- Email: [email protected]
Compositional and functional changes of low-density lipoprotein during hemodialysis in patients with ESRD1
Kidney International, Vol. 54 (1998), pp. 608 – 617
Compositional and functional changes of low-density lipoprotein during hemodialysis in patients with ESRD1 ANDREAS AMBROSCH, UTE DOMROESE, SABINE WESTPHAL, JUTTA DIERKES, WOLFGANG AUGUSTIN, KLAUS H. NEUMANN, and CLAUS LULEY Institute of Clinical Chemistry, Department of Pathobiochemistry, and Department of Medicine, Division of Nephrology, University Hospital of Magdeburg, Magdeburg, Germany
Compositional and functional changes of low-density lipoprotein during hemodialysis in patients with ESRD. Background. This study focused on the effects of hemodialysis on the atherogenic properties of low density lipoprotein (LDL) in patients with end-stage renal disease (ESRD). The impact of cholesterol ester transfer protein (CETP) activity and lipolysis on LDL composition, particularly the changes during hemodialysis, was investigated. Methods. Blood was drawn from 15 normotriglyceridemic (NTG) and 15 hypertriglyceridemic patients [HTG; triglycerides (TG) ,2.2 mmol/liter] before hemodialysis, during (1.5 hr after the beginning of anticoagulation) and at the end of treatment. In each sample, lipid values and CETP activity were measured. LDL was prepared and characterized by its components and diameters (2 to 16% PAGGE). To investigate the functional properties of LDL, fractions obtained from NTG and HTG patients were incubated with human skin fibroblasts and a cell line of murine macrophages (P388), and cholesterol ester formation rates were measured. Results. In comparison to LDL from NTG patients at baseline, HTG-LDL were enriched in triglycerides (P , 0.02), depleted in cholesterol proportion (P , 0.01) and small in size (P , 0.001). These LDL induced the cholesterol esterification rates (50 mg/mL LDL-protein) in a twofold greater unsaturation in macrophages when compared to LDL from NTG patients (P , 0.04). The rates in fibroblasts were reduced by approximately half (P , 0.05). During hemodialysis, LDL were decreased in size (P , 0.001) and depleted in TG (P , 0.01), particularly in the hypertriglyceridemic state. Although CETP activity increased during hemodialysis (P , 0.001), the cholesterol content remained unchanged. When HTG-LDL obtained during hemodialysis were incubated with cells, esterification rates particularly in macrophages were markedly accelerated in comparison to the unmodified lipoprotein at baseline (P , 0.05). Conclusion. LDL from HTG patients with ESRD was TGenriched, CH-depleted and smaller in size. As the intracellular 1
This work is dedicated to Professor Dr. Dietrich Seidel, Klinikum Grobhadern, University of Munich, on the occasion of his 60th birthday. Key words: end-stage renal disease, hypertriglyceridemia, low density lipoprotein, CETP activity, esterification rate, atherosclerosis. Received for publication June 30, 1997 and in revised form March 18, 1998 Accepted for publication March 18, 1998
© 1998 by the International Society of Nephrology
esterification rates induced by LDL were related to the cellular uptake, these LDL were a superior substrate for murine macrophages with the tendency of intracellular accumulation, and an inferior substrate for fibroblasts suggesting a decreased uptake by the specific receptor pathway. TG-depletion of LDL during hemodialysis, particularly in HTG patients due to a lipasemediated TG-hydrolysis, increased these effects in macrophages. We suggest that the alterations of LDL that occur during repeated hemodialysis in vivo could contribute to the high prevalence of premature atherosclerosis found in HTG patients with ESRD.
End-stage renal disease (ESRD) is associated with accelerated atherosclerosis and a high incidence of cardiovascular disease [1, 2]. In the United States, the death rate from myocardial infarction for patients with ESRD aged between 45 and 64 years is 24.1 per 1000 patient-years, which is far in excess of the general population [3, 4]. Although the course of increased atherosclerotic cardiovascular disease in these patients is probably multifactorial, lipid abnormalities are thought to play a major role [5] and a number of alterations in lipids have been described. The primary lipid abnormality found in ESRD is elevated triglyceride (TG) levels that are predominantly due to reduced lipolysis of triglyceride-rich lipoproteins [6 –9], which is often accompanied by a reduced high-density lipoprotein (HDL) cholesterol level [10, 11]. These abnormalities are present in patients with ESRD who are not on dialysis as well as patients treated either with hemodialysis or continuous ambulatory peritoneal dialysis [12, 13]. Elevated TG levels can indicate accumulation of triglyceriderich, potentially atherogenic remnant particles, a finding reported in patients treated with chronic hemodialysis [14 –16]. Low-density lipoprotein (LDL) cholesterol levels are not elevated in ESRD [17, 18], but changes in size and composition in association with hypertriglyceridemia could potentially result in increased atherogenicity of LDL particles [19]. Recently, Luley et al described a small-dense LDLpattern for dyslipidemic patients with ESRD and, in addition, an acute alteration in LDL size during the course of
608
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
hemodialysis [20]. These changes in LDL diameter were pronounced in dyslipidemic patients and accompanied by an increase in total LDL-cholesterol during each hemodialysis session. Consequently, it can be theorized that repeated hemodialysis in itself has an unfavorable effect on the atherogenic profile of LDL. In the present study, we determined the impact of cholesteryl ester transfer protein (CETP) activity and lipolytic activity on changes in LDL composition and size, before and during hemodialysis in normo- and dyslipidemic patients with ESRD. Furthermore, we investigated the functional properties of this lipoprotein species in vitro in relation to the compositional changes caused by hemodialysis. METHODS Study subjects A total of 30 patients with ESRD (17 male and 13 female; from the Department of Internal Medicine, Center for Kidney Disease, University Hospital Magdeburg, Germany) were enrolled in the study. The patients group consisted of 15 hypertriglyceridemic subjects (total triglycerides . 2.2 mmol/liter) and 15 subjects with normal plasma triglycerides. All patients were hemodialyzed against a bicarbonate bath for three to four hours in a treatment three times weekly. Underlying renal disease were: diabetic nephropathy (N 5 7), chronic glomerulonephritis (N 5 7), small kidneys of unknown cause (N 5 9), polycystic kidney disease (N 5 3), pyelonephritis (N 5 3), and Alport syndrome (N 5 1). A total of nine patients were diabetic: three in the NTG group and six in the HTG group (7 patients were on hemodialysis because of diabetic nephropathy, 1 because of polycystic kidney disease and 1 because of pyelonephritis). Two patients in the HTG group and one in the NTG group had a history of renal transplantation with a consecutive transplant dysfunction. In both groups, the prevalence of a history for coronary artery disease (CAD), peripheral vascular disease and for stroke was evaluated: history of CAD in HTG versus NTG patients 6 of 15 versus 4 of 15; for peripheral vascular disease 7 of 15 versus 3 of 15; for stroke 4 of 15 versus 0 of 15. Patients with a history of recent myocardial infarction (within 3 months), unstable angina, severe congestive heart failure, recent stroke and hypertension with systolic RR . 200 mm Hg (and/or diastolic RR . 105) were excluded. In addition, no patient had acute bacterial or viral infection except one patients who had a history of hepatitis C. Since most of our patients suffered from biochemical hyperparathyroidism (serum parathyroid hormone . 65 ng/liter), only patients with a symptomatic manifestation (hyperparathyroid bone disease, electrocardiographic signs of hypercalcemia) were excluded. Exclusion criteria did not allow to study patients with cancer (or a history of cancer). Informed consent was obtained from all patients.
