Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of variations in HDL composition on components of reverse cholesterol transport Sparks, Daniel Leslie 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1989_A1 S63_5.pdf [ 9.89MB ]
Metadata
JSON: 831-1.0098201.json
JSON-LD: 831-1.0098201-ld.json
RDF/XML (Pretty): 831-1.0098201-rdf.xml
RDF/JSON: 831-1.0098201-rdf.json
Turtle: 831-1.0098201-turtle.txt
N-Triples: 831-1.0098201-rdf-ntriples.txt
Original Record: 831-1.0098201-source.json
Full Text
831-1.0098201-fulltext.txt
Citation
831-1.0098201.ris

Full Text

THE EFFECT OF VARIATIONS IN HDL COMPOSITION ON COMPONENTS OF REVERSE CHOLESTEROL TRANSPORT by . DANIEL LESLIE SPARKS B.Sc, University of British Columbia, 1984 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Pathology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1989 ©Daniel Leslie Sparks, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada Date Ml* 27, /?$? DE-6 (2/88) ABSTRACT This thesis is concerned with the mechanism which underlies the apparent anti-atherogenic capacity of high density lipoproteins (HDL). Specifically, the relationship between HDL composition, cholesteryl ester transfer protein (CETP) and lecithin: cholesterol acyltransferase (LCAT) was investigated. A novel assay was developed which allowed for the determination of the rate of transfer of [3H]cholesteryl ester from agarose-bound HDL to the endogenous lipoproteins of a plasma sample. Comparison of this assay with an established method, which utilized exogenous lipoprotein substrates, indicated a significant positive correlation (r = 0.960) between the assays for the rates measured for normolipidemic subjects. When the same relationship was studied in 30 patients with lipid metabolic disorders, however, no such correlation could be demonstrated (r=0.147). This suggests that lipoprotein composition and content may not affect total cholesteryl ester transfer activity in normolipidemic, but may markedly affect the function of CETP in the plasma of hyperlipidemic patients. When cholesteryl ester transfer was determined in plasma of 50 patients with defined primary disorders of lipid metabolism, transfer to the plasma HDL pool was shown to be significantly reduced in almost all patient groups. Reduced transfer to HDL occurred in samples with altered HDL composition; particularly where HDL-triglyceride was significantly increased and HDL-cholesteryl esters were reduced. Transfer to LDL and VLDL was increased in patients with dysbetalipoproteinemia and hypoalphalipoproteinemia. In addition, HDL ii unesterified cholesterol content was significantly increased and cholesteryl ester transfer to H D L 3 was significantly reduced in patients with documented evidence of vascular disease. These findings indicate that impaired interaction of CETP with the HDL pool may contribute to the risk of coronary heart disease in patients with specific plasma lipid abnormalities. The mechanism underlying the results described above was investigated in vitro. Incubations utilized recombinant high density lipoproteins (rHDL) prepared by short duration co-sonication. Increasing the triglyceride content, relative to cholesteryl ester, significantly decreased the ability of the particles to accept cholesteryl esters transferred by CETP. When the free cholesterol content was increased relative to phospholipid, the ability of the particles to accept cholesteryl esters was also decreased in a similar manner. Since this corroborates the clinical observations, these findings indicate that altered HDL composition may have marked effects on the transfer and equilibration of cholesteryl esters within this lipoprotein pool. In addition, when rHDL were characterized as substrates for purified human LCAT, it was shown that increasing the triglyceride content relative to cholesteryl ester in rHDL also markedly decreased the maximum catalytic potential. These findings suggest that altered HDL composition results in abnormal interactions between HDL, CETP and LCAT. This may impair the equilibration of cholesteryl esters and their subsequent removal from the HDL pool. The consequence may be increased transfer and accumulation of cholesteryl esters in lower density lipoproteins; an event which promotes atherogenesis. iii TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS v i ABBREVIATIONS x LIST OF TABLES '. x i i i LIST OF FIGURES x i v ACKNOWLEDGEMENTS xv DEDICATION . . x v i 1 INTRODUCTION 1.1 A t h e r o s c l e r o s i s , l i p i d s , and HDL 1 1.2 L i p o p r o t e i n s 2 1.2.1 L i p o p r o t e i n s t r u c t u r e and m e t a b o l i s m 2 1.2.2 Reverse c h o l e s t e r o l t r a n s p o r t 3 1.2.3 HDL s t r u c t u r e and h e t e r o g e n e i t y 5 1.2.4 L e c i t h i n : c h o l e s t e r o l a c y l t r a n s f e r a s e 6 1.2.5 N e u t r a l l i p i d t r a n s f e r r e a c t i o n s 8 1.2.5.1 I n t r o d u c t i o n 8 1.2.5.2 B i o c h e m i c a l c h a r a c t e r i z a t i o n 8 1.2.5.3 Mechanism o f a c t i o n . . 9 1.2.5.4 Role i n l i p o p r o t e i n m e t a b o l i s m 12 1.2.5.5 Role i n h y p e r l i p i d e m i a 14 1.3. Recombinant l i p o p r o t e i n systems 17 1.3.1 H i s t o r i c a l p e r s p e c t i v e s 17 1.3.2 S t r u c t u r e and metabolism 18 1.4 R a t i o n a l e f o r t h i s study 19 1.5 S p e c i f i c aims 21 iv 2 MATERIAL AND METHODS 2.1 Materials 22 2.2 Lipoproteins 22 2.2.1 Isolation 22 2.2.2 Lipoprotein characterization 22 2.2.2.1 Composition analyses 23 2.2.2.2 Lipoprotein agarose electrophoresis 24 2.2.2.3 Polyacrylamide electrophoresis 24 2.2.3 Incorporation of radiolabeled l i p i d s 24 2.2.3.1 [ 3H]cholesterol 24 2.2.3.2 [ 3H]cholesteryl ester 25 2.3 Preparation of recombinant lipoproteins 25 2.3.1 Pu r i f i c a t i o n of apoproteins 25 2.3.2 Recombinant HDL preparation 26 2.3.3 Characterization of recombinant HDL 26 2.3.3.1 Gradient ultracentrifligation 26 2.3.3.2 Gel f i l t r a t i o n 27 2.3.3.3 Electron microscopy 27 2.3.3.4 Nuclear magnetic resonance spectroscopy..27 2.4 Metabolic assays 28 2.4.1 Cholesteryl ester transfer activity...........28 2.4.1.1 CETA using exogenous substrates 28 2.4.1.2 CETA in plasma 29 2.4.2 Lecithin-cholesterol acyltransferase activity.31 2.4.2.1 LCAT a c t i v i t y determination 31 2.4.2.2 LCAT with native and recombinant HDL 31 v .2.5 P u r i f i c a t i o n of LCAT and CETP 32 2.6 C l i n i c a l studies 32 2.6.1 CETA in severe HDL de f i c i e n t subjects 32 2.6.2 CETA in hyperlipidemic subjects 33 2.7 Incubations involving recombinant lipoproteins 34 2.7.1 Native and rHDL as acceptors of CE 34 2.7.2 Native and rHDL as substrates for LCAT 34 2.8 Mathematical analyses 35 - 2.8.1 Statistical-analyses 35 2.8.2 Kinetic analyses '. 35 3 RESULTS 3.1 Characterization of l i p i d transfer assays 36 3.1.1 Lipid transfer in lipoprotein depleted plasma.36 3.1.2 Lipid transfer in plasma 36 3.2 C l i n i c a l studies 38 3.2.1 Comparison of l i p i d transfer assays 38 3.2.2 CETP and patients with severe HDL deficiency..40 3.2.3 CETA and HDL composition in hyperlipidemia 42 3.2.3.1 Patient characterization 42 3.2.3.2 HDL composition 44 3.2.3.3 Cholesteryl ester transfer a c t i v i t y 45 3.2.3.4 S t a t i s t i c a l analyses. 47 3.3 P u r i f i c a t i o n of CETP and LCAT 48 3.4 Characterization of recombinant HDL 52 3.4.1 Recombinant HDL composition and size 52 vi 3.4.2 Electron microscopy 58 3.4.3 Nuclear magnetic resonance spectroscopy..58 3.5 Incubations involving native and recombinant HDI 58 3.5.1 Native and rHDL composition and size 58 3.5.2 Native and rHDL as substrates of CETP 68 3.5.3 Effect of rHDL l i p i d composition on CETA 72 3.5.4 Effect of rHDL apoprotein composition on CETA.76 3.5.5 Effect of rHDL l i p i d composition on LCAT 79 4 DISCUSSION 4.1 The determination of CETP a c t i v i t y 84 4.1.1 The measurement of CETP a c t i v i t y in plasma 84 4.1.2 CETP in patients with severe HDL deficiency.. .86 4.1.3 CETP and HDL composition in dyslipidemia 88 4.2 Use of recombinant HDL to study CETP and LCAT 93 4.2.1 Characterization of recombinant HDL 93 4.2.2 Use of recombinant HDL to study CETP 93 4.2.3 Effect of rHDL composition on CETP 94 4.2.3.1 Effect of rHDL l i p i d content 94 4.2.3.2 Effect of rHDL apoprotein content 97 4.2.4 Use of recombinant HDL to study LCAT 98 4.3 HDL composition, CETP and LCAT: concluding remarks.99 4.4 Proposal for future study 106 5 REFERENCES 108 6 APPENDIX A , 118 vii ABBREVIATIONS Apo apolipoprotein or apoprotein BSA bovine serum albumin CE cholesteryl ester CET cholesteryl ester transfer CETA cholesteryl ester transfer a c t i v i t y CETP cholesteryl ester transfer protein CHD coronary heart disease CM carboxymethyl cellulose DlFP di i sopropyl f 1 uorophosphate DYSB dysbetal ipoproteinemia DSS sodium 2,2-dimethyl silapentane 5-sulfonate EMTS ethyl mercurithiosalicylate EYPC egg yolk phosphatidylcholine FC free cholesterol FCH familial combined hyperl ipidemia FED.... f i s h eye disease FH fam i l i a l hypercholesterolemia FPLC fast protein l i q u i d chromatography FTG familial hypertriglyceridemia HA fami 1 i a l hypoal phal ipoproteinemi a HA+HTG hypoal phal i poprotei nemi a wi th hypertri glyceri demi a HDL high density lipoprotein HDL-TG high density lipoprotein t r i g l y c e r i d e viii HDL-CE high density lipoprotein cholesteryl ester HL hepatic lipase HSA human serum albumin HTP hydroxylapatite IDL intermediate density lipoprotein IEF i s o l e l e c t r i c focussing KH • HDL concentration at half maximal l i p i d transfer velocity LCAT lecithin:cholesterol acyltransferase LDL low density lipoprotein LDL-C low density lipoprotein cholesterol LTP-I l i p i d transfer protein-I or cholesterol ester transfer protein LTP-II l i p i d transfer protein-II or phospholipid transfer protein LPDP ..lipoprotein depleted plasma LPL lipoprotein lipase Mr molecular weight NMR nuclear magnetic resonance PAGE ; polyacrylamide gel electrophoresis pCMPS p-choromercuri phenyl sul fonate pHMB..' p-hydroxymercuribenzonate PL phosphol ipid QLS ;.. .quasielastic laser l i g h t scattering RCT reverse cholesterol transport rHDL recombinant HDL SDS sodium dodecylsulphate ix TCA t r i c h l o r o a c e t i c acid TD Tangier disease TG t r i g l y c e r i d e TLC thin layer chromatography T m a x maximum velocity of l i p i d transfer V velocity VHDL very high density lipoprotein VLDL very low density lipoprotein V m a x : maximum velocity of reaction x LIST OF TABLES Table 1. Plasma l i p i d and lipoprotein composition in severe HDL-deficiency 42 Table 2. Cholesteryl ester transfer a c t i v i t i e s in plasma 42 Table 3. Plasma l i p i d s and apoproteins 43 Table 4. HDL l i p i d composition 44 Table 5. Relative HDL-lipid composition 45 Table 6. Cholesteryl ester transfer from solid-phase-bound-HDL to plasma lipoproteins 47 Table 7. HDL composition and cholesteryl ester transfer a c t i v i t y in patients with vascular disease 48 Table 8. CETP p u r i f i c a t i o n table... 49 Table 9. LCAT p u r i f i c a t i o n table '. 52 Table 10. I n i t i a l and f i n a l l i p i d and protein composition of recombinant spherical and discoidal HDL 53 Table 11. Native HDL composition.. 62 Table 12. I n i t i a l l i p i d and apoprotein composition of recombinant HDL . 67 Table 13. Final l i p i d and apoprotein composition and size of recombinant HDL 67 Table 14. Kinetic parameters of native HDL as a substrate for CETP 72 Table 15. Effect of l i p i d s on the kinetic parameters of rHDL as a substrate for CETP 76 Table 16. Effect of apoproteins on the kinetic parameters of rHDL as a substrate for CETP 79 Table 17. Effect of rHDL size and l i p i d content on the k i n e t i c parameters of rHDL as a substrate for LCAT 81 xi LIST OF FIGURES Figure 1. Reverse cholesterol transport 4 Figure 2. Postulated mechanisms of action of CETP ...10 Figure 3. Solid-phase assay of cholesteryl ester transfer a c t i v i t y in plasma 30 Figure 4. Cholesteryl ester transfer to plasma and lipoprotein ' depleted plasma of normal and hyperlipidemic subjects 39 Figure 5. Cholesteryl ester transfer to plasma and lipoprotein depleted plasma of normolipidemic and HDL defi c i e n t subjects \ 41 Figure 6. Cholesteryl ester transfer in normal and dyslipidemic subjects 46 Figure 7. SDS polyacrylamide gel electrophoresis of partly p u r i f i e d CETP 50 Figure 8. SDS polyacrylamide gel electrophoresis of puri f i e d LCAT 51 Figure 9. Size exclusion chromatography of recombinant HDL before ultracentrifugal isolation 54 Figure 10. Size exclusion chromatography of native and recombinant HDL after ultracentrifugal isolation 55 Figure 11. Agarose lipoprotein electrophoresis 56 Figure 12. Equilibrium density gradient ultracentrifugation of rHDL ....57 Figure 13. Negative staining electron microscopy of HDL and rHDL 59 Figure 14. 1H-NMR spectroscopy of HDL and rHDL 60 Figure 15. 1H-NMR of the N-methyl choline region of vesicles, HDL, and rHDL 61 Figure 16. I s o l e l e c t r i c focusing of native HDL2, HDL3, and VHDL apoproteins 63 Figure 17. I s o l e l e c t r i c focusing of recombinant HDL apoproteins 64 xii Figure 18. Negative stain electron microscopy of rHDL with variations in core neutral l i p i d s 65 Figure 19. Negative stain electron microscopy of rHDL with variations in l i p i d s and apoproteins 66 Figure 20. rHDL and HDL as substrates for CETP 69 Figure 21. Cholesteryl ester transfer to HDL subfractions in the absence of exogenously added CETP 70 Figure 22. Cholesteryl ester transfer to HDL subfractions 71 Figure 23. Effect of rHDL neutral l i p i d composition on cholesteryl ester transfer 73 Figure 24. Effect of rHDL polar l i p i d composition on cholesteryl ester transfer 74 Figure 25. Effect of rHDL l i p i d composition on CETP reaction kinetics..75 Figure 26. Comparison of apoprotein A-I and A-II containing rHDL as substrates for CETP 77 Figure 27. Effect of rHDL apoprotein content on cholesteryl ester transfer to rHDL 78 Figure 28. rHDL and HDL as substrates for LCAT 80 Figure 29. Effect of rHDL l i p i d composition on LCAT a c t i v i t y 82 Figure 30. The relationship between rHDL composition, size, and LCAT reaction kinetics 83 Figure 31. Reverse cholesterol transport in hyperlipidemia. 105 xiii LIST OF FIGURES IN APPENDIX A Figure 1. Characterization of the transfer of cholesteryl ester from LDL to HDL in lipoprotein depleted plasma 119 Figure 2. Effect of substrate concentration on cholesteryl ester transfer from LDL by depleted plasma 120 Figure 3. Effect of partly pure CETP on cholesteryl ester transfer from LDL 121 Figure 4. Effect of incubation time and plasma concentration on cholesteryl ester transfer from solid-phase-bound HDL to plasma lipoproteins 122 Figure 5. The e f f e c t of the d>l.21 g/mL fraction on cholesteryl ester transfer from solid-phase-bound HDL 123 Figure 6. Transfer of cholesteryl ester to plasma and recombined 1 ipoproteins 124 Figure 7. Effect of partly pure CETP and solid-phase-bound HDL concentration on cholesteryl ester transfer 125 Figure 8. Effect of LCAT inactivation on transfer a c t i v i t y . 126 Figure 9. Effect of plasma storage at 4°C on cholesteryl ester transfer from s o l i d phase bound HDL 127 Figure 10. Effect of incubation temperature on cholesteryl ester transfer from solid-phase-bound HDL .128 Figure 11. Localization of CETP and LCAT in the d>1.21 g/mL frac t i o n 129 Figure 12. Phenyl-Sepharose column chromatography elution p r o f i l e s of CETP and LCAT a c t i v i t i e s 130 Figure 13. CM-cellulose column chromatography elution p r o f i l e s of CETP and LCAT a c t i v i t i e s 131 Figure 14. Kinetic analysis of l i p i d transfer to HDL subfractions 132 Figure 15. Kinetic analysis of l i p i d transfer to rHDL with various l i p i d compositions 133 Figure 16. Kinetic analysis of l i p i d transfer to rHDL with various apoprotein compositions 134 xiv ACKNOWLEDGEMENTS I would l i k e to express my gratitude to everyone who has contributed to the successful completion of these studies. In p a r t i c u l a r , I would l i k e to take this opportunity to express my sincerest thanks to my supervisor, Dr. Haydn Pritchard. Throughout these investigations, Haydn has been a d i l i g e n t and consistent contributor; providing invaluable support, a constant opportunity for discussion and a clear mind for innovation. In fact, Haydn's assistance in almost every aspect of my post graduate s c i e n t i f i c education has been well above and beyond the c a l l of duty. This is perhaps a small indication of the dedication of Dr. Pritchard to the pursuit of excellence in the f i e l d of medical research. Secondly, I would l i k e to thank the members of my supervisory committee for t h e i r valuable recommendations and in particular, Dr. J i r i Frohlich, for his support and guidance in the c l i n i c a l component of t h i s study. In addition, I would l i k e to express a special thanks to members of our laboratory who have provided both technical and i n t e l l e c t u a l contributions throughout these investigations. They are Roger McLeod, Cathy Mcguiness, Lida Adler and Dr. Yashpal Parmar. A debt of gratitude is also owed to Dr. Andras Lacko for his ongoing enthusiasm and helpful discussions. F i n a l l y , I would l i k e to express my gratitude to the B r i t i s h Columbia Heart Association for providing financial support during this period. xv DEDICATION I dedicate this work to my dear wife EWA and our son JUSTIN. J 1 INTRODUCTION 1.1 ATHEROSCLEROSIS, LIPIDS AND HDL Atherosclerosis was once thought to be simply the result of the aging process; however, it has now been characterized as a multifactorial disease, with many etiologic factors involved. While direct evidence is lacking which correlates atherosclerotic risk factors and the bio-molecular events involved, much information has been reported which links altered plasma lipid metabolism to the development of atherosclerotic plaques (1). Early morphological studies of atherosclerotic plaques identified a fibrous, hypercellular lesion that was, most notably, loaded with cholesterol (2). Subsequent epidemiological studies have since shown a strong correlation between the occurrence of these lesions and chronically elevated levels of plasma cholesterol (reviewed by 3). More recent studies have shown that hyperlipidemia may be a sufficient cause for atherosclerosis, via a mechanism that may involve aggravation of endothelial injuries (4); however, the precise sequence of events which promote and maintain hyperlipidemia still remain unclear. Ischemic heart disease, resulting from abnormal lipid metabolism, is one of the major causes of premature mortalities in developed nations (5). Eradication of this major disease will require a thorough understanding of all the factors involved in the regulation of plasma lipid metabolism. The level of cholesterol in the plasma high density lipoprotein (HDL) has been shown to be inversely related to incidence of ischemic heart disease (reviewed in 6). Consequently, increased plasma levels of this lipoprotein are believed to offer "protection" against vascular lipid accumulation. While the initial observations of this protective effect were based on measurement of the cholesterol content of HDL, it has become evident that the complete lipid/apoprotein composition of this lipoprotein plays an important role in directing its synthesis, modification and subsequent catabolism (7). This study was designed to investigate the role of HDL composition on the plasma proteins which regulate the metabolism of cholesterol in the plasma compartment. 1 1.2 L I P O P R O T E I N S 1.2.1 Lipoprotein structure and metabolism Cholesterol is transported i n the plasma in discrete complexes of l i p i d and protein called lipoproteins. Lipoproteins are multimolecular complexes composed of a series of different proteins, or apoproteins and four major classes of lipids: unesterified cholesterol, esterified cholesterol, t r i g l y c e r i d e , and phospholipid. Structural studies have shown that the neutral l i p i d components, esterified cholesterol and triglyceride, make up the core of a particle that is enveloped i n a phospholipid and unesterified cholesterol monolayer (8). The structural integrity of these particles is maintained by apoproteins which interdigitate w i t h the phospholipid monolayer and maintain the micellular conformation (8,9). Lipoproteins have been traditionally classified into f i v e major categories on the basis of their flotation densities: chylomicrons, very low density lipoproteins ( V L D L ) , low density lipoproteins ( L D L ) , high density lipoproteins (HDL), and very high density lipoproteins (VHDL). In general, the average size of each class of lipoprotein is inversely proportional to its hydrated density, with the least dense class, chylomicrons, being the largest i n size (10). Each class represents a continuum of related species which encompass a range of size and densities but which have in common a similar apoprotein composition. The major apoprotein of the small dense lipoproteins V H D L and H D L is apoprotein A-I and of V L D L and L D L is apoprotein B (10). While L D L primarily has only the one apoprotein, a l l the other lipoprotein classes have mixtures of several other proteins, designated as apoproteins A, C, D, E, F, and H (1,7,10). Lipoprotein cholesterol metabolism may be simplistically viewed as three separate components: 1) hepatic delivery of exogenous cholesterol, 2) extrahepatic cholesterol delivery and 3) reverse cholesterol transport. Dietary cholesterol is delivered to the liver i n the form of chylomicron remnant particles. Chylomicrons are produced by the intestinal mucosal cells from dietary lipids and secreted into the mesenteric lymph ducts, from which they enter the plasma (10,11). In the plasma vascular bed, the triglyceride core of 2 the chylomicron is rapidly hydrolyzed by the endothelial bound enzyme lipoprotein lipase, resulting in the formation of a remnant particle which is rapidly removed from the plasma by specific hepatic receptor-mediated uptake (11). The liver either disposes of the cholesterol by converting it to bile salts which are excreted into the intestines, or diverts it back into the plasma in the form of newly secreted lipoproteins. LDL is thought to be primarily responsible for the delivery of cholesterol to extrahepatic tissues (11). The liver is capable of either directly producing LDL or indirectly producing it via VLDL. VLDL is produced by the liver and is catabolized much like chylomicrons by lipoprotein lipase to an intermediate density particle which can be either directly removed by the liver or converted into LDL (10-12). LDL is a cholesteryl ester rich particle which delivers cholesterol to hepatic and extrahepatic cells by binding to specific LDL receptors on cell surfaces and subsequent receptor endocytosis (11). While some recent studies have suggested that HDL may also be capable of delivering cholesterol to extrahepatic tissues through receptor or non-receptor mechanisms, the overall importance of this pathway remains unclear (13). HDL can be either be produced directly by the liver and intestine or may originate as a degradation product of chylomicron catabolism (7). Apoprotein A-I synthesis and lipolysis of triglyceride-rich lipoproteins provide the proteins and surface remnant phospholipids (PL) and free cholesterol (FC) which combine to form the HDL precursor pool of nascent or discoidal particles. Mature, spherical HDL are thought to then result from the generation of cholesteryl esters by the enzyme lecithinxholesterol acyltransferase (LCAT)(7). 1.2.2 Reverse cholesterol transport (RCT) The mechanism behind the anti-atherogenic nature of HDL is thought to be related to its role in the removal of peripheral tissue cholesterol and transport to the liver through a process known as reverse cholesterol transport (14). This phenomenon, originally described by Glomset (15), is composed of four distinct elements: efflux of free cholesterol, esterification of cholesterol in HDL, transfer to other lipoproteins and transport to the liver (Figure 1). 3 PERIPHERAL CELLS 1. Cholesterol efflux 2. Esterification by LCAT 3. Transfer by CETP 4. Hepatic uptake Figure 1. Reverse cholesterol transport. HDL is thought to receive unesterified cholesterol from cell membranes either by aqueous diffusion, transfer during transient collision contacts, or by transfer within stable complexes formed as a result of affinity binding of HDL to specific sites on the surface of peripheral cells (14). Subsequent esterification of HDL cholesterol by LCAT is thought to maintain the cholesterol concentration gradient between the cell membrane and HDL (15). Some studies have shown that cholesterol efflux from cell membranes to HDL may be promoted by the action of L C A T (15,16). Transport of esterified cholesterol to the liver may occur by direct uptake of the entire HDL particle (13,14) or by selective uptake of cholesteryl ester (13). Alternatively, cholesteryl ester may be transferred to apoB-containing lipoproteins and then be cleared by hepatic uptake of IDL and LDL (14). This 4 has been popularly accepted as the major pathway by which cholesteryl esters are transported to the liver for three reasons. CETP has been shown to promote a net mass transfer of cholesteryl esters only to VLDL in vitro. This suggests that the net movement of the majority of cholesteryl esters, formed in HDL is to VLDL and thus to LDL (17). Since the turnover of LDL and V L D L apo B has been shown to be faster than HDL apo A-I, this suggests that hepatic uptake of lower density lipoprotein cholesteryl esters may be faster than HDL cholesteryl esters (18). Finally, accumulation of cholesteryl esters in HDL has been thought to inhibit the action of L C A T (19). Transfer of esters to low density lipoproteins and their subsequent hepatic uptake may therefore help to maintain the cholesterol concentration gradient (14). If cholesteryl esters are quantitatively transferred to apo B containing particles and these particles are removed more rapidly than higher density lipoproteins, this may be the major route by which cholesterol is removed from the plasma compartment. Recent reports, however, have raised major questions over the validity of this hypothesis. Studies by Glass et al have shown that the hepatic uptake of HDL cholesteryl esters may be much more rapid than its apoproteins (13). This work has prompted speculation that HDL may play an important role in the direct transport of cholesteryl ester to the liver (13,14,18). 1.2.3 HDL structure and heterogeneity. High density lipoproteins (HDL) are a heterogeneous group of particles of 7-10 nm in diameter with a molecular weight of 200-400 kDa (7). They are composed of approximately 50% by weight of protein (mainly apo A-I and apo A-II in human HDL) and 50% lipid (mainly phospholipid, free and esterified cholesterol and a small amount of triglyceride). Since HDL particles are distinctly different from other lipoproteins in terms of size, weight and composition, HDL has historically been isolated by ultracentrifugation and characterized on the basis of its hydrated density of 1.063-1.21 g/mL. While two major discrete HDL subclasses can be isolated at densities of 1.063-1.125 g/mL and 1.125-1.21 g/mL and called H D L 2 and H D L 3 , respectively, additional studies have identified several other discrete HDL populations which differ in both hydrated density and 5 particle size (7). H D L subclasses may also be characterized by their apoprotein content. Cheung and Albers (20) have employed immunoaffinity chromatography to isolate two populations of apo A-I containing lipoproteins from the plasma of normal individuals without ultracentrifugation. One subclass ( L p A ) contained apo A - I and A - I I and was composed of 3 distinct molecular sizes whereas the other subclass (LpA-I) contained apoA-I but no A-II. The latter group consisted of 2 distinct molecular sizes. Different proportions of apoproteins C, D and E and L C A T were detected i n both L p A - I subclasses isolated. H D L lipoproteins have also been isolated by Norfeldt et al (21) using a combination of ion-exchange, adsorption and immunoabsorption chromatography. Other workers have turned their attention to H D L particles containing apo E. They have used the binding properties of heparin towards apo E to selectively remove H D L particles containing this apoprotein (22). The isolation of this lipoprotein directly from plasma by this method, however, is complicated by the concomitant binding and subsequent elution of apo B containing lipoproteins. Separation of H D L subclasses on the basis of net charge has also indicated H D L subspeciation. Jahani and Lacko (23) used this method with whole plasma to isolate a number of H D L subclasses that differed i n their ability to interact with purified L C A T . Marcel et al (24) have also described up to 10 subspecies of H D L that may be separated by isoelectric focusing on agarose films. Gradient polyacrylamide gel electrophoresis has become a simple and yet valuable method of clearly identifying 5 different H D L subclasses, two associated with H D L 2 and three associated with H D L 3 (25). Clearly, the H D L pool is not structurally homogeneous. Stable subclasses of H D L exist which d i f f e r i n size, charge, l i p i d and protein composition and i n their hydrated density. Despite these increases i n our understanding of HDL, we have not yet been able to clearly identify any metabolic heterogeneity that this may represent. 