How cholesterol interacts with proteins and lipids during its intracellular transport Academic research paper on "Biological sciences"

PII: DOI:

Reference:

Daniel WUstner, Katarzyna Solanko

How cholesterol interacts with proteins and lipids during its intracellular transport

Accepted Manuscript

S0005-2736(15)00158-3

doi: 10.1016/j.bbamem.2015.05.010

BBAMEM 81904

To appear in: BBA - Biomembranes

Received date: 4 January 2015 Revised date: 14 April 2015

Accepted date: 13 May 2015

Please cite this article as: Daniel Wiistner, Katarzyna Solanko, How cholesterol interacts with proteins and lipids during its intracellular transport, BBA - Biomembranes (2015), doi: 10.1016/j.bbamem.2015.05.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

How cholesterol interacts with proteins and lipids during its

intracellular transport

Daniel Wustner and Katarzyna Solanko

department of Biochemistry and Molecular Biology University of Southern Denmark, DK-5230 Odense M, Denmark; Adress correspondence to: Daniel Wustner, Tel. +45-6550-2405, Fax +456550-2405, e-mail: wuestner@bmb.sdu.dk

Keywords: cholesterol, transport, kinetics, membrane, fluorescence, partition, Gibbs free energy, nonvesicular, flux, diffusion, correlation, tracking, molecular structure, endocytosis, feedback regulation

Abstract

Sterols, as cholesterol in mammalian cells and ergosterol in fungi, are indispensable molecules for proper functioning and nanoscale organization of the plasma membrane. Synthesis, uptake and efflux of cholesterol are regulated by a variety of protein-lipid and protein-protein interactions. Similarly, membrane lipids and their physico-chemical properties directly affect cholesterol partitioning and thereby contribute to the highly heterogeneous intracellular cholesterol distribution. Movement of cholesterol in cells is mediated by vesicle trafficking along the endocytic and secretory pathways as well as by non-vesicular sterol exchange between organelles. In this article, we will review recent progress in elucidating sterol-lipid and sterol-protein interactions contributing to proper sterol transport in living cells. We outline recent biophysical models of cholesterol distribution and dynamics in membranes and explain how such models are related to sterol flux between organelles. An overview of various sterol-transfer proteins is given, and the physico-chemical principles of their function in non-vesicular sterol transport are explained. We also discuss selected experimental approaches for characterization of sterol-protein interactions and for monitoring intracellular sterol transport. Finally, we review recent work on the molecular mechanisms underlying lipoprotein-mediated cholesterol import into mammalian cells and describe the process of cellular cholesterol efflux. Overall, we emphasize how specific protein-lipid and protein-protein interactions help overcoming the extremely low water solubility of cholesterol, thereby controlling intracellular cholesterol movement.

1. Introduction

Cholesterol attracts continuous interest of physicians, cell biologists, biochemists and biophysicists due to its uttermost importance in human pathobiology, its central role in regulating cellular functions, its complex metabolism and its impact on membrane structure and dynamics. The last three decades have witnessed a tremendous increase in our understanding of cholesterol transport between tissues and control of cholesterol homeostasis at an organism level [1-3]. Similarly, a lot of effort has been put into characterizing the membrane organization of cholesterol as well as in determining its effect on membrane proteins. In contrast to that progress, the dynamical aspects, energetics, protein-dependence and overall regulation of intracellular cholesterol transport are only beginning to become resolved. Cholesterol in mammalian cells and ergosterol in fungi play a

central role in regulating the permeability barrier and overall structural organization of the plasma membrane (PM) which is why both sterols are highly enriched in that membrane compared to most intracellular organelles. Intermediate cholesterol concentrations are found in early endosomes and the trans-Golgi network (TGN), while mitochondria and the endoplasmic reticulum (ER) harbor only low amounts of cholesterol (i.e. the amount of cholesterol in the ER is below 5% of total relative to the PM with about 60-70% of total). A central question is how the low abundance of cholesterol in the ER is maintained despite the fact that this is the site of cholesterol synthesis and esterification. The ER contains several proteins with essential function in cholesterol metabolism. For example, hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase catalyzes formation of mevalonate in the cytoplasm, which is the rate-limiting step in cholesterol synthesis. Its activity is regulated in a fascinating manner via a protein complex with sterol response element binding protein 2 (SREBP2), its escort proteins Scap and Insig, which all localize to the ER in cells with normal cholesterol levels. Cholesterol homeostasis is regulated in a convergent feedback loop involving these proteins, and sudden changes in ER cholesterol trigger metabolic responses [4-7]. If cholesterol levels in the ER drop, Insig dissociates from Scap/SREBP2 and gets degraded by the proteasome [8]. The Scap/SREBP2 complex is transported in COPII vesicles to the Golgi following normal anterograde membrane traffic, and in the Golgi, two serine proteases cleave the N-terminal domain of SREBP2. This domain travels to the nucleus, where it acts as transcription factor promoting the expression of several genes involved in cholesterol synthesis, uptake and transport [8]. Scap binds cholesterol directly, likely via its sterol sensing domain, while expression of Insig (at least Insig-1) is controlled by the SREBP2 transcription factor. Once normal cholesterol levels are restored and sufficient new Insig is synthesized, the Insig/Scap/SREBP2 complex reforms in the ER, and acquisition of new cholesterol by the cell (via synthesis or uptake of low density lipoprotein (LDL) ceases. Several excellent reviews have been published in the last few years about these processes [8-11], and we will not discuss the details further here. Instead, we continue with a brief overview of experimental approaches for analysis of cholesterol transport (section 2.1. and 2.2). This is especially important due to the fact that our limited knowledge about intracellular cholesterol trafficking is largely a consequence of technical hurdles. We will also briefly discuss several experimental approaches for studying protein-sterol interactions and protein-mediated intermembrane sterol transport (section 2.3.). Kinetic and thermodynamic aspects of cholesterol-lipid interaction and inter-membrane cholesterol transfer are explained in section 3. A survey of specific

protein-lipid and protein-protein interactions during intracellular cholesterol movement, uptake and efflux are provided in section 4.

2. Tools for studying sterol trafficking and sterol-protein interactions

2.1. Quantitative approaches for analysis of intracellular transport processes Intracellular transport studies in general demand specific and robust labeling of the transported entity with minimal alteration of the investigated transport process by the labeling strategy. This aspect is particularly critical for elucidation of lipid transport processes, in which tagging with a reporter moiety can significantly affect the properties of the studied molecule. The experimenter must also decide whether steady state or dynamic information will be gathered in a particular investigation. For example, the polyene macrolide filipin will report about the steady state distribution of cholesterol or other sterols bearing a free 3-hydroxy group, while no kinetic information can be inferred by that approach [12]. Also, filipin staining would not report about the origin of a particular sterol pool. Similar arguments apply to studies using sterol-binding proteins, as fluorescence tagged perifringolysin O (PFO) [12]. Dynamic information about cellular transport can be inferred in one of three ways; I) one deflects the cellular system from its steady state and follows the system response as a function of time. This approach is used in pulse-chase experiments of isotope- or fluorescence-labeled molecules as well as in fluorescence recovery after photobleaching (FRAP) of suitable fluorescent sterols [13, 14]. Pulse-chase studies are also widely used for deciphering lipoprotein-mediated sterol import into cells, and such studies can be easily coupled to metabolic experiments, for example of cholesteryl ester (CE) hydrolysis and reesterification of cholesterol [15] as well as of cholesterol synthesis and transport [16]. II) transport can be studied by creating a permanent sink as a consequence of constantly removing some tagged molecules from the system, as is done in fluorescence loss in photobleaching (FLIP) [17]. In contrast, to the first strategy, the system is permanently disturbed and no new global steady state can be re-established in FLIP studies. III) cells become steady-state labeled with the tagged molecule of interest, for example cholesterol with an attached BODIPY group in the side chain (BChol or TopFluor-cholesterol as trade name) and stochastic fluctuations around the steady state value are monitored as function of time [18, 19]. This approach is only applicable with bright and photostable fluorescent reporter groups allowing for either long term tracking of sterol-rich vesicles or for unbiased analysis of fluorescence fluctuations due to molecular transport [18]. According to

the fluctuation-dissipation theorem, comparable transport rates can be inferred from the response of the system to a perturbation (i.e., strategies I) and II), above) as well as from long-term analysis of fluctuations around a mean value (i.e., strategy III), above) [20]. In the latter case, transport coefficients, as diffusion constants, are derived from time correlation functions, as in fluorescence correlation spectroscopy (FCS) and its imaging variants [21, 22]. Alternatively, statistical ensembles are generated by single particle tracking (SPT) followed by moment analysis of the experimentally determined probability density function (PDF). For example, in SPT the second moment of the step length distribution for recorded trajectories gives the mean square displacement from which diffusion constants or flow speed can be extracted using adequate transport models [2326]. FRAP and SPT have been shown to provide comparable information in case of membrane diffusion [26], while similar transport coefficients were found for intracellular vesicles using either fluorescence fluctuation analysis or SPT [27].

2.2. Radioactive and fluorescent probes for studying sterol transport and metabolism

.... 3

Cholesterol synthesis can be investigated using H-acetate or other isotope-labeled precursors [16,

28]. Cholesterol esterification is often determined by feeding cells H-cholesterol or H-oleate and analyzing the amount of formed radioactive CEs [29, 30]. Isotope-labeled cholesterol is also used in cellular efflux studies for determining protein dependence and kinetics of sterol clearance from cells [31]. Cyclodextrin is often used in such studies, either to introduce the labeled sterols selectively into the plasma membrane or to allow for sterol extraction from cells in efflux experiments [31-33]. Cyclodextrin is similarly used in studies involving fluorescent cholesterol analogs, and cholesterol-containing cyclodextrin is useful for exchanging a labeled sterol pool or for acute cholesterol loading of cells [13, 34, 35]. Uptake of sterol-containing lipoproteins can be followed by

appropriate fluorescence-tagging of the apoprotein or by scintillation chromatography based on I-tagged apoproteins [36, 37]. Hydrolysis of isotope-labeled CEs will, combined with assessment of apoprotein degradation using trichloracetic acid precipitation, report about intracellular degradation

of the lipoprotein [36, 38]. Double-isotope studies using H- or C-sterols are also possible, either to assess re-esterification of LDL-liberated cholesterol or to determine the extent of esterification of different cholesterol pools [39]. Another tool to generate strong concentration gradients is to use hydrophobic amines, such as U18666A, and steroid hormones, as progesterone, to trap cholesterol in late endosomes and lysosomes (LE/LYSs) [15, 40]. Washout of these compounds allows for kinetic and metabolic analysis of the intracellular fate of cholesterol released from these compartments [32, 39]. Use of fluorescent analogs of cholesterol combined with suitable imaging

technology provides an alternative strategy for gaining insight into sterol transport pathways and dynamics in cells. Fluorescent cholesterol analogs bear either a covalently linked fluorescent dye of suitable color and brightness or are intrinsically fluorescent due to a small conjugated system in the steroid backbone of the sterols. Most cholesterol analogs with extrinsic reporter group fail to mimic cholesterol adequately in model membranes and in cellular studies [12, 41]. A somehow reasonable exception is TopFluor-cholesterol which has been successfully used in cells and model organisms, even though it is miss-targeted in cells with high fat content [35, 42]. However, BChol or TopFluor-cholesterol fails to order fatty acyl chains in model membranes, an important property of cholesterol to be mimicked by suitable cholesterol analogs [43, 44]. Ultraviolet (UV)-sensitive wide field (UV-WF) and multiphoton microscopy of intrinsically fluorescent sterols, such as cholestatrienol (CTL) and dehydroergosterol (DHE) as close analogs of cholesterol and ergosterol has provided new insight into cellular sterol trafficking [13, 17, 34, 45, 46]. Their poor fluorescence properties justify the quest for development of further sterol probes. Importantly, metabolic studies on intracellular esterification are also possible with fluorescent sterols [34, 35]. A different tagging strategy is to synthesize alkyne-labeled cholesterol or oxysterols, which after cellular uptake become chemically linked to a fluorescent dye, as BODIPY [47, 48]. A disadvantage here is the need for fixation of cells making live-cell imaging studies impossible. For further details, the reader is referred to recent technical reviews [12, 49-51].

2.3. Biophysical characterization of sterol-protein interactions

Non-vesicular transport of cholesterol and ergosterol in mammalian and yeast cells, respectively depends on cytosolic lipid transfer proteins (LTPs). Non-specific LTPs, as cytosolic sterol carrier protein 2 (SCP2) and fatty acid binding proteins (FABPs) transfer sterols and a variety of other lipids between model and cell membranes and bind their ligands with micromolar affinity [52, 53]. Specific LTPs bind and transfer only one or two types of lipids, sometimes in exchange against each other and often show affinities in the low nanomolar range [54, 55]. Examples of the latter category are members of the family of oxysterol binding proteins (OSBPs), steroidogenic acute regulatory (StAR) proteins and elicitins - plant defensins to which the Niemann Pick C2 protein (NPC2) shows high structural similarity [56-59]. All LTPs were shown to bind their respective ligand and to transfer it between liposomes. In case of sterols, typical binding assays consist of making an often aqueous solution of either radiolabeled or fluorescent sterol and adding the protein in the absence or presence of excess unlabeled sterol [54]. In such competition assays, apparent dissociation constants (KD) for cholesterol or other sterols, as oxysterols, to LTPs have been

determined. Sterol binding to LTPs can be also measured by NMR spectroscopy or by H/D exchange followed by mass spectrometry detection [60, 61]. When using fluorescent analogs, intrinsically fluorescent sterols as DHE or CTL are clearly preferable compared to tagged cholesterol, as NBD-, Dansyl or TopFluor-cholesterol, since binding equilibria are strongly affected by the presence of the fluorophore in the latter analogs [62]. For example, side-chain tagged NBD-cholesterol binds to Niemann Pick C1 protein (NPC1) but not to NPC2 [62]. This is likely a direct consequence of the opposite orientation of the sterol in the respective binding pocket; while in NPC1, sterols bind with their 3'-hydroxy group buried inside the binding pocket [63], in NPC2, the sterol side chain points into the binding region (see section 4.3., below) [64]. Accordingly, the fluorophore in NBD-cholesterol will interfere with binding to NPC2 but not to NPC1. On the other hand, DHE and/or CTL have been used to demonstrate sterol binding of NPC2 [65, 66], FABPs [52], SCP2 [67], NPC1 [62], yeast homologs of OSBPs [68, 69], StAR proteins [14] and elicitins [70]. Photo-crosslinking of a suitable cholesterol analog to the protein of interest can be also used as evidence for sterol-binding capacity [49, 62, 71]. This strategy can be combined with click-chemistry via a side-chain alkyne group to generate a bifunctional cholesterol probe [72]. Combined with stable-isotope labeling by amino acids in cell culture (SILAC) mass spectrometry, such a bifunctional cholesterol probe has been employed to identify over 250 sterol-interacting proteins [72]. Beside many known sterol-binding proteins, several new candidates were identified, which were formerly known only from their function in sugar and glycerolipid metabolism. To determine specific sterol binding to a reconstituted membrane protein, immuno-affinity tagging to sepharose beads has been used followed by adding radioactive cholesterol in a buffer solution, centrifugation of the bead-protein/ligand complex and extraction of bound sterol with organic solvent [73]. Estimation of comparable KD values is often impeded by the low monomeric solubility of cholesterol and its analogs, as DHE, which is far below 100 nM [74, 75]. Accordingly, cholesterol or fluorescent sterols will form crystals or micelles, which, however, are not taken explicitly into account in most published assays reporting KD values. Also, different solubility of cholesterol and its oxidized metabolites can bias the measured affinity values to LTPs [54, 66]. Such problems can be circumvented by inserting the sterol ligand into a supported bilayer followed by detection of protein binding using surface plasmon resonance [68]. Alternatively, the critical concentration of sterol aggregate formation is explicitly taken into account in the estimation of binding equilibria [76].

Binding stoichiometries of sterols to LTPs have been estimated by fluorescence approaches, calorimetry or circular dichroism [52, 53, 76]. Again, cholesterol analogs with extrinsic fluorophore, as NBD-cholesterol gave deviating binding stoichiometry (2:1) compared to native cholesterol (1:1) to the StAR protein, the mitochondrial cholesterol importer [76-78]. Chemical modification of the sterol structure can also have a large impact on inter-membrane sterol transfer catalyzed by LTPs. For example, the BODIPY-group in TopFluor-cholesterol was found to block interbilayer transport of that analog by StARD4, a cytoplasmic sterol transport protein in mammalian cells (Fred Maxfield, Cornell Medical College; personal communication). In contrast, StARD4 transferred DHE with similar efficiency as cholesterol between model membranes and between the PM and the endocytic recycling compartment (ERC) in living CHO cells [14]. Interbilayer exchange of DHE and CTL has become a standard method for determining transfer activity of a variety of LTPs. Two readouts of these sterols can be used for that purpose: (1) relief of Forster resonance energy transfer from DHE or CTL to Dansyl-tagged phosphatidyl ethanolamine (Dansyl-PE). This assay has been originally developed to assess DHE's flip-flop in model membranes by stopped flow-based monitoring of cyclodxtrin-mediated sterol extraction [79]. Two kinetic phases were found in such experiments, and the slow one could be associated with DHE transbilayer migration. Later, this assay was used to monitor DHE and CTL transfer between membranes by a variety of LTPs including NPC2 [80], StARD4 [14], the yeast OSBP homologue Osh 4 [69], OSBP [81] and OSBP related protein 9 (ORP9) [82]. (2) DHE shows a concentration-dependent decline in fluorescence polarization or anisotropy which has been used to predict sterol transbilayer dimers at low sterol concentration [75, 83]. This property was used first by Schroeder and colleagues to quantify sterol transport by SCP2 and FABPs between model and cellular membranes [84, 85]. Similarly, Slotte and colleagues has used anisotropy as readout for equilibrium partitioning of CTL between membranes [86]. Both assays do not depend on separation of donor and acceptor vesicles, a problem which otherwise has been tackled, when using radioactive or unlabeled cholesterol [87, 88].

