wild-type parasites. The mutant gene in the M-40 R line was cloned by transformation of the resistant strain with a genomic DNA library generated from a

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

miltefosine-sensitive wild-type strain [8]. A single cosmid clone was identified by its ability to revert the M40-R strain to miltefosine sensitivity, and subsequent subcloning and sequencing of rescuing plasmids identified a gene, which was named LdMT. The protein encoded by LdMT shared significant sequence identity with members of the P4-ATPase family, and a GFP-tagged LdMT protein was observed associated with the plasma membrane. These similarities make the LdMT protein the functional homolog of the yeast Dnf1p and Dnf2p transporters described above. Additional screening of L. donovanii clones that had been mutagenized with ethyl-methane sulfonate identified six more independently isolated strains that were resistant to miltefosine. These mutant isolates were defective in the uptake of miltefosine, and all six reverted to miltefosine sensitivity upon transformation with the wild-type LdMT gene. Thus, all seven of the known miltefosine resistant strains were mutants in the LdMT gene, indicating that this gene will be under immense negative selective pressure in the wild-type pathogen during miltefosine treatment. The apparent non-essential nature of this gene further enhances its susceptibility to negative selection during drug treatment, leading to the development of drug resistance. The apparent equivalence of LdMT to the yeast Dnf1p and Dnf2p transporters prompted additional studies aimed at the identification of a putative accessory β-subunit for LdMT related to the Cdc50p family [9]. It was presumed that this protein would be required for the proper localization and/or activity of LdMT, as in the case of the yeast proteins. Extensive mutagenesis and miltefosine-resistance screening produced a mutant line designated M-1M that was strongly resistant to the drug, but not rescued by an LdMT containing plasmid. The expression of the GFP-LdMT protein in the M-1M mutant strain resulted in a mislocalized fluorescence signal, with LdMT-GFP failing to be exported from the Golgi. This result is very similar to the situation in yeast, in which the absence of the Lem3p subunit disrupts the proper trafficking of Dnf1p and Dnf2p [14]. The authors used a candidate-gene approach and identified the LdMT β-subunit by amplifying and expressing each of the three L. donovanii CDC50 orthologs in the M-1M mutant. Only one of these genes restored the miltefosine sensitivity of this resistant mutant, and was designated LdRos3. As with the yeast ortholog, a GFP-tagged construct of LdRos3 was localized to the plasma membrane, and the M-1M mutant was also deficient in the uptake of NBD-labeled aminoglycerolipids. Taken together, these studies elucidate a critical role for the P4-ATPase complex of Leishmania spp. in the uptake and potency of alkylphosphocholine type drugs, and provide critical insight into one important mechanism of resistance to this class of drugs. In addition to the activity of alkylphosphocholine drugs against Leishmania spp., this class of compounds has been studied with respect to their potential anti-cancer activities [35], however they have not found wide use due to negative side effects and limited efficacy (e.g. [48]). The anti-neoplastic activity of drugs such as edelfosine and perifosine, have, under certain circumstances, been demonstrated to require the specific uptake of these compounds via ATP dependent transport mechanisms [10,49]. Not surprisingly, as in the case of Leishmania parasites, mammalian tumor cells will readily develop resistance to alkylphosphocholine drugs. Examples of the development of resistance include the kidney cancer cell line KD [49] and the mouse macrophage RAW cell line [10]. Interestingly, in these cases of drug resistance, the uptake of NBD-phospholipids was also severely reduced. In the edelfosine resistant RAW.R1 cell line, uptake of radiolabeled lyso-PtdEtn and lyso-PtdCho was also essentially abolished. The data suggest that the uptake of the drug or lysophospholipid is dependent on a single transporter with characteristics of the previously characterized P4-ATPases, but none of the genes affected in these alkylphosphocholine resistant cell lines have been cloned. This presents a prime opportunity for the identification of one of the uncharacterized human or mouse P4-family members (Fig. 2) as the lysophospholipid transporter.