1.2.4 Lecithin-.cholesterol acyltransferase L C A T is the enzyme responsible for the production of the majority of 6 cholesteryl esters in plasma via transfer and esterification of sn-2 fatty acids from phosphatidylcholine to the 3-hydroxyl group of cholesterol (26). This activity requires the presence of specific apoprotein cofactors in vitro. Several investigators have established that apo A-I is the principal activator of cholesterol esterification by L C A T under in vitro conditions (27-30). Furthermore, when apo A-I containing lipoproteins are removed from plasma by immunoaffinity chromatography, much of the cholesterol esterifying activity is also removed (31,32). This implies a physical association between apo A-I and L C A T and reflects their functional interdependence. In certain pathologic states and with some in vitro manipulations LCAT can be activated by other apolipoproteins. Apo A-IV, apo C-I, apo D and Apo E have been shown to activate LCAT in the absence of A-I (28,29,33-35). In the presence of suboptimal levels of A-I, A-II can also activate the enzyme despite being inhibitory at optimal A -I concentrations (36,37). The observation that other apoproteins can activate LCAT is of particular interest since several HDL deficiency states have been described where cholesterol esterification proceeds normally, or at slightly reduced rates, despite extremely low levels of apo A-I (38). L C A T has a dual catalytic function, acting both as a phospholipase A 2 and as an acyltransferase (18,26-37). The mechanism of action of L C A T has been recently elucidated in the laboratory of P. Dolphin and coworkers. They have proposed the fundamental chemical mechanism of L C A T to result in a series of transesterification reactions involving paired cysteine groups and close proximity serine and histidine groups (39). They have shown the reaction to be initiated by single serine and histidine groups which mediate the cleavage of the sn-2 ester bond of phosphatidylcholine and promote the formation of the serine oxyester. Subsequent transesterification of the fatty acyl group to either of the 2 cysteine residues results in the formation of a thioester and allows for the further transesterification of the acyl group to the 3-hydroxy group of cholesterol (39). Detailed characterization of the interaction between HDL and L C A T has been limited by the variable and uncontrollable lipid compositions of native 7 substrates. While many investigators have used discoidal or vesicular apoprotein/lipid mixtures to probe the biochemical interactions involved (37,39-41), little work has been carried out with protein/lipid mixtures which are structurally comparable to native mature HDL (42). Some studies with native lipoproteins, however, have indicated that the relative neutral lipid composition of HDL may be of particular importance in the interaction between HDL and LCAT (43-46). Fielding et al have suggested that cholesteryl ester may be a feedback inhibitor of LCAT as their studies showed that accumulation of this lipid in HDL resulted in the inhibition of LCAT activity (19). Other investigators have also suggested that HDL particle size may directly affect its ability to interact with L C A T (44-46). 1.2.5 Neutral lipid transfer reactions 1.2.5.1 Introduction In the late sixties and early seventies several investigators identified a human plasma activity capable of promoting a reciprocal exchange of cholesteryl ester in HDL for triglyceride from VLDL (47-49). This activity was shown to be independent of plasma L C A T activity, yet its action resulted in the accumulation of products of the L C A T reaction, cholesteryl esters, in VLDL and LDL (47,49). This has prompted speculation by some investigators (14,15,50) that this protein, known as CETP or lipid transfer protein (LTP-I), may play an integral role in reverse cholesterol transport (section 1.2.2), a process that may prevent intravascular cholesterol accumulation and retard atherosclerosis. 1.2.5.2 Biochemical characterization Early purification attempts by several investigators initially identified an activity with an apparent molecular weight of 70-80 kDa (51-53) and another at 30-35 kDa (54). Subsequent studies have shown the 30 kDa protein to be a common contaminant of CETP purification, apoprotein D, a minor protein of HDL with no antigenic similarity to CETP and no lipid transfer activity of its own (55). Partially purified preparations of CETP have been also been shown to be contaminated with a 41 kDa protein, known as LTP-II, that preferentially transfers phospholipids between lipoproteins (53,56). CETP has been shown to 8 have a M r of 74,000 Da on SDS-polyacrylamide gel electrophoresis after purification of 55,000 (57) or 100,000 (58) fold. Drayna et al have recently produced a CETP complementary DNA clone and have also sequenced the protein (59). They have reported the mature peptide to be made up of 476 amino acids with a translated M f of 53,108 Da. Their studies also showed the protein to contain four potential asparagine-linked glycosylation sites and have suggested that the observed M f of 74,000 Da may be a result of post-translational glycosylation. This group and others (57) have shown this protein to be more hydrophobic than other known apoproteins or lipoprotein metabolizing enzymes, with a high content of nonpolar residues (45%) and an overall index of hydropathy of +0.10 (59). CETP has been shown to have an isoelectric point (pi) within the pH range of 4.8-5.2 (52,53) and to be relatively heat stable, with no activity loss after incubation at 56°C for up to 2 h (60). CETP has been shown to be secreted by macrophages (61), human CaCo-2 enterocytes (62) and by HepG2 cells (63). Northern blot hybridization with the previously described CETP cDNA clone has further identified CETP mRNA in human liver, small intestine and adrenal gland tissues and also in isolated hepatocytes (59). 1.2.5.3 Mechanism of action CETP has been shown to have specific lipid binding sites for cholesteryl ester, triglyceride and phosphatidylcholine (64). Accordingly, CETP has been demonstrated to promote the transfer of all three lipids among all classes of plasma lipoproteins (56,60). Two different mechanisms have been proposed to describe the lipid transfer process (Figure 2). One concept suggests that CETP acts as a "reusable shuttle" (Figure 2A), a carrier protein which transports lipids between donor and acceptor lipoproteins (49,64). Recently CETP has been shown to have specific lipid binding sites (64), which may support this hypothesis; therefore, CETP has been compared to other well characterized lipid carrier proteins, bovine liver phosphatidylcholine transfer protein and cyotsolic nonspecific lipid transfer protein (64). Conversely, there is also evidence which refutes this hypothesis. In particular, most of the CETP in plasma has been shown to be bound to HDL (65), with very 9 CE t CE J Figure 2. Postulated mechanisms of action of CETP. A. CETP acts as a carrier (ie. reusable shuttle) of neutral lipids between lipoproteins. B. CETP promotes the formation of a ternary complex, in which neutral lipid transfer occurs. little existing in a free, unbound state (66). In addition, kinetic studies do not support the carrier protein concept but suggest that lipid transfer results from the formation of a ternary collision complex of CETP, donor and acceptor lipoproteins (67). One observation, reported by Morton and Zilversmit (68), which suggested that neutral lipids were transferred through a complex-mediated lipid exchange (Figure 2B), was that highly purified CETP is capable of promoting a reciprocal exchange, ie hetero-exchange, of cholesteryl ester for triglyceride between some classes of lipoproteins (68). This concept of a stable collision complex is further substantiated by earlier work which demonstrated that the binding of CETP to the lipoprotein surface was an integral component of subsequent lipid transfer (70). Several studies have now shown 10 that the binding of CETP to lipoproteins is charge dependent (69-71) and may be primarily regulated by an electrostatic interaction with surface phospholipid phosphate groups (70). This conclusion was derived from studies by Pattnaik and Zilversmit (70) which demonstrated that modification of the surface charge of lipoproteins by a change in pH, amino acid acylation or phospholipase A 2 treatment, resulting in an increase in the net negative charge, effectively increased the stability of CETP-lipoprotein complexes. Conversely, since removal of the choline phosphate groups by phospholipase C treatment effectively abolished the binding of CETP to HDL, this suggests a role of the phosphate head groups in complex formation (70). In similar experiments, Sammet and Tall (71) further demonstrated that phospholipase A 2 , bacterial lipase and lipoprotein lipase treatment all augmented cholesteryl ester transfer protein activity. Since they further showed that addition of sodium oleate also enhanced transfer activity and that reducing the pH abolished any enhancement, they concluded that it was probably the accumulation of fatty acid products of lipolysis which modified lipoprotein surface charge, augmented the binding of CETP to these lipoproteins and enhanced lipid transfer. Increased lipase activity and subsequent fatty acid release may in fact explain the apparent increase in CETP activity noted during alimentary lipemia (72,73). Binding studies have shown that CETP binds all lipoproteins with high affinity but binds HDL with the highest avidity (69). Other studies have recently demonstrated that in plasma most of the CETP is bound to a subfraction of HDL devoid of apoprotein A-II (65), and that only a very small amount exists as free CETP (66). Taken together, these results seem to unanimously discount the possibility that this extremely hydrophobic protein, with the high affinity it has for all lipoproteins, could transfer lipid as a dissociated shuttle. Inhibition of CETP activity in vitro can be accomplished by mercurial compounds (68). While several mercurials; p-chloromercuriphenylsulfonate (pCMPS), p-hydroxymercuribenzoate (pHMB) and ethyl mercurithiosalicylate (EMTS), have been shown to inhibit triglyceride transfer activities of CETP (52), only pCMPS has been shown to be capable of completely inhibiting both 11 triglyceride and cholesteryl ester transfer activities (68). Conversely, EMTS, 5,5-dithiobis-(2-nitrobenzoate) and diethyl-p-nitrophenyl phosphate were shown to enhance the transfer of cholesteryl esters (52). Work by other investigators has shown that apo A-I (74) and fatty acid free albumin (71) also have the capacity to inhibit CETP mediated cholesteryl ester transfer. Furthermore, a specific inhibitor of cholesteryl ester transfer protein has been isolated from human plasma (74,75,76). This protein, M f 29-32,000 Da, has recently been isolated from apo A-II rich subclasses of HDL and has been shown to inhibit cholesteryl ester, triglyceride and phospholipid transfer activities (76). The mechanisms by which many of these inhibitors act is still unclear; however, both fatty acid free albumin (71) and human CETP inhibitor protein (69) have been shown to decrease the binding of CETP to the lipoprotein surface. 1.2.5.4 The role of CETP in lipoprotein metabolism The major function of CETP in the plasma compartment is thought to be the redistribution of LCAT-derived cholesteryl esters from their sites of synthesis in HDL to the less dense triglyceride-rich lipoproteins . (49,50,77-79). This concept was originally developed from early studies which identified the accumulation of LCAT-derived cholesteryl esters in LDL and V L D L (47,78). Subsequent studies showed that CETP was capable of promoting the net transfer and hetero-exchange of triglyceride and cholesteryl ester between V L D L and LDL or VLDL and HDL (68,78,79). These studies showed that the net transfer of cholesteryl esters from HDL occurred primarily into particles with the lowest cholesteryl ester/triglyceride ratios (nascent VLDL and chylomicrons), whereas transfer to particles with higher ratios (LDL) predominantly involved simple exchange of cholesteryl esters. Thus, CETP was postulated to play a central role in the equilibration of cholesteryl esters into particles that were thought to have a faster turnover rate than HDL (18), but at the same time, were also considered to be atherogenic (4,12). Some investigators have gone on to speculate that if CETP promotes the movement of cholesteryl ester into particles that are actively removed by the liver, perhaps it plays a central role in the removal of plasma cholesterol via the previously described pathway called 12 reverse cholesterol transport (14,18,72). The factors which regulate this process are, however, still quite unclear. In fact, the whole process seems to promote the accumulation of cholesteryl esters in particles which contribute to atherogenesis rather than allow for their removal. The activity of CETP also results in the speciation of lipoproteins and their remodeling into smaller particles (72,80,81). Tall (72) suggests that CETP may collaborate with lipoprotein lipase in the remodeling of both LDL and HDL. In plasma, CETP has been shown to be capable of replacing the predominantly cholesteryl ester core of LDL and HDL with triglyceride (72). In vitro studies have shown that both lipoprotein lipase (72) and hepatic lipase (82,83) are capable of acting upon HDL and LDL triglycerides. Thus, the combined action of triglyceride lipases and CETP may result in the progressive depletion of core lipids and the formation of smaller LDL and HDL particles. Probably the major remodeling role for CETP is in the neutral lipid core of HDL, which is continuously being increased due to the action of L C A T (72). Both CETP and LCAT have been shown to be primarily associated with a specific subfraction of HDL in plasma (65). Accordingly, some investigators have suggested that CETP and L C A T may actually form a complex, with HDL, which readily accepts unesterified cholesterol and subsequently facilitates the production and redistribution of cholesteryl esters (31). In vitro studies of L C A T have shown that accumulation of the cholesteryl esters within the substrate lipoproteins may result in a feedback inhibition of the enzyme (19). As :such, the ability of CETP to redistribute cholesteryl esters may regulate L C A T activity by promoting the removal of these inhibitory products (50). Some studies have suggested that HDL may be a preferred substrate of CETP (69). While several studies have identified a major role of HDL as a donor of cholesteryl esters (67,69,72), very little has been reported characterizing the ability of HDL to receive these lipids. Consequently, it is still unclear whether CETP is capable of promoting the equilibration of cholesteryl esters within the HDL pool prior to transfer to less dense lipoproteins. Some studies have suggested that HDL may be capable of delivering cholesteryl esters directly 13 to the liver and other tissues by some process other than direct uptake of the intact HDL particle (13). Also, CETP has been shown to promote the efflux of cholesteryl esters from several different cell types in tissue culture experiments (84). These studies suggest that HDL may play a role in the delivery as well as removal of cholesteryl esters from peripheral tissues. This concept may have particular importance in reverse cholesterol transport when one considers that HDL may be readily capable of crossing the endothelial barrier (85). This may permit HDL to transport cholesteryl esters into the extravascular compartments. Conversely, previous studies have shown that LDL and V L D L may have more restricted movement across the endothelial barrier and, therefore, may be more localized in the vascular compartment (85). This suggests that if elevated cholesteryl ester levels are maintained in HDL, this may may act as a sink for cholesteryl ester and allow for their equilibration in vascular and extravascular compartments. However, if CETP promotes a net transport to LDL and VLDL in vivo, this may reduce equilibration with extravascular pools and may in fact accentuate hypercholesterolemia. 1.2.5.5 The role of CETP in hyperlipidemia. The capacity of CETP to promote the transfer and possibly accumulation of cholesteryl esters in L D L , as well as some recent studies which have identified abnormal cholesteryl ester transfer rates in hyperlipidemic patients, have resulted in some suggestions that CETP might play an important role in atherogenesis (18,72,86). Interestingly, several species which have been shown to have low plasma CETP activities, such as the dog, rat, pig, cow and sheep (87) , have also been shown to have low plasma LDL esterified cholesterol levels (88) and also tend to be resistant to the development of atherosclerosis (18). This concept has been further supported by studies in hypercholesterolemic rabbits which identified increased CETP mediated cholesteryl ester transfer from HDL to apo B-containing particles and subsequent accumulation of these cholesteryl esters in cultured macrophages (66). In the same report, Tall et al also demonstrated accelerated CETP mediated cholesteryl ester transfer from HDL to LDL and VLDL in dysbetalipoproteinemic patients (66). 14 These results have prompted Tall to speculate on a potential mechanism by which CETP may promote atherogenesis (72). He suggests that subjects with high HDL cholesterol levels may have concomitantly low cholesteryl ester transfer from HDL to apo B containing lipoproteins and instead have another mechanism for disposing of HDL cholesteryl esters in the liver or other tissues. Individuals with low HDL cholesterol, on the other hand, may have elevated rates of cholesteryl ester transfer to apo B containing lipoproteins. If the levels of apo B containing particles are already high due to some other enzymatic or receptor defect, cholesteryl esters may instead accumulate in these particles and be diverted from hepatic removal into tissue deposition. In this scenario, abnormal CETP activity would be secondary to abnormal HDL cholesterol levels and elevated apo B particle levels. The author suggests however, that in some individuals, reduced levels of HDL cholesterol may in fact be caused by abnormal CETP activity, perhaps as a result of altered binding to the lipoproteins. Some very recent reports have shown that CETP mediated cholesteryl ester transfer to HDL is inhibited when the free cholesterol content of HDL is increased (89). Impaired transfer to HDL was shown to be concomitant with increased net transport of cholesteryl esters to lower density lipoproteins. These results have prompted Morton to suggest that cholesteryl ester transfer activity may be governed by the free cholesterol/phospholipid ratios in donor and acceptor lipoproteins (89). This concept is supported in one study reported by Fielding et al (90) where patients with non-insulin-dependent diabetes mellitus were shown to have significantly reduced transfer of cholesteryl esters to apo B particles and concomitantly increased free cholesterol to phospholipid ratios in these particles. Since this disease is commonly associated with elevated apo B containing lipoproteins, reduced HDL cholesterol, and increased risk for coronary artery disease, reduced cholesteryl ester transfer to the apo B particles seems contrary to Tail's atherogenesis hypothesis. In fact, subsequent work by Fielding's group has suggested that other disease states such as dysbetalipoproteinemia, familial hypercholesterolemia and hypertriglyceridemia with atherosclerosis, may have absent or reversed 15 cholesteryl ester transfer between HDL and apo B,E containing lipoproteins (86). To date there is no explanation for this discrepancy between the observations of these two investigators. Tall (66), however, suggests that accelerated transfer of cholesteryl esters to the d< 1.006 g/mL lipoproteins is the most probable source of the esters which have been shown to accumulate in this lipoprotein fraction in patients with dysbetalipoproteinemia. Several studies have shown markedly altered HDL compositions in individuals at increased risk to atherosclerosis (82,91-93). Tall (72) has suggested that HDL composition may in fact be regulated by the action of CETP. Reports by Tall have shown that CETP may actually promote the transfer of triglycerides from chylomicrons into HDL, in instances when lipase activity is incapable of maintaining normotriglyceridemia (72). Interestingly, Little et al (92) have shown HDL triglyceride levels to be independently and significantly correlated with the incidence of cardiovascular disease in some patient groups. Since CETP has been shown to play a central role in remodeling the neutral lipid core of HDL, this suggests that normal HDL composition may be integral to appropriate cholesteryl ester equilibration and disposal. In addition, reports of a cholesteryl ester transfer inhibitor protein, found to be associated with apo A -II rich subclasses of HDL, may suggest that the interaction of HDL and CETP may have other regulatory agents involved (76). One report has recently identified a patient with homozygous familial hyperalphalipoproteinemia which presents with an HDL that displays impaired reactivity with CETP relative to normal HDL (94). Furthermore, another report has demonstrated marked cholesteryl ester and triglyceride transfer inhibitory activity in the d>1.21 g/mL plasma fraction of patients with abetalipoproteinemia (95). These results suggest that the incidence of cardiovascular disease could be related to an altered CETP action which results in an abnormal HDL composition and defective reverse cholesterol transport. 16 1.3 RECOMBINANT LIPOPROTEIN SYSTEMS I. 3.1 Historical perspectives Reconstitution of HDL-like complexes has been accomplished using a wide variety of methodologies. Discoidal rHDL can be very easily produced by the spontaneous interaction of apoproteins with lipid vesicles. Apoproteins A-I, A -II, E and C have been shown to spontaneously disrupt lipid vesicles and form micellar, discoidal complexes of apoprotein and phospholipid (reviewed in 96 and 97). The rates and extent of formation are, however, highly variable and dependent upon a number of factors such as apoprotein type, phospholipid type, temperature and vesicle size (97). These kinetic limitations to the spontaneous interaction of vesicles and apoproteins have been partly overcome by the inclusion of detergents such as sodium cholate or deoxycholate (97). While detergent-dialysis methods markedly increase yields of discoidal complexes, they too are dependent on the temperature and type of phospholipid used and have also been shown to be limited in their capability of forming complexes when additional lipids, such as cholesterol, are included (97). This limitation is of particular importance in the production of spherical HDL complexes (which contain a neutral lipid core) from three or more different lipid classes and no studies to date have reported detailed characterization of the production of spherical HDL by detergent-dialysis. Another possible limitation of methods which employ detergent is the potential biological effects of the residual detergent remaining in complexes, even after extensive dialysis (97). This may be a particularly important concern if complexes are to be used for biochemical assays. In the early seventies, studies by Hirz and Scanu (98) showed that spherical rHDL could be produced from repurified HDL lipids and apoproteins by sonication. Subsequently, Ritter and Scanu (99,100) went on to further characterize the conditions and complexes formed during reconstitution of HDL-like particles from individual purified apoproteins A-I and A-II. They showed that rHDL particles which were similar in composition, size and hydrated density to authentic HDL could be produced from either apoprotein by sonication. They 17 / clearly showed that, relative to spontaneous interaction, sonication markedly improved the complex formation of apo A-I or A-II and HDL lipids; however, they also demonstrated a requirement that apo A-I be in a monomeric form for optimal complex yields. This limitation seems to have discouraged many investigators from seriously employing these methods and may have prompted the statement by Jonas, in her recent review, that the use of sonication for microemulsion production is limited due to poor yields and reproducibility (97). Nonetheless, several investigators have recently utilized sonication methodologies to produce particles for studies of lipid-protein interactions (101) as well as for biological studies (102,103). 1.3.2 Structure and metabolism Ritter and Scanu (99,100) showed that rHDL could be prepared with a similar size and structure to authentic HDL by sonication. They showed that rHDL prepared from purely apo A-I were also similar to native HDL in size (6-8 nm) and composition, while rHDL prepared from pure apo A-II were much more heterogeneous in both size (9-20 nm) and composition. They further showed that the state of association of purified apo A-I has a marked influence on its ability to bind lipids and that optimal reassembly of HDL particles requires the presence in solution of monomer-dimer forms of apo A-I. Recently, Pittman et al (102) have attempted to overcome the problem of self-association of apo A-I by including 2.5-4 M urea in their apo A-I preparations; however, no improvement in yield was reported. Their studies employed rHDL prepared by sonication of commercially available lipids and rat apo A-I. They showed that rHDL interact with cultured cells comparable to native HDL particles and allow selective cholesteryl ester uptake without concomitant particle uptake. In in vivo studies, they further showed that the rHDL had almost identical tissue uptake and plasma decay kinetics as authentic HDL (102). In other studies employing rHDL prepared by sonication, Jones and Rogers (103) have shown that rHDL are also comparable to native particles in their capacity to inhibit lipoprotein lipase. Structural studies by Fenske et al (101) have further shown that the phospholipid monolayer of rHDL, produced by sonication, have essentially 18 identical acyl chain orientational order and relaxation rates as native HDL particles. Taken together, these studies suggest that synthetic HDL produced by sonication are in fact representative of authentic HDL both structurally and biologically. 1.4 RATIONALE FOR THIS STUDY The inverse relationship between high-density lipoprotein (HDL) and cardiovascular disease has led to an intensive investigation of the role of this lipoprotein in reverse cholesterol transport (6,14). Much of the recent research effort has been placed on attempts to understand the role of the lipoprotein apoproteins in regulating lipoprotein structure and function. Recently, however, Little et al (92) reported that the concentration of , T G in the HDL particle may to be a significant independent risk factor for ischemic heart disease. While the molecular mechanisms by which this is brought about are not known, their results suggest that the increased triglyceride content of HDL may interfere with its function in reverse cholesterol transport. As with any biological macromolecular assembly, the function of the lipoproteins must be intimately related to their structure. This, in turn, will be governed by the interactions of the various lipid and protein components. Thus, it is proposed that an increase in H D L - T G concentration modifies the ability of HDL to function normally. CETP-mediated transfer of cholesteryl ester and triglyceride between HDL and other lipoproteins has been suggested to play a central role in the process of reverse cholesterol transport (14). Moreover, the observation that CETP may collaborate with the triglyceride lipases in the remodeling of the neutral lipid core of lipoproteins (72) suggests that a well regulated process is operating to maintain the lipoprotein integrity throughout transitory postprandial lipemic periods. The role of CETP in the etiology of atherosclerosis is not yet clear; however, several reports have suggested that the abnormal lipoprotein profile in some hyperlipidemic states may be due to an altered rate of transfer of cholesteryl ester to LDL and VLDL (72,86). In addition, the factors which directly affect CETP function any also have a marked effect on L C A T since the 19 production of cholesteryl esters by LCAT must be integrally linked to their CETP-mediated equilibration in the plasma compartment. If we are to fully understand the relationship between cholesteryl ester transfer and atherosclerosis, we must elucidate the role of HDL composition in the regulation of both CETP and L C A T , in an attempt to reveal the causes and perhaps consequences of abnormal lipoprotein levels. The primary aim of this study is to elucidate the effect of elevated HDL triglyceride concentration, relative to cholesteryl ester, on the ability of these particles to act as substrates for CETP and LCAT. This work was carried out in two major sections. Firstly, cholesteryl ester transfer activities and HDL compositions were determined and compared in plasma from-normal and hyperlipidemic subjects. Secondly, an in vitro model lipoprotein system was developed to directly study the effect of variations in lipoprotein composition on the ability of these particles to act as a substrate for purified CETP and LCAT. These studies have provided novel information regarding the type of lipoproteins that are required for efficient cholesteryl ester production and transfer in the plasma compartment. 