3. Modes and models of intracellular cholesterol transport

3.1. Is the heterogeneous distribution of cholesterol in cells based on equilibrium thermodynamics? More than 30 years ago, Silbert and colleagues found that isotope-labeled cholesterol resides preferentially in purified PM fractions compared to membrane fractions made of ER or

mitochondria in LM cells [89]. The same tendency (but not the same numbers) was found in lipid extracts made from these membrane fractions. Given the highly different phospho- and sphingolipid composition of PM versus ER and mitochondria, the authors speculated that the properties of the host lipid species in a particular organelle determine the different membrane affinities of cholesterol (and related sterols) and consequently the heterogeneous intracellular cholesterol distribution [89]. This is plausible, given that cholesterol clearly shows preferred interaction with membranes consisting of saturated sphingomyelin (SM) and phosphatidylcholine (PC), as found in the PM compared to membranes made of unsaturated PC and other lipids, as found in the ER [90, 91]. Indeed, the half-time of passive cholesterol transfer between liposomes was found to decrease in the order dipalmitoyl-PC (DPPC; 4.7 hrs), dimyristoyl-PC (DMPC; 3.9 hrs), dioleoyl-PC (DOPC; 2.6 hrs), and soy-bean-PC (1.8 hrs) [92]. Similarly, Mitropoulos and co-workers found a strong correlation between net cholesterol transfer between liposomes and ER membranes purified from rat liver and the activity of ACAT in these microsomes [91]. Thus, the higher the measured partitioning of cholesterol into microsomes was, the larger the activity of ACAT in these membranes. The activation energy for cholesterol transfer from liposomes containing bovine brain SM was about 45% higher than that from liposomes made of egg yolk PC [91]. The transfer kinetics was first order with a half-time of ~28 hours versus ~3 hours at comparable liposome/microsome ratios for SM- and PC-containing membranes, respectively. Given, that bovine brain SM contains much more saturated fatty acyl species than egg yolk PC these observations could be a consequence of the more saturated acyl chain species in membranes made of SM than those made of PC. Alternatively, preferred interaction between cholesterol and SM compared to PC even for similar acyl chain order could explain the observations by Mitropoulos and colleagues [91]. Partition experiments have indeed shown that cholesterol has the highest affinity for liposomes made of SM followed by those made from phosphatidylserine (PS), PC and finally phosphatidylethanolamine (PE) [90]. Similar results were found using CTL by Slotte and colleagues using a fluorescence assay based on CTL [93]. These authors showed additionally that increased membrane order as assessed by diphenylhexatriene fluorescence in SM containing membranes is not the primary cause of higher sterol partitioning into SM compared to PC bilayers [93]. Thus, energetically favorable interfacial properties of SM compared to PC might stabilize cholesterol in sphingolipid containing bilayers. Structural differences in the ceramide compared to the glycerol backbone must account for such properties, as the phosphorylcholine head group of PC and SM is the same. For example, hydrogen bonding between the 3'-hydroxy group of cholesterol

and the NH-group of the ceramide part of SM could stabilize such interactions. Others, however, could not find evidence for a specific interaction of cholesterol with SM [94]. Independent of the molecular details, the observed partition differences of cholesterol for varying head group and acyl chain composition are fully in accordance with equilibrium thermodynamics. Let's consider a two-membrane system; membrane A and membrane B with differing phospholipid composition (for example A contains saturated PC and B contains unsaturated PC) and cholesterol as solute within such membranes. Cholesterol's mole fraction in both membranes reads

xA _ 'nChol J xB _ Chol n

XChol _ A A and XChol _ B B D)

nPhl + nChol nPhl + nChol

Here, n'cho,and riPM with i =A, B correspond to the number of cholesterol and phospholipid

molecules in the respective membrane A and B. Let's assume that cholesterol transfer between both membranes takes place via the aqueous solution, we can write the scheme

kx k2 A \—t water < * B

k, k-,

Here, k\ and k.\ is the desorption and insertion rate constant for cholesterol into membrane A, respectively. The aqueous sterol fraction is indicated by 'water' For membrane B, k2 is the insertion rate constant and k-2 is the corresponding desorption rate constant. Typical experimentally

determined values for DHE desorption from and insertion into POPC membranes are ki = 5.1-10

M- •s' and k-i = M0- s [95]. The equilibrium constants for the individual steps in Eq. 2 are given by q1 _ kjk_1 and q2 _ k2/k_2, respectively. Whether aqueous or collisional transfer dominates for cholesterol is still debated due to cholesterol's low water solubility [87, 96]. For example for DHE, the partition coefficient between water and POPC liposomes has been estimated after normalizing to phospholipid concentration to q=1. 3-106 corresponding to 13 million DHE molecules in the membrane for one molecule in water [95]. At increasing concentration of acceptor membranes, sterol transfer seems to be dictated by the frequency of vesicle collisions [96]. After sufficient time, the cholesterol distribution between membranes A and B will be independent of the aqueous sterol pool, anyway, according to:

water B B k ■ k9

k _ q1 ■ q2 _---_ — _ —11—(3)

A water A k_x ■ k_2

Thus, cholesterol solubility in the aqueous phase might play a role for the kinetics of sterol transfer between membranes but not for the equilibrium distribution. Accordingly, we can ignore an eventual intermediate cholesterol pool between the membranes in what follows. At thermodynamic equilibrium cholesterol's chemical potential, nchol, defined as change in Gibbs free energy due to an infinitesimal change in cholesterol abundance according to the Gibb-Duhem equation, is identical in both membranes, i.e.

A ÔG aO A B ÔG B,0 , 7 rri i B / A\

Vchol = = Mchol + kb ■ T ■ ln Xchol = Mchol = = Vchol + kb ■ T ■ ln XChol (4)

ÔnChol Ônchol

Here, kb is Boltzmann's constant and T is temperature. The partition coefficient of cholesterol between both membranes, assuming ideal mixing, will therefore become

K = ^ = exp

XA>~> (№c'hol Mc'hol)

kb ■ T

Thus, how cholesterol partitions, depends on its chemical potential in the respective membrane under standard conditions. This, in turn depends on the property of the host lipid matrix in the particular membranes. Lipid membranes, however, are not dilute isotropic solvents for cholesterol, but instead crowded, anisotropic media with many internal degrees of freedom. If we want to take such effects into account, we can introduce the activity coefficient, y, leading to a modification of the expression for the chemical potential:

Mahal = ¿Choi + kb ■ T ■ln XChol ■ Tchol (6)

From that, we define an apparent equilibrium constant, K', as:

XChol TChoi js

K= — = -T- ■K (7)

xChol T Choi

The activity coefficient for cholesterol determines the free volume accessible for cholesterol in each membrane [97]. Thus, if for example, Tchoi= 5 ■ Tchoi because the available volume for cholesterol in membrane A made of saturated PC is several-fold that in membrane B made of unsaturated PC, membrane c could accommodate five times the amount of cholesterol compared to membrane c at equilibrium.

For many membrane solutes, the activity coefficient can be directly related to the work required to create a cavity for the molecule in the bilayer, as can be derived from scaled particle theory [97, 98]. However, cholesterol reduces the available free volume in the bilayer, and consequently, lowers drastically the membrane permeability for many solutes [99-101]. So, why would a

membrane consisting of saturated PC molecules accommodate more cholesterol than a membrane made of unsaturated PC? The answer lies in cholesterol's condensing effect on phospholipid bilayers, a well-known phenomenon tracing back to experimental mapping of binary phase diagrams of PC and cholesterol and their theoretical interpretation [102, 103]. Due to its planar, non-polar structure, cholesterol reduces the conformational space available for fatty acyl chains, thereby reducing their area requirement under the PC head groups [103-105]. In membranes made of unsaturated PC, however, higher flexibility of the kinked fatty acyl chains prevents equally tight packing with cholesterol. Consequently, the space requirement of the non-polar bilayer region (i.e. sum of fatty acyl chains and cholesterol) in a membrane is higher, if the host phospholipid is unsaturated PC (e.g. DOPC in membrane B) than if it is saturated PC (e.g. DPPC in membrane A). Since both membranes have the same head group, more cholesterol gets exposed to the interfacial area in membrane B compared to membrane A, thereby making unfavorable contacts with water. Exposure of cholesterol to water lowers the entropy of the water molecules in vicinity of the sterol, which gives a positive contribution to the Gibbs free energy (i.e. the free enthalpy). This is therefore highly unfavorable, and there is a tendency to shield sterols under phospholipid head groups to lower the total free enthalpy. The efficiency of this shielding, though, depends also on the ability of cholesterol to pack with the adjacent phospholipid acyl chains, which in turn are ordered by the presence of cholesterol. The same reasoning applies to other host lipid species, as for example, the amount of sphingolipids will positively correlate with the partition preference of cholesterol in a given membrane [89, 90, 106]. Instead of free-volume theory, as for other membrane solutes, the extent of sterol tilt seems to be an adequate measure for cholesterol's effects on membrane properties [107, 108]. The lower the tilt, i.e. the more cholesterol is aligned with the bilayer normal, the larger is the extent of membrane condensation and the stronger is the interaction with the fatty acyl chains of the host phospho- or sphingolipids. A low tilt means also high mechanical stability and thickening of the membrane, thereby likely affecting protein interaction with the bilayer [107110]. Equally important, a low sterol tilt in the membrane is associated with lowered propensity for sterol flip-flop, which has been suggested to involve rotation of the sterol perpendicular to its long axis [111, 112]. The sterol tilt does not only depend on the host lipid composition but also on the sterol type and mole fraction in the bilayer. In full accordance with that are recent experiments and theoretical studies showing i) faster flip-flop and higher tilt of ketosterols than cholesterol [113], ii) decreasing cholesterol tilt for increasing cholesterol mole fraction in the membrane [108, 114], iii) higher partitioning of cholesterol in cholesterol-rich PC than in cholesterol-poor PC membranes

[111, 115, 116] and iv) decreasing flip-flop rates of cholesterol for increasing membrane cholesterol content [111].

Taken together, the specific properties of the host membrane including phospho- and sphingolipid composition and sterol content determine cholesterol's chemical activity in the respective membrane allowing for differing sterol mole fractions in two bilayers at thermodynamic equilibrium (Eq. 7). Deviation from thermodynamic equilibrium upon a small perturbation will cause a non-vesicular sterol flux between both membranes. This flux is proportional to the difference in chemical potentials between both membranes (i.e. to &tchol= tchoi ~tchoi). Thus, the chemical potential difference of cholesterol between the membranes acts as thermodynamic driving force for sterol flux and this flux vanishes at equilibrium when A^choi approaches zero. This connection between differences in chemical potential and flux is called a linear flux-force relationship in linear irreversible thermodynamics and can be derived from a Taylor expansion of the flux with respect to the force around a reference state. Of course, lipid membranes are anisotropic solvents, causing additional terms in the Gibbs free energy change associated with sterol transfer [98, 114]. In contrast, classical partition experiments upon which the above formalism is based use an oily phase, made of isotropic solvents as alkanes with no preferred orientation of the hydrocarbon chains. In cells, proteins could additionally affect the cholesterol solubility of membranes, as could be shown for transmembrane peptides affecting the partition coefficient of fluorescent sterols as CTL or TopFluor-cholesterol between liposomes [117].

3.2. How do membrane properties of cholesterol relate to its intracellular transport? Several lines of evidence indicate that intracellular sterol flux becomes non-linearly dependent on the membrane sterol mole fraction. For example, above certain threshold concentrations of total cell cholesterol a non-linear raise in cholesterol content in the ER accompanied by a halt in SREBP2 activation and acute proteolytic inactivation of HMG-CoA reductase has been observed [4, 6, 7]. An increase of cellular cholesterol beyond some threshold causes also abrupt stimulation of cholesterol esterification by ACAT [5, 7], sudden acceleration of cholestenone formation by cholesterol oxidase and enhanced extractability of cholesterol with cyclodextrin [7, 118], accelerated DHE influx in ATP-depleted macrophages [34] and increased accessibility of cholesterol to PFO binding [51, 57]. These observations cannot be explained by a linear flux-force relationship, since they point to a non-linear increase of ^chol at critical cholesterol mole fractions.

Three physico-chemical models have been invoked to explain such sudden changes in nChol at critical cholesterol mole fractions. In all models, as in our considerations above, two membranes can have largely differing cholesterol amounts and still be in thermodynamic equilibrium. In the superlattice model, cholesterol is supposed to prefer regular packing geometries in the phospholipid matrix due to a mismatch in the cross-sectional area of the different lipid species [119]. This causes long-range repulsive forces between cholesterol molecules and sudden spikes in ^Chol at critical mole fractions of the constituents, where preferred packing geometries are disturbed [120]. In the umbrella model, cholesterol is believed to avoid contacts with water due to its small polar OH-group resulting in preferred association with certain membrane phospholipids, which can shield the sterol under their head groups [121-123]. Due to increasing order of the fatty acyl chains and consequently condensing of the membrane, the ability to 'hide' cholesterol under the phospholipid head group diminishes as a non-linear function of cholesterol concentration. The umbrella model considers explicitly multi-body interactions in the energy functional and predicts several jumps in ^Chol at particular packing geometries of phospholipids and cholesterol at high sterol mole fractions in the bilayer [121, 122]. Each jump in ^Chol corresponds to a transition from one type of a regular cholesterol distribution to another. In the third model, the condensed complex model, stoichiometric interactions are assumed between cholesterol and phospholipids with saturated fatty acyl chains, called reactive phospholipid (e.g. DPPC), while no complex formation is assumed for unsaturated phospholipids (called 'unreactive; e.g. DOPC) [124]. The expression for the Gibbs free energy of the system contains an ideal term and a mixing term for all four species (i.e. cholesterol, reactive and unreactive phospholipid and the complex formed between reactive lipid and cholesterol). From a thermodynamic analysis of this system and its temperature dependence deuterium NMR quadrupole splittings and order parameters were predicted, and phase diagrams were constructured [124, 125]. The non-linear behavior in that model stems from formation of higher order complexes implying some form of cooperativity between cholesterol and certain phospholipid species [126, 127]. The condensed complex model predicts a jump in ^Chol at a stoichiometric composition for a particular cholesterol fraction ( i.e., 33 mol% to 45 mol%cholesterol, dependent on the exact model parameters).

Recently, many observations made on living cells are interpreted based on the condensed complex model by conjecturing two pools of cholesterol to reside in each intracellular membrane, a free, that is non-complexed and a lipid-complexed cholesterol pool [4, 127-129]. The equilibrium constant

between both intra-membranous pools is set by the differing cholesterol affinity of the host lipids in that model [127, 128]. The non-complexed pool, increasing substantially above the threshold concentration, is sometimes called 'active cholesterol' [129]. Although intriguing and certainly an attractive view point for understanding cellular homeostatic responses [9], several questions remain, as for example the physical identity of the active and condensed cholesterol pool in a biological membrane are not clear. Even though the authors of the condensed complex model emphasize, that observation of sudden increases in nChol in a membrane do not require phase separation [124, 125], suggestive figures in recent reviews on sterol transport indicate two lateral sterol pools (i.e., non-complexed and complexed) in cellular membranes [128, 130, 131]. However, lateral microscopic sterol-enriched domains have never been observed in membranes of living cells. In fact, recent results show a homogeneous lateral distribution of cholesterol and its fluorescent analogs in the PM of living cells [19, 132-134]. This was supported by high resolution secondary ion mass spectrometry reporting that cholesterol is homogeneously distributed and not enriched in SM domains in the PM of fibroblasts, even though the cells were fixed using glutaraldehyde in these experiments [134]. Stimulated emission-depletion based FCS allowing for testing of diffusion laws by spot-size variation found free diffusion but no trapping of TopFluor-cholesterol and other dye-tagged cholesterol probes down to less than 80 nm in the PM of living cells [19, 135]. Thus, if cholesterol forms indeed lateral inhomogeneities in cellular membranes, such domains must be highly dynamic and very small with diameters much less than 80 nm [19, 136]. Interestingly, Monte Carlo simulations and recent neutron scattering experiments indicate cholesterol-PC domains in binary mixtures at high sterol mole fractions with an upper diameter of 22 nm and an average lifetime of 100 nsec [52, 70]. Currently, no experimental technique is available to determine whether cholesterol forms similar domains in membranes of living cells. Could the two PM leaflets resemble the active versus condensed or restrained cholesterol pools? While this cannot be ruled out, the known rapid sterol flip-flop in lipid membranes as well as the numbers found in recent studies with only about 30% sterol in the SM-rich outer leaflet and the majority of sterol in the likely less packed inner leaflet of the PM make this explanation also questionable [106, 137, 138]. The prediction of jumps in ^Chol in the condensed complex and in the umbrella model, as applied to bilayers, occur for particularly high cholesterol mole fraction (i.e., about 40-50 mol% depending on the exact parameters used for lipid interaction potentials) [123, 124]. At these concentrations, a given membrane might simply have reached its capacity to solubilize cholesterol, in which case the 'active' cholesterol pool would resemble excess cholesterol beyond the membrane solubility limit

[122]. The cholesterol solubility limit depends of course on the phospholipid head group and the acyl chain length and saturation, such that all arguments of the above analysis apply. Above the solubility point, cholesterol would precipitate from the bilayer in form of crystals [122], unless other acceptor membranes with remaining capacity to solubilize cholesterol or soluble LTPs are available in the system. Thus, the active cholesterol could simply resemble partly water-exposed cholesterol, which cannot be shielded under phospho- and sphingolipid head groups and needs to be exported to other membranes for lowering the total system Gibbs free energy. Experiments on acutely cholesterol-loaded cells support that view, since the cholesterol content of intracellular membranes increases rapidly up to five-fold under these conditions [48]. Also, non-vesicular DHE influx to lipid droplets and other membranes as well as efflux to cyclodextrin is enhanced in cholesterol-loaded murine macrophages [34]. Interestingly, these cells do not sequester excess sterol in domains in the PM [133], but rather stimulate their PC synthesis to generate more membrane for accommodation of excess sterol [139]. After prolonged cholesterol loading in such macrophage foam cells cholesterol crystals have been observed, indicating that the sterol solubility limit of all membranes has been passed [140, 141]. Neutron scattering experiments support the view, that cholesterol can partially protrude from the bilayer, especially at high mole fractions [142]. Recent molecular dynamics simulations indicate that the distance of cholesterol's hydroxyl group from the bilayer center diminishes non-linearly for high sterol mole fractions [143]. This partially protruding cholesterol could resemble the active cholesterol pool without need for separate domains. Increased transverse excursion of cholesterol at high mole fractions could resemble the onset of cholesterol precipitation unless being picked up by LTPs to trigger non-vesicular sterol exchange to a given acceptor membrane. Several other thermodynamic models have been put forward to explain binary and ternary phase diagrams of cholesterol and phospholipids; for example [102, 103, 144-146]. Not all of them invoke formation of condensed complexes, indicating that existence of complexes is not a requirement for the observed mixing characteristics of cholesterol and phospholipids. Such models do not attempt to relate the proposed biophysical mechanisms of cholesterol-phosholipid interactions to cholesterol transfer between membranes. This could be an interesting topic for future research.

Interesting but also somehow contradictive evidence for the hypothesis of 'active' cholesterol comes from recent studies using the cholesterol dependence of PFO binding to model and cellular membranes as readout for changes in nChol in the bilayer. In model and cell membranes, there is a

threshold concentration of cholesterol above which binding of PFO and its derivatives becomes detectable [147, 148]. This threshold does not seem to relate to cooperative oligomer formation of the proteins, since it also occurs for truncated PFO variants not forming oligomers [149]. For the PM of intact cells, the threshold for PFO binding is around 35% cholesterol, while for ER membranes this threshold is only 5 mol% cholesterol [4, 148]. Intriguingly, the latter value coincides exactly with the threshold for SREBP2 processing and nuclear targeting as a switch for regulating cholesterol homeostasis [4, 150]. Based on these observations, it has been proposed that the threshold for PFO binding resembles the jump in nChol, as predicted in the condensed complex and umbrella model [143, 147, 148]. In strong support of this argument is the sequence of threshold cholesterol concentrations for PFO binding to PC model membranes of varying acyl chain saturation ranging from about 35 mol% for DOPC over 45 mol% for POPC to approx. 49 mol% for DPPC [143, 147, 148]. Also, the mean depth of cholesterol's hydroxyl group in MD simulations coincided convincingly with the PFO binding affinity measured in the same study [143]. Neither membrane fluidity as assessed by diphenylhexatriene fluorescence anisotropy nor detergent solubility correlated with the threshold of PFO binding [147]. Thus, one can argue that PFO detects the 'active' cholesterol pool in the membrane, when the bilayer approaches its cholesterol solubility limit [143]. The extremely low threshold value for PFO binding to ER membranes, however, could not be reconciled in any model membrane experiment and cannot be explained by the condensed complex model, either [4]. Since the same threshold was observed for liposomes made from ER lipids, a specific role of ER proteins can be excluded [4]. Liposomes made of ER lipids should have bulk biophysical properties (i.e. fluidity, bending rigidity and microviscosity) resembling those of liposomes made from unsaturated PC species, as DOPC [89, 91, 151]. Thus, the very different threshold values for cholesterol accessibility of liposomes made of DOPC compared to those made of ER-derived lipids as measured by PFO binding remains enigmatic [4]. Since some cholesterol-dependent cytolysins require cholesterol in both membrane leaflets [152], and cholesterol might form transbilayer dimers at very low concentrations [75, 153, 154], a specific transverse cholesterol organization in the ER membrane might be responsible for the low threshold in PFO binding. Alternatively, PFO binding might require a specific lipid co-factor found in the complex mixture of ER lipids but not in model membranes. Quantitative lipid mass spectrometry of purified ER membranes might help to clarify this issue.

