Abstract 393: Covalently Connected Apolipoprotein A-I Molecules Provide Support for the Double Belt Model in Reconstituted HDL

Abstract

Because of its critical role in high density lipoprotein (HDL) metabolism, significant work has been devoted to studying apolipoprotein (apo)A-I structure in its lipid-bound state, particularly reconstituted forms of HDL (rHDL). Unfortunately, high-resolution structural studies have not been successful because of HDL heterogeneity and compositional dynamics. There is general agreement that apoA-I adopts an antiparallel arrangement on rHDL with the majority of the molecule conforming to the “double belt” model. However, less is understood about the locations of the N- and C-termini in these particles. The classic double belt predicts that, in a particle containing two molecules of apoA-I, all four termini are localized to one area of the particle. However, other models predict that they are far apart. To address this issue, we generated a series of covalently connected apoA-I dimers and tested their ability to generate rHDL particles. The dimeric (d)-apoA-I C-N mutant, locked the C-terminus of one apoA-I molecule to the N-terminus of a second molecule with three intervening Ala residues. Linked by strategically introduced cysteine residues, the d-apoA-I C-C , and d-apoA-I N-N mutants lock the C-termini and the N-termini of two apoA-I molecules together, respectively. When all three of these mutants were reconstituted into rHDL particles using sodium cholate and synthetic phospholipids, the resulting particles closely resembled those generated with wild-type (WT) apoA-I. This indicates a ring-like structure predicted by the double belt. However, when testing the ability of these mutants to spontaneously clear lipid, we found that both the N- and C- termini required the ability to separate and move independently to form normal particles. Native gel analysis showed the dimerized lipid-free mutants present as a tight homogenous band compared to wild-type which runs as a mixture of oligomers. We are currently evaluating if the homogeneity of these mutants can allow for more facile crystallization of full-length apoA-I. Further studies with these stabilized mutants should help us understand the structural basis underlying HDL’s diverse functions, which is essential for developing HDL-based therapies for cardiovascular disease.