Juxta Membrane Domains

Juxta-membrane domains can be distinguished from transmembrane helices or lipidation in that these segments do not penetrate deeply into the bilayer. One type of juxta-membrane domain is a cluster of cationic amino-acid residues that can bind to the surface of anionic membranes (Ellena et al. 2004). There are some lipoproteins, such as NAP-22, in which a smaller cationic cluster on the protein is found close to the amino terminus that is post translationally modified by myristoylation. In such cases the cationic cluster may sequester adjacent to the membrane surface because of a combination of lipid anchoring and electrostatic interactions. In addition, many of these cationic clusters, such as MARCKS, also contain aromatic amino-acid residues that will also contribute to membrane binding. The consequence of electrostatic interactions between cationic segments of proteins and anionic lipids has been described in Sect. 4.3.1.

There are also a few examples of protein segments that are adjacent to transmembrane domains that have been shown to be important in sequestering the protein to raft domains in membranes. One example of this is the so-called scaffolding domain of caveolin-1. Caveolin-1 is a major protein component of ca-veolae which are small structures of 50-100 nm in the plasma membrane that bud inward toward the cell (Anderson 1998). The lipid composition of caveolae is enriched in cholesterol and sphingomyelin, the principal components of raft domains. The protein caveolin-1 has three palmitoyl chains that would facilitate its incorporation into raft-like domains. However, when the three sites of palmi-toylation are mutated, the resulting protein still translocates to caveolae (Dietzen et al. 1995). This suggests that segments of the protein itself facilitate the interaction of caveolin-1 with cholesterol-rich domains. Mutational studies indicate that the segment comprising residues 82-101 is necessary and sufficient for membrane binding and has been termed the scaffolding domain (Schlegel et al. 1999). A recent fluorescence microscopy study has shown that a synthetic peptide corresponding to the scaffolding domain of caveolin can recruit NBD-labeled forms of cholesterol and acidic lipids (Wanaski et al. 2003).

Another example of a juxta-membrane domain, adjacent to a transmembrane segment, is the Trp-rich region of the gp41 protein of HIV. This region has been shown to be important for the infectivity of the virus (Salzwedel et al. 1999) and for sequestering the protein to cholesterol-rich domains (Saez-Cirion et al. 2002; Epand et al. 2003). Evidence that HIV also buds from raft domains includes the finding that the viral envelope is enriched in both lipids (Aloia et al. 1993) and GPI-anchored proteins (Esser et al. 2001; Nguyen and Hildreth 2000) found in "raft" domains of mammalian membranes. Several components that facilitate the entry of HIV into cells are suggested to be associated with raft domains in membranes, including cholesterol, CD4, the coreceptors CXCR4 and CCR5, as well as glycosphingolipids (Viard et al. 2004). However, viral entry into cells does not necessarily require the presence of cholesterol (Viard et al. 2002), but the removal of cholesterol with methyl-P-cyclodextrin from cells with relatively low receptor densities reduces the capacity of HIV-1 to trigger fusion (Viard et al. 2002). Extraction of cholesterol from the envelope of HIV-1 or SIV also results in reduced infectivity (Graham et al. 2003).

There is a relationship between the juxta-membrane regions of caveolin-1 and of the gp41 of HIV. Both segments conform to the requirements for a Cholesterol Recognition/interaction Amino acid Consensus (CRAC) motif (Li and Pa-padopoulos 1998) that was developed by analyzing amino acid sequences of proteins that interact with cholesterol (Li et al. 2001). A CRAC sequence is defined by having the pattern -L/V-(X)(l-5)-Y-(X)(l-5)-R/K-, in which (X)(l-5) represents between one and five residues of any amino acid.

