Formation and Metabolism of Prebeta-Migrating, Lipid-Poor Apolipoprotein A-I

Formation and Metabolism of Prebeta-Migrating, Lipid-Poor Apolipoprotein A-I Kerry-Anne Rye, Philip J. Barter Abstract The preferred extracellular acceptor of cell phospholipids and unesterified cholesterol in the process mediated by the ATP-binding cassette A1 (ABCA1) transporter is a monomolecular, prebeta-migrating, lipid-poor or lipid-free form of apolipoprotein (apo) A-I. This monomolecular form of apoa-i is quite distinct from the prebeta-migrating, discoidal high-density lipoprotein (HDL) that contains two or three molecules of apoa-i per particle and which are present as minor components of the HDL fraction in human plasma. The mechanism of the ABCA1-mediated efflux of phospholipid and cholesterol from cells has been studied extensively. In contrast, much less attention has been given to the origin and subsequent metabolism of the acceptor lipid-free/lipid-poor apoa-i. There is a substantial body of evidence from studies conducted in vitro that a monomolecular, lipid-free/lipid-poor form of apoa-i dissociates from HDL during the remodeling of HDLs by plasma factors such as cholesteryl ester transfer protein, hepatic lipase, and phospholipid transfer protein. The rate at which apoa-i dissociates from HDL is influenced by the phospholipid composition of the particles and by the presence of apoa-ii. This review describes current knowledge regarding the formation, metabolism, and regulation of monomolecular, lipid-free/lipid-poor apoa-i in plasma. (Arterioscler Thromb Vasc Biol. 2004;24:421-428.) Key Words: apoa-i prebeta HDL CETP HDL remodeling PLTP High-density lipoproteins (HDL) are anti-atherogenic. 1 8 The precise mechanism of this effect is not known but presumably relates to one or more of the known functions of these lipoproteins. The most widely known of these functions is the involvement of HDL in the pathway of reverse cholesterol transport. 9 In this pathway, cholesterol in peripheral tissues is transferred via the plasma compartment to the liver, where it is either recycled back into plasma as a component of newly formed lipoproteins or it is excreted from the body via bile. The first step in the reverse cholesterol transport pathway is the efflux of phospholipids and unesterified cholesterol from cell membranes to a monomolecular, lipid-free or lipid-poor form of apolipoprotein A-I (apoa-i) in a process promoted by the ATP-binding cassette A1 (ABCA1). 10 12 Efflux of cholesterol from cells may also be mediated by an interaction of extracellular acceptors with the scavenger receptor-b1 (SR-B1). 10 A massive research effort has provided much information about the regulation and the mechanism of action of ABCA1 and SR-B1 and has identified these proteins as potential therapeutic targets. In contrast, investigation of the pool of lipid-free/lipid-poor apoa-i that accepts the cholesterol released by ABCA1 has been much more limited. The apoa-i in plasma is present in three potential forms (Figure 1). Most ( 90%) apoa-i circulates as a component of mature, spherical, -migrating HDL particles. 13 ApoA-I may also exist as a component of discoidal HDL, although the rapidity with which HDL discs are converted by lecithin:cholesterol acyltransferase (LCAT) into spherical HDL 14,15 ensures that such particles are normally present in plasma at only very low concentrations. Thirdly, apoa-i may exist in plasma in a lipid-free or lipid-poor form. The fact that discoidal HDL and lipid-free apoa-i both have a prebeta electrophoretic mobility has created much confusion about what is meant by the term prebeta HDL. As outlined below, many of the reported studies on prebetamigrating apoa-i have used techniques that do not have the capacity to differentiate between discoidal HDL (that contain at least two molecules of apoa-i in a complex that also contains phospholipids and possibly unesterified cholesterol) and lipid-free or lipid-poor apoa-i that exists in a monomolecular form. Given that it is the monomolecular, lipid-free/lipid-poor form of apoa-i that interacts with ABCA1, the distinction between these two forms of apoa-i is more than academic. This review is concerned primarily with the monomolecular form of lipid-free/lipid-poor apoa-i and outlines what is known about its formation and its subsequent metabolic fate. However, to see where such a pool of lipid-free/poor apoa-i fits into overall HDL transport, the structure, function, and metabolism of the total HDL fraction will first be summarized. Received September 21, 2003; revision accepted October 20, 2003. From the Heart Research Institute, Sydney, Australia. Correspondence to Prof Philip Barter, The Heart Research Institute, 145 Missenden Rd, Camperdown, Sydney, NSW 2050, Australia. E-mail p.barter@hri.org.au 2004 American Heart Association, Inc. Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org DOI: 10.1161/01.ATV.0000104029.74961.f5 421

422 Arterioscler Thromb Vasc Biol. March 2004 Figure 1. ApoA-I exists in three forms in plasma. Structure of HDL The HDL fraction in human plasma is heterogeneous in terms of size, shape, composition, and surface charge. 