609
Study design Venous blood was drawn after a 12-hour overnight fast to obtain baseline values. Then, a lipid load test was performed (a) because patients normally are advised not to fast before a hemodialysis procedure; (b) to standardize the postprandial lipoprotein metabolism of the patients [21]. Fifteen minutes after the lipid load, the hemodialysis procedure began. Furthermore, two additional blood samples were obtained, one 1.5 hours after the beginning of hemodialysis (peri-dialytically) and one at the end of each hemodialysis session (post-dialytically). The anticoagulant that was employed contained unfractionated heparin during each hemodialysis. All patients were within 2% of their fluid free weight, which was used to calculate the individual heparin doses. The anticoagulant was administered with a single intravenous bolus (60 IU/kg body wt) at the beginning of each hemodialysis session and with a subsequent infusion of 1000 IU/hr. Lipoproteins EDTA-plasma (EDTA 1 mg/liter) for lipoprotein analysis was separated without delay by centrifugation at room temperature and immediately processed or stored at 220°C. For baseline lipoprotein and apoprotein concentrations, very low density lipoprotein (VLDL) and chylomicrons were separated from LDL and HDL by a single centrifugation step over a time period of 18 hours at 100,000 g in a TI 50.4 rotor with a Beckman L-50 ultracentrifuge (all from Beckman, Munich, Germany). The layer of VLDL on the top of the tubes was removed with a Pasteur-pipette. In the infranatent, the LDL fraction was precipitated with phosphotungsten acid/MgCl2 reagent before measurement of HDL composition. The composition of the VLDL-, HDL-fractions and the infranatent were determined by measurement of CH, TG, FC and apoproteins on an automated Hitachi 911 analyzer (Boehringer, Mannheim, Germany). LDL isolation. After separation of chylomicrons by centrifugation at 100,000 g [density (d) , 1.006 g/ml] for 30 minutes, the infranatent was increased to d 5 1.019 mg/liter by the addition of solid potassium bromide (Sigma, Munich, Germany) and then centrifuged at 150,000 g (24 hr at 10°C). The floating intermediate density lipoproteins (IDL) and VLDL were collected. The infranatent was then brought to d 5 1.060 g/liter and centrifuged at 150,000 g (24 hr at 10°C). The floating LDL was obtained and dialyzed against a Tris-EDTA buffer for 24 hours at 8°C and subsequently characterized by their CH, FC, CE, TG and apo B content. Lipoprotein component assays Total apo AI, B, CIII and E were analyzed by immunturbidimetric assays using goat antihuman polyclonal antisera (Greiner, Munich, Germany). For measurement of
610
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
cholesterol, we used the CHOD-PAP method (Chol kit®; Boehringer), for triglycerides and free glycerol the GPOPAP method (Triglyceride GPO-PAP kit®; Boehringer) was used, for free cholesterol the COD-PAP method (free cholesterol C kit®; WAKO, Munich, Germany) and for measurement of free fatty acids the ACS-ACOD method (NEFA C®; WAKO). The CE content was calculated from total CH and total FC concentrations. The hematocrit was measured in each sample and used for corrections of lipoprotein levels. Nondenaturating polyacrylamide gradient gel electrophoresis To estimate LDL particle size, a nondenaturating polyacrylamide 2 to 16% gradient gel electrophoresis (PAGGE) was performed, modified from [22]. Briefly, the stock solutions in the TE-buffer (Tris 210 mM, sodium acid 3 mM, disodium EDTA 3 mM, pH 8.35; all from Sigma) included in following: the high limit solution acrylamide 152.3 g/liter, bis-acrylamide 7.7 g/liter (16% total, 4% crosslinker) and glycerol 260 g/liter (all from Sigma), and the low limit solution acrylamid, 16 g/liter and bis-acrylamide 4 g/liter (2% total, 4% crosslinker). The high and the low limit solutions were made immediately before the gradient was cast and contained 1 volume high or low limit solution, 0.0015 vol freshly prepared ammonium persulfate (100 g/liter; Sigma), and 0.00025 vol 3-dimethylaminopropionitril (TEMED; Sigma). The gradient was generated with a BioRad Econo-pump (2.6 ml/min; BioRad, Heidelberg, Germany), mixed in an external mixing chamber (BioRad), and cast in a gel slab casting apparatus (180 3 160 mm). The gels were prerun in TE buffer at 60 V for 30 minutes. Samples were made dense with glycerol (1:3 vol/vol), and a volume containing 20 ml plasma samples was loaded on the gel. Electrophoresis time was two hours at 60 V, 12 hours at 50 V and six hours at 300 V, with a constant voltage power supply (BioRad) and a constant temperature of 8°C. At the end of electrophoresis, gels were stained with Coomassie brilliant blue G (Sigma) for one hour and destained in a solution of 50% methanol, 10% acetic acid and 40% water. Latex beads 34 nm (Duke Scientific, Palo Alto, CA, USA), thyroglobulin 17 nm and ferritin 12.2 nm (Pharmacia, Freiburg, Germany) were run as standards. The apparent diameters of LDL were determined by using a calibration curve generated from standards according to Williams et al [23]. Reproducibility of the gel analysis of LDL size was evaluated by analyzing the same LDL preparation three times. Coefficient of variation of mean LDL diameters between runs was 5%. Cholesteryl ester transfer protein activity Cholesteryl ester transfer protein (CETP) activity was assessed in a defined system by using a fluorescent cholesteryl linoleate analogue (WAK, Bad Homburg, Germany). Exogenous donor and acceptor particles (human
VLDL) were combined with the CETP source (EDTAplasma) within a total volume of a 10 mM Tris buffer. The donor particle consisted of a nitrobenzooxadiol-fluorophor (NBD), which takes advantage of a self-quenching effect on microemulsion incorporated cholesteryl linoleate. The CETP-facilitated transfer of fluorescent cholesteryl linoleate from the self-quenching donor to the acceptor was determined by an increase in fluorescence intensity (incubation time 3 hr, 37°C; excitation wavelength 465 nm (EX 456 nm) and emission wavelength of 535 nm (EM 535 nm) using an LS50 fluorescence spectrometer (Perkin Elmer, Hannover, Germany). A standard curve to derive the relation between fluorescence intensity and mass transfer of cholesteryl linoleate (pmol/3 hr) was generated by dispersing donor particles in isopropanol in a linear dilution (cholesteryl linoleate concentration was 80.9 mg/ml). Spectrally pure (HPLC grade) isopropanol was used as the solvent to avoid background fluorescence of the blank (isopropanol). From each sample, triplicates were measured and the mean fluorescences were calculated. The intra-assay variance of the mean CETP activities was calculated with 10.8% and the inter-assay variance with 11.6%. The fluorimetric CETP assay was