Plasma exchange, which was first introduced over 10

Transcription

1 Plasma Exchange and Low Density Lipoprotein Apheresis in Watanabe Heritable Hyperlipidemic Rabbits Masakats Kano, Jnji Koizmi, Arvind Jadhav, and Gilbert R. Thompson Comparison of the effects of plasma exchange sing llpoproteln-deflclent plasma with those of LDL apheresis sing a dextran slphate colmn was ndertaken In two grops of Watanabe heritable hyperlipidemic rabbits pre-labelled with 3 H-cholesterol. Total and HDL cholesterol were redced more by plasma exchange bt HDL cholesterol rebonded higher after LDL apheresis; rises In HDL cholesterol correlated with preceding decreases In both HDL and total cholesterol. An Increase In the specific activity of HDL cholesterol occrred on the day after each procedre, being more marked after plasma exchange, and was accompanied by a decrease In the cholesterol/phosphollpld ratio of HDL. These reslts sggest that an Inflx of extravasclar HDL Into plasma occrred after both procedres, reslting In mobilization of tisse cholesterol. (Arteriosclerosis 7: , May/Jne 1987) Downloaded from by on Janary 6, 2019 Plasma exchange, which was first introdced over 10 years ago to treat homozygos familial hypercholesterolemia (FH), reslts in gradal regression of ctaneos and tendon xanthomata. 12 Indirect evidence that this reflects mobilization of tisse cholesterol was obtained by pre-labelling patients' tisses with radioactive cholesterol and demonstrating acte rises in the specific activity of plasma cholesterol after plasma exchange. 1 In one instance analysis of individal lipoproteins showed this rise to be localized to high density lipoprotein (HDL) cholesterol. 3 An Inverse relationship between the concentrations of HDL cholesterol in plasma and the size of the exchangeable pool(s) of tisse cholesterol has been proposed 4 and the possible role of HDL in transporting cholesterol from tisses into plasma was reviewed recently. 5 Several, 6 " 8 bt not all, 9 epidemiological srveys have shown an increased risk of developing coronary heart disease (CHD) in those with low HDL cholesterol levels, and angiographic data sggest that HDL cholesterol is inversely correlated with the severity of coronary arterial lesions, 10 ' 11 as well as with the likelihood that these will progress In the light of these findings and becase plasma exchange removes HDL as well as low density lipoprotein (LDL), techniqes designed to conserve HDL dring LDL removal have been developed. These inclde exposing whole blood to heparin-agarose beads 14 and perfsing plasma throgh affinity colmns containing LDL antibodies 15 or dextran slphate. 18 Each method specifically From the Medical Research Concil Lipoprotein Team, Hammersmith Hospital, London, United Kingdom. Jnji Koizmi was in receipt of an Overseas Visiting Fellowship from the British Heart Fondation. Address for reprints: Dr. Gilbert R. Thompson, MRC Lipoprotein Team, Hammersmith Hospital, Dcane Road, London W12 OHS, United Kingdom. Received September 29,1986; revision accepted Decembers, removes LDL bt not HDL, a process known collectively as LDL apheresis. The present stdy compares the effects of plasma exchange and LDL apheresis in Watanabe heritable hyperiipldemic (WHHL) rabbits, an animal model of FH. 17 Methods WHHL rabbits were bred locally from homozygos males and heterozygos females, kindly provided by Dr. Y. Watanabe. The progeny were classified as heterozygotes or homozygotes according to their serm cholesterol levels while on a normal diet; those with a vale above 100 mg/dl at 3 months were regarded as homozygotes, the remainder as obligate heterozygotes. The former remained on a normal diet bt heterozygotes received spplements that increased the cholesterol content of their diet to 0.2% for the 4 months prior to the start of the