625

4. P4-ATPases in other model systems In addition to the yeast, mammalian, and parasite systems described above, P4-ATPases have been identified as determinants of developmental and cell biological processes in other model organisms. Putative aminophospholipid translocases and Lem3p/Cdc50p family members have been identified and characterized in the plant Arabidopsis thaliana [11,50] and the nematode worm Caenorhabditis elegans [51,52]. These P-type ATPases are included with those from other model organisms, such as the mouse and fruit fly, in the phylogenetic tree presented in Fig. 2. An early study on the P4-ATPase family in plants revealed that the ALA1 gene (for aminophospholipid ATPase 1) is required for normal chilling tolerance [11]. When expressed in the yeast drs2Δ mutant, the ALA1 protein was able to rescue the cold-sensitive phenotype of the strain, and mediate the transport of NBD-PtdSer in microsomal vesicles, thereby demonstrating the conserved functions between yeast and plant Drs2p homologs. Interestingly, ALA1 expressed in yeast exerted these functions in the absence of a Cdc50p-like βsubunit, which either indicates that the protein does not require a Cdc50 homolog as a partner, or that one of the endogenous yeast Cdc50p-like proteins is able to pair with the ALA1 protein to promote its proper function. Another ALA homolog from Arabidopsis, ALA3, was recently characterized and shown to function in the Golgi of root cells involved in secretory processes in plants, most notably in the columella and border-like cells of the root tip, which are involved in the vesicular secretion of polysaccharides and other compounds into the rhizosphere [50]. The expression of ALA3 in the yeast dnf1Δ dnf2Δ drs2Δ strain had no effect on the internalization of NBD-labeled phospholipids. To test whether the inactivity of ALA3 was due to the lack of a functional accessory β-subunit, the authors individually co-expressed the five homologs of Cdc50p encoded in the Arabidopsis genome, which were denoted ALA interacting subunit 1–5 (ALIS1–5). Of these five proteins, ALIS1, 3, and 5 were shown to interact physically with ALA3 in the yeast split-ubiquitin two-hybrid interaction system. The co-expression of ALA3 and ALIS3 in the dnf1Δ dnf2Δ drs2Δ yeast strain resulted in the presence of the ALA3–ALIS1 complex in the plasma membrane, and facilitated the internalization of NBD-PtdEtn, and to a lesser extent, NBD-PtdSer and NBD-PtdCho. This indicates that the ALA3–ALIS1 complex shares functions with both Drs2p– Cdc50p and Dnf1/2p-Lem3p complexes. The further functional characterization of the other Arabidopsis ALA and ALIS family members with regard to substrate specificity, subcellular localization, and tissue distribution will be interesting given the more diverse composition of plant membranes, which consist not only of the glycerophospholipids typical of yeast and mammalian cells, but also of glycolipids and sulfolipids, and betaine lipids [53,54]. 5. Perspectives and unresolved questions Recent findings have begun to expand our understanding of the biochemical and cellular functions of the aminophospholipid translocating P4-ATPases. In the yeast system, the realization that the P4ATPases must physically pair with an accessory β-subunit of the Cdc50p family raises new questions about the roles of these proteins in determining the substrate specificity and localization of the ATPase components. Specifically, for the Dnf1p/Dnf2p and Lem3p complex, the relevant regions of the molecules that determine substrate specificity have remained elusive. Powerful new tools for addressing these questions now exist with the development of yeast strains that require exogenous lyso-PtdEtn or lyso-PtdCho for growth, contingent on uptake via functional Dnf1p or Dnf2p coupled to Lem3p. These findings also lead to questions of whether mammalian and other metazoan systems, such as the worm and fruit fly, also take up and utilize lysophospholipids for membrane biogenesis. Remarkably,

