Impact of LDL apheresis on atheroprotective reverse cholesterol transport pathway in familial hypercholesterolemia.

In familial hypercholesterolemia (FH), low HDL cholesterol (HDL-C) levels are associated with functional alterations of HDL particles that reduce their capacity to mediate the reverse cholesterol transport (RCT) pathway. The objective of this study was to evaluate the consequences of LDL apheresis on the efficacy of the RCT pathway in FH patients. LDL apheresis markedly reduced abnormal accelerated cholesteryl ester transfer protein (CETP)-mediated cholesteryl ester (CE) transfer from HDL to LDL, thus reducing their CE content. Equally, we observed a major decrease (-53%; P < 0.0001) in pre-β1-HDL levels. The capacity of whole plasma to mediate free cholesterol efflux from human macrophages was reduced (-15%; P < 0.02) following LDL apheresis. Such reduction resulted from a marked decrease in the ABCA1-dependent efflux (-71%; P < 0.0001) in the scavenger receptor class B type I-dependent efflux (-21%; P < 0.0001) and in the ABCG1-dependent pathway (-15%; P < 0.04). However, HDL particles isolated from FH patients before and after LDL apheresis displayed a similar capacity to mediate cellular free cholesterol efflux or to deliver CE to hepatic cells. We demonstrate that rapid removal of circulating lipoprotein particles by LDL apheresis transitorily reduces RCT. However, LDL apheresis is without impact on the intrinsic ability of HDL particles to promote either cellular free cholesterol efflux from macrophages or to deliver CE to hepatic cells.

Genetic mutations in genes encoding the LDLR (LDL receptor), APOB (apolipoprotein B-100), or PCSK9 (proprotein convertase subtilysin kexin 9) account for at least 80% of autosomal dominant hypercholesterolemia ( 1 ). Familial hypercholesterolemia (FH) is characterized by a selective increase of LDL particles in plasma, resulting in cholesterol deposition in the arteries, tendons, and skin xanthomas, arcus cornea, thereby increasing the risk of premature coronary heart disease ( 2 ).
In addition to elevated plasma LDL cholesterol (LDL-C) levels, the low HDL cholesterol (HDL-C) phenotype frequently observed in FH patients may also contribute to premature atherosclerosis ( 3,4 ). Indeed, epidemiological studies have demonstrated that an increase of 1 mg/dl in plasma HDL-C concentration yields a 2% to 3% reduction in cardiovascular risk ( 5 ). HDL particles possess multiple antiatherogenic functions, in particular those related to the reverse cholesterol transport (RCT) pathway, the physiological process by which excess cholesterol is removed from peripheral tissues and transported back to the liver for biliary excretion. Indeed, this pathway represents the primary mechanism by which HDL protects against atherosclerosis and by which it may induce plaque regression ( 6 ). In this context, we have Abstract In familial hypercholesterolemia (FH), low HDL cholesterol (HDL-C) levels are associated with functional alterations of HDL particles that reduce their capacity to mediate the reverse cholesterol transport (RCT) pathway. The objective of this study was to evaluate the consequences of LDL apheresis on the effi cacy of the RCT pathway in FH patients. LDL apheresis markedly reduced abnormal accelerated cholesteryl ester transfer protein (CETP) -mediated cholesteryl ester (CE) transfer from HDL to LDL, thus reducing their CE content. Equally, we observed a major decrease ( ؊ 53%; P < 0.0001) in pre-␤ 1-HDL levels. The capacity of whole plasma to mediate free cholesterol effl ux from human macrophages was reduced ( ؊ 15%; P < 0.02) following LDL apheresis. Such reduction resulted from a marked decrease in the ABCA1-dependent effl ux ( ؊ 71%; P < 0.0001) in the scavenger receptor class B type I-dependent effl ux ( ؊ 21%; P < 0.0001) and in the ABCG1-dependent pathway ( ؊ 15%; P < 0.04). However, HDL particles isolated from FH patients before and after LDL apheresis displayed a similar capacity to mediate cellular free cholesterol effl ux or to deliver CE to hepatic cells. We demonstrate that rapid removal of circulating lipoprotein particles by LDL apheresis transitorily reduces RCT. However, LDL apheresis is without impact on the intrinsic ability of HDL particles to promote either cellular free cholesterol effl ux from macrophages or to deliver CE to hepatic cells. Lipoprotein fractionation and pre-␤ 1 HDL quantifi cation Plasma lipoproteins were isolated from plasma by density gradient ultracentrifugation in a Beckman SW41 Ti rotor at 40,000 rpm for 48 h in a Beckman XL70 at 15°C as previously described ( 15 ). After centrifugation, gradients were collected from the top of the tubes with an Eppendorf precision pipette into 30 fractions of 0.4 ml in order to obtain VLDL (d<1.006 g/ml; fraction 1), IDL (d = 1.006-1.019 g/ml; fraction 2), LDL (d = 1.019-1.063 g/ml; fractions 3 to 12), HDL2 (d = 1.063-1.110 g/ml; fractions 13 to 17), and HDL3 (d = 1.110-1.179 g/ml; fractions 18 to 23). Plasma levels of pre-␤ 1-HDL were determined by using the pre-␤ 1-HDL Elisa kit from Sekisui Medical Co. (Tokyo, Japan) as previously described by Miyazaki et al. ( 16 ).