stdy. Dring this period blood was obtained from an ear vein each month for determination of serm cholesterol, and only heterozygotes with cholesterol levels that had risen into the homozygote range were sed. The body weight of the rabbits sed in these stdies was 3.5 to 4.5 kg. All were female. Each rabbit received an intravenos injection of (1a, 2an 3 H) cholesterol (Amersham International, United Kingdom) 100 to 600 ^Ci, which was first dissolved in ethanol and then mixed with 0.5 ml of Upostabil (Natterman, Cologne, West Germany) and 2 ml of atologos plasma, at 15 to 22 days before ndergoing plasma exchange or LDL apheresis. This interval was chosen after a preliminary stdy had shown that the specific activity of cholesterol in the plasma of a normocholesterolemic New Zealand white (NZW) rabbit became lower than that of skin, adipose tisse, mscle and red blood cells by 2 to 3 weeks after the injection of 3 H-cholesterol. At 21 days the specific activity of cholesterol in liver was similar to plasma bt that in the aorta was mch lower (Figre 1). Prior to ndertaking plasma exchange, blood was ob-

2 PLASMA EXCHANGE AND LDL APHERESIS IN WHHL RABBITS Kano et al. 257 Downloaded from by on Janary 6, 2019 Plasma A Adipose tisse Skin Okldney omscle Oliver Oior DAYS AFTER INTRAVENOUS INJECTION OF 3 H CHOLESTEROL Figre 1. Changes in specific activity of cholesterol In plasma, adipose tisse, and skin of a normal New Zealand white rabbit following an injection of 3 H-cholesterol. The specific activities of cholesterol in kidney, mscle, liver, and aorta at the time of sacrifice on Day 21 are also shown. tained from NZW rabbits and mixed with 0.01% EDTA for preparation of lipoprotein-deficient plasma (LPDP). Plasma was separated and ltracentrifged at 55,000 rpm in a Beckman 70 Tl rotor for 40 hors at 10 C after adjsting the density to with solid KBr. After tbe-slicing, the inf ranatant was dialyzed against 0.15 M NaCI containing 3 mm K+, then sterilized by Milllpore filtration (0.22 ^m) and kept at 4 C ntil se. LPDP prepared in this manner had a cholesterol content of 3 to 4 /og/ml. Dring plasma exchange or LDL apheresis, rabbits were anesthetized with fentanyl and flanisone (Janssen Pharmaceticals, Oxford) and were heparinized. These procedres were performed in a manner similar to that described by Yokoyama et al 18 sing an Apex II device (Kawasmi Laboratories, Tokyo, Japan) fitted with a Minicre S-5 filter (Kraray Limited, Osaka, Japan) to separate plasma and red cells. Blood was obtained from an artery in one ear and retrned via a vein in the opposite ear, at a flow rate of 6 to 10 ml/min. Dring plasma exchange plasma was removed and replaced with an eqivalent volme of LPDP, whereas dring LDL apheresis plasma was perfsed throgh an LA 01 colmn (Kanegafchi Chemical Indstry Limited, Osaka, Japan), containing dextran slphate linked to celllose, and then retrned to the rabbit. These colmns were 22 mm in diameter, 70 mm in length and contained 25 ml of wet-packed adsorbent. 18 The latter has the capacity to bind 10 to 12 mg cholesterol per ml, as VLDL and LDL, bt does not bind HDL, or albmin or Immnogloblins. 16 Experimental details of the 12 procedres that were sccessflly completed and that inclded at least 5 days of sbseqent observations are shown in Table 1. Blood samples were obtained at intervals before and after each procedre for determination of the concentration and specific activity of total and HDL cholesterol in plasma, and also the concentration of total and HDL phospholipid. HDL was obtained by adding 0.4% phosphotngstic acid and 2 mm MgCI 2 to plasma immediately after this had been separated from fresh blood samples at 4 C, 19 followed by centrifgation and filtration of the trbid spernatant. 