Plasma lipid transport in the preruminant calf, Bos spp: primary structure of bovine apolipoprotein A-I.

The preruminant calf (Bos spp.) is a model of considerable interest with regard to hepatic and intestinal lipoprotein metabolism (Bauchart et al., J. Lipid Res. (1989) 30, 1499-1514 and Laplaud et al., J. Lipid Res. (1990) 31, 1781-1792). As a preliminary step towards future experiments dealing with HDL metabolism in the calf, we have purified apoA-I from this animal and determined its complete amino acid sequence. Thus, approx. 10% of calf apoA-I was shown to contain a propeptide, with the sequence Arg-His-Phe-Trp-Gln-Gln. Enzymatic cleavage of apoA-I resulted in 10 proteolytic peptides. The complete apoA-I sequence was obtained after alignment of peptides on the basis of their homologies with those from rabbit apoA-I. Thus calf apoA-I consists of 241 amino acid residues, and exhibits high sequence homology with all mammalian apoA-I's studied to date. The bovine protein contained 10 hydrophobic amphipathic helical regions, occurring between residues 43-64, 65-86, 87-97, 98-119, 120-141, 142-163, 164-184, 185-206, 207-217 and 218-241. A computer-constructed phylogenetic tree showed that bovine apoA-I was more closely related to its dog counterpart, including the presence of a single methionine, than to the corresponding macaque and human proteins. Comparative predictions of the respective antigenic structures of human and bovine apoA-I's using the Hopp-Woods algorithm indicated similar positions for all 13 detectable antigenic sites, among which 7 were of identical, or closely related, amino acid composition. This finding was confirmed by demonstration of partial immunological identity between the two proteins upon immunodiffusion analysis, a result obtained using a monospecific rabbit antiserum against bovine apoA-I. Finally, comparison of sequence homology between bovine apoA-I and the lecithin:cholesterol acyl transferase (LCAT) activating region of human apoC-I suggests that several LCAT activating domains may be present in calf apoA-I.

Interaction of glycosaminoglycans with lipoproteins.

Apolipoproteins B-100 and E are protein constituents of human plasma chylomicrons, very low (VLDL), and low density lipoproteins (LDL). The interaction of lipoproteins with cell receptors is mediated by apoB and E. Lipoproteins also bind to the extracellular matrix, such as glycosaminoglycans (GAG), forming insoluble complexes in the presence of Ca2+. The purpose of this study was to identify the GAG-binding domains in apoB and E. By a combination of fragmentation of the intact proteins, peptide synthesis and quantitative GAG-binding, domains in apoB and apoE were identified and are shown below. These domains contain clusters of basic amino acids that we suggest are required for GAG-binding. table; see text.

Apolipoprotein E: phospholipid binding studies with synthetic peptides containing the putative receptor binding region.

To define the lipid and receptor binding regions of apolipoprotein E (apoE), we have synthesized four peptides beginning at residue 169 and continuing through the putative receptor binding region and ending at residue 129 so as to include a proposed lipid binding domain. The peptides were synthesized by solid-phase techniques, cleaved with anhydrous HF, and purified by ion-exchange and semipreparative reversed-phase high-performance liquid chromatography (HPLC). The peptides had the correct amino acid composition and were greater than 99% pure by analytical reversed-phase HPLC. The circular dichroic spectrum of each peptide was recorded before and after mixing with dimyristoylphosphatidylcholine. With apoE (148-169), apoE (144-169), and apoE (139-169), no changes were observed in the ellipticity at 222 nm. However, with apoE (129-169), an increase in alpha-helicity to approximately 42% was observed. Density gradient ultracentrifugation of the lipid-peptide mixture permitted isolation of a complex with apoE (129-169) with a molar ratio of lipid to peptide of 125:1, which was stable to recentrifugation. The alpha-helicity of the peptide in the complex was estimated to be 56%. No complexes were isolated from the gradients of the shorter peptides. Therefore, we conclude that the amphipathic helix formed by residues 130-150 contains one of the lipid binding regions of apoE.

Characterization and amino-terminal sequence of apolipoprotein AI from plasma high density lipoproteins in the preruminant calf, Bos spp.