compared to a recently described isotopic CETP assay [24], and a good correlation was found (N 5 25, r 5 0.90, P , 0.001) [25]. Cellular [3H]cholesteryl ester formation in cultured human fibroblasts and murine macrophages Cells. Human skin fibroblasts were cultured from biopsies of the skin from normal adult donors. The cells were plated at 7 3 104 cells/35 mm plastic dish in Dulbecco’s minimal essential medium (DMEM, Sigma), containing sodium hydrogen carbonate 25 mM, HEPES 20 mM, streptomycin 100 mg/liter, penicillin 105 IU/liter (Sigma) and fetal calf serum (FCS) 10% (Biochrom, Munich, Germany). Cultures of fibroblasts in passages 4 to 6 were washed with Dulbecco’s phosphate-buffered saline (PBS, Sigma) and incubated in fresh medium containing lipoproteindeficient serum 10% (LPDS) to up-regulate apoB/E receptor activity before experiments. The P388 murine macrophage-like cell line was purchased from American Type Culture Collection (ATCC). P388 cell were plated at 7 3 104 cells/35 mm dish in RPMI 1640 (Sigma) supplemented with FCS 10%, penicillin, streptomycin and glutamine. The cells were fed every three days and were used for experiments within seven days of plating. Metabolism of LDL in cell culture. For cellular experiments, pooled LDL (d 5 1.06) from six NTG and six HTG subjects, respectively, was prepared from fasting plasma (baseline LDL) and from plasma after heparinization (peridialytically; 1.5 hr after the bolus heparinization at the beginning of hemodialysis). LDL cholesterol uptake by cells was estimated by measurement of the stimulation of [3H]-oleate incorporation into cholesteryl ester [26]. After
611
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL Table 1. Anthropometric data and lipoprotein profile of the study groups Normolipidemic group, (NTG), N 5 15
Hyperlipidemic group (HTG), N 5 15
Significance P
Age years BMI kg/m2 Gender male/female Duration of dialysis months (range) Creatinine mmol/liter Urea mmol/liter
49.3 (38 –76) 23.7 (17.8 –29.9) 8/7 34 (15–107) 768 (389 –1373) 19.3 (14.0 –31.8)
51.6 (41–74) 19.4 (18.2–21.5) 9/6 37.5 (25–167) 746 (217–1304) 20.3 (14.9 –28.6)
NS NS NS NS NS
Triglycerides mmol/liter Cholesterol mmol/liter ApoCIII mg/dl ApoE mg/dl ApoAI mg/dl VLDL triglyceride mmol/liter cholesterol mmol/liter LDL cholesterol mmol/liter triglyceride mmol/liter cholesteryl ester mmol/liter unesterified chol. mmol/liter apoB g/liter HDL cholesterol mmol/liter triglyceride mmol/liter cholesteryl ester mmol/liter unesterified cholesterol mmol/liter FFA mmol/liter CETP activity pmol/3 hr
1.54 (0.43–2.04) 5.53 (3.84 – 8.81) 13.3 (9.1–28.7) 5.13 (2.62–9.17) 1.26 (0.89 –2.14)
3.67 (2.20 –7.41) 7.44 (4.32–9.71) 21.7 (13.4 –35.0) 6.80 (3.94 –9.72) 1.14 (0.85–1.57)
, 0.000 , 0.05 , 0.02 , 0.05 NS
1.15 (0.06 –2.18) 1.35 (0.18 –2.34)
2.44 (1.88 –5.98) 2.13 (1.36 – 4.86)
, 0.001 , 0.01
3.37 (2.23– 6.71) 0.39 (0.23– 0.53) 2.42 (1.46 –5.03) 0.96 (0.60 –1.72) 0.85 (0.58 –1.70)
3.16 (1.96 –5.32) 0.64 (0.30 –1.01) 2.29 (1.22–3.95) 0.81 (0.61–1.34) 1.04 (0.65–1.51)
NS , 0.02 NS NS NS
1.24 (0.86 – 4.17) 0.20 (0.13– 0.35) 1.02 (0.69 –3.27) 0.20 (0.03– 0.90) 0.36 (0.12–1.31) 22.0 (9.4 –37.1)
1.04 (0.69 –2.11) 0.25 (0.19 – 0.38) 0.87 (0.58 –1.81) 0.16 (0.08 – 0.35) 0.46 (0.21– 0.77) 24.2 (5.3– 45.6)
NS , 0.03 NS NS NS NS
Values are median (min-max). FFA is free fatty acids.
a 48 hour incubation with medium containing LPDS, cells were washed with PBS and fresh medium containing either LPDS alone or LPDS with LDL. After 24 hours of incubation with lipoproteins, cultures were washed with PBS three times and incubated with [3H]-oleate (0.2 mM, 10 mCi/ml; Amersham, Braunschweig, Germany) in the presence of fatty acid-free albumin (0.07 mM) for two hours at 37°C. The cells were then washed twice with PBS, damaged in an ultrasound bath and incubated with 1 ml of chloroform/methyl alcohol (2:1, vol/vol) to extract lipids. After two more washes with these solvents, the pooled lipid extract was dryed and resolubilized in chloroform/methyl alcohol. The labeled cholesteryl ester was isolated by thin-layer chromatography on silica gel plates using hexane/ diethyl ether/acetic acid (83:16:1). The protein content of cell extracts was measured by a modified method of Lowry et al [27]. All cell experiments were done for three times in triplicates. Statistical analysis The results were presented as medians with minimum and maximum when data were not normally distributed, or as mean 6 SD (intracellular esterification rates). Data were compared using the U-test for interindividual differences and the Wilcoxon test for intraindividual differences. Differences were considered significant at P , 0.05. Coefficients of skewness and kurtosis were calculated to test deviations from a normal distribution. Associations be-
tween lipoprotein parameters were determined by calculation of Pearsons correlation coefficients or by Spearman rank correlation analysis. For all analysis, the SPSS software package (version 6.13) was used. RESULTS Baseline characteristics and lipoprotein profile of the study subjects General characteristics of the subjects, parameters of renal function and results of the biochemical lipoprotein analyses are summarized in Table 1. There were no significant differences between the HTG and NTG groups resulting from sex, age, body mass index (BMI) and parameters of renal function. The lipoprotein profiles of the HTG subjects were characterized by elevated total triglycerides (TG) (P , 0.000), total cholesterol (CH; P , 0.05), VLDL-TG (P , 0.001) and VLDL-CH concentrations (P , 0.01). Within the HTG group, no differences were found between diabetic and non-diabetic patients with regard to the lipoprotein profile (data not shown). ApoCIII and apoE corresponded well to the VLDL-TG and varied significantly between both study groups (P , 0.02 and P , 0.05, respectively). The plasma free fatty acids (FFA) were not significantly different between both groups. Whereas total CH, FC, CE and apo AI in HDL did not differ between both groups (apolipoprotein data not shown), HDL-TG was significantly increased in the HTG group
612
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL Table 2. LDL composition and size in subjects with ESRD
LDL diameter nm CH/apoB ratio CE/apoB ratio FC/apoB ratio TG/apoB ratio
Normotriglyceridemic group (NTG)
Hypertriglyceridemic group (HTG)
Significance P
26.83 (23.91–29.37) 3.77 (2.12– 4.04) 2.97 (2.06 –3.43) 1.08 (0.74 –1.59) 0.40 (0.24 – 0.78)
23.76 (22.31–26.84) 3.16 (1.81–3.89) 2.63 (1.79 –3.22) 0.92 (0.65–1.12) 0.60 (0.25–1.23)
, 0.001 , 0.01 , 0.04 , 0.001 , 0.01
Values are median (min-max). Abbreviations are: LDL, low density lipoprotein; CH, cholesterol; CE, esterfied cholesterol; FC, unesterified cholesterol; TG, triglycerides; apo, apolipoprotein.