In the PM of living cells additional leaflet-specific cholesterol-lipid interactions might be at play in regulating sterol transport and triggering physiological responses to control cholesterol metabolism. For example, the majority of SM and all glycosphingolipids will be restricted to the outer PM leaflet. Treatment of fibroblasts with bacterial sphingomyelinase (SMase) causes the formation of ceramide in the PM paralleled by rapid ATP-independent cholesterol transport to and esterification in the ER [155, 156]. The same treatment increases cholesterol efflux to cyclodextrin, while degradation of PC with a specific phospholipase had a much lower effect on this process [157]. SMase treatment also triggered energy-independent formation of endocytic vesicles which, however, do not appear to be enriched in PM-derived sterol [34, 158]. These observations can be explained by a preferred interaction of cholesterol with sphingolipids, as discussed in section 3.1., above and further reviewed elsewhere [159]. Consumption of SM as substrate in the SMase-catalyzed reaction would reduce the number of cholesterol-SM pairs, thereby weakening cholesterol interactions in the PM. Alternatively, the non-polar reaction product, ceramide, competes with cholesterol for shielding by phospholipid head groups and pushes cholesterol out of the bilayer above some critical concentration. This explanation is fully in line with the umbrella model, described above and further substantiated by several experimental studies. For example, ceramide replaces cholesterol from ordered domains in model membranes and triggers cholesterol precipitation into the crystal phase [160, 161]. The maximum solubility of cholesterol in the latter study decreased concomitantly with the increase in ceramide concentration, completely in line with the umbrella model of cholesterol-lipid interactions. Ceramide also directly competes with cholesterol for association with SM in model membranes [86]. A recent study combined the PFO binding assay with SMase treatment of cells and other biochemical assays on human fibroblasts incubated with LDL [162]. PFO binding was increased after incubating cells with LDL for several hours, and this effect was enhanced when cells where treated with SMase prior to PFO binding. Similarly, SMase caused increased cholesterol esterification, in agreement with earlier results [155, 156, 162]. The authors argued that three pools of cholesterol exist in the PM; 1) a PFO-detectable pool comprising about 15 mol% of PM lipids, which is expanded upon LDL uptake. The other two pools are accordingly not accessible to PFO and were suggested to resemble 2) a SM-sequestered pool and 3) a so-called essential pool, not responding to any treatment but being important for cell viability. As discussed above for the 'active' vs. lipid-complexed cholesterol pool, the physical nature of these biochemically defined cholesterol pools remains to be determined, especially in light of accumulating evidence for a homogeneous lateral cholesterol distribution in the PM [162-164].

Experiments with collisional quenchers and intrinsically fluorescent cholesterol analogs suggest that the majority of cholesterol resides in the inner leaflet of the PM. This is a striking but also puzzling observation given the evidence for preferred interaction of cholesterol with SM and other sphingolipids in the outer leaflet (see above). However, evidence for preferred enrichment of cholesterol in the inner leaflet has been provided also in other much earlier studies [165-167]. Recently, a model has been put forward which suggests that cholesterol is drawn to the inner leaflet by the high abundance of PE in this leaflet [168]. PE has a small headgroup and by itself a high spontaneous curvature forming inverted hexagonal phases at physiologic temperature due to dominating entropy of its fatty acyl chains. This causes a strong bending energy being quadratic in the PE concentration, which would be counteracted by cholesterol [168]. At first glance, this suggestion is in contradiction to partition experiments, in which cholesterol has a lower affinity for liposomes made of PE compared to those made of PC and SM (see section 3.1., above) [90]. However, the important point in the model by Giang and Schick (2014) is the asymmetric PE distribution in the membrane causing an energy penalty due to negative bending and PE's spontaneous curvature [168]. Since cholesterol is shielded under the PE head group in accordance with the umbrella model, the spontaneous curvature caused by PE's molecular shape gets compensated when cholesterol is flipped to the inner leaflet [121, 168]. The model predicts about 48-58% of PM cholesterol in the inner leaflet, depending on the parametrization of the spontaneous curvature. These modelling results are in line with earlier theoretical studies on phospholipid translocation in the erythrocyte membrane by an ATP-dependent translocase, which also predicted about 50% of cholesterol in the inner leaflet [169]. Active translocation of PE and PS or adding exogeneous SM to the membranes outer leaflet would cause a transient increase of cholesterol in the inner leaflet to compensate for any area imbalance. Thus, fast cholesterol flip-flop in the PM could be an efficient mechanism for keeping the difference in lipid numbers between both leaflets small, especially under non-stationary conditions due to membrane fusion or bending [169]. The latter is indeed supported by experiments [170].

Taken together, passive processes and specific cholesterol-lipid interactions play a dominant role in cholesterol movement and distribution in the cell, especially when cellular cholesterol exceeds some threshold concentration. Understanding the detailed nature of such cholesterol-lipid interactions in individual cellular membranes and the specificity of LTPs carrying cholesterol along

seem to be the key aspects in determining the control of non-vesicular sterol transport in cells. How other biophysical properties of cellular membranes, as protein content [171], free fatty acids [172] or bilayer curvature [60, 68, 76] affect inter-organelle cholesterol transfer remains open. Similarly, understanding the impact of constant vesicle trafficking and ATP-consuming lipid remodeling on non-vesicular sterol fluxes awaits future research.

4. Protein-mediated uptake, intracellular targeting and efflux of cholesterol

In the following, we will give an overview of several cholesterol transport pathways and their protein dependence in living cells. We will focus on mammalian cells, since several excellent reviews have recently discussed sterol trafficking in yeast and other organisms [56, 130, 173].

4.1. How lipid transfer proteins mediate non-vesicular sterol transport - two examples. Several cases of non-vesicular cholesterol transport in cells have gained a lot of interest and insight in the last decades. One example is cholesterol import into mitochondria in steroidogenic cells, which involves StAR on the mitochondria and cholesterol-donating StARD3/MLN64 and NPC2 on late endosomes (LEs) [174-176]. StAR and MLN-64 are both involved in cholesterol transport, and they were found to contain the same domain named steroidogenic acute regulatory-related lipid transfer (START) domain. The START domain of MLN64 and StAR was shown to bind cholesterol at an equimolar ratio [77]. The C-terminal domain of MLN-64 contains 37% amino acids identical to the StAR protein and nearly 60% of amino acids, which show similarities to StAR [177]. Little is known about the mechanistic details of this transport route, and current knowledge has been reviewed elsewhere [174, 175]. Another case is transport of cholesterol from the ER, where cholesterol is synthesized, to the cell surface and the third case is cholesterol import from the PM to the ERC and ER. We will consider these latter two processes in the following section and use them for delineating basic principles of LTPs function in sterol transport. De novo synthesized cholesterol traffics to the PM in about 10 min, largely bypassing the Golgi apparatus [16, 28, 178, 179]. This transport required ATP and stopped below 15 C, while maintenance of the established cholesterol gradient between PM and ER was independent of metabolic energy [16, 28]. Thus, vesicular and non-vesicular transport modes seem to be at play for export of cholesterol from the ER. Sterol exchange between ER and Golgi depends on the activity of OSBP, whose two structural motifs create membrane contact sites between both organelles. The N-terminal plekstrin homology (PH) domain binds to phosphatidyl inositol-4-phosphate (PI-4-P) in the TGN, while the FFAT (two

phenylalanine in an acid tract) motif binds to VAMP associated protein (VAP) in the ER [56, 81]. The OSBP-related domain (ORD) of the protein transfers sterols between both membranes, as suggested by in-vitro experiments, in which about 30 DHE molecules were transferred per minute per OSBP tethered between donor and acceptor liposomes [81]. Addition of 25-hydroxycholesterol to cells recruits OSBP to the Golgi interface, while the PH-FFAT region induces tethering and formation of several membrane appositions of about 20 nm distance between ER and PM [81]. Such close apposition will enhance the likelihood of sterol transfer instead of reabsorption by the donor membrane (see Figure 1 and below) [180]. The ORD binds and transfers not only sterols but also PI-4-P, and a model has been put forward, in which ATP-consuming hydrolysis of PI-4-P in the ER by Sac1 drives directional lipid exchange between both organelles [81]. Further information about the function of OSBP and other family members in lipid transfer can be found in recent review articles [56, 130, 181]. Establishment of membrane contact sites by a putative sterol transporter has been recently also observed between LEs and the ER via interaction of MLN64 with VAP [182]. This result suggests that MLN64 can catalyze non-vesicular sterol transfer between these organelles by a similar mechanism, as proposed for OSBP (for further details on cholesterol transport from LEs, see section 4.3. and Figure 3, below).

Compelling evidence for non-vesicular sterol transport from the PM came from imaging studies of DHE in TRVb1 cells, a Chinese hamster ovarian (CHO) cell line expressing the human transferrin receptor [13]. DHE was inserted into the PM from a cyclodextrin complex, and sterol transport to the ERC was studied. Interestingly, energy-poisoning reduced the steady state level of DHE in the ERC only by 30%, and transport of DHE to the ERC continued even after cell fixation [13]. A simple flux-force relationship can explain these observations, as long as the influx rate is proportional to the size of the perturbation. Here, a strong gradient of fluorescent sterol is established at the moment, the tracer DHE is added to the cells (i.e., pulse-chase conditions; see 3.1., above). Note, that in this experiment DHE is exchanged against PM cholesterol, such that overall sterol equilibrium should not be perturbed. Specific properties of the endosomal membranes might create a sink for sterol transport resulting in DHE entrapment in the ERC [13]. This could be a consequence of the specific lipid composition of this organelle, as discussed in section 3.2. Nonvesicular cholesterol transport between PM and ERC/ER was found to be accelerated by overexpressing some LTPs, as SCP2, liver-specific FABP and ORP2, an OSBP homologue, as well as StARD4 [14, 85, 183-185]. The abundance of StARD4 had a pronounced effect on DHE

targeting to the ERC and ER, on cholesterol esterification and on proteolytic procession of SREBP2 [14]. FRAP experiments have shown that DHE transport to the ERC was rapid and largely ATP-independent with a half-time of t1/2 ~ 2.5 min in TRVb1 cells [13]. Reduced expression of StARD4 but not its over-expression slowed also the FRAP kinetics of DHE in the ERC of HepG2 cells and resulted in increased free cholesterol in the PM [186]. Importantly, injection of cyclodextrin mimicked the effects of StARD4 including acceleration of DHE transfer between PM and ERC [14]. In vitro, sterol transfer by StARD4 and by SCP2 was enhanced in the presence of negatively charged lipids, which likely interact electrostatically with surface residues on the proteins [14, 187]. StARD4 was also more efficient than cyclodextrin in transferring DHE between liposomes, when comparing the same carrier concentrations [14].

Given the extremely low water solubility of cholesterol, pulling one cholesterol molecule out of a bilayer into water requires overcoming an energy barrier of 80-90 kJ/mol corresponding to hydrolysis of about 1.5 ATP molecules [111]. So, what is the molecular mechanism underlying enhanced interbilayer sterol transport by these LTPs? Rapid binding of sterol to LTPs after cholesterol desorption from the bilayer would prevent reinsertion of the sterol into the donor membrane. The average distance, a cholesterol molecule diffuses into water before rebinding to the donor membrane can be quantified as mean diffusional excursion, dm, as proposed in a steady state transport model developed by Weisiger and Zucker (2002) to describe cytosolic fatty acid transport by FABPs [180]. Applied to non-vesicular sterol transport, the presence of a LTP with concentration [LTP] and dissociation constant for the ligand, KD, will increase dm according to

d. = 2 iDf +\LTPl ) (g)

Here, Df and Db are the diffusion constants of the free and protein-bound sterol, respectively, and Pmw is the permeability of the membrane-water interface defining the rate of rebinding [180, 188]. From Eq. 8, one can see that the largest values for dm results from highly abundant LTPs with high diffusion constants and strong binding affinity. The diffusive flux between two membranes at steady state, J, was calculated in that model as function of membrane separation distance, d, to [180, 188]:

j = Jmx (9)

1 + dldm

Here, jmax is half the rate of sterol dissociation (desorption) from the donor membrane, and thereby directly related to the chemical potential difference between the membranes, as discussed in section 3.2. For non-zero cholesterol concentration in the donor and acceptor membrane with concentration difference Ac, one gets Jmax = Ac-Pmw/2 [188]. Thus, the maximally achievable flux follows a simple flux-force relationship, as discussed in section 3.1., in which the concentration difference resembles the thermodynamic force, and Pmw is the generalized transport coefficient (e.g. 1. Fick's law on diffusion) [20]. Figure 1A illustrates the possible fates for a sterol molecule after desorption from a donor membrane, while Figure 1B shows, how the inter-bilayer flux depends on the intermembrane distance, d. Adding a soluble LTP will strongly increase dm, thereby preventing reabsorption of the sterol molecule (Figure 1C). If dm >> d, one sees from inspection of Eq. 9, that the steady state flux J approaches Jmax, as indeed found for membrane-inactive LTPs (e.g., fatty acid transport by L-FABP, Figure 1C) [188]. If the membrane separation distance, d, is much smaller than dm, J approaches Jmax, even without soluble LTPs, since the flux in this case is not diffusion-limited (Figure 1B) [188]. For cholesterol, which has a much lower desorption rate from membranes (i.e. hours, see section 3.1.) than fatty acids (i.e. milliseconds to seconds) [189], soluble LTPs might be not very efficient. Instead, for most known sterol carrier proteins and even for cyclodextrin, transient interaction with the donor membrane is instrumental for catalyzing inter-bilayer cholesterol transfer [14, 187, 190, 191]. In their pioneering modeling study, Weisiger and Zucker (2002) also determined the impact of membrane interactions of the LTPs on steady state flux of fatty acids [180, 188]. They showed that transient membrane associations of the LTPs also affects Jmax, as they lower the Gibbs free energy barrier for sterol desorption and thereby increase the permeability at the membrane-water interface, Pmw (Figure 1D). Interaction of LTPs with certain cellular membranes, either due to electrostatic interactions with acidic lipids as shown for StARD4 or due to dual binding motifs to proteins residing in different organelles as known from OSBP fall into this category [14, 81]. Membrane interaction of a LTP will therefore not only raise specificity but also increase the inter-bilayer sterol flux even if the membrane-membrane distance, d, is smaller than the mean diffusional excursion, dm. Interestingly, membrane-interacting LTPs as StARD4 show up to thousandfold higher per-molecule transfer activity of compared to cyclodextrin [14]. Similarly, absolute rates of intralysosomal sterol transfer by NPC2 are higher than for cyclodextrin [192]. As in case with StARD4, NPC2 was found experimentally to interact at least transiently with

membranes [191]. Molecular simulations including free energy calculations suggest also absorption of cyclodextrin to membranes [190], but whether this interaction is weaker compared to StARD4 or NPC2 awaits further comparative studies. Dual binding motifs to the donor and acceptor membrane, as found in OSBP or as recently suggested for tethering LEs to the ER via MLN64, might be a particularly efficient way of raising sterol flux, since beside of increasing Jmax, such proteins reduce the interbilayer distance by creating close membrane contact zones [56]. Via such membrane contacts, a situation might be generated in which dm ~ d, such that monomeric sterol diffusion is not limiting and very efficient sterol flux between organelles is ensured (Figure 1C and D, case (3)).

4.2. Cholesterol endocytosis and its functional importance

In addition to non-vesicular transport described above in section 3.2., cholesterol trafficking between PM and ERC takes place by vesicle traffic as well. For example, recycling of DHE from the ERC back to the cell surface required the vesicle export machinery involving the Rme-1 protein and occurred with a half-time of about 24 min in TRVb1 cells [13]. Transport of DHE from the PM to the ERC studied in pulse-chase experiments in the same study reached steady state values after about 15 min, which was similarly found in macrophages and human hepatoma HepG2 cells [17, 34]. Similarly, transport of DHE to the ERC took place with a half time of about 8 min in rat hepatoma cells [45]. We compared recently membrane partitioning and intracellular transport of DHE with that of TopFluor-cholesterol and showed that both sterols are targeted to the ERC of Baby hamster kidney (BHK) cells with identical kinetics [42]. Uptake of both sterols from the cell surface was strongly reduced in BHK cells overexpressing a dominant-negative clathrin heavy chain and after ATP depletion, suggesting that clathrin-dependent endocytosis makes a significant contribution to overall sterol uptake in these cells. Similarly, Ge et al., found that uptake of cholesterol from the plasma membrane of rat hepatoma cells depended on clathrin and required expression of human Niemann-Pick C1 like 1 (NPC1L1) which directly interacted with clathrin heavy chain in this study [193]. NPC1L1 is a target of ezetimibe, a cholesterol absorption inhibitor in the intestine and is thought to play a central role in intestinal cholesterol absorption [194, 195]. The sequence and structure similarity of NPC1L1 to NPC1 including sterol binding in the N-terminal domain would suggest a similar function of NPC1L1 in cholesterol export from early endosomes and/or LE/LYSs [196]. This, however, was not reported so far. NPC1L1 contains a sterol sensing domain, as NPC1 or HMG-CoA reductase and binds cholesterol at its N-terminal domain [197]. NPC1L1is found in the apical, canalicular membrane of hepatocytes and polarized hepatoma cells, where it is highly mobile, as shown by FRAP [198, 199]. This protein was found to

be important for cholesterol reabsorption from the bile compartment, probably to a subapical recycling compartment, thereby regulating cholesterol homeostasis also in the liver [200]. We observed recycling of NPC1L1 between cell surface and ERC and slightly enhanced vesicular targeting of DHE to this endocytic compartment in non-polarized rat hepatoma cells [45]. However, the majority of DHE uptake in this and other cell types takes place also in the absence of NPC1L1. Similarly, recycling of DHE or cholesterol from the ERC to the cell surface does not require NPC1L1, but rather depends on other proteins orchestrating endocytic recycling, as the ED domain protein RME-1 [13]. Several publications by the Song-group have provided evidence for a microtubule-based NPC1L1-mediated endocytic net cholesterol transport route from the PM to the ERC requiring flotilin and the clathrin adaptor Numb and for recycling of the cholesterol-NPC1L1 complex back to the cell surface in dependence of Myosin Vb, Rab11-FIP2 and the small GTPase Cdc42 [193, 201-203]. While this transport route is likely important for maintenance of the steady state distribution of NPC1L1, the used cholesterol transport assay is not beyond reproach; filipin has been used to detect cholesterol replenishment from a cyclodextrin-cholesterol complex, but first after acute cholesterol depletion of the cells for 60 min with cyclodextrin [193]. While this treatment allowed for establishing a filipin-based pulse-chase transport assay, it seems to be far from mimicking the demand for cholesterol absorption in enterocytes. Thus, eventual artefacts known to be caused by prolonged cholesterol depletion using empty cyclodextrin cannot be ruled out [204]. Indeed, intestinal cholesterol absorption is a complex process, and the involvement of NPC1L1 is not without question [205]. Given the high sterol mole fraction in the PM of mammalian cells, formation of endocytic vesicles will probably always cause some internalization of cholesterol, likely in several endocytic pathways (i.e., clathrin-dependent and -independent pathways). Thus, specific sterol recruitment to a particular membrane protein, as NPC1L1, in the PM, prior to internalization is likely not necessary for sterol endocytosis. Indeed, we observed endocytosis of DHE and TopFluor-cholesterol by time-lapse microscopy in cells not expressing NPC1L1 [18, 46]. Due to lack of conclusive experimental evidence, a convincing mechanistic model on involvement of NPC1L1 in sterol uptake and transport between PM and ERC is therefore currently not available. Interestingly, both fluids from which NPC1L1 has been proposed to mediate cholesterol absorption into cells, the digestive fluid in the intestine and the bile fluid in the liver, contain cholesterol derived detergents, i.e. bile salts. It would be interesting to study, how bile salts affect NPC1L1-mediated cholesterol absorption into cells, thereby eventually shedding light onto the function of this protein in these specialized tissues.