20 1.5 SPECIFIC AIMS 1. To develop a new assay for the determination of cholesteryl ester transfer activity in plasma. 2. To measure cholesteryl ester transfer activities and HDL composition in normolipidemic and hyperlipidemic subjects. 3. To develop an HDL-analogue (rHDL) in which the protein and lipid components may be accurately manipulated. 4. To characterize the effects of variations in lipid and apoprotein composition of the rHDL on its ability to act as a substrate for purified CETP. 5. To characterize the effects of variations in rHDL composition on its ability to act as a substrate for purified LCAT. 21 2 MATERIAL AND METHODS 2.1 MATERIALS [ H]cholesteryl oleate (specific activity 72 Ci/mmol) was obtained from NEN Research Products, Quebec. [ HJcholesterol (specific activity 5 Ci/mmol) was purchased from Amersham Corp., Arlington Heights, IL. Bovine and human serum albumin, beta-mercaptoethanol, cholesterol, cholesteryl linoleate, triolein, and MnSO^ were purchased from Sigma Chemical Co., St. Louis, MO. Deuterium oxide was purchased from MSD isotopes, Montreal. Egg phosphatidylcholine was purchased from Avanti Polar Lipids, Birmingham, Alabama. Sodium 2,2-dimethyl silapentane 5-sulfonate (DSS) was purchased from Stable Isotopes, Montreal, Quebec. Kits for the determination of total and free cholesterol and triglyceride were purchased from Boehringer, Mannheim, F.R.G. All other chemicals were analytical grade from BDH Chemicals Canada Ltd., Vancouver, B.C. 2.2 LIPOPROTEINS 2.2.1 Isolation Blood was collected from 16 h fasted, normolipidemic subjects into E D T A -containing (1.5 M) tubes. Plasma was recovered by centrifugation at 1750 X g for 10 min. V L D L , LDL, and HDL were isolated by sequential ultracentrifugation at densities 1.006, 1.006-1.063 and 1.063-1.210 g/mL, respectively (104). H D L 2 , H D L 3 , and VHDL were similarly isolated at densities 1.063-1.125, 1.125-1.21, and 1.21-1.25 g/mL respectively (104). Density adjustments were accomplished by adding, to 1 volume of plasma, 0.5 volume of salt solutions; 1.006, 1.182, and 1.478 g/mL (for V L D L , LDL and HDL respectively), containing NaCl, NaBr, 0.01% EDTA and 0.03% NaN^. In some cases, specific lipoprotein subclasses were directly isolated by addition of solid NaBr to obtain the required density. Ultracentrifugation at 114,000 x g was performed at 15°C for 18h (VLDL), 20h (LDL), and 48h (HDL). All lipoproteins were subsequently recentrifuged at the same density to remove any contaminating lipoproteins or plasma proteins. The washed lipoproteins were dialyzed four times for 4-12 h against 100-fold greater volumes of NaCl/Tris buffer (buffer A; 150 mM NaCl, 10 mM Tris/HCl, 0.3 mM EDTA, 4.6 mM NaN,, pH 7.4), before use. 22 2.2.2 Characterization of lipoproteins 2.2.2.1 Composition Total cholesterol, free cholesterol and triglycerides were determined enzymatically using Boehringer Mannheim kits and manufacturer's suggested procedures. Free cholesterol assays were based upon the oxidation of cholesterol with cholesterol oxidase to produce hydrogen peroxide, which is further reacted with catalase and methanol to produce formaldehyde. A final colorimetric reaction occurs when formaldehyde is reacted with ammonium ions and acetylacetone to produce a yellow colored compound 3,5-diacetyl-l,4-dihydrolutidine (105). Total cholesterol values were determined after cleaving fatty acids with cholesterol esterase and esterified cholesterol was then determined mathematically by subtracting unesterified from total cholesterol values. Triglyceride assays were based upon lipase hydrolysis of triglycerides to glycerol, subsequent action by glycerol kinase and glycerol phosphate oxidase to produce hydrogen peroxide, and a final colorimetric reaction with peroxidase, 4-aminophenazone and 4-chlorophenol to produce a pink colored compound, 4-(p-bensoquinone-mono-imino)-phenazone (106). Phospholipids were determined after an acidic Bligh and Dyer lipid extraction (107) by the method of Anderson et al (108) . This is based upon a phosphoric acid/molybdate complex formed by digestion with sulfuric acid and subsequent oxidation with hydrogen peroxide. Proteins were determined on TCA precipitable proteins by a modified Lowry method (109) . Plasma apolipoprotein A-I and B values were quantitated by radial immunodiffusion (Tago, Diffu-gen) which involved application of diluted sample to sample wells, incubation for 48h, and subsequent measurement of ring diameter. HDL lipid composition was also determined enzymatically, after precipitation of LDL and VLDL with heparin/MnCl2 (90 mM MnCl 2 , 10 mM NaCl, and 227 USP units/L of heparin) (110). HDL^ cholesterol was determined after further precipitation of the H D L 2 by dextran sulfate (72.7 uM)(110). HDL2-cholesterol was calculated by subtracting HDL^ cholesterol from HDL-cholesterol values. LDL cholesterol levels were calculated from the formula (111): LDL-cholesterol = Total cholesterol - [(Triglyceride/2.2) + HDL-cholesterol]. 23 2.2.2.2 Agarose electrophoresis 1 uL of plasma or isolated lipoprotein was layered into the sample wells of prepoured 1% agarose gels (Corning Universal) and then electrophoresed for 35 min using a Corning Universal PHAB buffer (barbital, pH 8.6). Agarose gels were then dried and stained with 0.023% fat red 7B in methanol. Gels were washed with in 50% methanol, redded and scanned on a Beckman Appraise 1123 densitometer. 2.2.2.3 Polyacrylamide electrophoresis of plasma proteins Apparent molecular weights of plasma proteins and apoproteins were determined by SDS gel electrophoresis on 7% or 12.5% polyacrylamide 0.75mm gels, pH 8.8, in the presence of beta-mercaptoethanol, with a 4% acrylamide, pH 6.8, stacking gel. The running buffer was 25 mM Tris-glycine, pH 8.3, containing 0.1% SDS and reference standards used were: phosphorylase b (M r 92,500), bovine serum albumin (M r 66,200), ovalbumin (M f 45,000), carbonic anhydrase (M r 31,000), soybean trypsin inhibitor (M f 21,500), and lysozyme (M f 14,400). Proteins were stained with 0.25% Coumassie R250 in acetic acid/methanol/H^O (1:4.5:4.5) and destained in acetic acid/methanol/TLjO (1:4.5:4.5). Isoelectric focusing of apoproteins (200 ug) was performed on 7.5% acrylamide tube gels (5 mm) containing pH 4-6 ampholytes as described by Warnick et al (112). Samples were focused for 16h at 450 volts, stained with 0.04% Coumassie Blue G250 in 3.5% perchloric acid, destained with 7.5% acetic acid, and scanned on a Beckman Appraise 1123 densitometer. 2.2.3 Incorporation of radiolabeled lipids into lipoproteins 2.2.3.1 [3H]cholesterol Plasma lipoproteins were labeled by drying [HJcholesterol under adding the lipoprotein sample, incubating at 4°C overnight, and then transferring the sample to a new container. Gel filtration of labeled lipoproteins showed that the [HJcholesterol was associated with the lipoprotein. Labeling efficiencies were usually between 10-20% with losses primarily due to adhesion of the labeled lipid to the labeling container. 24 2.2.3.2 [JH]cholesteryl ester Two different methods for labeling plasma lipoproteins with [ HJcholesteryl ester were employed. Lipoproteins for calibration of size exclusion columns • and for the preparation of the [ HJCE-LDL substrate for CETP assays, were prepared as follows: [3H]cholesteryl oleate and EYPC (100 uCi/7mg) were dried under rehydrated in buffer A, and then sonicated as described in section 2.3.2. Labeled vesicles were incubated with fresh plasma for 24h at 37°C and then the labeled lipoproteins were isolated by sequential ultracentrifugation at their appropriate densities. This method resulted in marked compositional modification of HDL (elevated triglyceride content) and therefore another method was developed for preparing HDL for use in the assay of CETP using solid-phase-bound HDL. [3H]cholesteryl oleate (10-50 uCi) was dried under N 2 with 50 uL of HDL (5 g/L). Additional HDL (20-60 mg) was added and the mixture gently vortex-mixed and sonicated twice (section 2.3.2) under N 2 . After storage overnight at 4°C, the labeled HDL was re-isolated by ultracentrifugation. Reisolated, labeled HDL had exactly the same protein and lipid composition as before the labeling procedure. Gel filtration through a Superose 6 column resulted in identical protein elution profiles for [ H]HDL and native HDL. Agarose electrophoresis further showed that electrophoretic mobility was unchanged. Labeling efficiencies were usually between 20 and 60%. 2.3 PREPARATION OF RECOMBINANT LIPOPROTEINS 2.3.1 Purification of apoproteins HDL was delipidated in 50 volumes ethanol:diethyl ether (3:2) overnight at -20°C (113) and then centrifuged to pellet the precipitate. The solvent was aspirated and the aspirant combined with excess diethyl ether (final ethanol/ether ratio; 3:5) and stored overnight at -20°C to allow precipitation of the ethanol soluble apoproteins. The precipitates were pooled, washed three times with ether, dried under N 2 and resolubilized in 6M guanidine HC1. The apo HDL was extensively dialyzed against buffer A. The absence of albumin was confirmed by SDS-PAGE. Purified apoproteins A-I and A-II were prepared by chromatofocusing (114). Briefly, apo HDL (100 mg) was solubilized in lOmM 25 Tris/HCL (pH 7.4) containing 7.2M urea and 10 mM dithiothreitol (DTT) and was loaded onto a 70 x 1.5 cm column of PBE 94 (Pharmacia, Canada) equilibrated with 25mM imidazole HCL, 7.2M urea, and ImM EDTA (pH 7.4). The pH gradient was formed by elution with Polybuffer 74 (pH 4.0) diluted 1:8 with 8M urea, at a flow rate of 25 mL/h. The &2S0 °* e a c n fraction was determined and peaks were pooled. Apoproteins E, C-I, and C-III were prepared in a similar manner from delipidated VLDL by Roger McLeod (113) and were further resolved by chromatography on a 60 x 2.6 cm column of Sephacryl S-200 (Pharmacia, Canada). Polybuffer was removed by chromatography on Biogel HTP (Biorad Labs., Canada). Each apoprotein fraction was applied directly onto a HTP column equilibrated with lOmM sodium phosphate, 0.1% SDS (pH 7.2) and eluted with 0.5M phosphate, 0.1% SDS (pH 6.8). A.2gg peaks were again pooled, dialyzed against buffer A, characterized by isoelectric focusing and stored at -70°C. 2.3.2 Recombinant HDL preparation rHDL were prepared with varied lipid and apoprotein compositions (Table 12) by a sonication method similar to that described by Hirz and Scanu (97). Purified lipids were solubilized in chloroform and the appropriate quantities were combined in a 12 x 75 mm test tube and dried under The desired amount of apo HDL (or purified apoproteins) was added with buffer A containing 0.8M NaBr, in a total of 3 ml. The mixture was suspended in a cold water bath (14°C) and sonicated, using a microtip probe, through eight cycles of 30 sec at 75 watts with 30 sec cooling periods. rHDL was immediately reisolated by sequential ultracentrifugation between densities 1.063-1.21 g/mL, dialyzed extensively against buffer A and stored at 4°C. 2.3.3 Characterization of rHDL 2.3.3.1 Gradient ultracentrifugation The density of rHDL was determined by ultracentrifugation in a discontinuous sucrose/NaCI gradient (115). 500 mg sucrose was placed into a 12.8 mL cellulose nitrate tube and overlayed with 5 mL 4M NaCl, lmL microemulsion, and 6.8 mL of a 0.67M NaCl solution containing 0.05% EDTA, pH 7.0. The gradient 26 was spun in an SW41 swinging bucket rotor for 48h at 38,000 rpm and 10°C and the rotor slowed with no brake. The contents of each tube was pumped out the bottom and collected in 10 fractions. The density and lipid composition of each fraction was subsequently determined. 2.3.3.2 Gel filtration Size exclusion chromatography of native and recombinant lipoproteins that had been previously labeled with [ Hjcholesteryl ester was performed on a 10 x 300 mm column of Superose 6 (Pharmacia, Canada) with buffer A as elution buffer and a flow rate of 40 mL/h. 35 x 0.5 mL samples were collected and analyzed for radioactivity and protein content. 2.3.3.3 Electron microscopy Negative staining lipoprotein electron microscopy was performed as described by Forte and Nordhausen (116). Native or reconstituted HDL were isolated and immediately dialyzed against a buffer consisting of 0.125 M ammonium acetate, 2.6 M ammonium carbonate and 0.26 mM tetrasodium EDTA at pH 7.4. Each sample was adjusted to a concentration of 0.5 mg protein/mL and stored at 4°C. Immediately prior to examination, the sample was mixed with an equal volume of 2% sodium phosphotungstate (pH 7.4). A small droplet (3 uL) was then applied to a Formvar and carbon coated grid (200-300 mesh) and after 30 sec, excess fluid was blotted with a wedge of filter paper. Electron microscopy was performed with the assistance of Dr. David Walker, Department of Pathology. Grids were immediately examined on a Phillips 400 electron microscope and micrographs were photographed at an instrument magnification of 77,000. Mean particle dimensions of 100 particles were determined from projection of each negative. 2.3.3.4 Nuclear magnetic resonance spectroscopy ' H - N M R spectroscopy was done in collaboration with Dr. Yashpal Parmar. Deuterium oxide exchanges were performed on native HDL and rHDL to suppress the water signal. An Amicon ultrafiltration unit equipped with PM30 (30,000 M f cutoff) membranes was used for concentration and subsequent dilution with deuterium oxide. A total of four such exchanges were performed. EYPC-27 cholesterol vesicles (45 mole percent cholesterol) were prepared by co-dissolving lipids in chloroform, drying under N 2 , rehydrating in deuterium oxide and by sonication as described in section 2.3.2. The vesicles were approximately 48 nm in diameter as determined by Nicomp submicron laser light scattering (117). The samples (0.4 ml) were placed in a 5 mm NMR tube and the Fourier transformed spectra were obtained with sample spinning using a Bruker WP-200 spectrometer operating at 200 MHz. The temperature was 30.0 + 0.5°C. The NMR parameters used were the same for all of the spectra collected: sweep width = 2000 Hz; data size 4000; pulse width = 3.5 usee (45° flip angle); number of acquisition = 512. Line broadening was not used. The spectra were obtained with an interpulse delay of 1.02 sec. This was sufficient time to obtain equilibrium for longitudinal magnetization. An external standard, sodium 2,2-dimethyl silapentane 5-sulfonate (DSS) in deuterium oxide, was used to reference the peaks. Proton peak assignments were according to Hamilton and Morrisett (118). MnSO j^ was used to some experiments quench the N-methyl choline signal and was therefore freeze dried once in deuterium oxide once to remove residual water. 2.4 METABOLIC ASSAYS 2.4.1 Cholesteryl ester transfer activity CETP activity was determined using two different assays. One assay used delipidated plasma as a source of CETP and exogenous lipoproteins as acceptors and donors of cholesteryl ester. This allowed for the determination of cholesteryl ester transfer activity (CETA) independently of lipoprotein composition or content. The other assay was designed to determine CETA in plasma without prior removal of endogenous lipoproteins. Determinations from this assay were therefore representative of the activity of CETP under the influence of any lipoprotein associated factors. 2.4.1.1 CETA determination using exogenous lipoprotein substrates CETA was determined with exogenous lipoprotein substrates (isolated normolipidemic LDL and HDL) and lipoprotein depleted plasma (LPDP), or purified CETP (section 2.5), as a source of CETP. LPDP was prepared by addition of solid NaBr to plasma to achieve a density of 1.21 g/mL, ultracentrifugation at 114,000 28 x g for 48h, tube slicing to remove the d<1.21 lipoprotein supernatant, and extensive dialysis of the infranatant. LDL was labeled with [ HJcholesteryl ester by the previously described vesicular incubation method (section 2.2.3.2). CETA was measured in the following manner: purified HDL (30 ug total cholesterol) was incubated with [ 3H]LDL (30 ug total cholesterol), LPDP (300 uL) or purified CETP, BSA (1 mg),. and buffer A (to a total of 700 uL) for 0-180 min at 37°C. Total [3H]cholesteryl ester transferred to HDL was determined by precipitation of [ HJLDL with Heparin/MnC^ (as described in section 2.2.2.1) and determination of radioactivity in the supernatant. Incubations lacking the CETP source were run as controls and the rate of transfer of radioactivity was subtracted from that measured for the test incubations. 2.4.1.2 CETA in plasma •3 [HJcholesteryl ester labeled HDL (section 2.2.3.2, specific activity 4200 dpm/ug HDL protein) was covalently linked to CNBr-activated Sepharose 4B by the manufacturer's suggested procedures (119). This involved incubating the radiolabeled HDL with hydrated CNBr-activated Sepharose 4B (8.5 mg protein/mL gel) in a 0.1M NaHC0 3 , 0.5M NaCl buffer (pH 8.3) for 2h at room temperature on a end over end mixer. After coupling, any remaining active groups were blocked by incubation in a 1 M ethanolamine solution for 2 h at room temperature. The immobilized HDL was washed 8 times with buffer A, then with buffer A containing 1 ug/mL LDL and 0.5% BSA and finally 4 times with buffer A. The LDL/BSA wash •3 significantly reduced nonspecific transfer of [HJcholesteryl ester to plasma 125 lipoproteins. Coupling of I-HDL to Sepharose showed that no protein was lost •3 during the washing or subsequent incubation procedures. The [ HJHDL-Sepharose beads were stored in NaCl/Tris buffer at 4 °C until use. Beads used as long as 2 months after preparation showed no significant differences as assay substrates from freshly prepared beads. CET from immobilized HDL to plasma lipoproteins was measured 800 uL of plasma with 200 ug solid-phase-bound HDL protein for 0-180 min at 37°C on an end over end mixer (Figure 3). After incubation, bound HDL was removed by centrifugation and an aliquot of the supernatant was counted to determine the total [^HJcholesteryl ester transferred to plasma. Transfer to HDL was 29 P L A S M A I N C U B A T E A T 3 7 C F O R 0 - 2 H O U R S v D E T E R M I N E T O T A L 7" 3 H - C E T R A N S F E R T O P L A S M A H E P A R I N / M N C L , P R E C I P I T A T I O N O F L D L & V L D L - W D E T E R M I N E 3 H - C E T R A N S F E R T O H D L Figure 3. Solid-phase assay of cholesteryl ester transfer activity in plasma. 800 uL of plasma was incubated with 200 ug of solid-phase-bound HDL protein at 37 °C for 0-180 min. The Sepharose-bound HDL was removed by centrifugation and the supernate radioactivity was determined. The percent [ H]cholesteryl ester transfer to HDL was determined by precipitation of L D L / V L D L with heparin/MnCL, followed by supernate radioactivity determination. Transfer to LDL+VLDL was determined by subtracting the value for HDL from the total percentage transferred. Relative transfer to H D L 2 and HDLj subfractions was further determined by selectively precipitating the H D L 2 with dextran sulfate, followed by determination of the radioactivity associated with each fraction. 30 determined in the remainder of the plasma sample by precipitating the V L D L and LDL with heparin/MnCI2 (section 2.2.2.1) and determination of radioactivity in the supernate. Transfer to HDL^ was determined by precipitation of H D L 2 from the heparin/MnC^ supernatant with dextran sulfate (section 2.2.2.1) and determination of radioactivity in the resulting supernate. Transfer to the precipitated lipoproteins, (VLDL/LDL or HDL 2 ) , was calculated from the difference between total and nonprecipitable radioactivity. Greater than 98% of the [ H] transferred during the assay procedure was associated with cholesteryl ester, as shown by lipid extraction (120) and thin layer chromatography (121). 2.4.2 Lecithin:cholesterol acyltransferase 2.4.2.1 Determination of L C A T activity L C A T activity in plasma and for LCAT purification was measured using a substrate of single bilayer vesicles prepared by the method of Batzri and Korn (122). EYPC, cholesterol, and [JH]cholesterol were dried under N 2 , resolubilized in 125 uL absolute ethanol, and rapidly injected into 10 mL of buffer A. The mixture was concentrated to 2.5 mL in an Amicon concentrator and 30 uL aliquots were preincubated with 15 ug apo A-I at 37°C for 30 min. Each assay contained 4.65 nmol of unesterified [3H]cholesterol, 20 nmol EYPC, 15 ug of apo A-I, 50 uL 8% BSA, and 10 uL 3-mercaptoethanol, made up to 140 uL with buffer A. Esterification rates were measured over a period of 30 min at 37°C using 15 uL of plasma or LCAT source. The reaction was stopped by extraction of the mixtures with 4 mL C H C l 3 / C H 3 O H (2:1). The lipids were dried under N 2 , resolubilized in 70 uL C H C I 3 and chromatographed on silica gel TLC plates (121). Free and esterified cholesterol bands were visualized, cut out and counted. Less than 10% of the cholesterol in the substrates was esterified under these conditions. 2.4.2.2 Determination of LCAT activity with native and recombinant lipoproteins Native and reconstituted HDL, labeled with [JH]cholesterol (final specific activity: 0.013 uCi/nmol cholesterol) were used as substrates for L C A T in the following assay: each assay tube contained 2.5-30 nmol of lipoprotein free cholesterol, purified LCAT (0.4 ug), BSA (2%), beta-mercaptoethanol (0.42 mmol), 31 and buffer A (to a total of 100 uL). The molar ratio of unesterified cholesterol to phosphatidylcholine was comparable for both HDL and rHDL. Esterification rates were measured over a period of 1 h at 37°C by essentially the same technique described in section 2.3.2.1. 2.5 PURIFICATION OF LCAT AND CETP LPDP was prepared as described in section 2.4.1.1. The albumin poor, upper 50% of the d> 1.21 infranate was pooled and loaded directly onto 2.5 x 60 cm column containing Phenyl-Sepharose CL-4B, equilibrated with buffer A containing 2M NaCl (123). The column was extensively washed with 1M NaCl (800 mL) and the CETP and L C A T activities were co-eluted with distilled H 2 0 . CETP activity was measured as described in section 2.4.1.1 and L C A T activity determined as in section 2.4.2.1. The pooled activities were dialyzed overnight against distilled H 2 0 , equilibrated with 0.01M sodium acetate (pH 4.5), and loaded onto a 1.6 x 20 cm CM-52 cellulose column with a flow rate of 40 mL/h (123). All fractions were collected into tubes containing 0.5 mL 1M Tris (pH 7.4), to immediately neutralize the pH. Fractions (approximately 25 x 6mL) were eluted with the run buffer, in which the L C A T activity was eluted in the column void volume. CETP was subsequently eluted from the CM-52 cellulose column with a 400 mL linear 0-400 mM NaCl gradient (60). The CETP and LCAT activity peaks were pooled and concentrated in Aquacide. Both preparations were dialyzed against buffer A and stored at -70°C for up to 4 mos. Immediately before use, L C A T was further purified by chromatography on hydroxylapatite (123). The pooled CM-52 cellulose L C A T activity was equilibrated with ImM sodium phosphate (pH 7.4), loaded onto a 1.5 x 5 cm HTP column, and then eluted with a linear gradient of 10-100 mM sodium phosphate (pH 7.4) at 25 mL/h. The pooled LCAT activity was dialyzed against buffer A and used immediately. 2.6 CLINICAL STUDIES 2.6.1 Cholesteryl ester transfer in severe HDL deficient subjects Patients with the homozygous forms of LCAT deficiency, FED, and Tangier Disease were recruited from Shaughnessy Hospital Lipid Clinic. Control plasma 32 samples were obtained from 16 healthy normolipidemic volunteers. Blood from 16 h fasted normal and HDL deficient subjects was collected into EDTA-containing (1.5M) tubes. Plasma was removed by centrifugation at 1750 x g for 10 min and analyzed immediately. Cholesteryl ester transfer to endogenous plasma lipoproteins was determined by the solid-phase-bound HDL method described in section 2.4.1.2. Cholesteryl ester transfer rates were also determined in a lipoprotein depleted plasma sample from each patient as described in section 2.4.1.1. L C A T activities were determined in each subject as described in section 2.4.2.1. Plasma and lipoprotein lipids and apoproteins were determined as described in section 2.2.2.1. 2.6.2 Cholesteryl ester transfer in hyperlipidemic subjects Blood was drawn from 27 normolipidemic volunteers and 50 patients with primary dyslipidemias attending the Shaughnessy Hospital Lipid Clinic. Venipuncture was performed after a 16 h fast and samples were collected into EDTA-containing (1.5M) tubes. Plasma was recovered by centrifugation at 1750 x g for 10 min and analyzed im mediately as described in the following scheme. Plasma Measure CE transfer to the total lipoprotein pool. Measure: Triglycerides Cholesterol HDL-cholesterol Apoprotein A-I and B Precipitate V L D L / L D L with Heparin/MnCU Measure CE transfer to HDL and L D L / V L D L Determine HDL lipid composition: HDL-triglycerides HDL-cholesteryl ester HDL-free cholesterol HDL-phospholipids HDL-protein Precipitate H D L 2 with dextran sulfate Measure CE transfer to HDL-) and HDL, Determine: HDL^-cholesterol HDL2~cholesterol 33 Patients were grouped into six categories according to the clinical diagnosis: dysbetalipoproteinemia, familial hypercholesterolemia, familial combined hyperlipidemia, familial hypertriglyceridemia, hyperalphalipoproteinemia and combined hypoalphalipoproteinemia with hypertriglyceridemia. The diagnostic criteria used for these classifications are described in the Results, section 3.2.3.1. Evidence of both individual or familial history of vascular disease in these patients was determined from a retrospective chart review. 2.7 INCUBATIONS INVOLVING RECOMBINANT LIPOPROTEINS 2.7.1 Native and recombinant HDL as acceptors of cholesteryl ester rHDL were prepared with varied lipid and apoprotein compositions and compared to native normolipidemic HDL ( H D L 2 , H D L 3 and VHDL) as acceptors of cholesteryl ester from LDL, in incubations similar to that described in section 2.4.1.1, except that partly pure CETP (section 2.5, 1.3 ug protein) was used in place of the LPDP. In three different series, rHDL were prepared in which: 1) core lipid (cholesteryl ester:triglyceride) 2) surface lipid (phospholipidxholesterol) and 3) apoprotein (apo A-I, apo A-II, apo E , apo C-I and apo C-III) compositions were varied. Starting lipid concentrations were varied for core and surface lipid studies; however, for apoprotein characterizations the starting concentrations were kept constant at 48% protein, 30% phospholipid, 5% triglyceride, 5% cholesterol and 12% cholesteryl ester by weight (Table 12). Both native HDL and rHDL were utilized in assays of cholesteryl ester transfer and compared as acceptors of cholesteryl ester. Initial characterization studies of all rHDL indicated that less, than 5% of the rHDL protein or lipids were precipitable by heparin/MnC^, under the conditions described. This was similar to the value obtained for native HDL and indicates that differences in the radioactivity recovered in the non-precipitable fractions were due to differences in the amount transferred. 2.7.2 rHDL as substrates for LCAT Native and recombinant HDL were compared as substrates for purified human L C A T in the assay described in section 2.4.2.2. rHDL were prepared with varying 34 cholesteryl ester/triglyceride content as described in Table 12, series A. 2.8 MATHEMATICAL ANALYSES 2.8.1 Statistical analyses Significance of difference between population means for the control group and the specific disorders of lipid metabolism were determined by an unpaired Student's t-test. Correlation coefficients were determined by linear regression analyses and P values were calculated by Fisher Z-transformation. 