The Role of Cholesterol and Glycosylphosphatidylinositol-anchored Proteins of Erythrocyte Rafts in Regulating Raft Protein Content and Malarial Infection

Abstract

Human erythrocytes are terminally differentiated, nonendocytic cells that lack all intracellular organelles. Here we show that their plasma membranes contain detergent-resistant membrane rafts that constitute a small fraction (4%) of the total membrane protein, with a complex mixture of proteins that differentially associate with rafts. Depletion of raft-cholesterol abrogates association of all proteins with no significant effect on cholesterol:protein ratios in the rest of the membrane, lipid asymmetry, deformability, or transport properties of the bilayer, indicating that cholesterol is critical for protein assembly into rafts and suggesting that rafts have little influence on several erythrocyte functions. Erythrocytes from patients with paroxysmal nocturnal hemoglobinuria, which lack glycosylphosphatidylinositol-anchored proteins, show significant elevation in raft-cholesterol but no increase in raft protein association, suggesting that raft assembly does not require glycosylphosphatidylinositol-anchored proteins, raft proteins do not bind directly to cholesterol, and only threshold levels of raft-cholesterol are critical for protein recruitment. Loss of glycosylphosphatidylinositol-anchored proteins had no effect on erythrocytic infection by malarial parasite or movement of raft markers into the parasite's vacuole. However, infection is blocked following raft-cholesterol disruption, suggesting that erythrocyte rafts can be functionally exploited and providing the first evidence for the involvement of host rafts in an apicomplexan infection. Human erythrocytes are terminally differentiated, nonendocytic cells that lack all intracellular organelles. Here we show that their plasma membranes contain detergent-resistant membrane rafts that constitute a small fraction (4%) of the total membrane protein, with a complex mixture of proteins that differentially associate with rafts. Depletion of raft-cholesterol abrogates association of all proteins with no significant effect on cholesterol:protein ratios in the rest of the membrane, lipid asymmetry, deformability, or transport properties of the bilayer, indicating that cholesterol is critical for protein assembly into rafts and suggesting that rafts have little influence on several erythrocyte functions. Erythrocytes from patients with paroxysmal nocturnal hemoglobinuria, which lack glycosylphosphatidylinositol-anchored proteins, show significant elevation in raft-cholesterol but no increase in raft protein association, suggesting that raft assembly does not require glycosylphosphatidylinositol-anchored proteins, raft proteins do not bind directly to cholesterol, and only threshold levels of raft-cholesterol are critical for protein recruitment. Loss of glycosylphosphatidylinositol-anchored proteins had no effect on erythrocytic infection by malarial parasite or movement of raft markers into the parasite's vacuole. However, infection is blocked following raft-cholesterol disruption, suggesting that erythrocyte rafts can be functionally exploited and providing the first evidence for the involvement of host rafts in an apicomplexan infection. detergent-resistant membrane, GPI, glycosylphosphatidylinositol methyl-β-cyclodextrin phosphate-buffered saline paroxysmal nocturnal hemoglobinuria fluorescein isothiocyanate polyacrylamide gel electrophoresis parasitophorous vacuolar membrane Membrane microdomains or rafts reflect lateral heterogeneities in lipid bilayers of cellular membranes (1Simons K. Ikonen E. Nature. 1997; 387: 569-572Crossref PubMed Scopus (8157) Google Scholar). They have been implicated in a variety of sorting and signaling processes in cells (2Brown D.A. London E. Annu. Rev. Cell Dev. Biol. 1998; 14: 111-136Crossref PubMed Scopus (2557) Google Scholar). They are presumably present in all cells and can be functionally isolated as detergent-resistant membranes (DRMs)1 by virtue of their property of being insoluble in cold, nonionic detergents (3Brown D.A. Rose J.K. Cell. 1992; 68: 533-544Abstract Full Text PDF PubMed Scopus (2618) Google Scholar). Despite some earlier concerns about DRMs, as reviewed by Brown and London (2Brown D.A. London E. Annu. Rev. Cell Dev. Biol. 1998; 14: 111-136Crossref PubMed Scopus (2557) Google Scholar), DRMs reflect at least a subset of raft-sorting properties in the membrane and have been very useful tools for studying rafts. But it is difficult to separate DRMs isolated from different compartments of a cell and analyze their resident proteins, which underlie the catalytic properties of interest ascribed to rafts. Mature human erythrocytes are unique in that they are terminally differentiated cells that comprise a plasma membrane and dense cytoplasm enriched in hemoglobin. They are devoid of intracellular organelles, lack endocytic machinery, and are incapable of de novoprotein or lipid biosynthesis (4Chasis J.A. Prenant M. Leung A. Mohandas N. Blood. 