626

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627

the uptake and direct acylation of lyso-PtdCho in rats has been known for over 40 years from work originating in the laboratory of Stein and Stein [55]. These early studies on organismal lysophospholipid metabolism revealed that the half-life for lyso-PtdCho and lysoPtdEtn injected intravenously in rats was between 5 and 11 min, and that these compounds were taken up and directly acylated to form the diacyl phospholipid, primarily by the liver and intestine. These results clearly indicate an active and underappreciated lipid transport process that is likely to have implications for phospholipid trafficking between organs. As in the case of alkylphosphocholine drug transport, the P4ATPases and accessory β-subunits involved in this process are unknown and their elucidation and characterization would add greatly to our understanding of whole-body phospholipid trafficking and metabolism in metazoan eukaryotes. References [1] D.L. Daleke, Phospholipid flippases, J. Biol. Chem. 282 (2007) 821–825. [2] T.R. Graham, Flippases and vesicle-mediated protein transport, Trends Cell Biol. 14 (2004) 670–677. [3] C.C. Paulusma, R.P. Oude Elferink, The type 4 subfamily of P-type ATPases, putative aminophospholipid translocases with a role in human disease, Biochim. Biophys. Acta 1741 (2005) 11–24. [4] W.R. Riekhof, D.R. Voelker, Uptake and utilization of lyso-phosphatidylethanolamine by Saccharomyces cerevisiae, J. Biol. Chem. 281 (2006) 36588–36596. [5] W.R. Riekhof, J. Wu, M.A. Gijon, S. Zarini, R.C. Murphy, D.R. Voelker, Lysophosphatidylcholine metabolism in Saccharomyces cerevisiae: the role of P-type ATPases in transport and a broad specificity acyltransferase in acylation, J. Biol. Chem. 282 (2007) 36853–36861. [6] T. Pomorski, R. Lombardi, H. Riezman, P.F. Devaux, G. van Meer, J.C. Holthuis, Drs2p-related P-type ATPases Dnf1p and Dnf2p are required for phospholipid translocation across the yeast plasma membrane and serve a role in endocytosis, Mol. Biol. Cell 14 (2003) 1240–1254. [7] F.J. Perez-Victoria, S. Castanys, F. Gamarro, Leishmania donovani resistance to miltefosine involves a defective inward translocation of the drug, Antimicrob. Agents Chemother. 47 (2003) 2397–2403. [8] F.J. Perez-Victoria, F. Gamarro, M. Ouellette, S. Castanys, Functional cloning of the miltefosine transporter. A novel P-type phospholipid translocase from Leishmania involved in drug resistance, J. Biol. Chem. 278 (2003) 49965–49971. [9] F.J. Perez-Victoria, M.P. Sanchez-Canete, S. Castanys, F. Gamarro, Phospholipid translocation and miltefosine potency require both L. donovani miltefosine transporter and the new protein LdRos3 in Leishmania parasites, J. Biol. Chem. 281 (2006) 23766–23775. [10] R.A. Zoeller, M.D. Layne, E.J. Modest, Animal cell mutants unable to take up biologically active glycerophospholipids, J. Lipid Res. 36 (1995) 1866–1875. [11] E. Gomes, M.K. Jakobsen, K.B. Axelsen, M. Geisler, M.G. Palmgren, Chilling tolerance in Arabidopsis involves ALA1, a member of a new family of putative aminophospholipid translocases, Plant Cell 12 (2000) 2441–2454. [12] A. Zachowski, J.P. Henry, P.F. Devaux, Control of transmembrane lipid asymmetry in chromaffin granules by an ATP-dependent protein, Nature 340 (1989) 75–76. [13] C.Y. Chen, M.F. Ingram, P.H. Rosal, T.R. Graham, Role for Drs2p, a P-type ATPase and potential aminophospholipid translocase, in yeast late Golgi function, J. Cell Biol. 147 (1999) 1223–1236. [14] K. Saito, K. Fujimura-Kamada, N. Furuta, U. Kato, M. Umeda, K. Tanaka, Cdc50p, a protein required for polarized growth, associates with the Drs2p P-type ATPase implicated in phospholipid translocation in Saccharomyces cerevisiae, Mol. Biol. Cell 15 (2004) 3418–3432. [15] P. Natarajan, J. Wang, Z. Hua, T.R. Graham, Drs2p-coupled aminophospholipid translocase activity in yeast Golgi membranes and relationship to in vivo function, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 10614–10619. [16] A.M. Grant, P.K. Hanson, L. Malone, J.W. Nichols, NBD-labeled phosphatidylcholine and phosphatidylethanolamine are internalized by transbilayer transport across the yeast plasma membrane, Traffic 2 (2001) 37–50. [17] L.S. Kean, R.S. Fuller, J.W. Nichols, Retrograde lipid traffic in yeast: identification of two distinct pathways for internalization of fluorescent-labeled phosphatidylcholine from the plasma membrane, J. Cell Biol. 123 (1993) 1403–1419. [18] N. Alder-Baerens, Q. Lisman, L. Luong, T. Pomorski, J.C. Holthuis, Loss of P4 ATPases Drs2p and Dnf3p disrupts aminophospholipid transport and asymmetry in yeast post-Golgi secretory vesicles, Mol. Biol. Cell 17 (2006) 1632–1642. [19] S. Chen, J. Wang, B.P. Muthusamy, K. Liu, S. Zare, R.J. Andersen, T.R. Graham, Roles for the Drs2p–Cdc50p complex in protein transport and phosphatidylserine asymmetry of the yeast plasma membrane, Traffic 7 (2006) 1503–1517. [20] X. Tang, M.S. Halleck, R.A. Schlegel, P. Williamson, A subfamily of P-type ATPases with aminophospholipid transporting activity, Science 272 (1996) 1495–1497. [21] K. Wakamatsu, S.Y. Choung, T. Kobayashi, K. Inoue, T. Higashijima, T. Miyazawa, Complex formation of peptide antibiotic Ro09-0198 with lysophosphatidylethanolamine: 1H NMR analyses in dimethyl sulfoxide solution, Biochemistry 29 (1990) 113–118. [22] S.Y. Choung, T. Kobayashi, J. Inoue, K. Takemoto, H. Ishitsuka, K. Inoue, Hemolytic activity of a cyclic peptide Ro09-0198 isolated from Streptoverticillium, Biochim. Biophys. Acta 940 (1988) 171–179.