20 Measrement of HDL cholesterol by this method gave mean ± SE vales of 19.8 ± 7.1 and 7.0 ± 0.5 mg/dl respectively in six NZW and six WHHL rabbits compared with vales of 18.9 ± 3.9 and 7.6 ± 3.1 mg/dl obtained by assaying the infranatant of samples ltracentrifged at d = Cholesterol was measred by GLC 21 and phospholipid by enzymatic assay (Wako Chemicals, West Germany). Radioactivity was determined in a Beckman LS 1801 liqid scintillation conter, after dissolving 0.05 to 0.2 ml of sample in 10 ml Beckman Ready Solve MP. Tisse samples were extracted with chloroform/methanol, 2M (vol/vol) 22 and the chloroform phase was separated, dried down, and saponified before determination of specific activity. Reslts The possibility that the lipoprotein particles present in the plasma of cholesterol-fed rabbits pass throgh the plasma separator less easily or adsorb to the LA 01 colmn less well than those in the plasma of WHHL homozygotes 18 was catered for by constrcting two similar grops of rabbits, each comprising for homozygotes and two cholesterol-fed heterozygotes. Changes In Plasma Cholesterol Concentrations The effects of plasma exchange and LDL apheresis on serm total and HDL cholesterol concentrations are shown in Figre 2. Mean vales of serm total and HDL cholesterol were similar in the two grops before each procedre althogh levels tended to be higher in cholesterol-fed heterozygotes than in homozygotes. Percentage decreases in total cholesterol were slightly greater after plasma exchange than after LDL apheresis (66% and 59%, respectively), despite the larger volme of plasma processed dring the latter procedre (see Table 1). As expected, the percentage " TOTAL CHOLESTEROL LDL APHERESIS PLASMA EXCHANGE Figre 2. Concentration of plasma total and HDL cholesterol In Watanabe heritable hyperllpldemic homozygotes ( ) and cholesterol-fed heterozygotes (o) before and after LDL apheresis and plasma exchange.

3 258 ARTERIOSCLEROSIS VOL 7, No 3, MAY/JUNE 1987 Table 1. Experimental Details Grop (n) LDL apheresis WHHL Hmz* (4) WHHL CF-Htzf (2) Plasma exchange WHHL Hmz* (4) WHHL CF-Htzf (2) Weight (kg) Plasma processed (ml) Heparin given (I.U.) "Watanabe heritable hyperlipidemic rabbit homozygote, tchclesterol-fed heterozygote. 1 3 /' i 7 DAVS Downloaded from by on Janary 6, 2019 decrease in HDL cholesterol was mch more marked after plasma exchange (62%) than after LDL apheresis (28%). Seqential changes in total and HDL cholesterol dring the week after each procedre are shown in Rgres 3 and 4. The greater decrease in total cholesterol after plasma exchange was followed by a faster rate of rebond than after LDL apheresis (58 mg/dl/d verss 32 mg/dl/d). In contrast, althogh HDL cholesterol fell less after LDL apheresis, the sbseqent rise above baseline on Day 5 exceeded that seen after plasma exchange (27% verss 8%). Rises in HDL cholesterol concentration after each procedre, expressed as the difference between immediate post-procedre and sbseqent peak vales, correlated well with preceding decreases in total and HDL cholesterol, as shown in Figre 5. Correlation coefficients were higher when increments in HDL cholesterol were correlated with previos decreases in HDL cholesterol (B) rather than total cholesterol (A) bt the differences were slight. Changes In Specific Activity The effects of LDL apheresis on the specific activities of total and HDL cholesterol in plasma are illstrated in Figre 6. The specific activity of HDL cholesterol became higher than that of total cholesterol dring the first few days PLASMA EXCHAKCE Figre 4. Changes in HDL cholesterol after LDL apheresis and plasma exchange, expressed In a manner similar to Figre 3. after injection of 3 H-cholesterol and rose even more after LDL apheresis. Similar changes occrred after plasma exchange, bt to an even more marked extent, as shown in Figre 7. The rise in the relative specific activity of HDL