1989; 74: 1112-1120Crossref PubMed Google Scholar). We have recently shown that from mature erythrocytes, glycosylphosphatidylinositol (GPI)-anchored proteins, a transmembrane receptor, and a cytoplasmically associated signaling molecule can be isolated in DRMs that float with densities expected for rafts on sucrose gradients (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). But a basic characterization of the complexity of protein composition of isolated plasma membrane rafts, their association with cytoskeletal proteins, the relative abundance of cholesterol, its regulation of raft protein composition, and the role of rafts in erythrocyte function are not known. Biophysical studies using fluorescence anisotropy and chemical cross-linking data show rafts as submicron (70–200-nm) clusters of GPI-anchored proteins on cellular plasma membranes maintained at high levels of cholesterol (6Varma R. Mayor S. Nature. 1998; 394: 798-801Crossref PubMed Scopus (1031) Google Scholar, 7Friedrichson T. Kurzchalia T.V. Nature. 1998; 394: 802-805Crossref PubMed Scopus (480) Google Scholar). Rafts were proposed by Simons and Ikonen (1Simons K. Ikonen E. Nature. 1997; 387: 569-572Crossref PubMed Scopus (8157) Google Scholar) to provide a model for sphingolipid-based sorting of apical proteins and basolateral proteins in polarized epithelial cells. Later, this model of hydrogen-bonded sphingolipid domains incorporated a basis for cholesterol-based microheterogeneity seen in membranes by proposing that cholesterol fills gaps between glycosphingolipids. Independent studies by Brown and co-workers (2Brown D.A. London E. Annu. Rev. Cell Dev. Biol. 1998; 14: 111-136Crossref PubMed Scopus (2557) Google Scholar, 8Brown D.A. London E. J. Biol. Chem. 2000; 275: 17221-17224Abstract Full Text Full Text PDF PubMed Scopus (2065) Google Scholar, 9Ahmed S.N. Brown D.A. London E. Biochemistry. 1997; 36: 10944-10953Crossref PubMed Scopus (615) Google Scholar, D.A. London E. J. Biol. 1998; PubMed Scopus Google Scholar) have to the of a model that that between lipid are critical for raft as domains in a that are not for raft Brown D.A. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). been shown that proteins associated with rafts in the of cholesterol S. J. 1998; PubMed Scopus Google Scholar), but cholesterol is for raft assembly in membranes not been the role of GPI-anchored proteins, to be for raft not been GPI-anchored proteins by virtue of the of on the that the lipid in the membrane to the protein in the GPI-anchored proteins have which their into by virtue of and GPI-anchored proteins to raft composition and sorting in cells. Rafts have been implicated in a variety of processes (1Simons K. Ikonen E. Nature. 1997; 387: 569-572Crossref PubMed Scopus (8157) Google Scholar, D.A. London E. Annu. Rev. Cell Dev. Biol. 1998; 14: 111-136Crossref PubMed Scopus (2557) Google Scholar, Simons K. Cell Biol. 1997; PubMed Scopus Google Scholar). been proposed that at cellular plasma rafts to of the to into membrane of T. Simons K. Cell Biol. 1997; PubMed Scopus Google Scholar). However, studies that rafts and that this is the of their sorting and in cells S. S. EMBO J. 1998; PubMed Scopus Google Scholar). erythrocytes are they can be by a of vacuolar the We have shown that lipid and protein of rafts are into the the and the erythrocyte and cytoskeletal proteins are not (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). cholesterol is critical for the but cholesterol in the erythrocyte membrane is for malarial and this is by GPI-anchored proteins is not known. the present we show that rafts from erythrocyte plasma membranes contain a small but complex of proteins and cholesterol levels with of the plasma of raft-cholesterol all raft protein association, indicating that cholesterol is critical for all protein assembly into rafts. Loss of GPI-anchored proteins is associated with cholesterol but this does not