2.8.2 Kinetic analyses Reaction kinetics for experiments with CETP were estimated from double reciprocal and direct linear (Cornish-Bowden) plots (124) in an attempt to numerically characterize variations in the ability of different HDL to accept cholesteryl ester from LDL. In this context, estimated values, analogous to V m a x and K m , represent the maximum velocity of transfer, T m a x , and the HDL concentration required for half maximal activity, K J J . Thus, these terms are merely operational terms since CETP is not an enzyme in the classical sense, nor can HDL be considered an enzyme substrate. Reaction kinetics were calculated as an average of estimations on double reciprocal and direct linear plots since estimation on double reciprocal plots alone may be biased by the linearity of the data (124) and in some cases the linearity of the data presented may be questioned (particularly in cases of apparent substrate inhibition, section 3.5.2). Determinations on direct linear plots were independent of data linearity and usually corroborated estimations from double reciprocal plots. In this study, the term T m a x is used to rank the ability of different HDL to accept cholesteryl ester. Similarly, the term K J J is used merely to rank differences in the ability of CETP to interact with the HDL population studied. Kinetic analysis for experiments with LCAT were also estimated from both double reciprocal and direct linear plots. In this case though, since L C A T is well established as an enzyme, kinetic determinations are presented in their classical manner, as V m n v and K m . 35 3 RESULTS 3.1 CHARACTERIZATION OF LIPID TRANSFER ASSAYS 3.1.1 Lipid transfer in lipoprotein depleted plasma Transfer of the [HJcholesteryl ester from radiolabeled LDL to HDL, in the presence of LPDP, was unstaturated between 0-180 min (Appendix A; Figure 1A). Approximately 11% of the initial radioactivity in LDL was transferred to HDL in 1.5h and 16% in 3h. There was a linear relationship between the total amount amount of radiolabeled lipid transferred and the volume of LPDP present in the incubation mixtures (Appendix A; Figure IB) up to 500 ul LPDP. In the absence •3 of LPDP, approximately 1.5% of the [JH] was transferred to HDL, BSA and buffer. When HDL was omitted, 7% of the radiolabel was transferred to the d>1.21 alone (data not shown). Increasing HDL concentrations above 10 ug total cholesterol •3 HDL did not affect the total [HJcholesteryl ester transferred (Appendix A; •3 Figure 2A). Similarly, increasing [ JH]LDL concentrations above 15 ug total cholesterol LDL, resulted in only small increases in cholesteryl ester transferred, indicating a reduced dependence on LDL concentrations (Appendix A; Figure 2B). When donor and acceptor concentrations were maintained at 30 ug total cholesterol each, addition of 2000-fold purified CETP (section 3.3) resulted in linear increases in transfer rates up to 5 ug/mL, after which the rates plateaued at approximately 23 %/mL/h (Appendix A; Figure 3). 3.1.2 Lipid transfer in plasma •3 The transfer of [HJcholesteryl ester from solid-phase HDL to plasma lipoproteins was linear for 3 h and from 0.2-1.00 ml of plasma (Appendix A; Figure 4A and 4B, respectively). Transfer to specific lipoprotein classes (HDL or VLDL/LDL) was linear up to 2 h and was dependent on a factor (CETP) associated with the d>1.21 g/mL fraction of plasma (Appendix A; Figure 5A and 5B). Thus, this method simultaneously measures initial rates of transfer from the immobilized HDL to specific lipoprotein classes. After 1 h, the transfer of cholesteryl esters to HDL was approximately 80% lower in the absence of the 1.21 g/mL fraction (Appendix A; Figure 5A), while lipid transfer to V L D L / L D L was reduced only to 40% (Appendix A; Figure 5B) relative to plasma values. Similar 36 results were observed when purified VLDL, LDL and HDL were recombined, in the presence of 10 mg/mL human serum albumin, at concentrations similar to those found in normal plasma: 0.1 mg/mL VLDL, 1.0 mg/mL LDL, and 1.8 mg/mL HDL (lipoprotein protein/mL) and then assayed as for plasma (Appendix A; Figure 6). While the total transfer to the lipoprotein mixture was less than half that observed for normal plasma, this reduction was not equally distributed in the rates of transfer to HDL and L D L /V LDL. Transfer rates to HDL of the lipoprotein mixture was 75% lower than to HDL in plasma whereas transfer to L D L / V L D L was decreased by only 25%. These results may confirm the observations of other investigators which have shown that a significant portion of the newly formed cholesteryl esters incorporated into LDL may be independent of the action of CETP (125). The lipid transfer activity observed in the absence of the d>1.21 g/mL fraction suggests that some CETP activity may have remained with the isolated lipoproteins during ultracentrifugal isolation, and was responsible for the cholesteryl ester transfer measured in the recombined lipoprotein incubations. In fact, subsequent studies have shown that a significant amount of CETP activity is associated with the HDL^ fraction even after ultracentrifugal isolation (section 3.5.2). The rates of [ HJcholesteryl ester transfer to endogenous plasma lipoproteins were also shown to be dependent on the addition of exogenous CETP (Appendix A; Figure 7A) and on the donor substrate concentration (Appendix A; Figure 7B). Conversely, when exogenous HDL was titrated into plasma (up to 0.6mg HDL protein/mL), no significant changes in transfer rates were observed (data not shown). Transfer rates also remained unchanged when either V L D L or LDL were added to plasma up to a final concentration of 1.55 mg/mL exogenous V L D L -triacylglycerol and 1.20 mg/mL exogenous LDL-cholesterol (data not shown). Therefore, since reaction rates were well within the linear range and were independent of acceptor substrate concentrations at 1 h, all clinical determinations were based on 1 h incubations. The addition of 1 mmol/L diisopropyl fluorophosphate to plasma was shown to inhibit 98% of the LCAT activity (data not shown), as previously reported (17). 37 Under these conditions, however, no significant effect on the cholesteryl ester transfer rates was observed (Appendix A; Figure 8). Storage of plasma at 4°C for up to 4 weeks was shown to markedly reduce cholesteryl ester transfer rates (Appendix A; Figure 9), while storage at -20°C for up to 4 months had no significant effects on transfer rates (data not shown). Cholesteryl ester transfer from solid-phase-bound HDL to plasma was further shown to be temperature dependent (Appendix A; Figure 10). Transfer rates.at 7 or 22°C were only approximately 10% of the rates at 37°C, while cholesteryl ester transfer rates at 42°C were increased by 150% over those measured at 37°C. 3.2 CLINICAL STUDIES 3.2.1 Comparison of lipid transfer assays Cholesteryl ester transfer activities were measured in the plasma and LPDP of 12 normolipidemic subjects and 30 patients with primary disorders of lipid metabolism. When the transfer of cholesteryl ester from immobilized HDL to endogenous lipoproteins was plotted against the activity determined in LPDP, a significant correlation (r=0.960) was observed for normolipidemic subjects (Figure 4A). This indicates that the total transfer rates measured by the solid-phase-bound HDL method are directly proportional to the rates determined in the d> 1.21 g/ml fraction. It further suggests that this measured activity is protein (CETP) dependent, but is independent of acceptor lipoprotein composition or content in normolipidemic individuals. When the same relationship was studied in 30 patients with primary disorders of lipid metabolism, no such correlation could be demonstrated (r=0.147; Figure 4B). This could be explained by differences in the activity of CETP or in the quantity and/or composition of the endogenous lipoproteins which may have altered the observed rate of cholesteryl ester transfer in the plasma of these patients. However, since many of the hyperlipidemic patients were shown to have elevated activities in the d> 1.21 fraction, relative to plasma, it is possible that the reduced plasma activity may have resulted from of some inhibitor associated with the lipoproteins in these patients, but not present in the d>I.21 fraction. Interestingly, many of the patients with low plasma CETP 38 C E T A IN P L A S M A ( % / m L / h ) Figure 4. Cholesteryl ester transfer to plasma and lipoprotein depleted plasma of normal and hyperlipidemic subjects. Cholesteryl ester transfer activities were determined in 800 uL plasma (x-axis), as described in Appendix A, Figure 7 and 300 uL patients LPDP (labeled d>1.21 on y-axis), with exogenous substrates and incubation time as described in Figure 6, from normolipidemic subjects (A) and dyslipidemic patients (B). Regression analyses indicated a significant positive relationship (solid line, r=0.960) between assays in normolipidemic subjects (n=12) but no relationship (r=0.147) in hyperlipidemic subjects (n=30). 39 activity also had documented evidence of vascular disease. While the etiology of vascular disease is probably very diverse in this extremely heterogeneous group of patients, perhaps CETP inhibition may be in some way involved in the development of vascular disease in the patients studied. 3.2.2 CETP and patients with severe HDL deficiency Cholesteryl ester transfer activities were measured in the plasma and LPDP of three patients with severe HDL deficiencies: LCAT deficiency, Tangier disease and FED. Plasma lipid values for severely HDL deficient patients are shown in Table 1. All three patients presented with HDL-cholesterol below the 10th percentile and both Tangier and FED patients had moderate hypertriglyceridemia. Total cholesterol values were more than 2 SD lower than normals in both Tangier disease and LCAT deficiency, but were significantly elevated in FED. L C A T activity was not detectable in LCAT deficiency, while FED was reduced 94% and Tangiers disease was reduced 42% as compared to average normal (Table 1). These results agree with previously reported data (126,127,128). Cholesteryl ester transfer from solid-phase bound HDL to V L D L / L D L in Tangier plasma was significantly (>2 SD) increased by 145% over controls (Table 2); however, transfer to V L D L / L D L in LCAT deficiency and FED was normal. Transfer to HDL was approximately 25% of normal values for L C A T deficiency and FED and 8.6% of normal in Tangier disease (Table 2). In addition, the increased transfer of cholesterol ester to endogenous lipoproteins in Tangier disease was associated with an increased transfer rate in LPDP (Figure 15). A similar relationship was observed for the low transfer rates in both plasma and LPDP from the FED and LCAT deficient subjects. Linear regression analysis of lipid transfer rates estimated in plasma and in LPDP (normolipidemic subjects from section 3.2.1 and HDL deficient patients) gave a positive correlation, with r=0.950, between the transfer rates determined by the different assays (Figure 15). This suggests that altered transfer rates measured were not totally associated with endogenous lipoprotein content and composition but may have been associated with changes in, CETP mass or degree of inhibition. 40 50 Q 4 0 -30-| > 20. Ld o 10' 0 TANGIER DISEASE NORMAL L C A T D E F I C I E N C Y FISH EYE DISEASE r = 0 . 9 5 0 —« 1 « 1 ' 1 « I ! 0 4 8 12 16 20 CE TRANSFER TO PLASMA (%/mL/h) Figure 5. Cholesteryl ester transfer to plasma and lipoprotein depleted plasma of normolipidemic and HDL deficient subjects. The rate of cholesteryl ester transfer was measured in plasma and LPDP of 12 normolipidemic subjects (Figure 4A) and 3 patients with severe HDL deficiency. Normolipidemic values are plotted as a mean+SD. HDL deficient values are the average of two independent duplicate determinations. Linear regression analysis indicated a significant relationship (solid line, r=0.950) between assays of cholesteryl ester transfer activities in plasma (x-axis) and lipoprotein depleted plasma (y-axis) for all subjects studied. 41 Table 1. Plasma l ipid and lipoprotein composition in severe HDL-deficiency Controla Tangier Disease Plasma CE (mg/mL) FC (mg/mL) T G (mg/mL) HDL CE (mg/mL) FC (mg/mL) T G (mg/mL) L C A T ACTIVITY (nmol est'd/mL/h) LCAT deficiency 160±25 48+ 8 85+24 43±10 15+ 5 28+ 4 31.6+4.6 60* 42* 320 2* 11 18.3 * 12* 82 75 Fish Eye Disease 118* 71* 276 * 3* 4* 10 0.0 1.8 *n=16 greater than 2 SD difference from the mean of the control subjects. Table 2. Cholesteryl ester transfer activities iri plasma Cholesteryl ester transfer (% transfer/mL/h) Controla Tangier LCAT Fish Eye Disease Deficiency Disease Acceptor: HDL 5.8+1.1 0.5 V L D L / L D L 5.5+0.9 13.3* 1.2 7.1 1.4 7.1 aSD, n=16 greater than 2 SD difference from the mean of the control subjects. 3.2.3 CETA and HDL composition in hyperlipidemia 3.2.3.1 Patient characterization All dyslipoproteinemic patients had primary lipid disorders and neither were severely obese nor had other concomitant disorders likely to affect lipoprotein metabolism. Total lipid and apoprotein A-I and B results are shown in Table 3. Familial combined hyperlipidemic patients (FCH) presented with elevated cholesterol, triglycerides, LDL-cholesterol and apo B levels and with a 42 documented familial history of hyperlipidemia. Familial hypercholesterolemic patients (FH) were defined by demonstrating evidence of genetic transmission as well as elevated cholesterol, LDL-cholesterol and apo B levels, but normal triglyceride levels (<2 mmol/L). In most cases, tendon xanthomas were present. Dysbetalipoproteinemic (DYSB) subjects had the apolipoprotein ^2/2 Phenotype and elevations of B-mobility VLDL. Hypertriglyceridemia (FTG) was defined as triglyceride elevation with normal cholesterol (<6 mmol/L) and apolipoprotein levels. Hypoalphalipoproteinemic patients were those with significantly reduced apolipoprotein A-I and HDL-cholesterol levels. These patients were subgrouped into two subclasses; hypoalphalipoproteinemia with hypertriglyceridemia (HA&HTG) and hypoalphalipoproteinemia with normal triglyceride and cholesterol levels (HA). The presence of vascular disease was determined from retrospective chart review and was defined as documented evidence of one or more of the following: myocardial infarction, stroke, or peripheral vascular disease. Table 3. Plasma lipids and apoproteins in dyslipidemic patients Triglyceride Total LDL Apo-B Apo-AI cholesterol cholesterol n (mmol/L) (mg/mL) Normal (27) 1.0±0.32 4.6+0.6 2.7+0.5 0.63+0.24 1.28±0.25 FCH (19) 3.1±1.4 7.2+1.7 4.7+1.8 1.11+0.3 1.29±0.34 FH (7) 1.5±0.83 7.5+2.0 5.5+1.7 1.19+0.29 1.26+0.49 FTG (4) 5.3±0.91 5.4+0.4 2.1+0.6 0.80+0.12 1.28+0.15 DYSB (5) 7.3±6.9 7.9±1.0 5.3+0.2 0.94+0.31 1.15+0.13 HA (6) 1.8+0.46 5.0+0.3 3.5+0.4 0.89+0.04 0.96±0.08 HA+HTG (9) 5.3±3.0 5.8+1.1 2.6±0.8 0.90±0.14 1.00+0.13 Abbreviations: DYSB, dysbetalipoproteinemia; HA+HTG, hypoalphalipoproteinemia with hypertriglyceridemia; HA, hypoalphalipoproteinemia; F T G , familial hypertriglyceridemia; FH, familial hypercholesterolemia; FCH, familial combined hyperlipidemia; NORMAL, normolipidemia. 43 3.2.3.2 HDL composition HDL-cholesterol and HDL^-cholesterol levels were significantly reduced in all patient groups except in those with hypercholesterolemia (Table 4). H D L 2 -cholesterol levels, however, were reduced only in patients with hypoalphalipoproteinemia and hypertriglyceridemia. HDL lipid mass, calculated as the sum of all the HDL lipid components, was significantly reduced in patients with hypertriglyceridemia and hypoalphalipoproteinemia with (or without) hypertriglyceridemia. The relative lipid composition of HDL is shown in Table 5. In all patient groups, HDL-triglyceride was significantly increased while HDL-cholesteryl ester were decreased relative to normals. Furthermore, H D L -unesterified cholesterol was increased in hypercholesterolemia whereas H D L -phospholipid was decreased in both hypercholesterolemia and familial combined hypercholesterolemia. Table 4. Lipid composition of HDL from dyslipidemic patients HDL cholesterol n Total H D L 2 H D L 3 Total HDL lipid1 (mmol/L) Normal (27) 1.44+0.30 0.38+0.18 1.07+0.35 2.83+0.54 FCH (19) 1.26±0.40* 0.46+0.18 0.80+0.38* 2.74+0.79 FH (7) 1.42+0.62 0.57+0.56 0.85+0.14 2.72+1.2 FTG (4) *** 0.89±0.11 0.36+0.06 0.53±0.08** 2.22+0.13* DYSB (5) 1.01±0.26** 0.49±0.10 *** 0.52+0.17 2.62+0.71 HA (6) 0.69±0.26 0.29±0.18 *** 0.40+0.17 2.22+0.19** HA+HTG (9) *** 0.76+0.05 0.21+0.12* *** 0.55+0.12 *** 1.87+0.16 Significance of difference from controls: *+p<0.05 (determined by Student's t test) ***p<0.005 p<0.001 *sum of HDL-triglycerides, unesterified cholesterol, cholesteryl ester and phospholipid Abbreviations: DYSB, dysbetalipoproteinemia; HA+HTG, hypoalphalipoproteinemia with hypertriglyceridemia; HA, hypoalphalipoproteinemia; F T G , familial hypertriglyceridemia; FH, familial hypercholesterolemia; FCH, familial combined hyperlipidemia; NORMAL, normolipidemia. 44 Table 5. Relative HDL-lipid composition Triglyceride Unesterified cholesterol Cholesteryl ester Phospholipid (mole %] ) Normal (27) 6.5±1.4 7.3±1.6 43.6+4.4 42.6+4.4 FCH (19) *** 19.3+11 7.0+1.4 *** 36.8+6.4 *** 36.9±7.2 FH (7) *** 10.7±3.2 *** 10.3+1.7 39.5+5.0* 39.6+3.1* FTG (4) *** 18.9±0.3 6.9+0.3 *** 37.4+3.4 36.8+3.4 DYSB (5) *** 15.9±8.0 7.0±1.6 *** 32.3±6.6 44.8+7.3 HA (6) *** 20.6±4.5 7.4+1.7 *** 27.9±1.3 44.1+5.4 HA+HTQ9) *** 14.6±5.0 8.2+2.1 *** 33.3+3.2 43.9+6.7 Significance of differences from controls: **p<0.05 (determined by Student's t test) ++*p<0.005 p<0.001 Abbreviations: DYSB, dysbetalipoproteinemia; HA+HTG, hypoalphalipoproteinemia with hypertriglyceridemia; HA, hypoalphalipoproteinemia; F T G , familial hypertriglyceridemia; FH, familial hypercholesterolemia; FCH, familial combined hyperlipidemia; NORMAL, normolipidemia. 3.2.3.3 Cholesteryl ester transfer activity Total cholesteryl ester transfer rates (from solid-phase-bound HDL to endogenous lipoproteins) were significantly reduced in the plasma of patients with familial combined hyperlipidemia, familial hypercholesterolemia and hypoalphalipoproteinemia with hypertriglyceridemia, but were normal for all other classes (Table 6 ) . Cholesteryl ester transfer to LDL and V L D L was significantly increased in patients with dysbetalipoproteinemia and hypoalphalipoproteinemia (Figure 6 A ) while transfer to HDL was significantly reduced in all patient groups except hypertriglyceridemic patients (Figure 6 B ) . Determinations of the transfer of cholesteryl ester to HDL subclasses indicated a significant reduction in transfer to H D L 3 in all dyslipoproteinemic patients (Figure 6 C ) , relative to normals, and an increase in cholesteryl ester transfer to H D L 2 in hypertriglyceridemic patients (Figure 6 D ) . 45 \ _1 E 10-4-10. b o 6-2-LDL t VUX. E33 HDL3 I I I I 1 1s NORMAL DYSB HA-t- HA, FTC FH FCH HDL 24S - r x x X X X X X x x X X X X X X X X X X X X X X X X X X X ESS HDI-2 83 X X NORUAL DYSB HA+ HA FTC FH FCH Figure 6. Cholesteryl ester transfer in normal and dyslipidemic subjects. Cholesteryl ester transfer from solid-phase-bound HDL to 800 uL plasma from normal and dyslipidemic patients is shown. Lipid transfer rates to specific lipoprotein subclasses; LDL and VLDL (A), HDL (B), HDL3 (C), and H D L 2 (D) were determined as described in section 2.4.1.2 and are expressed as the mean+lSD of patient groups: DYSB, dysbetalipoproteinemia (n=5); HA+, hypoalphalipoproteinemia with hypertriglyceridemia (n=9); HA, hypoalphalipoproteinemia (n=6); F T G , familial hypertriglyceridemia (n=4); FH, familial hypercholesterolemia (n=7); FCH, familial combined hyperlipidemia (n=19); NORMAL, normolipidemia (n=27). Significance of difference from controls: **p<0.05 (determined by Student's t test) „*p<0.005 v p<0.001 46 Table 6. Cholesteryl ester transfer from solid-phase-bound HDL to plasma lipoproteins. Cholesteryl ester transfer activity (% CE transferred/ml plasma/h) Normal (27) 11.2+1.4 * Familial combined hypercholesterolemia (19) 10.5+1.3 * Familial hypercholesterolemia (7) 9.4+1.3 Familial hypertriglyceridemia (4) 10.9+0.5 Dysbetalipoproteinemia (5) 11.7+1.9 Hypoalphalipoproteinemia (6) 10.7+1.4 * Hypoalphalipoproteinemia (9) 9.5+2.0 with hypertriglyceridemia Transfer rates were determined in 800 uL of subjects plasma as described in section 2.4.1.2. Values are the mean+lSD. * Significance of difference for controls: %+p<0.05 (determined by Student's't test) p<0.005 3.2.3.4 Statistical Analyses. The relationship between cholesteryl ester transfer activity and all lipid and apolipoprotein data in dyslipidemic subjects was studied by correlation analysis and P values were determined by Fisher z transformation. Analyses showed that cholesteryl ester transfer to HDL was significantly related to HDL apoprotein A-I levels (P<0.001) and all HDL lipids (P<0.005) except triglyceride. Lipid transfer to LDL and VLDL was significantly but inversely proportional to HDL-cholesterol (P<0.005), esterified cholesterol (P<0.005), unesterified cholesterol (P<0.05), HDL3-cholesterol (P<0.001) and total HDL lipid mass (P<0.05). This suggests that reduced lipid transfer to HDL was associated with decreased levels of HDL-cholesterol and a general reduction in HDL-lipid mass. This type of analysis may be misleading, however, since specific factors associated with particular disease groups may be overlooked. In fact, one group with apparently normal HDL mass and apoprotein A-I levels, dysbetalipoproteinemia, was shown to have significantly reduced cholesteryl ester transfer to HDL and increased transfer to L D L / V L D L . 47 Characterization of patients with documented vascular disease identified some significant differences from subjects with no evidence of disease (Table 7). Patients with vascular disease were shown to have significantly increased levels of HDL-unesterified cholesterol (P<0.005) and reduced cholesteryl ester (P<0.05) and HDL^-cholesterol (P<0.05) levels. In addition, these patients were also shown to have significantly reduced cholesteryl ester transfer to the total lipoprotein pool (P<0.05), to L D L / V L D L (P<0.05) and to H D L 3 (P<0.05), when compared to all other subjects. This prompts further speculation that perhaps, inhibition of CETP activity may be in some way associated with the incidence of vascular disease. Table 7. HDL lipid composition and cholesteryl ester transfer activity in patients with vascular disease. No disease Vascular disease (n=65) (n=12) (mole %) ^ 7.3+1.6 8.9+2.3 12.5+8.9 14.8+6.6* 39.1±7.3 34.2+4.8 41.1+6.2 42.1+6.5 (%/mL/h) * 10.8±1.4 9.6+2.0* 6.2+1.2 5.4+1.5 4.6±1.3 4.2+1.1 1.5±0.9 1.8±0.9* 3.1±0.8 2.4+0.7 Significance of difference between groups: **p<0.05 (determined by Student's t test) p<0.005 Presence of vascular disease was determined from retrospective chart review as described section 3.2.3.1. CE transfer rates were determined in 800 uL of subjects plasma as described in section 2.4.1.2. Values are the mean+lSD. 3.3 Purification of CETP and LCAT Fractionation of the d>1.21 g/mL fraction of a normolipidemic subject and determination of both L C A T and CETP activities in each fraction showed that the bulk of both activities were primarily associated with the density region 1.21-1.23 g/mL (Appendix A; Figure 11). Since this region corresponded with the albumin poor, clear zone, immediately beneath the lipoprotein layer of a d=1.21 HDL lipid composition unesterified cholesterol triglyceride cholesteryl ester phospholipid CE transfer to: all lipoproteins L D L / V L D L HDL H D L . H D L 3 48 g/mL ultracentrifugal spin, this fraction (1.21 mid) was used as a source of both LCAT and CETP for purification. Phenyl-Sepharose chromatography elution profiles (Appendix A; Figure 12) of CET activity (A) and L C A T activity (B) showed that both activities co-eluted in fractions 10-22. Conversely, C M -cellulose chromatography (Appendix A; Figure 13) separated L C A T activities, (B) which eluted in the void volume before the NaCl gradient, from CET activity (A), which eluted at approximately 150 mM NaCl. CETP was routinely purified approximately 1600-fold over lipoprotein depleted plasma (d>l .21 g/mL fraction) and contained no LCAT activity (Table 8, CM 1). This partially purified preparation was used as the CETP source in all subsequent incubations. SDS PAGE of this partially purified CETP resulted in a major band of approximately 70,000 daltons, corresponding to CETP, and a minor band of 30,000 daltons which was tentatively identified as apoprotein D (Figure 7). Further purification of CETP was not carried out due to the increasing instability of the protein at highly purified states (60). In one experiment, however, the pooled CETP activity fraction from the CM cellulose column (section 2.5) was rechromatographed on the same column resulting in an increased purification to 6800 fold over lipoprotein depleted plasma (Table 8). Table 8. CETP purification table. Fraction Volume (mL) Total protein(mg) Activity (%/mL/h) Spec.Act (act/mg) Fold purified % recovery d>1.21 400 30800 40400 1.31 1 100 1.21 mid 200 7600 49080 6.46 5 121 P-Sepharose 120 169 31056 185 141 77 CM l a 100 6.4 13390 2092 1595 33 CM 2 b 80 0.96 8568 8925 6804 21 CM cellulose elution #1 and #2 49 B 92 , 5 0 0 6 6 , 2 0 0 4 5 , 0 0 0 3 1 , 0 0 0 2 1 , 5 0 0 1 4 , 4 0 0 Figure 7. SDS-PAGE of partly purified CETP. A. delipidated HDL, d=1.063-1.21 g/mL (19 ug), B. pooled CETP active fractions from Phenyl-Sepharose (20 ug), C. pooled CETP active fractions from C M -cellulose, CM1 Table 8 (6 ug), and D. reference standards (2 ug): phosphorylase b (92,500), BSA (66,200), ovalbumin (45,000), carbonic anhydrase (31,000), soybean trypsin inhibitor (21,500), and lysozyme (14,400). The samples were electrophoresed in 12.5% polyacrylamide with SDS and in the presence of beta-mercaptoethanol. Proteins were stained with 0.25% Coumassie R250 in acetic acid/methanol/^O (1:4.5:4.5) and destained in acetic acid/methanol/^O (1:4.5:4.5). 50 9 2 , 5 0 0 6 6 , 2 0 0 4 5 , 0 0 0 3 1 , 0 0 0 2 1 , 5 0 0 Figure 8. SDS-PAGE of purified LCAT. Reference standards (2 ug, left lane) and purified LCAT (HTP-fraction) (0.20 ug) were electrophoresed on 7.0% polyacrylamide with SDS and in the presence of beta-mercaptoethanol. Proteins were stained and destained as described in Figure 20. 51 L C A T was routinely purified approximately 200-fold after CM-cellulose chromatography and then stored at -70°C until needed. Immediately prior to using, L C A T was. purified to greater than 10,000-fold through chromatography on hydroxylapatite (Table 9). SDS-PAGE of this preparation revealed a single band at approximately 68,000 daltons (Figure 8), which suggests that the protein was purified to near homogeneity (123). Table 9. L C A T purification table. Fraction Vdume (mL) Total protein(mg) activity (%/mL/h) Spec.Act. (act./mg) Fold purified % recovery d>1.21 300 21300 1108855 52 1 100 P-Sepharose 50 121 307582 2532 49 38 CM l a 71 23.4 269624 11508 221 24 H T P b 12 0.48 296983 618714 11885 27 ^CM-cellulose Hydroxylapatite 3.4 CHARACTERIZATION OF NATIVE AND RECOMBINANT HDL 3.4.1 Recombinant HDL composition and size The composition of the rHDL was determined after reisolation by ultracentrifugation. The unesterified and esterified cholesterol, triglyceride, phospholipid and protein content for rHDL of similar composition to normal HDL is shown in Table 10. Total protein and lipid recoveries ranged from 58-94% (Table 10). The spherical recombinant particles contained a higher neutral lipid and reduced protein content relative to starting values. The most significant loss of apoprotein occurred as a result of the production of dense, neutral lipid poor particles in the d> 1.21 g/ml range which were removed during reisolation. Compositional analyses of rHDL prepared without neutral lipids is also shown in Table 10. Recoveries of these particles at 1.063-1.21 g/mL was less than for spherical rHDL, ranging between 45% for protein and 65% for unesterified cholesterol. This suggests that about one third of the discoidal preparations had densities greater than 1.21 g/mL. 52 Table 10. Initial and final lipid and protein composition of recombinant spherical and discoidal HDL. Composition (% by weight) Spherical Discoidal-Initial Final Recovery Initial Final T G 4.8 6.1+0.1b 88 FC 4.8 4.2+0.8 73 CE 12.8 18.5+0.4 95 PL 29.8 29.5+1.0 73 Protein 47.9 41.5+1.1 60 3.8 57.7 38.5 A B 4.7 4.7 58.8 55.3 36.5 39.9 HDL apoproteins and the indicated amount of lipids were combined as described in section 2.3.2. The data indicate the initial composition prior to sonication and the final composition after purification of the complexes by ultracentrifugation. -Protein component of preparation A was all HDL apoproteins and preparation B was apoprotein A-I Mean+SD for 3 separate preparations. Size exclusion chromatography of rHDL on Superose 6B prior to ultracentrifugal reisolation is shown in Figure 9. Based on [HJcholesteryl ester and protein (not shown) elution profiles, increased sonication was shown to reduce the relative amount of large particles, which eluted in the void volume (fraction 7), and increase the amount of smaller particles with comparable retention times to native HDL (fraction 13-16). The reduced total radioactivity eluted in panel A was due to retention of the very large particles in the column prefilter. After reisolation, recombinant particles had almost identical [^HJcholesteryl ester elution profiles to native HDL (Figure 10). Protein elution patterns (not shown) for both native and rHDL were also comparable and were directly comparable with cholesteryl ester profiles. Similar reisolated rHDL, prepared with a full complement of HDL apoproteins, were further shown to have an equivalent electrophoretic mobility (alpha) on agarose to native, isolated HDL (Figure 11). Density gradient ultracentrifugation of rHDL containing [^HJcholesteryl ester indicated that the bulk of the microemulsion cholesteryl ester and protein was associated with particles having a similar density range to native HDL, 1.07-1.25 g/mL (Figure 12). A small amount of labeled cholesteryl ester was 53 F A C T I O N NUMBER Figure 9. Size exclusion chromatography of recombinant HDL before ultracentrifugal isolation. Spherical rHDL containing a neutral lipid core (Table 10) and [3H]cholesteryl ester was prepared as described (section 2.3.2) but with only 2, 4, or 6 x 30 sec sonication (sonic) periods. Prior to ultracentrifugal isolation, rHDL (200 uL) was chromatographed on a 10 x 300 mm Superose 6 gel filtration FPLC column at a flow rate of 40 mL/h and with buffer A as the running buffer. For each run 35 x 500 uL fractions were collected and measured for total radioactivity. 