Diverse relations between ABC transporters and lipids: An overview

Diverse relations between ABC transporters and lipids: An overview

BBAMEM-82320; No. of pages: 14; 4C: 6, 7, 9 Biochimica et Biophysica Acta xxx (2016) xxx–xxx Contents lists available at ScienceDirect Biochimica et...

2MB Sizes 0 Downloads 203 Views

BBAMEM-82320; No. of pages: 14; 4C: 6, 7, 9 Biochimica et Biophysica Acta xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamem

Diverse relations between ABC transporters and lipids: An overview Jennifer Neumann, Dania Rose-Sperling, Ute A. Hellmich ⁎ Institute of Pharmacy and Biochemistry, Johannes Gutenberg University Mainz, Johann-Joachim-Becher-Weg 30, 55128 Mainz, Germany

a r t i c l e

i n f o

Article history: Received 2 August 2016 Received in revised form 24 September 2016 Accepted 26 September 2016 Available online xxxx Keywords: ATP binding cassette protein-lipid interactions lipid bilayer lipid flopping drug-lipid interactions

a b s t r a c t It was first discovered in 1992 that P-glycoprotein (Pgp, ABCB1), an ATP binding cassette (ABC) transporter, can transport phospholipids such as phosphatidylcholine, −ethanolamine and -serine as well as glucosylceramide and glycosphingolipids. Subsequently, many other ABC transporters were identified to act as lipid transporters. For substrate transport by ABC transporters, typically a classic, alternating access model with an ATPdependent conformational switch between a high and a low affinity substrate binding site is evoked. Transport of small hydrophilic substrates can easily be imagined this way, as the molecule can in principle enter and exit the transporter in the same orientation. Lipids on the other hand need to undergo a 180° degree turn as they translocate from one membrane leaflet to the other. Lipids and lipidated molecules are highly diverse, so there may be various ways how to achieve their flipping and flopping. Nonetheless, an increase in biophysical, biochemical and structural data is beginning to shed some light on specific aspects of lipid transport by ABC transporters. In addition, there is now abundant evidence that lipids affect ABC transporter conformation, dynamics as well as transport and ATPase activity in general. In this review, we will discuss different ways in which lipids and ABC transporters interact and how lipid translocation may be achieved with a focus on the techniques used to investigate these processes. © 2016 Published by Elsevier B.V.

Contents 1. 2. 3.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . ABC transporters . . . . . . . . . . . . . . . . . . . . . . . . Human ABC transporters in lipid transport . . . . . . . . . . . . 3.1. Transport of phospholipids and sphingolipids . . . . . . . 3.2. Transport of sterols and bile salts by ABC transporters . . . 3.3. Transport of fatty acids . . . . . . . . . . . . . . . . . 3.4. Transport of specialty lipids . . . . . . . . . . . . . . . 4. Lipids influence ABC transporter function . . . . . . . . . . . . 5. Are lipids integral parts of ABC transporter structures? . . . . . . 6. Substrate promiscuity in ABC lipid transporters . . . . . . . . . . 7. Lipids as reservoirs for ABC transporter substrates . . . . . . . . . 8. Lipids and substrates affect structure and dynamics of ABC exporters 9. Different models for lipid translocation by ABC exporters . . . . . Conclusion and outlook . . . . . . . . . . . . . . . . . . . . . . . Transparency document . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1. Introduction ⁎ *Corresponding author at: Johannes Gutenberg University Mainz, Department of Pharmacy and Biochemistry, Johann-Joachim-Becher-Weg 30, 55128, Mainz, Germany. E-mail address: [email protected] (U.A. Hellmich).

Lipids make up about 5–10% of the dry mass of a cell [1], and mammalian cells dedicate about 5% of their genes to the synthesis of lipids [2]. Lipids are chemically much more diverse than implied by

http://dx.doi.org/10.1016/j.bbamem.2016.09.023 0005-2736/© 2016 Published by Elsevier B.V.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

2

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

the simple structural classification into hydrophobic tail and hydrophilic head groups. Significant methodological advances, mainly in mass spectrometry, have given rise to the field of “lipidomics”. A 2005 lipidomics survey classified eight basic types of lipids (glycerophospholipids, glycerolipids, sphingolipids, saccharolipids, fatty acyls, sterol lipids, prenol lipids and polyketides), which are present in cellular membranes [3]. These main categories can then be further subdivided into classes and subclasses. For the most abundant lipid category in mammalian cells, glycerophospholipids, at least 20 additional subcategories exist. A single cell, with its ability to create diverse lipid chains and head groups, may therefore well contain more than a thousand chemically different lipids [2] and in entire tissues or organisms this number may rise to tens of thousands [4]. However, complexity does not end there. Lipids are not homogenously distributed across a cell membrane, but compartmentalized, i.e. to different organelles [5]. In addition, the lipid distribution between the two lipid bilayer leaflets is asymmetric. At the plasma membrane, phosphatidylethanolamine, phosphatidylserine or phosphatidic acid are mostly located in the inner (cytosolic) membrane leaflet, while phosphatidylcholine, sphingomyelin or glycolipids have a preference for the outer (exo-cytoplasmic) leaflet [2]. In some cases, these preferences stem from the physical properties of the lipids themselves, i.e. their chain length and degree of saturation, as well as their head group properties that influence membrane curvature and fluidity. Furthermore, proteins are responsible for heterogeneity in lipid distribution (e.g. [5]). Although passive diffusion of lipids from one bilayer leaflet to the other is very slow (~10−15 cm2/s) compared to their lateral movement (~10−8 cm2/s), active transport is required to counteract diffusionbased lipid movement and the resulting homogenization of leaflet content. Several types of membrane proteins mediate lipid transport: (i) Secondary active transporters belonging to the MOP (multidrug, oligosaccharidyl-lipid, polysaccharide) superfamily require a proton or sodium gradient [6]. (ii) Primary active transporters such as ABC (ATP binding cassette) transporters usually “flop” lipids from the inner membrane leaflet to the outer leaflet, energized by ATP hydrolysis [7]. The reverse “flipping” process from the outer to the inner leaflet is carried out by P4-type ATPases [8,9]. Lipid transport in this direction is also carried out by ABCA4, which is the only eukaryotic ABC transporter described to date to function as a flippase and thereby as an importer (see below, [10]). Finally, (iii) scramblases are not direction-specific and act energy independently [11,12]. In addition to being substrates of transporters, lipids can also serve as scaffolds for membrane proteins, and they can be embedded permanently in a protein's structure. The lipid environment can also influence protein activity, i.e. ATPase or transport activity in the case of ABC transporters. Finally, e.g. for multidrug (ABC) transporters with a strong preference for amphipathic and hydrophobic molecules, lipids can serve as a reservoir for these non-lipid hydrophobic substrates due to their enrichment in the membrane bilayer. We will give an overview over human ABC lipid transporters first, followed by a discussion of a few specific questions how lipids influence ABC transporter functions, how ABC transporters may translocate lipids and how this has been studied with different biophysical techniques. Due to the available data, the latter section will also strongly feature bacterial transporters. 2. ABC transporters ABC transporters are ubiquitous in all three phyla of life. They share the same core structure consisting of two transmembrane regions (TMDs) and two soluble nucleotide binding domains (NBDs). The TMDs bind and translocate substrates across lipid bilayers while the NBDs bind and hydrolyze ATP. ABC transporters can function as exporters (i.e. removing substrates from the cytoplasm) or importers (i.e. translocating substrates into the cytoplasm) [13]. Mammalian ABC

proteins belong to seven subfamilies, ABCA-G: Members of the families A-D and G code for membrane transporters (Fig. 1), while the members of the ABCE and ABCF families are soluble proteins consisting only of NBDs, which are involved in, e.g. ribosome recycling and transcription regulation [14,15]. The NBDs of all ABC proteins are highly conserved both on the level of sequence and 3D structure. They consist of a RecA-like domain including the Walker A (P-loop) and B motifs and a helical domain including the ABC signature motif (C-loop) signifying their ABC protein family affiliation. Eukaryotic transporters of the ABCA and C families are expressed with all four domains on a single polypeptide chain (so-called full transporters), while ABCD and G members (as well as the majority of prokaryotic exporters) are “half-transporters” where one TMD is fused to one NBD. A unique feature of the ABCG subfamily is their domain topology with an N-terminally leading NBD followed by the TMD. In the ABCB family, both full and half transporters can be found (Fig. 1). Two half-transporters will homo- or hetero-dimerize to form a full transporter [16]. In bacteria, many other ABC transporter domain organizations, such as all four domains on separate peptides, exist.

3. Human ABC transporters in lipid transport Twenty out of the 48 human ABC transporter proteins have been implicated in the transport of lipids or lipid-like molecules, such as steroids (including cholesterol and bile acids), phospholipids and sphingolipids (for a detailed description please refer to [17]). These “lipid translocators” belong to all ABC transporter subfamilies (A, B, C, D, G), thus no subfamily specific trait seems to be responsible for lipid recognition and transport. Mutations in these proteins lead to disruptions in lipid metabolism and distribution. Consequences are lipid-associated diseases such as Sitosterolemia [18], Stargardt disease [19] or Tangier disease [20], thus underlining the importance of well-organized cellular lipid translocation.

3.1. Transport of phospholipids and sphingolipids ABCA1, ABCA3, ABCA4, ABCA7, ABCB1 (Pgp), ABCB4 and ABCC1 can translocate phospholipids across the plasma membrane (Fig. 1) (e.g. [17,21]). ABCA1 and ABCA7 are also important for the formation of high density lipoproteins (HDL) [22,23] required for cholesterol transport to the liver. While the physiological significance of phospholipid and glycosphingolipid transport by ABCB1 remains unclear [24], one putative physiological function may be the secretion of PAF (platelet-activating factor), a phospholipid involved in inflammation and allergic responses [25]. ABCB4 is responsible for phosphatidylcholine extrusion from the liver into the bile [26,27]. ABCC1 was initially identified as a glutathione-conjugate transporter [28], but lipid transport has also been demonstrated (e.g. [29]). ABCA3 expression peaks before birth and coincides with the expression of surfactant proteins. Its presence is restricted to the lung alveolar type II pneumocytes. These are responsible for the storage and excretion of a phospholipid/cholesterol/protein mixture that acts as the lung surfactant. For this, the lipid/protein mixture is stored in so-called lamellar bodies. ABCA3 is responsible for the uptake of lipids into these specialized organelles [30,31]. Mutations in ABCA12, expressed in epidermal keratinocytes, lead to harlequin-type ichthyosis [32], caused by dysfunctional lipid translocation into lamellar granules, which act as stores for lipids such as ceramides and phospholipids [33–35]. These secretory organelles are released to form a water-impermeable, protective skin barrier. In harlequin-type ichthyosis patients, the dermis cannot properly form (before birth) and at birth newborns display a harlequin-pattern of dried, cracked skin [36]. Patients are extremely vulnerable to pathogens and many newborns die shortly after birth.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

3

Fig. 1. Human ABC transporters involved in lipid transport. The topology for the different ABC families as well as their substrates and cellular distribution are shown.

3.2. Transport of sterols and bile salts by ABC transporters ABCB11 (bile salt export pump, BSEP) and members of the ABCC family (C1-C4) translocate conjugated or unconjugated bile salts [37–39]. ABCA2, ABCG1, ABCG4, macrophage-expressed ABCA1 and the heterodimeric ABCG5/G8 (also referred to as Sterolin-1 and Sterolin-2), translocate sterols across the endosomal membrane (A2, G1, G4) [40–43] or the plasma membrane (A1, G5/G8) [44,45]. While the ABCG family member ABCG2 (BCRP, breast cancer resistance protein or MXR, mitoxantrone resistance protein) is more commonly associated with drug extrusion [46], a function for it in sterol transport has also been demonstrated and its ATPase activity can be stimulated with cholesterol [47,48]. In addition, ABCB1 has been implicated in sterol translocation: Cholesterol interacts with and influences ABCB1 conformation as indicated by the altered interaction with an antibody [49] and fluorescently labeled sterols [50]. However, it should be kept in mind here, and also for other data obtained with fluorescently labelled lipids, that it can be difficult to extract the impact of the fluorophore on binding and transport. Cholesteryl hemisuccinate (CHS), an acidic cholesterol ester that can self-assemble into bilayers at neutral pH [51], was able to stabilize ABCB1 in thermal unfolding experiments [50] and to increase its ATPase rate after reconstitution into liposomes [52]. Recently, the 3.9 Å X-ray structure of the human ABCG5/G8 heterodimer has been determined ( [53], PDB ID: 5DO7). It represents the first structure of a member of the ABCG family and only the second structure with reasonable resolution of a human ABC membrane transporter besides the mitochondrial transporter ABCB10 [54]. ABCG5/G8 mediates the excretion of neutral sterols in liver and intestines [18]. Mutations can lead to premature coronary atherosclerosis as ABCG5/G8 limits the uptake of dietary sterol in the intestine by promoting its secretion into the bile in healthy humans [18,55]. In order to purify the ABCG5/G8 complex for structure determination, cholate (a bile acid and cholesterol derivate) and CHS were supplemented during two initial affinity chromatography steps, purification tag removal and subsequent size exclusion chromatography purification [53]. After a reductive methylation step (to methylate surface lysine amino groups and to render the protein more easily crystallizable due to changes in hydropathy, solubility and ultimately crystal packing properties [56]), ABCG5/G8 dimers were re-lipidated by adding DOPC/DOPE