[23] S.Y. Choung, T. Kobayashi, K. Takemoto, H. Ishitsuka, K. Inoue, Interaction of a cyclic peptide, Ro09-0198, with phosphatidylethanolamine in liposomal membranes, Biochim. Biophys. Acta 940 (1988) 180–187. [24] G. Machaidze, A. Ziegler, J. Seelig, Specific binding of Ro 09-0198 (cinnamycin) to phosphatidylethanolamine: a thermodynamic analysis, Biochemistry 41 (2002) 1965–1971. [25] U. Kato, K. Emoto, C. Fredriksson, H. Nakamura, A. Ohta, T. Kobayashi, K. Murakami-Murofushi, T. Kobayashi, M. Umeda, A novel membrane protein, Ros3p, is required for phospholipid translocation across the plasma membrane in Saccharomyces cerevisiae, J. Biol. Chem. 277 (2002) 37855–37862. [26] K. Iwamoto, S. Kobayashi, R. Fukuda, M. Umeda, T. Kobayashi, A. Ohta, Local exposure of phosphatidylethanolamine on the yeast plasma membrane is implicated in cell polarity, Genes Cells 9 (2004) 891–903. [27] L.R. Poulsen, R.L. Lopez-Marques, M.G. Palmgren, Flippases: still more questions than answers, Cell. Mol. Life Sci. 65 (2008) 3119–3125. [28] N. Furuta, K. Fujimura-Kamada, K. Saito, T. Yamamoto, K. Tanaka, Endocytic recycling in yeast is regulated by putative phospholipid translocases and the Ypt31p/32p-Rcy1p pathway, Mol. Biol. Cell 18 (2007) 295–312. [29] M.K. Storey, K.L. Clay, T. Kutateladze, R.C. Murphy, M. Overduin, D.R. Voelker, Phosphatidylethanolamine has an essential role in Saccharomyces cerevisiae that is independent of its ability to form hexagonal phase structures, J. Biol. Chem. 276 (2001) 48539–48548. [30] W.R. Riekhof, J. Wu, J.L. Jones, D.R. Voelker, Identification and characterization of the major lysophosphatidylethanolamine acyltransferase in Saccharomyces cerevisiae, J. Biol. Chem. 282 (2007) 28344–28352. [31] K. Liu, K. Surendhran, S.F. Nothwehr, T.R. Graham, P4-ATPase requirement for AP1/clathrin function in protein transport from the trans-Golgi network and early endosomes, Mol. Biol. Cell 19 (2008) 3526–3535. [32] K. Saito, K. Fujimura-Kamada, H. Hanamatsu, U. Kato, M. Umeda, K.G. Kozminski, K. Tanaka, Transbilayer phospholipid flipping regulates Cdc42p signaling during polarized cell growth via Rga GTPase-activating proteins, Dev. Cell 13 (2007) 743–751. [33] K. Liu, Z. Hua, J.A. Nepute, T.R. Graham, Yeast P4-ATPases Drs2p and Dnf1p are essential cargos of the NPFXD/Sla1p endocytic pathway, Mol. Biol. Cell 18 (2007) 487–500. [34] H.C. Stevens, L. Malone, J.W. Nichols, The putative aminophospholipid translocases, DNF1 and DNF2, are not required for 7-nitrobenz-2-oxa-1,3-diazol-4-ylphosphatidylserine flip across the plasma membrane of Saccharomyces cerevisiae, J. Biol. Chem. 283 (2008) 35060–35069. [35] W.J. van Blitterswijk, M. Verheij, Anticancer alkylphospholipids: mechanisms of action, cellular sensitivity and resistance, and clinical prospects, Curr. Pharm. Des. 14 (2008) 2061–2074. [36] J. Mishra, A. Saxena, S. Singh, Chemotherapy of leishmaniasis: past, present and future, Curr. Med. Chem. 14 (2007) 1153–1169. [37] I. Baburina, S. Jackowski, Apoptosis