After internalization, the dynamics of vesicles carrying TopFluor-cholesterol has been studied in CHO cells using two-photon excited SPT [18]. Such vesicles were independently shown to contain DHE and TopFluor-cholesterol avoiding any artefacts caused by eventual miss-targeting of TopFluo-cholesterol [12, 42]. We found that the majority of sterol-rich vesicles moved slowly by anomalous diffusion with D~ 2-10-3 |im2/seca, and a ~0.6 [18]. Both, actin and microtubule disruption affected the diffusion of such vesicles in the cytosol. A sub-pool of vesicles, however, showed faster, directed motion, and endosome fission and fusion was found in time-lapse sequences of TopFluor-cholesterol [18]. Slow confined diffusion was also found for DHE-rich vesicles by SPT on a wide field set up in macrophages and by temporal image correlation spectroscopy on a multiphoton microscope in HepG2 cells [46, 206]. We showed also recently that SPT and TICS provide comparable results on diffusion and directed transport of vesicles [27]. Together, recent data suggest that under normal physiological conditions sterol trafficking between PM and the sterol-rich ERC takes place on a time scale of 15 to 20 minutes. Vesicular and non-vesicular modes seem to contribute each about half to that sterol transport, but exact numbers depend on cell type and used transport assay.

4.3. Cholesterol import via low density lipoprotein and intracellular processing of its sterol load This section is dedicated to cholesterol import into cells via lipoproteins. Since lipoproteins are very diverse and their physiology is a huge field of research, we will focus on the paradigm of LDL-mediated cholesterol uptake. This pathway and its discovery represents one important example, in which a genetic disease, here familial hypercholesterolemia, led to important mechanistic insights into a fundamental biochemical process [36, 207]. There are, of course, many earlier examples in which genetic mutations were mechanistically linked to human diseases, as sickle cell disease or phenylketonuria, but the history of the discovery of the LDL pathway follows that tradition [208]. LDL is about 22-27 nm in diameter, and each particle contains abundant CEs and triacylglycerides (TAGs) in its core. The surface coat of the LDL particle contains phospholipids (25% by weight), a single 550 kDa apoB-100 protein (25%) and non-esterified cholesterol. TAGs, cholesterol and CEs represent the remaining 50% by weight of each LDL particle [209]. LDL binds to the LDL receptor (LDL-R) at the cell surface at neutral pH, upon which the ligand-receptor complex gets recruited to clathrin-coated pits via interaction with adaptor protein 2 (AP2). After endocytosis and uncoating, the newly formed vesicle fuses with sorting endosomes (SEs), a sub-population of the early

endocytic pathway. SEs loose their fusion competence for incoming vesicles and mature into LEs, a process accompanied by a switch in endosome-associated rab proteins, from rab5 to rab7 and by acquisition of intraluminal vesicles (ILVs) (Figure 2) [210, 211]. The lysosomal pathway can be divided into LEs, a subpopulation of which is sometimes called multivesicular bodies, and lysosomes (LYSs) - acidic organelles with large content of hydrolytic enzymes [212]. Formation of ILVs inside LEs depends on specific lipids such as bis(monoacylglycero)phosphate (BMP), ceramide and cholesterol. This process is also strictly dependent on the endosomal sorting complex required for transport (ESCRT), as recently reviewed in detail [212]. The ILVs fulfil diverse functions in endosome maturation, protein degradation and lipid metabolism [212, 213]. LE/LYSs can also secrete material from cells, often in form of small vesicles called exosomes. These secreted vesicles likely form from ILVs and have diverse functions in cell-to-cell communication and lipid metabolism [214]. Traffic between LEs and LYSs is extensive and several theories exist about the relation of both organelles [212]. Degradation of ingested cargo seems to take place in LEs or hybrid organelles between LEs and LYSs, while LYSs fulfil a role as enzyme reservoir [212, 215]. However, as assignment of protein markers and molecular function is not unequivocal in the literature, we will use in the following the term LE/LYSs to address transport through the degradative organelles, unless otherwise specified. The LDL-R contains six building domains: the binding domain, an epidermal growth factor domain (EGF), a tyrosine-tryptophan-threonine-aspartic acid (YWTD) rich-domain, an O-linked sugar domain, the transmembrane domain and the cytosolic NPxY motif [216]. There are around 1400 LDL-R mutations known, which lead to fatal diseases such as hypercholesterolemia, tendinous xanthoma and premature coronary heart disease [3]. The ligand binding domain in the extracellular portion of the LDL-R contains 7 cysteine-rich regions, which were shown to be negatively charged. Therefore, they can bind to positively charged residues on the apoB-100 of LDL and VLDL. Lipoprotein binding to the LDL-R can be blocked by chemical modification of R1-R7 in LDL-R [3]. The extracellular domain of the LDL-R undergoes a conformational change upon switching the pH from 7.4 to 5.0, which is a precondition of ligand dissociation within SEs in cells [217]. Fass et al. (1997) showed binding of calcium ions into the binding site on the R5 at pH=5.0, and the crystal structure of this complex could be resolved [218]. Recent results show that, beside an acidic pH, low endosomal calcium is essential for release of LDL from its receptor, likely as a consequence of structural changes in R1-R7 upon calcium release [219].

The ligand-freed LDL-R returns to the cell surface, either directly from SEs or from the ERC for another round of LDL uptake [210, 220]. Each LDL-R can import about 120 LDL particles and thereby about 200.000 CE molecules. Within a population of LEs, acid lipase hydrolyzes LDL-associated CEs to cholesterol and fatty acids. Lack or reduced expression of the acid lipase causes Wolman disease and CE storage disease, respectively [221]. These diseases are characterized by accumulation of CEs in LE/LYSs. Export of the LDL-derived cholesterol from LEs is only partly understood, but important insight came from another genetic disease. NPC disease is a rare neurodegenerative disorder caused by a mutation in one of two proteins. NPC1 protein contains several transmembrane helices and locates mostly to LE/LYSs but is also found in the TGN [222]. NPC2 protein is small and soluble in the lysosomal lumen, but it is also present in several body fluids including milk, bile and epididymal fluid [59, 222]. Loss of function in either NPC1 or NPC2 protein has been shown to cause severe lysosomal accumulation of cholesterol, glycosphingolipids, sphingosine, and SM [222, 223]. The resulting cellular phenotype is characterized by large amounts of lipid-laden crescent shaped LE/LYSs, sometimes called lysosomal storage organelles [222]. Fibroblasts lacking functional NPC1 protein hydrolyze LDL derived CEs normally but have a strongly reduced ability to elicit normal regulatory responses, as stimulation of esterification and suppression of synthesis of cholesterol [59, 222]. It is unclear whether cholesterol liberated from ingested LDL is first targeted to the PM, the TGN or to the ER, where most cholesterol metabolizing enzymes reside [224]. NPC1 and NPC2 bind cholesterol, various oxysterols as well as intrinsically fluorescent sterols like DHE with nanomolar to micromolar affinity [62, 65, 66, 225]. Infante et al. (2008) purified and characterized the N-terminal domain (NTD) of NPC1, later called NPC(NTD) [225]. This loop contained 240 amino acids, and it was isolated as a highly water soluble protein with tendency to form a homodimer, as suggested by gel filtration [54, 225]. One cholesterol molecule is able to bind to NPC(NTD), but 25-hydroxycholesterol, was found to be favored in the binding site in comparison to cholesterol, eventually due to formation of additional hydrogen interactions of the oxysterol in the binding site [225]. In fact, Infante et al. (2008) found that the majority of the oxysterol binding protein found in the liver is NPC1 [54]. The crystal structure of bovine NPC2 was first determined by Friedland et al in 2003 [65]. Crystals suitable for the X-Ray experiment were grown from NPC2 purified from bovine milk using a hanging drop method at neutral pH stabilized by Tris buffer. The structure contains 7 // strands, which are arranged into two // sheets. The hydrophobic interior responsible for sterol binding in NPC2 is loosely packed with an available space for one sterol molecule (i.e. with a volume of 84 A3 [64,

65]. The mechanisms underlying NPC1/ NPC2 function in export of LDL-derived cholesterol from LE/LYS are not known. Mutations in both genes cause inhibition of LDL-stimulated cholesterol re-esterification in the ER and failed suppression of SREBP- and LXR-dependent gene expression [226]. These defects take place despite increased total cellular free cholesterol and are consistently more severe for mutations in NPC2. Treatment of NPC disease fibroblasts with the oxysterol, 25-hydroxycholesterol can partially overcome the cholesterol loading in these cells [226]. It has been proposed that NPC1 and NPC2 protein work in tandem in mediating lysosomal cholesterol export [59]. Pioneering studies by Goldstein & Brown and co-workers published from 2008 on have shown that the N-terminal soluble luminal domain of NPC1, named NPC1(NTD) can bind cholesterol and oxysterols with high affinity [54, 225]. Elegand binding and transfer assays showed that NPC2 can efficiently transfer cholesterol to and from liposomes, while transfer between vesicles and NPC1(NTD) was very slow. This transfer was, however, more than a hundredfold accelerated in the presence of NPC2 [227]. Subsequent alanine mutagenesis studies combined with X-ray crystallography revealed distinct subdomains in NPC1(NTD) for cholesterol binding and transfer [63]. Key amino acids on the surface of NPC1(NTD) and NPC2 were identified for interaction of both proteins and for cholesterol transfer [228]. The suggested hydrophobic hand-off of cholesterol between both proteins resembles substrate channeling in metabolic reactions and has been investigated further using free energy calculations based on molecular dynamics simulations [227, 229]. The latter study used an implicit solvent description based on the generalized Born approximation and a simulation technique called nudge elastic band calculations to suggest likely (i.e. energetically favourable) conformational transitions during cholesterol transfer between NPC1(NTD) and NPC2 [229]. Importantly, the authors found that the protein complex is most stable if cholesterol associates with NPC2 and least stable if the apoproteins NPC1(NTD) and NPC2 were simulated [229]. However, based on existing experimental and computational data, the directionality of cholesterol transport between both proteins could not be unequivocally established, yet [227, 229]. The cholesterol binding pockets of both proteins are bent with respect to each other during cholesterol transfer in the simulation study by Wiest and colleagues, indicating that either the sterol ligand or one of the NPC proteins needs to change its conformation during sterol transfer (Figure 3, inset A) [229]. Another computational study on the NPC 1(NTD)-NPC2 complex was published recently by Elghobashi-Meinhardt (2014), in which the possibility of conformational changes in the side chain of cholesterol during sliding from the NPC2-pocket to the NPC1(NTD) binding site was explored [230]. Using a quantum mechanical (QM) description of the sterol ligand

and a classical molecular mechanics (MM) force field for the proteins, the energy barrier for rotation of the C17-C20-C22-C23 dihedral angle was determined during sliding of cholesterol from NPC2 to NPC1(NTD) (Figure 3, inset B) [230]. This was motivated by the earlier simulation study showing that this dihedral angle is 71.6° for cholesterol in NPC2 but -157.3° in the NPC1(NPD) binding pocket, while in the respective crystal structures, this torsion angle is nearly identical with -164° [63, 64, 229, 230]. Using QM/MM simulations, the likely 'reaction pathway' for cholesterol transfer and the energy barrier along that path was calculated giving a barrier of ~22 kcal/mol in total. Structural changes in several torsion angles in the cholesterol side chain were suggested to allow for its isomerization during transfer [230]. In both simulation studies, point mutations found in the alanine mutagenesis screen were found to reduce either the efficiency of transfer or the stability of the complexes [63, 228-230]. Additional interactions between NPC2 and other intraluminal loops of NPC1 have been recently determined by surface plasmon resonance [231].

Despite this progress and mechanistic insight, the hydrophobic handoff model for cholesterol transfer between NPC2 and NPC1 is based solely on in vitro experiments but not (yet) grounded in thorough cellular studies. For example, a direct interaction between NPC1 and NPC2 protein has never been demonstrated in living cells. NPC1 seems to have a rather complex distribution between several organelles including not only LEs but also early endosomes, the Golgi and even partly the PM [232-234]. In contrast, NPC2 is restricted to LE/LYSs, and a recent quantitative imaging study showed that internalized bovine NPC2 can remove cholesterol efficiently from a subset of the sterol storage compartments visualized using filipin [25]. Also, export pathways for internalized cargo from LE/LYSs have been described, which depend on NPC2 but not on the NPC1 protein [235237]. For example, the StAR-domain containing MLN64 was found to export cholesterol from LE/LYSs independent of NPC1 and was found to reside in a distinct population of LE/LYSs [235, 238]. How such results can be incorporated into the structural model of NPC2-NPC1-mediated cholesterol transfer described above and in Figure 3 is a challenge for future research. Maybe, several exit pathways exist from LE/LYSs for cholesterol; e.g. a NPC1-mediated route as well as a pathway dependent on MLN64-and/or the ATP-binding cassette transporter A1 (ABCA1, discussed below in section 4.3.; Figure 2). NPC2 could feed all these pathways by efficiently shuttling cholesterol from ILVs to such transporters in the limiting endosomal membrane which then transfer them to various acceptor organelles. On the cytoplasmic side, OSBP-related protein (ORP5) and

OSBP-related protein like protein 1 (ORPL1) were additionally implicated in shuttling of LDL-derived cholesterol between LE/LYSs and the ER [239, 240]. Recent multi-color imaging studies suggest functional sub-compartmentalization of LEs in 'early LEs' containing LDL, an ABC transporter (ABCA3) and MLN64 and 'late LEs' harboring most NPC1 and ORPL1 [238]. ORP1L1 seems to be essential to target LEs to the perinuclear region in proximity of the ER, which would be a precondition for cholesterol transfer [239]. Again a virus trafficking study provided direct evidence for concerted action of NPC1 and ORPL1 in targeting cholesterol from LEs to the ER; adenovirus RIDa can rescue the cholesterol storage phenotype in cells lacking NPC1 [241]. This pathway led to increased formation of CEs derived from hydrolysis of LDL and in formation of ER-derived lipid droplets [242]. Importantly, this transport depended strictly on functional NPC2 supporting the view that NPC2, NPC1 and ORPL1 act in tandem [242]. Direct observation of endocytic processing of LDL-derived CEs was made possible using a TopFluor-cholesteryl ester reconstituted into LDL [243]. Egress of hydrolyzed TopFluor-cholesterol from LE/LYSs was found to depend not only on acid lipase and NPC1 activity but also on the small GTPase rab8. This supports earlier results by the same group showing that rab8 overexpression can partially restore the NPC1 lipid storage phenotype [244]. The motility of LEs containing the fluorescent CE analog depended on actin, similar as previously found again by Ikonen's group for MLN64 [245]. Members of the ESCRT family, as Hrs/Vps27, and rab7, a key GTPase for LEs formation and positioning, are also involved in processing of LDL and release of its cholesterol after CE hydrolysis [246, 247]. In addition, evidence has been provided for an export route of LDL-derived cholesterol from LEs over the Golgi to the ER [248]. Further information about the protein machinery regulating trafficking through the endosomal pathway and thereby affecting also cellular cholesterol homeostasis can be found elsewhere [212, 224, 249, 250].

Some evidence has been provided that the primary offending metabolite in NPC1 disease is SM, while other studies point to gangliosides or accumulation of sphingosine (the product of ceramide degradation by acid ceramidase) [223]. Sphingolipids, as SM, were found to block calcium import into LE/LYSs, thereby impairing endosome/lysosome fusion and lysosomal exocytosis, a mechanism suggested for NPC1 disease [223, 251]. Agonists of the transient receptor potential (TRP) channel in LE/LYSs have been shown to restore normal calcium levels in LE/LYSs (~150 |iM) and to rescue the cholesterol and sphingolipid storage phenotype in NPC1 disease cells [251].

Similar effects were described for the polyphenol curcumin found in yellow curry and the blocker of the ER calcium pump thapsigargin [252]. It was speculated that lysosomal exocytosis was stimulated by such treatments, resulting in enhanced secretion of cholesterol-rich exosomes from the NPC1 disease cells [253]. How calcium homeostasis is linked to function of the NPC2 protein and eventual lysosomal exocytosis remains to be determined. In in vitro assays, NPC2 protein was found to enhance inter-membrane cholesterol transfer in a ceramide- and BMP-dependent manner [213]. Production of ceramide from SM by an acid SMase (aSMase) seems to be essential for cholesterol efflux from LE/LYSs [213, 254]. In fact, treatment of NPC1 disease cells with aSMase could also ameliorate the cholesterol storage phenotype [254]. Similar experiments have not been performed for NPC2 disease cells, but in vitro experiments indicate, that NPC2 and aSMase act in concert in transferring cholesterol between liposomes [213, 255]. However, though much argues for cholesterol, the offending metabolite remains to be unequivocally identified.