(dioleoylphosphatidylcholine/−ethanolamine) mixtures. All this was done while the protein was still in the presence of CHS and cholate. After yet another desalting step, samples were concentrated and then reconstituted into DMPC (dimyristoylphosphatidylcholine) bicelles that had to be supplemented with 5 mol% cholesterol to obtain well diffracting crystals. It is remarkable that a human membrane protein can survive such a long and arduous purification protocol. This may very likely be a testament to the extreme importance of lipids as stabilizers for membrane proteins. Although crystallized in the presence of 10 mM ATP, the ABCG5/G8 structure is in the inward-facing, nucleotide-free conformation. It is unclear in which step of the transport cycle the ABCG5/G8 transporter has been captured. It could either be the true apo state (without nucleotide and substrate) or rather a substrate bound form since electron density at symmetrical “vestibules” on opposing faces of the ABCG5/G8 heterodimer, which open to the bilayer, was observed [53]. The authors hypothesized that these vestibules may represent binding surfaces or entry paths for sterols to access the core of the heterodimer interface and that the observed density might be cholesterol. This would indicate that both ABCG5 and ABCG8 have cholesterol binding sites, compatible with the notion that the other sterol-transporting members of the ABCG family are homodimers. However, it is unclear if ABCG5 and ABCG8 have equivalent substrate binding sites. With the structure as a valuable blueprint, many new and exciting avenues of research now await being explored. Many cholesterol binding proteins contain characteristic amino acid sequences, such as the CRAC (cholesterol recognition/interaction amino acid consensus) and the CARC (inverted CRAC) motifs [57]. The CRAC motif consensus sequence is L/V-X(1–5)-Y-X(1–5)-R/K, the CARC motif follows the reverse order R/K-X(1–5)-Y/F-X(1–5)-L/V (with X = any amino acid). Such motifs have been found in many ABC transporters, including ABCG1, G2 and G5/G8 as well as B1 [50,58,59]. Because of their rather loose definition, experimental data is always required to verify sterol binding postulated based on the presence of a CRAC motif. ABCG1 is involved in the delivery of cholesterol and sphingolipids from cells to HDL [42]. It contains more than a dozen CRAC and CARC motifs located in the extracellular loops between TMH5 and TMH6, in the intracellular loop between TMH2 and TMH3 as well as in the N-

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

4

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

terminus [58]. The corresponding large extracellular loops between TMH5 and 6 are resolved in the ABCG5/G8 structure and fold into a distinct secondary structure [53]. Interestingly, residues mutated in Sitosterolemia [18], a lipid metabolic disorder caused by the decreased biliary excretion of nutrient (plant) sterols, are in contact with this extracellular region in ABCG5/G8. This supports the idea that this region is important for sterol recognition and/or translocation. In ABCG1, mutagenesis of the CRAC motif in TMH6 led to reduced cholesterol efflux from cells as determined by measuring accumulation of radiolabeled substrate in the medium [58]. Proteasomal degradation of ABCG1 can be inhibited by the addition of cholesterol to the medium, but this effect is lost when the TMH6 CRAC motif is mutated [60]. TMH6 of ABCG1 was also shown to be important for drug binding and translocation [58]. If the ABCG1 homodimer follows the architecture of the ABCG5/G8 heterodimer, TMH6 will face the lipid bilayer, while TMH2 and 5 line the transporter pore. In the case of ABCG drug transporters, it could be possible that recognition of substrates is carried out through TMH6, potentially even at the same site identified as a putative cholesterol binding site in ABCG5/G8. The different putative drug/lipid binding site localization in ABCG compared to ABCB could hint at a very different substrate recognition (and possibly tra nslocation) mechanism for these protein families. In Pgp (and bacterial MDR transporters) it has been proposed that substrates are bound directly from the inner leaflet of the membrane and that TMH6 (and its corresponding TMH12 in Pgp) are important for drug recognition and translocation [61]. These helices are located on the inside of the transporter cavity. Another member of the ABCG subfamily, ABCG2, behaves more like ABCB1 when considering its broad substrate specificity and the two share partial substrate overlap [62]. ABCG2 is mostly located in tissues with barrier functions such as the intestine, skin, placenta, liver, kidney, or blood brain barrier, where it likely confers protective functions and is overexpressed in certain types of cancer [63]. Among its substrates are usually hydrophobic or negatively charged molecules with a flat architecture [62,64]. ABCG2 contains a number of CRAC motifs in addition to a putative steroid-binding element (SBE) LxxL in TMH5 which could explain sterol recognition by ABCG2 [65,66]. For ABCG2 expressed in cholesterol-poor insect Sf9 cell membranes, significantly reduced substrate stimulated ATPase activity was observed. At the same time, cholesterol enrichment in insect cell membranes (by addition of cholesterol-loaded cyclodextrin) resulted in a reduced basal ATPase activity but an increased substrate stimulated ATPase activity [48] while cholesterol depletion in mammalian cell membranes inhibited the ABCG2 transport activity reversibly [67]. Functional reconstitution also required cholesterol [68]. Other physiological ABCG2 substrates include heme, riboflavin, sulfated or glucuronidated conjugates of sex hormones like estrone-3-sulfate, free estradiol, bile acids or xenobiotics such as methotrexate (e.g. [47,69]). Due to the very different folds of their ligand recognizing transmembrane domain, it will be fascinating to see if there are any functional commonalities between ABCB1 and ABCG2. The structural template of ABCG5/G8 may additionally help to expand our knowledge about the diverse transport functions of ABCG2, and how point mutations, such as the infamous variants of R482 lead to the observed differences in drug recognition and translocation [69]. 3.3. Transport of fatty acids ABCD1, D2 and D3 are responsible for the import of fatty acyl-CoA (coenzyme A) into peroxisomes and are thus an integral part of the cellular lipid metabolism [70,71]. Mutations in ABCD1 (also called adrenoleukodystrophy protein, ALDP) result in the accumulation of unbranched, saturated very long chain fatty acids (VLC-FAs, i.e. with an aliphatic chain of 22 or more carbon residues) in the cytoplasm. VLC-FA cannot be metabolized in mitochondria and thus need to be subjected to β-oxidation in peroxisomes. The VLC-FA accumulation is therefore

particularly severe in the nervous tissue and leads to neurodegeneration [72]. ABCD2 (adrenoleukodystrophy related protein, ALDR) prefers unsaturated, unbranched VLC-FAs [73,74]. ABCD3 (peroxisomal membrane protein of 70 kDa, PMP70) translocates branched, unsaturated, long-chain decarboxylic acids [71]. It has been a matter of debate whether ABCD transporters translocate free fatty acids or the CoAlinked molecules. Recently, it was demonstrated that comatose (an Arabidopsis thaliana ABCD transporter) has an intrinsic thioesterase activity [75]. If the same holds true for other (mammalian) ABCD transporters, it would present an elegant solution to the problem of translocating amphiphilic molecules: the fatty acyl-CoA molecule is cleaved, the hydrophilic CoA moiety is transported by the ABCD transporter and the fatty acid group can potentially cross the lipid bilayer on its own. Once inside the peroxisomal lumen, the two molecules are re-esterified by acyl-CoA synthetases (for a recent review, see [76]). However, this modus operando leaves a few details open for discussion: Where does substrate selectivity in particular regarding acyl chain length come from? What are the time scales and/or stoichiometry of CoA versus fatty acid translocation across the bilayer? Is it conceivable that both molecules cross the bilayer separately, but via a mechanism mediated by the ABCD transporter, e.g. the hydrophilic moiety passes through the transporter core while the hydrophobic part slides alongside it and thereby stays in contact with the bilayer? This mechanism could be a crossover between the credit card sliding model proposed for lipid scramblases [8,77,78] and the lipid-linked oligosaccharide (LLO) ABC transporter PglK's whipping mechanism ( [79], see below, Fig. 4B). Alternatively, the intriguing example of the ABCD proteins may show that an ABC transporter can be intricately linked to lipid translocation, without having to be a bona fide lipid transporter itself. 3.4. Transport of specialty lipids ABCA4 (also called photoreceptor rim protein, ABCR) is currently the only described ABC importer in mammals [10]. Mutations in ABCA4 lead to Stargardt disease, which results in vision loss due to lipofuscin accumulation in the retina [19,80]. 11-cis and all-trans retinal stimulate the ATPase activity of ABCA4 [81,82] and ABCA4 knockout mice accumulate all-trans retinal, N-retinylidene-PE (NRPE) and PE in rod outer segments [83]. In a regular photocycle, photons stimulate the switch from 11-cis retinal bound to the G-protein coupled receptor (GPCR) rhodopsin via a Schiff base to all-trans retinal which activates the protein. In order to deactivate rhodopsin, its Schiff base is hydrolyzed and the alltrans retinal removed, leaving an empty GPCR called opsin. Opsin can bind a new molecule of 11-cis retinal, thus becoming rhodopsin again and ready to enter a new photocycle. When all-trans retinoids are released from rhodopsin in a mechanism called photobleaching, they can react with PE to form NRPE. ABCA4 flips NRPE to the outer leaflet of the disc membrane where it is hydrolyzed to release all-trans retinal. After a few chemical modifications and translocation steps, the retinal is ultimately returned to opsin [84]. 4. Lipids influence ABC transporter function Many insights into the interplay between proteins and lipids can already be garnered from biochemical assays. For ABC transporters, monitoring ATPase or substrate transport activity has for a long time been the first pass to assess activity. Stimulated ATPase assays, where the presence of a substrate increases ATP turnover, has been an additional approach to determine substrate preferences. A number of ABC transporters show different basal and/or stimulated ATPase activity in detergent micelles when compared to activity measurements in liposomes or nanodiscs, complicating the interpretation of the functional data. Pgp for example has a negligible/non-existent basal ATPase rate in DDM (n-dodecyl-ß-D-maltoside) micelles, but regains basal activity when reconstituted in liposomes [85]. Similarly, ABCG1 does not exhibit ATPase activity unless reconstituted into a lipid bilayer [86], and even

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

here, ATPase activity depends on the type of detergent used for the previous purification. ABCG1 purified with DDM was active in a reconstituted system, while protein purified with Fos-choline-14 was not [86]. A similar observation was made for the Lactococcus lactis exporter LmrA [87]. A potential explanation for this observation may be the removal of important lipids by too harsh solubilization conditions (see below). All these transporters' basal ATPase activity can be further stimulated with amphipathic drugs [87–90]. In addition, for human Pgp reconstituted into nanodiscs, the ATPase rate was even higher than in proteoliposomes made from the same lipids [91]. Indeed, the increased rate was more than twice the rate observed in liposomes (both for the basal and the stimulated activity) and thus must have a different explanation than a simple issue of ATP-accessible NBDs. In liposomes, when the protein is randomly integrated, it can be expected that up to 50% of the protein is ATPase incompetent due to the fact that the NBDs are on the inside of the liposome. This is not the case for nanodiscs, which on the other hand have the disadvantage that one cannot measure transport rates. Potentially, the bilayer lateral pressure is higher in nanodiscs than in liposomes and this may explain differences in the observed ATPase activity of ABC transporters [91,92]. Likewise, MsbA's ATPase activity in nanosdiscs was increased compared to detergent and liposomes [93,94]. Here, MsbA activity depended on lipid type as well as the nanodisc diameter [93]. The homodimeric ABC transporter MsbA is an essential protein in E. coli that mediates the translocation of lipopolysaccharides (LPS) (including lipid A as their lipid component) from the inner to the outer leaflet of the inner membrane [95]. Most, if not all ABC transporters display detergent and lipid sensitive ATPase and/or substrate transport activity. Other described examples are the transporter associated with antigen processing (TAP) that moves peptides across the ER membrane [96] and the mitochondrial ABC transporter Atm1 [97]. Interestingly, many ABC exporters seem to be very dependent on the presence of lipids to display basal ATPase activity, while for numerous importers, the opposite has been reported: The type I maltose transporter MalFGK2 for instance, has a maltose binding proteinindependent, uncoupled ATPase activity in detergent [98], which is lost in proteoliposomes or nanodiscs [99]. Similarly, the type II importers BtuCD and MolBC show a significant decrease in basal ATPase rate when being transferred from detergent to liposomes [100,101]. For the mammalian ABC lipid flippase ABCA4, reconstitution into liposomes also led to a decrease in ATPase activity, but since phospholipids were required throughout the purification to obtain an ATPase competent transporter in detergent [81], these data may be difficult to compare to that of bacterial importers. Nonetheless, one might speculate that if (lipid) ABC exporters indeed display a high ATPase activity while ABC importers display a low “basal” ATPase rate in lipid bilayers, this could be a hint that ABC exporters and importers interact with lipids differently or a testament to the very defined substrate specificity of most importers and the broader substrate range of many exporters. This seems reasonable based on their very different transmembrane architecture [13]. However, structure alone cannot be the only divide as the ABCG family also has a very different TMD fold than all other previously known ABC(B) exporters and it is possible that members of the ABCA, ABCC and ABCD families look different still. Finally, the comparison of ATPase activity between detergentsolubilized and even different lipid-reconstituted states may not be as straight-forward as it might seem at first glance. For instance, many lipids and detergents are confirmed or at least putative ABC transporter substrates (e.g. [24]). In this case, measuring a true basal ATPase activity in the presence of these molecules is extremely difficult. For functional studies carried out with detergent-solubilized proteins, additional thought needs to be given to the precise solubilizing conditions. For MsbA it was observed that short solubilization conditions that are typically used in many protocols (i.e. 1 h at 4 °C) left detectable amounts of lipid in the sample [102]. Similarly, the heterodimeric ABC transporter TmrAB from Thermus thermophilus was shown to be consecutively

