Structure-function correlates in human apolipoprotein A-1

Apolipoprotein (apo) A-I is the major protein component of human high density lipoprotein (HDL). The plasma level of HDL or apo A-I is inversely related to the risk of developing atherosclerosis and the protective effect of HDL is believed to be related to its ability to promote the movement of cholesterol from peripheral cells to the liver. Apo A-I is involved in the remodelling of HDL that accompanies this reverse” cholesterol transport but the elements of its structure that are responsible for its function are not well understood. In this thesis, in vitro mutagenesis and eucaryotic expression were used to study the structure and function of recombinant human apo A-I. Four expression systems were developed and were utilized to express the wild-type and mutant cDNA constructs. In vitro translation studies in rabbit reticulocyte lysate established that the cDNA encodes the precursor, preproapo A-I. This precursor was proteolytically processed to proapo A-I on addition of microsomal membranes, simulating in vivo translocation and processing on the membrane of the endoplasmic reticulum (ER). Apo A-I was synthesized and secreted constitutively from the precursor cDNA by three eucaryotic cell, types: transformed simian kidney (COS), baby hamster kidney (BHK) and chinese hamster ovary (CHO) cells. The apo A-I secreted by COS and BHK cells was proapo A-I, while CHO cells secreted mature apo A-I, indicating that the latter possess propeptide proteolytic activity. Low level propeptide hydrolysis was also detected during long term culture collections from BHK cells. In defined, serum free culture conditions, much of the apo A-I synthesized was eventually degraded. Long term collections of medium were found to contain the following levels of apo A-I: COS ≈10ng/mI, BHK ≈100 ng/ml and CHO ≈130ng/ml per 24 hours. Proapo A-I secreted from COS cells was readily integrated into the HDL density fraction of fetal bovine serum. The majority of the apo A-I secreted by CHO cells was lipid-free or lipid-poor but retained its ability to integrate into liposomes in vitro. A large portion of the apo A-I within COS cells retained its signal peptide sequence following translation, indicating that ER processing of preproapo A-I was inefficient. It was concluded that COS cells were a poor model for large scale apo A-I expression and CHO cells were the best model system for this purpose. The role of the apo A-I propeptide was investigated in BHK cells expressing apo A-I. The results indicated that the propeptide was required for efficient cellular transport and secretion of apo A I. Removal of the propeptide from the cDNA sequence had no effect on the rate of apo A-I synthesis or on the fidelity of signal peptide hydrolysis, but the altered protein remained in the cells in large vesicular structures which had some morphologic features of the ER. This change also appeared to reduce the rate of apo A-I degradation. The observations suggested that non-hepatic cells expressing apo A-I degrade a substantial portion of this protein. Furthermore, removing the propeptide caused much of the apo A-I to remain in the cell, perhaps by preventing the movement of the protein out of the ER. The functional roles of the middle and C-terminal regions of the apo A-I sequence were also investigated by generating mutants in these regions. Deletion of L10y7s (a naturally occurring mutation with functional abnormalities) had minimal influence on the ability of these proteins to bind to liposomes, although the resulting complexes were more heterogeneous by density gradient centrifugation. Deletion of one amphipathic helix from the C-terminus of apo A-I altered its ability to form stable lipidprotein complexes when compared to wild-type. While recombinant wild-type apo A-I was approximately 80% as effective a lecithin:cholesterol acyltransferase (LCAT) activator, the Lys’° 7 and a -helix deletion mutants were extremely poor LCAT activators. In conclusion, the results indicate that the propeptide portion of apo A-I is involved in the cellular transport of apo A-I and might regulate the movement of proapo A-I between the ER and the Golgi apparatus. Furthermore, low affinity amphipathic helices in the middle hinge region and the C terminal region of the apo A-I sequence may play a significant role in the LCAT activating mechanism. These elements do not appear to be major structural determinants of initial phospholipid binding but may be involved in subsequent transformations of HDL. Extension of these studies will provide important insight into the mechanisms underlying the anti-atherogenic properties of HDL.Medicine, Faculty ofPathology and Laboratory Medicine, Department ofGraduat