Caveolin-1 has a juxta-membrane segment, residues 94-101, with the sequence VTKYWFYR that conforms to a CRAC motif. A shortened version of this segment that does not fulfill the requirements of the consensus sequence, KY-WFYR, was found to sequester the protein to membranes, but is not sufficient to target the protein to the cholesterol-rich domain of caveolae (Woodman et al. 2002). In addition, the synthetic peptide, N-acetyl-KYWFYR-amide did not induce the formation of cholesterol-rich domains in liposomes despite the segment having four sequential aromatic residues (Epand et al. 2003). The segment N-acetyl-FTVTKYWFYR-amide was shown to bind weakly to lipid bilayers in the absence of cholesterol (Arbuzova et al. 2000). Two longer peptide segments of caveolin that do correspond to the requirements of the consensus sequence of Li and Papadopolus (1998), i.e. N-acetyl-VTKYWFYR-amide (residues 94-101) and

N-acetyl-GIWKASFTTFTVTKYWFYRL-amide (residues 83-102), exhibit preferential interaction with cholesterol (Epand et al. 2005c).

The Trp-rich juxta-membrane domain of the gp41 of HIV has the sequence LWYIK. The peptide N-acetyl-LWYIK-amide, also having a CRAC motif, has preferential interaction with cholesterol-rich domains (Epand et al. 2003). The sequence LWYIK is located at the carboxyl-terminus of a longer membrane-proximal region that is rich in Trp. It has been shown by mutational studies that this region is important for membrane fusion and virus infectivity (Salzwedel et al. 1999). This longer membrane-proximal sequence has also been studied as an isolated 20 amino acid, synthetic peptide and found to aggregate and promote fusion of liposomes containing sphingomyelin and cholesterol, the principal lipid components of membrane rafts (Saez-Cirion et al. 2002; Shnaper et al. 2004). The amino-terminal region of this peptide is responsible for the formation of peptide oligomers (Saez-Cirion et al. 2003).

There is not a great deal of information regarding the nature of the molecular interactions of the various residues of the CRAC motif with membranes. To begin with, the consensus sequence is rather general, having two segments that vary in length between one and five amino acid residues. This would suggest that the CRAC motif does not have a specific conformation since a unique folding pattern would require that there be a certain number of amino acids between one required residue and another. There are only three required residues, L/V, Y and R/K. In the case of LWYIK it was suggested that the Y residue interacts with the A ring of cholesterol (Epand et al. 2003), possibly through hydrophobic interactions of two conformationally restrained moieties. It is not known, however, if the Y can be replaced by another aromatic amino-acid residue. With regard to the requirement for an L or a V at the amino terminal side of the CRAC sequence, it is possible that this facilitates non-specific sequestering to a membrane by increasing the hydrophobicity of the segment. Might then an I also be able to substitute in this position? The last requirement, an R or K on the carboxyl side of the CRAC motif, suggests that a charge interaction is important. We have shown that peptides with a CRAC motif will promote the formation of cholesterol-rich domains even in the absence of anionic lipids. It is possible that on the cytoplasmic surface of mammalian cell membranes the cationic residue in the CRAC motif is important for interacting with anionic lipids; alternatively this group may play a role in modulating the depth of insertion of the peptide into the membrane. In that regard, the fact that the CRAC motif allows up to 10 residues of any type (as many as five before and five after the Y) without restriction, would allow for a wide range of peptide hydrophobicities. It seems likely that these 10 residues are not completely undetermined, but that the requirements for these segments have not yet been elucidated.

We have studied several peptides with variations of the CRAC motifs of the scaffolding region of caveolin-1 and ofTrp-rich region of the gp41 ofHIV. In the case of caveolin-1, shortening the segment to KYWFYR, a sequence that has four successive aromatic amino acids and a C-terminal R, does not result in preferential interaction with cholesterol (Epand et al. 2003) and this segment is not sufficient to target caveolin to the cholesterol-rich domain of caveolae (Woodman et al. 2002). Lengthening this segment to make N-acetyl-VTKYWFYR-amide re-