13 When viewed in the electron microscope, HDL can appear as either spherical or discoidal particles. In normal plasma, most are spherical. Spherical HDL have a hydrophobic core (mainly cholesteryl esters plus a small amount of triglyceride) surrounded by a surface monolayer that consists of phospholipids, unesterified cholesterol, and apolipoproteins. Discoidal HDL consist of a bilayer of phospholipids and a small amount of unesterified cholesterol. The hydrophobic phospholipid acyl chains in the bilayer are surrounded by an annulus of apolipoproteins that renders the particles water-soluble. Most discoidal HDL contain two or three molecules of apoa-i and do not contain a hydrophobic core. When isolated on the basis of density by ultracentrifugation, human HDL separate into two major subfractions, which have been designated HDL 2 (1.063 d 1.125 g/ml) and HDL 3 (1.125 d 1.21 g/ml). Nondenaturing polyacrylamide gradient gel electrophoresis has been used to separate HDL on the basis of particle size into five distinct subpopulations 16 with diameters ranging from 7.6 nm to 10.6 nm. The HDL in human plasma contain two main apolipoproteins, apoa-i and apoa-ii, that account for 70% and 20%, respectively, of the total HDL protein. Alphamigrating, spherical HDL have been classified into two main subpopulations on the basis of their apolipoprotein composition. The first subpopulation contains apoa-i but not apoa-ii (A-I HDL), whereas the other contains both apoa-i and apoa-ii (A-I/A-II HDL). 17,18 In most human subjects, apoa-i is divided approximately equally between A-I HDL and A-I/ A-II HDL, whereas virtually all of the apoa-ii is in A-I/A-II HDL. 17,18 A-I/A-II HDL are mostly found in the HDL 3 density range, whereas A-I HDL are prominent components of both HDL 2 and HDL 3. 17,18 The heterogeneity of HDL also extends to their surface charge. When HDL are separated on the basis of surface charge by agarose gel electrophoresis, they exhibit either alpha, prealpha, prebeta, or gamma migration. Alphamigrating HDL are spherical particles that account for the major proportion of HDL in plasma. They include the HDL 2 and HDL 3 subfractions as well as most of the A-I HDL and A-I/A-II HDL subpopulations. Prebeta HDL represent either discoidal particles consisting of two or three molecules of apoa-i complexed with phospholipids and unesterified cholesterol or monomolecular, lipid-free/poor apoa-i 19. Discoidal HDL may also have a prealpha mobility. 20 22 A minor subpopulation of gamma-hdl consisting of discoidal particles containing apoe and phospholipids has also been reported. 23 Functions of HDL The best-known function of HDL relates to their role in reverse cholesterol transport. 9 This term is used to describe the pathway in which cholesterol in peripheral tissues is transferred, via the plasma, to the liver for either recycling or excretion from the body in bile. This pathway involves several identifiable steps. (1) Phospholipids and unesterified cholesterol efflux from cells to a monomolecular form of lipid-free/poor apoa-i in a process mediated by ABCA1. The resulting lipidation of the apoa-i results in the formation of prebeta-migrating discoidal complexes containing at least two molecules of apoa-i plus phospholipids and unesterified cholesterol. (2) A portion of the unesterified cholesterol in the newly formed discoidal HDL is delivered to the liver before it becomes esterified. (3) The remaining unesterified cholesterol is esterified by LCAT in a process that converts the prebeta-migrating discs into alpha-migrating spheres. (4) A portion of the newly formed cholesteryl esters is transferred by cholesteryl ester transfer protein (CETP) from HDL to very low-density lipoproteins (VLDL) and low-density lipoproteins (LDL). These cholesteryl esters are subsequently taken up by the liver as components of LDL and VLDL. (5) The remaining HDL cholesteryl esters are delivered directly to the liver. There is circumstantial evidence that the involvement of HDL in reverse cholesterol transport is one of the mechanisms by which HDL protect against atherosclerosis. It should be noted, however, that HDL have additional antiinflammatory, antioxidant and antithrombotic properties, apparently unrelated to their role in plasma lipid transport, that may also contribute to their antiatherogenic properties. 24 Differentiating Between Prebeta-Migrating Discoidal HDL and Prebeta-Migrating, Monomolecular, Lipid-Poor ApaA-I Both monomolecular, lipid-free/lipid-poor apoa-i and discoidal apoa-i-containing HDL have a prebeta mobility when subjected to agarose gel electrophoresis. 10,25,26 Thus, the discovery that plasma contains prebeta-migrating apoa-i could reflect the presence of either monomolecular, lipidpoor apoa-i or discoidal HDL or both. Although there is compelling evidence that prebeta-migrating, discoidal HDL are present in plasma in vivo, it has not yet been proven that apoa-i circulates in a monomolecular form in vivo.

Rye and Barter Metabolism of Lipid-Poor ApoA-I 423 The reported concentration of prebeta-migrating apoa-i varies widely, in part reflecting biological variation but also because of differences in the techniques that are used to quantify them. Most of the studies have employed twodimensional gel electrophoresis (agarose gel electrophoresis followed by nondenaturing polyacrylamide gradient gel electrophoresis) to separate prebeta-migrating HDL from alphamigrating particles and to subfractionate the prebeta HDL into smaller (designated prebeta-1 HDL) and larger (designated prebeta-2 and prebeta-3 HDL) particles. Prebeta-1 HDL has a reported apparent molecular mass in the range of 60 to 71 kda, 19,27 29 whereas prebeta-2 and prebeta-3 HDL have molecular masses in excess of 300 kda. 29,30 Given that the molecular weight of apoa-i is 28 300 kda, it is apparent that an HDL particle with a molecular mass exceeding 60 kd must contain at least two molecules of apoa-i. The unequivocal presence in plasma of prebeta-migrating apoa-i in a monomolecular form has not been demonstrated by nondenaturing gradient gel electrophoresis. But the fact that any monomolecular apoa-i may have run off the end of the gel also means that the existence of a monomolecular form of apoa-i in plasma has not been excluded. Indeed, one study using crossed immunoelectrophoresis has identified a pool of plasma apoa-i that has the same size and electrophoretic mobility as purified free A-1. 