4.4. ATP-binding cassette transporters and their role in cholesterol efflux from cells Plasma concentrations of high density lipoprotein (HDL) are negatively correlated with the risk for developing cardiovascular disease. This is, because HDL-mediated reverse cholesterol transport to the liver is an efficient way to remove excess cholesterol from the circulation. Formation of HDL starts with secretion of lipid-poor apoprotein A1 (apoA1) by the intestine and liver, and apoA1 gets subsequently lipidated by interaction with peripheral tissue cells. ApoA1 receives cellular cholesterol and phospholipids in a process dependent on ABCA1. Within the nascent HDL particles, cholesterol is esterified by lecithin-cholesteryl acyl transferase (LCAT) associated with the growing lipoprotein. Using a fatty acyl chain of apoprotein-associated PC, LCAT esterifies cholesterol in the nascent HDL particles thereby creating a sink for further cholesterol efflux from cells. Patients with mutated ABCA1, as observed in Tangier disease, have an increased risk of developing cardiovascular diseases due to a strongly impaired cellular lipid efflux to apoprotein A1 (apoA1), and consequently dramatically reduced plasma HDL levels [256, 257]. Cholesterol efflux to mature HDL is primarily mediated by the ABC half transporter ABCG1 and takes place either on the cell surface or during endocytic recycling of HDL [258]. Targeted disruption of the gene coding for ABCG1 in mouse results in massive accumulation of CEs, TAGs and phospholipids in hepatocytes and macrophages [259]. Recent evidence indicates ABCGl's involvement in regulating endosomal cholesterol levels, thereby indirectly affecting the transbilayer sterol distribution in the PM upon fusion of endocytic vesicles with the cell surface [260]. Expression of ABCA1 and ABCG1 is under control of the Liver X receptor (LXR) and can be stimulated by LXR ligands, like

25-hydroxycholesterol [261]. It is currently not known, which cellular cholesterol pool is dominantly used for ABCA1-mediated sterol efflux. The prevailing hypothesis is that ABCA1 acts in concert with apoA1 on the cell surface to remove PM cholesterol being constantly replenished by intracellular sources [256, 262]. Other studies indicate that ABCA1, which follows a complex intracellular trafficking scheme [257, 263, 264], mediates lipidation of apoA1 during its passage through the cell, likely by a retroendocytic pathway [262]. Interestingly, ABCA1-mediated cholesterol efflux is not only impaired in fibroblasts from patients with Tangier disease but also in fibroblasts from patients suffering from NPC disease and CE storage disorder [221, 265, 266]. Reduced conversion of LDL-derived cholesterol into oxysterols is likely causing the diminished expression of ABCA1 in these diseases, since oxysterols act as LXR agonists and induce expression of ABCA1 [226, 261]. Importantly, activation of ABCA1 using LXR agonists is sufficient to trigger cholesterol efflux to apoA1 in cells lacking functional NPC1 protein [265], while ABCA1-mediated cholesterol efflux is strictly dependent on the NPC2 protein [236]. These results demonstrate the importance of the lysosomal cholesterol pool for ABCA1-dependent cholesterol efflux and indicate again different mechanistic roles for NPC1 and NPC2 in cholesterol mobilization from lysosomes. Interestingly, recent work by Molday and colleagues on reconstituted ABCA1 shows that this transporter, though being so central for cellular cholesterol efflux, does not bind cholesterol; in fact, its ATPase activity in phospholipid liposomes actually declines in the presence of cholesterol [73]. Reconstituted ABCA1 actively translocated fluorescent analogs of PC, SM and PS from the cytoplasmic to the exoplasmic leaflet of membranes [267]. In line with this observation, active translocation of PS analogs to the outer leaflet of the PM by ABCA1 has been shown in intact cells [268]. This causes altered inner surface membrane potential and reduced rate of endocytosis [268270]. Side-specific quenchers of DHE and CTL have been used to demonstrate ABCA1/ABCG1-dependent sterol redistribution from the inner to the outer leaflet of the PM in living cells [271]. Similar experiments on CHO cells which lack ABCA1 revealed that most sterol resides in the inner leaflet [138], while sterol flip-flop across membranes is very rapid [79, 272, 273]. Together with the absence of cholesterol binding to ABCA1 and the inhibitory effect on its ATPase activity [73], a mechanistic understanding of the enhanced ABCA1-dependent sterol translocation to the exoplasmic leaflet of the PM is lacking. One can hypothesize that active flopping of PS and other phospholipids 'pulls' some cholesterol to the outer leaf of the PM. In line with this idea are recent observations by Smith and colleagues, showing that depletion of SM by treating cells with SMase or by inhibiting sphingolipid synthesis cause PM remodeling with enhanced PS exposure on the

outer leaflet and higher cholesterol efflux from cells by ABCA1-dependent and -independent mechanisms [274]. Importantly, LXR-dependent expression of ABCA1 depends on another ABC transporter, ABCA12, which was first described to mediate formation of lipid lamellae in the skin. Mutations in ABCA12 cause harlequin ichthyosis, a devastating skin disease with abnormal lipid granules in keratinocytes resulting in impairment of the skin barrier function [275]. ABCA12 deficiency caused foam cell formation and decreased reverse cholesterol transport in mice due to a posttranslational impact on ABCA1 [276]. Thus, as discussed for the NPC proteins, mechanisms underlying ABCA1 mediated cholesterol transport are complex and regulated on several levels. Further discussions of ABC transporters and scavenger receptors involved in intestinal cholesterol absorption and biliary cholesterol secretion can be found elsewhere [277, 278].

Acknowledgement

DW acknowledges funding from the Villum foundation and the Novo Nordisk foundation. References

[1] P. Linsel-Nitschke, A.R. Tall, HDL as a target in the treatment of artherosclerotic cardiovascular disease., Nat. Rev. Drug Disc., 4 (2005) 193-205.

[2] I. Tabas, Cholesterol in health and disease., J. Clin. Invest., 110 (2002) 583-590.

[3] J.L. Goldstein, M.S. Brown, The LDL receptor., Arterioscler. Thromb. Vasc. Biol. , 29 (2009) 431-438.

[4] A. Sokolov, A. Radhakrishnan, Accessibility of cholesterol in endoplasmic reticulum membranes and activation of SREBP-2 switch abruptly at a common cholesterol threshold, J Biol Chem, 285 (2010) 2948029490.

[5] X.X. Xu, I. Tabas, Lipoproteins activate acyl-coenzyme A:cholesterol acyltransferase in macrophages only after cellular cholesterol pools are expanded to a critical threshold level., J. Biol. Chem., 266 (1991) 1704017048.

[6] Y. Lange, J. Ye, M. Rigney, T.L. Steck., Regulation of endoplasmic reticulum cholesterol by plasma membrane cholesterol., J. Lipid Res., 40 (1999) 2264-2270.

[7] Y. Lange, D.S. Ory, J. Ye, M.H. Lanier, F.F. Hsu, T.L. Steck, Effectors of rapid homeostatic responses of endoplasmic reticulum cholesterol and 3-hydroxy-3-methylglutaryl-CoA reductase., J. Biol. Chem., 283 (2008) 1445-1555.

[8] J.L. Goldstein, R.A. DeBose-Boyd, M.S. Brown, Protein sensors for membrane sterols., Cell, 124 (2006) 35-46.

[9] Y. Lange, T.L. Steck, Cholesterol homeostasis and the escape tendency (activity) of plasma membrane cholesterol., Prog. Lipid Res., 47 (2008) 319-332.

[10] B. Mesmin, F.R. Maxfield, Intracellular sterol dynamics, Biochim Biophys Acta, 1791 (2009) 636-645.

[11] T.Y. Chang, B.L. Li, C.C. Chang, Y. Urano, Acyl-coenzyme A:cholesterol acyltransferases., Am. J. Physiol. Endocrinol. Metab. , 297 (2009) E1-9.

[12] F.R. Maxfield, D. Wüstner, Analysis of cholesterol trafficking with fluorescent probes., Methods Cell Biol., 108 (2012) 367-393.

[13] M. Hao, S.X. Lin, O.J. Karylowski, D. Wüstner, T.E. McGraw, F.R. Maxfield, Vesicular and non-vesicular sterol transport in living cells. The endocytic recycling compartment is a major sterol storage organelle., J. Biol. Chem., 277 (2002) 609-617.

[14] B. Mesmin, N.H. Pipalia, F.W. Lund, T.F. Ramlall, A. Sokolov, D. Eliezer, F.R. Maxfield, STARD4 abundance regulates sterol transport and sensing., Mol. Biol. Cell, 22 (2011) 4004-4015.

[15] L. Liscum, Pharmacological inhibition of the intracellular transport of low-density lipoprotein-derived cholesterol in Chinese hamster ovary cells., Biochim. et Biophys. Acta, 1045 (1990) 40-48.

[16] M.R. Kaplan, S. R.D., Transport of cholesterol from the endoplasmic reticulum to the plasma membrane., J. Cell Biol., 101 (1985) 446-453.

[17] D. Wüstner, A. Herrmann, M. Hao, F.R. Maxfield, Rapid nonvesicular transport of sterol between the plasma membrane domains of polarized hepatic cells., J. Biol. Chem., 277 (2002) 30325-30336.

[18] F.W. Lund, M.A. Lomholt, L.M. Solanko, R. Bittman, D. Wüstner, Two-Photon Time-Lapse Microscopy of BODIPY-Cholesterol Reveals Anomalous Sterol Diffusion in Chinese Hamster Ovary Cells. , BMC Biophysics., 5:20 (2012).

[19] L.M. Solanko, A. Honigmann, H.S. Midtiby, F.W. Lund, J.R. Brewer, V. Dekaris, R. Bittman, C. Eggeling, D. Wüstner, Membrane orientation and lateral diffusion of BODIPY-cholesterol as a function of probe structure. Biophys. J., 105 (2013) 2082-2092.

[20] R. Zwanzig, Nonequilibrium statistical mechanics., Oxford University Press, Oxford, UK, 2001.

[21] M.A. Digman, E. Gratton, Lessons in fluctuation correlation spectroscopy., Annu. Rev. Phys. Chem. , 62 (2011) 645-668.

[22] D.L. Kolin, P.W. Wiseman, Advances in image correlation spectroscopy: measuring number densities, aggregation states, and dynamics of fluorescently labeled macromolecules in cells., Cell Biochem. Biophys., 49 (2007) 141-164.

[23] M.J. Saxton, Lateral diffusion in an archipelago. , Biophys. J., 64 (1983) 1766-1780.

[24] M. Weiss, and Nilsson, T., In a mirro dimly: tracing the movements of molecules in living cells., Trends in Cell Biol., 14 (2004) 267-273.

[25] F.W. Lund, M.L. Jensen, T. Christensen, G.K. Nielsen, C.W. Heegaard, D. Wüstner, SpatTrack: An Imaging Toolbox for Analysis of Vesicle Motility and Distribution in Living Cells., Traffic, 15 (2014) 14061429.

[26] H. Qian, M.P. Sheetz, E.L. Elson, Single particle tracking. Analysis of diffusion and flow in two-dimensional systems., Biophys. J., 60 (1991) 910-921.

[27] F.W. Lund, D. Wüstner, A comparison of single particle tracking and temporal image correlation spectroscopy for quantitative analysis of endosome motility., J. Microscopy, 252 (2013) 169-188.

[28] R.F. DeGrella, and Simoni, R.D., Intracellular transport of cholesterol to the plasma membrane., J. Biol. Chem., 257 (1982) 14256-14262.

[29] H.S. Kruth, M.E. Comly, J.D. Butler, M.T. Vanier, J.K. Fink, D.A. Wenger, S. Patel, P.G. Pentchev, Type C Niemann-Pick disease. Abnormal metabolism of low density lipoprotein in homozygous and heterozygous fibroblasts., J. Biol. Chem., 261 (1986) 16769-16774.

[30] P.G. Pentchev, M.E. Comly, H.S. Kruth, M.T. Vanier, D.A. Wenger, S. Patel, R.O. Brady, A defect in cholesterol esterification in Niemann-Pick disease (type C) patients., Proc Natl Acad Sci U S A, 82 (1985) 8247-8251.

[31] M.P. Haynes, M.C. Phillips, G.H. Rothblat, Efflux of cholesterol from different cellular pools., Biochemistry, 39 (2000) 4508-4517.

[32] E.B. Neufeld, A.M. Cooney, J. Pitha, E.A. Dawidowicz, N.K. Dwyer, P.G. Pentchev, E.J. Blanchette-Mackie, Intracellular trafficking of cholesterol monitored with a cyclodextrin., J. Biol. Chem., 271 (1996) 21604-21613.

[33] M. Hölttä-Vuori, Tanhuanpää, K., Möbius, W., Somerharju, P., and Ikonen, E., Modulation of cellular cholesterol transport and homeostasis by Rab11., Mol. Biol. Cell, 13 (2002) 3107-3122.

[34] D. Wüstner, M. Mondal, I. Tabas, F.R. Maxfield, Direct observation of rapid internalization and intracellular transport of sterol by macrophage foam cells., Traffic, 6 (2005) 396-412.

[35] M. Hölttä-Vuori, R.L. Uronen, J. Repakova, E. Salonen, I. Vattulainen, P. Panula, Z. Li, R. Bittman, E. Ikonen, BODIPY-cholesterol: a new tool to visualize sterol trafficking in living cells and organisms., Traffic, 9 (2008)1839-1849.

[36] S.K. Basu, Goldstein, J.L., Anderson, R. G. W., and Brown, M.S., Degradatio n of cationized low density lipoprotein and regulation ofcholesterol metabolism in homozygous familial hypercholesterolemia fibroblasts., Proc. Natl. Acad. Sci. U. S. A., 73 (1976) 3178-3182.

[37] R.N. Ghosh, D.L. Gelman, F.R. Maxfield, Quantification of low density lipoprotein and transferrin endocytic sorting in HEp2 cells using confocal microscopy., J. Cell Sci., 107 (1994) 2177-2189.

[38] X. Buton, Z. Mamdouh, R.N. Ghosh, H. Du, G. Kuriakose, N. Beatini, G.A. Grabowski, F.R. Maxfield, I. Tabas, Unique cellular events occurring during the initial interaction of macrophages with matrix-retained or methylated aggregated low density lipoprotein (LDL). Prolonged cell-surface contact during which ldl-cholesteryl ester hydrolysis exceeds ldl protein degradation., J. Biol. Chem., 274 (1999) 32112-32121.

[39] Y. Lange, J. Ye, J. Chin, The Fate of Cholesterol Exiting Lysosomes., J. Biol. Chem., 272 (1997) 1701817022.

[40] J.L. Suhalim, C.Y. Chung, M.B. Lilledahl, R.S. Lim, M. Levi, B.J. Tromberg, E.O. Potma, Characterization of cholesterol crystals in atherosclerotic plaques using stimulated Raman scattering and second-harmonic generation microscopy., Biophys. J., 102 (2012) 1988-1995.

[41] D. Wüstner, Fluorescent sterols as tools in membrane biophysics and cell biology., Chem. Phys. Lipids, 146 (2007) 1-25.

[42] D. Wüstner, L.M. Solanko, E. Sokol, F.W. Lund, O. Garvik, Z. Li, R. Bittman, T. Korte, A. Herrmann, Quantitative Assessment of Sterol Traffic in Living Cells by Dual Labeling with Dehydroergosterol and BODIPY-cholesterol., Chem. Phys. Lipids, 164 (2011) 221-235.

[43] S. Milles, T. Meyer, H.A. Scheidt, R. Schwarzer, L. Thomas, M. Marek, L. Szente, R. Bittman, A. Herrmann, T. Günther Pomorski, D. Huster, P. Müller, Organization of fluorescent cholesterol analogs in lipid bilayers - Lessons from cyclodextrin extraction., Biochim Biophys Acta, 1828 (2013) 1822-1828.

[44] H.A. Scheidt, P. Müller, A. Herrmann, D. Huster, The potential of fluorescent and spin-labeled steroid analogs to mimic natural cholesterol., J. Biol. Chem., 278 (2003) 45563-45569.

[45] N. Hartwig Petersen, N.J. Fœrgeman, L. Yu, D. Wüstner, Kinetic imaging of NPC1L1 and sterol trafficking between plasma membrane and recycling endosomes in hepatoma cells., J. Lipid Res., 49 (2008) 20232037.

[46] D. Wüstner, N.J. Fœrgeman, Spatiotemporal analysis of endocytosis and membrane distribution of fluorescent sterols in living cells., Histochem. Cell Biol., 130 (2008) 891-908.

[47] K. Hofmann, C. Thiele, H.F. Schött, A. Gaebler, M. Schoene, Y. Kiver, S. Friedrichs, D. Lütjohann, L. Kuerschner, A novel alkyne cholesterol to trace cellular cholesterol metabolism and localization., J. Lipid Res., 55 (2014) 583-591.

[48] S.M. Peyrot, S. Nachtergaele, G. Luchetti, L.K. Mydock-McGrane, H. Fujiwara, D. Scherrer, A. Jallouk, P.H. Schlesinger, D.S. Ory, D.F. Covey, R. Rohatgi, Tracking the Subcellular Fate of 20(S)-Hydroxycholesterol with Click Chemistry Reveals a Transport Pathway to the Golgi, J. Biol. Chem., 289 (2014) 11095-11110.

[49] G. Gimpl, Cholesterol-protein interaction: methods and cholesterol reporter molecules, Subcell Biochem, 51 (2010) 1-45.

[50] D. Wüstner, Following intracellular cholesterol transport by linear and non-linear optical microscopy of intrinsically fluorescent sterols., Curr. Pharm. Biotechnol., 13 (2012) 303-318.

[51] L. Kuerschner, C. Thiele, Multiple bonds for the lipid interest., Biochim. Biophys. Acta, 1841 (2014) 1031-1037.

[52] G. Nemecz, F. Schroeder, Selective binding of cholesterol by recombinant fatty acid binding proteins., J. Biol. Chem., 266 (1991) 17180-17186.

[53] F. Schroeder, P. Butko, G. Nemecz, T.J. Scallen, Interaction of fluorescent delta 5,7,9(11),22-ergostatetraen-3 beta-ol with sterol carrier protein-2., J. Biol. Chem., 265 (1990) 151-157.

[54] R.E. Infante, L. Abi-Mosleh, A. Radhakrishnan, J.D. Dale, M.S. Brown, J.L. Goldstein, Purified NPC1 protein. I. Binding of cholesterol and oxysterols to a 1278-amino acid membrane protein., J. Biol. Chem., 283 (2008) 1052-1063.

[55] D.C. Ko, J. Binkley, A. Sidow, M.P. Scott, The integrity of a cholesterol-binding pocket in Niemann-Pick C2 protein is necessary to control lysosome cholesterol levels., Proc. Natl. Acad. Sci. U. S. A., 100 (2003) 2518-2525.

[56] B. Mesmin, B. Antonny, G. Drin, Insights into the mechanisms of sterol transport between organelles., Cell. Mol. Life Sci., 70 (2013) 3405-3421.

[57] S. Lev, Non-vesicular lipid transport by lipid-transfer proteins and beyond., Nat. Rev. Mol. Cell Biol. , 11

(2010) 739-750.

[58] V. Mikes, M.L. Milat, M. Ponchet, F. Panabieres, P. Ricci, J.P. Blein, Elicitins, proteinaceous elicitors of plant defense, are a new class of sterol carrier proteins., Biochem. Biophys. Res. Commun., 245 (1998) 133139.

[59] J. Storch, Niemann-Pick C2 (NPC2) and intracellular cholesterol trafficking., 1791 (2009) 671-678.

[60] C.S. Cheng, D. Samuel, Y.J. Liu, J.C. Shyu, S.M. Lai, K.F. Lin, P.C. Lyu, Binding mechanism of nonspecific lipid transfer proteins and their role in plant defense., Biochemistry, 43 (2004) 13628-13636.

[61] N. Stolowich, Frolov, A., Petrescu, A.D., Scott, A.I., Billheimer, J.T., and Schroeder, F., Holo-sterol carrier protein-2. (13)C NMR investigation of cholesterol and fatty acid binding sites., J. Biol. Chem., 274 (1999) 35425-35433.

[62] R. Liu, P. Lu, J.W. Chu, F.J. Sharom, Characterization of fluorescent sterol binding to purified human NPC1., J. Biol. Chem., 284 (2009) 1840-1852.

[63] H.J. Kwon, L. Abi-Mosleh, M.L. Wang, J. Deisenhofer, J.L. Goldstein, M.S. Brown, R.E. Infante, Structure of N-terminal domain of NPC1 reveals distinct subdomains for binding and transfer of cholesterol., Cell, 137 (2009) 1213-1224.

[64] S. Xu, B. Benoff, H.L. Liou, P. Lobel, A.M. Stock, Structural basis of sterol binding by NPC2, a lysosomal protein deficient in Niemann-Pick type C2 disease., J. Biol. Chem., 282 (2007) 23525-23531.

[65] N. Friedland, Liou, H-L., Lobel, P., and Stock, A.M., Structure of a cholesterol-binding protein deficient in Niemann-Pick type C2 disease., Proc. Natl. Acad. Sci. U. S. A., 100 (2003) 2512-2517.

[66] H.L. Liou, S.S. Dixit, S. Xu, G.S. Tint, A.M. Stock, P. Lobel, NPC2, the protein deficient in Niemann-Pick C2 disease, consists of multiple glycoforms that bind a variety of sterols., J. Biol. Chem., 281 (2006) 3671036723.

[67] F. Schroeder, M.E. Dempsey, R.T. Fischer, Sterol and squalene carrier protein interactions with fluorescent delta 5,7,9(11)-cholestatrien-3 beta-ol, J. Biol. Chem., 260 (1985) 2904-2911.