The major apolipoprotein of calf plasma high-density lipoproteins, apo-AI, has been isolated and characterized. Apolipoprotein AI (apo-AI) was separated from the protein moiety of high-density lipoproteins (d 1.090-1.180 g/ml) by preparative electrophoresis in SDS-polyacrylamide gels followed by electrophoretic elution. Purified calf apo-AI had an Mr of approx. 27,000-28,000 in SDS-polyacrylamide gels, resembling human apo-AI. The amino acid composition of calf apo-AI displayed an overall similarity to that of its human and other mammalian counterparts (baboon, dog, badger, rabbit, rat and mouse), but differed in having higher proportions of glutamic acid, alanine and isoleucine. Amino-terminal amino acid sequence analysis up to the 47th residue showed close homology between calf apo-AI and those of the mammals with which it was compared. However, residues 2, 7, 20 and 22 in calf AI (i.e. aspartic acid, serine, glutamic acid and isoleucine, respectively) were substituted by glutamic acid, proline or glutamine, aspartic acid, and valine or leucine respectively, in the other mammals.

Crystallization and preliminary x-ray diffraction study of synthetic apolipoprotein E fragment (residues 129-169).

Apolipoprotein E is a plasma protein comprised of a lipid binding region (which together with other apoproteins maintains the structure of lipoprotein particles) and a receptor binding domain (which interacts with cellular receptors for control of triglyceride and cholesterol metabolism). A peptide, comprising residues 129-169 of human apolipoprotein E, which contains both a putative lipid-binding region and receptor binding domain, has been synthesized by solid phase techniques. Diffraction quality crystals of the synthetic apolipoprotein E fragment129-169 have been obtained at room temperature by vapor diffusion with polyethylene glycol in the presence of the nonionic detergent beta-octylglucoside. The crystals have been characterized with x-radiation as orthorhombic, space group I222 or I2(1)2(1)2(1), with unit cell dimensions a = 61.91, b = 30.84, and c = 42.79 A. There are eight molecules per unit cell, with one molecule (Mr = 4771) in each asymmetric unit. Precession photographs show that crystals diffract beyond 2.7-A resolution and are stable in the x-ray beam at room temperature for at least 200 h; thus, they can be used to collect three-dimensional data for a detailed crystallographic analysis.

Apolipoprotein E: phospholipid binding studies with synthetic peptides from the carboxyl terminus.

We have previously shown that the synthetic peptide apoE(129-169) forms lipid-peptide complexes with dimyristoylphosphatidylcholine (DMPC) with an L:P molar ratio of 125:1; the peptide in the isolated complex contains approximately 56% alpha-helicity. These results verify the presence of an amphipathic alpha-helix in this region of apoE as predicted by Chou-Fasman analysis and hydrophobicity calculations. To further define the lipid binding regions of apoE, we have synthesized four peptides, apoE(211-243), -(202-243), -(267-286), and -(263-286), from the carboxyl terminus of apoE and studied their lipid binding properties; apoE(202-243) contains two potential amphipathic helices. Although all four peptides formed alpha-helices in the helix-forming solvent 30% hexafluoropropanol, we found that only apoE(263-286) formed a stable complex with DMPC. The peptide contained approximately 80% alpha-helicity, and its Trp fluorescence spectrum was blue-shifted by 20 nm in the complex which had an L:P ratio of 163:1. We conclude that this sequence is a newly identified lipid binding region of apoE and that the amphipathic helices 203-221 and 226-243 are too hydrophilic to bind phospholipid.

Determination of the molecular mass of apolipoprotein B-100. A chemical approach.

Apolipoprotein B-100 (apo B-100) is the protein ligand in low-density lipoproteins that binds to a specific cell-surface receptor. Its molecular mass has been a subject of controversy. We have determined the molecular mass of the protein by a chemical approach. After complete CNBr cleavage, the C-terminal fragment of apo B-100 was purified by reverse-phase h.p.l.c. Amino acid N- and C-terminal analyses confirm that this peptide represents the C-terminal peptide as deduced from the DNA sequence of a human apo B-100 cDNA clone. A chemically synthesized peptide was used to determine the recovery of the peptide (74.72%). On the basis of these data, the molecular mass of apo B-100 was determined to be 496.82 +/- 24.84 kDa.

Plasma lipid transport in the hedgehog: partial characterization of structure and function of apolipoprotein A-I.