54 LU FRACTION NUMBER Figure 10: Size exclusion chromatography of native and recombinant HDL after ultracentrifugal isolation. Spherical rHDL containing a neutral lipid core (Table 10) and [ H]cholesteryl ester tracer was prepared and reisolated by ultracentrifugation. The rHDL (circle) was chromatographed on a Superose 6 gel filtration column which was precalibrated with [ HJcholesteryl ester labeled HDL (diamond) and a mixture of labeled VLDL, LDL and HDL (triangle). Elution profiles of total protein (data not shown) for HDL and rHDL were superimposable with radioactivity profiles. 55 o r i g i n Figure 11: Agarose lipoprotein electrophoresis. Densitometric scans of fresh plasma (A), native (B) and rHDL (C) following electrophoresis in 1% agarose and staining with Fat red 7B/methanol. 56 DENSITY (g /mL) Figure 12: Equilibrium density gradient ultracentrifugation of rHDL rHDL containing a neutral lipid core labeled with [3H]cholesteryl ester was centrifuged in an equilibrium NaCl/sucrose gradient as described in section 2.3.3.1. The contents of each tube were pumped out from the bottom and collected in 10 fractions. The density, protein content and cholesteryl ester radioactivity of each fraction were determined. 57 associated with low density particles (d<1.07 g/mL) but no protein was detectable in this fraction. 3.4.2 Electron microscopy of native and recombinant HDL Electron microscopy of native HDL and rHDL containing a neutral lipid core indicated that the particles were spherical with mean particle sizes of 9.02+0.98 nm and 9.37+1.29 nm, respectively (Figure 13). rHDL prepared from all HDL apoproteins or apo A-I with no neutral lipid core, were observed as stacked discs. The average sizes were: 13.02+2.22 by 4.38+0.45 (apo HDL) and 15.58+3.55 by 4.93±0.68 (apo A-I). 3.4.3 Nuclear magnetic resonance spectroscopy The 200 MHz *H-NMR spectra for rHDL with a normal lipid and apoprotein composition was shown to be practically identical to that for native HDL (Figure 14). The addition of 1 mM MnSO^ was shown to broaden and abolish the N-methyl choline peak at 3.25 p^pm in both native and reconstituted HDL spectra. This indicates the lack of any vesicular structures in both native and reconstituted systems. Their presence would have prevented the complete abolition of the N -methyl choline peaks since vesicles are impermeable to M n + + and therefore prevent the broadening and abolition of the signal associated with the inner phospholipid monolayer (129). Additional evidence for the lack of vesicles in rHDL is shown in Figure 15. rHDL and native HDL have comparable N-methyl choline resonance at approximately 3.25 ppm whereas EYPC vesicles (top spectrum) show two resonances which are broader and chemically shifted, a product of inner and outer phospholipid monolayers. 3.5 INCUBATIONS INVOLVING NATIVE AND RECOMBINANT HDL 3.5.1 Native and recombinant HDL composition and size Total cholesterol, free cholesterol, triglyceride, phospholipid and protein content were determined for native HDL, HDL-,, HDL^ and VHDL (Table 11) and for all reconstituted lipoproteins (Table 12 and 13). Final lipid compositions were comparable to starting concentrations but some differences were noted that seemed to be dependent on both lipids and apoproteins incorporated. Differences in neutral and polar lipid ratios (Series A & B) were 58 N A T I V E H D L R E C O N S T I T U T E D H D L + N E U T R A L L I P I D - N E U T R A L L I P I D - 1 0 ran Figure 13: Negative staining electron microscopy of HDL and rHDL Native and rHDL were dialyzed against a 0.125M ammonium acetate buffer (section 2.3.3.3), mixed with an equal volume of 2% sodium phosphotungstate (pH 7.4) and applied to a Formvar and carbon coated grid (200 mesh). Electron micrographs of native HDL and rHDL, with and without a neutral lipid core, were photographed at an instrument magnification of 77,000 x. Mean particle dimensions of 100 particles were determined from projection of each negative (section 3.4.2). 59 1 1 j : : ! 1— 3 . 0 2 . 0 l . o 3 . 0 2 . 0 1 . 0 p p m f r o m D S S p p m f r o m D S S Figure 14: A H-NMR spectroscopy of HDL and rHDL. Native and rHDL were equilibrated in deuterium oxide, placed in a 5mm NMR tube and the Fourier transformed spectra were obtained with sample spinning using a Bruker WP-200 spectrometer operating at 200 MHz (section 2.3.3.4).*H-NMR spectroscopy was carried out in the presence (lower spectra) and absence (upper spectra) of 1 mM MnSO^. Broadening and abolition of the N-methyl signal at 3.25 ppm after addition of MnSO^ indicates the absence of vesicular structures in both preparations. 60 J j J .—, , J , — 3.50 3.25 3.00 ppm f r o m DSS Figure 15: *H-NMR of the N-methyl choline region of vesicles, HDL and rHDL. The 200 MHz J H - N M R spectra of EYPC-cholesterol (45 mole % cholesterol) vesicles, spherical rHDL and native HDL were obtained as described in section 2.3.3.4. N-methyl choline resonances at 3.25-3.40 ppm are shown. 61 maintained in the final complexes and the protein and lipid recoveries ranged from 50-95%, depending upon the apoproteins/lipid mixture used. The presence of specific apoproteins and normal apoprotein isoforms after sonication and ultracentrifugal re-isolation was confirmed by isoelectric focusing (Figures 16 and 17) and SDS-PAGE. Resolution of the smaller peptides (A-II, C-I, and C-III) on SDS-PAGE was poor and therefore only densitometric scans of the isoelectric gels are shown. The final apoprotein molar ratios in complexes containing mixtures of purified apoproteins were not determined. Particles prepared with purified apoproteins (Table 12, Series C) had similar lipid compositions to rHDL prepared from apo HDL except that rHDL prepared from pure apo A-I had a reduced cholesteryl ester content compared to those prepared from apo HDL (Series A, batch II). In addition, as the apo A-II content was increased in rHDL, the relative recovery of cholesteryl ester in the rHDL also increased (Series C). Negatively stained micrographs of rHDL with different lipid and protein compositions are shown in Figure 18 and 19. Particles prepared with variations in neutral or polar lipid composition showed significant alterations in particle size (Table 13, series A and B). Those with a purely triglyceride or cholesteryl ester core were significantly (p<0.001) smaller than rHDL with both neutral lipids or native HDL (Table 13, series A). Furthermore, increases in rHDL free cholesterol relative to phospholipid resulted in increased particle heterogeneity and an increased mean particle size (series B). rHDL prepared with a neutral lipid core and either purified apo A-I or apo A-II, had comparable mean particle sizes, 8.7+1.9 nm and 8.3+1.6 nm respectively, and rHDL prepared from mixtures of apo A-I and apo A-II also showed similar mean particle size (series C). Table 11. Protein and lipid composition of native HDL. Composition (% by weight) unesterified cholesterol cholesteryl ester triglyceride phospholipid protein H D L H D L -H D L : VHD 2.1 3.0 3.2 1.1 20.1 23.9 21.2 11.9 2.0 3.6 2.9 2.3 24.0 30.7 26.7 22.7 51.8 38.8 46.1 61.9 Values are the average of duplicate determinations. 62 A C - l Figure 16. Isolelectric focusing of native H D L 2 , H D L 3 and VHDL apoproteins. Ultracentrifugally isolated HDL subfractions were delipidated in ethanol:diethyl ether (3:2) and the apoproteins were thoroughly dried under N 2 (section 2.3.1). Isolelectric focusing of apoproteins (200 ug) was performed on 7.5% acrylamide tube gels (5mm) containing pH 4-6 ampholytes. After focusing for 16 h at 450 volts, gels were stained with Coumassie Blue G250 in 3.5% perchloric acid and destained with 7.5% acetic acid. Densitometer scans of H D L 2 (A), HDL, (B), and VHDL (C) are shown. Apoprotein labels indicate the approximate vicinity of the major isoforms. 63 Figure 17. Isoelectric focusing of recombinant HDL apoproteins. Ultracentrifugally isolated rHDL prepared from mixtures of purified apoproteins were delipidated in ethanol:diethyl ether (3:2) and dried under N 2 . Isolelectric focusing of apoproteins (200 ug) was performed on 7.5% acrylamide tube gels as described in section 2.2.2.3. Densitometer scans of rHDL with the following apoprotein compositions (initial molar ratios) are shown: A) apoprotein A-I, B) apoprotein A-I + C-l (3:1), C) apoprotein A-I + C-III (3:1), D) apoprotein A-I + E (6:1), E) apoprotein A-I + A-II (1:1), and F) apoprotein A-I + A-II (3:1). Arrows indicate the location of additional apoproteins. 64 A B Figure 18. Negative stain electron microscopy of rHDL with variations in core neutral lipids. Electron micrographs of rHDL prepared from apo HDL with various cholesteryl ester to triglyceride molar ratios: 4:1 (A), 0:1 (B), 1:0 (C) and 1:1 (D) are shown. Instrument magnification was 77,000. Mean particle dimensions of 100 particles were determined from projection of each negative, as described in section 2.3.3.3. A B Figure 19. Negative stain electron microscopy of rHDL with variations in lipids and apoproteins.. Electron micrographs of rHDL prepared from purified apoprotein A-I (A), apo A-I and A-II (B), apo A-II (C), and apo HDL with a molar free cholesterol to phospholipid ratio of 1:0.6 (D) are shown. Instrument magnification was 77,000. Mean particle dimensions of 100 particles were determined from projection of each negative. 66 Table 12. Initial lipid and apoprotein composition of recombinant HDL. Composition (% by weight) Apoproteins Series Batch FC CE T G PL Protein Present \ B I 4.6 17.4 0.0 30.0 48.0 II 4.6 12.8 4.6 30.0 48.0 III 4.6 4.6 12.8 30.0 48.0 IV 4.6 0.0 17.4 30.0 48.0 I 6.5 12.8 4.7 28.0 48.0 II 8.7 12.8 4.7 25.8 48.0 III 11.5 12.8 4.7 23.0 48.0 IV 17.3 12.8 4.7 17.2 48.0 I 4.6 12.8 4.6 30.0 48.0 II 4.6 12.8 4.6 30.0 48.0 III 4.6 12.8 4.6 30.0 48.0 IV 4.6 12.8 4.6 30.0 48.0 V - 4.6 12.8 4.6 30.0 48.0 VI 4.6 12.8 4.6 30.0 48.8 VII 4.6 12.8 4.6 30.0 48.8 Molar ratio = ^Molar ratio = ^Molar ratio = ^Molar ratio = Molar ratio = Apo HDL Apo HDL Apo HDL Apo HDL Apo HDL Apo HDL Apo HDL Apo HDL A-I A-II , A-I/A-II* A-I /AjII z A - I / E 3 A - I / C - I 4 _ A-I/C-III 5 Table 13. Final lipid and apoprotein composition and size of recombinant HDL Composition (% by weight) Apoproteins Particle Batch FC CE TG PL Protein Present Size (nm)1 A I 4.1 20.2 0.0 34.7 41.0 Apo HDL 8.55+0.88 II 3.2 18.2 6.2 29.7 42.6 Apo HDL 9.37+1.29 III 3.3 11.7 15.5 29.1 40.5 Apo HDL 10.52+1.74 IV 6.0 0.0 17.7 33.0 40.5 Apo HDL 7.25+1.00 B I 4.5 19.0 6.2 28.3 41.9 Apo HDL 8.66+1.24 II 6.0 20.9 5.7 22.2 45.3 Apo HDL 8.69+1.77 III 8.5 20.4 6.5 20.3 44.3 Apo HDL 9.18+2.07 IV 12.3 18.9 13.2 15.0 40.5 Apo HDL 10.03±3.07 C I 3.8 12.5 6.7 32.7 44.3 A-I 8.69+1.94 II 1.5 20.9 8.0 27.5 42.2 A-II 8.27+1.58 III 4.3 16.1 5.1 28.0 46.6 A-I/A-II 9.06+1.55 IV 1.6 21.6 5.4 30.8 40.5 A-I/A-II 8.44+1.77 V 4.8 16.9 6.0 29.4 42.9 A-I /E not determined VI 4.6 17.7 6.5 25.8 45.4 A-I/C-I not determined VII 5.1 18.8 6.5 25.7 43.9 A-I/C-III not determined aDetermined by negative staining electron microscopy. Values are the mean+lSD of 100 particles per micrograph. 67 3.5.2 Native and recombinant HDL as acceptors of cholesteryl ester CETP mediated transfer of cholesteryl ester from LDL to the native HDL was a saturable process (Figure 20) and therefore the reaction kinetics were estimated by double reciprocal and direct linear plots (124) in an attempt to numerically characterize variations in the ability of different HDL to accept cholesteryl ester from LDL. Values, analogous to V m a x and K m , represent the maximum velocity of transfer, T „ „ , and the HDL concentration required for half maximal activity, Kpj. Reaction kinetics were calculated as an average of estimations on double reciprocal and direct linear plots as described in section 2.8.2. Determination of the reaction kinetics for the total HDL fraction indicated an apparent Kpj of 17.0 ug protein/mL and T m a x of 10.0 % transferred/mL/h. rHDL of comparable apoprotein and lipid composition to native HDL were shown to be functionally equivalent as a substrate for CETP (Figure 20), with almost identical reaction kinetics to native particles. Cholesteryl ester transfer to discoidal rHDL was significantly reduced when compared to native and spherical rHDL. Lipid transfer to apo HDL discs was undetectable while transfer to discs containing only apo A-I was not saturated and was less than 30% of that for native HDL (Figure 20). Cholesteryl ester transfer to native HDL subfractions is shown in Figures 21 and 22. Lipid transfer rates determined in the absence of exogenously added CETP (Figure 21) showed that a significant amount of endogenous CETP activity was associated with HDL^ and to a lesser extent, VHDL. Conversely, endogenous cholesteryl ester transfer activity was markedly reduced in the total HDL and H D L 2 subfractions. Lipid transfer rates to native HDL subfractions, determined in the presence of exogenous CETP, is shown in Figure 22, and as a double reciprocal plot in Appendix A, Figure 14. CETP mediated lipid transfer to the total HDL fraction was saturable with an apparent K J J of 17.0 ug protein/mL and T m a x of 10.0 %/mL/h (Table 14). Reaction kinetics for HDL^ demonstrated a similar K J J to for total HDL but an increased T M A X . At high substrate concentrations of HDL^, there was an apparent substrate inhibition of the reaction catalyzed by CETP (Figure 22). This effect was not observed with H D L 7 68 - HDL rHDL CONCENTRATION (ug prote in /mL) Figure 20. rHDL and HDL as substrates for CETP. The transfer of [3H]cholesteryl ester from labeled LDL to the indicated amounts of HDL or spherical rHDL (Table 10) is shown. Mixtures of partially purified CETP (2.6 ug), LDL (30 ug "total cholesterol), BSA (1 mg) and HDL or rHDL were incubated at 37°C for 1.5 h and the lipid transfer was determined as described section 2.4.1.1. Values are the mean+SD of quadruplicate determinations. In addition, the ability of discoidal rHDL, prepared from HDL apoproteins or pure apo A-I, to accept [3H]cholesteryl ester was determined. The results represent means of duplicate incubations. 69 HDL CONCENTRATION (ug prote in/mL) Figure 21 . Cholesteryl ester transfer to HDL subfractions in the absence of exogenously added CETP. The transfer of [3H]cholesteryl ester from labeled LDL to different HDL subfractions is shown. HDL subfraction compositions are shown in Table 11. Mixtures containing LDL (30 ug total cholesterol), BSA (1 mg) and the indicated amounts of HDL, were incubated at 37°C for 1.5 h and the lipid transfer was determined as described in section 2.4.1.1. Values are the average of duplicate determinations. 70 HDL CONCENTRATION (ug pro te in /mL) Figure 22. Cholesteryl ester transfer to HDL subfractions. The transfer of 3 [ HJcholesteryl ester from labeled LDL to different HDL subfractions is shown. HDL subfraction compositions are shown in Table 11. Mixtures containing partially purified CETP (1.3 ug), LDL (30 ug total cholesterol), BSA (1 mg) and the indicated amounts of HDL were incubated at 37°C for 1.5 h and the lipid transfer was determined as described in section 2.4.1.1. Values are the mean+SD of quadruplicate determinations. 71 and VHDL; however, both substrates demonstrated a slightly increased T m . v and a 1X1 d.X marked increase in apparent K H relative to total HDL (Table 14 and Appendix A, Figure 14). Table 14. Kinetic parameters of native HDL as a substrate for CETP. molar ratios K H a T m a x a Fraction C E / T G FC/PL (ug/mL) (%/mL/h) HDL 9:1 1:4 17.0 10.0 HDL2 11:1 1:3 60.7 11.4 HDL3 9:1 1:5 19.9 14.0 VHDL 7:1 1:6 68.3 12.1 aValues are the average of estimations from double reciprocal and direct linear plots, as explained in section 2.8.2. 3.5.3 Effect of rHDL lipid composition on lipid transfer rHDL with a cholesteryl ester rich core and comparable apoprotein composition to native HDL had similar reaction kinetics with CETP to native particles (Table 14 and 15). Increases in the triglyceride content, relative to cholesteryl ester, in particles with all HDL apoproteins resulted in significant decreases in the ability of these particles to accept cholesteryl ester (Figure 23), as indicated by a reduction in the reaction T m a x (Table 15 and Appendix A, Figure 15A). This was also observed when the free cholesterol content was increased relative to phospholipid (Figure 24; Table 15). Regardless of changes in lipid composition, all particles that were prepared with an identical apoprotein content to native HDL, had a similar apparent Kpj for the reaction with CETP as did native HDL (Table 15 and Appendix A, Figure 15B). Regression analysis identified a significant inverse relationship between the CETP reaction Tmax a n c * t l i e r r I D L relative triglyceride content (-0.965, Figure 25A) and relative free cholesterol content (-0.988, Figure 25B). 72 EFFECT OF CE:TG IN rHDL ON CETA rHDL CONCENTRATION (ug p r o t e i n / m L ) Figure 23: Effect of rHDL neutral lipid composition on cholesteryl ester transfer. The transfer of [3H]cholesteryl ester from LDL to rHDL is shown. rHDL were prepared, as described in section 2.3.2, from a full complement of HDL apoproteins and purified lipids, with the indicated cholesteryl ester to triglyceride molar ratio. rHDL final compositions are shown in Table 13, Series A. Incubations were as described for Figure 22 with values being the mean+lSD of quadruplicate determinations. 73 EFFECT OF FC:PL IN rHDL ON CETA 10-rHDL CONCENTRATION (ug p r o t e i n / m L ) Figure 24: Effect of rHDL polar lipid composition on cholesteryl ester transfer. The transfer of [HJcholesteryl ester from LDL to rHDL is shown. rHDL were prepared as described in section 2.3.2 from a full complement of HDL apoproteins and purified lipids, with the indicated free cholesterol to phospholipid molar ratio. rHDL compositions are shown in Table 13, Series B. Incubations were as described for Figure 22 with values being the mean+lSD of quadruplicate determinations. 74 RELATIVE TRIGLYCERIDE CONTENT (% by weight) RELATIVE FREE CHOLESTEROL CONTENT (% by weight ) Figure 25: Effect of rHDL lipid composition on CETP reaction kinetics. T m o v in 3.x values were estimated as described in section 2.8.2. from the data of the incubations described in Figure 23 and 24 and are plotted against rHDL triglyceride (A) and free cholesterol (B) contents (shown in Table 13, series A and B). Correlation coefficients (r) were determined by regression analysis. 75 Table 15. Effect of lipids on the kinetic parameters of rHDL a as a substrate for CETP. Apoproteins molar ratios ^max^ Series Batch present C E / T G FC/PL (ug/mL) (%/mL/h) A I apo HDL 1:0 1:4.2 19.6 10.7 II apo HDL 4:1 1:4.6 16.1 8.9 III apo HDL 1:1 1:4.4 19.0 7.4 IV apo HDL 0:1 1:2.7 16.2 5.4 B I apo HDL 4.2:1 1:4.6 16.1 9.8 II apo HDL 5.1:1 1:1.9 19.7 8.1 III apo HDL 4.3:1 1:1.2 21.4 6.3 IV apo HDL 2.0:1. 1:0.6 19.0 4.3 a rHDL prepared as described in Table 13. Values are the average of esti plots, as explained in section 2.8.2. mations from double reciprocal and direct linear 3.5.4 Effect of rHDL apoprotein composition on lipid transfer rHDL prepared with purified apo A-I only, had a significantly higher Kj_£ and T m a x for the reaction with CETP than did particles made with the full complement of HDL apoproteins or native HDL (Table 16). Since the ratio of T m a x / K j j was the similar for both apo HDL and apo A-I rHDL (0.3) with similar lipid compositions, it seems possible that some other component of the apo HDL was uncompetitive inhibitor (124), equally reducing both the T m a x and K g in the interaction between rHDL with a full complement of apoproteins and CETP. Interestingly, at high substrate concentrations (greater than 100 ug protein/mL), there was an apparent substrate inhibition in the interaction between apo A-I rHDL and CETP, which was similar to that for HDL^ (Figure 26). Particles prepared with only purified apoprotein A-II also demonstrated this apparent substrate inhibition, with a comparable T m a x to A-I rHDL but a reduced K H (Figure 26; Table 16). The ability of rHDL with different mixtures of purified apoproteins to receive cholesteryl esters is shown in Figure 27. rHDL prepared from apo A-I and C-I had a comparable T m a x to apo A-I rHDL but demonstrated a markedly elevated Kpj for the interaction with CETP (Table 16). When apo E was combined with apo A -I, both K J J and T m a x values were reduced relative to those for particles containing only apo A-I. Conversely, rHDL prepared from a combination of 76 0- f 1 1 1 — — i 1 1 > 1 ' i ' 0 50 100 150 200 250 300 rHDL CONCENTRATION (ug protein/mL) Figure 26: Comparison of apoprotein A-I and A-II containing rHDL as substrates for CETP. [3H]choIesteryl ester transfer from LDL to rHDL prepared from purified apoprotein A-I, apo A-II, or delipidated HDL apoproteins is shown. Particle compositions and size are shown in Table 13. Incubations were as described in Figure 22 and values are the mean+lSD of quadruplicate determinations. 77 0 - i , 1 , 1 • { . 1 • i 1 0 50 100 150 200 250 300 r H D L CONCENTRATION (ug p r o t e i n / m L ) ' Figure 27: Effect of rHDL apoprotein content on cholesteryl ester transfer to rHDL. The transfer of [JH]cholesteryl ester from LDL to rHDL, prepared with the indicated mixtures of apoproteins is shown. rHDL compositions and size are shown in Table 13. Incubations are as described in Figure 22 and values are the means of quadruplicate determinations. 78 apoproteins A-I and A-II (initial molar ratio 3:1) were shown to have elevated T m a x and K H but reduced allosteric inhibition, relative to A-I rHDL (Figure 27 and Table 16). Increased A-II content (initial molar ratio of A-I to A-II 1:1) in recombinant particles was shown to further increase K J J and T M A X values for the interaction with CETP (Table 16). Conversely, particles prepared from apoproteins A-I and C-III were shown to have a comparable T m a x but a markedly reduced K J J for their interaction with CETP (Table i6 and Appendix A, Figure 16). Interestingly, inclusion of apo C-III resulted in almost an identical K J J as for particles containing the full compliment of HDL apoproteins (Table 16), which suggests that apoprotein C-III may play a major role in the regulation of the interaction of CETP and apo C-III containing particles. Table 16. Effect of apoproteins on the kinetic parameters of rHDL as a substrate for CETP. Apoproteins molar ratios T max Series Batch present C E / T G FC/PL (ug/mL) (%/mL/ C I A-I 2.5:1 1:4.3 57.9 17.8 II A-II 3.6:1 1:9.0 35.7 18.0 III A-I/A-II 4.4:1 1:3.2 107.1 18.3 IV . A-I/A-II 5.5:1 1:9.3 164.3 19.1 V A-I /E 3.8:1 1:3.1 36.1 15.6 VI A-I/C-I 3.7:1 1:2.8 72.3 18.2 VII A-I/C-III 3.9:1 1:2.5 19.0 17.6 aValues are the average of estimations from double reciprocal and direct linear plots, as explained in section 2.8.2. Estimated directly from Figure 26 and 27. 3.5.5 Effect of rHDL lipid composition on LCAT Comparison of recombinant and native HDL as substrates for L C A T also showed a marked similarity in their ability to interact with this enzyme. rHDL with a comparable lipid and protein composition to native particles were almost identical as substrates for LCAT (Figure 28). However, recombinant particles that were prepared without a neutral lipid core (discoidal) were markedly better substrates for LCAT than were either native HDL or rHDL containing a neutral lipid core (Figure 28). This is comparable to observations by other investigators (130). 79 rHDL CONCENTRATION ( n m o l c h o l / m L ) Figure 28: rHDL and HDL as substrates for LCAT. [-^cholesterol labeled rHDL with and without a cholesteryl ester/triglyceride core were compared with native HDL as substrates for LCAT. Mixtures containing native or rHDL (25-300 nmol free cholesterol/mL), purified LCAT (4.0 ug/mL), BSA (17 mg/mL), beta-mercaptoethanol (4.2 mM), and buffer A (in a total of 100 uL), were incubated for lh at 37°C. The results for native and spherical rHDL are the mean+1/2 range of triplicate determinations from two separate preparations while the data for discoidal rHDL is the average of duplicate determinations. 80 rHDL prepared with a lipid composition comparable to native HDL was also shown to have similar kinetics in their reaction with LCAT (Table 17). Reaction kinetics were estimated as described in section 2.8.2. Increasing the triglyceride content, relative to cholesteryl ester in rHDL, markedly impaired the ability of LCAT to catalyze the esterification of cholesterol in rHDL (Figure 29), as indicated by a reduction in the reaction V m a x (Table 17). Interestingly, a change in the core composition of rHDL also markedly affected the K m (for cholesterol) of the interaction with LCAT. Particles that contained both cholesteryl ester and triglyceride in their core had substantially higher K m values (Table 17) than those with a single neutral lipid. Regression analysis showed that K m values were significantly related to protein (r=0.93), phospholipid (r=-0.99) and unesterified cholesterol (r=-0.96) content and particle size (r=0.96, Figure 30); however, there was no relationship between K m and neutral lipid content. Conversely, V m a x values were not related to size, surface lipid or protein content but were significantly related to triglyceride (Figure 30) and cholesteryl ester contents (r=-0.96 and 0.80 respectively). Table 17. Effect of rHDL size and lipid content on the kinetic parameters of rHDL as a substrate for LCAT. Particle molar ratios Sample Batch Size (nm)a C E / T G FC/PL K m b V ' max HDL 9.02+0.98 9:1 1:4.2 88 22 rHDL I 8.55+0.88 1:0 1:3.4 69 - 47 II 9.37+1.29 4:1 1:3.2 125 31 III 10.52+1.74 1:1 1:3.2 146 22 IV 7.25+1.00 0:1 1:3.9 51 19 Kinetic values are the average of estimations from double reciprocal and direct linear plots, as explained in section 2.8.2. ^Determined by negative staining electron microscopy as described in section 2.3.3.3. (nmol unesterified cholesterol/mL) c(nmol cholesterol esterified/mL/h) 81 rHDL UNESTERIFIED CHOLESTEROL ( U M ) Figure 29: Effect of rHDL lipid composition on LCAT activity. The esterification of [ HJcholesterol of rHDL by LCAT is shown. rHDL were prepared as described in section 2.3.2 from a full complement of HDL apoproteins and purified lipids, with the indicated cholesteryl ester to triglyceride molar ratios. rHDL compositions are shown in Table 13. Incubations were as described in Figure 28 with values being the mean+1/2 range of triplicate determinations from two separate preparations. 