(P , 0.03). In addition, LDL from HTG patients were enriched in triglycerides in comparison to LDL from NTG patients (P , 0.05; Table 1) and differed in their triglyceride content per apoB (P , 0.01; Table 2). Furthermore, the apparent diameter of the major plasma LDL subfraction as determined by PAGGE was significantly decreased in the HTG group (P , 0.05; Table 2). According to the classification for LDL-phenotypes by Krauss and Burke [28], 11 of 15 of the HTG patients were found to have a phenotype B-pattern of LDL (,25.5 nm) in contrast to only 4 of 15 NTG participants. The smaller LDL particles were characterized by a decreased CE content (P , 0.04) and FC content (P , 0.001) per apoB (Table 2) and, as mentionend above, by an increased TG proportion (P , 0.01, Table 2). Therefore, LDL diameters correlated inversely with the TG/apoB ratios (r 5 20.42, P , 0.02; Fig. 1A) and positively with the FC content per particle (r 5 0.43; P , 0.03; Fig. 1B). In addition, a significant inverse correlation was found between the CE and TG proportion per apoB (r 5 20.45; P , 0.02; Fig. 1C). Triglyceride depletion of LDL during hemodialysis One and a half hours (1.5 hr) after the bolus heparinization at the beginning of hemodialysis (peri-dialytically), the LDL-TG/apoB ratios decreased significantly in HTG patients [LDL-TG/apoB baseline vs. postheparinic: 0.60 (0.25 to 1.23) versus 0.40 (0.22 to 0.97); P , 0.02], but remained unchanged in the NTG individuals [LDL-TG/apoB baseline vs. postheparinic: 0.40 (0.24 to 0.91) versus 0.39 (0.23 to 0.78); NS]. The changes in the LDL-TG/apoB ratio (DLDL-TG/apoB; calculated as the differences between the baseline LDL-TG/apoB ratios and postheparinic LDLTG/apoB ratios) were positively related (r 5 0.38, P , 0.05; Fig. 2) to the decrease in the LDL diameter (DLDLdiameter; calculated as differences between baseline LDL diameter and postheparinic LDL diameter). Up to the end of the hemodialysis therapy, the LDL-TG/apoB ratios increased in the HTG patients [LDL-TG/apoB at the end of therapy, 0.50 (0.18 to 1.08); versus postheparinic values; P , 0.05], but remained significantly different from baseline (LDL-TG/apoB ratio on baseline vs. end of therapy, P , 0.05). The LDL-CE and -FC, and their ratios per apoB were unaltered in both study groups during the course of hemodialysis (data not shown).
Baseline CETP activities and CETP activation during hemodialysis Baseline CETP activities tended to be increased in the HTG subjects, but this was not statistically significant (Table 1). Shortly after the bolus heparinization at the beginning of the hemodialysis procedure (1.5 hr), CETP values were increased in the respective study groups in comparison to baseline (P , 0.001). Nevertheless, the extent of the peri-dialytical CETP activation was nearly twice as high in the HTG group when compared with the NTG subjects (P , 0.001) (Fig. 3). Furthermore, the scope of the CETP activation (DCETP; calculated as differences between peri-dialytical CETP activities and baseline CETP levels) was related (r 5 0.39, P , 0.04) to the extent of the peri-dialytical FFA increase (DFFA; calculated as differences between postheparinic FFA concentrations and baseline FFA levels). Until the end of hemodialysis, CETP activities decreased in both study groups but remained elevated when compared with baseline levels (P , 0.01, respectively; Fig. 3). Furthermore, a multiple regression analysis was performed to determine a possible effect of CETP activities on the LDL composition and diameter. No correlation was found between CETP and LDL-TG (0.26, NS) or LDL-CE proportion (r 5 20.26, NS), or the LDL diameter (r 5 20.16, NS). Cellular cholesteryl ester formation in fibroblasts and murine macrophages P388 induced by LDL Baseline LDL derived from HTG subjects induced the esterification rate in fibroblasts in a concentration-dependent, saturable manner but to a minor (60%) degree (with 50 mg lipoprotein) compared to baseline LDL levels from NTG individuals (P , 0.05; Fig. 4A). On the other hand, twice as many cholesteryl esters were formed in macrophages after preincubation with baseline LDL (50 mg) from HTG subjects when compared with LDL from NTG individuals (P , 0.04; Fig. 4B). The esterification rates in macrophages were unsaturable up to LDL-protein concentrations of 100 mg (Fig. 4B). When fibroblasts and macrophages were preincubated with LDL preparations derived during hemodialysis from hypertriglyceridemic patients, intracellular [3H]-cholesteryl
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
613
Fig. 1. Inverse correlations between the size of the major low-density lipoprotein (LDL) subfractions (nm) and the triglyceride (TG) content per particle (A; r 5 0.42, P < 0.02) and the relationship between LDL size and unesterified cholesterol (FC) per apoB ratios (B; r 5 0.40, P < 0.04) in normo- and hypertriglyceridemic patients. (C) Inverse correlation between cholesteryl ester content per particle and the triglyceride content in LDL within normo- (NTG; Œ) and hypertriglyceridemic (HTG; M) study participants (r 5 20.37, P , 0.05). 4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™
Fig. 2. Changes in the triglyceride content per apoB during hemodialysis (expressed as DLDL-TG/apoB) are positively related to the changes in the LDL diameter (expressed as DLDL-diameter). Symbols are: (M) hypertriglyceridemic and (Œ) normoglyceridemic subjects. r 5 0.38, P , 0.05.
Fig. 3. CETP activities before, during (1.5 hr after the bolus heparinization at the beginning of hemodialysis) and after hemodialysis therapy in normo- (NTG; Œ) and hypertriglyceridemic (HTG; M) patients. Each time point represents the median values with minimum and maximum. #P , 0.001, NTG vs. HTG group; *P , 0.01 and **P , 0.001 vs. baseline levels within each study group.