Lipid and protein analyses
The lipid content of plasma and isolated lipoprotein fractions, total protein and apoA-I, and apoB were quantifi ed with an Autoanalyzer (Konelab 20). Total cholesterol and triglyceride (TG) levels were measured by using reagent kits from Roche diagnostics and ThermoElectron, respectively. Quantifi cation of free cholesterol and phospholipids was performed with reagent kits (Wako Diagnostics). CE mass was calculated as (TC-FC) × 1.67 and as previously described ( 17 ). BCA reagent from Pierce was utilized for total protein quantifi cation. Fasting plasma LDL-C was calculated using the Friedewald formula. HDL-C levels were determined after dextran sulfate-magnesium precipitation of apoBcontaining lipoproteins. Plasma apolipoprotein concentrations (apoA-I and apoB) were determined using ThermoElectron reagents and calibrators.

Determination of endogenous CETP-mediated CE transfer from HDL to apoB-containing lipoprotein activity and plasma CETP concentration
Determination of endogenous CE transfer from HDL to apoBcontaining lipoproteins was assayed by the method of Guerin, Dolphin, and Chapman ( 18 ), which estimates net physiological CE transfer between lipoprotein donor and acceptor particles in the plasma of individual patients. Radiolabeled HDLs were obtained from the d>1.063 g/ml plasma fraction by ultracentrifugation at 100,000 rpm for 3 h 30 min at 15°C with a Beckman TL100 centrifuge. The d>1.063 g/ml fraction was then labeled with [ 3 H] cholesterol overnight (4 µCi/ml). Radiolabeled [ 3 H]HDLs were isolated from the d>1.063 g/ml plasma fraction by centrifugation at 100,000 rpm for 5 h 30 min at 15°C after adjustment of the density at 1.21 g/ml by addition of dry solid KBr. CE transfer was determined after incubation of whole plasma (500 µl) from individual subjects at 37°C or 0°C for 3 h in the presence of radiolabeled HDL (25 µg HDL-CE) and iodoacetate (fi nal concentration 1.5 mmol/l) for inhibition of LCAT. After incubation, plasma lipoproteins were fractionated by isopycnic density gradient ultracentrifugation. The radioactive content of lipoprotein subfractions was quantifi ed by liquid scintillation spectrometry with a Trilux 1450 (Perkin Elmer). The CETP-dependent CE transfer was calculated from the difference between the radioactivity transferred at 37°C and 0°C. The rate of CE transfer was calculated from the known specifi c radioactivity of radiolabeled HDL-CE and expressed as micrograms CE transferred.h CETP mass was determined by using a two-antibody sandwich immunoassay as previously described ( 19 ). Briefl y, this assay involves use of 96-well plates coated with a combination of TP1 and TP2 monoclonal antibodies. Plates were blocked with 1% BSA to prevent nonspecifi c binding. Antidigoxigenin Fab fragments of TP20 coupled to peroxidase were used as the second antibody; 3,3',5,5'-tetramethylbenzidine and H 2 O 2 were added, and the absorbance of the mixture read at 450 nm. recently demonstrated that the atheroprotective RCT pathway is defective in FH patients as a result of functional anomalies of HDL particles that alter their capacity to mediate key steps of RCT. Such anomalies contribute signifi cantly to accelerating atherosclerosis in FH patients ( 7 ).