triggered by 1-O-octadecyl-2-O-methyl-racglycero-3-phosphocholine is prevented by increased expression of CTP:phosphocholine cytidylyltransferase, J. Biol. Chem. 273 (1998) 2169–2173. [38] K. Boggs, C.O. Rock, S. Jackowski, The antiproliferative effect of hexadecylphosphocholine toward HL60 cells is prevented by exogenous lysophosphatidylcholine, Biochim. Biophys. Acta 1389 (1998) 1–12. [39] K.P. Boggs, C.O. Rock, S. Jackowski, Lysophosphatidylcholine and 1-O-octadecyl-2O-methyl-rac-glycero-3-phosphocholine inhibit the CDP-choline pathway of phosphatidylcholine synthesis at the CTP:phosphocholine cytidylyltransferase step, J. Biol. Chem. 270 (1995) 7757–7764. [40] W. Yang, K.P. Boggs, S. Jackowski, The association of lipid activators with the amphipathic helical domain of CTP:phosphocholine cytidylyltransferase accelerates catalysis by increasing the affinity of the enzyme for CTP, J. Biol. Chem. 270 (1995) 23951–23957. [41] V. Zaremberg, C. Gajate, L.M. Cacharro, F. Mollinedo, C.R. McMaster, Cytotoxicity of an anti-cancer lysophospholipid through selective modification of lipid raft composition, J. Biol. Chem. 280 (2005) 38047–38058. [42] T. Nieto-Miguel, R.I. Fonteriz, L. Vay, C. Gajate, S. Lopez-Hernandez, F. Mollinedo, Endoplasmic reticulum stress in the proapoptotic action of edelfosine in solid tumor cells, Cancer Res. 67 (2007) 10368–10378. [43] Q. Chen, M. Kazachkov, Z. Zheng, J. Zou, The yeast acylglycerol acyltransferase LCA1 is a key component of Lands cycle for phosphatidylcholine turnover, FEBS Lett. 581 (2007) 5511–5516. [44] P.K. Hanson, L. Malone, J.L. Birchmore, J.W. Nichols, Lem3p is essential for the uptake and potency of alkylphosphocholine drugs, edelfosine and miltefosine, J. Biol. Chem. 278 (2003) 36041–36050. [45] F.J. Perez-Victoria, M.P. Sanchez-Canete, K. Seifert, S.L. Croft, S. Sundar, S. Castanys, F. Gamarro, Mechanisms of experimental resistance of Leishmania to miltefosine: implications for clinical use, Drug Resist. Updat. 9 (2006) 26–39. [46] K. Seifert, S. Matu, F. Javier Perez-Victoria, S. Castanys, F. Gamarro, S.L. Croft, Characterisation of Leishmania donovani promastigotes resistant to hexadecylphosphocholine (miltefosine), Int. J. Antimicrob. Agents 22 (2003) 380–387. [47] J.M. Perez-Victoria, F.J. Perez-Victoria, A. Parodi-Talice, I.A. Jimenez, A.G. Ravelo, S. Castanys, F. Gamarro, Alkyl-lysophospholipid resistance in multidrugresistant Leishmania tropica and chemosensitization by a novel P-glycoprotein-like transporter modulator, Antimicrob. Agents Chemother. 45 (2001) 2468–2474. [48] N.B. Leighl, S. Dent, M. Clemons, T.A. Vandenberg, R. Tozer, D.G. Warr, R.M. Crump, D. Hedley, G.R. Pond, J.E. Dancey, M.J. Moore, A Phase 2 study of perifosine in advanced or metastatic breast cancer, Breast Cancer Res. Treat. 108 (2008) 87–92. [49] F. Munoz-Martinez, C. Torres, S. Castanys, F. Gamarro, The anti-tumor alkylphospholipid perifosine is internalized by an ATP-dependent translocase activity