614
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
Fig. 4. Cellular [3H]-oleate incorporation into cholesteryl ester in fibroblasts (A) and murine macrophages P388 (B) induced by LDL prepared from NTG and HTG patients before [HTG-LDL (f) and NTG-LDL () and during hemodialysis [HTG-LDL (hemodialysis) (M) and NTG-LDL (hemodialysis) (‚)]. #P , 0.05, between NTG-LDL and the corresponding HTG-LDL concentration; *P , 0.05, between hemodialysis-modified LDL and baseline LDL within each group and LDL concentration. Values are mean 6 SD.
ester formation was significantly increased in comparison to the baseline LDL (P , 0.05, respectively, in Fig. 4 A and B). Nevertheless, esterification rates in fibroblasts did not reach the values obtained with LDL from normolipidemic patients. No differences in the esterification rates were observed when cells were preincubated with normotriglyceridemic LDL derived during hemodialysis or at baseline (Fig. 4). DISCUSSION This study demonstrates that triglyceride-enriched and cholesterol-depleted LDL from hypertriglyceridemic patients with ESRD leads to an increased and unsaturable cholesterol esterification rate in macrophages and a decreased rate in fibroblasts when compared with LDL from normolipidemic patients. In addition, modification of LDL lipids during hemodialysis obviously due to a lipase-mediated triglyceride hydrolysis and not to an increased CETP activity enforces these effects particularly in macrophages. An atherogenic aspect of hypertriglyceridemia in ESRD is the prevalence of small and dense LDL fractions known to be associated with marked changes in the LDL composition [16, 29 –31]. In our hypertriglyceridemic patients, decreasing LDL diameters were related to decreasing amounts of unesterified cholesterol and an exchange between cholesteryl esters and triglycerides, thus proposing a role of lipid transfer in LDL size reduction. Since accelerated CETP activity might explain the exchange between triglycerides and cholesteryl esters in LDL, Deckelbaum and collegues proposed an in vitro model for the size reduction of isolated LDL involving lipid transfer [32].
Indeed, the baseline CETP activity in our hypertriglyceridemic patients tends to be increased but differences did not reach statistical significance. In addition, CETP did not correlate with LDL components or particle diameter. This finding is consistent with results of several other investigators who have also failed to demonstrate a significant correlation between CETP activity and LDL particle diameter [33–35]. In this context, it must be taken into consideration that alterations of the cholesterol ester transfer (CET) are not only determined by CETP levels, but also by the quantity and the quality of the interacting lipoprotein particles. There is evidence that both an increase in triglyceride-rich lipoproteins and compositional changes in VLDL or HDL affected their participation in CET and potentially drives a higher exchange with LDL at the same CETP level [36 –39]. In this context, it is noteworthy that our HTG group had increasing triglycerides in HDL and an increase of cholesterol in VLDL. Both compositional changes might not only be a result of an accelerated cholesterol ester transfer, but might also trigger their process [38, 39]. Furthermore, the increase of cholesterol in the VLDL-fraction additionally suggests a large amount of remnant particles in the plasma of our HTG patients. Although the VLDL-CH levels in our patients are quite high in comparison to other studies, there is evidence that HTG patients with ESRD have a severe defect in clearing cholesterol and triglyceride enriched remnants [9, 14, 40 – 42]. In this context, Weintraub and colleagues described a threefold accumulation of remnants after more than 10 hours postprandially in HTG patients with ESRD in comparison to dyslipidemic patients without renal failure [42].
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
One hypothesis concerning an active role of small LDL particles in the atherogenesis came from observations about their increased susceptibility to in vitro oxidation. The increased susceptibility is obviously related to the LDL composition and might be explained in part by the lower antioxidant content in small dense LDL [42– 44]. Nevertheless, only a few studies investigated the oxidizability of LDL in patients with renal failure, and the results were somewhat contradictory [45– 47]. Another hypothesis about the potential atherogenicity of hypertriglyceridemic LDL might be gained from our cell culture experiments. In fact, triglyceride-enriched LDL enhanced the cholesteryl ester formation in cultured fibroblasts to a lesser degree than LDL from normolipidemic subjects. Since apoB-containing lipoproteins are taken up in fibroblasts by the classical LDL receptor, decreased intracellular cholesteryl ester formation might be caused by a decrease in the receptor affinity. Such a circumstance might be attributed to the higher triglyceride content in LDL altering the conformation of apoB [36, 48, 49], or to an oxidative modified recognition site. On the other hand, the unsaturable esterification rates in murine macrophages– our in vitro model for the nonspecific receptor pathway– demonstrated that LDL from HTG subjects enhanced the uptake more likely than LDL from NTG patients. From all these facts, it could be proposed that a decreased uptake of hypertriglyceridemic LDL by the specific pathway in vivo leads to a prolonged clearance of these particles, followed by an intracellular deposition of LDL-cholesterol in macrophages [50, 51]. In comparison to baseline levels, the lipid profile in patients with ESRD was adversely affected during hemodialysis obviously due to the peri-dialytical anticoagulation [20]. Particularly in hypertriglyceridemic patients, a shortterm decrease in triglyceride rich lipoproteins and a corresponding excess of FFA concentrations illustrated a postheparin activation of the lipolytic system. Besides these effects, an increase in the total LDL cholesterol concentrations and a significant corresponding decrease in the LDL particle diameter have been observed and recently described [20]. The present data suggest that the decrease in LDL diameters during hemodialysis is caused by a substantial reduction of the core triglycerides (Fig. 2). These observations are consistent with findings that lipoprotein lipase (LPL) as well as the hepatic lipase (HL) induce alterations in the LDL composition that are characterized by a reduction in core triglycerides [37, 45, 52], and a conversion towards subpopulations of smaller size [53, 54]. On the other hand, Lagrost, Gambert and Lallement [55] demonstrated in vitro that not lipolysis alone, but sequential effects of lipolysis and lipid transfer reactions induced dramatic changes in the mean size of LDL. In our study, CETP activity increased during hemodialysis with a peak at 1.5 hours after the beginning of anticoagulation and lipid