31 The classification of prebeta-migrating apoa-i has been complicated by the fact that the electrophoretic approach used to separate these particles has not been standardized. To determine the molecular mass and the size of prebetamigrating HDL, it is necessary for the nondenaturing gradient gel electrophoresis to be run to equilibrium. Under these circumstances, the particles migrate into the gel until they reach a limiting pore size and cannot migrate any further. It is only when equilibrium is reached (after 3000 Vh of electrophoresis on a gel with a gradient of 3% to 30% to 35% acrylamide) that particle size can be determined with any degree of confidence. Some of the reported studies have used electrophoresis in gels with acrylamide gradients ranging from 2% to 16% to 20%. 32,33 Under these circumstances, the smallest HDL subpopulations, as well as any monomolecular apoa-i, would migrate off the end of the gels if the electrophoresis were to be run to equilibrium. The fact that in many cases the electrophoresis is not run to equilibrium 32,33 means that molecular masses and particle size cannot be determined. In fact, in many studies, the migration of particles designated as prebeta-1 HDL is similar to that of the bulk, alphamigrating HDL. Under these conditions, it is not possible to draw any conclusions about whether apoa-i with a prebeta mobility is present as a component of discoidal HDL or whether it is present in a monomolecular, lipid-free/lipid poor form. In those studies in which samples have been electrophoresed to equilibrium on 3% to 30% nondenaturing polyacrylamide gradient gels, it is apparent that the particles designated prebeta-1 HDL have a molecular mass and size considerably smaller than the smallest of the alpha-migrating HDL. 30,34 In these studies, size and the molecular mass of such particles has been reported to be in excess of 60 kd. In none of the studies using gradient gel electrophoresis has there been evidence on the gradient gels of a pool of monomolecular apoa-i. The presence of lipid-free apoa-i in plasma has been reported, 31,35,36 although only in the study using crossed immunoelectrophoresis 31 was the apoa-i shown to be of the same size and electrophoretic mobility as purified lipid-free A-1. Thus, at the current time, it remains unknown how much monomolecular apoa-i exists in human plasma. It is possible that monomolecular, lipid-free/lipid-poor apoa-i exists only transiently in vivo, and is relipidated or excreted through the kidneys at a rate that limits its accumulation to below detection limits in plasma. It is also possible that monomolecular apoa-i is not generated in vivo at all. However, given the clear evidence in vitro that a monomolecular, lipid-free/ lipid-poor form of apoa-i is generated during the remodeling of HDL, and given that such apoa-i may play a fundamental role in the efflux of cholesterol from cells, it is clearly important to review what is known about its generation and subsequent metabolism. Metabolism of ApoA-1 in Plasma Studies conducted in vitro have shown that apoa-i in plasma cycles between lipid-poor and lipid-rich forms. 37 In vitro, this cycling has been shown to involve several identifiable steps: (1) generation of monomolecular, prebeta-migrating, lipidpoor apoa-i from alpha-migrating, spherical HDL; (2) acquisition by the lipid-poor apoa-i of phospholipids and unesterified cholesterol from cell membranes 38 and plasma lipoproteins to form prebeta-migrating, discoidal A-I HDL; (3) the conversion of prebeta-migrating, discoidal A-I HDL into small alpha-migrating, spherical A-I HDL; (4) the conversion of small alpha-migrating, spherical A-I HDL into larger alpha-migrating, spherical A-I HDL; and (5) the remodeling of the large alpha-migrating, spherical A-I HDL by plasma factors in processes that result in the dissociation of monomolecular, prebeta-migrating, lipid-poor apoa-i from the particles, thus completing the cycle. Generation of Monomolecular, Lipid-Poor ApoA-1 in Plasma For the remainder of this review, the term lipid-poor apoa-i will be used to describe monomolecular apoa-i that contains either no or only a very small amount of phospholipid. It has a prebeta electrophoretic mobility and a Stokes diameter and molecular weight ( 28 Kd comparable to that of purified lipid-free apoa-i) when subjected to nondenaturing polyacrylamide gel electrophoresis. It is clearly distinct from the prebeta-migrating, discoidal HDL that contain two or more molecules of apoa-i per particle. There are three potential sources of lipid-poor apoa-i in plasma: (1) it may be released as lipid-poor protein after its synthesis in the liver and intestine; 39,40 (2) it may be released from triglyceride-rich lipoproteins that are undergoing hydrolysis by lipoprotein lipase; 41 and (3) it may be generated within the plasma during the remodeling of mature, spherical HDL particles. 13 The remodeling of HDL has been shown in vitro to generate substantial amounts of lipid-poor apoa-i. It remains to be established, however, whether HDL remodeling also generates lipid-poor apoa-i in vivo.