cholesterol was evident within 2 hors after plasma exchange bt was not apparent ntil 6 hors after LDL apher io t 5 U _J Q I -1 O to O a. 15 APH (r= A TOTAL CHOLESTEROL (PRE - POST) MC/DL 500 _l a x Figre 3. Changes in plasma total cholesterol after LDL apheresis and plasma exchange. Mean ± SE vales are shown relative to the pre-exchange vales, inclding the average level dring the week before each procedre HDL-C (PRE - POST) MC/DL Figre 5. Correlations between the rebond in HDL cholesterol in Watanabe heritable hyperlipidemic homozygotes (circles) and cholesterol-fed heterozygotes (triangles) after plasma exchange (PE, open symbols) or LDL apheresis (APH, closed symbols) and the preceding decrease in HDL or total cholesterol. The correlation coefficients (r) for each procedre are shown. 25

4 PLASMA EXCHANGE AND LDL APHERESIS IN WHHL RABBITS Kano et al. 259 Downloaded from by on Janary 6, 2019 o i a o i > I- <. 10 plasma O 0 HDL DAYS AFTER INTRAVENOUS INJECTION OF J H CHOLESTEROL Figre 6. Effect of LDL apheresis on specific activity of plasma total and HDL cholesterol in a Watanabe heritable hyperiipidemic homozygos rabbit injected with 3 H-cholesterol 21 days previosly. esis, becoming maximal at 24 hors. In contrast, the relative specific activity of total cholesterol showed a slight decrease dring the 6 hors after each procedre. Changes in HDL Composition Changes in HDL composition, as jdged from the cholesterol/phospholipid ratio, occrred after both plasma exchange and LDL apheresis. However, baseline HDL com- O a. I o Z LDL APHERESIS (n=6) PLASMA EXCHANCE (n=6) Table 2. HDL Cholesterol/Phosphollpid Mass Ratios after Plasma Exchange or LDL Apheresis Time Baseline 6 hors 1 day 2 days 3 days 5 days Vales are mean ± SE. Homozygotes (n = 4) 0.18± ± ± ± ±0.02 Cholesterol-fed heterozygotes (n = 4) 0.49 ± ± ± ± ± ±0.05 position differed markedly between homozygotes and cholesterol-fed heterozygotes and this, rather than the type of procedre ndergone, was the major determinant of sbseqent changes in composition. As shown in Table 2 the mass ratio of cholesterol/phospholipid in HDL was mch higher in cholesterol-fed heterozygotes than in homozygotes. Decreases became evident on the first and second days after plasma exchange or LDL apheresis, the change being relatively slight in homozygotes bt qite marked in cholesterol-fed heterozygotes. Changes in this ratio on Day 1 were negatively correlated with changes in HDL specific activity, when each variable was expressed as a percentage of its pre-exchange vale (r = ), as shown in Figre 8. Ratios retrned to near their initial vales between the third and fifth days after either procedre. Discssion These stdies set ot to compare the respective capabilities of plasma exchange and LDL apheresis to indce mobilization of tisse cholesterol. The greater redction in total cholesterol achieved by plasma exchange sggests that it is more efficient than a single LA 01 colmn at removing LDL from a given volme of plasma. This presmably reflects the limited capacity of the colmn to bind cholesterol, which is estimated at 1 g per 100 ml of adsorbent. 18 However, LA 01 colmns do not bind HDL, the small redctions in the latter which were observed being A HDL %p. act (Day I - Prat. 1 < HDL-C 50-1 Figre 7. Change in relative specific activities (means ± SE) of total cholesterol (TC) and HDL cholesterol (HDL-C) after LDL apheresis and plasma exchange, expressed as a percentage of the specific activity of total cholesterol immediately before each procedre. - a -w -20 -to HDL C/P {Day I - Prt, 1 Figre 8. Correlation between changes in HDL cholesterol specific activity and cholesterol/phospholipid (C/P) ratio on the day after plasma exchange or LDL apheresis in Watanabe heritable hyperiipidemic homozygos ( ) and cholesterol-fed heterozygos rabbits ( ).