82 7 8 9 10 11 0 4 8 12 16 20 PARTICLE SIZE (nm) RELATIVE TRIGLYCERIDE CONTENT (% by weight) Figure 30: The relationship between rHDL composition, size and L C A T reaction kinetics. L C A T reaction K m and V " m a x values were estimated as described insection 2.8.2 from data of the incubations described in Figure 29 and are plotted against rHDL particle size (A) and triglyceride (B) content (Table 13, series A). The correlation coefficients (r) shown were determined by linear regression analysis. ' 83 4.0 DISCUSSION 4.1 T H E DETERMINATION OF CETP ACTIVITY 4.1.1 The measurement of CETP activity in plasma The major objective of the first part of this project was to develop a novel assay which could be used to measure the ability of endogenous plasma lipoproteins, in a patient's plasma sample, to receive cholesteryl esters transferred from an exogenous HDL-donor. The overall objective was to determine how variations in endogenous HDL composition would affect its ability to act as an acceptor of transferred cholesteryl esters. Characterization of this assay showed that [ Hjcholesteryl ester was transferred from Sepharose-bound HDL to plasma lipoproteins with approximately equal distribution to HDL and VLDL+LDL. In normolipidemic plasma, lipid transfer was shown to be independent of acceptor lipoproteins, since addition of VLDL, LDL or HDL to the plasma before incubation with the immobilized HDL did not affect the total amount of label transferred or the amount recovered in the different lipoprotein classes. This suggested that the acceptor particle was not rate limiting in the lipid transfer measured and that some other plasma factor regulates the process. Transfer activity was, however, dependent on the amount of plasma present, and may therefore have been associated with increases in the amount of lipid transfer proteins present. In fact, the addition of partly pure CETP to plasma prior to incubation markedly increased transfer rates. This suggests that CETP was responsible for a substantial amount of transfer of cholesteryl ester in this system. When a mixture of VLDL, LDL and HDL was substituted for whole plasma as acceptor, the proportion of label recovered in HDL was greatly decreased, even though the label recovered in L D L / V L D L was only slightly reduced. This suggests that CETP may transfer lipids to HDL in preference to the lower density lipoproteins. Since it was also observed that 50% of the total transfer may occur in the absence of the d>1.21 g/mL fraction, contaminating transfer proteins may have been present in one or more of the purified lipoproteins. In fact, subsequent studies (Figure 34) indicated that H D L 3 retained a significant 84 amount of lipid transfer activity even after extensive ultracentrifugation. Therefore, the high rates of lipid transfer to the L D L / V L D L fraction in the absence of the d>1.21 g/mL fraction suggest that at very low levels of CETP, cholesteryl esters are preferentially transferred to lower density lipoproteins. However, whether this is through a protein or non-protein mediated mechanism is unclear. Barter et al (126) have made similar observations in which they showed that only 30% of L C A T derived cholesteryl esters incorporated in LDL were dependent on the action of CETP. As with my studies, however, the studies of Barter et al did not take into account endogenous CETP that may have been associated with the purified lipoproteins. To resolve this question of whether this transfer is protein or non-protein mediated will require studies with lipoprotein substrates that are completely devoid of endogenous transfer activities. The recovery of considerable amounts of label in the HDL fraction indicates that cholesteryl ester may be transferred to this lipoprotein with the same or better efficiency as to the lower density lipoproteins. Morton (69) has also shown HDL to be a major substrate of CETP. He demonstrated that binding of CETP to lipoproteins was required for lipid transfer and that although all lipoproteins bind CETP with the same high affinity, the complex formed with HDL is much more stable than that with other lipoprotein classes. While Morton and others have shown HDL to be an important donor substrate of CETP (51,52,53), the results of this study clearly show that HDL also may be a effective acceptor of cholesteryl ester. Our observations may reflect the potential for cholesteryl ester to be transferred from the HDL subclass in which it is synthesized, to other HDL subclasses prior to subsequent transfer to VLDL or LDL. Presumably, the absolute rates of transfer of [HJcholesteryl ester to the plasma lipoproteins could be affected by changes in the specific radioactivity of the donor cholesteryl ester pool, for example, by endogenous synthesis of cholesteryl ester by LCAT. However, carrying out the incubations in the presence of diisopropyl fluorophosphate indicated that inhibiting this enzyme had no effect on the rate of transfer from the HDL-Sepharose to plasma HDL or 85 VLDL+LDL, for incubations of 60 min or less. In addition, characterizations of the effects of plasma storage on lipid transfer showed that while storage of plasma at 4°C resulted in progressive reductions in the measured lipid transfer activities, this effect could be minimized by freezing the plasma at -20°C. Comparison of lipid transfer rates determined in plasma and LPDP of normolipidemic subjects showed a significant correlation between the two assays. This indicates that the differences noted in lipid transfer rates may not have been due to changes in lipoprotein composition and/or concentration, but may instead have been related to changes in mass or degree of inhibition of CETP. Accordingly, it is proposed that the transfer of [ Hjcholesteryl ester from solid-phase-bound HDL to plasma lipoproteins may be a viable method to estimate the rate of cholesteryl ester transport in intact plasma samples. Any attempt to correlate these activities with CETP mass, however, would be difficult, as net neutral lipid transfer activity has been shown to be a result of both the activity of CETP and the effect of specific inhibitor proteins (73,75,76). Determination of cholesteryl ester transfer rates in patient's whole plasma may therefore be more appropriate than in LPDP since ultracentrifugation may disassociate CETP from the" plasma lipoproteins to a different extent than its inhibitors and subsequent determination of lipid transfer activities may be biased. In summary, a rapid reproducible assay for the determination of cholesteryl ester transfer rates from immobilized HDL to the plasma lipoproteins has been developed. The assay is not affected by endogenous L C A T activity and works equally well with fresh or frozen plasma samples. This method allows for the estimation cholesteryl ester equilibration in samples of whole plasma without the need for prior removal of the lipoproteins by ultracentrifugation. 4.1.2 CETP activity in patients with severe HDL deficiency This study identified a significant correlation between assays of cholesteryl ester transfer in unaltered plasma and LPDP of HDL-deficient patients. As with normolipidemic subjects, this suggests that the differences noted in the rate of lipid transfer in these patients may not have been due to 86 variations in lipoprotein composition and/or concentration, but may have been related to changes in mass or degree of inhibition of CETP. Results from this study further showed that cholesteryl ester transfer activity may differ amongst patients with severe HDL-deficiency. Decreased activity in LCAT deficiency and FED and the increased activity in Tangier disease indicate that the level of HDL does not directly affect the total cholesteryl ester transfer rates in these individuals. A possible relationship between L C A T activity and cholesteryl ester transfer may exist since patients with FED and L C A T deficiency also have a markedly reduced L C A T activity. This may result in a decreased requirement for the transport of cholesteryl esters and a subsequent decreased transfer activity. There is one report of normal cholesteryl ester transfer activity in two patients with FED (131). These observations, however, cannot be directly compared to this investigation since they used the d>1.25 g/mL plasma fraction as a source of cholesteryl ester transfer activity and it is well established that a significant portion of plasma CETP is recovered at 1.21-1.25 g/mL (53). Thus, it seems likely that the transfer rates reported by Calvert and Carlson (131) were not representative of the total plasma activity in their patients. Our laboratory has also studied cholesteryl ester synthesis and transport in L C A T deficient whole plasma to which purified L C A T was added (128). We determined that the net transfer of newly synthesized cholesteryl ester was impaired and esters accumulated in HDL suggesting that HDL in L C A T deficiency may be unable to release cholesteryl ester to other lipoproteins. This indicates that the reduced transfer activity found in LCAT-deficient plasma may also have been partly due to changes in HDL composition. Concomitant reductions in transfer activity determined in LPDP, however, suggests that protein mass may indeed be reduced. In fact, some studies have indicated that low levels of transfer may result in hyperalphalipoproteinemia (93,94). Tangier disease, on the other hand, is associated with only slightly reduced L C A T activity despite the HDL deficiency. Thus, the increased transfer activity may reflect the need to transport newly synthesized cholesteryl ester 87 to VLDL and LDL since there is no pool of HDL to act as an acceptor. This suggests that the changes in CETP activity in severe hypoalphalipoproteinemias may only be adaptive mechanisms that are developed to maintain effective reverse cholesterol transport. Conversely, it may also be possible that the increased activity may play a more central role in the HDL deficiency itself, however, a mechanism by which this is brought about cannot be elucidated by the present study. 4.1.3 The relationship between HDL composition and CETP activity in patients with disorders of lipoprotein metabolism. HDL has been shown to be an excellent donor as well as an acceptor of cholesteryl ester (65,67,68). Thus, it is proposed that a major function of CETP may be to promote the equilibration of cholesteryl esters within the HDL subclasses in addition to catalyzing transfer to VLDL and LDL. Impairment of this equilibration of cholesteryl esters within the HDL pool may be related to reduced HDL cholesterol levels, since approximately 75% of HDL cholesterol content is cholesteryl ester. This is significant since many studies have demonstrated a relationship between low levels of HDL-cholesterol and the incidence of atherosclerosis (1,2,14,16). This relationship was confirmed in the present study in which decreased levels of HDL2-cholesterol were correlated with the presence of vascular disease. Several studies have shown that changes in the lipid composition of HDL may be independently associated with risk of heart disease. For example, decreased HDL-unesterified cholesterol (91), H D L -phospholipid (132) and increased HDL-triglyceride (92) have all been positively correlated with premature cardiovascular disease. This study has identified a significant correlation between the changes in HDL composition, in patients with disorders of lipoprotein metabolism, and an altered ability of this lipoprotein class to function as an acceptor of cholesteryl esters. In studies where cholesterol ester transfer activity was determined in the plasma and LPDP of normolipidemic and severely HDL deficient patients, the strong correlation between assays observed suggested that lipid transfer rates may not be affected by the presence or absence of endogenous lipoproteins in 88 these subjects. This is strong evidence that this measurement of lipid transfer activity in plasma samples is indeed representative of plasma cholesteryl ester transfer activity. However, since no significant relationship was shown to exist between cholesteryl ester transfer activity measured in the presence and absence of endogenous lipoproteins in hyperlipidemic patients, it seems that the composition of the lipoproteins, or some factor associated with the lipoproteins, may modify the rate of lipid transfer in these subjects. Interestingly, this lack of a relationship between assays seemed to be partly a result of markedly elevated transfer rates in the LPDP when the observed rate in plasma was low. This suggests that some component associated with the lipoproteins in these patients, that could not be disassociated by ultracentrifugation, may have had an inhibitory effect on CETP. It further suggests that measurement of lipid transfer activity in intact plasma may more closely depict the actual activity of CETP in plasma since determinations with the LPDP samples may be partially or completely devoid of inhibitors of CETP. Cholesteryl ester transfer to the total plasma lipoprotein pool was significantly reduced in patients with hypercholesterolemia, familial combined hyperlipidemia and in subjects with hypoalphalipoproteinemia and hypertriglyceridemia. However, the magnitude of these changes was small and many of the dyslipidemic subjects in this study had normal total lipid transfer rates. This measurement represents the sum of the transfer of cholesteryl ester to both the HDL and V L D L / L D L pools. Thus, analysis of the transfer to individual classes of lipoproteins may provide more meaningful data. Transfer to L D L / V L D L was significantly increased in patients with dysbetalipoproteinemia which agrees with a recent report by Tall et al (66) . Of more significance, however, is the observation that there was reduced cholesteryl ester transfer in all subjects except those with familial hypertriglyceridemia. This appeared to be due to a reduction in transfer to H D L 3 while transfer to H D L 2 was relatively unaffected. In subsequent studies with recombinant HDL particles, it was demonstrated that changes in neutral lipid core composition will affect the ability of the HDL analog to act as a substrate for the transfer protein. Furthermore, other 89 investigators have shown that a change in surface charge of an HDL particle directly affects the ability of , CETP to bind and subsequently catalyze the transfer of lipid (70,71). Thus, changes in HDL composition may be integrally related to variations in the rates of cholesteryl ester transfer to these particles. While total HDL mass was evidently an important factor affecting the total cholesteryl ester transfer to HDL, it is interesting to note that normal HDL lipid mass and apo A-I levels were observed in the three diseases with significantly reduced cholesteryl ester transfer rates to HDL (ie. familial hypercholesterolemia, familial combined hyperlipidemia and dysbetalipoproteinemia). This suggests that total HDL mass was not the only factor affecting lipid transfer to HDL. Since HDL-triglyceride levels were significantly elevated in all of these groups, the composition of HDL may be as important as the total mass in the regulation of. reverse cholesterol transport. Furthermore, since patients with documented evidence of vascular disease were shown to have significantly reduced cholesteryl ester transfer to H D L 3 , impaired cholesteryl ester transfer which results from abnormal HDL composition may be indirectly involved in the development of atherosclerosis. In this study, it was demonstrated that significant deviations from normal HDL composition are found in dyslipidemic patients. The major change was shown to be a decreased ratio of cholesteryl ester to triglyceride. Since almost all groups demonstrated significantly reduced HDL^-cholesterol, it is proposed that this change in core neutral lipid ratio may have been associated with HDL^ subfraction. Two recent studies have demonstrated marked elevations of HDL-triglyceride during postprandial lipemia (133,134) and in fact showed that the triglyceride content in both H D L 3 and H D L 2 was positively correlated with the magnitude of lipemia (23). These findings support the hypothesis that the neutral lipid core of H D L 3 may be modified in patients with plasma lipid abnormalities. Only one group in this study, however, showed significant changes in H D L 2 composition: hypoalphalipoproteinemia with hypertriglyceridemia. This group had significantly reduced HDL2-cholesterol and interestingly, also / 90 contained 60% of the patients with vascular disease. These findings are consistent with other investigator's observations (82,110) and support the widely accepted view that HDL^-cholesterol levels may be a valuable risk indicator for vascular disease. The cause of elevated HDL-triglyceride as a result of hypertriglyceridemia has been extensively reviewed (134,82). One widely accepted hypothesis suggests that H D L 3 accumulates the breakdown products of the catabolism of triglyceride-rich particles and form mature H D L 2 particles, when lipoprotein lipase activity is high. Conversely, when lipase activity is low, HDL may accumulate triglyceride via CETP mediated transfer from chylomicrons. Both hypertriglyceridemic H D L 2 and H D L 3 have been shown to be highly susceptible to catabolism by hepatic lipase (134) and as such, it has been postulated that, under normal circumstances, hepatic lipase is operational in the reconversion of hypertriglyceridemic H D L 2 and H D L 3 back to smaller H D L 3 particles. Accordingly, hepatic lipase very likely plays a central role in the regulation of HDL triglyceride content. In fact, a deficiency of this enzyme has been shown (135.136) to result in the accumulation of triglycerides in HDL. Conversely, the involvement of CETP in this interconversion of HDL particles is widely accepted (134.137) , but whether or not CETP could be responsible for T G accumulation in HDL has been questioned. Patsch et al (134) have suggested that it is unlikely that individual differences in lipid transfer activity could be responsible for differences in HDL-triglyceride. This conclusion was based upon their observation of a strong correlation between the magnitude of postprandial lipemia (hypertriglyceridemia) and the abundance of triglyceride in HDL, which suggests that the enrichment of triglyceride may be governed primarily by the concentration of donor and acceptor lipoproteins. In vitro studies by Hopkins and Barter (137), on the other hand, have demonstrated an integral role for CETP in the formation of hypertriglyceridemic HDL and have further suggested that in hypertriglyceridemia there must be a dynamic balance between the action of CETP in the formation of hypertriglyceridemic HDL, and its subsequent catabolism by hepatic lipase. In the present study, two patient groups (familial 91 hypercholesterolemia and hypoalphalipoproteinemia) had significantly elevated HDL-triglyceride levels despite normal plasma triglyceride levels. Clearly, some other factor as well as hypertriglyceridemia must be contributing to the accumulation of triglyceride in HDL. Hepatic clearance of cholesteryl esters derived from HDL has been thought to be dependent upon the movement of cholesteryl esters to lower density lipoproteins for several reasons (18). Firstly, LDL and V L D L contain approximately 70% of the plasma esterified cholesterol. Secondly, they have been shown to be efficient acceptors of LCAT-derived cholesterol esters transferred by CETP in vitro, and thirdly, the plasma turnover rates of their apoproteins are higher than for HDL. Studies by Glass et al (13,138), however, have raised important questions regarding the interpretation of HDL turnover studies. In the rat, they showed that the turnover of HDL cholesteryl ester may be up to 7-fold greater than that for apo A-I. This implies that esterified cholesterol formed in HDL may be selectively removed from the plasma by some process other than endocytosis of the entire lipoprotein or transfer to lipoproteins with faster clearance rates (18). Other studies have shown that HDL may be readily capable of crossing the vascular endothelial wall (85) and that the concentration of discoidal HDL in interstitial fluid of cholesterol fed dogs may be up to 4-times greater than in plasma (139). Clearly, the dynamics of cholesteryl ester in HDL is more complex than previously thought. These studies suggest that major alterations in HDL composition may have profound effects on hepatic uptake of HDL cholesteryl esters or equilibration of esters between the vascular and extravascular compartments. Results from this study showed that CETP mediated cholesteryl ester equilibration among the plasma lipoproteins may be markedly affected by changes in HDL lipid composition. Other studies have shown that the apoprotein composition of the HDL, may also be of particular importance (65). CETP has recently been shown to be primarily associated with a subfraction of HDL which is devoid of apo A-II (65). By contrast, an inhibitor of CETP has been shown to be associated with an apo A-II rich subclass of HDL (76). Interestingly, this 92 study has shown significantly decreased cholesteryl ester transfer to HDL^ in almost all dyslipidemic patients and characterization of this nonprecipitable HDL-j by other investigators (140) has shown this subclass to be rich in apo A -II. This may suggest that perturbations in lipid metabolism may be associated with increases in a protein which inhibits CETP, alters H D L 3 composition, and inhibits cholesteryl ester transfer within the HDL pool. 4.2 USE OF RECOMBINANT HDL TO STUDY CETP AND L C A T IN VITRO 4.2.1 Characterization of recombinant HDL In this study, it was demonstrated that rHDL could be rapidly produced in vitro by short duration co-sonication of purified lipids and apoproteins. The methods described are simple, reproducible and result in acceptable yields. This study also showed that the physical characteristics of the rHDL were similar to native HDL with respect to size, charge and hydrated density. In addition, the recombinant and native particles were essentially identical in their ability to interact with CETP and LCAT. Previous studies by Pittman et al have shown that rHDL prepared by similar methods interact with cultured cells and permit selective lipid uptake in a similar manner to native HDL (102). Further, in vivo studies showed that these rHDL had almost identical tissue uptake and plasma decay kinetics to authentic HDL (102). Taken together these studies suggest that rHDL produced by co-sonication are structurally and metabolically identical to authentic HDL. In studies with purified human apoproteins, it was further shown that by varying the apoprotein and lipid components of these rHDL it is possible to produce homogeneous preparations which have well defined compositions representative of specific HDL subfractions. Particles of this type may be more physiologically relevant than vesicular or discoidal complexes for in vitro studies since they are structurally representative of mature, spherical HDL and provide the potential to study both the role of the neutral lipid core as well as the function of specific apoproteins. 4.2.2 Use of recombinant HDL as substrates of CETP In this study, rHDL prepared with and without a neutral lipid core were compared to native HDL. While particles prepared with a similar lipid and 93 protein composition were almost identical to native HDL as substrates for CETP and LCAT, changes in lipid composition were shown to markedly affect their interaction with both proteins. Since variations in HDL lipid composition have been shown to be correlated with the incidence of vascular disease (91,92,132), a detailed understanding of how altered HDL composition may affect both L C A T and CETP may help to elucidate the mechanism of the anti-atherogenic effect of HDL. While much evidence exists which suggests that L C A T is capable of transforming nascent HDL precursors into spherical particles (7), no studies have yet demonstrated whether CETP is capable of transferring newly synthesized cholesteryl esters produced by LCAT into discoidal HDL particles. In this study, it was shown that discoidal HDL are in fact very poor acceptors of cholesteryl esters. This supports previous observations which have shown that the transfer of neutral lipid may require a reciprocal exchange of either triglyceride or cholesteryl ester (68). By contrast, preliminary observations in our laboratory have shown that CETP avidly binds to discoidal HDL (data not shown). Therefore, since both LCAT and CETP may be associated with the same subfraction of HDL (65), it is possible that nascent HDL primarily exists as a complex with CETP and L C A T (31) which readily accepts free cholesterol (141) and promotes the synthesis and redistribution of cholesteryl esters to other lipoproteins. Thus, CETP may play a role in the speciation of HDL as suggested by others (81). 4.2.3 The effect of recombinant HDL composition on CETP 4.2.3.1 The effect of recombinant HDL lipid content The purpose of the modeling component of this project was to identify some of the lipid and apoprotein parameters that regulate the interaction of CETP with reconstructed HDL particles. This work was designed to corroborate the observations from the clinical studies which demonstrated that an increase in the triglyceride to cholesteryl ester ratio of HDL was associated with a reduced rate of transfer of CE into HDL. The observation that this may be exacerbated in patients with vascular disease has led us to suggest that, in normal individuals, CE transfer within the HDL pool prevents excessive accumulation within an atherogenic LDL pool. In studies with recombinant lipoproteins, it 94 was demonstrated that the neutral and polar lipid composition of rHDL both have a marked effect on CETP activity. Specifically, it was shown that an increase in rHDL triglyceride or free cholesterol (relative to cholesteryl ester or phospholipid respectively) decreased the capability of CETP to transfer cholesteryl ester to rHDL. Previous studies have suggested that CETP mediates a simple exchange of one molecule of cholesteryl ester for another between HDL and LDL or smaller V L D L and a net transfer of cholesteryl ester into nascent VLDL or chylomicrons via a reciprocal exchange of cholesteryl ester for triglyceride (72). Since rHDL with a pure triglyceride core were poor but nonetheless viable acceptors of cholesteryl ester, this suggests that a reciprocal exchange of cholesteryl ester for triglyceride may occur between LDL and triglyceride-rich HDL. If all the cholesteryl esters transferred from LDL were preferentially exchanged for triglycerides in the rHDL, the resultant changes in donor composition may have been responsible for the observed reduction in transfer rates, as Morton et al (68) have previously shown. This seems unlikely, however, since very little change in the donor composition would have occurred, even if triglycerides were preferentially exchanged, because very little lipid was transferred during incubations. It is proposed, therefore, that the observed reduction in CETP activity as triglyceride content was increased was likely a result of altered CETP catalytic ability rather than altered donor composition. During the preparation of this thesis, Morton (89) also reported that the transfer of cholesteryl ester to HDL is inhibited when the free cholesterol content of HDL is increased. He showed that this decreased transfer of cholesteryl ester to HDL was concomitant with a increased mass transfer of cholesteryl esters out of HDL to the lower density lipoproteins with a reciprocal equimolar increase in triglyceride transferred to HDL from VLDL. Thus, an increase in HDL free cholesterol promoted the accumulation of cholesteryl esters in lower density lipoproteins and triglycerides in HDL. These results suggest that cholesteryl ester transfer activity may be governed by the free cholesterol/phospholipid ratios in donor and acceptor lipoproteins 95 (89). Observations from this study suggest that changes in both core or surface lipids of an HDL particle affect its interaction with CETP (Figure 47). Since only the T _ 0 „ was affected, this suggests that HDL lipid content may regulate III a A the catalytic activity of CETP, apparently without modifying the interaction (Kpj) between CETP and HDL. Abnormal HDL lipid composition may, therefore, markedly affect cholesteryl ester equilibration within the HDL pool. The mechanism by which HDL lipid composition regulates the interaction with CETP may involve changes in the physical arrangement of the outer polar lipid monolayer. Evidence for this comes from studies of model lipoproteins which have shown that rHDL with a purely cholesteryl ester core have increased order of the phospholipid acyl Ghains relative to rHDL with a triglyceride core (101). As this study indicates that an increased cholesteryl ester content of HDL results in increased catalytic ability of CETP, it is proposed that this protein may require a highly ordered outer lipid monolayer. This may further explain the observation that HDL^ was a better substrate for CETP, when compared to H D L 2 , since HDL^ has also been shown to have more highly ordered phospholipid acyl chains (101). Conversely, the observation that increased free cholesterol inhibits the activity of CETP cannot be explained by this hypothesis if this lipid orders the phospholipid acyl chains as it does in model membrane bilayers (142). There is no evidence, however, that free cholesterol will order the surface lipid of a complex which is very small and is already highly ordered (101). In fact, it i s tempting to speculate that the cholesterol would disorder such a complex as it does in highly ordered gel state model membranes (143). Clearly, in order to test this hypothesis experiments on the direct effect of free cholesterol on acyl chain order in HDL must be carried out. Alternatively, some studies have suggested that the solubility of different neutral lipids in the phospholipid layer may also be affected by the free cholesterol content (144,145). It has been shown that increases in free cholesterol markedly reduce the solubility of cholesteryl ester, relative to triglyceride, in emulsion phospholipid monolayers (144,145). If triglyceride is also able to reduce the solubility of cholesteryl esters in phospholipid monolayers, this may explain 96 why elevated triglyceride levels, as well as free cholesterol levels, impair the ability of rHDL to receive transferred esters. 