424 Arterioscler Thromb Vasc Biol. March 2004 Generation of Lipid-Poor apoa-i Through Hydrolysis of Triglyceride-Rich Lipoproteins Chylomicrons have been reported to be the source of significant amounts of the apoa-i in plasma. 42,43 Furthermore, the apoa-i in chylomicrons has been shown to be a precursor of the pool of apoa-i in HDL. 44,45 It is not known, however, whether apoa-i dissociates from chylomicrons in a lipid-poor form or as a complex of apoa-i with phospholipids and other lipids. Dissociation of Lipid-Poor apoa-i From Mature HDL There is now abundant evidence from studies conducted in vitro to show that HDL are subject to extensive remodeling by a range of plasma factors. 13,46,47 Processes that reduce the size of HDL, and those that promote HDL particle fusion, also have the capacity to promote the dissociation of lipidpoor apoa-i. It should, however, be noted that lipid-poor apoa-i does not dissociate from HDL unless the remodeling is accompanied by a reduction in the core lipid content of the particles. 48 Generation of Lipid-Poor apoa-i by CETP Cholesteryl ester transfer protein (CETP) promotes the transfer of cholesteryl esters from HDL to other lipoproteins and triglyceride from triglyceride-rich lipoproteins to HDL. These processes deplete the HDL core of cholesteryl esters and enrich them with triglycerides. When HDL are incubated in vitro with triglyceride-rich lipoproteins in the presence of CETP, the magnitude of the transfer of cholesteryl esters out of HDL may be greater than that of the transfer of triglyceride into the HDL. Under these circumstances there is a net reduction in HDL core lipid content and a reduction particle size. 49 This results in an excess of constituents on the HDL surface that is alleviated by the dissociation of prebetamigrating, lipid-poor apoa-i from the particles. Although there is every reason to suspect that comparable processes also operate in vivo, this has not yet been shown, possibly because the dissociated apoa-i is lipidated as rapidly as it is generated. As outlined in greater detail below, the relipidation of apoa-i may involve accepting lipids either from cells 10 12 or from pre-existing plasma lipoproteins. 50 There is also evidence to suggest that the lipid-poor apoa-i that dissociate from HDL during remodeling by CETP in vitro are also rapidly reincorporated back into pre-existing HDL. 13,46 Generation of Lipid-Poor apoa-i by Hepatic Lipase Lipid-poor apoa-i dissociates from HDL during the remodeling of HDL by triglyceride-rich lipoproteins, CETP, and hepatic lipase (Figure 2). This has been demonstrated in studies conducted in vitro in which human plasma has been supplemented with additional triglyceride-rich lipoproteins, CETP, and hepatic lipase and incubated at 37 C. 51 53 Such incubation results in the HDL becoming depleted of cholesteryl esters and reduced in particle size. Within 2 hours, about 30% of the apoa-i dissociates from the HDL. 53 However, with extension of the incubation beyond 2 hours, the apoa-i returns progressively to the HDL fraction. By 8 hours, the concentration of apoa-i in HDL is identical to that in nonincubated samples. The return of apoa-i to the HDL Figure 2. Dissociation of lipid-poor apoa-i from HDL accompanying the remodeling of HDL by triglyceride-rich lipoproteins, CETP, and hepatic lipase. CETP promotes the exchange of HDL cholesteryl esters for triglyceride in triglyceride-rich lipoproteins to form cholesteryl ester-depleted, triglyceride -enriched HDL particles. Subsequent hydrolysis of HDL triglyceride by hepatic lipase reduces the size of the HDL core. The consequent redundancy of surface constituents results in a dissociation of a proportion of the apoa-i from the HDL surface and thus the generation of a pool of monomolecular, lipid-free/lipid-poor apoa-i. CE indicates cholesteryl esters; TG, triglyceride; TGR-LP, triglyceride-rich lipoproteins; and HL, hepatic lipase. density range is accompanied by a progressive appearance in electron micrographs of discoidal HDL particles. 53 Thus, although the depletion of the core lipid content and the reduction in particle size of HDL promoted by lipid transfers and HL activity led to a shedding of apoa-i from the particle, the apoa-i was subsequently formed into new discoidal particles, presumably as a consequence of acquiring phospholipids released during the hydrolysis of triglyceride-rich lipoproteins. 50 Studies conducted in rabbits have provided strong evidence that the processes of HDL core lipid depletion and hepatic lipase-mediated hydrolysis of HDL triglyceride may also operate in vivo. Rabbits have a high level of activity of CETP 54,55 but are naturally deficient in hepatic lipase. 56 Introduction of the gene for human hepatic lipase into rabbits reduces HDL particle size 26,57 and increases the rate of apoa-i catabolism. 26,58 However, the appearance of lipidpoor apoa-i comparable to what has been reported in vitro has not been observed in vivo in these rabbits. This may reflect no more than the rapid relipidation of lipid-poor apoa-i, as has been observed in vitro. 50 Generation of Lipid-Poor apoa-i by PLTP Plasma factor that transfers phospholipids between plasma lipoproteins (PLTP) 59 most likely acts by forming a ternary complex between donor and acceptor particles 60 in a process that results not only in a redistribution of phospholipids between the two types of particles, but also in the remodeling of HDL. PLTP converts HDL 3 into larger and smaller particles and mediates the release of lipid-poor apoa-i. The physiological importance of this process is uncertain but in

Rye and Barter Metabolism of Lipid-Poor ApoA-I 425 Figure 3. Metabolic fate of lipid-poor apoa-i in plasma. The monomolecular form of prebeta-migrating, lipid-poor apoa-i in plasma may either be excreted through the kidney and be irreversibly lost or it may be relipidated and retained in the plasma. Relipidation may be achieved by direct incorporation of lipidpoor apoa-i into pre-existing, alpha-migrating, spherical HDL particles in a process linked to LCAT-mediated particle expansion. Lipid-poor apoa-i may also accept phospholipids and unesterified cholesterol whether by transfer from plasma lipoproteins or from cells in the process mediated by ABCA1, to form prebeta-migrating discoidal HDL. The prebeta-migrating, discoidal HDL are subsequently converted into alpha-migrating spherical HDL by the action of LCAT. species such as rats and mice that lack activity of CETP, PLTP may be responsible for generating most of the lipidpoor apoa-i that is required for initiating reverse cholesterol transport. The possibility that PLTP also mediates the dissociation of apoa-i in vivo is supported by studies in mice in which overexpression of human PLTP is accompanied by an increase in the level of prebeta-migrating apoa-i. 61,62 Metabolic Fate of Lipid-Poor apoa-i Any lipid-poor apoa-i that is generated in plasma may either be excreted through the kidney 63 and be irreversibly lost or it may be relipidated and retained in the plasma (Figure 3). Relipidation may be achieved by direct reincorporation of lipid-poor apoa-i into pre-existing HDL particles in a process linked to LCAT-mediated particle expansion. 