Apart from exhibiting the presence of lipoprotein [a] in its plasma, another interest of the European hedgehog in lipoprotein research lies in the quantitative prominence of a complex spectrum of high density lipoproteins (HDL) and very high density lipoproteins (VHDL) as cholesterol transporters in plasma (Laplaud, P. M. et al. 1989. Biochim. Biophys. Acta. 1005: 143-156). We, therefore, initiated studies in the field of reverse cholesterol transport in the hedgehog. As a first step, we characterized apolipoprotein A-I (apoA-I), the main protein component of hedgehog HDL and VHDL. Proteolytic cleavage of apoA-I (M(r) approx. 27 kDa) using two different enzymes resulted in two sets of peptides that were subsequently purified by high performance liquid chromatography, and that allowed us determination of the complete protein sequence. Hedgehog apoA-I thus consists of 241 amino acid residues and exhibits an overall 58% homology to its human counterpart, i.e., the lowest value observed to date among mammalian species. However, it retained the general organization common to all known apoA-Is, i.e., a series of amphipathic helical segments punctuated by proline residues. Circular dichroism experiments indicated a helical content of approx. 45%, increasing to approx. 58% in the presence of lecithin unilamellar liposomes. Apart from other differences, amino acid composition analysis shows that hedgehog apoA-I contains four isoleucine residues, while this amino acid is totally absent from the corresponding protein in higher mammals. Polyclonal antibodies raised against hedgehog apoA-I failed to detect any cross-reactivity between the animal and human proteins, although comparative prediction of the respective antigenic structures using the Hopp-Woods algorithm indicated that several potentially antigenic sites may occur in similar regions of the protein. Finally, hedgehog apoA-I was shown to be able to activate lecithin:cholesterol acyl transferase, although it was 4 to 5 times less efficient in this respect than the human protein.

A nonexchangeable apolipoprotein E peptide that mediates binding to the low density lipoprotein receptor.

ApoE is a 34-kDa apoprotein that mediates lipoprotein binding to the low density lipoprotein (LDL) receptor and to the LDL receptor-related protein. Receptor binding is mediated by a highly basic, alpha-helical sequence of approximately 15 amino acids that interacts with cysteine-rich repeat regions of the receptor. To determine the relationship between the receptor binding and lipid associating properties of apoE, we have synthesized a series of apoE peptides containing all (residues 129-169) or part (residues 139-169, 144-169, and 148-169) of the receptor-binding domain. The lipophilicity of these peptides was increased by modification of their N termini by acylation with either palmitic acid (C16-apoE peptide) or the N,N-distearyl derivative of glycine (diC18-Gly-apoE peptide). The unmodified peptides demonstrated low affinity for lipid surfaces (Kd > 10(-5) M) and moderate alpha-helicity in the presence of lipid (40%) and had no effect on LDL uptake by fibroblasts. N-Palmitoyl peptides had increased affinity for lipid (Kd approximately 10(-6) M) and increased alpha-helicity (55%) in the presence of lipid. The addition of the C16-apoE-(129-169)-peptide to 125I-LDL enhanced its uptake and degradation by fibroblasts 8-10-fold; however, < 50% of the degradation was mediated by the LDL receptor. By contrast, the diC18-Gly-apoE-(129-169)-peptide was essentially nonexchangeable (Kd < or = 10(-9) M) and highly helical (78%) in the presence of lipid. The addition of the diC18-Gly-apoE-(129-169)-peptide to 125I-LDL enhanced the specific uptake and degradation of LDL by both LDL receptor-mediated and non-LDL receptor-mediated mechanisms. Uptake and degradation of methylated LDL containing diC18-Gly-apoE-(129-169) revealed that the lipoprotein-bound peptide is the active agent. In agreement with this finding, a mutant diC18-Gly-apoE peptide (Arg142-->Gln) was much less effective than the wild-type peptide in potentiating binding, uptake, and degradation of 125I-LDL. Complexes of diC18-Gly-apoE-(129-169), apoA-I, and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine containing four to six copies of the peptide/particle displayed an affinity for the LDL receptor similar to that of apoE-L-alpha-dimyristoylphosphatidylcholine discs containing four copies of apoE.(ABSTRACT TRUNCATED AT 400 WORDS)

Structure and orientation of apo B-100 peptides into a lipid bilayer.