increase raft association of proteins, suggesting that raft assembly does not require GPI-anchored proteins and that raft proteins do not bind directly to of rafts no effect on the deformability, or transport properties of the it can be to malarial suggesting that rafts provide of vacuolar in the erythrocyte were from or J. T. were using with the and were at or were in human in at as PubMed Scopus Google Scholar). were with K. A. Cell Biol. PubMed Scopus Google Scholar). following were to from to and or and as to directly to Cell were in at and to to with for at cells were in and with following which using in cells were with with blocked with by with the to and to with with were with or with as and were on an fluorescence and a by from as (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). were the of the and to and D.A. Cell Biol. Google Scholar). the of data and data from all of the in its and it to an data can be for of in the and the of the and were by that the fluorescence a of and that the fluorescence with the parasite as as the cell were this with a as as with and from at least cells of from to mature in the of were and were to the parasite and by that had no cells. and fluorescence levels were by with or not the and fluorescence and densities associated with the parasite and erythrocytes were in of the fluorescence by the vacuolar we associated with the erythrocyte membrane and the for the in the of and on erythrocytes, the data are as a fraction of the total fluorescence by the parasite as a function of of of at least cells on a parasite were to and shown in of cells and malarial infection as (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). erythrocytes, of were in of methyl-β-cyclodextrin and for at following which and cells were with for in with human at at the of which by or to cholesterol, which as Biochemistry. PubMed Scopus Google Scholar). of at least in cells of infection and of at the of were as K. A. Cell Biol. PubMed Scopus Google Scholar), with an of erythrocytes from paroxysmal nocturnal hemoglobinuria patients or in of cell at of erythrocyte of the at and to were with or erythrocyte as the be to the as that for the erythrocyte and were maintained the of for GPI-anchored on and cells by of erythrocyte in and in of of to human for at cells were with and in in for following which parasite were with for an cell in were on a of were on a total of at least to erythrocyte were with and fluorescence and were on a or erythrocyte from cells were by Blood. 1998; PubMed Google Scholar) and on in of in saline and and to sucrose (3Brown D.A. Rose J.K. Cell. 1992; 68: 533-544Abstract Full Text PDF PubMed Scopus (2618) Google Scholar, S. J. 1998; PubMed Scopus Google Scholar). at the with of sucrose and with of sucrose and of sucrose in saline and to in a at for at as the first the were as and their protein were the proteins were by in and were to electrophoresis by and with the in of proteins in the were by or the and the were from were on of erythrocyte membranes or of sucrose using a and from is a in which cholesterol is to with cholesterol is a is on an that cholesterol and and to the on and human erythrocytes were with a of to and the is as a function of at a of using an of at is to the of the erythrocytes and to membrane Mohandas N. Blood. PubMed Google Scholar). Cell were for and to the of were and on a were with erythrocyte as Blood. 1998; PubMed Google Scholar) or in the of and were using erythrocytes in the of a of studies have shown that DRMs GPI-anchored proteins by and signaling proteins can be isolated from a variety of cells. studies on the of rafts from mature erythrocytes have been Blood. 1998; PubMed Google Scholar) the of GPI-anchored proteins in detergent-resistant membrane isolated at We have recently shown that detergent-resistant with densities expected for rafts were enriched in GPI-anchored proteins and a transmembrane as and associated with the of the erythrocyte membrane (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). rafts were isolated on gradients that comprise a of with of and sucrose is to be enriched in rafts, the sucrose total cellular that the protein of the fraction is unique and it with that of the cell from the and the rafts were to and the total protein of the fraction is with fraction as by Despite the fraction protein between and is as in the fraction as in the protein between and between and at and at to be and to fraction erythrocyte membrane to of and and as as the seen in the are in fraction data that the erythrocyte fraction is enriched in a subset of proteins and a we had shown that erythrocyte raft are enriched in GPI-anchored proteins, the transmembrane and the signaling molecule but not S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google and we the of raft and have been recently as markers for rafts, and been to J. Biol. Chem. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar). to be present in erythrocytes J. Cell Biol. 1997; PubMed Scopus Google Scholar). to its in erythrocytes show that proteins and earlier markers are associated with rafts in fraction which to the sucrose membrane rafts of in and that DRMs be J. Biol. Chem. 1997; Full Text Full Text PDF PubMed Scopus Google Scholar) a of in the raft reflect in raft the of proteins S. K. T. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). transmembrane proteins, as the and in or DRMs in a variety of cells E. J. A. J. J. 1998; Google Scholar, J. Google Scholar, J. Biol. Chem. 2000; 275: Full Text Full Text PDF PubMed Scopus Google Scholar), are enriched in fraction not that plasma membrane DRMs from erythrocytes contain transmembrane as We were to levels of erythrocyte proteins as and in suggesting the of between erythrocyte DRMs and data show that erythrocyte rafts have complex protein composition with several transmembrane and all of the signaling protein their protein is relative to the total membrane protein of the erythrocyte and cholesterol is to be enriched in rafts and critical to their we in shown in of the total cellular cholesterol is in the to a of of of protein is the cholesterol:protein in the the is ratios in the are with the of the data show that DRMs are enriched in cholesterol relative to the rest of the plasma of a protein or lipid isolated in DRMs at not reflect the in which it is in rafts at However, the is at the cholesterol:protein ratios that we in isolated DRMs present an for ratios in cells a E. Brown D.A. 2000; PubMed Scopus Google Scholar). data that cholesterol:protein ratios in rafts have the to increase by as as relative to that in domains of the plasma the data that and raft properties are in cells and that erythrocytes provide an model in which to plasma membrane rafts. cholesterol been shown to an role in raft D.A. London E. J. Biol. 1998; PubMed Scopus Google Scholar, E. Brown D.A. 2000; PubMed Scopus Google Scholar), we the of its on erythrocyte rafts. We erythrocytes of cholesterol by with which in a in cellular cholesterol levels and the cells to and sucrose of the cholesterol of the that the cholesterol of fraction from cells to the and cholesterol from and cells to and the total of cholesterol from and the of cholesterol in the raft of the protein of that raft are of protein as shown in the cholesterol:protein of the membrane is of the by that raft proteins as and to float as DRMs, indicating that a critical of cholesterol is to association with plasma membrane rafts and the of association of a protein with rafts its to bind to cholesterol have properties of the we the on cell of transport of and as N. Haldar K. 1997; PubMed Scopus Google Scholar, K. de J. Cell Biol. 1989; PubMed Scopus Google Scholar). shown in cholesterol had no effect on membrane transport of and of or membrane by Mohandas N. Blood. PubMed Google Scholar, N. Google Scholar). only effect of cholesterol is a very small in membrane of the cells. that raft association of cholesterol and proteins no significant on membrane of erythrocytes, the of rafts are in the erythrocyte of on erythrocyte deformability, lipid asymmetry, and on and human erythrocytes of at is to the of the cells and membrane of in membrane deformability, at of as to the were with erythrocyte as or in the of and and were on a as for is shown for to with no significant between and on and human erythrocytes of at is to the of the cells and membrane Mohandas N. Blood. PubMed Google Scholar). of in membrane deformability, at of as N. Google to the were with erythrocyte as Blood. 