64 Lipid-poor apoa-i may also accept phospholipids and unesterified cholesterol from plasma lipoproteins. 50 And finally, apoa-i may acquire phospholipids and cholesterol that are released from cells that express ABCA1. 10 12 The association of lipid-poor apoa-i with lipids from other lipoproteins and from cells generates prebeta-migrating discoidal HDL. Incorporation of Lipid-Poor apoa-i Into Pre-Existing Mature HDL It has been shown in vitro that when apoa-i dissociates from HDL, it may be reincorporated into other HDL in a process that is driven by LCAT. 64 In those studies, the lipoprotein fraction (d 1.21 g/ml) of human plasma was incubated with CETP under conditions that led to a reduction in particle size and the dissociation of about 30% of the apoa-i from the HDL. When the incubation was complete, the d 1.21 g/ml fraction was reisolated, supplemented with purified lipid-free apoa-i, and reincubated in either the presence or absence of LCAT. In the absence of LCAT, HDL size did not increase and lipid-free apoa-i was not incorporated into the HDL density range. However, when LCAT was included in the incubation mixture, the size of the HDL increased, lipid-free apoa-i was incorporated into the particles, and the apoa-i content of the HDL was restored to that of the original, unmodified particles. Evidence that incorporation of lipid-free apoa-i into preexisting HDL also occurs in vivo has been clearly demonstrated in recent studies in which rabbits were injected with 125 I-labeled lipid-free apoa-i. 26 Before injection, the 125 I- labeled apoa-i had a prebeta electrophoretic mobility and was smaller in size than native HDL. Within minutes after injection, the 125 I-labeled apoa-i appeared quantitatively in HDL-sized, alpha-migrating particles. Furthermore, the 125 I- labeled apoa-i decayed from plasma with a fractional catabolic rate that was identical to what was observed when it was injected into the rabbits as a component of intact spherical HDL particles. 26 Transfer of Lipids From Other Lipoproteins to Lipid-Poor apoa-i in Plasma Lipid-free apoa-i has the capacity to accept phospholipids and cholesterol from plasma lipoproteins in a process that is dependent on the presence of nonesterified fatty acids. 50 In studies conducted in vitro, incubation of lipid-free apoa-i with VLDL or LDL and a source of nonesterified fatty acids (whether as exogenous sodium oleate or generated endogenously by the interaction of lipoprotein lipase on VLDL) resulted in a significant proportion of the apoa-i being incorporated into discoidal particles within the HDL density range. These particles contained apoa-i, phospholipids, and unesterified cholesterol, had a prebeta mobility on agarose gel electrophoresis, and apparent diameters ranging from 7.6 to 10.6 nm when subjected to nondenaturing gradient gel electrophoresis. 50 Whether a comparable process operates in vivo, or if it is physiologically relevant, remains to be determined. Transfer of Lipids From Cells to Lipid-Poor apoa-i in Extracellular Space Lipid-poor apoa-i may enter the interstitial space and function as an acceptor of cell cholesterol in the process mediated by ABCA1. 10 12,35,65 ABCA1 translocates phospholipids and cholesterol from the inner to the outer leaflets of cell membranes where they are picked up by the lipid-poor apoa-i in the extracellular space. This interaction is limited to apoa-i that contains no or very little lipid. Precisely how lipid-poor apoa-i removes cholesterol from cells is not known. According to one view, apoa-i simultaneously removes both phospholipids and cholesterol from cell membranes. 66 Alternately, it has been suggested that lipid-poor apoa-i first acquires phospholipids from cells in an ABCA1- mediated process that results in the formation of prebeta migrating, discoidal HDL that are efficient acceptors of additional cell membrane unesterified cholesterol in a diffusion process that appears not to require ABCA1. 10,67 The prebeta-migrating, discoidal HDL then transport the cell cholesterol, via the lymphatics, to the plasma compartment.

426 Arterioscler Thromb Vasc Biol. March 2004 Factors That Impact the Dissociation of Lipid-Poor apoa-i From Mature HDL The generation of apoa-i during remodeling of HDL is influenced by factors in addition to CETP, hepatic lipase, and PLTP. These include the phospholipid composition of the particles and the presence of apoa-ii. Phospholipid Composition of HDL We have reported recently that phospholipids regulate lipidpoor apoa-i formation and the CETP-mediated remodeling of HDL. 68 These studies were performed using well-defined, monodisperse preparations of spherical reconstituted HDL (rhdl) that contained apoa-i as the only apolipoprotein and either 1-palmitoyl-2-oleoyl phosphatidylcholine (POPC), 1-palmitoyl-2-linoleoyl phosphatidylcholine (PLPC), 1-palmitoyl-2-arachidonyl phosphatidylcholine (PAPC), or 1-palmitoyl-2-docosahexanoyl phosphatidylcholine (PDPC) as the only phospholipid. When these rhdl were incubated in vitro with CETP and appropriate phospholipid/triolein microemulsions, core lipid depletion and particle remodeling varied widely. 