Peptide-Induced Formation of Cholesterol-Rich Domains † Richard M. Epand,* ,‡ Brian G. Sayer, § and Raquel F. Epand ‡ Department of Biochemistry, McMaster UniVersity, Hamilton, Ontario L8N 3Z5, Canada, and Department of Chemistry, McMaster UniVersity, Hamilton, Ontario L8S 4M1, Canada ReceiVed September 3, 2003; ReVised Manuscript ReceiVed October 9, 2003 ABSTRACT: The peptide N-acetyl-LWYIK-amide causes the reorganization of bilayers of phosphatidylcholine and cholesterol to produce domains enriched in cholesterol. At a cholesterol mol fraction of 0.5, addition of N-acetyl-LWYIK-amide results in the formation of cholesterol crystallites. Addition of this peptide to mixtures of 1-stearoyl-2-oleoylphosphatidylcholine with lower mol fractions of cholesterol results in an increase in the enthalpy of the chain melting transition of the phospholipid, indicating the depletion of cholesterol from a domain in the membrane. The peptide binds to membranes both with and without cholesterol. However, 1 H magic-angle spinning (MAS) nuclear Overhauser effect spectroscopy (NOESY) indicates that in the presence of cholesterol the peptide has greater penetration into the bilayer. 13 C MAS NMR indicates that the peptide has stronger interactions with the A ring of cholesterol than it does with the interior of the bilayer. These results are in contrast with those of another peptide, N-acetyl-KYWFYR- amide, which does not promote the formation of cholesterol crystallites and does not show preferential interaction with cholesterol by NMR. Therefore, cholesterol can promote the insertion of N-acetyl-LWYIK- amide into a membrane and this peptide will sequester cholesterol into domains. These properties help to explain the observation that this sequence is found to be important in causing the fusion protein of human immunodeficiency virus (HIV) to sequester into raft domains in biological membranes. There is considerable current interest in the formation and properties of cholesterol-rich domains in biological mem- branes (for a recent series of reviews, see ref 1). One type of cholesterol-rich domain is the caveolae that are enriched in the lipids cholesterol and sphingomyelin. Domains of similar lipid composition, termed rafts, are thought to also be floating in the plasma membrane. These domains seques- ter certain proteins but not others. Most GPI-anchored proteins and many palmitoylated proteins are found in cholesterol-rich domains (2). There is also evidence to suggest that both the mechanical properties and the avoidance of length mismatch between transmembrane segments and bilayer thickness contribute to the partitioning of proteins into different membrane domains (3). In addition, it has been suggested that some amino acid sequences also favor partitioning into raft domains. There is a structural motif present in the V3 loop of HIV-1 1 gp 120, the human prion protein (PrP), and the Alzheimer -amyloid peptide (4) as well as other proteins (5) that is believed to target these proteins to raft domains by having affinity for sphingomyelin. However, cholesterol is also required for the high-affinity binding of one of these segments to membranes, the V3 loop of gp120 (6). There may also be other cholesterol-rich domains in membranes that do not fit the definition of rafts because they do not have a high concentration of sphingo- myelin, they are not in the liquid-ordered phase, and they are not insoluble in 1% Triton X-100 at 4 °C. Certain proteins and peptides have affinity for cholesterol-rich domains. It has been shown that a protein found in neuronal rafts, NAP- 22, binds to membranes containing cholesterol (7, 8). It has also been found that a toxic peptide, perfringolysin O, binds to cholesterol-rich domains in membranes (9-11). For another group of proteins, there is evidence for the existence of a consensus sequence, having the pattern -L/V-(X)(1- 5)-Y-(X)(1-5)-R/K-, in which (X)(1-5) represents 1-5 residues of any amino acid that recognizes cholesterol (12). The HIV-1 fusion protein gp41 has a segment consistent with this consensus sequence that is adjacent to the transmembrane anchor. This segment contains the sequence LWYIK and has been shown to promote membrane fusion by mutational studies of the intact viral protein (13) as well as with the use of a 20-amino acid synthetic peptide (14). This synthetic peptide contains the LWYIK sequence at the carboxyl terminal end, but the amino terminal segment of this peptide is also required for membrane fusion, perhaps by facilitating oligomerization of gp41 (50). Cholesterol was found to be required for HIV infection (15-17) as well as for fusion promoted by the synthetic peptide (14). Depletion of cholesterol from HIV results in loss of their infectivity (51). † This work was supported by a grant from the Canadian Institutes of Health Research, Grant MT-7654. * To whom correspondence should be addressed: e-mail epand@ mcmaster.ca. ‡ Department of Biochemistry, McMaster University. § Department of Chemistry, McMaster University. 1 Abbreviations: PC, phosphatidylcholine; DO, dioleoyl; PO, 1-palmit- oyl-2-oleoyl; SO, 1-stearoyl-2-oleoyl; DNS-PE, N-[5-(dimethylamino)- naphthalene-1-sulfonyl]-1,2-dihexadecanoyl-sn-glycero-3-phosphoetha- nolamine, triethylammonium salt; CP, cross polarization; MAS, magic angle spinning; MLV, multilamellar vesicle; LUV, large unilamellar vesicle; SUV, small unilamellar vesicle; DSC, differential scanning calorimetry; T m, transition temperature; ∆Hcal, calorimetric enthalpy; HIV-1, human immunodeficiency virus type 1; EDTA, ethylenedi- aminetetraacetic acid; Hepes, N-(2-hydroxyethyl)piperazine-N′-2- ethanesulfonic acid; NOESY, nuclear Overhauser effect spectroscopy; PIPES, piperazine-N,N′-bis(2-ethanesulfonic acid); 1-D and 2-D, one- and two-dimensional. 14677 Biochemistry 2003, 42, 14677-14689 10.1021/bi035587j CCC: $25.00 © 2003 American Chemical Society Published on Web 11/14/2003