[68] B. Wiltschi, M. Schober, S.D. Kohlwein, D. Oesterhelt, E.K. Sinner, Sterol binding assay using surface plasmon fluorescence spectroscopy., Anal. Biochem., 78 (2006) 547-555.

[69] M. de Saint-Jean, V. Delfosse, D. Douguet, G. Chicanne, B. Payrastre, W. Bourguet, B. Antonny, G. Drin, Osh4p exchanges sterols for phosphatidylinositol 4-phosphate between lipid bilayers., J. Cell Biol., 195

(2011) 965-978.

[70] S. Vauthrin, V. Mikes, M.L. Milat, M. Ponchet, B. Maume, H. Osman, J.P. Blein, Elicitins trap and transfer sterols from micelles, liposomes and plant plasma membranes., Biochim. Biophys. Acta, 1419 (1999) 335-342.

[71] C. Assanasen, Mineo, C., Seetharam, D., Yuhanna, I.S., Marcel, Y.L., Connelly, M.A., Williams, D.L., de la Llera-Moya, M., Shaul, P.W., and Silver, D.L., Cholesterol binding, efflux, and a PDZ-interacting domain of scavenger receptor-BI mediate HDL-initiated signaling., J. Clin. Invest., 115 (2005) 969-977.

[72] J.J. Hulce, A.B. Cognetta, M.J. Niphakis, S.E. Tully, B.F. Cravatt, Proteome-wide mapping of cholesterol-interacting proteins in mammalian cells., Nat. Methods, 10 (2013) 259-264.

[73] E. Reboul, F.M. Dyka, F. Quazi, R.S. Molday, Cholesterol transport via ABCA1: new insights from solidphase binding assay., Biochimie, 95 (2013) 957-961.

[74] M.A. Castanho, W. Brown, M.J. Prieto, Rod-like cholesterol micelles in aqueous solution studied using polarized and depolarized dynamic light scattering., Biophys. J., 63 (1992) 1455-1461.

[75] L.M. Loura, M. Prieto, Dehydroergosterol structural organization in aqueous medium and in a model system of membranes., Biophys. J., 72 (1997) 2226-2236.

[76] A. Roostaee, E. Barbar, J.G. Lehoux, P. Lavigne, Cholesterol binding is a prerequisite for the activity of the steroidogenic acute regulatory protein (StAR). Biochem. J., 412 (2008) 553-562.

[77] Y. Tsujishita, J.H. Hurley, Structure and lipid transport mechanism of a StAR-related domain., Nat. Struct. Biol., 7 (2000) 408-414.

[78] A.D. Petrescu, A.M. Gallegos, Y. Okamura, r. Strauss, J.F.,, F. Schroeder, Steroidogenic acute regulatory protein binds cholesterol and modulates mitochondrial membrane sterol domain dynamics., J. Biol. Chem., 276 (2001) 36970-36982.

[79] K. John, J. Kubelt, P. Muller, D. Wustner, A. Herrmann, Rapid transbilayer movement of the fluorescent sterol dehydroergosterol in lipid membranes, Biophysical Journal, 83 (2002) 1525-1534.

[80] Z. Xu, W. Farver, S. Kodukula, J. Storch, Regulation of sterol transport between membranes and NPC2., Biochmistry, 47 (2008) 11134-11143.

[81] B. Mesmin, J. Bigay, J. Moser von Filseck, S. Lacas-Gervais, G. Drin, B. Antonny, A four-step cycle driven by PI(4)P hydrolysis directs sterol/PI(4)P exchange by the ER-Golgi tether OSBP., Cell, 155 (2013) 830-843.

[82] X. Liu, N.D. Ridgway, Characterization of the sterol and phosphatidylinositol 4-phosphate binding properties of Golgi-associated OSBP-related protein 9 (ORP9). PLoS One, 9 (2014 ) e108368.

[83] M. Pourmousa, T. Rog, R. Mikkeli, I. Vattulainen, L. Solanko, D. Wüstner, N. List, J. Kongsted, M. Karttunen, Dehydroergosterol as an analogue for cholesterol: why it mimics cholesterol so well or does It?, J. Phys. Chem., In press. (2014).

[84] A. Frolov, J.K. Woodford, E.J. Murphy, J.T. Billheimer, F. Schroeder, Spontaneous and protein-mediated sterol transfer between intracellular membranes., J. Biol. Chem., 271 (1996) 16075-16083.

[85] A. Frolov, J.K. Woodford, E.J. Murphy, J.T. Billheimer, F. Schroeder, Fibroblast membrane sterol kinetic domains: modulation by sterol carrier protein-2 and liver fatty acid binding protein., J. Lipid Res., 37 (1996) 1862-1874.

[86] T.K. Nyholm, P.M. Grandell, B. Westerlund, J.P. Slotte, Sterol affinity for bilayer membranes is affected by their ceramide content and the ceramide chain length., Biochim. Biophys. Acta, 1798 (2010) 1008-1013.

[87] J.M. Backer, E.A. Dawidowicz, Mechanism of cholesterol exchange between phospholipid vesicles., Biochemistry, 20 (1981) 3805-3810.

[88] J.O. Babalola, M. Wendeler, B. Breiden, C. Arenz, G. Schwarzmann, S. Locatelli-Hoops, K. Sandhoff, Development of an assay for the intermembrane transfer of cholesterol by Niemann-Pick C2 protein., Biol. Chem., 388 (2007) 617-626.

[89] B.W. Wattenberg, D.F. Silbert, Sterol partitioning among intracellular membranes. Testing a model for cellular sterol distribution., J. Biol. Chem., 258 (1983) 2284-2289.

[90] S.-L. Niu, Litman, B.J., Determination of membrane cholesterol partition coefficient using a lipid vesicle-cyclodextrin binary system: effect of phospholipid acyl chain unsaturation and headgroup composition., Biophys. J., 83 (2002) 3408-3415.

[91] C. Bhuvaneswaran, K.A. Mitropoulos, Effect of liposomal phospholipid composition on cholesterol transfer between microsomal and liposomal vesicles., Biochem. J., 238 (1986) 647-652.

[92] B. Bloj, D.B. Zilversmit, Complete exchangeability of cholesterol in phosphatidylcholine/cholesterol vesicles of different degrees of unsaturation., Biochemistry, 16 (1977) 3943-3948.

[93] M. Lönnfors, J.P. Doux, J.A. Killian, T.K. Nyholm, J.P. Slotte, Sterols have higher affinity for sphingomyelin than for phosphatidylcholine bilayers even at equal acyl-chain order., Biophys. J., 100 (2011) 2633-2641.

[94] J.M. Holopainen, A.J. Metso, J.P. Mattila, A. Jutila, P.K. Kinnunen, Evidence for the lack of a specific interaction between cholesterol and sphingomyelin., Biophys. J., 86 (2004) 1510-1520.

[95] L.M. Estronca, M.J. Moreno, W.L. Vaz, Kinetics and thermodynamics of the association of dehydroergosterol with lipid bilayer membranes., Biophys. J., 93 (2007) 4244-4253.

[96] T.L. Steck, F.J. Kezdy, Y. Lange, An activation-collision mechanism for cholesterol transfer between membranes., J. Biol. Chem., 263 (1988) 13023-13031.

[97] E. Sackmann, R. Merkel, Lehrbuch der Biophysik, Wiley VCH Weinheim, 2010.

[98] S. Mitragotri, M.E. Johnson, D. Blankschtein, R. Langer, An analysis of the size selectivity of solute partitioning, diffusion and permeation across lipid bilayers., Biophys. J., 77 (1999) 1268-1283.

[99] E. Falck, M. Patra, M. Karttunen, M.T. Hyvonen, I. Vattulainen, Impact of cholesterol on voids in phospholipid membranes., J. Chem. Phys., 121 (2004) 12676-12689.

[100] A.V. Krylov, P. Pohl, M.L. Zeidel, W.G. Hill, Water permeability of asymmetric planar lipid bilayers: leaflets of different composition offer independent and additive resistance to permeation., J. Gen. Physiol., 118 (2001) 333-339.

[101] W.K. Subczynski, J.S. Hyde, A. Kusumi, Oxygen permeability of phosphatidylcholine--cholesterol membranes., Proc. Natl. Acad. Sci. U. S. A., 86 (1989) 4474-4478.

[102] M.R. Vist, J.H. Davis, Phase equilibria of cholesterol/dipalmitoylphosphatidylcholine mixtures: 2H nuclear magnetic resonance and differential scanning calorimetry., Biochemistry, 29 (1990) 451-464.

[103] J.H. Ipsen, G. Karlstrom, O.G. Mouritsen, H. Wennerstrom, M.J. Zuckermann, Phase equilibria in the phosphatidylcholine-cholesterol system., Biochim. Biophys. Acta, 905 (1987) 162-172.

[104] J.H. Ipsen, O.G. Mouritsen, M. Bloom, Relationships between lipid membrane area, hydrophobic thickness, and acyl-chain orientational order. The effects of cholesterol., Biophys. J., 57 (1990) 405-412.

[105] P.L. Yeagle, Cholesterol and the cell membrane., Biochim. Biophys. Acta, 822 (1985) 267-287.

[106] R. Leventis, J.R. Silvius, Use of cyclodextrins to monitor transbilayer movement and differential lipid affinities of cholesterol., Biophys. J., 81 (2001) 2257-2267.

[107] J. Aittoniemi, T. Rog, P. Niemela, M. Pasenkiewicz-Gierula, M. Karttunen, I. Vattulainen, Tilt: major factor in sterols' ordering capability in membranes., J. Phys. Chem. B, 110 (2006) 25562-25564.

[108] G. Khelashvili, D. Harries, How cholesterol tilt modulates the mechanical properties of saturated and unsaturated lipid membranes., J. Phys. Chem. B, 117 (2013) 2411-2421.

[109] 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).

[110] M.O. Jensen, O.G. Mouritsen, Lipids do influence protein function-the hydrophobic matching hypothesis revisited., Biochim. Biophys. Acta, 1666 (2004) 205-226.

[111] W.F. Bennett, J.L. MacCallum, M.J. Hinner, S.J. Marrink, D.P. Tieleman, Molecular view of cholesterol flip-flop and chemical potential in different membrane environments., J. Am. Chem. Soc., 131 (2009) 12714-12720.

[112] S. Jo, H. Rui, J.B. Lim, J.B. Klauda, W. Im, Cholesterol flip-flop: insights from free energy simulation studies., J. Phys. Chem. B, 114 (2010) 13342-13348.

[113] T. Rog, L.M. Stimson, M. Pasenkiewicz-Gierula, I. Vattulainen, M. Karttunen, Replacing the cholesterol hydroxyl group with the ketone group facilitates sterol flip-flop and promotes membrane fluidity., J. Phys. Chem. B, 112 (2008) 1946-1952.

[114] A. Kessel, N. Ben-Tal, S. May, Interactions of cholesterol with lipid bilayers: the preferred configuration and fluctuations., Biophys. J., 81 (2001) 643-658.

[115] A. Tsamaloukas, H. Szadkowska, P.J. Slotte, H. Heerklotz, Interactions of cholesterol with lipid membranes and cyclodextrin characterized by calorimetry., Biophys. J., 89 (2005) 1109-1119.

[116] Z. Zhang, L. Lu, M.L. Berkowitz, Energetics of cholesterol transfer between lipid bilayers., J. Phys. Chem. B, 112 (2008) 3807-3811.

[117] J.H. Nystrom, M. Lonnfors, T.K. Nyholm, Transmembrane peptides influence the affinity of sterols for phospholipid bilayers., Biophys. J., 99 (2010) 526-533.

[118] Y. Lange, J. Ye, T.L. Steck, How cholesterol homeostasis is regulated by plasma membrane cholesterol in excess of phospholipids., Proc. Natl. Acad. Sci. U. S. A., 101 (2004) 11664-11667.

[119] P. Somerharju, J.A. Virtanen, K.H. Cheng, Lateral organisation of membrane lipids. The superlattice view., Biochim. Biophys. Acta, 1440 (1999) 32-48.

[120] J.P. Blein, P. Coutos-Thevenot, D. Marion, M. Ponchet, From elicitins to lipid-transfer proteins: a new insight in cell signalling involved in plant defence mechanisms., Trends Plant Sci., 7 (2002) 293-296.

[121] J. Huang., J.T. Buboltz, G.W. Feigenson, Maximum solubility of cholesterol in phosphatidylcholine and phosphatidylethanolamine bilayers., Biochim. et Biophys. Acta, 1417 (1999) 89-100.

[122] J. Huang., G.W. Feigenson, A microscopic interaction model of maximum solubility of cholesterol in lipid bilayers., Biophys. J., 76 (1999) 2142-2157.

[123] M.R. Ali, K.H. Cheng, J. Huang, Assess the nature of cholesterol-lipid interactions through the chemical potential of cholesterol in phosphatidylcholine bilayers., Proc. Natl. Acad. Sci. USA, 104 (2007) 5372-5377.

[124] A. Radhakrishnan, H. McConnell, Condensed complexes in vesicles containing cholesterol and phospholipids., Proc. Natl. Acad. Sci. USA, 102 (2005) 12662-12666.

[125] H. McConnell, A. Radhakrishnan, Theory of the deuterium NMR of sterol-phospholipid membranes., Proc. Natl. Acad. Sci. USA, 103 (2006) 1184-1189.

[126] H.M. McConnell, and Radhakrishnan, A., Condensed complexes of cholesterol and phospholipids., Biochim. et Biophys. Acta, 1610 (2003) 159-173.

[127] A. Radhakrishnan, H.M. McConnell, Chemical activity of cholesterol in membranes., Biochemistry, 39 (2000) 8119-8124.

[128] F.R. Maxfield, A.K. Menon, Intracellular sterol transport and distribution., Curr. Opin. Cell Biol., 18 (2006) 379-385.

[129] T.L. Steck, Y. Lange, Cell cholesterol homeostasis: mediation by active cholesterol., Trends Cell Biol., 20 (2010) 680-687.

[130] C.T. Beh, C.R. McMaster, K.G. Kozminski, A.K. Menon, A Detour for Yeast Oxysterol-Binding Proteins., J. Biol. Chem., In press. (2012).

[131] E. Ikonen, Cellular cholesterol trafficking and compartmentalization., Nat. Rev. Mol. Cell Biol., 9 (2008) 125-138.

[132] D. Wüstner, Plasma membrane sterol distribution resembles the surface topography of living cells., Mol. Biol. Cell, 18 (2007) 211-228.

[133] D. Wüstner, Free-cholesterol loading does not trigger phase separation of the fluorescent sterol dehydroergosterol in the plasma membrane of macrophages., Chem. Phys. Lipids, 154 (2008) 129-136.

[134] J.F. Frisz, H.A. Klitzing, K. Lou, I.D. Hutcheon, P.K. Weber, J. Zimmerberg, M.L. Kraft, Sphingolipid domains in the plasma membranes of fibroblasts are not enriched with cholesterol., J. Biol. Chem., 288 (2013) 16855-16861.

[135] A. Honigmann, V. Mueller, H. Ta, A. Schoenle, E. Sezgin, S.W. Hell, C. Eggeling, Scanning STED-FCS reveals spatiotemporal heterogeneity of lipid interaction in the plasma membrane of living cells., Nat. Commun., 5 (2014) 5412.

[136] V. Mueller, C. Ringemann, A. Honigmann, G. Schwarzmann, R. Medda, M. Leutenegger, S. Polyakova, V.N. Belov, S.W. Hell, C. Eggeling, STED nanoscopy reveals molecular details of cholesterol- and cytoskeleton-modulated lipid interactions in living cells., Biophys. J., 101 (2011) 1651-1660.

[137] K. John, J. Kubelt, P. Müller, D. Wüstner, A. Herrmann, Rapid transbilayer movement of the fluorescent sterol dehydroergosterol in lipid membranes., Biophys. J., 83 (2002) 1525-1534.

[138] M. Mondal, B. Mesmin, S. Mukherjee, F.R. Maxfield, Sterols are mainly in the cytoplasmic leaflet of the plasma membrane and the endocytic recycling compartment in CHO cells., Mol. Biol. Cell, 20 (2009) 581-588.

[139] Y. Shiratori, Okwu, A.K., and Tabas, I., Free cholesterol loading of macrophages stimulates phosphatidylcholine biosynthesis and up-regulation of CTP: phosphocholine cytidylyltransferase., J. Biol. Chem., 269 (1994) 11337-11348.

[140] G. Kellner-Weibel, Yancey, P.G., Jerome, W.G., Walser, T., Mason, R.P., Phillips, M.C., and Rothblat, G.H., Crystallization of free cholesterol in model macrophage foam cells., Arterioscler. Thromb. Vasc. Biol., 19(1999)1891-1898.

[141] R.K. Tangirala, Jerome, W.G., Jones, N.L., Small, D.M., Johnson, W.J., Glick, J.M., Mahlberg, F.H., and Rothblat, G.H., Formation of cholesterol monohydrate crystals in macrophage-derived foam cells., J. Lipid Res., 35 (1994) 93-104.

[142] C. Gliss, O. Randel, H. Casalta, E. Sackmann, R. Zorn, T. Bayerl, Anisotropic motion of cholesterol in oriented DPPC bilayers studied by quasielastic neutron scattering: the liquid-ordered phase., Biophys. J., 77 (1999) 331-340.

[143] B.N. Olsen, A.A. Bielska, T. Lee, M.D. Daily, D.F. Covey, P.H. Schlesinger, N.A. Baker, D.S. Ory, The structural basis of cholesterol accessibility in membranes., Biophys. J., 105 (2013) 1838-1847.

[144] S.A. Pandit, G. Khelashvili, E. Jakobsson, A. Grama, H.L. Scott, Lateral organization in lipid-cholesterol mixed bilayers., Biophys. J., 92 (2007) 440-447.

[145] R. Elliott, I. Szleifer, M. Schick, Phase diagram of a ternary mixture of cholesterol and saturated and unsaturated lipids calculated from a microscopic model., Phys. Rev. Letters, 96 (2006) 098101.

[146] P.F. Almeida, A simple thermodynamic model of the liquid-ordered state and the interactions between phospholipids and cholesterol., Biophys. J., 100 (2011) 420-429.

[147] J.J. Flanagan, R.K. Tweten, A.E. Johnson, A.P. Heuck, Cholesterol exposure at the membrane surface is necessary and sufficient to trigger Perifringolysin O binding., Biochemistry, 48 (2009) 3977-3987.

[148] A. Das, J.L. Goldstein, D.D. Anderson, M.S. Brown, A. Radhakrishnan, Use of mutant 125I-perfringolysin O to probe transport and organization of cholesterol in membranes of animal cells., Proc. Natl. Acad. Sci. USA, 110 (2013) 10580-10585.

[149] A. Gay, D. Rye, A. Radhakrishnan, Switch-like Responses of Two Cholesterol Sensors Do Not Require Protein Oligomerization in Membranes., Biophys. J., 108 (2015) 1459-1469.

[150] A. Radhakrishnan, J.L. Goldstein, J.G. McDonald, M.S. Brown, Switch-like control of SREBP-2 transport triggered by small changes in ER cholesterol: a delicate balance., Cell Metab., 8 (2008) 512-521.

[151] Y. Li, M. Ge, L. Ciani, G. Kuriakose, E.J. Westover, M. Dura, D.F. Covey, J.H. Freed, F.R. Maxfield, J. Lytton, I. Tabas, Enrichment of endoplasmic reticulum with cholesterol inhibits sarcoplasmic-endoplasmic reticulum calcium ATPase-2b activity in parallel with increased order of membrane lipids: implications for depletion of endoplasmic reticulum calcium stores and apoptosis in cholesterol-loaded macrophages., J. Biol. Chem., 279 (2004) 37030-37039.