615
ingestion. By these facts, it is difficult to determine whether the increase in CETP activity is the result of postprandial changes or the effect of hemodialysis procedure. We prefere the latter interpretation, since several studies in patients without ESRD usually demonstrated a postprandial CETP peak in correlation to the remnant peak four to six hours after lipid ingestion [48, 56]. Therefore, our observations of an early CETP peak more likely illustrated an effect of hemodialysis than a “physiological” time course. It might be possible that the increase in CETP activity during the early phase of hemodialysis is triggered by the postheparinic excess of FFA. This interpretation is supported by recent findings that triglyceride-lipolysis in general enhanced the catalytic efficiency of CETP between triglyceride-rich and cholesterol-rich lipoproteins [50, 57, 58]. These observations may be result of an increase in lipid-bound FFA that influenced the binding properties of CETP to lipoproteins, resulting in an accelerated transfer rate. In addition, the increased presence of lipids available for the exchange with cholesterol esters during lipolysis may promote the formation of shuttle complexes containing CETP [57]. However, the physiological significance of the short-term CETP activation during hemodialysis on the LDL size pattern remains uncertain, since the LDL cholesterol content was unaltered and the depletion of triglycerides was due to the hydrolytic activity. From these observations, we propose that compositional changes of LDL during hemodialysis may also affect the properties of the lipoprotein surface, thereby modifying the interaction of the lipoprotein with the receptor pathway. Indeed, LDL derived from HTG patients during hemodialysis led to an accelerated cholesteryl ester formation in both fibroblasts and murine macrophages. These findings are of physiological significance in several respects: since the LDL-receptor in fibroblasts is regulated by the intracellular cholesterol content, an accelerated uptake of LDL usually results in a down-regulation of the receptor expression [51]. An increase in the number of circulating LDL in vivo may be the respective result and contributes to the observed increase of LDL-cholesterol [20] during hemodialysis. Since the scavenger receptor activity in macrophages is not regulated by the cell cholesterol content, increased uptake can lead to an unsaturable intracellular accumulation of cholesterol and foam cell generation [50, 51]. In conclusion, we believe that small and triglycerideenriched LDL particles derived from dyslipidemic patients with ESRD are an inferior substrate for the specific pathway in fibroblasts and a superior substrate for the nonspecific pathway in macrophages. In addition, modifications of the core triglycerides of LDL during hemodialysis may provide an alternative means of causing macrophages to take up and accumulate cholesterol. Since these effects potentially occur in vivo, our data might contribute to the
616
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
understanding of the high prevalence of premature atherosclerosis in hypertriglyceridemic patients with ESRD on the maintenance of hemodialysis. Reprint requests to Dr. Andreas Ambrosch, Institute of Clinical Chemistry, University Hospital of Magdeburg, Leipziger Str. 44, 39120 Magdeburg, Germany. E-mail: [email protected]
APPENDIX Abbreviations used in this article are: apo, apolipoprotein; BMI, body mass index; CE, esterified cholesterol; CETP, cholesteryl ester transfer protein; CH, cholesterol; ESRD, end-stage renal disease; FC, unesterified cholesterol; FFA, free fatty acids; HDL, high density lipoprotein; HL, hepatic lipase; HTG, hyperlipidemic patients; LDL, low density lipoprotein; LPL, lipoprotein lipase; NTG, normolipidemic patients; PAGGE, nondenaturating polyacrylamide gradient gel electrophoresis; TG, triglyceride; VLDL, very low density lipoprotein.
REFERENCES 1. KJELLSTRAND CM, HYLANDER B, COLLINS A: Mortality on dialysis. On the influence of early start, patients characteristics and transplantation and acceptance rates. Am J Kidney Dis 15:483– 490, 1990 2. GOKAL R, KING J, BOGLE S, MARSH F, OLIVER D, JAKUBOWSKI C, HUNT L, BAILLOD R, OGG C, WARD M, WILKINSON R: Outcome in patients on continuous ambulatory peritoneal dialysis and hemodialysis: 4 years analysis of a prospective multicentric study. Lancet 2:1105–1109, 1987 3. UNITED STATES RENAL DATA SYSTEM: Annual data report 1990. Am J Kidney Dis 16(Suppl):S44 –S52, 1990 4. World Health Annual Statistics. Geneva, Switzerland, World Health Organization, 1990, pp 166 –169 5. MA KW, GREEN EL, RAIJ L: Cardiovascular risk factors in the chronic renal failure and hemodialysis population. Am J Kidney Dis 19:505– 513, 1992 6. BRUNZELL JD, ALBERS JJ, HAAS LB, GOLDBERG AP, AGADOA L, SHERRARD DJ: Prevalence of serum lipid abnormalities in chronic hemodialysis. Metabolism 26:903–910, 1977 7. BAGDADE JD, ALBERS JJ: Plasma high density lipoprotein concentrations in chronic hemodialysis and renal transplant patients. N Engl J Med 296:1436 –1449, 1977 8. HAAS LB, WAHL PW, SHERRARD DJ: A longitudinal study of lipid abnormalities in renal failure. Nephron 33:145–159, 1983 9. CAMEJO G, RIERA G, LEE M, LOPEZ F: Lipoprotein structural abnormalities in chronic failure with and without hemodialysis. BBA 47:239 –245, 1988 10. RUBIES-PRAT J, ESPINEL E, JOVEN J, RAS MR, PIRA L: High density lipoprotein subfractions in chronic renal failure. Am J Kidney Dis 9:60 – 65, 1987 11. SENTI M, ROMERO R, PEDRO-BOTET J, PELEGRI A, NOGUES X, RUBIED-PRAT J: Lipoprotein abnormalities in hyperlipidemic and normolipidemic men on hemodialysis with chronic renal failure. Kidney Int 41:1394 –1399, 1992 12. CHAN MK, PERSAUD J, VARGHESE Z, MOORHEAD JF: Pathogenic roles of postheparin lipases in lipid abnormalities in hemodialysis patients. Kidney Int 25:812– 818, 1984 13. LINDHOLM B, NORBECK HE: Serum lipids and lipoproteins during continuous ambulatory peritoneal dialysis. Acta Med Scand 220:143– 151, 1986 14. NESTEL PJ, FIDGE NH, TAN MH: Increased lipoprotein remnant formation in chronic renal failure. N Engl J Med 307:329 –333, 1982 15. RON D, OREN I, AVIRAM M, BELTER DS, BROOK JG: Accumulation of lipoprotein remnants in patients with chronic renal failure. Atherosclerosis 46:67–75, 1983 16. MCNEIL D, LEE P, MURPHY B, BEST J: Low density lipoprotein particle size distribution in end-stage renal disease treated with hemodialysis or peritoneal dialysis. Am J Kidney Dis 27:84 –91, 1996 17. ATTMAN PO, ALAUPOVIC P, GUSTAVSON A: Serum apolipoprotein profile of patients with chronic renal failure. Kidney Int 32:368 –375, 1987