68 After 24 hours, CETP had remodeled 38% of the (POPC)rHDL and 70% of the (PLPC)rHDL into small particles. In the case of (PAPC)rHDL and (PDPC)rHDL, 65% and 78% of the particles, respectively, had decreased in size. Although lipid-poor apoa-i did not dissociate from the (POPC)rHDL, it was apparent by 24 hours in the (PLPC)rHDL incubation, and by 12 hours in the (PAPC)rHDL and (PD- PC)rHDL incubations. Physicochemical studies established that the more rapid dissociation of lipid-poor apoa-i from (PAPC)rHDL and (PDPC)rHDL was because of enhanced core lipid depletion and the destabilization and progressive exclusion of apoa-i from the surface of the particles. 68 Presence of apoa-ii in HDL Until recently, the evidence as to whether apoa-ii affects the ability of lipid-poor apoa-i to dissociate from HDL has been conflicting. Using HDL isolated from human plasma, some investigators have reported that lipid-poor apoa-i is generated during the in vitro remodeling of (A-I/A-II)HDL by CETP, 69 whereas others have found this not to be the case. 70 This is not surprising given that human plasma HDL contain apolipoproteins other than apoa-i and apoa-ii, as well as plasma factors, such as paraoxonase. Although these constituents are likely to influence lipid-poor apoa-i formation, they have yet to be investigated systematically. It should also be noted that variations in HDL phospholipid composition are likely to contribute to the conflicting results outlined above. A recent study has provided unequivocal evidence that apoa-ii inhibits the dissociation of lipid-poor apoa-i from HDL. 71 This conclusion was reached by studying the CETPmediated remodeling of well-characterized preparations of spherical rhdl. These preparations were identical in size and phospholipid composition, had comparable lipid/apolipoprotein ratios, and contained either apoa-i as the only apolipoprotein, (A-I)rHDL, or apoa-i and apoa-ii, (A-I/A- II)rHDL. When the (A-I)rHDL were incubated with CETP, their core lipid content decreased by 25%. This was accompanied by a concomitant formation of lipid-poor apoa-i and the remodeling of all of the (A-I)rHDL into large and small particles. This was not the case for the (A-I/A-II)rHDL, in which only 30% of the (A-I/A-II)rHDL were remodeled into small particles. Large particles were not generated in these incubations. Furthermore, lipid-poor apoa-i did not dissociate from the (A-I/A-II)rHDL. The inability of lipid-poor apoa-i to dissociate from (A-I/A-II)rHDL was subsequently shown to be due to salt bridge formation between apoa-ii and the C-terminal domain of apoa-i. This enhanced the stability of the apoa-i, attenuated the remodeling of the (A-I/A- II)rHDL, and inhibited lipid poor apoa-i formation. 71 Conclusion The apoa-i in plasma is present mainly as a component of alpha-migrating, spherical HDL particles, with a small proportion existing as a component of prebeta-migrating discoidal HDL. ApoA-I may also exist transiently in a monomolecular form that contains little or no lipid. This lipid-free/ lipid-poor apoa-i has a prebeta mobility but is distinct from the prebeta-migrating discoidal HDL. This pool of lipid-free/ lipid-poor apoa-i functions as the preferred acceptor of cell lipids released in the process mediated by the ABCA1 transporter. There is compelling evidence that the apoa-i in plasma is subject to continual cycling between its lipid-rich and lipid-poor forms in processes that may impact substantially on HDL function and the operation of the reverse cholesterol transport pathway. By understanding what regulates the dissociation of monomolecular, lipid-free/lipid-poor apoa-i from spherical HDL and the subsequent metabolic fate of the dissociated apoa-i, it may be possible to identify potential new targets for therapies designed to enhance the anti-atherogenic potential of HDL. References 1. Gordon DJ, Knoke J, Probstfield JL, Superko R, Tyroler HA. Highdensity lipoprotein cholesterol and coronary heart disease in hypercholesterolemic men: the Lipid Research Clinics Coronary Primary Prevention Trial. Circulation. 1986;74:1217 1225. 2. Enger SC, Hjermann I, Foss OP, Helgeland A, Holme I, Leren P, Norum KR. High density lipoprotein cholesterol and myocardial infarction or sudden coronary death: a prospective case-control study in middle-aged men of the Oslo study. Artery. 1979;5:170 181. 3. Miller NE, Thelle DS, Forde OH, Mjos OD. The Tromso heart-study. High density lipoprotein and coronary heart-disease: a prospective casecontrol study. Lancet. 1977;1:965 968. 4. Goldbourt U, Medalie JH. High density lipoprotein cholesterol and incidence of coronary heart disease: the Israeli Ischemic Heart Disease Study. Am J Epidemiol. 1979;109:296 308. 5. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study. Am J Med. 1977;62:707 714. 6. Jacobs DR, Jr., Mebane IL, Bangdiwala SI, Criqui MH, Tyroler HA. High density lipoprotein cholesterol as a predictor of cardiovascular disease mortality in men and women: the follow-up study of the Lipid Research Clinics Prevalence Study. Am J Epidemiol. 1990;131:32 47. 7. Miller M, Seidler A, Kwiterovich PO, Pearson TA. Long-term predictors of subsequent cardiovascular events with coronary artery disease and desirable levels of plasma total cholesterol. Circulation. 1992;86: 1165 1170. 8. Pekkanen J, Linn S, Heiss G, Suchindran CM, Leon A, Rifkind BM, Tyroler HA. Ten-year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med. 1990;322:1700 1707. 9. Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995;36:211 228.