4.2.3.2 The effect of recombinant HDL apoprotein content In this study, rHDL was used to investigate the role of specific apoproteins on the activity of CETP. Discoidal rHDL prepared from apo A-I were shown to be better acceptors of cholesteryl esters than rHDL prepared from a full complement of HDL apoproteins. This indicates that some component of the apo HDL may inhibit the catalytic ability of CETP and govern the equilibration of cholesteryl esters within the HDL pool. Results with spherical particles further showed that rHDL prepared from apo HDL had a significantly reduced T m o v . and K J J , relative to particles prepared pure apo A-I, A-II, or both apoproteins. Since the ratio T m a x / K p j was similar to that for native HDL and apo HDL rHDL, this suggests that some component(s) of the HDL apoproteins was acting as a uncompetitive inhibitor of CETP, which equally reduced the reaction K u and T „ „ v XT ITlaX (Figure 27). Preliminary studies with other purified apoproteins have shown that while inclusion of apo A-II in A-I rHDL resulted in increases in reaction K J J and T m a x , inclusion of either apo E or apo C-III resulted in reduced reaction K J J and T m a x . Interestingly, inclusion of apo C-III substantially reduced the apparent K J J of the reaction to that comparable to rHDL prepared from apo HDL. Thus, it is postulated that some component of the HDL apoproteins (possibly apo C-III) is associated with H D L 3 and regulates the affinity with which CETP binds to HDL. However, inclusion of apo C-III did not completely normalize the T m a x or abolish the apparent substrate inhibition observed at high substrate concentrations. Therefore, some other protein which is associated with H D L 2 and VHDL, may be involved in a non-competitive inhibition and act by reducing the reaction T m a x values and preventing the apparent substrate inhibition at high substrate concentrations: This may be the CETP-inhibitor protein described by others (73,75,76). The presence of this protein on H D L 2 and VHDL may prevent the formation of inactive HDL-CETP-HDL complexes which could be analogous to the inactive substrate-enzyme-substrate complexes which may result in substrate inhibition. The lack of this inhibitor on HDL^ may allow for the maturation of 97 these particles by permitting the influx of cholesteryl esters from other lipoproteins. 4.2.4 The effect of recombinant HDL lipid composition on L C A T activity The study of the interaction between L C A T and HDL has been limited by the complex nature of native substrates. However, details of the reaction mechanism (39), substrate specificity (43,45) and the apoprotein activation parameters (40,42) for LCAT have been elucidated using model lipoproteins in vitro. Most of the model substrates used have been mixtures of polar lipid and apoprotein whereas there have been no systematic studies of the interaction of L C A T with model substrates containing a neutral lipid core. Consequently, the factors regulating the interaction of LCAT with HDL still remain unclear. Previous studies using native lipoproteins have shown that smaller HDL^ particles are preferred substrates of LCAT, with lower apparent K „ and higher V—,,,, values than III Ilia. X for H D L 2 (44). In this study, the estimated kinetic parameters for rHDL were similar to those estimated for native HDL. It was demonstrated that a significant relationship exists between the size of an HDL particle, its relative protein content and the apparent K m for cholesterol as a L C A T substrate. It was also shown that, regardless of the size or relative protein content of an rHDL particle, increased triglyceride relative to cholesteryl ester reduced the V m a x for the reaction with LCAT. This suggests that while HDL particle size may regulate the interaction between LCAT and HDL, the relative neutral lipid content of the particle may modulate the catalytic potential of the enzyme. This hypothesis is supported by the observations of Barter et al (45) from studies with HDL^ modified in vitro by incubation with intralipid. They showed that homogeneous preparations of hypertriglyceridemic HDL^ were very poor substrates for LCAT and had a low V m o v for. the L C A T reaction relative to Hid. A native HDL-j. Interestingly, they further showed that inclusion of a subpopulation of very small particles resulted in a heterogeneous mixture that was a much better substrate for LCAT with an elevated V m Q V and reduced K ^ , II ldA III relative to untreated HDL-j (45). Barter et at (44,45) have provided evidence that particle size determines 98 the reactivity of HDL, as a substrate for LCAT, in a manner which was independent of the cholesteryl ester content. Conversely, it has also been suggested that the ability of LCAT to esterify cholesterol is determined by the cholesteryl ester content of the substrate since cholesteryl ester may act as a feedback inhibitor (19,43). Results from the present study suggest that HDL enriched in cholesteryl ester is not a poor substrate for L C A T and that the ability of HDL to act as a substrate is negatively related to the particle size. Therefore, we propose that any feedback inhibition observed as L C A T acts upon a lipoprotein substrate is related to an increase in particle size, rather than a direct effect of the cholesteryl ester. The release from inhibition, facilitated by CETP and acceptor lipoproteins (19,43), may also be a result of the changes in particle size since it has also been shown in vitro that CETP may promote the formation of smaller HDL particles (72,81). Thus, it is possible that HDL^ from hypertriglyceridemic subjects may have greater than normal reactivity with LCAT primarily because of a subpopulation with a very small particle size, rather than their decreased cholesteryl ester content (44,45). 4.3 HDL COMPOSITION, CETP AND LCAT: CONCLUDING REMARKS This study has indicated that HDL composition is markedly altered in individuals at increased risk for coronary heart disease. Results from both clinical studies and studies employing rHDL showed that increases in the triglyceride or free cholesterol content of HDL markedly impair the ability of HDL to receive cholesteryl ester transferred by CETP. What then is the physiological significance of reduced cholesteryl ester transfer into HDL? The proposed net movement of cholesteryl ester in reverse cholesterol transport is from HDL to lower density lipoproteins; however, as stated earlier, the assumptions this theory is based upon may be questioned. In this thesis, it is suggested that a net movement of cholesteryl esters into lower density lipoproteins is not a normal physiological event, except during postprandial lipemia. Conversely, it is now proposed that a net transport of cholesteryl esters into lower density lipoproteins results from a reduced ability of HDL to receive cholesteryl esters 99 and may be an atherogenic event. The previously cited report by Morton (89) clearly showed that HDL may act as both a donor and recipient of cholesteryl ester. This is graphically illustrated in the following scheme. The illustration is a composite of a series of incubations, through which, Morton showed that increased free cholesterol content markedly affected the ability of HDL to receive cholesteryl ester but had no effect on the ability of lower density lipoproteins to receive this lipid. A B F C RICH L D L V L D L 23 CE 32 F C R I C H HDL Each panel represents the sum of four different incubations where cholesteryl ester transfer was estimated via isotope transfer from LDL to HDL, V L D L to HDL, HDL to LDL and from HDL to VLDL. Units represent lipid transfer rates (kt x 100) as defined by Morton. Incubations with unmodified lipoproteins (A) showed that cholesteryl ester transfer rates into HDL slightly exceeded that leaving. When free cholesterol content was increased (B), however, the ability of HDL to receive cholesteryl ester markedly fell while the ability of lower density lipoproteins to receive this lipid was unaffected. If the amount of cholesteryl ester coming into the HDL pool falls but the amount leaving remains constant, one would expect a net movement of cholesteryl ester into lower 100 density lipoproteins. In fact, Morton further showed that increased free cholesterol in VLDL resulted in a increase in the net mass transfer of cholesteryl esters from HDL to lower density lipoproteins (89). In the present study, it was shown that increased free cholesterol content will impair the ability of HDL to receive cholesteryl ester; however, it was futher demonstrated that triglyceride content will impair cholesteryl ester transfer in a similar manner. It was also demonstrated that the rate of cholesteryl ester transferred into HDL was significantly reduced in patients with disorders of lipid metabolism, but the rate of transfer into LDL and VLDL was normal or slightly increased. Consequently, it is now proposed that a change in HDL composition impairs the ability of HDL to equilibrate cholesteryl esters within the HDL pool and results in a net increased transfer to more atherogenic lipoproteins. This may subsequently promote atherogenesis and an increased risk of coronary heart disease. Furthermore, the apparent paradox of the low incidence of ischemic heart disease in some patients with severe HDL deficiencies (FED, Tangier disease, and LCAT deficiency) and the high incidence in some patients with normal HDL levels (hepatic lipase deficiency) suggests that it may be the composition of HDL, rather than the mass, that is central to its anti-atherogenic capacity. The factors which lead to the accumulation of triglycerides in HDL in instances of hypertriglyceridemia are established and have been reviewed extensively (82,83). However, the factors involved in the accumulation of triglycerides in HDL of normotriglyceridemic subjects are unclear. In studies in the early seventies, Fielding (146) showed that increased free cholesterol in triglyceride emulsions effectively inhibits lipoprotein lipase activity. Patsch et al (134) have shown that impaired lipase activity results in the transfer of triglycerides to HDL. Therefore, if the free cholesterol content of chylomicrons is elevated in hypercholesterolemia this may inhibit lipoprotein lipase catabolism of triglycerides and thus lead to transfer of these lipids into HDL. This may be the mechanism that initially promotes the accumulation of triglycerides in HDL and may also be a mechanism by which hypercholesterolemia 101 may aggravate the lipoprotein modifying effects of hypertriglyceridemia. Hepatic lipase is thought to play a central role in the regulation of HDL triglyceride content. (134). Therefore, if hepatic lipase activity is also inhibited by an increased free cholesterol content in HDL (from chylomicron degradation products) this could prevent the catabolism of HDL triglycerides and further accentuate the accumulation of triglycerides in the HDL pool. Interestingly, a deficiency of hepatic lipase has been shown to result in the accumulation of triglycerides in HDL and LDL, pronounced coronary heart disease, but no significant changes in HDL mass (135,136). In this study, it was shown that elevated HDL triglyceride or free cholesterol content may impair the equilibration of cholesteryl esters in the HDL pool. Since approximately 75% of HDL cholesterol content is cholesteryl ester, it seems possible that an impairment in this equilibration of cholesteryl esters within the HDL pool and increased net transfer to lower density lipoproteins is in fact what causes reduced HDL cholesterol levels. This would suggest that the widely accepted risk factor for heart disease, reduced HDL-cholesterol, may be directly related to an elevation in HDL-triglycerides. Perhaps, hypertriglyceridemia on its own may not directly result in the accumulation of cholesteryl esters in lower density lipoproteins since an active hepatic lipase may prevent the excessive accumulation of. triglycerides in the HDL pool which is responsible for production and equilibration of cholesteryl esters. However, if free cholesterol levels are elevated and hepatic lipase is subsequently inhibited, triglycerides may accumulate in HDL and result in a diversion of the L C A T derived cholesteryl esters to lower density lipoproteins. In this study, it was shown that increased rHDL triglyceride content resulted in an impaired catalytic potential of LCAT. This observation suggests that there may be a connection between triglyceride and cholesterol metabolism. Conceivably, decreased LCAT activity may lead to an increase in HDL free cholesterol content as well as an increased net transport of cholesteryl esters to more atherogenic lipoproteins. Interestingly, Kuksis et al (147) have identified a significant relationship between increased plasma free cholesterol 102 levels and the incidence of ischemic heart disease. This study has shown that both the lipid and apoprotein components of HDL are integrally involved in the regulation of CETP. Results from the clinical component of this study showed that the decreased transfer to patient's HDL was primarily a result of decreased transfer to H D L 3 and as such it was postulated that this may result from increased inhibitor content in the HDL of these subjects. In studies with native and rHDL, it was shown that some component of the HDL apoproteins, probably associated with H D L 2 and VHDL, may be involved in a non-competitive inhibition of CETP and reduce the maximum velocity of lipid transfer into these fractions. If this is in fact the CETP-inhibitor protein described by others (73,75,76), this protein may play an important role in regulating the influx/efflux of cholesteryl ester in particular HDL subclasses. The lack of this inhibitor on H D L 3 may allow for the maturation of these particles by permitting the influx of cholesteryl esters from other lipoproteins. An elevation of the level of this inhibitor in HDL^, however, may markedly affect the composition of this HDL fraction by directly impairing the influx and increasing the efflux of cholesteryl esters to other lipoproteins. This may also result in decrease in HDL-cholesterol and may have profound effects on reverse cholesterol transport. The central role that CETP plays in the equilibration and transport of cholesteryl esters has resulted in a great deal of speculation on the role CETP in atherogenesis (15,53,72). Fielding et ai (53) have proposed that the transfer of cholesteryl ester from HDL to more rapidly turning over pools of lipoproteins may be anti-atherogenic since this promotes the transfer of cholesterol from peripheral tissues to the liver. However, the • capacity of CETP to transfer CE into LDL suggests that CETP may be directly involved in elevating plasma levels of this potentially atherogenic lipoprotein (72). In the clinical component of this study, evidence is provided that an increase in the ratio of triglyceride/cholesteryl ester in HDL is concomitant with a net reduction of cholesteryl ester transferred to the HDL pool. This may result in defective transport of cholesteryl ester within the HDL pool and it is proposed that this 103 could impair reverse cholesterol transport by three possible mechanisms. Firstly, an increase in overall HDL-triglyceride (or free cholesterol) content might promote an increased net movement of LCAT derived cholesteryl esters to lower density lipoproteins and lead to increased levels of LDL. Secondly, increased transfer of cholesteryl ester to lower density particles may directly impede the normal maturation of HDL and subsequently prevent the formation of a cholesteryl ester rich HDL particle which is rapidly removed by direct hepatic uptake or depleted of cholesteryl esters by selective lipid uptake by the liver. Finally, if reduced lipid transfer to HDL impairs the equilibration of HDL cholesteryl esters between vascular and extravascular compartments, the cholesteryl esters may accumulate' in the vascular compartment in potentially atherogenic particles instead of being temporarily stored in the nonatherogenic HDL pool. The overall conclusion of this thesis is that HDL composition regulates the interaction of L C A T and CETP with this lipoprotein. This has the potential to play a major regulatory role in the process of reverse cholesterol transport. Furthermore, since variations in HDL lipid composition have been shown to be directly correlated with the incidence of ischemic heart disease, the work presented in this thesis supports the hypothesis that it may be HDL composition, rather than mass, that is central to the anti-atherogenic capacity of HDL. As shown in Figure 31, abnormal interactions between HDL, CETP, and LCAT may impair the equilibration of cholesteryl esters and their subsequent removal from the HDL pool. The consequence may be increased transfer and accumulation of cholesteryl esters in lower density lipoproteins; an event which promotes atherogenesis. 104 NORMAL HYPERLIPIDEMIA PERIPHERAL CELLS Figure 31: Reverse cholesterol transport in hyperlipidemia. Cholesterol transport in normolipidemia has be proposed to involve the efflux of free cholesterol from peripheral cells to HDL, esterification by LCAT, interlipoprotein transfer by CETP and finally hepatic uptake. We have shown that patients with hyperlipidemia often present with elevated HDL triglyceride levels. We further showed that transfer of cholesteryl ester to this hypertriglyceridemic HDL pool was impaired in these individuals. Conversely, in some patient groups, cholesteryl ester transfer to LDL and VLDL was shown to be elevated. As such, we have postulated that if hepatic uptake of these lipoproteins is rate limiting, cholesteryl esters may accumulate in these lower density lipoproteins and this may then accentuate the development of atheroma in these individuals. 105 4.4 PROPOSAL FOR FUTURE STUDY The work presented in this thesis has identified an important relationship between altered HDL composition and impaired CETP and L C A T action. Since abnormal HDL composition has been correlated with the incidence of ischemic heart disease, we have proposed that this observed abnormal CETP and L C A T function may have a central role in the atherogenic process. Future studies of the factors which regulate both CETP and LCAT activity are necessary if we are to understand the biomolecular events which lead to atherosclerosis. These studies could be grouped into three related classes; 1. Clinical studies Further questions that must be answered by clinical studies are as follows: a. Does altered HDL lipid composition affect the molar apoprotein ratios of the individual HDL subfractions? b. Is altered HDL composition associated with variations in the level of CETP inhibitor proteins? c. Is altered CETP mass associated any specific disease states? d. How does altered HDL composition affect extravascular lipid metabolism and specifically, does elevated triglyceride content of HDL prevent vascular and extravascular cholesteryl ester equilibration? e. Is chylomicron free cholesterol content elevated in hypercholesterolemia (or other disorders) and, if so, does this impair lipoprotein lipase activity in these subjects? f. What conditions lead to elevated HDL free cholesterol content and does this impair hepatic lipase catabolism of HDL-triglycerides? 2. Biochemical studies We are currently undertaking a continuation of the present study, with the objective to further clarify the effect of different apoprotein (particularly apo C-III) and inhibitor proteins, in recombinant lipoproteins, on CETP and L C A T activity. In addition, isolation of native HDL subclasses, from patients with disorders of lipid metabolism, and their characterization as substrates for both 106 CETP and L C A T may elucidate why the HDL of some patients seem to be more efficient in the process of reverse cholesterol transport. In future studies, we plan to extend this design to include a detailed characterization of hepatic lipase in much the same way as we have done for CETP and LCAT. 3. Biophysical studies Biophysical characterization of microstructure of an HDL particle may help to elucidate the mechanisms behind the observed altered activities of CETP and LCAT. With this in mind, we are presently undertaking NMR studies to further characterize the effect of variations in HDL neutral or polar lipid composition on the phospholipid acyl chain order. These studies will involve deuterium NMR studies of rHDL prepared from phospholipids that were selectively deuterated in their acyl chains. We have hypothesized that optimal CETP (and perhaps LCAT) activity may require highly ordered phospholipid acyl chains. In initial experiments, we intend to study the effects of increased rHDL free cholesterol on acyl chain mobility. In subsequent work, we plan to undertake a detailed characterization of the effects of different apoproteins, inhibitor proteins, CETP and L C A T on the orientational order of rHDL phospholipid acyl chains. 107 5.0 REFERENCES 1. Havel R.J., Goldstein J.L. and Brown M.S. Lipoproteins and lipid transport. In "Metabolic Control and Disease (P.K. Bondy and L.E. Rosenberg, eds.) 8th Ed,, pp.393-494. Saunders, Philadelphia, 1980. 2. McGill H.C. Jr., ed. The geographic pathology of atherosclerosis. Baltimore: Williams & Wilkins, 1968:pp.l-193. 3. Inkeles S. and Eisenberg D. Hyperlipidemia and coronary atherosclerosis: a review. Medicine (Baltimore) 1981;60:110-23. 4. Steinberg D. Lipoproteins and atherosclerosis: a look back and a look ahead. Arteriosclerosis 1983;3:283-301. 5. Report of the Working Group of Arteriosclerosis of the National Heart, Lung, and Blood Institute. Vol. 2. Washington, D.C.: Government Printing Office, 1981. (DHEW publication no. (NIH) 82-2035). 6. Miller G.J. and Miller N.E. Plasma-high-density-lipoprotein concentration and development of ischaemic heart-disease. Lancet 1975;1:707-14. 7. Eisenberg S. High density lipoprotein metabolism. J. Lipid Res. 1984;25:1017-1066. 8. Morrisett J.D., Jackson R.L. and Gotto A .M. Jr. Lipid-protein interactions in the plasma lipoproteins. Biochim. Biophys. Acta. 1977;472:93-133. 9. Segrest J.P., Jackson R.L., Morrisett J.D. and Gotto A.M. Jr. A molecular theory of lipid-protein interactions in the plasma lipoproteins. Febs. Lett. 1974;38:247-253. 10. Smith L.C., Pownall H.J. and Gotto A.M. Jr. The plasma lipoproteins: structure and metabolism. Ann. Rev. Biochem. 1978;47:751-77. 11. Brown M.S., Kovanen P.T. and Goldstein J.L. Regulation of plasma cholesterol by lipoprotein receptors. Science 1981;212:628-35. 12. Bilheimer D., Eisenberg S. and Levy R.I. The metabolism of very low density lipoprotein proteins. I. Preliminary in vitro and in vivo observations. Biochim. Biophys. Acta 1972;260:212-21. 13. Glass C , Pitmann R.C., Weinstein D.B. and Steinberg D. Dissociation of tissue uptake of cholesteryl ester from that of apoprotein A-I of rat plasma high density lipoprotein: selective delivery of cholesterol to liver, adrenal, and gonad. Proc. Natl. Acad. Sci. U.S.A. 1983;80:5435-5439. 14. Miller N.E. Current concepts of the role of HDL in reverse cholesterol transport. In: Miller N.E., Miller G.J. eds. Clinical and Metabolic aspects of high density lipoproteins. Elsevier Science Publishers B.V. 1984:pp.187-216. 15. Glomset J.A. Physiological role of lecithin:cholesterol acyltransferase. Amer. J. Clin. Nutr. 1970;23:1129-36. 16. Fielding C.J. and Fielding P.E. Evidence for a lipoprotein carrier in human plasma catalyzing sterol efflux from cultured fibroblasts and its relationship to lecithin:cholesterol acyltransferase. Proc. Natl. Acad. Sci. USA 1981;78:3911-14. 108 17. Barter P.J., Hopkins G.J. and Calvert G.D. Transfers and exchanges of esterified cholesterol between plasma lipoproteins. Biochem. J. 1982;208:1-7. 18. Calvert G. and Abbey M. Plasma lipoproteins, apolipoproteins, and proteins concerned with lipid metabolism. Advances in Clinical Chemistry 1985;24:217-298. 19. Chajek T. and Fielding C.J. Interaction of lecithinxholesterol acyltransferase and cholesteryl ester transfer protein in the transport of cholesteryl ester into sphingomyelin liposomes. Biochemistry 1980:19;3673-3677. 20. Cheung M.C. and Albers J.J. Characterization of lipoprotein particles isolated by immunoaffinity chromatography. J. Biol. Chem. 1984;259:12201-12209. 21. Norfeldt P.P., Olofsson S.O., Fager G. and Bondjers G. Isolation and partial characterization of the lipoprotein families A and A-I from High-density lipoproteins of human serum. Eur. J. Biochem. 1981;118:1-8. 22. Weisgraber K.H. and Mahley R.W. Subfractionation of human high density lipoproteins by heparin-Sepharose affinity chromatography. J. Lipid Res. 1980;21:316-325. 23. Jahani M. and Lacko A.G. A study of the interaction of lecithin:cholesterol acyltransferase with subfractions of high density lipoproteins. J. Lipid Res. 1981;22:1102-1110. 24. Marcel Y.L. , Weech P.K.,. Nguyen T.D., Milne R.W. and McConathy W.J. Apolipoproteins as the basis for heterogeniety in high-density lipoprotein2 and high-denstiy lipcprotein3. Studies by isolelectric focusing on agarose films. Eur. J. Biochem. 1984;143:467-478. 25. Blanche P.J., Gong E.L., Forte T.M. and Nichols A.V. Characterization of human high density lipoproteins by gradient gel electrophoresis. Biochim. Biophys. Acta 1981;665:408-419. 26. Glomset J.A. The plasma lecithin:cholesterol acyltransferase reaction. J. Lipid Res. 1968;9:155-67. 27. Soutar A.K., Garner C.W., Baker H.N., Sparrow J.T., Jackson R.L., Gotto A.M. and Smith L.C. Effect of the human plasma apolipoproteins and phosphatidylcholine acyl donor on the activity of lecithinxholesterol acyltransferase. Biochemistry 1975;14:3057-3063. 28. Fielding C.J., Shore V.G. and Fielding P.E. A protein cofactor of lecithin-.cholesterol acyltranserase. Biochem. Biophys. Res. Commun. 1972;46:1493-8. 29. Albers J.J., Lin J.T. and Roberts G.P. Effect of human plasma apolipoproteins on the activity of purified lecithinxholesterol acyltransferase. Artery 1979;5:61-75. 30. Matz C.E. and Jonas A. Reaction of human lecithinxholesterol acyltransferase with synthetic micellular complexes of apolipoprotein A-I, phosphatidylcholine, and cholesterol. J. Biol. Chem. 1982;257:4541-6. 31. Fielding P.E. and Fielding C.J. A cholesteryl ester transfer complex in human plasma. Proc. Natl. Acad. Sci. USA 1980;77:3327-30. ; 109 32. Cheung M.C., Wolf A.C., Lum K.D., Tollefson J.H. and Albers J.J. Distribution and localization of lecithinxholesterol acyltransferase and cholesteryl ester transfer activity in A-I-containing lipoproteins. J. Lipid Res. 1986;27:1135-44. 33. McLeod R., Lacko A.G. , Pritchard P.H. and Frohlich J. Purification of biologically active apolipoproteins by chromatofocussing. J. Chrom. 1986;381:271-83. 34. Steinmetz A. and Utermann G. Activation of lecithin:cholesterol acyltransferase by human apolipoprotein A-IV. J. Biol. Chem. 1985;260:2258-64. 35. Steinmetz A., Kaffarnik H., and Utermann G. Activation of phosphatidylcholine-sterol acyltransferase by human apolipoprotein E isoforms. Eur. J. Biochem. 1985;152:747-51. 36. Chen C.H. and Albers J.J. Stimulation of lecithinxholesterol acyltransferase activity by apolipoprotein A-II in the presence of apolipoprotein A-I. Eur. J. Biochem. 1986;155:589-94. 37. Nishida H.I., Nakanishi T., Yen E.A., Arai H., Yen F.T. -and Nishida T. Nature of the enhancement of lecithinxholesterol acyltransferase. J. Biol. Chem. 1986;261:12028-35. 38. Schaeffer E.J. Clinical, biochemical, and genetic features in familial disorders of high density lipoprotein deficiency. Arteriosclerosis 1984;4:303-22. 39. Jauhiainen M. and Dolphin P.J. Human plasma lecithinxholesterol acyltransferase. An elucidation of the catalytic mechanism. J. Biol. Chem. 1986;15:7032-43. 40. Ho W.K.K. and Nichols A.V. Interaction of lecithinxholesterol acyltransferase with sonicated dispersions of lecithin. Biochim. Biophys. Acta 1971;231:185-193. 41. Pownall H.J., Van Winkle B., Pao Q., Rohde M . , and Gotto A .M. Action of lecithinxholesterol acyltransferase on model lipoproteins. Preparation and characterization of model nascent high density lipoprotein. Biochim. Biophys. Acta 1982;713:494-5 42. Jonas A. Synthetic substrates of lecithinxholesterol acyltransferase. J. Lipid Res. 1986;27:689-698. 43. Fielding C.J. and Fielding P.E. Regulation of human plasma lecithinxholesterol acyltransferase activity by lipoprotein acceptor cholesteryl ester content. J. Biol. Chem. 1981;256:2102-4. 44. Barter P.J., Hopkins G.J., Gorjatschko L. and Jones M.E. Competitive inhibition of plasma cholesterol esterification by human high density lipoprotein subfraction 2. Biochim. Biochem. Acta 1984;793:260-268. 45. Barter P.J., Hopkins G.J. and Gorjatschko L. Lipoprotein substrates for plasma cholesterol esterification. Influence of particle size and composition of the high density lipoprotein subfraction 3. Atherosclerosis 1985;58:97-107. 46. Hopkins G.J. and Barter P.J. An effect of very low density lipoproteins on the rate of cholesterol esterification in human plasma. Biochim. Biophys. Acta 1982;794:152-60. 