5 260 ARTERIOSCLEROSIS VOL 7, No 3, MAY/JUNE 1987 Downloaded from by on Janary 6, 2019 de to losses of whole plasma dring the filling and emptying of the extracorporeal circit. Evidence that HDL plays a role in transporting tisse cholesterol 5 and the knowledge that more HDL is lost dring plasma exchange than dring LDL apheresis sggested that the latter procedre might be the more advantageos. This likelihood was strengthened by observations that LDL apheresis of whole blood sing heparin-agarose 23 or of plasma sing an immnoaffinity colmn 24 reslts in an evental increase in HDL cholesterol after each procedre. On the other hand, the proven ability of plasma exchange to indce regression of xanthomata 2 and, possibly, atheroma 2-25 implies that it possesses considerable cholesterol-mobilizing ability, despite the loss of HDL entailed. Administration of radioactive cholesterol has long been sed to label tisse cholesterol in both hmans 26 and experimental animals. 27 Radioactivity In plasma eqilibrates rapidly with cholesterol in liver and red cells, more slowly with fat and mscle, and least rapidly with the arterial intjma. 26 By waiting long enogh, it is possible to ensre that all tisses other than brain and arteries will have a higher specific activity than plasma. Under these circmstances, any sbseqent rise in plasma specific activity mst indicate movement of cholesterol from tisses into plasma, de either to exchange or net transfer. If the rise follows manevers that do not entail the introdction into plasma of nlabelled cholesterol, then it can be assmed to indicate net transfer rather than exchange. This phenomenon was observed in WHHL rabbits both after plasma exchange with LPDP, which contains only very small amonts of cholesterol, and after LDL apheresis, sggesting that both procedres cased a net efflx of tisse cholesterol into plasma. Rises in HDL specific activity, which were maximal at 24 hors, were more marked after plasma exchange than after LDL apheresis, possibly reflecting the greater redction of the intravasclar pool of HDL. However, it is nlikely that mch of the mobilized cholesterol came from arteries since trnover of cholesterol there is so mch slower than in most other tisses, inclding xanthomata. 28 ' ^ Similar findings have been observed after plasma exchange in hmans 1 ' 3 and pigs 30 and after fasting in obese sbjects. 31 The rise in HDL specific activity dring the 2 days after each procedre was accompanied by a redction in the cholesterol/phospholipid ratio of HDL. Both findings are compatible with the entry into plasma of HDL derived from the extravasclar compartment, in that the HDL of peripheral lymph has a higher phospholipid content than HDL in plasma 32 and has been reported to be the major acceptor of free cholesterol derived from tisses. 33 The mechanism whereby plasma exchange with LPDP indces efflx of tisse cholesterol cold relate to its content of very high density lipoprotein and apo A-l, extrapolating from data obtained sing hman lipoprotein deficient serm (LPDS) in vitro. 34 ' x However, LDL apheresis did not involve administration of LPDP, bt despite this, the rebond in HDL cholesterol on Day 5 was even more marked than after plasma exchange. In both instances the magnitde of the overall increases in HDL cholesterol dring the 5 days after each procedre were proportional to the preceding decreases, which were greater after plasma exchange. The likelihood that removal of small amonts of HDL by LDL apheresis stimlates HDL synthesis has been discssed by Parker et al. 24 If tre, it wold be expected that HDL synthesis wold be stimlated to an even greater extent by plasma exchange, althogh direct evidence of this will be hard to obtain, owing to the non-steady-state conditions pertaining. If LDL apheresis and plasma exchange do indeed promote movement of HDL into plasma from the extravasclar compartment, this cold be secondary to the redction in non-hdl cholesterol that accompanies these procedres, since the rebond in HDL cholesterol was proportional to preceding decreases in not only HDL bt also total cholesterol, comprising mainly very low and low density lipoproteins. Recent data sggest that HDL is bond to cells by specific receptors, the expression of which is directly proportional to celllar cholesterol content. 