Rye and Barter Metabolism of Lipid-Poor ApoA-I 427 10. Yancey PG, Bortnick AE, Kellner-Weibel G, de la Llera-Moya M, Phillips MC, Rothblat GH. Importance of different pathways of cellular cholesterol efflux. Arterioscler Thromb Vasc Biol. 2003;23:712 719. 11. Oram JF. HDL Apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol. Arterioscler Thromb Vasc Biol. 2003;23: 720 727. 12. Wang N, Tall AR. Regulation and mechanism of ATP-binding cassette transporter A1-mediated cellular cholesterol efflux. Arterioscler Thromb Vasc Biol. 2003;23:1178 1184. 13. Rye KA, Clay MA, Barter PJ. Remodeling of high density lipoproteins by plasma factors. Atherosclerosis. 1999;145:227 238. 14. Jonas A, von Eckardstein A, Kezdy KE, Steinmetz A, Assmann G. Structural and functional properties of reconstituted high density lipoprotein discs prepared with six apolipoprotein A-I variants. J Lipid Res. 1991;32:97 106. 15. Nichols AV, Blanche PJ, Gong EL, Shore VG, Forte TM. Molecular pathways in the transformation of model discoidal lipoprotein complexes induced by lecithin:cholesterol acyltransferase. Biochim Biophys Acta. 1985;834:285 300. 16. Blanche PJ, Gong EL, Forte TM, Nichols AV. Characterization of human high-density lipoproteins by gradient gel electrophoresis. Biochim Biophys Acta. 1981;665:408 419. 17. Cheung MC, Albers JJ. Characterization of lipoprotein particles isolated by immunoaffinity chromatography: particles containing A-I and A-II and particles containing A-I but no A-II. J Biol Chem. 1984;259: 12201 12209. 18. Cheung MC, Albers JJ. Distribution of high density lipoprotein particles with different apoprotein composition: particles with A-I and A-II and particles with A-I but no A-II. J Lipid Res. 1982;23:747 753. 19. Barrans A, Jaspard B, Barbaras R, Chap H, Perret B, Collet X. Pre-b HDL: structure and metabolism. Biochim Biophys Acta. 1996;1300: 73 85. 20. Norum KR, Glomset JA, Nichols AV, Forte T. Plasma lipoproteins in familial lecithin: cholesterol acyltransferase deficiency: physical and chemical studies of low and high density lipoproteins. J Clin Invest. 1971;50:1131 1140. 21. Forte TM, Bielicki JK, Knoff L, McCall MR. Structural relationships between nascent apoa-i-containing particles that are extracellularly assembled in cell culture. J Lipid Res. 1996;37:1076 1085. 22. Asztalos B, Zhang W, Roheim PS, Wong L. Role of free apolipoprotein A-I in cholesterol efflux: formation of pre-alpha-migrating high-density lipoprotein particles. Arterioscler Thromb Vasc Biol. 1997;17: 1630 1636. 23. Huang Y, von Eckardstein A, Wu S, Maeda N, Assmann G. A plasma lipoprotein containing only apolipoprotein E and with gamma mobility on electrophoresis releases cholesterol from cells. Proc Natl Acad Sci U S A. 1994;91:1834 1838. 24. Barter PJ and Rye K-A. High density lipoproteins and coronary heart disease. Atherosclerosis. 1996;121:1 12. 25. Rye K-A, Hime N, Barter PJ. The influence of sphingomyelin on the structure and function of reconstituted high density lipoproteins. J Biol Chem. 1996;271:4243 4250. 26. Kee P, Rye K-A, Taylor JL, Barrett PH, Barter PJ. Metabolism of apoa-i as lipid-free protein or as component of discoidal and spherical reconstituted HDLs: studies in wild-type and hepatic lipase transgenic rabbits. Arterioscler Thromb Vasc Biol. 2002;22:1912 1917. 27. Kunitake ST, La Sala KJ, Kane JP. Apolipoprotein A-I-containing lipoproteins with pre-beta electrophoretic mobility. J Lipid Res. 1985;26: 549 555. 28. Castro GR, Fielding CJ. Early incorporation of cell-derived cholesterol into pre-beta-migrating high-density lipoprotein. Biochemistry. 1988;27: 25 29. 29. O Connor PM, Zysow BR, Schoenhaus SA, Ishida BY, Kunitake ST, Naya-Vigne JM, Duchateau PN, Redberg RF, Spencer SJ, Mark S, Mazur M, Heilbron DC, Jaaffe RB, Malloy MJ, Kane JP. Prebeta-1 HDL in plasma of normolipidemic individuals: influences of plasma lipoproteins, age, and gender. J Lipid Res. 1998;39:670 678. 30. Asztalos BF, Sloop CH, Wong L, Roheim PS. Two-dimensional electrophoresis of plasma lipoproteins: recognition of new apo A-I-containing subpopulations. Biochim Biophys Acta. 1993;1169:291 300. 31. Neary RH, Gowland E. Stability of free apolipoprotein A-1 concentration in serum, and its measurement in normal and hyperlipidemic subjects. Clin Chem. 1987;33:1163 1169. 32. Miida T, Kawano M, Fielding CJ, Fielding PE. Regulation of pre high density lipoprotein in normal plasma by cell membranes and lecithincholesterol acyltransferase activity. Biochemistry. 1992;31:11112 11117. 33. Miida T, Ozaki K, Murakami T, Kashiwa T, Yamadera T, Tsuda T, Inano K, Okada M. Pre 1-high density lipoprotein concentration can change low-density lipoprotein-cholesterol independent of cholesteryl ester transfer protein. Clin Chim Acta. 2000;292:69 80. 34. Ishida BY, Frolich J, Fielding CJ. Prebeta-migrating high density lipoprotein: quantitation in normal and hyperlipidemic plasma by solid phase radioimmunoassaay following electrophoretic transfer. J Lipid Res. 1987;28:778 786. 35. Daerr WH, Minzlaff U, Greten H. Quantitative determination of apolipoprotein A-I in high-density lipoproteins and free apolipoprotein A-I by two-dimensional agarose gel lipoprotein- rocket immunoelectrophoresis of human serum. Biochim Biophys Acta. 1986;879:134 139. 36. Borresen AL, Berg K. Presence of free apoa-i in serum: implications for immunological quantification of HDL and its apoproteins. Artery. 1980;7:139 160. 37. Liang H-Q, Rye K-A, Barter PJ. Cycling of apolipoprotein A-I between lipid-associated and lipid-free pools. Biochim Biophys Acta. 1995;1257: 31 37. 38. Hara H, Yokoyama S. Role of apolipoproteins in cholesterol efflux from macrophages to lipid microemulsion: proposal of a putative model for the pre- -high density lipoprotein pathway. Biochemistry. 1992;31: 2040 2046. 39. Dixon JL, Ginsberg HN. Hepatic synthesis of lipoproteins and apolipoproteins. Semin Liver Dis. 1992;12:364 372. 40. Danielsen EM, Hansen GH, Poulsen MD. Apical secretion of apolipoproteins from enterocytes. J Cell Biol. 1993;120:1347 1356. 41. Schaefer EJ, Wetzel MG, Bengtsson G, Scow RO, Brewer HB Jr, Olivecrona T. Transfer of human lymph chylomicron constituents to other lipoprotein density fractions during in vitro lipolysis. J Lipid Res. 1982; 23:1259 1273. 