[152] O.V. Krasilnikov, L.N. Yuldasheva, Transmembrane cholesterol migration in planar lipid membranes measured with Vibrio cholerae cytolysin as molecular tool., Biochimie, 91 (2009) 620-623.

[153] R. Rukmini, Rawat, S.S., Biswas, S.C., and Chattopadhyay, A., Cholesterol organization in membranes at low concentrations: effects of curvature stress and membrane thickness., Biophys. J., 81 (2001) 21222134.

[154] J.S. Harris, D.E. Epps, S.R. Davio, F.J. Kezdy, Evidence for transbilayer, tail-to-tail cholesterol dimers in dipalmitoylglycerophosphocholine liposomes., Biochemistry, 34 (1995) 3851-3857.

[155] P.J. Skiba, Zha, X., Maxfield, F.R., Schissel, S.L., and Tabas, I., The distal pathway of lipoprotein-induced cholesterol esterification, but not sphingomyelinase-induced cholesterol esterification, is energy-dependent., J. Biol. Chem., 271 (1996) 13392-13400.

[156] J.P. Slotte, and Bierman, E.L., Depletion of plasma-membrane sphingomyelin rapidly alters the distribution of cholesterol between plasma membranes and intracellular cholesterol pools in cultured fibroblasts., Biochem. J., 250 (1988) 653-658.

[157] H. Ohvo, C. Olsio, J.P. Slotte, Effects of sphingomyelin and phosphatidylcholine degradation on cyclodextrin-mediated cholesterol efflux in cultured fibroblasts., Biochim. Biophys. Acta, 1349 (1997) 131141.

[158] X. Zha, Pierini, L.M., Leopold, P.L., Skiba, P.J., Tabas, I., and F.R. Maxfield., Sphingomyelinase treatment induces ATP-independent endocytosis., J. Cell Biol., 140 (1998) 39-47.

[159] H. Ohvo-Rekila, B. Ramstedt, P. Leppimaki, J.P. Slotte, Cholesterol interactions with phospholipids in membranes., Prog. Lipid Res., 41 (2002) 66-97.

[160] M. London, E. London, Ceramide Selectively Displaces Cholesterol from Ordered Lipid Domains (Rafts) Implications for lipid raft structure and function., J. Biol. Chem., 279 (2004) 9997-10004.

[161] M.R. Ali, K.H. Cheng, J. Huang, Ceramide drives cholesterol out of the ordered lipid bilayer phase into the crystal phase in 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine/cholesterol/ceramide ternary mixtures., Biochemistry, 45 (2006) 12629-12638.

[162] A. Das, M.S. Brown, D.D. Anderson, J.L. Goldstein, A. Radhakrishnan, Three pools of plasma membrane cholesterol and their relation to cholesterol homeostasis., eLife, 3 (2014) eLife.02882.

[163] S. Munro, Lipid rafts: elusive or illusive?, Cell, 115 (2003) 377-388.

[164] M.L. Kraft, Plasma membrane organization and function: moving past lipid rafts., Mol. Biol. Cell, 24 (2013) 2765-2768.

[165] J.E. Hale, F. Schroeder, Asymmetric transbilayer distribution of sterol across plasma membranes determined by fluorescence quenching of dehydroergosterol., Eur. J. Biochem., 122 (1982) 649-661.

[166] F. Schroeder, G. Nemecz, W.G. Wood, C. Joiner, G. Morrot, M. Ayraut-Jarrier, P.F. Devaux, Transmembrane distribution of sterol in the human erythrocyte., Biochim. Biophys. Acta., 1066 (1991) 183192.

[167] D.L. Brasaemle, A.D. Robertson, A.D. Attie, Transbilayer movement of cholesterol in the human erythrocyte membrane., J. Lipid Res., 29 (1988) 481-489.

[168] H. Giang, M. Schick, How cholesterol could Be drawn to the cytoplasmic leaf of the plasma membrane by phosphatidylethanolamine, Biophys. J., 107 (2014) 2337-2344.

[169] R. Heinrich, Brumen, M., Jaeger, A., Müller, P., and Herrmann, A., Modelling of phospholipid translocation in the erythrocyte membrane: a combined kinetic and thermodynamic approach., J. Theor. Biol., 185 (1997) 295-312.

[170] R.J. Bruckner, S.S. Mansy, A. Ricardo, L. Mahadevan, J.W. Szostak, Flip-Flop-Induced Relaxation of Bending Energy: Implications for Membrane Remodeling Biophys. J., 97 (2009) 3113-3122.

[171] H.A. Scheidt, T. Meyer, J. Nikolaus, J.B. Dong, R. Bittman, P. Müller, A. Herrmann, D. Huster, Cholesterol's aliphatic side chain modulates membrane properties., Angew. Chem., 52 (2013) 12848-12851.

[172] R. Johnson, Hamilton, J.A., Worgall, T.S., and Deckelbaum, R.J., Free fatty acids modulate intermembrane trafficking of cholesterol by increasing lipid mobilities: novel 13C NMR analyses of free cholesterol partitioning., Biochemistry, 42 (2003) 1637-1645.

[173] S. Reiner, D. Micolod, R. Schneiter, Saccharomyces cerevisiae, a model to study sterol uptake and transport in eukaryotes., Biochem. Soc. Trans., 33 (2005) 1186-1188.

[174] R.E. Soccio, J. Breslow, StAR-related lipid transfer (START) proteins: mediators of intracellular lipid metabolism., J. Biol. Chem., 278 (2003) 22183-22186.

[175] A. Rigotti, D.E. Cohen, S. Zanlungo, STARTing to understand MLN64 function in cholesterol transport., J. Lipid Res., 51 (2010) 2015-2017.

[176] M. Charman, B.E. Kennedy, N. Osborne, B. Karten, MLN64 mediates egress of cholesterol from endosomes to mitochondria in the absence of functional Niemann-Pick Type C1 protein., J. Lipid Res., 51 (2010) 1023-1034.

[177] H. Watari, F. Arakane, C. Moog-Lutz, C.B. Kallen, C. Tomasetto, G.L. Gerton, M.C. Rio, M.E. Baker, J.F. Strauss, 3rd, MLN64 contains a domain with homology to the steroidogenic acute regulatory protein (StAR) that stimulates steroidogenesis, Proceedings of the National Academy of Sciences of the United States of America, 94 (1997) 8462-8467.

[178] S. Heino, S. Lusa, P. Somerharju, C. Ehnholm, V.M. Olkkonen, E. Ikonen, Dissecting the role of the golgi complex and lipid rafts in biosynthetic transport of cholesterol to the cell surface., Proc. Natl. Acad. Sci. U. S. A., 97 (2000) 8375-8380.

[179] L. Urbani, R.D. Simoni, Cholesterol and vesicular stomatitis virus G protein take separate routes from the endoplasmic reticulum to the plasma membrane., J. Biol. Chem., 265 (1990) 1919-1923.

[180] R.A. Weisiger, S.D. Zucker, Transfer of fatty acids between intracellular membranes: role of soluble binding proteins, distance and time., Am. J. Physiol. Gastrointest. Liver Physiol., 282 (2002) G105-G115.

[181] V.M. Olkkonen, In: Cellular lipid metabolism., Ed. Ehnholm, C., Springer press. (2009).

[182] F. Alpy, A. Rousseau, Y. Schwab, F. Legueux, I. Stoll, C. Wendling, C. Spiegelhalter, P. Kessler, C. Mathelin, M.C. Rio, T.P. Levine, C. Tomasetto, STARD3 or STARD3NL and VAP form a novel molecular tether between late endosomes and the ER., J. Cell Sci., 126 (2013) 5500-5512.

[183] J.R. Jefferson, J.P. Slotte, G. Nemecz, A. Pastuszyn, T.J. Scallen, F. Schroeder, Intracellular sterol distribution in transfected mouse L-cell fibroblasts expressing rat liver fatty acid-binding protein., J. Biol. Chem., 266 (1991) 5486-5496.

[184] B.P. Atshaves, O. Starodub, A. Mcintosh, A. Petrescu, J.B. Roths, A.B. Kier, F. Schroeder, Sterol carrier protein-2 alters high density lipoprotein-mediated cholesterol efflux., J. Biol. Chem., 275 (2000) 3685236861.

[185] M. Jansen, Y. Ohsaki, L. Rita Rega, R. Bittman, V.M. Olkkonen, E. Ikonen, Role of ORPs in sterol transport from plasma membrane to ER and lipid droplets in mammalian cells., Traffic, 12 (2011) 218-231.

[186] J. Garbarino, M. Pan, H.F. Chin, F.W. Lund, F.R. Maxfield, J.L. Breslow, STARD4 knockdown in HepG2 cells disrupts cholesterol trafficking associated with the plasma membrane, ER, and ERC., J. Lipid Res., 53 (2012) 2716-2725.

[187] P. Butko, I. Hapala, T.J. Scallen, F. Schroeder, Acidic phospholipids strikingly potentiate sterol carrier protein 2 mediated intermembrane sterol transfer., Biochemistry, 29 (1990) 4070-4077.

[188] R.A. Weisiger, Cytosolic fatty acid binding proteins catalyze two distinct steps in intracellular transport of their ligands., Mol. Cell. Biochem. , 239 (2002) 35-42.

[189] J.A. Hamilton, Fast flip-flop of cholesterol and fatty acids in membranes: implications for membrane transport proteins., Curr. Opin. Lipidol., 14 (2003) 263-271.

[190] C.A. López, A.H. de Vries, S.J. Marrink, Computational microscopy of cyclodextrin mediated cholesterol extraction from lipid model membranes., Sci. Rep., 3 (2013) srep02071.

[191] S.R. Cheruku, Z. Xu, R. Dutia, P. Lobel, J. Storch, Mechanism of cholesterol transfer from the Niemann-Pick type C2 protein to model membranes supports a role in lysosomal cholesterol transport., J. Biol. Chem., 281 (2006) 31594-31604.

[192] L.A. McCauliff, Z. Xu, J. Storch, Sterol transfer between cyclodextrin and membranes: similar but not identical mechanism to NPC2-mediated cholesterol transfer., Biochemistry, 50 (2011) 7341-7349.

[193] L. Ge, W. Qi, H.H. Miao, J. Cao, Y.X. Qu, B.L. Li, B.L. Song, The cholesterol absorption inhibitor ezetimibe acts by blocking the sterol-induced internalization of NPC1L1., Cell Metabolism, 7 (2008) 508519.

[194] J.P. Davies, C. Scott, K. Oishi, A. Liapis, Y.A. ioannou, Inactivation of NPC1L1 causes multiple lipid transport defects and protects against diet-induced hypercholesterolemia., J. Biol. Chem., 280 (2005) 12710-12720.

[195] M. Garcia-Calvo, J. Lisnock, H.G. Bull, B.E. Hawes, D.A. Burnett, M.P. Braun, J.H. Crona, H.R.J. Davis, D.C. Dean, P.A. Detmers, M.P. Graziano, M. Hughes, D.E. Macintyre, A. Ogawa, K.A. O'neill, S.P. Iyer, D.E. Shevell, M.M. Smith, Y.S. Tang, A.M. Makarewicz, F. Ujjainwalla, S.W. Altmann, K.T. Chapman, N.A. Thornberry, The target of ezetimibe is Niemann-Pick C1-Like 1 (NPC1L1). Proc. Natl. Acad. Sci. U. S. A., 102 (2005) 8132-8137.

[196] H.J. Kwon, M. Palnitkar, J. Deisenhofer, The structure of the NPC1L1 N-terminal domain in a closed conformation., PLoS One, 6 (2011) e18722.

[197] Zhang J.H., L. Ge, W. Qi, L. Zhang, H.H. Miao, B.L. Li, M. Yang, B.L. Song, The N-terminal domain of NPC1L1 protein binds cholesterol and plays essential roles in cholesterol uptake., J. Biol. Chem. , 286 (2011) 25088-25097.

[198] L. Yu, S. Bharadwaj, J.M. Brown, Y. Ma, W. Du, M.A. Davis, P. Michaely, P. Liu, M.C. Willingham, L.L. Rudel, Cholesterol-regulated translocation of NPC1L1 to the cell surface facilitates free cholesterol uptake., J. Biol. Chem., 281 (2006) 6616-6624.

[199] N.H. Petersen, N.J. Faegeman, L. Yu, D. Wustner, Kinetic imaging of NPC1L1 and sterol trafficking between plasma membrane and recycling endosomes in hepatoma cells, Journal of Lipid Research, 49 (2008) 2023-2037.

[200] R.E. Temel, W. Tang, Y. Ma, L.L. Rudel, M.C. Willingham, Y.A. Ioannou, J.P. Davies, L.M. Nilsson, L. Yu, Hepatic Niemann-Pick C1-like 1 regulates biliary cholesterol concentration and is a target of ezetimibe., J. Clin. Invest., 117 (2007) 1968-1978.

[201] J. Wei, Z.Y. Fu, P.S. Li, H.H. Miao, B.L. Li, Y.T. Ma, B.L. Song, The Clathrin Adaptor Proteins ARH, Dab2, and Numb Play Distinct Roles in Niemann-Pick C1-Like 1 Versus Low Density Lipoprotein Receptor-mediated Cholesterol Uptake., J. Biol. Chem., 289 (2014) 33689-33700.

[202] C. Xie, N. Li, Z.J. Chen, B.L. Li, B.L. Song, The small GTPase Cdc42 interacts with Niemann-Pick C1-like 1 (NPC1L1) and controls its movement from endocytic recycling compartment to plasma membrane in a cholesterol-dependent manner., J. Biol. Chem., 286 (2011) 35933-35942.

[203] L. Ge, W. Qi, L.J. Wang, H.H. Miao, Y.X. Qu, B.L. Li, B.L. Song, Flotillins play an essential role in Niemann-Pick C1-like 1-mediated cholesterol uptake., Proc. Natl. Acad. Sci. USA, 108 (2011) 551-556.

[204] R. Zidovetzki, I. Levitan, Use of cyclodextrins to manipulate plasma membrane cholesterol content: evidence, misconceptions and control strategies., Biochim. Biophys. Acta, 1768 (2007).

[205] M. Knöpfel, J.P. Davies, P.T. Duong, L. Kv^rn0, E.M. Carreira, M.C. Phillips, Y.A. Ioannou, H. Hauser, Multiple plasma membrane receptors but not NPC1L1 mediate high-affinity, ezetimibe-sensitive cholesterol uptake into the intestinal brush border membrane., Biochim. Biophys. Acta, In press. (2007).

[206] D. Wüstner, J.R. Brewer, L.A. Bagatolli, D. Sage, Potential of ultraviolet widefield imaging and multiphoton microscopy for analysis of dehydroergosterol in cellular membranes. , Microsc. Res. Tech., 74 (2011) 92-108.

[207] J.L. Goldstein, M.S. Brown, Binding and degradation of low density lipoproteins by cultured human fibroblasts. Comparison of cells from a normal subject and from a patient with homozygous familial hypercholesterolemia., J. Biol. Chem., 249 (1974).

[208] M.S. Brown, J.L. Goldstein, Cholesterol feedback: from Schoenheimer's bottle to Scap's MELADL., J. Lipid Res., Suppl. (2009) S15-S27.

[209] T. Teerlink, P.G. Scheffer, LDL particles are nonspherical: consequences for size determination and phenotypic classification., Clin. Chem., 53 (2007) 361-362.

[210] F.R. Maxfield, McGraw, T.E., Endocytic recycling., Nat. Rev. Mol. Cell Biol., 5 (2004) 121-132.

[211] J. Rink, E. Ghigo, Y. Kalaidzidis, M. Zerial, Rab conversion as a mechanism of progression from early to late endosomes., Cell, 122 (2005) 735-749.

[212] J. Huotari, A. Helenius, Endosome maturation., EMBO J., 30 (2011) 3481-3500.

[213] H.D. Gallala, B. Breiden, K. Sandhoff, Regulation of the NPC2 protein-mediated cholesterol trafficking by membrane lipids., J. Neurochem., 116 (2011) 702-707.

[214] M. Record, K. Carayon, M. Poirot, S. Silvenet-Poirot, Exosomes as new vesicular lipid transporters involved in cell-cell communication and various pathophysiologies., Biochim. Biophys. Acta, 1841 (2014) 108-120.

[215] J.P. Luzio, N.A. Bright, P.R. Pryor, The role of calcium and other ions in sorting and delivery in the late endocytic pathway., Biochem. Soc. Trans., 35 (2007) 1088-1091.

[216] J. Gent, I. Braakman, Low-density lipoprotein receptor structure and folding., Cell. Mol. Life Sci., 61 (2004) 2461-2470.

[217] S. Mukherjee, Ghosh, R.N., and Maxfield, F.R., Endocytosis, Physiol. Rev., 77 (1997) 759-803.

[218] D. Fass, S. Blacklow, P.S. Kim, J.M. Berger, Molecular basis of familial hypercholesterolaemia from structure of LDL receptor module., Nature, 388 (1997) 691-693.

[219] Z. Zhao, P. Michaely, The role of calcium in lipoprotein release by the low-density lipoprotein receptor., Biochemistry, 48 (2009) 7313-7324.

[220] M. Hao, F.R. Maxfield, Characterization of rapid membrane internalization and recycling., J. Biol. Chem., 275 (2000) 15279-15286.

[221] K.L. Bowden, N.J. Bilbey, L.M. Bilawchuk, E. Boadu, R. Sidhu, D.S. Ory, H. Du, T. Chan, G.A. Francis, Lysosomal acid lipase deficiency impairs regulation of ABCA1 gene and formation of high density lipoproteins in cholesteryl ester storage disease., J. Biol. Chem., 286 (2011) 30624-30635.

[222] S. Mukherjee, F.R. Maxfield, Lipid and cholesterol trafficking in NPC., Biochim. Biophys. Acta, 1685 (2004) 28-37.

[223] E. Lloyd-Evans, F.M. Platt, Lipids on trial: the search for the offending metabolite in Niemann Pick type C disease., Traffic, 11 (2010) 419-428.

[224] D. Wüstner, Intracellular cholesterol transport., In: Cellular lipid metabolism., Ed. Ehnholm, C., Springer press. (2009) p. 157-190.

[225] R.E. Infante, A. Radhakrishnan, L. Abi-Mosleh, L.N. Kinch, M.L. Wang, N.V. Grishin, J.L. Goldstein, M.S. Brown, Purified NPC1 protein: II. Localization of sterol binding to a 240-amino acid soluble luminal loop., J. Biol. Chem., 283 (2008) 1064-1075.

[226] A. Frolov, S.E. Zielinski, J.R. Crowley, N. Dudley-Rucker, J.E. Schaffer, D.S. Ory, NPC1 and NPC2 regulate cellular cholesterol homeostasis through generation of low density lipoprotein cholesterol-derived oxysterols., J. Biol. Chem., 278 (2003) 25517-25525.

[227] R.E. Infante, M.L. Wang, A. Radhakrishnan, H.J. Kwon, M.S. Brown, J.L. Goldstein, NPC2 facilitates bidirectional transfer of cholesterol between NPC1 and lipid bilayers, a step in cholesterol egress from lysosomes., Proc. Natl. Acad. Sci. U. S. A., 105 (2008) 15287-15292.

[228] M.L. Wang, M. Motamed, R.E. Infante, L. Abi-Mosleh, H.J. Kwon, M.S. Brown, J.L. Goldstein, Identification of surface residues on Niemann-Pick C2 essential for hydrophobic handoff of cholesterol to NPC1 in lysosomes., Cell Metabolism, 12 (2010) 166-173.