The treatment of FH includes a low-lipid diet with PUFAs and more vegetables and fruits, a lipid-lowering therapy with a statin, and, if necessary, an intestinal cholesterol absorption inhibitor such as ezetimibe ( 8,9 ). Such pharmacological therapies are in some cases not suffi cient to reduce LDL-C to the therapeutic goal ( 10 ). Thus, the latter FH patients need to undergo LDL apheresis sessions every 2 or 3 weeks. Indeed, LDL apheresis represents a method of choice, in addition to drug therapy, to reduce cardiovascular risk in severe FH patients by dramatically reducing atherogenic LDL particle levels ( 11 ). Various LDL apheresis techniques are currently used: immunoadsorption, de xtran sulfate-cellulose adsorption, heparin extracorporal LDL precipitation system, and the direct adsorption of lipoprotein (DALI ® ) system (Fresenius Medical Care; Germany) using hemoperfusion ( 12 ). These different techniques facilitate reduction in all classes of apoB-containing lipoprotein particles, including VLDL, LDL, and lipoprotein [a] (Lp[a]) (up to 70%) and signifi cantly contribute to enhancing the impact of diet and drug therapy on the progression of coronary artery disease ( 11 ). However, the LDL apheresis procedure also results in a signifi cant decrease in plasma HDL-C levels (9%-25%), depending on the LDL apheresis system ( 13 ). This effect refl ects preferential reduction in plasma levels of large HDL2 subfractions ( Ϫ 30%) and, to a lesser extent, in the smaller HDL3 particles ( Ϫ 20%) ( 14 ). In the present study, we evaluated the consequences of LDL apheresis on the effi cacy of the RCT pathway in FH patients by assessing the ability of HDL particles to mediate free cholesterol effl ux from macrophages, CETPmediated cholesteryl ester (CE) transfer from HDL to apoBcontaining lipoproteins, and hepatic HDL-CE delivery.

Patients
Eighteen patients displaying FH (10 men and 8 women) regularly undergoing LDL apheresis at intervals of 2 to 3 weeks were selected for this study. Diagnosis of FH was validated by DNA analysis for 16 patients: all the patients displayed a molecular defect in the LDLR gene. All patients were treated once daily with lipid-lowering drugs in combination therapy: atorvatstatin/ezetimibe (80 mg/10 mg; n = 16), simvastatin/ezetimibe (8 mg/10 mg; n = 1), or rosuvastatin/ezetimibe (20 mg/10 mg; n = 1). The treatment had been unchanged for 3 months before the sampling. Major clinical and biological characteristics of patients before and after LDL apheresis are presented in Table 1 .
Blood samples were obtained by venipuncture before and immediately after the LDL apheresis procedure using the DALI ® system and were collected into sterile EDTA-containing tubes for isolation of plasma. Plasma was immediately separated from blood cells by low-speed centrifugation at 2,500 rpm for 20 min at 4°C and frozen at 80°C until used. The study was approved by the Human Subjects Review Committee of the hospital, and written informed consent was obtained from all subjects.
Fu5AH cells were maintained in Eagle's MEM containing 5% newborn calf serum. Cells were seeded into 24 multiwell plates at a density of 40,000 cells/ml. Two days after plating, cells were labeled by incubation with [ 3 H]cholesterol (1 Ci/ml) in culture medium for 48 h. Subsequently, Fu5AH cells were incubated for an additional period of 24 h in serum-free medium supplemented with BSA (0.5%). Then an effl ux experiment was performed by incubating cells for 4 h at 37°C in the presence of serum-free medium containing cholesterol acceptors (40-fold-diluted plasma or isolated HDL particles at 10 µg phospholipid/ml, equivalent to approximately 15 µg protein/ml).