5

delipidated with longer and longer detergent solubilization conditions, but even then tightly bound negatively charged lipids remained [103]. For TmrAB as well as MsbA, increased delipidation led to a concomitant decrease in ATPase activity [102,103]. Previous studies had observed the same effect for Pgp [104,105]. In summary, the presence or absence of lipids has an influence on ABC transporter activity, but it is difficult to distinguish between the various reasons for which a lipid may cause such a change. Is it because the lipid (or detergent) is really a substrate? Do lipids act as structural stabilizers that allow e.g. a more efficient coupling between the TMD and the NBD? Does this imply that lipids are integral parts of ABC transporter structures? Are ABC transporters less structurally stable in the absence of lipids? Or could lipids influence ABC transporters through more indirect effects, such as changes in membrane fluidity, hydrophobic (mis)match and/or lateral membrane pressure? These could then in turn also affect how ABC transporters interact with other hydrophobic molecules, such as drugs. Lipids can act as reservoirs for these drugs and many ABC transporters display surprising substrate promiscuity. The solubility of transporter substrates could differ substantially between lipid environments, strongly affecting dose-response behavior. Finally, functional differences between environments (e.g. different ATPase activities in proteoliposomes or detergent micelles) should be coupled to structural and/or dynamic alterations within the transporter. We will discuss examples for these cases below. 5. Are lipids integral parts of ABC transporter structures? Potentially the most straight-forward ways to monitor protein-lipid interactions are high-resolution X-ray crystallography or cryo-electron microscopy (cryoEM) structures in the presence of lipid molecules. The challenge then is only to determine whether the observed lipid molecule presents a substrate, an annular lipid or a structural component. However, if crystals of a protein in a detergent micelle had been obtained by vapor diffusion, it is also not clear whether the protein is in a fully native conformation, compared to a structure in a lipid bilayer. In crystallography, bicelle technology and lipid cubic phases (LCP) were developed to overcome this limitation [106–109]. Bicelles are phospholipid membrane-mimetic discs (e.g. DMPC-CHAPSO) that allow to keep proteins in a lipid-bilayer like environment [106]. Lipid cubic phases are highly stable three-dimensional lipid packing arrays, penetrated by a system of aqueous channels. During reconstitution the native protein structure is conserved and addition of a precipitant generates a local microenvironment around the protein that supports crystal formation. For ABC transporters, most X-ray structures to date have been determined in detergent micelles, but the structures of ABCG5/G8 (see above, [53], PDB ID: 5DO7), the Clostridium thermocellum polypeptide exporter PCAT1 ( [110], PDB ID: 4RY2) and the E. coli maltose importer in complex with a regulatory protein ( [111], PDB ID: 4JBW) have been solved in bicelles. In the inward-open crystal structure of the Caenorhabditis elegans Ppg, two detergent molecules are caught in the apex of the transmembrane domain cleft ( [112], PDB ID: 4F4C). These molecules are in roughly the same position as cyclic peptides observed in mouse Pgp ( [113], PDB ID: e.g. 4M2S) and detergent molecules in ABCB10 ( [54], PDB ID: 4AYT/3ZDQ). It is tempting to speculate that the observed position of detergent molecules in C. elegans Pgp point to a putative lipid transport mechanism, where lipid/detergent acyl chains are only partially in contact with hydrophobic amino acid side chains and mostly reside in the lipid bilayer while the hydrophilic head groups are tucked into the cavity (Fig. 4C). (We will discuss putative pitfalls of this hypothesis later). In the X-ray structure of ABCB10, cardiolipin and dodecylmaltoside molecules were observed binding to the outside of the TMD ( [54], PDB ID: 4AYT) (Fig. 2). Cardiolipin may be at a similar position to where it may reside in LmrA and LmrCD, L. lactis ABC transporters that bind cardiolipin as detected by intact protein mass spectrometry [114, 115]. Mass spectrometry (MS) has developed to become a versatile

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

6

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

Fig. 2. ABC transporters and their lipophilic substrates. The three ABC transporters structures are: C. elegans Pgp (PDB ID: 4F4C) with (A) undecyl-4-O-alpha-D-glucopyranosyl-1-thiobeta-D-glucopyranoside (gold) in the inward open cavity [112]; human ABCB10 (PDB ID: 3ZDQ) with (B) cardiolipin (gold) attached to the outside of the TMD [54]; C. jejuni PglK (PDB ID: 5C73) with the characteristic extracytoplasmic helices [79]. Attached lipids and detergents: (1) DEPC (C22:1/C22:1; 1,2-dierucoyl-sn-glycero-3-phosphocholine); (2) POPC (C16:0/C18:1; 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine); (3) NBD-PE (C18:1/C18:1; 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl)), fluorescent NBD in yellow; (4) daunorubicin; (5) DMPC (C14:1/C14:1; 1,2-dimyristoleoyl-sn-glycero-3-phosphocholine); (6) cyclosporine A; (7) cholesterol; (8) DOPC (C18:1/C18:1; 1,2-oleoylsn-glycero-3-phosphocholine); (9) Lipid A; (10) Lipid-linked oligosaccharide (LLO). As a size comparison: (11) ATP; (12) water.

and powerful tool to study cellular lipid compositions as well as proteinlipid interactions [116]. These latter studies became possible with the discovery that intact membrane proteins, with correct subunit assembly and even non-covalently attached ligands, could be released into the gas phase from a detergent solution [117]. The applicability of this technique was originally demonstrated on the E. coli vitamin B12 importer BtuCD where increased heterodimer stability in the presence of bound ATP and posttranslational modifications were detected [117]. Of note, no bound lipid molecules were detected for BtuCD using MS, but LDAO (lauryl dimethylamine N-oxide) molecules are present in a crystal structure on the outside of the TMD ( [118], PDB ID: 4R9U) and EPR spectroscopy showed that spin labels attached in proximity to the cytoplasmic and periplasmic gates showed increased flexibility in the detergent-solubilized state compared to liposomes [119]. In subsequent MS studies focusing on ABC exporters, tightly bound lipids could indeed be identified: MacB, a macrolide extrusion ABC transporter that interacts with the proteins MacA and TolC to form a tripartite extrusion complex in E. coli, binds to zwitterionic phosphatidylethanolamine [114]. In a systematic intact mass spectrometry study to identify lipid binding preferences for mouse Pgp, it was found that it prefers lipids over detergent molecules [120]. The head group influenced lipid binding preferences much more than acyl chain variation. A total of four zwitterionic or six negatively charged lipids could be detected in complex with the protein in the absence of nucleotides. It was found that negatively charged head groups bind more strongly than zwitterionic ones, and that smaller lipids bind more favorably than larger ones. Lipid binding to Pgp thus seems to depend on a combination of electrostatic and steric effects. For the T. thermophilus protein TmrAB, a total of 24 lipids in the annular belt could be detected by the same MS approach [103]. All these lipids had C16 to C18 carbon chains and were mostly mono-unsaturated. While both phosphatidylethanolamine and phosphatidylglycerol were detected in the annular belt before delipidation, after more rigorous delipidation, a remainder of an average of five phosphatidylglycerol species were found to bind tightly to TmrAB. Thus, just like Pgp, LmrA and LmrCD, TmrAB preferentially binds negatively charged lipids. In addition to its annular lipids, TmrAB also interacts with Lipid A molecules with various acyl chain species. TmrAB shares 30% sequence identity with E. coli MsbA, and just like MsbA, it can act as a multidrug transporter [121] that

presumably also transports lipid A, as this lipid is displaced during ATP hydrolysis (while ATP binding was not sufficient) [103]. Another technique that has made amazing progress in the last couple of years in obtaining high resolution structures of mammalian membrane proteins is cryoEM (e.g. [122,123]). For ABC transporters, an ~8 Å structure of the TmrAB heterodimer bound to a Fab fragment [124], and a 6.5 Å structure of the human transporter associated with antigen processing (TAP1/2) bound to a viral inhibitor are currently available [125]. At this resolution, no lipids can be discerned, but in a recent example for the nanodisc-reconstituted, agonist-bound TRPV1 ion channel, the cryoEM structure was of high enough quality (2.9 Å) to elucidate how a phosphoinositide and annular lipids interact with the channel [122]. For ABC transporters, nanodiscs have been used in functional studies before (e.g. [91,99,126]), therefore it is presumably only a matter of time before we can expect high-resolution structures of ABC transporters in lipid bilayers at atomic resolution. 6. Substrate promiscuity in ABC lipid transporters Interestingly, ABC transporters that function as lipid floppases, meaning they translocate lipids from the inner to the outer membrane leaflet, seem to be relatively promiscuous with regards to their lipid preferences. Many ABC transporters can interact with and translocate a large number of lipids as well as sterols and modified lipids. In addition, ABC transporters have, from the earliest days of their discovery, always been studied in the context of being able to convey “multidrug resistance” to cells, the ability to render these cells insensitive to a variety of otherwise harmful drugs. Indeed, Juliano and Ling described in their seminal 1976 paper the presence of a glycosylated protein whose presence at the cell surface directly correlated with the inverse of the permeability of a variety of structurally unrelated drugs, hence terming the name P-glycoprotein, which is now also called MDR1 (multidrug resistance protein 1) [127]. The ability of Pgp to extrude drugs encompasses hundreds of compounds. Transport is contingent on ATP hydrolysis and also dependent on the lipid environment (e.g. [104,128], reviewed in [24]). Since the discovery of Pgp, it has become apparent that other mammalian ABC transporters can also function as MDR transporters. For example, similar to Pgp, the lipid transporters ABCB4 and ABCC2 are also known by their alternative names (MDR3,

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

multidrug resistance protein 3) and MRP2 (multidrug resistance associated protein 2), pointing to their role in drug translocation. Many bacterial lipid transporters are also able to translocate drugs, including MsbA, LmrA [129] and TmrAB [121]. Such promiscuity in transport is interesting and raises the question whether all lipid transporters are drug transporters and vice versa. For various reasons, this question is difficult to answer, beginning with trying to define what a true MDR substrate really is. Initially, MDR substrates were defined as molecules that cause cells to overexpress a higher number of (ABC-)efflux pumps in response to drug exposure and are thus rendered resistant against this compound. Once it had been realized that Pgp and other proteins play an important role in cancer resistance to chemotherapy, numerous studies aimed to find unifying properties of molecules that made them MDR substrates. It was first hypothesized that a basic nitrogen atom and two planar aromatic moieties are the important chemical structure elements [130]. Subsequently, a more comprehensive compound screening revealed that molecules without a basic nitrogen can also interact with Pgp [131] and transported substrates can include steroid hormones [132]. In an early systematic study, Pgp substrates were classified according to their electron donor group patterns. Substrates fell into two groups, type I with two electron donor groups, spatially separated by 2.5 ± 0.3 Å and type II unit with two donor groups spatially separated by 4.6 ± 0.6 Å or three donor groups when the spatial separation of the outer two groups also reached 4.6 ± 0.6 Å [131] (Fig. 3). In a follow-up study, similar substrate recognition patterns were identified for MRP1 (multidrug resistance associated protein 1, ABCC1) [133]. Substrate binding to Pgp or MRP1 was found to be enhanced with the increase in number of electron donor or hydrogen bonding acceptor groups forming the patterns for type I and type II recognition units. Electron donor groups of type I are present in the lipid interface region between the polar head groups and the hydrophobic core of lipids, such as ester groups [131,134]. Thus, promiscuous lipid transport by ABC transporters can potentially be explained by this substrate model, where a negatively charged type I unit (phosphatidyl moiety) and two type I electron donating ester groups are complemented by the hydrogen and electron donor networks of the respective lipid head groups [133] (Fig. 3).

Fig. 3. Drugs intercalate into the glycerol backbone of lipid bilayers and drugs and lipids may share similar recognition patterns for ABC lipid/MDR transporters. Using solid-state NMR, the precise localization of MDR pump substrates has been determined within lipid bilayers [142] as shown here exemplary for daunorubicin. A specific electron donor pattern for Pgp substrates has been observed (type I donor pattern) [131] that may be similar between drugs and lipids. This could be an explanation why so many drug transporters translocate both lipids and drugs.