42. Glickman RM, Green PHR, Lees RS, Tall A. Apoprotein A-I synthesis in normal intestinal mucosa and in Tangier disease. N Engl J Med. 1978; 299:1424 1427. 43. Schonfeld G, Bell E, Alpers DH. Intestinal apoproteins during fat absorption. J Clin Invest. 1978;61:1539 1550. 44. Tall AR, Green PHR, Glickman RM, Riley JW. Metabolic fate of chylomicron, phospholipids and apoproteins in the rat. J Clin Invest. 1979; 64:977 989. 45. Schaefer EJ, Jenkins LL, Brewer HB Jr. Human chylomicron apolipoprotein metabolism. Biochem Biophys Res Commun. 1978;80: 405 412. 46. Barter PJ. Hugh Sinclair lecture: the regulation and remodeling of HDL by plasma factors. Atheroscler Suppl. 2002;3:39 47. 47. Huuskonen J, Olkkonen VM, Jauhiainen M, Ehnholm C. The impact of phospholipid transfer protein (PLTP) on HDL metabolism. Atherosclerosis. 2001;155:269 281. 48. Rye KA, Duong MN. Influence of phospholipid depletion on the size, structure, and remodeling of reconstituted high density lipoproteins. J Lipid Res. 2000;41:1640 1650. 49. Rye K-A, Hime NJ, Barter PJ. Evidence that CETP-mediated reductions in reconstituted high density lipoprotein size involve particle fusion. J Biol Chem. 1997;272:5953 5960. 50. Clay MA, Barter PJ. Formation of new HDL particles from lipid-free apolipoprotein A-I. J Lipid Res. 1996;7:1722 1732. 51. Barrans A, Collet X, Barbaras R, Jaspard B, Manent J, Vieu C, Chap H, Perret B. Hepatic lipase induces the formation of pre 1-high density lipoprotein (HDL) from triacylglycerol-rich HDL 2. J Biol Chem. 1994; 269:11572 11577. 52. Clay MA, Newnham HH, Barter PJ. Hepatic lipase promotes a loss of apolipoprotein A-I from triglyceride-enriched human high density lipoproteins during incubation in vitro. Arterioscler Thromb. 1991;11: 415 422. 53. Clay MA, Newnham HH, Forte TM, Barter PJ. Cholesteryl ester transfer protein and hepatic lipase activity promote shedding of apo A-I from HDL and subsequent formation of discoidal HDL. Biochim Biophys Acta. 1992;1124:52 58. 54. Ha YC, Barter PJ. Differences in plasma cholesteryl ester transfer activity in sixteen vertebrate species. Comp Biochem Physiol B. 1982;71: 265 269. 55. Tsutsumi K, Hagi A, Inoue Y. The relationship between plasma high density lipoprotein cholesterol levels and cholesteryl ester transfer protein

428 Arterioscler Thromb Vasc Biol. March 2004 activity in six species of healthy experimental animals. Biol Pharm Bull. 2001;24:579 581. 56. Clay MA, Hopkins GJ, Ehnholm CP, Barter PJ. The rabbit as an animal model of hepatic lipase deficiency. Biochim Biophys Acta. 1989;1002: 173 181. 57. Fan J, Wang J, Bensadoun A, Lauer SJ, Dang Q, Mahley RW, Taylor JM. Overexpression of hepatic lipase in transgenic rabbits leads to a marked reduction of plasma high density lipoproteins and intermediate density lipoproteins. Proc Natl Acad Sci U S A. 1994;91:8724 8728. 58. Rashid S, Trinh DK, Uffelman KD, Cohn JS, Rader DJ, Lewis G. Expression of human hepatic lipase in the rabbit model preferentially enhances the clearance of triglyceride-enriched versus native high-density lipoprotein apolipoprotein A-I. Circulation. 2003;107:3066 3072. 59. van Tol A. Phospholipid transfer protein. Curr Opin Lipidol. 2002;13: 135 139. 60. Lusa S, Jauhiainen M, Metso J, Somerharju P, Ehnholm C. The mechanism of human plasma phospholipid transfer protein-induced enlargement of high-density lipoprotein particles: evidence for particle fusion. Biochem J. 1996;313:275 282. 61. van Haperen R, van Tol A, Vermeulen P, Jauhiainen M, van Gent T, van den Berg P, Ehnholm S, Grosveld F, van der Kamp A, de Crom R. Human plasma phospholipid transfer protein increases the antiatherogenic potential of high density lipoproteins in transgenic mice. Arterioscler Thromb Vasc Biol. 2000;20:1082 1088. 62. Jiang X, Francone OL, Bruce C, Milne R, Mar J, Walsh A, Breslow JL, Tall AR. Increased prebeta-high density lipoprotein, apolipoprotein AI, and phospholipid in mice expressing the human phospholipid transfer protein and human apolipoprotein AI transgenes. J Clin Invest. 1996;98: 2373 2380. 63. Horowitz BS, Goldberg IJ, Merab J, Vanni TM, Ramakrishnan R, Ginsberg HN. Increased plasma and renal clearance of an exchangeable pool of apolipoprotein A-I in subjects with low levels of high density lipoprotein cholesterol. J Clin Invest. 1993;91:1743 1752. 64. Liang H-Q, Rye K-A, Barter PJ. Remodeling of reconstituted high density lipoproteins by lecithin: cholesterol acyltransferase. J. Lipid Res. 1996; 37:1962 1970. 65. Francis GA, Knopp RH, Oram JF. Defective removal of cellular cholesterol and phospholipids by aapolipoprotein A-I in Tangier disease. J Clin Invest. 1995;96:78 87. 66. Oram JF, Lawn RM. ABCA1: the gatekeeper for eliminating excess tissue cholesterol. J Lipid Res. 2001;42:1173 1179. 67. Fielding PE, Nagao K, Hakamata H, Chimini G, Fielding CJ. A two-step mechanism for free cholesterol and phospholipid efflux from human vascular cells to apolipoprotein A-1. Biochemistry. 2000;39: 14113 14120. 68. Rye K-A, Duong M, Psaltis MK, Curtiss LK, Bonnet DJ, Stocker R, Barter PJ. Evidence that phospholipids play a key role in pre- apoa-i formation and high-density lipoprotein remodeling. Biochemistry. 2002; 41:12538 12545. 69. Hennessy LK, Kunitake ST, Kane JP. Apolipoprotein A-I-containing lipoproteins, with or without apolipoprotein A-II, as progenitors of pre-beta high-density lipoprotein particles. Biochemistry. 1993;32: 5759 5765. 70. Calabresi L, Lucchini A, Vecchio G, Sirtori CR, Franceschini G. Human apolipoprotein A-II inhibits the formation of pre-beta high density lipoproteins. Biochim Biophys Acta. 1996;1304:32 42. 71. Rye K-A, Wee K, Curtiss L, Bonnet D, Barter P. Apolipoproteins A-II inhibits high density lipoprotein remodeling and lipid-poor apolipoproteins A-I formation. J Biol Chem. 2003;278:22530 22536.