[229] G. Estiu, N. Khatri, O. Wiest, Computational studies of the cholesterol transport between NPC2 and the N-terminal domain of NPC1 (NPC1(NTD)). Biochemistry, 52 (2013) 6879-6891.

[230] N. Elghobashi-Meinhardt, Niemann-Pick type C disease: a QM/MM study of confrmational changes in cholesterol in the NPC1(NTD) and NPC2 binding pockets., Biochemistry, 53 (2014) 6603-6614.

[231] M.S. Deffieu, S.R. Pfeffer, Niemann-Pick type C 1 function requires lumenal domain residues that mediate cholesterol-dependent NPC2 binding., Proc. Natl. Acad. Sci. USA, 108 (2011) 18932-18936.

[232] T.S. Blom, M.D. Linder, K. Snow, H. Pihko, M.W. Hess, E. Jokitalo, V. Veckman, A.-C. Syvänen, E. Ikonen, Defective endocytic trafficking of NPC1 and NPC2 underlying infantile Niemann-Pick type C disease., Hum. Mol. Genet., 12 (2003) 257-272.

[233] D.C. Ko, M.D. Gordon, J.Y. Jin, M.P. Scott, Dynamic movements of organelles containing Niemann-Pick C1 protein: NPC1 involvement in late endocytic events., Mol. Biol. Cell, 12 (2001) 601-614.

[234] M. Zhang, N.K. Dwyer, D.C. Love, A. Cooney, M. Comly, E. Neufeld, P.G. Pentchev, E.J. Blanchette-Mackie, J.A. Hanover, Cessation of rapid late endosomal tubulovesicular trafficking in Niemann-Pick type C1 disease., Proc. Natl. Acad. Sci. U. S. A., 98 (2001) 4466-4471.

[235] B.E. Kennedy, M. Charman, B. Karten, Niemann-Pick Type C2 protein contributes to the transport of endosomal cholesterol to mitochondria without interacting with NPC1., J. Lipid Res., 53 (2012) 2632-2642.

[236] E. Boadu, R.C. Nelson, G.A. Francis, ABCA1-dependent mobilization of lysosomal cholesterol requires functional Niemann-Pick C2 but not Niemann-Pick C1 protein., Biochim Biophys Acta, 1821 (2012) 396-404.

[237] S.D. Goldman, J.P. Krise, Niemann-Pick C1 functions independently of Niemann-Pick C2 in the initial stage of retrograde transport of membrane-impermeable lysosomal cargo., J. Biol. Chem., 285 (2010) 49834994.

[238] R. van der Kant, I. Zondervan, L. Janssen, J. Neefjes, Cholesterol-binding molecules MLN64 and ORPL1 mark distinct late endosomes with transporters ABCA3 and NPC1., J. Lipid Res., 54 (2013) 2153-2164.

[239] N. Rocha, C. Kuijl, R. van der Kant, L. Janssen, D. Houben, H. Janssen, W. Zwart, J. Neefjes, Cholesterol sensor ORP1L contacts the ER protein VAP to control Rab7-RILP-p150 Glued and late endosome positioning., J. Cell Biol., 185 (2009) 1209-1225.

[240] X. Du, J. Kumar, C. Ferguson, T.A. Schulz, Y.S. Ong, W. Hong, W.A. Prinz, R.G. Parton, A.J. Brown, H. Yang, A role for oxysterol-binding protein-related protein 5 in endosomal cholesterol trafficking., J. Cell Biol., 192 (2011) 121-135.

[241] N.L. Cianciola, C.R. Carlin, Adenovirus RID-alpha activates an autonomous cholesterol regulatory mechanism that rescues defects linked to Niemann-Pick disease type C., J. Cell Biol., 187 (2009) 537-552.

[242] N.L. Cianciola, D.J. Greene, R.E. Morton, C.R. Carlin, Adenovirus RIDa uncovers a novel pathway requiring ORP1L for lipid droplet formation independent of NPC1., Mol. Biol. Cell, 24 (2013) 3309-3325.

[243] K. Kanerva, R.L. Uronen, T. Blom, S. Li, R. Bittman, P. Lappalainen, J. Peränen, G. Raposo, E. Ikonen, LDL cholesterol recycles to the plasma membrane via a Rab8a-Myosin5b-actin-dependent membrane transport route., Dev. Cell, 27 (2013) 249-262.

[244] M.D. Linder, R.L. Uronen, M. Holtta-Vuori, P. van der Sluijs, J. Peranen, E. Ikonen, Rab8-dependent Recycling Promotes Endosomal Cholesterol Removal in Normal and Sphingolipidosis Cells., Mol. Biol. Cell, 18. (2007) 47-56.

[245] M. Hölttä-Vuori, F. Alpy, K. Tanhuanpaa, E. Jokitalo, A.L. Mutka, E. Ikonen, MLN64 is involved in actin-mediated dynamics of late endocytic organelles., Mol. Biol. Cell, 16 (2005) 3873-3886.

[246] X. Du, A.S. Kazim, A.J. Brown, H. Yang, An essential role of Hrs/Vps27 in endosomal cholesterol trafficking., Cell Reports, 1 (2012) 29-35.

[247] C. Bucci, P. Thomsen, P. Nicoziani, J. McCarthy, B. van Deurs, Rab7: A key to lysosome biogenesis., Mol. Biol. Cell, 11 (2000) 467-480.

[248] Y. Urano, H. Watanabe, S.R. Murphy, Y. Shibuya, Y. Geng, A.A. Peden, C.C. Chang, T.Y. Chang, Transport of LDL-derived cholesterol from the NPC1 compartment to the ER involves the trans-Golgi network and the SNARE protein complex., Proc. Natl. Acad. Sci. U. S. A., 105 (2008) 16513-16518.

[249] J. Neefjes, R. van der Kant, Stuck in traffic: an emerging theme in diseases of the nervous system., Trends Neurosci., 37 (2014) 66-76.

[250] X. Du, H. Yang, Endosomal cholesterol trafficking: protein factors at a glance., Acta Biochim Biophys Sin (Shanghai). , 45 (2013) 11-17.

[251] D. Shen, X. Wang, X. Li, X. Zhang, Z. Yao, S. Dibble, X. Dong, T. Yu, Liebermann A.P., H.D. Showalter, H. Xu, Lipid storage disorders block lysosomal trafficking by inhibiting a TRP channel and lysosomal calcium release., Nat. Commun., 3 (2012) 1-11.

[252] E. Lloyd-Evans, A.J. Morgan, X. He, D.A. Smith, E. Elliot-Smith, D.J. Sillence, G.C. Churchill, E.H. Schuchman, A. Galione, F.M. Platt, Niemann-Pick disease type C1 is a sphingosine storage disease that causes deregulation of lysosomal calcium., Nat. Med., 14 (2008) 1247-1255.

[253] M.A. Samie, H. Xu, Lysosomal exocytosis and lipid storage disorders., J. Lipid Res., 55 (2014) 995-1009.

[254] C. Devlin, N.H. Pipalia, X. Liao, E.H. Schuchman, F.R. Maxfield, I. Tabas, Improvement in lipid and protein trafficking in Niemann-Pick C1 cells by correction of a secondary enzyme defect., Traffic, 11 (2012) 601-615.

[255] V.O. Oninla, B. Breiden, J.O. Babalola, K. Sandhoff, Acid sphingomyelinase activity is regulated by membrane lipids and facilitates cholesterol transfer by NPC2., J. Lipid Res., 55 (2014) 2606-2619.

[256] N. Wang, Silver, D.L., Thiele, C., and Tall, A.R., ATP-binding cassette transporter A1 (ABCA1) functions as a cholesterol efflux regulatory protein., J. Biol. Chem., 276 (2001) 23742-23747.

[257] E.B. Neufeld, Stonik, J.A., Demosky, S.J. Jr., Knapper, C.L., Combs, C.A., Cooney, A., Comly, M., Dwyer, N., Blanchette-Mackie, J., Remaley, A.T., Santamarina-Fojo, S., and Brewer, H.B. Jr., The ABCA1 transporter modulates late endocytic trafficking: insights from the correction of the genetic defect in Tangier disease., J. Biol. Chem., 279 (2004) 15571-15578.

[258] C. Cavelier, I. Lorenzi, L. Rohrer, A. von Eckardstein, Lipid efflux by the ATP-binding cassette transporters ABCA1 and ABCG1., Biochim. Biophys. Acta, 1761 (2006) 655-666.

[259] M.A. Kennedy, G.C. Barrera, K. Nakamura, A. Baldan, P. Tarr, M.C. Fishbein, J. Frank, O.L. Francone, P.A. Edwards, ABCG1 has a critical role in mediating cholesterol efflux to HDL and preventing cellular lipid accumulation., Cell Metab., 1 (2005) 121-131.

[260] E.J. Tarling, P.A. Edwards, ATP binding cassette transporter G1 (ABCG1) is an intracellular sterol transporter., Proc. Natl. Acad. Sci. USA, 108 (2011) 19719-19724.

[261] C.P. Sparrow, J. Baffic, M.H. Lam, E.G. Lund, A.D. Adams, X. Fu, N. Hayes, A.B. Jones, K.L. Macnaul, J. Ondeyka, S. Singh, J. Wang, G. Zhou, D.E. Moller, S.D. Wright, J.G. Menke, A potent synthetic LXR agonist is more effective than cholesterol loading at inducing ABCA1 mRNA and stimulating cholesterol efflux., J. Biol. Chem., 277 (2002) 10021-10027.

[262] M. Denis, Y.D. Landry, X. Zha, ATP-binding cassette A1-mediated lipidation of apolipoprotein A-I occurs at the plasma membrane and not in the endocytic compartments., J. Biol. Chem., 283 (2008) 1617816186.

[263] E.B. Neufeld, Demosky, S.J. Jr., Stonik, J.A., Combs, C., Remaley, A.T., Duverger, N., Santamarina-Fojo, S., and Brewer, H.B. Jr., The ABCA1 transporter functions on the basolateral surface of hepatocytes., Biochem. Biophys. Res. Commun., 297 (2002) 974-979.

[264] X. Zha, A. Gauthier, J.J. Genest, R. McPherson, Secretory vesicular transport from the Golgi is altered during ATP-binding cassette protein A1 (ABCA1)-mediated cholesterol efflux., J. Biol. Chem., 278 (2003) 10002-10005.

[265] E. Boadu, H.Y. Choi, D.W. Lee, E.I. Waddington, T. Chan, B. Asztalos, J.E. Vance, A. Chan, G. Castro, G.A. Francis, Correction of apolipoprotein A-I-mediated lipid efflux and high density lipoprotein particle formation in human Niemann-Pick type C disease fibroblasts., J. Biol. Chem., 281 (2006) 37081-37090.

[266] H.Y. Choi, B. Karten, T. Chan, J.E. Vance, W.L. Greer, R.A. Heidenreich, W.S. Garver, G.A. Francis, Impaired ABCA1-dependent lipid efflux and hypoalphalipoproteinemia in human Niemann-Pick type C disease., J. Biol. Chem., 278 (2003) 32569-32577.

[267] 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.

[268] N. Alder-Baerens, P. Müller, A. Pohl, T. Korte, Y. Hamon, G. Chimini, T. Pomorski, A. Herrmann, Headgroup-specific exposure of phospholipids in ABCA1-expressing cells., J. Biol. Chem., 280 (2005) 17751785.

[269] A. Zarubica, A.P. Plazzo, M. Stöckl, T. Trombik, Y. Hamon, P. Müller, T. Pomorski, A. Herrmann, G. Chimini, Functional implications of the influence of ABCA1 on lipid microenvironment at the plasma membrane: a biophysical study., FASEB J., 23 (2009) 1775-1785.

[270] X. Zha, J.J. Genest, R. McPherson, Endocytosis is enhanced in Tangier fibroblasts: possible role of ATP-binding cassette protein A1 in endosomal vesicular transport., J. Biol. Chem., 276 (2001) 39476-39483.

[271] T.A. Pagler, M. Wang, M. Mondal, A.J. Murphy, M. Westerterp, K.J. Moore, F.R. Maxfield, A.R. Tall, Deletion of ABCA1 and ABCG1 impairs macrophage migration because of increased Rac1 signaling., Circ. Res., 108 (2011) 194-200.

[272] P. Müller, A. Herrmann, Rapid transbilayer movement of spin-labeled steroids in human erythrocytes and in liposomes., Biophys. J., 82 (2002) 1418-1428.

[273] T.L. Steck, J. Ye, Y. Lange, Probing red cell membrane cholesterol movement with cyclodextrin., Biophys. J., 83 (2002) 2118-2125.

[274] K. Gulshan, G. Brubaker, S. Wang, S.L. Hazen, J.D. Smith, Sphingomyelin depletion impairs anionic phospholipid inward translocation and induces cholesterol efflux., J. Biol. Chem., 288 (2013) 37166-37179.

[275] 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.

[276] Y. Fu, N. Mukhamedova, S. Ip, W. D'Souza, K.J. Henley, T. DiTommaso, R. Kesani, M. Ditiatkovski, L. Jones, R.M. Lane, G. Jennings, I.M. Smyth, B.T. Kile, D. Sviridov, ABCA12 regulates ABCA1-dependent cholesterol efflux from macrophages and the development of atherosclerosis., Cell Metab., 18 (2013) 225238.

[277] E. Levy, S. Spahis, D. Sinnett, N. Peretti, F. Maupas-Schwalm, E. Delvin, M. Lambert, M.A. Lavoie, Intestinal cholesterol transport proteins: an update and beyond., Curr. Opin. Lipidol. , 18 (2007) 310-318.

[278] D.M. Small, Role of ABC transporters in secretion of cholesterol from liver into bile., Proc. Natl. Acad. Sci. U. S. A., 100 (2003) 4-6.

Figure legends

Figure 1. Possible mechanisms underlying non-vesicular sterol transport by transfer proteins.

A, possible outcomes of desorption of a cholesterol molecule from a donor membrane: 1) the molecule performs a random walk from the donor membrane (left) to the acceptor membrane (right), where it gets inserted giving a successful excursion, indicated by a smiley. This process is likely very inefficient and requires transfer proteins to take place at significant rates. 2) the cholesterol molecule desorbs and diffuses up to approx. its mean excursion distance, dm, before it gets reabsorbed by the donor membrane. The definition of dm is given in Eq. 8 of the main text. 3) membrane structures located in between the donor and acceptor membrane might bind the sterol transiently allowing for its lateral diffusion in that membrane and desorption closer to the acceptor membrane. Together, this allows for extending dm so much as to enable the sterol molecule to travel to the acceptor membrane, where it gets inserted (smiley). 4) An intermediate membrane, as there are plenty in the cytoplasm, might extend dm somehow by allowing for transient absorption of the sterol , but this is not sufficient to prevent its reabsorption by the donor membrane. B, Sterol transport rate as function of distance between the membranes is given in units of dm. If d < dm the sterol molecule can move to the acceptor membrane approaching the maximal flux, Jmax, for very short distances. In this regime, sterol flux is limited by partitioning of sterol between the membranes. For inter-membrane distances larger than dm the diffusion constant of cholesterol in the medium limits its flux. C, D, soluble transfer proteins (i.e., carriers which do not interact with membranes; case (1) - upper part in panel D) will increase dm several fold without changing Jmax (red dotted line in C). In case (2), membrane-binding transfer proteins will increase both, dm and Jmax (cyan dotted line in C and middle part of panel D). In case (3), transfer proteins tether two membranes together, thereby increasing Jmax and pulling the membranes to a distance d ~dm , such that diffusion will never limit the net sterol flux between membranes (orange dotted line in C and lower part of panel D). Adapted from [188].

Figure 2. Schematic diagram for endosomal cholesterol transport. (1) LDL particles bind to the LDL receptor (LDL-R) at the cell surface and become internalized by clathrin-dependent endocytosis followed by (2) fusion with sorting endosomes (SEs). (3) in the SEs, LDL dissociates from its receptor and some hydrolysis by acid lipase starts. (4) SEs mature into late endosomes (LEs). (5) the majority of LDL degradation including hydrolysis of cholesteryl esters to cholesterol

and fatty acid occurs in LEs. LDL-liberated cholesterol can exit the LEs via several pathways; (6a) to the Golgi/TGN in transport vesicles which fuse with the acceptor compartment via TGN-specific SNAREs (e.g. syntaxin 16); (6b) to the endoplasmic reticulum (ER) in a pathway involving NPC2, NPC1 and ORP1L1; (6c) to mitochondria in steroidogenic cells in a pathway involving NPC2 and MLN64 but not NPC1 (a similar pathway might also operate between LEs and ER, not shown); (6d) to the plasma membrane (PM) via a poorly defined pathway. (7) some sterol follows also the normal endocytic recycling route from SEs via the endocytic recycling compartment (ERC) to the PM. (8) from the PM, some cholesterol can be internalized by a non-vesicular mechanism involving StARD4. In this process, cholesterol gets targeted to the ERC and ER, respectively (dashed line). Excess cholesterol arriving in the ER becomes esterified by acyl-CoA acyl transferase (ACAT) and stored in lipid droplets (LDs). See text for further explanations.

Figure 3. Complex between the N-terminal loop of NPC1 (NPC1(NTD)) and NPC2 in endosomes and its suggested conformational changes allowing for cholesterol transfer. Based on the crystal structure of NPC1(NTD) (PDB: 3GKI) and NPC2 (PDB: 2HKA), molecular simulations have been carried out to uncover the energetic and conformational changes associated with cholesterol transfer from NPC2 to NPC1 [230]. A putative complex forms between NPC2 and NPC1, located in the late-endosomal lumen and limiting membrane, respectively [228]. In NPC2, the sterol (blue stick representation) tail is buried in the binding pocket, and the 3P-hydroxyl group (red ball) extends toward the NPC1(NTD) binding site. In NPC1(NTD), the orientation of the ligand (green stick representation) is reversed, with the ligand 3P-hydroxyl group (red ball) pointing toward the interior of the protein, while the isooctyl sterol tail points toward the sterol opening. The transfer of the cholesterol ligand from one binding pocket to the other requires a conformational change in the ligand-protein complex. One possible mechanism includes the conformational rearrangement of the cholesterol ligand inside the NPC2 binding pocket (blue arrow), for example, isomerization of the cholesterol side chain dihedral torsion angle C17-C20-C22-C23, as shown in inset (1). Via an intermediate structure (yellow), the sterol ligand could slide to the binding pocket in NPC1(NTD) (yellow to green arrow). Inset (2) shows, how the NPC2 binding pocket (in blue) and the NPC1(NTD) binding pocket (in green) are bent with respect to each other in this scenario during cholesterol transfer [229]. Inset (3) shows an alternative scenario in which the protein moieties are rotated with respect to each other such that the binding pockets are aligned, allowing

for transfer of cholesterol with negligible change in its conformation. Reproduced and adapted from [230] with permission.

Wustnerand Solanko, Fig. 1

Wustner and Solanko, Fig. 2

Wustnerand Solanko, Fig. 3

The authors, Dr. Daniel Wüstner and Dr. Katarzyna Solanko, declare to have no conflict of interest.

Graphical abstract

Highlights

• Cholesterol has a very heterogeneous distribution in living cells.

• Cholesterol moves by vesicular and non-vesicular mechanisms.

• Specific cholesterol-lipid interactions control intracellular sterol gradients.

• Lipid transfer proteins facilitate non-vesicular cholesterol transport.

• Lipoprotein-mediated cholesterol transport is regulated by a network of proteins.