7

7. Lipids as reservoirs for ABC transporter substrates For drug extrusion by ABC transporters, different functional models such as the “vacuum cleaner” and “solvation exchange” have been discussed [135,136]. Regardless of the details of each proposed model, they all assume incorporation of the drugs into the lipid bilayer as the first step of the transport process. Therefore, to understand multidrug resistance caused by (ABC) transporters, uptake of the drug molecules from a membrane environment must be elucidated. An interesting observation has been that the Km and Vmax values of Pgp ATPase activation by drugs are not only related to the hydrogenbond pattern but also the air-water partitioning coefficient of the substrates [137,138]. This work thus postulated that the interaction of the drug with the lipid bilayer is the rate-limiting step for substratePgp interaction. Using substrates with an intrinsic fluorescence and tryptophan fluorescence quenching within Pgp, it was discovered that substrates preferentially partition into the liquid crystalline phase of a lipid bilayer compared to the gel phase [139] while generally binding affinity of drugs to Pgp was higher in the gel phase. Drug transport rates were higher in the liquid crystalline phase than in the gel phase [139,140]. There is clearly a complicated interplay between drugs and the membrane, which may affect lipid fluidity and subsequently drug mobility as well as transporter dynamics. Likewise, drug and transporter as well as lipid and transporter interactions play an important role, as they may affect protein conformation and subsequently transport efficiency. The relevance of the membrane phase for Pgp activity is also apparent from the fact that membrane fluidizers and surfactants can reverse multidrug resistance [141]. Frustratingly, Pgp's activity is extremely sensitive towards its environment, i.e. lipid composition or detergent micelles, thus rendering questions regarding the specific effect of a lipid type or membrane fluidity, etc. so hard to study. But where are MDR substrates localized in the membrane? Can their location and interaction with ABC transporters inform on mechanisms how lipids are being recognized? Using 1H solid-state nuclear magnetic resonance (NMR) spectroscopy, the localization of nine Pgp substrates and modulators was investigated in a liquid crystalline phase of DMPC and DMPC/DMPG [142]. It was found that these compounds do not distribute evenly across the lipid bilayer but rather accumulate below the lipid head group in the vicinity of the glycerol backbone. Another NMR study with a different set of ABC substrates, such as flavonoids, found that they are also located most frequently in the interface of the lipid head group and the hydrocarbon tails [143]. The position identified by NMR for drugs is in the same area that would fall into a type I electron donor pattern as postulated for lipid head group recognition (Fig. 3). Could it then be that drug recognition by MDR transporters is an accident? That they really want to grab on a lipid head group but the drug presents itself in the right location of the membrane with the correct electron donor group pattern? Or just that nature, at least in the case of ABC transporters, has taken advantage of those specialized on lipids, to carry out another, highly similar job, that of an MDR pump? A huge advantage of NMR is that both drugs and lipids can be monitored simultaneously without the need of artificial labeling. The advantage of not requiring exogenous labelling of a substrate is also realized by Raman scattering [144]. This technique can distinguish between the orientations (gauche or trans) of the lipid acyl chains by detection of the C-C (gauche: ~ 1086 cm− 1 and trans: ~ 1065–1114 cm− 1) stretching modes of the methylene regions when applied to phospholipid vesicles. Changes in gauche or trans are observed via peak intensity changes between the gauche to trans peaks. Analysis of the effect of drugs (ibuprofen and salicylate) on DMPC lipids were indeed indicative of changes in the lipid environment. Similarly, changes in membrane fluidity were observed upon drug addition to lipids by differential scanning calorimetry (DSC) (e.g. [145]). Because ABC transporter activity is dependent on lipid phase, membrane thickness and fluidity [24], a change in lipid acyl chain conformation

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

8

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

may in turn also influence its behavior, i.e. ATPase activity, even if the drug causing the effect is not a substrate itself. 8. Lipids and substrates affect structure and dynamics of ABC exporters The conformational space accessible to an ABC transporter under various conditions (e.g. in a lipid versus a detergent environment) is a question of ongoing debate. To put it to rest, high resolution structures of ABC transporters will need to be coupled with experiments to describe functional dynamics of these systems in lipid environments. Magnetic resonance and optical techniques, such as EPR (electron paramagnetic resonance), NMR, fluorescence and luminescence studies are currently used for this purpose. For soluble proteins, even membrane associated ones, solution NMR spectroscopy is a well-established technique to determine both structure and dynamics of a protein of interest. A number of ABC transporter NBDs have been investigated by solution NMR, regarding, e.g. the interaction with nucleotides [146,147]. For membrane proteins on the other hand, due to their size and tumbling behavior, this is often not a trivial feat. In order to study ABC transporters at atomic resolution in a lipid environment, solid-state or magic angle spinning (MAS) NMR, which is not size limited, can be used. Solid-state NMR studies were undertaken on the full-length bacterial exporters LmrA, MsbA and BmrA [102,148–150], as well as on the importer ArtMP [151]. For reconstituted LmrA, it was observed that the phase transition temperature was reduced compared to empty liposomes [149] in agreement with earlier studies with DSC on reconstituted Pgp [152]. The protein displayed high flexibility in liposomes (in agreement with previous EPR experiments in detergent [153]) and this could only partially be reduced by nucleotide trapping, although specific residues become motionally restricted upon nucleotide binding [149]. Under the right conditions, NMR can measure kinetics (such as ATPase activity), enzymatic substrate turnover and even transport [154,155]. 31P solid-state NMR has been used to probe the ATPase activity of LmrA and to detect the presence of lipids bound to MsbA [102,156]. DEER (Double electron-electron resonance, also referred to as PELDOR, pulsed electron-electron double resonance) experiments to determine distances between spin labels, has been a powerful tool to dissect the transport and ATPase cycles of many ABC transporters [157,158], but MsbA remains the best studied ABC exporter. For pulsed EPR experiments, typically cysteine residues are introduced into the protein of interest and nitroxide radical spin labels are attached to the thiol-sidechain (this approach is referred to as site-directed spin labeling, SDSL). For DEER/PELDOR experiments, the sample is frozen and the distance between two such spin labels is then determined. Pulsed EPR studies have been carried out on MsbA in detergent micelles, in proteoliposomes and in so-called NABBS (nanoscale apolipoprotein-bound bilayers), that resemble nanodiscs [159–161]. Attaching spin label pairs in both the MsbA TMD and NBD, interspin distance distributions for MsbA in detergent micelles and liposomes were determined. The overall distance distribution and conformational changes upon ATP hydrolysis (established by vanadate trapping) were the same for MsbA in detergent and lipid [159,160]. Likewise, LRET (luminescence resonance energy transfer) studies have been carried out in detergent, proteoliposomes and nanodiscs [162,163]. While FRET (Förster resonance energy transfer) studies make use of two fluorophores and measure the distance between them, LRET uses luminescent lanthanide ions with long life times, e.g. terbium. Compared to FRET, LRET signals tend to have very high signal to noise ratio and sharper linewidths [164]. A single cysteine mutant of MsbA was labeled with a terbium chelator (Tb3+ acts as the donor) and a fluorescent dye (acceptor) on the C-terminal end of the NBD. In detergent, apo MsbA exists almost exclusively in an inward open conformation [163]. However, in nanodiscs and liposomes, the protein exists in two conformations, inward open and outward open, that are

almost equally populated. ~ 50% distribution between two protein populations indicates very similar free energies for these states. Since an equal population of inward and outward open conformations is contingent on the presence of lipids (but not on liposome versus nanodisc) and NBD closure rates are faster in presence of lipids [162], it may be speculated whether lipids contribute to the reduction of the thermodynamic threshold for the transition between states and thus allow for faster exchange between states. Additionally, reducing the difference between ground state energies of the two states would lead to equal population sizes. Lipids could achieve this by changing the relative stability of the outward open compared to the inward open state. While this may potentially be an explanation why lipids lead to enhanced basal ATPase activity, the structural basis of this finding remains unclear. Because the “stimulating” effect of lipids on MsbA's ATPase activity has been observed in detergent micelles in biochemical studies [165], it seems likely that this is due to a direct interaction between lipid and protein, i.e. that the lipid(s) act(s) as a kind of molecular grease to ease conformational transitions. This could potentially happen through lipids inserting themselves between transmembrane domains, rather than a long-range effect based, e.g. on lateral membrane pressure. Intriguingly, lipid-like molecules wedged between the transmembrane helices 4 and 5 crossing over to the trans transporter half and the TMH1–3,6 cis helix bundle have been observed within the crystal structure of Atm1 ( [166], PDB ID: 4MRP) and Atm1’s ATP hydrolysis is lipid-dependent [97]. By using LRET on MsbA, it was also observed that the NBDs were much closer to each other in lipid environments than in the detergent solubilized state even in the absence of ATP [162]. Combining stopped-flow technology with LRET measurements, the effects of MgATP on MsbA in nanodiscs under continuous hydrolysis conditions have been studied, and distinct conformations rather than a continuum were observed [162]. Two distinct conformational populations have also been seen for MsbA and Pgp in amphiphile-lipid mixtures [167] or Pgp in detergent [168] with cyroEM. Of note, an earlier EM study of Pgp in lipid monolayers observed a conformation with closed NBDs in the apo state [169]. In the recent study, for Pgp in detergent, the two conformations (inward and outward open) were equally populated. For Pgp in lipid-amphiphile mixtures, however, the protein was mostly in the inward open conformation, and addition of ATP or MgATP did not significantly alter the population distributions. Only the co-incubation of MgATP with vanadate and an ATPase stimulating drug led to an obvious increase in an outward open conformation, prompting the authors to speculate that the outward open state of Pgp may be a high energy conformation [167]. In contrast, MsbA already showed a small population (1–3%) in the outward open conformation even in the absence of nucleotides and this was increased to 58% upon addition of ATP, thus showing similar results to the LRET data, where ~10% of detergent-solubilized apo MsbA was observed in the outward open state which increased to ~80% in the presence of ATP [162,167]. Interestingly, in an intact protein ion mobility MS study, detergentsolubilized apo Pgp showed only one major arrival time, thus indicating a single (or closely related) conformation [120]. According to numerous crystal structures, this should be the inward open conformation. In addition, a less populated second population was observed, which corresponds to the theoretical flight time expected for an outward open MsbA dimer (MsbA was chosen as a reference because structural data exists for the outward open conformation and it is structurally similar to Pgp). It was thus speculated that Pgp exists in two distinct conformations in the apo state, with the majority in an inward open state and a smaller population in an outward open state. Importantly, no indication for a continuum of conformers was observed, which is in agreement with cryoEM data for Pgp in detergent ( [168] and cryoEM and LRET on MsbA in lipid environments [163,167], but at odds with cryoEM data of Pgp in a lipid environment [167]. In the MS experiments, separate addition of cardiolipin, ATP or cyclosporine A (CsA) did not change the distribution of conformational states. Only simultaneous addition of ATP and CsA or simultaneous addition of cardiolipin and CsA led to a

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

strong increase of the second outward open population, similar to what has been observed with cryoEM [167]. Of note, the second population (believed to be the outward open state) always only reached the same occupancy as the inward open state. It is possible that different ABC transporters show different switching behavior and may have different thresholds for this switching to occur. When comparing Pgp with MsbA for instance, care should be taken to note that MsbA is a homodimer, while Pgp has sequentially asymmetric halves that are tethered by a linker that may restrict conformational freedom. Regardless of the details, however, it seems clear that

9

lipids play an important part in defining the conformational space of ABC exporters. 9. Different models for lipid translocation by ABC exporters Small solute or drug transport can easily be envisioned by a transporter enclosing the substrate completely and binding and releasing it in the same orientation via an alternating access mechanism (Fig. 4A). Although such a restriction in substrate orientation is of course not a requirement for solute transport, a 180° turn with regards to its entry

Fig. 4. Potential transport modes for ABC transporters and lipids. Different models for substrate translocation by ABC transporters are depicted. Conformations with available structural data are indicated with a blue background. (A) Solute/ drug transport. Solute transport can occur via the alternating access model. The substrate can principally enter from the cytoplasm or the membrane and leave the transporter in the same orientation it entered. A representative structure for the inward open drug-bound state is given by mouse Pgp in complex with cyclic peptides (PDB ID: 4M2S) [113], while the outward open conformation was observed for Sav1866 (PDB ID: 2HYD) [170]. (B) Whipping mechanism for lipid-linked oligosaccharyls (LLO) by PglK following the model by Perez et al. [79]. The long acyl chain attaches to extracytoplasmic helices and transport is mediated solely through outward-facing conformations. LLO polar regions are shielded from the lipid bilayer by the transporter cavity, while the acyl chain never leaves the hydrophobic environment. The outward occluded ADP-bound structure is given by (PDB ID: 5C73). (C) Lipid threading mechanism inspired by the finding that in C. elegans Pgp (PDB ID: 4F4C) [112] detergent molecule headgroups are located in the transporter apex in the inward open state while the acyl chains remain outside. Lipid transport would follow an alternating access model, however, due to the different surfaces presented by the transporter in the inward- and outward-facing states, lipid translocation in this way would require threading through the transporter's extracytosolic loops. (D) Lipid gymnastics in the occluded cavity allow a solute-like transport mechanism. In contrast to the solute, however, the lipid has to change its orientation upon translocation. This can potentially be achieved through anchor points for the nonpolar acyl chains or the polar head group within the transporter, thus keeping unfavorable interactions to a minimum.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