Mouse macrophage Raw264.7 cells were maintained in DMEM supplemented with 10% FBS. Cells were seeded into 24-multiwell plates at a density of 225,000 cells/ml. The day after cell plating, cells were loaded and labeled with acetylated LDL (50 g/ml) and 0.5 Ci/ml [

Determination of LCAT-mediated cholesterol esterifi cation rate and plasma PLTP activity
LCAT-mediated cholesterol esterifi cation rate was measured using a nonradioactive method as previously described ( 20 ). This methodology evaluates the decrease in plasma free cholesterol content with time of incubation. Briefl y, aliquots of plasma (40 µl) were incubated at 37°C for 2 h in the presence or absence of iodoacetamide (150 mmol/l). Following incubation, free cholesterol content was measured as described above. The cholesterol esterifi cation rate was determined from the decrease in plasma free cholesterol content following incubation and is expressed as nanomoles of CE formed.h Ϫ 1 .ml Ϫ 1 .
A phospholipid transfer protein (PLTP) activity assay kit from Roar Biomedical (New York, NY) was used for determination of plasma PLTP activity according to the manufacturer's instructions.
The effects of LDL apheresis on the mean weight chemical composition of native lipoprotein subspecies are presented in Table 2 . Analysis of TG-rich lipoprotein particles failed to reveal an effect of LDL apheresis on the weight percent chemical composition of the VLDL and IDL fractions. By contrast, LDL particles displayed an enrichment in TG content (+33%; P < 0.0004) as well as a depletion in CE content ( Ϫ 17%; P < 0.002). In addition, we observed a decrease in both HDL2-CE ( Ϫ 19%; P < 0.0001) and in HDL3-CE ( Ϫ 12%; P < 0.02) content in type IIa patients after LDL apheresis as compared with patients before LDL apheresis. In addition, both HDL2 and HDL3 subfractions from FH patients were significantly enriched in TG (HDL2, +17%; P < 0.02; HDL3, +36%; P < 0.002) following LDL apheresis. Such concomitant increase in the relative proportion of TG associated with reduction in that of CE in both HDL2 and HDL3 resulted in signifi cant reductions in the CE/TG ratio in these particles ( Ϫ 31% and Ϫ 35% in HDL2 and HDL3, respectively; P < 0.004) after LDL apheresis. Interestingly, HDL subfractions were signifi cantly enriched in PL content following LDL apheresis (HDL2, +5%; P < 0.05 and HDL3, +10%; P < 0.05). calculated as the difference between fractional cholesterol effl ux to cells in the presence or absence of 8Br-cAMP.
CHO-K1 cells (wild-type and hABCG1-transfected cells) were maintained in Ham's F-12 medium containing 10% FBS. Cells were seeded into 24-multiwell plates at a density of 60,000 cells/ml. Two days after plating, cellular cholesterol was labeled by incubation of cells with culture medium and 1 Ci/ml [ 3 H]cholesterol for 24 h. Equilibration of the label was performed for 90 min in serum-free medium-containing BSA (0.1%). Then cholesterol acceptors (40-fold-diluted plasma or isolated HDL particles at 5 µg phospholipid/ml, equivalent to approximately 8 µg protein/ml) were added to the cells in serum-free medium containing BSA (0.1%) for 4 h at 37°C. The ABCG1-dependent effl ux was calculated as the difference between effl ux to hABCG1-transfected CHO-K1 cells and effl ux to wild-type CHO-K1 cells.
THP-1 monocytic cells were maintained in RPMI 1640 medium supplemented with 10% heat-inactivated FBS and were differentiated into macrophage-like cells with 50 ng/ml PMA for 3 days. Human THP-1 macrophages were cholesterol loaded with acetyl LDL (50 µg protein/ml) and labeled with [ 3 H]cholesterol (1 µCi/ml) in serum-free RPMI 1640 supplemented with 50 mM glucose, 2 mM glutamine, and 0.2% BSA (RGGB) for 24 h. The labeling medium was removed, and human macrophages were then equilibrated in RGGB for an additional 24 h period. After equilibration, free cholesterol acceptors (40-fold-diluted plasma or isolated HDL particles at 15 µg phospholipid/ml, equivalent to approximately 25 µg protein/ml) diluted in serum-free medium were added for 4 h at 37°C.
Optimal plasma dilutions or HDL concentrations to use were determined on the basis of dose response curves for the release of free cholesterol from each cellular model ( 22 ). Fractional cholesterol effl ux was calculated as the amount of the label recovered in the medium divided by the total label in each well (radioactivity in the medium + radioactivity in the cells). The radioactivity in the cells was obtained after lipid extraction from cells in a mixture of hexane-isopropanol (3:2; v/v). The background cholesterol effl ux obtained in the absence of any acceptor was subtracted from the effl ux values obtained with the test samples. All effl ux experiments were performed in triplicate for each sample.