18. BERGESIO F, MONZANI G, CIUTI R, SERRUTO A, BENUCCI A, FRIZZI V, SALVADORI M: Lipids and apolipoprotein changes during the progression of chronic renal failure. Clin Nephrol 38:264 –270, 1992 19. CORESH J, KWITEROVICH PO, SMITH HH, BACHORIC PS: Association of TG concentrations, LDL particle diameter, density and chemical composition with premature coronary heart disease in men and women. J Lipid Res 34:1687–1697, 1993 20. LULEY C, NEUMANN KH, WESTPHAL S, AMBROSCH A: Atherogenic effects of hemodialysis: Increased generation of small and cholesterol poor LDL particles. (abstract) Atherosclerosis 129:141, 1997 21. ORTH M, WAHL S, HANISCH M, FRIEDRICH I, WIELAND H, LULEY C: Clearance of postprandial lipoproteins in normolipidemics: Role of the apolipoprotein E phenotype. BBA 1303:22–30, 1996 22. RAINWATER DL, ANDRES DW, FORD AL, LOWE F, BLANCHE PJ, KRAUSS RM: Production of polyacrylamide gradient gels for electrophoretic resolution of lipoproteins. J Lipid Res 33:1876 –1881, 1992 23. WILLIAMS PT, KRAUSS RM, NICHOLS AV, VRANIZAN KM, WOODS PDS: Identifying the predominant peak diameter of high-density and low-density lipoproteins by electrophoresis. J Lipid Res 31:1131–1139, 1990 24. TATO F, VEGA GL, GRUNDY SM: Determinants of plasma HDLcholesterol in hypertriglyceridemic patients. Arterioscler Thromb Vasc Biol 17:56 – 63, 1997 25. AMBROSCH A, TATO F, LULEY C: Fluorescent determination of CETP activity in plasma. (abstract) Atherosclerosis 134:100, 1997 26. BROWN MS, HO YK, GOLDSTEIN JL: Cholesteryl ester cycle in macrophages. J Biol Chem 255:9344 –9352, 1980 27. LOWRY OH, ROSEBROUGH NJ, FARR AL, RANDALL AJ: Protein measurement with the Folin-phenol reagent. J Biol Chem 193:265– 275, 1951 28. KRAUSS RM, BURKE DJ: Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res 23:236 –244, 1982 29. STEWART MW, LAKER MF, DYER RG, GAME F, MITCHESON J, WINCOUR PH, ALBERT KGM: Lipoprotein compositional abnormalities and insulin resistance in type II diabetic patients with mild hypertlipidemia. Arterioscler Thromb 12:1046 –1052, 1993 30. HOKANSON JE, KRAUSS RM, ALBERS JJ, AUSTIN MA, BRUNZELL JD: LDL physical and chemical properties in familial combined hyperlipidemia. Arterioscler Throm Vasc Biol 15:452– 459, 1995 31. CAPELL WH, ZAMBON A, AUSTIN MA, BRUNZELL JD, HOKANSON JE: Compositional differences of LDL particles in normal subjects with LDL subclass phenotype A and LDL subclass phenotype B. Arterioscler Thromb Vasc Biol 16:1040 –1046, 1996 32. DECKELBAUM RJ, EISENBERG S, FAINARU M, BARENHOLZ Y, OLIVECRANO T: In-vitro production of human low density lipoprotein-like particles: A model for very low density lipoproteoin catabolism. J Biol Chem 254:6079 – 6087, 1979 33. KARPE F, TORNVALL T, OLIVECRANO T, STEINER G, CARSON LA, HAMSTON A: Composition of human low-density lipoprotein: Effect of postprandial triglyceride-rich lipoproteins, lipoproteine lipase, hepatic lipase and cholesterol ester transfer protein. (abstract) Atherosclerosis 98:33, 1993 34. TAN KCB, COOPER MB, LING KLE, GRIFFIN BA, FREEMAN DJ, PACKERD CJ, SHEPHERD J, HALES CN, BATTERIDGE DJ: Fasting and postprandial determinants for the occurrence of small dense LDL species in NIDDM patients with and without hypertriglyceridemia: The involvement of insulin, insulin precursor species and insulin resistance. (abstract) Atherosclerosis 113:275, 1995 35. WATSON TDG, CASLAKE DJ, FREEMAN BA, GRIFFIN BA, HINNIE J, PACKARD CJ, SHEPHERD J: Determinants of LDL subfraction distribution and concentration in young normolipidemic subjects. (abstract) Arterioscler Thromb 14:902, 1994 36. AVIRAM M, BIERMAN EL, CHAIT A: Modification of low density lipoprotein by lipoprotein lipase or hepatic lipase induces enhanced uptake of cholesterol accumulation in cell. J Biol Chem 263:15416 – 15422, 1988 37. DECKELBAUM RJ, EISENBERG S, OSCHRY Y, BUTBUL E, SHARON I, OLIVECRONA T: Reversible modification of human plasma low density lipoproteins towards triglyceride-rich precursors: A mechanismn for losing excess cholesterol esters. J Biol Chem 257:6509 – 6517, 1982 38. BAGDADE JD, RITTER MC, SUBBAIAH PV: Accelerated cholesteryl
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
39. 40. 41. 42.
43.
44. 45. 46.
47.
48.
ester transfer in patients with insulin-dependent diabetes mellitus. Eur J Clin Invest 21:161–167, 1991 EISENBERG S: Preferential enrichment of large-sized very low density lipoprotein populations with transferred cholesteryl esters. J Lipid Res 26:487– 494, 1985 WANNER C, FROMMHERZ K, HORL WH: Hyperlipoproteinemia in chronic renal failure: Pathophysiological and therapeutical aspects. Cardiology 78:202–217, 1991 CHEUNG A, WU LL, KABLITZ C, LEYPOLDT K: Atherogenic lipids and lipoproteins in hemodialysis patients. Am J Kidney Dis 22:271–276, 1993 WEINTRAUB M, BURSTEIN A, RASSIN A, LIRON M, RINGEL Y, CABILI S, BLUM M, PEER G, IAINA A: Severe defect in clearing postprandial chylomicron remnants in dialysis patients. Kidney Int 42:1247–1252, 1992 DEGRAAF J, HAK-LEMMERS HL, HECTORS PC, DEMARKER PNM, HENDRIKS JCM, STALENHOEF AFH: Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects. Arterioscler Thromb 11:298 –306, 1991 TRIBBLE DL, HOLL LG, WOOD PD, KRAUSS RM: Variation of oxidative susceptibility among six low density lipoprotein subfractions of varying size and density. Atherosclerosis 93:189 –199, 1992 SUTHERLAND W, WALKER RJ, BALL MJ, STAPLEY S, ROBERTSON MC: Oxidation of low density lipoproteins from patients with renal failure or renal transplants. Kidney Int 48:227–236, 1995 LOUGHREY CM, YOUNG IS, MCENENY J, MCDOWELL FW, MCMASTER C, MCNAMEE PT, TRIMBLE ER: Oxidation of low density lipoprotein in patients on regular hemodialysis. Atherosclerosis 110:185–193, 1994 MAGGI E, BELLAZZI R, FALASCHI F, FRATTONI A, PERANI G, FINARDI G, GAZO A, NAI M, ROMANINI D, BELLOMO G: Enhanced LDL oxidation in uremic patients: An additional mechanism for accelerated atherosclerosis. Kidney Int 45:876 – 8883, 1994 AGNANI G, BARD JM, CANDELIER L, DELATTRE S, FRUCHARD JC,
49. 50. 51. 52. 53. 54.
55. 56.
57.
58.