10

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

orientation is absolutely necessary for lipids, as unfavorable exposure of the polar head group to the lipid bilayer core or exposure of lipid acyl chains to the aqueous environment should be avoided. Any “flopping” of the lipid within an occluded transporter cavity would either require a serious energetic penalty due to unfavorable (transient) interactions and/or major gymnastics by the transported substrate (Fig. 4D). At least in bacteria, the Campylobacter jejuni lipid-linked oligosaccharide (LLO) floppase PglK seems to have developed an intriguing solution for moving long-chain lipid molecules from one side of the membrane to the other. By both binding to and translocating the substrate in an outward open conformation, PglK transfers the LLO in a whip-like manner [79] (Fig. 4B). In contrast, it is generally believed that MsbA, Pgp and other related ABC exporters cycle between a high substrate affinity inward open state and an ATP-mediated low-affinity outward open state (Fig. 4A, C, D). However, this alone does not explain how they manage to translocate lipids from one bilayer leaflet to the other. The structures of PglK show a similar fold to other ABC exporters, such as Sav1866 and MsbA [170,171], with long, slender transmembrane domains composed of six transmembrane helices each, and the typical swapping over of transmembrane helices 4 and 5. However they also revealed a feature previously not observed for ABC transporters: a small extracytoplasmic helix oriented in parallel to the membrane plane between TMH1 and TMH2 [79] (Fig. 2, Fig. 4B). This helix forms a hydrophobic groove with the transmembrane helices. Its mutation or deletion led to loss of ATPase stimulation by LLO. Because only long-chain LLOs could stimulate the PglK's ATPase, it was postulated that the short extracellular helix is involved in lipid substrate recognition. PglK was crystallized in two inward open apo conformations and an ADP-bound outward occluded conformation. Since occluding the inward open cavity had no effect on transport, Perez et al. suggested that only the outward open states are transport competent. The relatively small cavity in the crystallized ADP-bound state would not be able to accommodate the LLO. In contrast, in a fully open ATP-bound Sav1866 like state the cavity dimensions could potentially provide sufficient space and expose arginine residues to interact with the phosphate groups of the LLO. This would force the oligosaccharide group to follow, thus explaining its negligible role in substrate specificity, while the LLO polyprenyl tail snorkels up to the external helices. As ATP is hydrolyzed and the NBDs and successively the TMDs change their conformations, the pyrophosphate group is “squeezed” out of the transporter, released and the hydrophobic tail diffuses off or follows as the polar groups of the LLO are grabbed by subsequent acceptor proteins. Throughout this process, the polyprenyl tail remains outside of the transporter (Fig. 4B). Given an anchor point (similar to the external helices in PglK) for their hydrophobic regions, it is conceivable that other lipid transporters could use a mechanism similar to that of the LLO-floppase. This mechanism would, however, require extraordinarily long fatty acid tails by the substrate. This may be the case for some specialty transporters, such as e.g. the ABCD transporter family or bacterial O-antigen polysaccharide ABC transporters. Also, the anchor points may not be structurally realized through appendage helices, but could be hydrophobic pockets on the surface of the transporter. For MsbA, it seems unlikely that it follows the same transport mechanism as PglK. First, it lacks PglK's small helices or a similar hydrophobic groove and lipid A's acyl chains are much shorter than LLO's (Fig. 2). ATP-bound MsbA adopts an outward open conformation [160] and binding affinity for lipid A or drugs was reduced considerably when ATP was pre-bound [172]. This is the opposite of what should be expected for the PglK translocation mechanism and importantly, negates the idea of an outward open ATP-bound state as the transport competent state for MsbA. For both transporters, however, substrate translocation strictly depends on ATP hydrolysis [79,173]. The ADP-bound state found to stabilize outward open PglK is usually believed to be a transient state and either inward open or moving towards an inward open state, as Pi release and increase in negative charges at the NBD interface is

believed to drive the NBDs apart [174]. The ADP-bound state has also been somewhat underappreciated in biophysical studies of ABC transporters, although it is an important transition state within the ATPase catalytic cycle. Despite this negligence, it is not inconceivable that for some ABC transporters, the ADP-bound state could also be an essential part of the substrate translocation cycle, used e.g. to block re-binding of the just released substrate. This notion could hold true regardless of the specific mode of action for substrate translocation. Indeed, our own data using EPR spectroscopy have found an ADP-bound outward open state for the ABC exporter LmrA [153] and of note, in a recent paper studying MsbA with cryoEM, a small increase in the outward open conformation upon addition of ADP was also observed [167]. It will be exciting to see if more examples of stable ADP-bound states can be observed for other ABC transporters and if its influence on substrate interaction can be further elucidated in future studies. Despite the progress in understanding MsbA function, little is known about how and where MsbA binds its substrates, how the substrate translocation process takes place and even whether the pathway is equal for all substrates. One putative transport mechanism for various lipid species could be evoked from the picture of the C. elegans Pgp structure ( [112], PDB ID: 4F4C, Fig. 2). Here, detergent molecule head groups are bound in the inward open cavity while the acyl chains protrude outward toward the bilayer center (Fig. 4C. In the inward open conformation, the lipid head group can easily enter the transporter cavity. As the lipid's acyl chains stay bilayer exposed, the head group can move up into the apex of the cavity and would thus become the determining factor for specificity. The difficulty with this model is how to move the lipid from this position to the other membrane side, as different helices line the cavity surface in the inward and outward open conformations (see e.g. [175]. This means that there is no undisturbed path for a lipid through the transporter if its acyl chains remain bilayer exposed (Fig. 4C, bottom) The original lipid binding site between TMH4 and 6 in the transporter becomes restricted by the extracytosolic loops between TMH 3/4 and 5/6 as the transporter moves into an outward occluded and subsequently into an outward open state (Fig. 4C). To exit the transporter, the lipid molecule would have to thread through the opening formed by the transporter, similar to a piece of twine being threaded through the eye of a needle. It is possible that, as the outward open cavity begins to form, the hydrophilic head group is also already being tucked at by favorable interactions with the aqueous phase. At some point, it is then energetically more favorable for the lipid head group to move forward rather than backward. Thereby, it will pull the acyl chains with it similarly to an air-filled balloon that is submerged under water and breaks free. Such a mechanism should penalize lipids with too long or too inflexible acyl chains and this may potentially be tested in a systematic approach. Unfortunately, we are currently lacking enough detailed structural pictures of the intermediate conformations (preferably with lipids or detergent molecules bound) required between the presumed endpoints of the transporter trajectory to support or refute this model. Alternately, initial lipid binding could still happen as described above, with the substrate entering the inward open cavity, but in the occluded state, the entire lipid becomes enclosed by the transporter (Fig. 4D). If either the head group, or one or both acyl chains act as anchor points within the protein chamber interior, a 180° flip can be imagined after some lipid gymnastics as lipids are surprisingly flexible molecules [176–178]. This mechanism should, however, have penalties for more sterically demanding lipids. Again, we are in dire need of structural intermediates to support or refute possible mechanisms. Potentially, ABC transporters may even use various mechanisms (or combinations thereof) for different substrates, thus making it ever so hard to figure out how they function in mechanistic detail. Conclusion and outlook ABC transporters, lipids and their functional and structural interplay are an exciting topic to study, especially with so many new technical

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

advancements becoming available. Nonetheless, structures for a number of different ABC transporter subfamilies that translocate lipids are still missing. For others, only a single conformation has been characterized. Thus, it is clear that this journey is only at the beginning. With so many, sometimes contradictory data accumulating, we will need to carefully sift through the evidence and dissect the one or the many ways, how lipids are recognized and transported by ABC transporters. We believe that focusing of the dynamic features of these exquisite molecular machines will go a long way to disentangle the mysteries surrounding one of the most fundamental cellular processes, the lipid flip. Also, we should not forget that ABC transporters do not only consist of a membrane domain, but also of a nucleotide binding domain. An aspect that we have not covered in this review at all is the dependence of substrate transport on the NBD. For a number of ABC transporters, it is very obvious that mutations in the TMD affect NBD function, namely ATP hydrolysis. But the opposite is true as well: the NBD strongly influences which substrates are selected and how they are translocated. For ABCG5/G8 for instance, substitutions in NBD1 abolished cholesterol transport, although this NBD is ATPase deficient, while its mutagenesis did not affect plant sterol transport [179]. In TAP, a single mutation in the D-loop turned a strictly unidirectional transporter into a concentration-dependent facilitator [180] and in the yeast multidrug transporter Pdr5, mutation of the H-loop affected rhodamine transport but not that of other substrates [181]. Understanding how lipids and ABC transporters interact will therefore also hinge on a detailed understanding of how the NBD and TMD communicate. Transparency document The Transparency document associated with this article can be found, in online version. For more details regarding all putative lipid transport mechanisms, please refer to the main text. Acknowledgements We thank Dr. Thomas Stockner, Benedikt Goretzki, Franziska von Hammerstein and Erika Pfeifer for critical reading and insightful comments. Funding in our laboratory is generously provided by the Carl-Zeiss Foundation. Due to the wealth of literature on the topic of ABC transporters and lipids, it was impossible to cite every researcher and we apologize for this. References [1] F. Feijo Delgado, N. Cermak, V.C. Hecht, S. Son, Y. Li, S.M. Knudsen, S. Olcum, J.M. Higgins, J. Chen, W.H. Grover, S.R. Manalis, Intracellular water exchange for measuring the dry mass, water mass and changes in chemical composition of living cells, PLoS ONE 8 (2013) e67590. [2] G. van Meer, D.R. Voelker, G.W. Feigenson, Membrane lipids: where they are and how they behave, Nat. Rev. Mol. Cell Biol. 9 (2008) 112–124. [3] E. Fahy, S. Subramaniam, H.A. Brown, C.K. Glass, A.H. Merrill Jr., R.C. Murphy, C.R. Raetz, D.W. Russell, Y. Seyama, W. Shaw, T. Shimizu, F. Spener, G. van Meer, M.S. VanNieuwenhze, S.H. White, J.L. Witztum, E.A. Dennis, A comprehensive classification system for lipids, J. Lipid Res. 46 (2005) 839–861. [4] L. Yetukuri, K. Ekroos, A. Vidal-Puig, M. Oresic, Informatics and computational strategies for the study of lipids, Mol. BioSyst. 4 (2008) 121–127. [5] T. Pomorski, S. Hrafnsdottir, P.F. Devaux, G. van Meer, Lipid distribution and transport across cellular membranes, Semin. Cell Dev. Biol. 12 (2001) 139–148. [6] R.N. Hvorup, B. Winnen, A.B. Chang, Y. Jiang, X.F. Zhou, M.H. Saier Jr., The multidrug/oligosaccharidyl-lipid/polysaccharide (MOP) exporter superfamily, Eur. J. Biochem. 270 (2003) 799–813. [7] F. Quazi, R.S. Molday, Lipid transport by mammalian ABC proteins, Essays Biochem. 50 (2011) 265–290. [8] C. Montigny, J. Lyons, P. Champeil, P. Nissen, G. Lenoir, On the molecular mechanism of flippase- and scramblase-mediated phospholipid transport, Biochim. Biophys. Acta (2015). [9] R.D. Baldridge, T.R. Graham, Two-gate mechanism for phospholipid selection and transport by type IV P-type ATPases, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) E358–E367. [10] F. Quazi, S. Lenevich, R.S. Molday, ABCA4 is an N-retinylidenephosphatidylethanolamine and phosphatidylethanolamine importer, Nat. Commun. 3 (2012) 925.