In vitro selective hepatic uptake of HDL-CE
In vitro selective HDL-CE liver uptake via scavenger receptor class B type I (SR-BI) was performed by using Fu5AH cell model as previously described ( 25 ). Confl uent cells were incubated in the presence of [ 3 H]CE-labeled HDL (60 µg protein/ml) at 37°C for 5 h. At the end of incubation, the medium was removed and cells were washed four times with PBS and incubated in the presence of an excess of unlabeled HDL (100 µg protein/ml) for 30 min. Cells were then washed four times with PBS and solubilized with 200 µl of NaOH 0.2 N for 15 min at room temperature with gentle mixing. The radioactive content of 100 µl of each cell lysate was measured by liquid scintillation counting, and the protein content in each well was determined. Selective uptake was calculated from the known specifi c radioactivity of radiolabeled HDL-CE and is expressed in micrograms HDL-CE/micrograms cell protein.

Statistical analysis
All data are presented as mean ± SE, or as mean ± SD where indicated. Experimental data were analyzed using the SAS package software (SAS/STAT User's Guide, v 8 by SAS Institute, Inc.; Cary, NC). Differences between patients before and after LDL apheresis were determined by comparing these parameters at baseline with those after LDL apheresis by ANOVA using Student's paired t -test. The results were considered to be statistically signifi cant at P < 0.05. , before and after LDL apheresis, respectively). Plasma PLTP activity was signifi cantly increased (+41%; P < 0.004) in FH patients following LDL apheresis, as compared with baseline (23.8 ± 2.7 and 33.6 ± 2.7 pmol transferred/min, before and after LDL apheresis, respectively).

Determination of cellular free cholesterol effl ux
As shown in Fig. 2 , we observed a signifi cantly reduced capacity of whole plasma to mediate the cellular free cholesterol effl ux from cholesterol-loaded human THP-1 macro phages ( Ϫ 15%; P < 0.02) following LDL apheresis ( Fig. 2A ). Using our different cellular models, each representative of one specifi c effl ux pathway, we observed that following LDL apheresis, the capacity of whole plasma to remove cholesterol from cells was signifi cantly reduced through ABCA1, SR-BI, and ABCG1. Indeed, evaluation of plasma effl ux capacities using Fu5AH cells revealed that LDL apheresis signifi cantly reduced the SR-BI-dependent effl ux by 21% ( P < 0.0001; Fig. 2A ) . Also, using macrophage RAW cells, we observed that plasma samples obtained following LDL apheresis displayed a marked reduction in their capacity to remove cholesterol from cells via ABCA1 ( Ϫ 71%; P < 0.0001; Fig. 2C ). In addition, LDL apheresis induced a reduction in the capacity of whole plasma to mediate cellular cholesterol effl ux via ABCG1 ( Ϫ 15%; P < 0.04; Fig. 2B ). Interestingly, reduction in plasma apoA-I concentrations following LDL apheresis was signifi cantly correlated with the capacity of

Impact of LDL apheresis on LCAT-mediated cholesterol esterifi cation rate and plasma PLTP activity
LDL apheresis was associated with a major decrease ( Ϫ 42%; P < 0.005) in LCAT-mediated cholesterol esterifi cation rate  accelerated CETP-mediated neutral lipid redistribution between plasma lipoprotein particles. Reduction in both plasma CETP concentration and activity leads to reduction in CE content in the remaining circulating LDL particles following LDL apheresis. This procedure also contributed to diminished plasma levels of both mature HDL and small pre-␤ 1-HDL particles, and in this way, transitorily reduced the removal of cholesterol from peripheral cells and its transport back to the liver. However, this quantitative impact of LDL apheresis on HDL was not associated with modifi cation in the ability of HDL particles to promote cellular free effl ux cholesterol from macrophages or deliver CE to hepatic cells. We observed in this study that FH patients displayed a plasma CETP concentration within the normal range. Indeed it has been previously demonstrated that circulating CETP levels typically vary from 1 to 3 g/ml in normolipidemic subjects ( 26 ). However, our current observations appear to be in contrast with those previously reported showing elevated plasma CETP concentration in various forms of hyperlipidemia, including hypercholesterolemia alone or in combination with hypertriglyceridemia ( 27 ). It is relevant to consider that FH patients recruited here were treated once daily with lipid-lowering drugs in combination therapy involving a statin and ezetimibe, which probably contributed to normalize plasma CETP levels.