110 47. Nichols A.V. and Smith J. Effect of very-low density lipoproteins on lipid transfer in incubated serum. J. Lipid Res. 1965;6:206-210. 48. Nestel P.J., Reardon M. and Billington T. In vivo transfer of cholesteryl esters from high-density to very-low-density lipoproteins in man. Biochim. Biophys. Acta 1979;573:403-407. 49. Barter P.J. and Jones M.E. Kinetic studies of the transfer of esterified cholesterol between human plasma low and high density lipoproteins. J. Lipid Res. 1980;21:238-249. 50. Hopkins G.J. and Barter P.J. Role of esterified cholesterol transfers in the regulation of plasma cholesterol esterification. Atherosclerosis 1983;49:177-85. 51. Ihm I., Ellsworth J.L., Chataing B. and Harmony J.A.K. Plasma protein-facilitated coupled exchange of phosphatidylcholine and cholesteryl ester in the absence of cholesterol esterification. J. Biol. Chem. 1982;257:4818-4827. 52. Morton R.E. and Zilversmit D.B. Purification and characterization of lipid transfer protein(s) from - human lipoprotein-deficient plasama J. Lipid Res. 1982;23:1058-1067. 53. Albers J.J., Tollefson J.H., Chen C. and Steinmetz A. Isolation and characterization of human plasma lipid transfer proteins. Arteriosclerosis 1984;4:49-58. 54. Chajek T. and Fielding C.J. Isolation and characterization of a human serum cholesteryl ester transfer protein. Proc. Natl. Acad. Sci. USA 1978;75:3445-49. 55. Morton R.E. and Zilversmit D.B. The separation of apoD from cholesteryl ester transfer protein. Biochim. Biophys. Acta. 1981;663:350-55. 56. Tall A.R., Abreu E. and Shuman J. Separation of a plasma plasma phospholipid transfer protein from cholesteryl ester/phospholipid exchange protein. J. Biol. Chem. 1983;258:2174-80. 57. Hesler C.B., Swenson T.L. and Tall A.R. Purification and characterization of human plasma cholesteryl ester transfer protein. J. Biol. Chem. 1987;262:2275-2282. 58. Jarnagin A.S., Kohr W. and Fielding C.J. Isolation and specificity of a Mr 74,000 cholesteryl ester transfer protein from human plasma. Proc. Natl. Acad. Sci. USA 1987;84:1854-57. 59. Drayna D., Jarnagin A.S., McLean J., Henzel W., Kohr W., Fielding C. and Lawn R. Cloning and seqencing of human cholesteryl ester transfer protein cDNA. Nature 1987;327:632-634. 60. Tollefson J.H. and Albers J.J. Isolation, characterization and assay of plasma lipid transfer proteins. Methods in Enzymology. 1986:129;797-817. 61. Tollefson J . H , Faust R., Albers J.J. and Chait A. Secretion of a lipid transfer protein by human monocyte-derived macrophages. J. Biol. Chem. 1985;260:5887-90. 62. Faust R.A. and Albers J.J. Regulated vectorial secretion of cholesteryl ester transfer protein (LTP-I) by the CaCo-2 model of human enterocyte epithelium. J. Biol. Chem. 1988;263:8786-89. I l l 63. Swenson T.L., Simmons J.S., Hesler C.B., Bisgaier C. and Tall A.R. Cholesteryl ester transfer protein is secreted by Hep G2 cells and contains asparagine-linked carbohydrate and sialic acid. J. Biol. Chem. 1987;262:16271-74. 64. Swenson T.L., Brocia R.W. and Tall A.R. Plasma cholesteryl ester transfer protein has binding sites for neutral lipids and phospholipids. J. Biol. Chem. 1988;263:5150-57. 65. Cheung M.C., Wolf A.C., Lum K.D., Tollefson J.H. and Albers J.J. Distribution and localization of lecithinxholesterol acyltransferase and cholesteryl ester transfer activity in A-I-containing lipoproteins. J. Lipid Research 1986;27:1135-1144. 66. Tall A.R., Granot E., Brocia R., Tabas I., Hesler C , Williams K. and Denke M. Accelerated transfer of cholesteryl esters in dyslipidemic plasma. Role of cholesteryl ester transfer protein. J. Clin. Invest. 1987;79:1217-25. 67. Ihm J., Quinn D.M., Busch S.J., Chataing B. and Harmony J.A.K. Kinetics of plasma protein-catalyzed exchange of phophotidylcholine and cholesteryl ester between plasma lipoproteins. J. Lipid Res. 1982;23:1328-41. 68. Morton R.E. and Zilversmit D.B. Inter-relationship of lipids transferred by the lipid-transfer protein isolated from human lipoprotein-deficient plasma. J. Biol. Chem. 1983;258:11751-11757. 69. Morton R.E. Binding of plasma-derived lipid transfer protein to lipoprotein substrates. J. Biol. Chem. 1985;260:12593-12599. 70. Pattnaik N.M. and Zilversmit D.B. Interaction of cholesteryl ester exchange protein with human plasma lipoproteins and phospholipid vesicles. J. Biol. Chem. 1979;254:2782-86. 71. Sammet D. and Tall A.R. Mechanisms of enhancement of cholesteryl ester transfer protein activity by lipolysis. J. Biol. Chem. 1985;260:6687-97. 72. Tall A.R. Plasma lipid transfer proteins. J. Lipid Res. 1986;27:361-367. 73. Castro G.R. and Fielding C.J. Effects of postprandial lipemia on plasma cholesterol metabolism. J. Clin. Invest. 1985;75:874-882. 74. Son Y.C. and Zilversmit D.B. Purification and characterization of human proteins that inhibit lipid transfer activities. Biochim. Biophys. Acta. 1984;795:473-480. 75. Morton R.E. and Zilversmit D.B. A plasma inhibitor of triglyceride and cholesteryl ester transfer activities. J. Biol. Chem. 1981:256;11992-11995. 76. Nishide T., Tollefson J.H. and Albers J.J. Inhibition of lipid transfer by a specific high density lipoprotein subclass containing an inhibitor protein. Arteriosclerosis 1987;7:538a. 77. Glomset J.A. and Norum K.R. The meteabolic role of lecithinxholesterol acyltransferase: perspectives from pathology. Adv. Lipid Res. 1973;1:1-65. 78. Marcel Y.L. , Vezina C , Teng B. and Sniderman A. Transfer of cholesteryl esters between human high density lipoproteins and triglyceride-rich lipoproteins controlled by a plasma protein factor. Atherosclerosis 1980;35:127-133. 112 79. Eisenberg S. Preferential enrichment of large-sized very low density lipoprotein populations with transferred cholesteryl esters. J. Lipid Res. 1985;26:487-94. 80. Deckelbaum R.J., Eisenberg S., Oschry Y., Butbul E. , Sharon I. and Olivecrona T. Reversible modification of human plasma low density lipoproteins towards triglyceride-rich precursors. A mechanism for losing excess cholesteryl esters. J. Biol. Chem. 1982;257:6509-6517. 81. Ellsworth J.L., Kashyap M.L., Jackson R.L. and Harmony J.A.K. Human plasma lipid transfer protein catalyzes the speciation of high density lipoproteins. Biochim. Biophys. Acta 1987;918:260-266. 82. Nikkila E.A., Taskinen M.R. and Sane T. Plasma high-density lipoprotein concentration and subfraction distribution in relation to triglyceride metabolism. Am. Heart J. 1987;543-548. 83. Patsch J.R., Prasad S., Gotto A.M. and Bengtsson-Olivcrona G. Postprandial lipemia. A key for the conversion of high density lipoprotein into high density lipoproteins by hepatic lipase. J. Clin. Invest. 1984;74:2017-23. 84. Stein Y., Stein O., Olivecrona T. and Halperin G. Putative role of • cholesteryl ester tansfer protein in removal of cholesteryl ester from vascular interstitium, studied in a model system in cell culture. Biochim. Biophys. Acta. 1985;834:336-45. 85. Reichl D., Postiglione A., Myant N.B., Pflug J.J. and Press M. Observations on the passage of apoproteins from plasma lipoproteins into peripheral lymph in two men. Clin. Sci. Mol. Med. 1975;49:419-26. 86. Fielding P.E., Fielding C.J., Havel R.J., Kane J.P. and Tun P. Cholesterol net transport, esterification, and transfer in human hyperlipidemic plasma. J. Clin. Invest. 1983;71:449-460. 87. Ha Y.C. and Barter P.J. Differences in plasma cholesteryl ester transfer activity in sixteen vertebrate species. Comp. Biochem. Physiol. 1982;71b:265-69. 88. Calvert G. and Abbey M. Plasma lipoproteins, apolipoproteins, and proteins concerned with lipid metabolism. Advances in Clinical Chemistry 1985;24:217-298. 89. Morton R.E. Free cholesterol is a potent regulator of lipid transfer protein function. J. Biol. Chem. 1988:263(25);12235-12241. 90. Fielding C.J., Reaven G.M., Liu G. and Fielding P.E. Increased free cholesterol in plasma low and very low density lipoproteins in non-insulin-dependent diabetes mellitus: its role in the inhibition of cholesteryl ester transfer. Proc. Natl. Acad. Sci. USA 1984;81:2512-16. 91. Moshides J.S. High density lipoprotein free cholesterol and other lipids in coronary heart disease. Arteriosclerosis 1987;7:262-266. 92. Little A., Kakis G., Leather T. and Breckenridge C.W. Plasma high density lipoprotein triglyceride as a risk factor for ischemic vascular disease in a prospective study. Clin. Invest. Med. 1984;7 (suppl 2);68a. 93. Kurasawa T., Yokoyama S., Miyake Y., Yamamura T. and Yamamoto A. Rate of cholesteryl ester transfer between high and low density lipoproteins in human serum and a case with decreased transfer rate in association with hyperalphalipoproteinemia. J. Biochem. 1985;98:1499-1508. 113 94. Yokoyama S., Kurasawa T., Nishikawa O. and Yamamoto A. High density lipoproteins with poor reactivity to cholesteryl ester transfer reaction observed in a homozygote of familial hyperalphalipoproteinemia. Artery 1986;14:43-51. 95. Pappu A.S. and Illingworth D.R. Neutral lipid transfer activities in the plasma of patients with abetalipoproteinemia. Atherosclerosis 1988;71:1-7. 96. Atkinson D. and Small D. Recombinant lipoproteins: Implications for structure and assembly of native lipoproteins. Ann. Rev. Biophys. Biophys. Chem. 1986:15;403-56. 97. Jonas A. Reconstitution of high-density lipoproteins. Methods in Enzymology 1986:128;553-567 98. Hirz R. and Scanu A.M. Reassembly in vitro of a serum high-density lipoprotein. Biochim. Biophys. Acta 1970;207:364-367. 99. Ritter M.C. and Scanu A.M. Role of apolipoprotein A-I in the structure of human serum high density lipoproteins. J. Biol. Chem. 1977;252(4):1208-1216. 100. Ritter M.C. and Scanu A.M. Apolipoprotein A-II- and structure of human serum high density lipoproteins. An approach by reassembly techniques. J. Biol. Chem. 1979;254:2517-25. 101. Fenske D.B., Parmar Y.I., Treleaven W.D., Chana R.S. and Cushley R.J. Lipid-lipid interactions in reconstituted high-density lipoproteins by deuterium nuclear magnetic resonance. Biochemistry 1988;27:4491-4500. 102. Pittman R.C., Glass C.K., Atkinson D. and Small D.M. Synthetic high density lipoprotein particles. J. Biol. Chem. 1987;262(6):2435-2442. 103. Jones S.M.A. and Rogers M.P. Reconstituted high density lipoprotein behavior is modified by the nature of the lipid vesicle. FEBS. Lett. 1985;180:66-68. 104. Havel R.J., Eder H.A. and Bragdon J.H. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 1955;34:1345-1353. 105. Allain C.C., Poon L.S., Chan C.S. and Richmond W~. Enzymatic determination of total serum cholesterol. Clin. Chem.. 1974;20:470-475. 106. Bucolo G. and David H. Quantitive determination of serum triglycerides by the use of enzymes. Clin. Chem. 1973;19:476-482. 107. Bligh E.G. and Dyer W.J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 1979;37:911-915. 108. Anderson R.L. and Davis S. An organic phosphorus assay which avoids the use of hazardous perchloric acid. Clin. Chim. Acta. 1982;121:111-6. 109. Peterson G.L. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem. 1977;83:346-356. 110. Gidez L.I., Miller G.J., Burnstein M. , Slagle S. and Eder H.A. Separation and quantitation of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J Lipid Res 1982;23:1206-1223. 111. Friedewald W.T., Levy R.I. and Fredrickson D.S. Estimation of the plasma low density lipoprotein cholesterol concentration without use of the preparative ultracentrifugation. Clin. Chem. 1972;18:499-502. 114 112. Warnick G.R., Mayfield C , Albers J.J., and Hazzard W.R. Gel isoelectric focusing method for specific diagnosis of familial hyperlipoproteinemia type 3. Clin. Chem. 1979;25:279-84. 113. Scanu A. Forms of human serum high density lipoprotein protein. J. Lipid Res. 1966;7:295-306. 114. Scanu A.M. and Edelstein C. Solubility in aqueous solutions of ethanol of the small molecular weight peptides of the serum very low density and high density lipoproteins. Analytical Biochem. 1971;44:576-588. 115. Foreman J.R., Karlin J.B., Edelstein C , Juhn D.J., Rubenstein A.H. and Scanu A.M. Fractionation of human serum lipoproteins by single-spin gradient ultracentrifugation: quantification of apolipoproteins B and A-I and lipid components. J. Lipid Res. 1977;18:759-767. 116. Forte T.M. and Nordhausen R.W. Electron microscopy of negatively stained lipoproteins. Methods in Enzymology 1986;128:442-457. 117. Mayer L.D., Hope M.J. and Culiis P.R. Vesicles of variable size produced by a rapid extrusion procedure. Biochim. Biophys. Acta 1986;858:161-168. 118. Hamilton J.A. and Morrisett J.D. Nuclear magnetic resonance studies of lipoproteins. Methods in Enzymology 1986;128:472-515. 119. Axen R., Porah J. and Ernback S. Chemical coupling of peptides and proteins to polysaccharides by means of cyanogen halides. Nature (London) 1967;214:1302-4. 120. Folch J Lees M. and Stanley G.H.S. A simple method for the purification of total lipides form animal tissues. J. Biol. Chem. 1956;222:497-509. 121. Lacko A.G. , Rutenberg H.L. and Soloff L.A. Measurement of the initial rate of serum cholesterol esterification. Biochem Med. 1973;7:178-183. 122. Batzri S. and Korn E.D. Single bilayer liposomes prepared without sonication. Biochim Biophys Acta 1973;298:1015-1019. 123. Albers J.J., Chen C.H. and Lacko A.G. Isolation, characterization, and assay of lecithin-cholesterol acyltransferase. Methods in Enzymology 1986;129:763-783. 124. Cornish-Bowden A. Fundamentals of enzyme kinetics. London: Butterworth and Co (Publishers) Ltd., 1979:pp.l-223. 125. Rajaram O.V. and Barter P.J. Reactivity of lipoproteins with purified lecithinxholesterol acyltransferase during incubations in vitro. Biochim. Biophys. Acta. 1985;835:41-9. 126. Pritchard P.H., McLeod R., Frohlich J. and Lacko A. Lecithinxholesterol acyltransferase in familial HDL deficiency (Tangier Disease). Biochim. Biophys. Acta. 1988;958:227-234. 127. Frohlich J., Hoag G., McLeod R., Hayden M. , Godin D.V., Wadsworth L.D., Critchley D. and Pritchard P.H. Hypoalphalipoprotein resembling fish eye disease. Acta Med. Scand. 1987;221:291-8. 128. Park M . C , Kudchodkar B.J., Frohlich J., Pritchard H. , and Lacko A. Study of the components of reverse cholesterol transport in lecithinxholesterol acyltransferase deficiency. Arch. Biochem. Biophys. 1987;258:545-554. 115 129. Bystrov V.F., Dubrovina N.I., Barsukov L.I. and Bergleson L.D. NMR differentiation of the« internal .and external phospholipid membrane surfaces using paramagnetic Mn and Eu ions. Chem. Phys. Lipids 1971:6;343-350. 130. Pownall H.J., Van Winkle B., Pao Q., Rohde M. and Gotto A.M. Action of lecithinxholesterol acyltransferase on model lipoproteins. Preparation and characterization of model nascent high density lipoprotein. Biochim. Biophys. Acta 1982:713;494-503. 131. Calvert G.D. and Carlson L.A. Plasma lipid transfer in fish-eye disease. Acta Med. Scand. 1983;213:253-4. 132. Schriewer H., Schulte H. and Assmann G. HDL phosphatidylcholine and risk factors of coronary heart disease. J. Clin. Chem. Clin. Biochem. 1984;22:515-521. 133. Taskinen M.R. and Kuusi T. High density lipoproteins in postprandial lipemia relation to sex and lipoprotein lipase activity. Atherosclerosis 1986;59:121-130. 134. Patsch J.R., Prasad S., Gotto A.M. and Patsch W. Relationship of the plasma levels of this lipoprotein species to its composition, to the magnitude of postprandial lipemia, and to the activities of lipoprotein lipase and hepatic lipase. J. Clin. Invest. 1987;80:341-347. 135. Breckenridge W.C., Little J.A. and Gardiner G. Lipoprotein abnormalities associated with a familial deficiency of hepatic lipase. Atherosclerosis 1982;45:161-179. 136. Carlson L.A., Holmquist L. and Nilsson-Ehle P. Deficiency of hepatic lipase activity in post-heparin plasma in familial hyper-a-triglyceridemia. Acta Med. Scand. 1986;219:435-447. 137. Hopkins G.J. and Barter P.J. Role of triglyceride-rich lipoproteins and hepatic lipase in determining the particle size and composition of HDL. J. Lipid Res. 1986;27:1265-1277. 138. Glass C , Pittman R., Keller G. and Steinberg D. Tissue sites of degradation of apoprotein A-I in the rat. J. Biol. Chem. 1983;258:7161-7167. 139. Dory L. , Boquet L., Hamilton R., Sloop C. and Roheim P. Heterogeneity of dog interstitial fluid (peripheral lymph) high density lipoproteins: implications for a role in reverse cholesterol transport. J. Lipid Res. 1985;26:519-527. 140. Kostner G.M., Molinari E. and Pichler P. Evaluation of a new H D L 2 / H D L 3 quantitation method based on precipitation with polyethylene glycol. Clin. Chem. Acta 1985;148:139-147. 141. Picard M., Massey J.B., Kuhn D.E., Gotto Jr. A .M. , Gianturco S.H., and Pownall H.J. Partially reassembled high density lipoproteins. Arteriosclerosis 1986;6:434-441. 142. Cullis P.J., and Hope M.J. Physical properties and functional roles of lipids in membranes. In: Vance D.E. and Vance J.E., eds. Biochemistry of lipids and membranes. California: The Benjamin/Cummings Publishing Co. Inc, 1985:pp.25-72. 143. Bloch K. Cholesterol: evolution of structure and function. In: Vance D.E. and Vance J.E., eds. Biochemistry of lipids and membranes. California: The Benjamin/Cummings Publishing Co. Inc, 1985:pp.l-24. 116 144. Miller K.W. and Small D.M. Triolein-cholesteryl oleate-cholesterol-lecithin emulsions: structural models of triglyceride-rich lipoproteins. Biochemistry 1983;22:443-451. 145. Miller K.W. and Small D.M. The phase behavior of triolein, cholesterol, and lecithin emulsions. J. Colloid Interface Sci. 1982;89:466-478. 146. Fielding C.J. Human lipoprotein lipase inhibition of activity by cholesterol. Biochim. Biophys. Acta. 1970;218:221-6. 147. Kuksis A., Myher J.J., Geher K., Jones G.L., Breckenridge W.C., Feather T., Hewitt D. and Little J.A. Decreased plasma phosphatidylcholine/free cholesterol ratio as an indicator of risk for ischemic vascular disease. Arteriosclerosis 1982;2:296-302. 117 APPENDIX A 118 20 0 30 60 90 120 150 180 0 100 200 300 400 500 600 TIME (min) VOLUME LP DP/ASSAY (uL) Figure 1. Characterization of the transfer of cholesteryl ester from L D L to HDL in lipoprotein depleted plasma 300 ul of LPDP was incubated with [ 3H]LDL and HDL (30 ug total cholesterol each) and buffer A (total volume 700 uL) at 37°C for 0-180 min (A). Incubations for experiment B were identical except the volume of LPDP was varied as indicated and the incubation time was 1.5h. The percent [ HJcholesteryl ester transferred to HDL was calculated from the amount of radioactivity remaining after precipitation of LDL with heparin/MnC^. Control incubations that lacked a source of cholesteryl ester transfer activity were subtracted to determine CETP dependent lipid transfer. Values are the average of duplicate determinations. 119 Figure 2. Effect of substrate concentration on cholesteryl ester transfer from L D L by lipoprotein depleted plasma 300 uL of LPDP was incubated with [ 3H]LDL (30 ug total cholesterol) and the indicated amount of HDL (panel A), or with HDL (30 ug total cholesterol) and the indicated amount of [3H] LDL (Panel B) at 37°C for 60 min. Transfer rates for panel A were determined as in Figure 1 and blanks which lacked exogenous HDL were subtracted from indicated values. Mass transfer rates for panel B were calculated from the initial specific radioactivity of the donor [ 3H]LDL. Each data point represents the average of duplicate determinations. 120 PARTLY PURE CETP ( u g / m L ) Figure 3. E f f e c t of partly pure C E T P on cholesteryl ester transfer from L D L . Cholesteryl ester transfer activities were determined in mixtures containing [ HJLDL and HDL (30 ug total cholesterol each) and the indicated amount of 2000 fold purified CETP (section 3.3). After a 1.5 h incubation at 37°C, transfer rates were determined and are expressed as the average of duplicate determinations. 121 TIME ( m i n ) PLASMA VOLUME ( m L ) Figure 4. Effect of incubation time and plasma concentration on cholesteryl ester transfer from solid-phase-bound HDL to plasma lipoproteins. 800 uL of plasma was incubated with 200 ug of solid-phase-bound HDL protein at 37 °C for 0-120 min (A) or with the indicated volume of plasma and NaCl/Tris buffer (total volume, 2 mL) for 60 min (B). The solid-phase-bound HDL was removed by centrifugation and the radioactivity in a 200 uL aliquot of supernate was determined. Results are expressed as the percentage of the initial [JH]cholesteryl ester that was transferred to plasma. Values are means+SD of three normolipidemic individuals (A) and means of duplicate determinations (B). 122 25-• — • p lasma d<1.21 g / m L - i 1 1 1 1 r 0 30 60 90 120 150 180 TIME (min) 25 20-15-10-• — • p l a s m a » - • d<l.21 g / m L B LDL&VLDL 90 120 150 180 TIME (mm) Figure 5. Cholesteryl ester transfer from solid phase bound HDL to plasma and the d<1.21 g/mL fraction. 800 uL of whole plasma or the d<1.21 g/mL fraction was assayed as described in Figure 7A. The percent [3H]cholesteryl ester transfer to HDL (A) was determined by precipitation of LDL/VLDL with heparin/MnCl2> followed by supernate radioactivity determination. Transfer to LDL+VLDL (B) was determined by subtracting the value for HDL from the total percentage transferred. Values are means+SD of three normolipidemic individuals (plasma) or the average of duplicate determinations (d<1.21 g/mL). 123 ESS PLASMA LZZ1 LP-MIX Figure 6. Transfer of cholesteryl ester to plasma and recombined lipoproteins. The rate of cholesteryl ester transfer from solid-phase-bound HDL to a mixture of VLDL, LDL, and HDL (per mL, 0.1, 1.0, and 1.8 mg of the respective lipoprotein proteins) in human serum albumin, 10 mg/mL, was determined after 60 min incubations at 37°C and compared with the rate of transfer (mean+SD) to, normal plasma (n=7). Values for the lipoprotein mixtures are the mean+SD of quadruplicate determinations. 124 PARTLY PURE CETP ( u g / m L ) > < I -UJ o 0.0 0.2 0.4 0.6 0.8 SEPHAROSE—HDL ( m g p r o t e i n / m L ) Figure 7. Effect of partly pure CETP and solid phase bound HDL concentration on cholesteryl ester transfer. 800 uL plasma was incubated with 200 ug of solid-phase-bound HDL protein and 0-220 ug of 250-fold purified CETP for 60 min at 37 °C (A). Incubation conditions were identical for experiment B, however, the solid phase substrate was varied as indicated and no exogenous CETP was added. Transfer rates in A were determined as described in Figure 7, while mass transfer rates in B were calculated from the initial specific radioactivity of the donor solid-phase-bound HDL. Values are the average of duplicate determinations. 125 TIME (min) Figure 8. Effect of L C A T inactivation on transfer activity. Cholesteryl ester transfer from solid-phase-bound HDL to 800 uL of plasma was determined as described for Figure 7A in the presence or absence of 1 mM diisopropyl fluorophosphate (DIFP). Values are the average of duplicate determinations. 126 TIME (min) Figure 9. Effect of plasma storage at 4°C on cholesteryl ester transfer from solid phase bound HDL. Cholesteryl ester transfer from solid-phase-bound HDL to 800 uL of plasma was determined as described in Figure 4A after 2 and 4 weeks storage at 4°C and compared to that for fresh plasma. Values are the average of duplicate determinations. 127 5-E 4-c n > O < 3 2-Ld O 1 0 J ^ H D L Z D L D L / V L D L 7 22 37 42 O T E M P E R A T U R E ( C ) Figure 10. Effect of incubation temperature on cholesteryl ester transfer from solid-phase-bound HDL. Cholesteryl ester transfer from solid-phase-bound HDL to 800 uL of plasma was determined after 60 min incubations at the indicated temperatures. Mass transfer rates to HDL and LDL+YLDL were determined from the initial specific radioactivity of the donor solid-phase-bound HDL and are expressed as the average of duplicate determinations. 128 DENSITY (g /mL) Figure 11. Localization of CETP and LCAT in plasma at a density of d=1.21 g/mL. Fresh plasma was adjusted to a density of 1.21 g/mL by the addition of solid NaBr and ultracentrifuged for 48 h at 114,000 x g. The contents of each tube were pumped out from the bottom and collected in 14 fractions. The density, CETP, and the L C A T activities were determined in each fraction. CETP activity was determined by utilizing [ 3H]CE-LDL as a donor, HDL as an acceptor, and 300 uL of each fraction (section 2.4.1.1). LCAT activity was determined using single bilayer vesicle substrates and 50 uL of each fraction (section 2.4.2.1). Values are the average of duplicate determinations. 129 FRACTION NUMBER FRACTION NUMBER Figure 12. Phenyl-Sepharose column chromatography of CETP and L C A T activities. LPDP was loaded onto a Phenyl-Sepharose CL-4B and washed extensively with buffer A. Bound proteins were subsequently eluted with H 2 0 and 40 x 20 mL fractions were collected at a flow rate of 40 mL/h. The A^OQ (solid line), NaCl concentration (determined from conductivity), C E T F activity (A) and L C A T activity (B) were determined in alternate fractions. CETP and L C A T activities were determined as described in Figure 11 and sections 2.4.1.1 and 2.4.2.1 respectively. 130 -0.5 -0.4 E > ••3-D o to <u £ o OLi-_ l _ o E c -0.3 O 03 -0.2 -0.1 i 10 20 30 40 50 60 ' 70 FRACTION NUMBER 80 100 r250 -200 -150^ Q O L 5 0 r250 -200 -150^ •100^  •50 FRACTION NUMBER Figure 13. CM-cellulose column chxomatography of CETP and L C A T activities. Pooled CETP/LCAT active fractions from the Phenyl-Sepharose column were dialyzed against J^O, equilibrated with 0.01M sodium acetate (pH 4.5), and loaded onto a CM-52 cellulose column at a flow rate of 40 mL/h. Unretained protein was eluted with 150 mL of the same buffer and then bound proteins were eluted with a linear 0-400 mM NaCl gradient. All fractions (6 mL) were collected into tubes containing 0.5 mL 1M Tris (pH 7.4) to neutralize the pH. The A2gn (solid line), NaCl concentration, CETP activity (A) and LCAT activity (B) were determined as described in Figure 12. 131 - . 0 9 - . 0 6 - . 0 3 0.00 0.03 0.06 0.09 1/[HDL] Figure 14: Kinetic analysis of lipid transfer to HDL subfractions. Data from incubations described in Figure 22 is presented as double reciprocal plots where V = CET activity (%/mL/h) and [HDL] = HDL concentration (ug protein/mL). Lines are drawn through the K H and T m a x estimated from the indicated data (section 2.8.2). 132 CETG 0:1 / • A 0.30- • / * / / f 1 : 1 1 0.20- / 4:1 1:0 V • i 0^r> ' • i • 1 | 1 1 | -.09 -.06 -.03 0.00 0.03 0.06 0.09 1 / [rHDL] FC:PL / 1:0.6 B 1 0.40-0.30-• / j . 1:1.1 / • y 1:1.9 i • • i V i 0.20-i i ^ » 1:4.6 i I i i I -.09 -.06 ' -.03 0.00 0.03 0.06 0.09 1 / [rHDL] Figure 15: Kinetic analysis of lipid transfer to rHDL with various lipid compositions. Data from incubations containing rHDL with the indicated molar ratios of cholesteryl ester to triglyceride (A, described in Figure 23) or free cholesterol to phospholipid (B, described in Figure 24) are presented as double reciprocal plots where V = CET activity (%/mL/h) and [HDL] = HDL concentration (ug protein/mL). Lines are drawn through the Kjj and T m a x estimated from the indicated data (section 2.8.2). 133 -.09 -.06 -.03 0.00 0.03 0.06 0.09 1 / [ r H D L ] Figure 16: Kinetic analysis of lipid transfer to rHDL with various apoprotein compositions. Data from incubations containing rHDL with the indicated mixtures of apoproteins is presented as double reciprocal plots where V = CET activity (%/mL/h) and [HDL] = HDL concentration (ug protein/mL). Lines are drawn through the K J J and T m a x estimated from the indicated data (section 2 .8 .2) . 134 Daniel L. SPARKS PUBLICATIONS 1. Sparks D, Frohlich J & Pritchard PH. "The assay of cholesteryl ester transfer in plasma using solid phase high density lipoprotein", Clin Chem 33: 390-393 (1987) 2. Sparks DL, Frohlich J and Pritchard PH. "Cholesteryl ester transfer activity in plasma of patients with familial HDL deficiency". Clin Chem 34: 1812-1815 (1988) 3. Parmar YI, Sparks DL, Cullis PR and Pritchard PH. "The detection of vesicular lipoproteins in LCAT deficient plasma by 1H-NMR spectroscopy". J Lipid Res 30: in press (1989) 4. Sparks DL, Frohlich J, Lacko AG and Pritchard PH. "The relationship between cholesteryl ester transfer activity and high density lipoprotein composition in hyperlipidemic patients". Atherosclerosis, in press (1989) 5. Sparks DL and Pritchard PH. "The neutral lipid composition of recombinant high density lipoproteins regulates lecithinxholesterol acyltransferase activity". Biochem Cell Biol, in press (1989) 6. Sparks DL, Pritchard PH. "Transfer of cholesteryl ester into high density lipoprotein by cholesteryl ester transfer protein: Effect of HDL lipid and apoprotein content". J Lipid Res, in press (1989) 7. Frohlich J, Westerlund J, Sparks D and Pritchard PH. "Familial Hypoalphalipoproteinemias". Clin Invest Med, in press (1989) 8. Samborski RS, Sparks DL, Frohlich J and Pritchard PH. "The exchange of cholesteryl ester for triglyceride in normal high lipoproteins incubated in the plasma of a patient with Tangier disease". Arteriosclerosis, submitted (1989) 9. Sparks DL and Pritchard PH. "Generation and characterization of recombinant high density lipoprotein for use as substrates for lecithinxholesterol acyltransferase and cholesteryl ester transfer protein". J Lipid Res, submitted (1989) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098201/manifest

Comment

Related Items