36 Depletion of celllar cholesterol by manevers sch as LDL apheresis and plasma exchange might be expected to redce receptor-mediated binding of HDL in tisses and ths promote the movement of extravasclar HDL back into plasma. Whether postprocedral rises in HDL reflected the presence of apo E-rich particles was not determined. 37 On the basis of the evidence presented here, plasma exchange seems comparable to LDL apheresis in its ability to redce total cholesterol and ths promote efflx of tisse cholesterol, despite the loss of HDL entailed. However, the procedre was tolerated less well by WHHL rabbits than LDL apheresis, and early indications are that this preference also applies to man. For this and other reasons, sch as cost and safety, it seems likely that LDL apheresis may eventally spplant plasma exchange in the treatment of most forms of severe hypercholesterolemia. References 1. Thompson QR, Lowenthal R, Myant NB. Plasma exchange in the management of homozygos familial hypercholesterolaemia. Lancet 1975;1: Thompson QR, Myant NB, Kllpatrlck D, Oakley CM, Raphael MJ, Stelner RE. Assessment of long-term plasma exchange for familial hypercholesterolaemia. B. Heart J 1980; 43: Thompson QR. Plasma exchange for hypercholesterolaemia. Lancet 1981 ;1: Miller GJ, Miller NE. Plasma-high-density-lipoprotein concentration and development of Ischaemic heart disease. Lancet 1975;1:16 5. Miller NE. Crrent concepts of the role of HDL In reverse cholesterol transport. In: Miller NE, Miller GJ, eds. Clinical and metabolic aspects of high-density lipoproteins. Amsterdam: Elsevier, 1984; Rhoads QG, Glbrandsen CL, Kagan A. Serm lipoproteins and coronary heart disease in a poplation stdy of Hawaii Japanese men. New Eng J Med 1976;294: Castelll WP, Doyle JT, Gordon T, et al. HDL cholesterol and other lipids In coronary heart disease. The cooperative lipoprotein phenotyping stdy. Circlation 1977;55: Gordon T, Castelll WP, Hjortland MC, Kanrtel WB, Dawber TR. High density lipoprotein as a protective factor against coronary heart disease. Am J Med 1977;62: Pocock SJ, Shaper AG, Phillips AN, Walker M, Whrtebead TP. High density lipoprotein cholesterol is not a major risk factor for ischaemic heart disease in British men. Br Med J 1986;292:

6 PLASMA EXCHANGE AND LDL APHERESIS IN WHHL RABBITS Kano et al. 261 Downloaded from by on Janary 6, Jenkins PJ, Harper RW, Nestel PJ. Severity of coronary atherosclerosis related to lipoprotein concentration. Br Med J 1978;2: Pearson TA, Blkley BH, Achff SC, Kwiterovich PO, Gordis L. The association of low levels of HDL cholesterol and arteriographically defined coronary artery disease. Am J Epidemiol 1979;109: Nikkila EA, Viikinkoski P, Valle M, Frick MH. Prevention of progression of coronary atherosclerosis by treatment of hyperlipidaemia: a seven year prospective angiographic stdy. Br Med J 1984;289: Levy Rl, Brensike JF, Epstein SE, et al. The inflence of changes in lipid vales indced by cholestyramine and diet on progression of coronary artery disease: reslts of the NHLBI type II coronary intervention stdy. Circlation 1984; 69: Lplen P-J, Moorjani S, Awad J. A new approach to the management of familial hypercholesterolaemia: removal of plasma cholesterol based on the principle of affinity chromatography. Lancet 1976;1: Stoffel W, Borberg H, Greve V. Application of specific extracorporeal removal of low density lipoprotein in familial hypercholesterolaemia. Lancet 