Indeed, it has been demonstrated that atorvastatin, which was used in 16 of the 18 patients of the study, reduces whole plasma to mediate cellular free cholesterol effl ux via ABCA1 ( r = 0.492; P = 0.0023), SR-BI ( r = 0.406; P = 0.0140), and ABCG1 ( r = 0.509; P = 0.002). However, we observed that individual HDL subfractions, HDL2 or HDL3, isolated from FH patients before and after LDL apheresis, displayed an equivalent capacity to mediate cellular free cholesterol effl ux ( Fig. 3 ).
Taken together, these observations clearly indicate that the reduction in plasma effl ux capacity following LDL apheresis primarily resulted from a reduction in circulating cellular cholesterol acceptor number rather than modifi cation in the intrinsic capacity of HDL particles for cholesterol effl ux.

In vitro determination of selective HDL-CE uptake
The capacity of HDL particles isolated from type IIa patients before and after LDL apheresis to deliver CEs to hepatic cells was evaluated in vitro using Fu5AH cells. We did not observe any signifi cant impact of LDL apheresis on the capacity of HDL particles from FH patients to deliver CEs to hepatic cells (12.4 ± 4.4 and 12.3 ± 4.3 µg HDL-CE/µg cell protein before and after LDL apheresis, respectively).

DISCUSSION
In the present study, we demonstrate for the fi rst time that rapid removal of circulating LDL particles by LDL apheresis in FH patients dramatically reduces the abnormal  properties. The concomitant action of CETP, PLTP, and HL favors the interconversion of large HDL into smaller HDL particles, allowing apoA-I recycling ( 37 ). The latter becomes rapidly relipidated by cellular phospholipid and cholesterol to form pre-␤ 1-HDL particles that constitute the initial acceptors of cellular cholesterol through the ABCA1 pathway ( 34,38 ). Following LDL apheresis, we observed a reduction in plasma LCAT activity that is associated with reduction in CE content in HDL particles. Consistent with that observation, Franceschini et al. ( 14 ) previously reported a 26% reduction in plasma LCAT activity following LDL apheresis. It is likely that following LDL apheresis, LCAT reaction is limited as a result of the marked reduction in plasma pre-␤ 1-HDL concentration. Indeed, LCAT progressively transforms large pre-␤ -migrating HDLs that are complexed with several phospholipid molecules and contain signifi cant amounts of unesterifi ed cholesterol into spherical HDL particles. Through its continuous action, LCAT is responsible for the maturation of the HDL pool as a result of hydrophobic CE retention in the HDL core and thus contributes to the formation of larger HDL particles. Thus, the reduction in plasma LCAT activity following LDL apheresis contributes to the observed reduction in plasma levels of mature HDL particles. Following LDL apheresis, a signifi cant reduction in CETP-mediated CE transfer from HDL to LDL lowers CE content in LDL particles and reduces the formation of pre-␤ 1-HDL particles. Indeed, CETP mediates the transfer of CEs produced by LCAT to other lipoproteins, mainly TG-rich lipoproteins, with simultaneous reciprocal transfer of TGs. Then, TG-enriched HDL particles formed are catabolised by the extracellular hepatic TG lipase with concomitant release of lipid-poor apoA-I ( 37 ). The marked diminution in pre-␤ 1-HDL levels observed following LDL apheresis is therefore entirely consistent with a reduced release of apoA-I from HDL as a result of the reduction in CETP activity. However, we cannot rule out the possibility that some of circulating pre-␤ -HDL particles might have been physically removed by the LDL apheresis procedure itself by a mechanism which remains to be identifi ed.
plasma CETP concentration in a dose-dependent manner ( 28,29 ). Equally, it has been demonstrated that combination simvastatin/ezetimibe therapy reduces plasma CETP levels by 21% in patients with metabolic syndrome ( 30 ).