W.R. Riekhof, D.R. Voelker / Biochimica et Biophysica Acta 1791 (2009) 620–627 across the plasma membrane of human KB carcinoma cells, Biochim. Biophys. Acta 1778 (2008) 530–540. [50] L.R. Poulsen, R.L. Lopez-Marques, S.C. McDowell, J. Okkeri, D. Licht, A. Schulz, T. Pomorski, J.F. Harper, M.G. Palmgren, The Arabidopsis P4-ATPase ALA3 localizes to the golgi and requires a beta-subunit to function in lipid translocation and secretory vesicle formation, Plant Cell 20 (2008) 658–676. [51] N.N. Lyssenko, Y. Miteva, S. Gilroy, W. Hanna-Rose, R.A. Schlegel, An unexpectedly high degree of specialization and a widespread involvement in sterol metabolism among the C. elegans putative aminophospholipid translocases, BMC Dev. Biol. 8 (2008) 96.

627

[52] A.F. Ruaud, L. Nilsson, F. Richard, M.K. Larsen, J.L. Bessereau, S. Tuck, The C. elegans P4-ATPase TAT-1 regulates lysosome biogenesis and endocytosis, Traffic (2008). [53] C. Benning, C. Xu, K. Awai, Non-vesicular and vesicular lipid trafficking involving plastids, Curr. Opin. Plant Biol. 9 (2006) 241–247. [54] W.R. Riekhof, B.B. Sears, C. Benning, Annotation of genes involved in glycerolipid biosynthesis in Chlamydomonas reinhardtii: discovery of the betaine lipid synthase BTA1Cr, Eukaryotic Cell 4 (2005) 242–252. [55] Y. Stein, O. Stein, Metabolism of labeled lysolecithin, lysophosphatidyl ethanolamine and lecithin in the rat, Biochim. Biophys. Acta 116 (1966) 95–107.