Arrangement of apolipoprotein A-I in reconstituted high-density lipoprotein disks: An alternative model based on fluorescence resonance energy transfer experiments

The folding and organization of apolipoprotein A-I (apoA-I) in discoidal, h igh-density lipoprotein (HDL) complexes with phospholipids are not yet comp letely resolved. For about 20 years, it was generally accepted that the amp hipathic helices of apoA-I lie parallel to the acyl chains of the phospholi pids ("picket fence" model). However, based on the X-ray crystal structure of a large, lipid-free fragment of apoA-I, a "belt model" was recently prop osed. In this model, the helices of two antiparallel apoA-I molecules are e xtended in a circular arrangement and lie perpendicular to the phospholipid acyl chains. To obtain conclusive information on the spatial organization of apoA-I in discoidal HDL, we engineered three separate cysteine mutants o f apoA-I (D9C, A124C, A232C) for specific labeling with the fluorescence pr obes ALEXA-488 or ALEXA-546 (fluorescein and rhodamine derivatives). The la beled apoA-I was reconstituted into well-defined HDL complexes containing t wo molecules of protein and dipalmitoylphosphatidylcholine, and the complex es were used in three quantitative fluorescence resonance energy transfer ( FRET) experiments to determine the distances between two specific sites in an HDL particle. Comparison of the distances measured by FRET (4.7-7.8 nm) with those predicted from the existing models indicated that neither the pi cket fence nor the belt model can account for the experimental results; rat her, a hairpin folding of each apoA-I monomer with most helices perpendicul ar to the phospholipid acyl chains and a random head-to-tail and head-to-he ad arrangement of the two apoA-I molecules in the HDL particles are strongl y suggested by the distance and lifetime data.