Peptides corresponding to lipid binding domains of Apo B-100 were synthesized, purified, and incubated with dimyristoylphosphatidylcholine (DMPC) liposomes. The secondary structure of the apo B-100 peptide-lipid complexes was evaluated by attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR). Those peptides belonging to the hydrophobic "core" domain of apo B-100 when associated with phospholipids were rich in beta sheet structure; a predominant alpha helical conformation was shown to be associated with one peptide located in a surface region of apo B-100. IR dichroic spectra revealed, in the case of the "core" peptides, that the beta sheet component is the only oriented structure with respect to the phospholipid acyl chains. This orientation of the beta sheet was recently found in LDL particles after proteolytic digestion by trypsin (Goormaghtigh, E., Cabiaux, V., De Meutter, J., Rosseneu, M., and Ruysschaert, J. M., 1993, Biochemistry 32, 6104-6110). Altogether, the data suggest that beta sheet, present in a high proportion in the native apo B-100, is probably another protein structure in addition to the amphipathic helix which strongly interacts with the lipid outer layer surrounding the LDL particle.

Kinetics and mechanism of transfer of synthetic model apolipoproteins.

The effect of hydrophobicity on the rate and mechanism of transfer of a synthetic amphiphilic peptide between phosphatidylcholine single bilayer vesicles has been evaluated. These peptides, which had the sequence Cn-SSLKEYWSSLKESFS (where Cn represents a saturated acyl chain of n carbons that is attached to the amino terminus of the peptide and n = 8, 12, or 16), were distinguished by the length of the saturated acyl chain of n carbons that was covalently bonded to the amino terminus. The transfer of the peptides was monitored by following the rate of change of the intrinsic tryptophan fluorescence that followed mixing of donor vesicles, which contained peptide, phosphatidylcholine, and a fluorescence quencher, with acceptors composed only of phosphatidylcholine. The transfer rates were independent of the structure and concentration of the acceptor. The kinetics were biexponential with the contribution of the fast and slow components being nearly equal. The rates of both components decreased with increasing acyl chain length; the respective free energies of activation were linear with respect to the acyl chain length. These results showed that, unlike lipid transfer, peptide transfer is not always a simple unimolecular process. However, like lipid transfer, the transfer rates are a predictable function of hydrophobicity. It is proposed that the peptides exist as dimers on the phospholipid surface and that the two components of transfer are due to sequential transfer of each molecule in a dimer.

Binding of a high reactive heparin to human apolipoprotein E: identification of two heparin-binding domains.

Ligand-blotting and dot-blotting procedures were used to investigate the binding of [125I]-heparin to apolipoprotein E, its thrombin fragments E22 (residues 1-191) and E12 (residues 192-299), and to nine apolipoprotein E synthetic fragments. E22 and E12 bound [125I] heparin indicating multiple heparin-binding domains. Synthetic peptides of apoE corresponding to residues 129-169, 139-169, and 144-169, but not 148-169, bound [125I] heparin suggesting that residues 144-147 (Leu-Arg-Lys-Arg) in E22 are important for binding. Peptide 202-243 and 211-243 but not 219-243 bound [125I] heparin suggesting that residues 211-218 (Gly-Glu-Arg-Leu-Arg-Ala-Arg-Met) comprise a portion of the E12 heparin-binding domain.

Sequence, structure, receptor-binding domains and internal repeats of human apolipoprotein B-100.

Apolipoprotein (apo) B-100, the major protein component in low density lipoprotein (LDL), is the ligand that binds to the LDL receptor. It is important in the metabolism of LDL and elevated plasma levels of LDL-apo B are strongly associated with increased risk of coronary artery disease. Although apo B-100 is of great clinical and biological importance its primary structure has defied chemical elucidation, mainly because of its enormous size, insolubility, and tendency to aggregate. Less than 5% of the apo B-100 sequence has been reported, despite the efforts of many laboratories over the past twenty years. Here we report the complete amino acid sequence of human apo B-100 as deducted by sequence analysis of complementary DNA clones; 2,366 of the 4,536 residues were also confirmed by direct sequencing of apo B-100 tryptic peptides. The distribution of trypsin-accessible and -inaccessible peptides of the protein on LDL is non-random and they can be grouped into 5 hypothetical domains. Of 20 potential N-glycosylation sites identified in the sequence, 13 were found by direct peptide sequencing to be glycosylated, and 4 unglycosylated. Examination of the primary structure of apo B-100 reveals that it contains a large number of long (greater than 70 residues) internal repeats and an even larger number of shorter ones, suggesting that the apo B-100 sequence was derived largely from internal duplications. Finally, using synthetic peptides of a specific region of apo B-100, we have identified a potential LDL receptor-binding domain (residues 3,345-3,381) which can bind to the LDL receptor and suppress 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase activities in cultured human fibroblasts.