1998; PubMed Google or in the of and and were on a as N. Haldar K. 1997; PubMed Scopus Google Scholar) for is shown for to with no significant between and in a this we of cholesterol a role in malarial infection. studies have that erythrocytes with can a in from to K. PubMed Scopus Google Scholar). cells show an in malarial However, the studies not rafts or membrane properties of the erythrocytes or cellular properties by cholesterol we have shown that raft are from the erythrocytes into the malarial (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). we were in of association of protein and cholesterol had effect on of vacuolar infection the movement of into the vacuole. shown in cells with of as as cells show levels of parasite with which cholesterol and by that raft-cholesterol is for of on studies of parasite of erythrocytes, we are that the of in little or no effect on the of to cells. We do not on cholesterol and is seen in cells. is cholesterol can and high for rafts, and some DRMs cellular infection which are at for in the of that raft-cholesterol is for malarial by rafts. is a from a in the in or of GPI-anchored proteins on the cell 1997; PubMed Scopus Google Scholar). studies of of erythrocytes from patients have J. PubMed Scopus Google Scholar). in of of cells were to be in GPI-anchored proteins, as by using as the GPI-anchored protein on for a and of the cell are shown in of total membrane from the with that they to contain total cholesterol and on relative to that seen for erythrocytes at that be a between and cholesterol We for between the of and cholesterol shown in is a in cholesterol levels and of We the cholesterol total cellular cholesterol is of cells with the is shown in DRMs isolated from cells show a increase in their cholesterol of the cholesterol:protein ratios that fraction a of which is a increase with that in cells. a cholesterol:protein of and show to the in cells. that a of cholesterol with rafts. We cells and their isolated rafts to the of for raft-cholesterol on association and malarial infection. shown in the protein of fraction the with cells. to of we were to of the by and protein from cells. But of can be isolated from erythrocytes and were by as shown in the of and does not show in cells with their and are in DRMs of cells is in the of markers and are seen in association of raft proteins, as as to erythrocyte proteins as or cytoskeletal as into DRMs that raft protein composition is not by the of GPI-anchored raft and the associated of GPI-anchored proteins increase cholesterol levels in this does not have a significant effect on resident protein composition and of cholesterol in fraction is to that seen in the fraction from cells. However, from fraction in raft be is no significant in the of raft proteins in the between and cells. that cholesterol is to be in rafts, an increase levels does not resident proteins of rafts. does it association with or in the erythrocyte to or in erythrocyte DRMs the of and erythrocyte rafts the mature erythrocytes do not their membrane to and is no evidence for movement in However, the human parasite a in erythrocyte N. Haldar K. 1997; PubMed Scopus Google Scholar). and with an of membranes raft proteins and are and into the (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar), and we this to provide a of their movement from the erythrocyte membrane into an cholesterol not the composition of proteins, it the of movement of raft in the We malarial infection and vacuolar of raft markers in cells as with cells. studies have that cells are by the parasite PubMed Google Scholar, J. H. M. M. J. 1997; PubMed Scopus Google Scholar). However, the of of the parasite the and the of proteins and infection not been to analyze parasite by the of using A. S. M. 14: PubMed Scopus Google Scholar). for erythrocyte and parasite we are to infection by the We and infection of erythrocytes by its and the of the into mature increase in but do not the of a increase in and of which is in the fluorescence with for erythrocyte and parasite we were to between and in erythrocytes and cells from the using shown in no in is seen between and cells. the mature into no is seen between different and cells. were to for the in fluorescence in the seen between and cells. no in parasite of or to between cells and cells in the cells with be to analyze the of cholesterol on the movement of raft in erythrocytic cell sorting of cells and and and show different of infection of cells at different levels of in different of and and cells were in the as all a of and in and which to that seen in cells and the not in a show different of infection of cells at different levels of in different of and and cells were in the as all a of and in and which to that seen in cells and the not cells with were with that with cholesterol to shown in cholesterol is seen in the of the vacuolar in the membrane indicating that it is as in cells (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). for cholesterol