Interestingly, plasma CETP levels were signifi cantly reduced following LDL apheresis. In human plasma, CETP is mostly bound to lipoproteins, with only 1% of plasma CETP in a free form ( 31,32 ). Approximately 74% of CETP is associated with the HDL fraction, 24% with LDL, and only 1% with VLDL. Thus, the reduction in plasma CETP levels observed following LDL apheresis mainly resulted from the rapid removal of circulating LDL particles, which accounted for at least 70% of the total CETP mass lost. CETP-mediated transfers of CE and TG between different lipoprotein particles are dependent on the relative concentrations of the donor and acceptor particles, and on their core lipid content ( 33 ). Despite normal, even reduced, plasma CETP concentration observed following LDL apheresis, total CE transferred from HDL to apoB-containing lipoprotein was markedly lower, indicating that very low levels of apoB-containing lipoprotein particles represent the limiting factor in the CETP-mediated CE transfer reaction.
It is well established that ABCA1 mediates cellular free cholesterol effl ux to lipid-poor apoA-I similar to pre-␤ -HDL particles ( 34 ), whereas SR-BI and ABCG1 preferentially promote the effl ux of cellular cholesterol toward more mature HDL particles ( 35,36 ). Therefore, the reduced ABCA1dependent plasma effl ux capacity following LDL apheresis was entirely consistent with the concomitantly observed reduction in pre-␤ 1-HDL levels. In addition, we demonstrated that LDL apheresis did not signifi cantly modify the ability of isolated HDL particles, HDL2 or HDL3, to mediate cellular free cholesterol effl ux from macrophages through both SR-BI and ABCG1. We conclude that the observed reduction in plasma effl ux capacity for SR-BI-and ABCG1-dependent effl ux primarily resulted from the reduction in large HDL particle number observed following LDL apheresis.
Within human plasma, HDL particles are continuously remodelled by several lipid transfer proteins or enzymes, leading to the generation of multiple HDL subclasses characterized by distinct physico-chemical and functional this way, transitorily impacts both the quantitative and qualitative features of HDL particles, but is, however, without any signifi cant effect on their defective functional properties.
Finally, we observed a signifi cant increase in plasma PLTP activity following LDL apheresis that equally contributes to reduced plasma levels of large HDL particles. In plasma, PLTP mediates the fusion of two large HDL particles to form an unstable intermediate product that can rearrange into three small particles or can be converted into a larger particle with dissociation of two molecules of apoA-I ( 39 ). It is important to note that LDL apheresis-induced reduction in both HDL-C levels and LCAT activity occurs in a transitory manner. Indeed, 24 h after LDL apheresis, basal levels of HDL-C and of LCAT activity were restored, whereas those of LDL-C were increased by up to 10% and thus still remained markedly reduced as compared with baseline ( 14 ). Our present observations, in addition to those of the literature, allow us to suggest that the increase in plasma PLTP activity observed following LDL apheresis contributes to rapidly restoring the plasma pool of pre-␤ -HDL and subsequently, through the action of LCAT, those of mature HDL particles.
Our present observations allow us to propose an integrated mechanism of the impact of LDL apheresis on the RCT pathway in patients displaying FH ( Fig. 4 ). Key steps of the RCT, including cellular free cholesterol effl ux, LCAT-mediated cholesterol esterifi cation rate, CETP-mediated CE transfer between plasma lipoproteins, and hepatic CE delivery, are signifi cantly reduced following the LDL apheresis procedure as a result of the rapid removal of circulating lipoprotein particles. Transient reduction in the effi cacy of the RCT following LDL apheresis primarily results from a reduction in the number of HDL particles, with no signifi cant alteration of their intrinsic capacity to mediate either cellular free cholesterol or hepatic HDL-CE delivery. In severe FH patients, defective biological activities of HDL particles are associated with an altered effi cacy of the RCT pathway that contributes signifi cantly to accelerated atherosclerosis progression ( 7 ). Our present observations demonstrate that LDL apheresis, used to selectively remove apoB-containing lipoprotein particles and to reduce cardiovascular risk in severe FH patients, markedly reduces the abnormal accelerated CETP-mediated neutral lipid transfer and contributes to reducing the circulating CE pool within the remaining circulating LDL particles. In addition, we demonstrate that LDL apheresis is also associated with modifi cation in the action of three key proteins involved in intravascular HDL remodeling, and, in Dashed arrows indicate steps of the RCT pathway that are known from the literature to be reduced but that have not been directly assessed in the present study.