11

[11] S.K. Sahu, S.N. Gummadi, N. Manoj, G.K. Aradhyam, Phospholipid scramblases: an overview, Arch. Biochem. Biophys. 462 (2007) 103–114. [12] E.M. Bevers, P.L. Williamson, Phospholipid scramblase: an update, FEBS Lett. 584 (2010) 2724–2730. [13] J. ter Beek, A. Guskov, D.J. Slotboom, Structural diversity of ABC transporters, J. Gen. Physiol. 143 (2014) 419–435. [14] D. Barthelme, S. Dinkelaker, S.V. Albers, P. Londei, U. Ermler, R. Tampe, Ribosome recycling depends on a mechanistic link between the FeS cluster domain and a conformational switch of the twin-ATPase ABCE1, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 3228–3233. [15] G. Boel, P.C. Smith, W. Ning, M.T. Englander, B. Chen, Y. Hashem, A.J. Testa, J.J. Fischer, H.J. Wieden, J. Frank, R.L. Gonzalez Jr., J.F. Hunt, The ABC-F protein EttA gates ribosome entry into the translation elongation cycle, Nat. Struct. Mol. Biol. 21 (2014) 143–151. [16] G.E. Tusnady, B. Sarkadi, I. Simon, A. Varadi, Membrane topology of human ABC proteins, FEBS Lett. 580 (2006) 1017–1022. [17] E.J. Tarling, T.Q. de Aguiar Vallim, P.A. Edwards, Role of ABC transporters in lipid transport and human disease, Trends Endocrinol. Metab. 24 (2013) 342–350. [18] K.E. Berge, H. Tian, G.A. Graf, L. Yu, N.V. Grishin, J. Schultz, P. Kwiterovich, B. Shan, R. Barnes, H.H. Hobbs, Accumulation of dietary cholesterol in sitosterolemia caused by mutations in adjacent ABC transporters, Science 290 (2000) 1771–1775. [19] R. Allikmets, N. Singh, H. Sun, N.F. Shroyer, A. Hutchinson, A. Chidambaram, B. Gerrard, L. Baird, D. Stauffer, A. Peiffer, A. Rattner, P. Smallwood, Y. Li, K.L. Anderson, R.A. Lewis, J. Nathans, M. Leppert, M. Dean, J.R. Lupski, A photoreceptor cell-specific ATP-binding transporter gene (ABCR) is mutated in recessive Stargardt macular dystrophy, Nat. Genet. 15 (1997) 236–246. [20] M.L. Fitzgerald, Z. Mujawar, N. Tamehiro, ABC transporters, atherosclerosis and inflammation, Atherosclerosis 211 (2010) 361–370. [21] F. Quazi, R.S. Molday, Differential phospholipid substrates and directional transport by ATP-binding cassette proteins ABCA1, ABCA7, and ABCA4 and disease-causing mutants, J. Biol. Chem. 288 (2013) 34414–34426. [22] C. Vedhachalam, P.T. Duong, M. Nickel, D. Nguyen, P. Dhanasekaran, H. Saito, G.H. Rothblat, S. Lund-Katz, M.C. Phillips, Mechanism of ATP-binding cassette transporter A1-mediated cellular lipid efflux to apolipoprotein A-I and formation of high density lipoprotein particles, J. Biol. Chem. 282 (2007) 25123–25130. [23] N. Wang, D. Lan, M. Gerbod-Giannone, P. Linsel-Nitschke, A.W. Jehle, W. Chen, L.O. Martinez, A.R. Tall, ATP-binding cassette transporter A7 (ABCA7) binds apolipoprotein A-I and mediates cellular phospholipid but not cholesterol efflux, J. Biol. Chem. 278 (2003) 42906–42912. [24] F.J. Sharom, Complex Interplay between the P-Glycoprotein Multidrug Efflux Pump and the Membrane: Its Role in Modulating Protein Function, Front Oncol. 4 (2014) 41. [25] P.D. Eckford, F.J. Sharom, P-glycoprotein (ABCB1) interacts directly with lipidbased anti-cancer drugs and platelet-activating factors, Biochem. Cell Biol. 84 (2006) 1022–1033. [26] K.J. Linton, Lipid flopping in the liver, Biochem. Soc. Trans. 43 (2015) 1003–1010. [27] Y. Zhao, M. Ishigami, K. Nagao, K. Hanada, N. Kono, H. Arai, M. Matsuo, N. Kioka, K. Ueda, ABCB4 exports phosphatidylcholine in a sphingomyelin-dependent manner, J. Lipid Res. 56 (2015) 644–652. [28] S.P. Cole, R.G. Deeley, Transport of glutathione and glutathione conjugates by MRP1, Trends Pharmacol. Sci. 27 (2006) 438–446. [29] R.J. Raggers, A. van Helvoort, R. Evers, G. van Meer, The human multidrug resistance protein MRP1 translocates sphingolipid analogs across the plasma membrane, J. Cell Sci. 112 (Pt 3) (1999) 415–422. [30] G. Yamano, H. Funahashi, O. Kawanami, L.X. Zhao, N. Ban, Y. Uchida, T. Morohoshi, J. Ogawa, S. Shioda, N. Inagaki, ABCA3 is a lamellar body membrane protein in human lung alveolar type II cells, FEBS Lett. 508 (2001) 221–225. [31] N. Ban, Y. Matsumura, H. Sakai, Y. Takanezawa, M. Sasaki, H. Arai, N. Inagaki, ABCA3 as a lipid transporter in pulmonary surfactant biogenesis, J. Biol. Chem. 282 (2007) 9628–9634. [32] S. Rajpopat, C. Moss, J. Mellerio, A. Vahlquist, A. Ganemo, M. Hellstrom-Pigg, A. Ilchyshyn, N. Burrows, G. Lestringant, A. Taylor, C. Kennedy, D. Paige, J. Harper, M. Glover, P. Fleckman, D. Everman, M. Fouani, H. Kayserili, D. Purvis, E. Hobson, C. Chu, C. Mein, D. Kelsell, E. O'Toole, Harlequin ichthyosis: a review of clinical and molecular findings in 45 cases, Arch. Dermatol. 147 (2011) 681–686. [33] Y. Zuo, D.Z. Zhuang, R. Han, G. Isaac, J.J. Tobin, M. McKee, R. Welti, J.L. Brissette, M.L. Fitzgerald, M.W. Freeman, ABCA12 maintains the epidermal lipid permeability barrier by facilitating formation of ceramide linoleic esters, J. Biol. Chem. 283 (2008) 36624–36635. [34] M. Akiyama, The roles of ABCA12 in epidermal lipid barrier formation and keratinocyte differentiation, Biochim. Biophys. Acta 1841 (2014) 435–440. [35] M. Akiyama, Y. Sugiyama-Nakagiri, K. Sakai, J.R. McMillan, M. Goto, K. Arita, Y. Tsuji-Abe, N. Tabata, K. Matsuoka, R. Sasaki, D. Sawamura, H. Shimizu, Mutations in lipid transporter ABCA12 in harlequin ichthyosis and functional recovery by corrective gene transfer, J. Clin. Invest. 115 (2005) 1777–1784. [36] D.P. Kelsell, E.E. Norgett, H. Unsworth, M.T. Teh, T. Cullup, C.A. Mein, P.J. DoppingHepenstal, B.A. Dale, G. Tadini, P. Fleckman, K.G. Stephens, V.P. Sybert, S.B. Mallory, B.V. North, D.R. Witt, E. Sprecher, A.E. Taylor, A. Ilchyshyn, C.T. Kennedy, H. Goodyear, C. Moss, D. Paige, J.I. Harper, B.D. Young, I.M. Leigh, R.A. Eady, E.A. O'Toole, Mutations in ABCA12 underlie the severe congenital skin disease harlequin ichthyosis, Am. J. Hum. Genet. 76 (2005) 794–803. [37] T. Gerloff, B. Stieger, B. Hagenbuch, J. Madon, L. Landmann, J. Roth, A.F. Hofmann, P.J. Meier, The sister of P-glycoprotein represents the canalicular bile salt export pump of mammalian liver, J. Biol. Chem. 273 (1998) 10046–10050. [38] H. Akita, H. Suzuki, K. Ito, S. Kinoshita, N. Sato, H. Takikawa, Y. Sugiyama, Characterization of bile acid transport mediated by multidrug resistance associated protein 2 and bile salt export pump, Biochim. Biophys. Acta 1511 (2001) 7–16. [39] T. Hirohashi, H. Suzuki, H. Takikawa, Y. Sugiyama, ATP-dependent transport of bile salts by rat multidrug resistance-associated protein 3 (Mrp3), J. Biol. Chem. 275 (2000) 2905–2910.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

12

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

[40] J.T. Mack, D.M. Townsend, V. Beljanski, K.D. Tew, The ABCA2 transporter: intracellular roles in trafficking and metabolism of LDL-derived cholesterol and sterolrelated compounds, Curr. Drug Metab. 8 (2007) 47–57. [41] E.J. Tarling, P.A. Edwards, ATP binding cassette transporter G1 (ABCG1) is an intracellular sterol transporter, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 19719–19724. [42] N. Wang, D. Lan, W. Chen, F. Matsuura, A.R. Tall, ATP-binding cassette transporters G1 and G4 mediate cellular cholesterol efflux to high-density lipoproteins, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 9774–9779. [43] T. Engel, S. Lorkowski, A. Lueken, S. Rust, B. Schluter, G. Berger, P. Cullen, G. Assmann, The human ABCG4 gene is regulated by oxysterols and retinoids in monocyte-derived macrophages, Biochem. Biophys. Res. Commun. 288 (2001) 483–488. [44] A.M. Vaughan, J.F. Oram, ABCA1 redistributes membrane cholesterol independent of apolipoprotein interactions, J. Lipid Res. 44 (2003) 1373–1380. [45] J. Wang, F. Sun, D.W. Zhang, Y. Ma, F. Xu, J.D. Belani, J.C. Cohen, H.H. Hobbs, X.S. Xie, Sterol transfer by ABCG5 and ABCG8: in vitro assay and reconstitution, J. Biol. Chem. 281 (2006) 27894–27904. [46] L.A. Doyle, W. Yang, L.V. Abruzzo, T. Krogmann, Y. Gao, A.K. Rishi, D.D. Ross, A multidrug resistance transporter from human MCF-7 breast cancer cells, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 15665–15670. [47] T. Janvilisri, H. Venter, S. Shahi, G. Reuter, L. Balakrishnan, H.W. van Veen, Sterol transport by the human breast cancer resistance protein (ABCG2) expressed in Lactococcus lactis, J. Biol. Chem. 278 (2003) 20645–20651. [48] A. Pal, D. Mehn, E. Molnar, S. Gedey, P. Meszaros, T. Nagy, H. Glavinas, T. Janaky, O. von Richter, G. Bathori, L. Szente, P. Krajcsi, Cholesterol potentiates ABCG2 activity in a heterologous expression system: improved in vitro model to study function of human ABCG2, J. Pharmacol. Exp. Ther. 321 (2007) 1085–1094. [49] Z. Gutay-Toth, F. Fenyvesi, O. Barsony, L. Szente, K. Goda, G. Szabo, Z. Bacso, Cholesterol-dependent conformational changes of P-glycoprotein are detected by the 15D3 monoclonal antibody, Biochim. Biophys. Acta 1861 (2015) 188–195. [50] A.T. Clay, P. Lu, F.J. Sharom, Interaction of the P-Glycoprotein Multidrug Transporter with Sterols, Biochemistry 54 (2015) 6586–6597. [51] I.M. Hafez, P.R. Cullis, Cholesteryl hemisuccinate exhibits pH sensitive polymorphic phase behavior, Biochim. Biophys. Acta 1463 (2000) 107–114. [52] A. Kodan, H. Shibata, T. Matsumoto, K. Terakado, K. Sakiyama, M. Matsuo, K. Ueda, H. Kato, Improved expression and purification of human multidrug resistance protein MDR1 from baculovirus-infected insect cells, Protein Expr. Purif. 66 (2009) 7–14. [53] J.Y. Lee, L.N. Kinch, D.M. Borek, J. Wang, J. Wang, I.L. Urbatsch, X.S. Xie, N.V. Grishin, J.C. Cohen, Z. Otwinowski, H.H. Hobbs, D.M. Rosenbaum, Crystal structure of the human sterol transporter ABCG5/ABCG8, Nature (2016). [54] C.A. Shintre, A.C. Pike, Q. Li, J.I. Kim, A.J. Barr, S. Goubin, L. Shrestha, J. Yang, G. Berridge, J. Ross, P.J. Stansfeld, M.S. Sansom, A.M. Edwards, C. Bountra, B.D. Marsden, F. von Delft, A.N. Bullock, O. Gileadi, N.A. Burgess-Brown, E.P. Carpenter, Structures of ABCB10, a human ATP-binding cassette transporter in apo- and nucleotide-bound states, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 9710–9715. [55] G.A. Graf, L. Yu, W.P. Li, R. Gerard, P.L. Tuma, J.C. Cohen, H.H. Hobbs, ABCG5 and ABCG8 are obligate heterodimers for protein trafficking and biliary cholesterol excretion, J. Biol. Chem. 278 (2003) 48275–48282. [56] Y. Kim, P. Quartey, H. Li, L. Volkart, C. Hatzos, C. Chang, B. Nocek, M. Cuff, J. Osipiuk, K. Tan, Y. Fan, L. Bigelow, N. Maltseva, R. Wu, M. Borovilos, E. Duggan, M. Zhou, T.A. Binkowski, R.G. Zhang, A. Joachimiak, Large-scale evaluation of protein reductive methylation for improving protein crystallization, Nat. Methods 5 (2008) 853–854. [57] J. Fantini, F.J. Barrantes, How cholesterol interacts with membrane proteins: an exploration of cholesterol-binding sites including CRAC, CARC, and tilted domains, Front. Physiol. 4 (2013) 31. [58] L.J. Sharpe, G. Rao, P.M. Jones, E. Glancey, S.M. Aleidi, A.M. George, A.J. Brown, I.C. Gelissen, Cholesterol sensing by the ABCG1 lipid transporter: Requirement of a CRAC motif in the final transmembrane domain, Biochim. Biophys. Acta 1851 (2015) 956–964. [59] Z. Gal, C. Hegedus, G. Szakacs, A. Varadi, B. Sarkadi, C. Ozvegy-Laczka, Mutations of the central tyrosines of putative cholesterol recognition amino acid consensus (CRAC) sequences modify folding, activity, and sterol-sensing of the human ABCG2 multidrug transporter, Biochim. Biophys. Acta 1848 (2015) 477–487. [60] V. Hsieh, M.J. Kim, I.C. Gelissen, A.J. Brown, C. Sandoval, J.C. Hallab, M. Kockx, M. Traini, W. Jessup, L. Kritharides, Cellular cholesterol regulates ubiquitination and degradation of the cholesterol export proteins ABCA1 and ABCG1, J. Biol. Chem. 289 (2014) 7524–7536. [61] B. Woebking, S. Velamakanni, L. Federici, M.A. Seeger, S. Murakami, H.W. van Veen, Functional role of transmembrane helix 6 in drug binding and transport by the ABC transporter MsbA, Biochemistry 47 (2008) 10904–10914. [62] J. Cramer, S. Kopp, S.E. Bates, P. Chiba, G.F. Ecker, Multispecificity of drug transporters: probing inhibitor selectivity for the human drug efflux transporters ABCB1 and ABCG2, ChemMedChem 2 (2007) 1783–1788. [63] P. Krishnamurthy, J.D. Schuetz, Role of ABCG2/BCRP in biology and medicine, Annu. Rev. Pharmacol. Toxicol. 46 (2006) 381–410. [64] M.J. Szafraniec, M. Szczygiel, K. Urbanska, L. Fiedor, Determinants of the activity and substrate recognition of breast cancer resistance protein (ABCG2), Drug Metab. Rev. 46 459-474. [65] S. Velamakanni, T. Janvilisri, S. Shahi, H.W. van Veen, A functional steroid-binding element in an ATP-binding cassette multidrug transporter, Mol. Pharmacol. 73 (2008) 12–17. [66] A. Telbisz, C. Hegedus, A. Varadi, B. Sarkadi, C. Ozvegy-Laczka, Regulation of the function of the human ABCG2 multidrug transporter by cholesterol and bile acids: effects of mutations in potential substrate and steroid binding sites, Drug Metab. Dispos. 42 (2014) 575–585. [67] A. Telbisz, M. Muller, C. Ozvegy-Laczka, L. Homolya, L. Szente, A. Varadi, B. Sarkadi, Membrane cholesterol selectively modulates the activity of the human ABCG2 multidrug transporter, Biochim. Biophys. Acta 1768 (2007) 2698–2713. [68] A. Telbisz, C. Ozvegy-Laczka, T. Hegedus, A. Varadi, B. Sarkadi, Effects of the lipid environment, cholesterol and bile acids on the function of the purified and reconstituted human ABCG2 protein, Biochem. J. 450 (2013) 387–395.