suits in a peptide that exhibits some preference for cholesterol-rich domains. This peptide, unlike N-acetyl-KYWFYR-amide, fulfills the requirements of a CRAC sequence (Li and Papadopoulos 1998). A longer peptide, N-acetyl-GIWKASFT-TFTVTKYWFYRL-amide, is even more potent in promoting the formation of cholesterol crystals compared with N-acetyl-VTKYWFYR-amide. The ability of CRAC sequences to promote the formation of cholesterol crystals cannot be a consequence of the peptide binding to cholesterol. Such an interaction would inhibit formation of cholesterol crystals by stabilizing the cholesterol that is bound to the peptide. Apparently, however, the CRAC peptide can alter the properties of the membrane in such a manner as to recruit more molecules of cholesterol to a specific domain, thus passing its solubility limit in the membrane. In the case of the Trp-rich region of the gp41 of HIV, we find the opposite, i.e. the longer the segment of the juxta-membrane region, the less able is the peptide to sequester cholesterol into domains. We find that the short peptide N-acetyl-LWYIK-amide is more effective than a peptide, N-acetyl-KWASLWNWFNITNWLWYIK-amide, corresponding to the full Trp-rich, membrane-proximal segment of HIV-1Hxb2> residues 665-683 of the gp41 protein, in promoting the formation of cholesterol-rich domains (Epand et al. 2005a). The results with the membrane proximal region of HIV gp41 are opposite to thoes found with the caveolin scaffolding domain. In the case of gp41, the longer peptide, N-acetyl-KWASLWNWFNITNWL-WYIK-amide, with more aromatic groups present, is less capable of promoting the formation of cholesterol-rich domains. Perhaps being more hydrophobic it can no longer distinguish between specific lipid molecules in sequestering to membranes (Epand et al. 2005a).

In the HIV-1 sequence database there are variants having a Ser at position 681 and that therefore do not formally have a CRAC motif. In addition, HIV-2 as well as several strains of SIV are also devoid of a CRAC motif. For example, for HIV-2CBL24, SIVcpZ(Q88oo4), SIVmac25i, SIVagm and SIVsm84, the segment 679-683, adjacent to the transmembrane segment, has the sequence LASWIK (Vincent et al. 2002), similar to the sequence LWYIK of HIV-1, but not fulfilling the requirements of the CRAC motif because it has no Tyr (Li and Papadopoulos 1998). Altering the sequence of LWYIK to LASWIK reduced the lipid preference of the peptide (Epand et al. 2005a).

The presence of a CRAC motif does not strictly correlate with the ability of a peptide to sequester cholesterol into a domain. Nevertheless, a CRAC motif is found in the juxta-membrane region of two proteins that associate with cholesterol-rich domains of biological membranes, i.e. caveolin-1 and HIV gp41. This suggests that segments that conform to the CRAC motif may have greater cholesterol-sequestering ability than similar peptides, such as LASWIK, that do not conform to the requirements of this domain. Of course, LASWIK is very similar to a CRAC sequence and it has partial cholesterol-sequestering activity. There are also caveats in comparing the results of isolated peptides with liposomes and the interaction of full-length proteins with biological membranes. The peptide may fold into a different conformation in the full-length protein. In the case of the gp41 fragment, N-acetyl-KWASLWNWFNITNWLWYIK-amide, the amino-terminal region of this peptide is responsible for the formation of oligomers (Saez-Cirion et al. 2003). This may occur more readily in the intact proteins, whereas in the isolated peptide this region is freer to interact non-selectively with a membrane.

It is clear that juxta-membrane segments of a protein can cause the sequestering of cholesterol into domains. The precise correlation between amino acid sequence and length and the ability to recruit cholesterol is yet to be determined, as is the functional role of each amino-acid side-chain. Although the interaction does not appear to be very specific, the peptide N-acetyl-LWYIK-amide exhibits chiral specificity for cholesterol (Epand et al. 2005b). The aromatic side-chains by themselves are not chiral. One would therefore, a priori, not expect an effect of peptide chirality on the interaction with cholesterol. This is supported by the finding that the chirality of the LWYIK peptide has little to do with the extent of segregation of cholesterol. Cholesterol chirality has a small effect on lipid-lipid interactions that give rise to the formation of cholesterol-rich domains in the absence of peptide (Lalitha et al. 2001a,b). However, the ability of the L-enantiomer of N-acetyl-LWYIK-amide to sequester cholesterol into domains is strongly dependent on the chirality of the sterol (Epand et al. 2005b). Our results indicate that the membrane interactions of peptides and proteins that induce the formation of cholesterol-rich domains are altered by the small change in the arrangement and physical properties of cholesterol-containing bilayers as a result of changes in cholesterol chirality. As a consequence, the peptide-induced segregation of cholesterol into domains is strongly affected by cholesterol chirality, even though the peptide-lipid interactions that are involved are not very specific.

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook

Post a comment