1981 ;2: Yokoyama S, Hayashi R, Satanl M, Yamamoto A. Selective removal of low density lipoprotein by plasmapheresis in familial hypercholesterolemia. Arteriosclerosis 1985;5: Watanabe Y. Serial inbreeding of rabbits with heriditary hyperlipidemia (WHHL-rabbit). Atherosclerosis 1980; 36: Yokoyama S, Hayashi R, Klkkawa T, et al. Specific sorbent of apolipoprotein B-containing lipoproteins for plasmapheresis. Characterization and experimental se in hypercholesterolemic rabbits. Arteriosclerosis 1984;4: Brsteln M, Scholnick HR. Lipoprotein-polyanion-metal interactions. Adv Lipid Res 1973;11: Warnlck GR, Albers JJ. Heparin-Mn qantitation of highdensity-lipoprotein cholesterol: an ltrafiltration procedre for lipemic samples. Clin Chem 1978;25: Bromhoff JP. Serm cholesterol determination by gas liqid chromatography. Clin Chim Acta 1973;43: Folch J, Lees M, Sloane-Stanley GH. A simple method for the isolation and prification of total lipids from animal tisses. J Biol Chem 1957;226: Lpien P-J, Moorjani S, Gagne C, Brn L-D, Lo M, Dagenais G. Long term treatment of two familial hypercholesterolemic heterozygote patients with batch affinity chromatography (BAC). Artery 1982;10: Parker TS, Gordon BR, Saal SD, Rbin AL, Ahrens EH. Plasma high density lipoprotein is increased in man when low Index Terms: cholesterol efflx activity HDL composition density lipoprotein (LDL) is lowered by LDL-pheresis. Proc Natl Acad Sci USA 1986;83: Berger GMB, Bonnili F, Joffe HS, Dbovsky DW. Plasma exchange in the treatment of familial hypercholesterolemia. In: Gotto AM, Smith LC, Allen B, eds. Atherosclerosis V. New York: Springer-Verlag 1980:458-^* Chobanlan AV, Hollander W. Body cholesterol metabolism in man. I. The eqilibration of serm and tisse cholesterol. J Clin Invest 1962;41: Motafis CD, Myant NB. The distribtion of [ 14 C] cholesterol in mscle and skin of Rhess monkeys after intravenos injection. Clin Sci Mol Med 1976;50: Jagannathan SN, Connor WE, Baker WH, Bhattacharyya AK. The trnover of cholesterol in hman atherosclerotic arteries. J Clin Invest 1974;54: Bhattacharyya AK, Connor WE, Masolf FA, Flatt AE. Trnover of xanthoma cholesterol in hyperlipoproteinemia patients. J Clin Lab Med 1976;87: Angel A, Thanabalaslngham S, Reichl D, Pflg JJ, Thompson GR, Myant NB. Effects of starvation and plasma exchange on lecithin :cholesterol acyltransferase activity and cholesterol efflx in cholesterol-fed pigs. Res Exp Med 1984;184: Nestel P, Miller NE. Mobilization of adipose tisse cholesterol in high density lipoprotein dring weight redction in man. In: Gotto AM, Miller NE, Oliver MF, eds. High density lipoproteins and atherosclerosis. Amsterdam: Elsevier 1978: Sloop CH, Dory L, Hamilton R, Krase BR, Rohelm PS. Characterization of dog peripheral lymph lipoproteins: the presence of a disc-shaped 'nascent' high density lipoprotein. J Lipid Res 1983;24: Reichl D, Myant NB, Rdra DN, Pflg JJ. Evidence for the presence of tisse free cholesterol in low density and high density lipoproteins of hman peripheral lymph. Atherosclerosis 1980;37: Stein O, Vanderhoek J, Stein Y. Cholesterol content and sterol synthesis in hman skin fibroblasts and rat aortic smooth msle cells exposed to lipoprotein-depleted serm and high density apolipoprotein/phospholipid mixtres. Biochim Biophys Acta 1976;431: Oram JF, Albers JJ, Cheng MC, Bierman EL. The effects of sbtractions of high density lipoprotein on cholesterol efflx from cltred fibroblasts. J Biol Chem 1981 ;256: Oram JF, Brlnton EA, Bierman EL. Reglation of high density lipoprotein receptor activity in cltred hman skin fibroblasts and hman arterial smooth mscle cells. J Clin Invest 1983;72: Miller NE, La Ville A, Crook D. Direct evidence that reverse cholesterol transport is mediated by high-density lipoprotein in rabbit. Natre 1985;314: lipoprotein-deficient plasma dextran slphate affinity colmn HDL specific