[69] C. Ozvegy-Laczka, G. Koblos, B. Sarkadi, A. Varadi, Single amino acid (482) variants of the ABCG2 multidrug transporter: major differences in transport capacity and substrate recognition, Biochim. Biophys. Acta 1668 (2005) 53–63. [70] C.W. van Roermund, W.F. Visser, L. Ijlst, A. van Cruchten, M. Boek, W. Kulik, H.R. Waterham, R.J. Wanders, The human peroxisomal ABC half transporter ALDP functions as a homodimer and accepts acyl-CoA esters, FASEB J. Off. Publ. Feder. Amer. Socie. Exper. Biol. 22 (2008) 4201–4208. [71] M. Morita, T. Imanaka, Peroxisomal ABC transporters: structure, function and role in disease, Biochim. Biophys. Acta 1822 (2012) 1387–1396. [72] H.W. Moser, A. Mahmood, G.V. Raymond, X-linked adrenoleukodystrophy, Nat. Clin. Pract. Neurol. 3 (2007) 140–151. [73] C.W. van Roermund, W.F. Visser, L. Ijlst, H.R. Waterham, R.J. Wanders, Differential substrate specificities of human ABCD1 and ABCD2 in peroxisomal fatty acid betaoxidation, Biochim. Biophys. Acta 1811 (2011) 148–152. [74] E.C. Genin, F. Geillon, C. Gondcaille, A. Athias, P. Gambert, D. Trompier, S. Savary, Substrate specificity overlap and interaction between adrenoleukodystrophy protein (ALDP/ABCD1) and adrenoleukodystrophy-related protein (ALDRP/ABCD2), J. Biol. Chem. 286 (2011) 8075–8084. [75] C. De Marcos Lousa, C.W. van Roermund, V.L. Postis, D. Dietrich, I.D. Kerr, R.J. Wanders, S.A. Baldwin, A. Baker, F.L. Theodoulou, Intrinsic acyl-CoA thioesterase activity of a peroxisomal ATP binding cassette transporter is required for transport and metabolism of fatty acids, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 1279–1284. [76] F.L. Theodoulou, D.J. Carrier, T.A. Schaedler, S.A. Baldwin, A. Baker, How to move an amphipathic molecule across a lipid bilayer: different mechanisms for different ABC transporters? Biochem. Soc. Trans. 44 (2016) 774–782. [77] J.P. Andersen, A.L. Vestergaard, S.A. Mikkelsen, L.S. Mogensen, M. Chalat, R.S. Molday, P4-ATPases as Phospholipid Flippases-Structure, Function, and Enigmas, Front. Physiol. 7 (2016) 275. [78] J.D. Brunner, N.K. Lim, S. Schenck, A. Duerst, R. Dutzler, X-ray structure of a calcium-activated TMEM16 lipid scramblase, Nature 516 (2014) 207–212. [79] C. Perez, S. Gerber, J. Boilevin, M. Bucher, T. Darbre, M. Aebi, J.L. Reymond, K.P. Locher, Structure and mechanism of an active lipid-linked oligosaccharide flippase, Nature 524 (2015) 433–438. [80] E.E. Biswas-Fiss, S. Affet, M. Ha, S.B. Biswas, Retinoid binding properties of nucleotide binding domain 1 of the Stargardt disease-associated ATP binding cassette (ABC) transporter, ABCA4, J. Biol. Chem. 287 (2012) 44097–44107. [81] J. Ahn, J.T. Wong, R.S. Molday, The effect of lipid environment and retinoids on the ATPase activity of ABCR, the photoreceptor ABC transporter responsible for Stargardt macular dystrophy, J. Biol. Chem. 275 (2000) 20399–20405. [82] H. Sun, R.S. Molday, J. Nathans, Retinal stimulates ATP hydrolysis by purified and reconstituted ABCR, the photoreceptor-specific ATP-binding cassette transporter responsible for Stargardt disease, J. Biol. Chem. 274 (1999) 8269–8281. [83] J. Weng, N.L. Mata, S.M. Azarian, R.T. Tzekov, D.G. Birch, G.H. Travis, Insights into the function of Rim protein in photoreceptors and etiology of Stargardt's disease from the phenotype in abcr knockout mice, Cell 98 (1999) 13–23. [84] R.S. Molday, M. Zhong, F. Quazi, The role of the photoreceptor ABC transporter ABCA4 in lipid transport and Stargardt macular degeneration, Biochim. Biophys. Acta 1791 (2009) 573–583. [85] S.V. Ambudkar, I.H. Lelong, J. Zhang, C. Cardarelli, Purification and reconstitution of human P-glycoprotein, Methods Enzymol. 292 (1998) 492–504. [86] H. Hirayama, Y. Kimura, N. Kioka, M. Matsuo, K. Ueda, ATPase activity of human ABCG1 is stimulated by cholesterol and sphingomyelin, J. Lipid Res. 54 (2012) 496–502. [87] N. Infed, N. Hanekop, A.J. Driessen, S.H. Smits, L. Schmitt, Influence of detergents on the activity of the ABC transporter LmrA, Biochim. Biophys. Acta 1808 (2011) 2313–2321. [88] F.J. Sharom, Characterization and functional reconstitution of the multidrug transporter, J. Bioenerg. Biomembr. 27 (1995) 15–22. [89] S.V. Ambudkar, I.H. Lelong, J. Zhang, C.O. Cardarelli, M.M. Gottesman, I. Pastan, Partial purification and reconstitution of the human multidrug-resistance pump: characterization of the drug-stimulatable ATP hydrolysis, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 8472–8476. [90] A. de Athayde Moncorvo Collado, N. Corbalan, L. Homolya, R. Morero, C. Minahk, Resveratrol modulates ATPase activity of liposome-reconstituted ABCG1, FEBS Lett. 587 (2013) 2359–2363. [91] T.K. Ritchie, Y.V. Grinkova, T.H. Bayburt, I.G. Denisov, J.K. Zolnerciks, W.M. Atkins, S.G. Sligar, Chapter 11 - Reconstitution of membrane proteins in phospholipid bilayer nanodiscs, Methods Enzymol. 464 (2009) 211–231. [92] A. Karasawa, L.J. Swier, M.C. Stuart, J. Brouwers, B. Helms, B. Poolman, Physicochemical factors controlling the activity and energy coupling of an ionic strength-gated ATP-binding cassette (ABC) transporter, J. Biol. Chem. 288 (2013) 29862–29871. [93] T. Kawai, J.M. Caaveiro, R. Abe, T. Katagiri, K. Tsumoto, Catalytic activity of MsbA reconstituted in nanodisc particles is modulated by remote interactions with the bilayer, FEBS Lett. 585 (2011) 3533–3537. [94] W.T. Doerrler, C.R. Raetz, ATPase activity of the MsbA lipid flippase of Escherichia coli, J. Biol. Chem. 277 (2002) 36697–36705. [95] A. Polissi, C. Georgopoulos, Mutational analysis and properties of the msbA gene of Escherichia coli, coding for an essential ABC family transporter, Mol. Microbiol. 20 (1996) 1221–1233. [96] C. Scholz, D. Parcej, C.S. Ejsing, H. Robenek, I.L. Urbatsch, R. Tampe, Specific lipids modulate the transporter associated with antigen processing (TAP), J. Biol. Chem. 286 (2011) 13346–13356. [97] G. Kuhnke, K. Neumann, U. Muhlenhoff, R. Lill, Stimulation of the ATPase activity of the yeast mitochondrial ABC transporter Atm1p by thiol compounds, Mol. Membr. Biol. 23 (2006) 173–184. [98] R. Reich-Slotky, C. Panagiotidis, M. Reyes, H.A. Shuman, The detergent-soluble maltose transporter is activated by maltose binding protein and verapamil, J. Bacteriol. 182 (2000) 993–1000. [99] F.J. Alvarez, C. Orelle, A.L. Davidson, Functional reconstitution of an ABC transporter in nanodiscs for use in electron paramagnetic resonance spectroscopy, J. Am. Chem. Soc. 132 (2010) 9513–9515.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx [100] E.L. Borths, B. Poolman, R.N. Hvorup, K.P. Locher, D.C. Rees, In vitro functional characterization of BtuCD-F, the Escherichia coli ABC transporter for vitamin B12 uptake, Biochemistry 44 (2005) 16301–16309. [101] A.J. Rice, F.J. Alvarez, A.L. Davidson, H.W. Pinkett, Effects of lipid environment on the conformational changes of an ABC importer, Channels (Austin) 8 (2014) 327–333. [102] H. Kaur, A. Lakatos, R. Spadaccini, R. Vogel, C. Hoffmann, J. Becker-Baldus, O. Ouari, P. Tordo, H. McHaourab, C. Glaubitz, The ABC exporter MsbA probed by solid state NMR - challenges and opportunities, Biol. Chem. 396 (2015) 1135–1149. [103] C. Bechara, A. Noll, N. Morgner, M.T. Degiacomi, R. Tampe, C.V. Robinson, A subset of annular lipids is linked to the flippase activity of an ABC transporter, Nat. Chem. 7 (2015) 255–262. [104] C.A. Doige, X. Yu, F.J. Sharom, The effects of lipids and detergents on ATPase-active P-glycoprotein, Biochim. Biophys. Acta 1146 (1993) 65–72. [105] R. Callaghan, G. Berridge, D.R. Ferry, C.F. Higgins, The functional purification of Pglycoprotein is dependent on maintenance of a lipid-protein interface, Biochim. Biophys. Acta 1328 (1997) 109–124. [106] S. Faham, J.U. Bowie, Bicelle crystallization: a new method for crystallizing membrane proteins yields a monomeric bacteriorhodopsin structure, J. Mol. Biol. 316 (2002) 1–6. [107] E.M. Landau, J.P. Rosenbusch, Lipidic cubic phases: a novel concept for the crystallization of membrane proteins, Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 14532–14535. [108] M. Caffrey, D. Li, A. Dukkipati, Membrane protein structure determination using crystallography and lipidic mesophases: recent advances and successes, Biochemistry 51 (2012) 6266–6288. [109] V. Cherezov, Lipidic cubic phase technologies for membrane protein structural studies, Curr. Opin. Struct. Biol. 21 (2011) 559–566. [110] D.Y. Lin, S. Huang, J. Chen, Crystal structures of a polypeptide processing and secretion transporter, Nature 523 (2015) 425–430. [111] S. Chen, M.L. Oldham, A.L. Davidson, J. Chen, Carbon catabolite repression of the maltose transporter revealed by X-ray crystallography, Nature 499 (2013) 364–368. [112] M.S. Jin, M.L. Oldham, Q. Zhang, J. Chen, Crystal structure of the multidrug transporter P-glycoprotein from Caenorhabditis elegans, Nature 490 (2012) 566–569. [113] J. Li, K.F. Jaimes, S.G. Aller, Refined structures of mouse P-glycoprotein, Protein Sci. 23 (2014) 34–46. [114] N.P. Barrera, S.C. Isaacson, M. Zhou, V.N. Bavro, A. Welch, T.A. Schaedler, M.A. Seeger, R.N. Miguel, V.M. Korkhov, H.W. van Veen, H. Venter, A.R. Walmsley, C.G. Tate, C.V. Robinson, Mass spectrometry of membrane transporters reveals subunit stoichiometry and interactions, Nat. Methods 6 (2009) 585–587. [115] S. Velamakanni, C.H. Lau, D.A. Gutmann, H. Venter, N.P. Barrera, M.A. Seeger, B. Woebking, D. Matak-Vinkovic, L. Balakrishnan, Y. Yao, U. E.C., R.A. Shilling, C.V. Robinson, P. Thorn, H.W. van Veen, A multidrug ABC transporter with a taste for salt, PLoS ONE 4 (2009) e6137. [116] C. Bechara, C.V. Robinson, Different modes of lipid binding to membrane proteins probed by mass spectrometry, J. Am. Chem. Soc. 137 (2015) 5240–5247. [117] N.P. Barrera, N. Di Bartolo, P.J. Booth, C.V. Robinson, Micelles protect membrane complexes from solution to vacuum, Science 321 (2008) 243–246. [118] V.M. Korkhov, S.A. Mireku, D.B. Veprintsev, K.P. Locher, Structure of AMP-PNPbound BtuCD and mechanism of ATP-powered vitamin B12 transport by BtuCDF, Nat. Struct. Mol. Biol. 21 (2014) 1097–1099. [119] B. Joseph, G. Jeschke, B.A. Goetz, K.P. Locher, E. Bordignon, Transmembrane gate movements in the type II ATP-binding cassette (ABC) importer BtuCD-F during nucleotide cycle, J. Biol. Chem. 286 (2011) 41008–41017. [120] J. Marcoux, S.C. Wang, A. Politis, E. Reading, J. Ma, P.C. Biggin, M. Zhou, H. Tao, Q. Zhang, G. Chang, N. Morgner, C.V. Robinson, Mass spectrometry reveals synergistic effects of nucleotides, lipids, and drugs binding to a multidrug resistance efflux pump, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 9704–9709. [121] A. Zutz, J. Hoffmann, U.A. Hellmich, C. Glaubitz, B. Ludwig, B. Brutschy, R. Tampe, Asymmetric ATP hydrolysis cycle of the heterodimeric multidrug ABC transport complex TmrAB from Thermus thermophilus, J. Biol. Chem. 286 (2011) 7104–7115. [122] Y. Gao, E. Cao, D. Julius, Y. Cheng, TRPV1 structures in nanodiscs reveal mechanisms of ligand and lipid action, Nature 534 (2016) 347–351. [123] J. Du, W. Lu, S. Wu, Y. Cheng, E. Gouaux, Glycine receptor mechanism elucidated by electron cryo-microscopy, Nature 526 (2015) 224–229. [124] J. Kim, S. Wu, T.M. Tomasiak, C. Mergel, M.B. Winter, S.B. Stiller, Y. RoblesColmanares, R.M. Stroud, R. Tampe, C.S. Craik, Y. Cheng, Subnanometreresolution electron cryomicroscopy structure of a heterodimeric ABC exporter, Nature 517 (2014) 396–400. [125] M.L. Oldham, R.K. Hite, A.M. Steffen, E. Damko, Z. Li, T. Walz, J. Chen, A mechanism of viral immune evasion revealed by cryo-EM analysis of the TAP transporter, Nature 529 (2016) 537–540. [126] S. Eggensperger, O. Fisette, D. Parcej, L.V. Schafer, R. Tampe, An annular lipid belt is essential for allosteric coupling and viral inhibition of the antigen translocation complex TAP (transporter associated with antigen processing), J. Biol. Chem. 289 (2014) 33098–33108. [127] R.L. Juliano, V. Ling, A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants, Biochim. Biophys. Acta 455 (1976) 152–162. [128] I.L. Urbatsch, A.E. Senior, Effects of lipids on ATPase activity of purified Chinese hamster P-glycoprotein, Arch. Biochem. Biophys. 316 (1995) 135–140. [129] G. Reuter, T. Janvilisri, H. Venter, S. Shahi, L. Balakrishnan, H.W. van Veen, The ATP binding cassette multidrug transporter LmrA and lipid transporter MsbA have overlapping substrate specificities, J. Biol. Chem. 278 (2003) 35193–35198. [130] J.M. Zamora, H.L. Pearce, W.T. Beck, Physical-chemical properties shared by compounds that modulate multidrug resistance in human leukemic cells, Mol. Pharmacol. 33 (1988) 454–462. [131] A. Seelig, A general pattern for substrate recognition by P-glycoprotein, Eur. J. Biochem. 251 (1998) 252–261. [132] K. Ueda, N. Okamura, M. Hirai, Y. Tanigawara, T. Saeki, N. Kioka, T. Komano, R. Hori, Human P-glycoprotein transports cortisol, aldosterone, and dexamethasone, but not progesterone, J. Biol. Chem. 267 (1992) 24248–24252.