and erythrocyte GPI-anchored using and to human to that cholesterol seen in erythrocytes of and the of not and that are and protein raft which are into the malarial in erythrocytes (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar), are in cells and of in cells the as seen for and in cells data that movement of raft into an in the erythrocyte by malarial infection is not by at and cholesterol GPI-anchored proteins have no effect on movement of raft into the malarial vacuole. We provide a of the protein and cholesterol of rafts from Erythrocytes do not contain intracellular DRMs from cells contain only plasma membrane rafts. show that rafts are with the rest of the membrane, they contain a unique subset of erythrocyte proteins that and signaling proteins, with their function in signaling and sorting processes in cells. rafts associate with as their do not to and at the of rafts associate with the and they studies in the that of lipid rafts, and at T. Simons K. J. PubMed Scopus Google Scholar), suggesting that in cells movement of rafts require between lipid raft and the high for on were shown to be by J. Cell 2000; PubMed Google Scholar). cross-linking of at at of not by that association with lipid rafts is a and to be and not to the is no evidence for erythrocyte raft with of the but association with be be that require an an by shown in cells M. K. A. Biol. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar) and to membrane raft domains T. Simons K. J. PubMed Scopus Google Scholar). erythrocytes are they can be by the human parasite earlier and present data on membrane transport into show that transmembrane host proteins can this the erythrocyte plasma membrane We have shown that only proteins are (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). present data indicating that of cholesterol raft assembly and the of that rafts a critical role in of the vacuole. raft are and it is that rafts be at or parasite at the of of erythrocytic rafts with the to their movement into the vacuole. is a complex studies have that is a receptor, that be and in infection J. PubMed Scopus Google Scholar, M. Google Scholar, PubMed Google Scholar). is not suggesting that is by this receptor, the of rafts of the parasite proteins be into the to erythrocyte rafts and vacuolar rafts. We have shown that a protein which been shown to to the by is that rafts in the (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar). and rafts be as a model to the at the it is that host raft proteins as directly with and in erythrocytes, providing a for the of in cells been shown to the of S. K. T. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). is the first transmembrane protein of apicomplexan to be in DRMs (5Lauer S. VanWye J. Harrison T. McManus H. Samuel B.U. Hiller N.L. Mohandas N. Haldar K. EMBO J. 2000; 19: 3556-3564Crossref PubMed Scopus (195) Google Scholar), but and their and characterization parasite that this in or is to require signaling and not be erythrocytes are to in is that the plasma membrane cholesterol in cells is an to the cells from to in an with plasma membrane cholesterol in a been in and its cholesterol, have been to be in processes 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). plasma membrane cholesterol increase the and the of by complex in cells have been shown to be to proteins T. M. H. PubMed Scopus Google Scholar, H. T. J. 1997; Google Scholar, H. J. Biol. Chem. Full Text PDF PubMed Google Scholar). by been shown to be to membrane and cholesterol J. Google Scholar). the of erythrocytes to by with cholesterol of the membrane J. Google Scholar), indicating that cholesterol with the of S. A. PubMed Scopus Google Scholar), involvement of protein T. M. H. PubMed Scopus Google Scholar, H. T. J. 1997; Google Scholar, H. J. Biol. Chem. Full Text PDF PubMed Google Scholar), and present it is that complex membrane rafts. of raft composition and assembly to studies show that are not for rafts Brown D.A. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). We that in cholesterol is and rafts with protein composition and that is from cells. do not influence raft assembly plasma membrane rafts be in cholesterol of the erythrocyte membrane, suggesting that rafts lipid gradients to be in the membrane However, cholesterol is to raft protein association, and levels of cholesterol had no on raft assembly or movement in that properties in microdomains at a threshold be for of proteins to rafts. We are to for and R. for to human