13

[133] A. Seelig, X.L. Blatter, F. Wohnsland, Substrate recognition by P-glycoprotein and the multidrug resistance-associated protein MRP1: a comparison, Int. J. Clin. Pharmacol. Ther. 38 (2000) 111–121. [134] A. Seelig, G. Gerebtzoff, Enhancement of drug absorption by noncharged detergents through membrane and P-glycoprotein binding, Expert Opin. Drug Metab. Toxicol. 2 (2006) 733–752. [135] C.F. Higgins, M.M. Gottesman, Is the multidrug transporter a flippase? Trends Biochem. Sci. 17 (1992) 18–21. [136] H. Omote, M.K. Al-Shawi, A novel electron paramagnetic resonance approach to determine the mechanism of drug transport by P-glycoprotein, J. Biol. Chem. 277 (2002) 45688–45694. [137] A. Seelig, E. Landwojtowicz, Structure-activity relationship of P-glycoprotein substrates and modifiers, Eur. J. Pharm. Sci. 12 (2000) 31–40. [138] E.E. Chufan, K. Kapoor, S.V. Ambudkar, Drug-protein hydrogen bonds govern the inhibition of the ATP hydrolysis of the multidrug transporter P-glycoprotein, Biochem. Pharmacol. 101 (2016) 40–53. [139] A.T. Clay, F.J. Sharom, Lipid bilayer properties control membrane partitioning, binding, and transport of p-glycoprotein substrates, Biochemistry 52 (2013) 343–354. [140] Y. Romsicki, F.J. Sharom, The membrane lipid environment modulates drug interactions with the P-glycoprotein multidrug transporter, Biochemistry 38 (1999) 6887–6896. [141] R. Callaghan, A. Stafford, R.M. Epand, Increased accumulation of drugs in a multidrug resistant cell line by alteration of membrane biophysical properties, Biochim. Biophys. Acta 1175 (1993) 277–282. [142] A. Siarheyeva, J.J. Lopez, C. Glaubitz, Localization of multidrug transporter substrates within model membranes, Biochemistry 45 (2006) 6203–6211. [143] H.A. Scheidt, A. Pampel, L. Nissler, R. Gebhardt, D. Huster, Investigation of the membrane localization and distribution of flavonoids by high-resolution magic angle spinning NMR spectroscopy, Biochim. Biophys. Acta 1663 (2004) 97–107. [144] C.B. Fox, R.A. Horton, J.M. Harris, Detection of drug-membrane interactions in individual phospholipid vesicles by confocal Raman microscopy, Anal. Chem. 78 (2006) 4918–4924. [145] H. Maswadeh, C. Demetzos, I. Daliani, I. Kyrikou, T. Mavromoustakos, A. Tsortos, G. Nounesis, A molecular basis explanation of the dynamic and thermal effects of vinblastine sulfate upon dipalmitoylphosphatidylcholine bilayer membranes, Biochim. Biophys. Acta 1567 (2002) 49–55. [146] E.D. de Araujo, C.P. Alvarez, J.P. Lopez-Alonso, C.R. Sooklal, M. Stagljar, V. Kanelis, Phosphorylation-dependent changes in nucleotide binding, conformation, and dynamics of the first nucleotide binding domain (NBD1) of the sulfonylurea receptor 2B (SUR2B), J. Biol. Chem. 290 (2015) 22699–22714. [147] U.A. Hellmich, E. Duchardt-Ferner, C. Glaubitz, J. Wohnert, Backbone NMR resonance assignments of the nucleotide binding domain of the ABC multidrug transporter LmrA from Lactococcus lactis in its ADP-bound state, Biomol. NMR Assign. 6 (2012) 69–73. [148] A. Siarheyeva, J.J. Lopez, I. Lehner, U.A. Hellmich, H.W. van Veen, C. Glaubitz, Probing the molecular dynamics of the ABC multidrug transporter LmrA by deuterium solid-state nuclear magnetic resonance, Biochemistry 46 (2007) 3075–3083. [149] U.A. Hellmich, L. Monkemeyer, S. Velamakanni, H.W. van Veen, C. Glaubitz, Effects of nucleotide binding to LmrA: A combined MAS-NMR and solution NMR study, Biochim. Biophys. Acta 1848 (2015) 3158–3165. [150] B. Kunert, C. Gardiennet, D. Lacabanne, D. Calles-Garcia, P. Falson, J.M. Jault, B.H. Meier, F. Penin, A. Bockmann, Efficient and stable reconstitution of the ABC transporter BmrA for solid-state NMR studies, Front. Mol. Biosci. 1 (2014) 5. [151] V. Lange, J. Becker-Baldus, B. Kunert, B.J. van Rossum, F. Casagrande, A. Engel, Y. Roske, F.M. Scheffel, E. Schneider, H. Oschkinat, A MAS NMR study of the bacterial ABC transporter ArtMP, Chembiochem 11 (2010) 547–555. [152] Y. Romsicki, F.J. Sharom, Interaction of P-glycoprotein with defined phospholipid bilayers: a differential scanning calorimetric study, Biochemistry 36 (1997) 9807–9815. [153] U.A. Hellmich, S. Lyubenova, E. Kaltenborn, R. Doshi, H.W. van Veen, T.F. Prisner, C. Glaubitz, Probing the ATP hydrolysis cycle of the ABC multidrug transporter LmrA by pulsed EPR spectroscopy, J. Am. Chem. Soc. 134 (2012) 5857–5862. [154] S.J. Ullrich, U.A. Hellmich, S. Ullrich, C. Glaubitz, Interfacial enzyme kinetics of a membrane bound kinase analyzed by real-time MAS-NMR, Nat. Chem. Biol. 7 (2011) 263–270. [155] D.D. Mowrey, T. Cui, Y. Jia, D. Ma, A.M. Makhov, P. Zhang, P. Tang, Y. Xu, Openchannel structures of the human glycine receptor alpha1 full-length transmembrane domain, Structure 21 (2013) 1897–1904. [156] U.A. Hellmich, W. Haase, S. Velamakanni, H.W. van Veen, C. Glaubitz, Caught in the act: ATP hydrolysis of an ABC-multidrug transporter followed by real-time magic angle spinning NMR, FEBS Lett. 582 (2008) 3557–3562. [157] T. Stockner, A. Mullen, F. MacMillan, Investigating the dynamic nature of the ABC transporters: ABCB1 and MsbA as examples for the potential synergies of MD theory and EPR applications, Biochem. Soc. Trans. 43 (2015) 1023–1032. [158] U.A. Hellmich, C. Glaubitz, NMR and EPR studies of membrane transporters, Biol. Chem. 390 (2009) 815–834. [159] P.P. Borbat, K. Surendhran, M. Bortolus, P. Zou, J.H. Freed, H.S. McHaourab, Conformational motion of the ABC transporter MsbA induced by ATP hydrolysis, PLoS Biol. 5 (2007) e271. [160] P. Zou, M. Bortolus, H.S. McHaourab, Conformational cycle of the ABC transporter MsbA in liposomes: detailed analysis using double electron-electron resonance spectroscopy, J. Mol. Biol. 393 (2009) 586–597. [161] P. Zou, H.S. McHaourab, Increased sensitivity and extended range of distance measurements in spin-labeled membrane proteins: Q-band double electron-electron resonance and nanoscale bilayers, Biophys. J. 98 (2010) L18–L20. [162] R.S. Cooper, G.A. Altenberg, Association/Dissociation of the Nucleotide-binding Domains of the ATP-binding Cassette Protein MsbA Measured during Continuous Hydrolysis, J. Biol. Chem. 288 (2013) 20785–20796. [163] M.E. Zoghbi, R.S. Cooper, G.A. Altenberg, The Lipid Bilayer Modulates the Structure and Function of an ATP-binding Cassette Exporter, J. Biol. Chem. 291 (2016) 4453–4461.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023

14

J. Neumann et al. / Biochimica et Biophysica Acta xxx (2016) xxx–xxx

[164] P.R. Selvin, T.M. Rana, J.E. Hearst, Luminescence Resonance Energy-Transfer, J. Am. Chem. Soc. 116 (1994) 6029–6030. [165] P.D. Eckford, F.J. Sharom, Functional characterization of Escherichia coli MsbA: interaction with nucleotides and substrates, J. Biol. Chem. 283 (2008) 12840–12850. [166] J.Y. Lee, J.G. Yang, D. Zhitnitsky, O. Lewinson, D.C. Rees, Structural basis for heavy metal detoxification by an Atm1-type ABC exporter, Science 343 (2014) 1133–1136. [167] A. Moeller, S.C. Lee, H. Tao, J.A. Speir, G. Chang, I.L. Urbatsch, C.S. Potter, B. Carragher, Q. Zhang, Distinct conformational spectrum of homologous multidrug ABC transporters, Structure 23 (2015) 450–460. [168] G.A. Frank, S. Shukla, P. Rao, M.J. Borgnia, A. Bartesaghi, A. Merk, A. Mobin, L. Esser, L.A. Earl, M.M. Gottesman, D. Xia, S.V. Ambudkar, S. Subramaniam, Cryo-EM Analysis of the Conformational Landscape of Human P-glycoprotein (ABCB1) During its Catalytic Cycle, Mol. Pharmacol. 90 (2016) 35–41. [169] J.Y. Lee, I.L. Urbatsch, A.E. Senior, S. Wilkens, Projection structure of P-glycoprotein by electron microscopy - Evidence for a closed conformation of the nucleotide binding domains, J. Biol. Chem. 277 (2002) 40125–40131. [170] R.J. Dawson, K.P. Locher, Structure of a bacterial multidrug ABC transporter, Nature 443 (2006) 180–185. [171] A. Ward, C.L. Reyes, J. Yu, C.B. Roth, G. Chang, Flexibility in the ABC transporter MsbA: Alternating access with a twist, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 19005–19010. [172] A. Siarheyeva, F.J. Sharom, The ABC transporter MsbA interacts with lipid A and amphipathic drugs at different sites, Biochem. J. 419 (2009) 317–328.

[173] P.D.W. Eckford, F.J. Sharom, The reconstituted Escherichia coli MsbA protein displays lipid flippase activity, Biochem. J. 429 (2010) 195–203. [174] P.C. Smith, N. Karpowich, L. Millen, J.E. Moody, J. Rosen, P.J. Thomas, J.F. Hunt, ATP binding to the motor domain from an ABC transporter drives formation of a nucleotide sandwich dimer, Mol. Cell 10 (2002) 139–149. [175] J.K. Zolnerciks, E.J. Andress, M. Nicolaou, K.J. Linton, Structure of ABC transporters, Essays Biochem. 50 (2011) 43–61. [176] I. Langmuir, The Shapes of Group Molecules Forming the Surfaces of Liquids, Proc. Natl. Acad. Sci. U. S. A. 3 (1917) 251–257. [177] G. Brannigan, P.F. Philips, F.L. Brown, Flexible lipid bilayers in implicit solvent, Phys. Rev. E Stat. Nonlinear Soft Matter Phys. 72 (2005) 011915. [178] A. Vogel, K.T. Tan, H. Waldmann, S.E. Feller, M.F. Brown, D. Huster, Flexibility of ras lipid modifications studied by 2H solid-state NMR and molecular dynamics simulations, Biophys. J. 93 (2007) 2697–2712. [179] J. Wang, N. Grishin, L. Kinch, J.C. Cohen, H.H. Hobbs, X.S. Xie, Sequences in the nonconsensus nucleotide-binding domain of ABCG5/ABCG8 required for sterol transport, J. Biol. Chem. 286 (2011) 7308–7314. [180] N. Grossmann, A.S. Vakkasoglu, S. Hulpke, R. Abele, R. Gaudet, R. Tampe, Mechanistic determinants of the directionality and energetics of active export by a heterodimeric ABC transporter, Nat. Commun. 5 (2014) 5419. [181] R. Ernst, P. Kueppers, C.M. Klein, T. Schwarzmueller, K. Kuchler, L. Schmitt, A mutation of the H-loop selectively affects rhodamine transport by the yeast multidrug ABC transporter Pdr5, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 5069–5074.

Please cite this article as: J. Neumann, et al., Diverse relations between ABC transporters and lipids: An overview, Biochim. Biophys. Acta (2016), http://dx.doi.org/10.1016/j.bbamem.2016.09.023