617
CLAVEY V: Interaction of LpB, Lpb:E, LpB:CIII, LpB:CIII:E lipoproteins with the low density lipoprotein receptor of HeLa cells. Arterioscler Thromb 11:1021–1029, 1991 NICOLL A, LEWIS B: Evaluation of the roles of lipoprotein lipase and hepatic lipase in lipoprotein matabolism: In-vivo and in-vitro studies in men. Eur J Clin Invest 10:487– 495, 1980 ROSS RN: The pathogenesis of atherosclerosis — An update. N Engl J Med 314:488 –500, 1986 BROWN MS, GOLDSTEIN JL: The low density lipoprotein pathway and is relation to atherosclerosis. Annu Rev Biochem 52:223–261, 1983 MUSLINER TA, HERBERT PN, KINGSTON NJ: Lipoprotein substrates of lipoprotein lipase and hepatic triglycerol lipase from human postheparin plasma. BBA 575:277–288, 1979 HOMMA, Y, NAKAYA N, NAKAMURA H, GOTO Y: Increase in the density of lighter low density lipoprotein by hepatic triglyceride lipase. Artery 13:19 –31, 1985 LAGROST L, FLORENTIN E, GUYARD-DANGREMONT V, ATHIAS A, GANDJINI H, LALLEMANT C, GAMBERT P: Evidence for nonesterified fatty acids as modulators of neutral lipid transfer in normolipidemic human plasma. Arterioscler Thromb Vasc Biol 15:1388 –1396, 1995 LAGROST L, GAMBERT P, LALLEMENT C: Combined effects of lipid transfer and lipolysis on gradient gel patterns of human LDL. Arterioscl Thromb 14:1327–1336, 1994 KARPE F, TORNVALL P, OLIVECRONA T, STEINER G, CARLSON LA, HAMSTEN A: Composition of human low density lipoprotein: Effect of postprandial triglyceride-rich lipoproteins, lipoprotein lipase, hepatic lipase and cholesteryl ester transfer protein. Atherosclerosis 98:33– 49, 1993 TALL AR, SAMMET D, VITA G, DECKELBAUM RJ, OLIVECRONA T: Lipoprotein lipase enhances the cholesteryl ester transfer proteinmediated transfer of cholesteryl esters from high density lipoproteins to very low density lipoproteins. J Biol Chem 260:6687– 6697, 1985 SAMMET D, TALL AR: Mechanismns of enhancement of cholesteryl ester transfer protein activity by lipolysis. J Biol Chem 260:6687– 6697
Compositional and functional changes of low-density lipoprotein during hemodialysis in patients with ESRD1 ANDREAS AMBROSCH, UTE DOMROESE, SABINE WESTPHAL, JUTTA DIERKES, WOLFGANG AUGUSTIN, KLAUS H. NEUMANN, and CLAUS LULEY Institute of Clinical Chemistry, Department of Pathobiochemistry, and Department of Medicine, Division of Nephrology, University Hospital of Magdeburg, Magdeburg, Germany
Compositional and functional changes of low-density lipoprotein during hemodialysis in patients with ESRD. Background. This study focused on the effects of hemodialysis on the atherogenic properties of low density lipoprotein (LDL) in patients with end-stage renal disease (ESRD). The impact of cholesterol ester transfer protein (CETP) activity and lipolysis on LDL composition, particularly the changes during hemodialysis, was investigated. Methods. Blood was drawn from 15 normotriglyceridemic (NTG) and 15 hypertriglyceridemic patients [HTG; triglycerides (TG) ,2.2 mmol/liter] before hemodialysis, during (1.5 hr after the beginning of anticoagulation) and at the end of treatment. In each sample, lipid values and CETP activity were measured. LDL was prepared and characterized by its components and diameters (2 to 16% PAGGE). To investigate the functional properties of LDL, fractions obtained from NTG and HTG patients were incubated with human skin fibroblasts and a cell line of murine macrophages (P388), and cholesterol ester formation rates were measured. Results. In comparison to LDL from NTG patients at baseline, HTG-LDL were enriched in triglycerides (P , 0.02), depleted in cholesterol proportion (P , 0.01) and small in size (P , 0.001). These LDL induced the cholesterol esterification rates (50 mg/mL LDL-protein) in a twofold greater unsaturation in macrophages when compared to LDL from NTG patients (P , 0.04). The rates in fibroblasts were reduced by approximately half (P , 0.05). During hemodialysis, LDL were decreased in size (P , 0.001) and depleted in TG (P , 0.01), particularly in the hypertriglyceridemic state. Although CETP activity increased during hemodialysis (P , 0.001), the cholesterol content remained unchanged. When HTG-LDL obtained during hemodialysis were incubated with cells, esterification rates particularly in macrophages were markedly accelerated in comparison to the unmodified lipoprotein at baseline (P , 0.05). Conclusion. LDL from HTG patients with ESRD was TGenriched, CH-depleted and smaller in size. As the intracellular 1
This work is dedicated to Professor Dr. Dietrich Seidel, Klinikum Grobhadern, University of Munich, on the occasion of his 60th birthday. Key words: end-stage renal disease, hypertriglyceridemia, low density lipoprotein, CETP activity, esterification rate, atherosclerosis. Received for publication June 30, 1997 and in revised form March 18, 1998 Accepted for publication March 18, 1998
© 1998 by the International Society of Nephrology
esterification rates induced by LDL were related to the cellular uptake, these LDL were a superior substrate for murine macrophages with the tendency of intracellular accumulation, and an inferior substrate for fibroblasts suggesting a decreased uptake by the specific receptor pathway. TG-depletion of LDL during hemodialysis, particularly in HTG patients due to a lipasemediated TG-hydrolysis, increased these effects in macrophages. We suggest that the alterations of LDL that occur during repeated hemodialysis in vivo could contribute to the high prevalence of premature atherosclerosis found in HTG patients with ESRD.
End-stage renal disease (ESRD) is associated with accelerated atherosclerosis and a high incidence of cardiovascular disease [1, 2]. In the United States, the death rate from myocardial infarction for patients with ESRD aged between 45 and 64 years is 24.1 per 1000 patient-years, which is far in excess of the general population [3, 4]. Although the course of increased atherosclerotic cardiovascular disease in these patients is probably multifactorial, lipid abnormalities are thought to play a major role [5] and a number of alterations in lipids have been described. The primary lipid abnormality found in ESRD is elevated triglyceride (TG) levels that are predominantly due to reduced lipolysis of triglyceride-rich lipoproteins [6 –9], which is often accompanied by a reduced high-density lipoprotein (HDL) cholesterol level [10, 11]. These abnormalities are present in patients with ESRD who are not on dialysis as well as patients treated either with hemodialysis or continuous ambulatory peritoneal dialysis [12, 13]. Elevated TG levels can indicate accumulation of triglyceriderich, potentially atherogenic remnant particles, a finding reported in patients treated with chronic hemodialysis [14 –16]. Low-density lipoprotein (LDL) cholesterol levels are not elevated in ESRD [17, 18], but changes in size and composition in association with hypertriglyceridemia could potentially result in increased atherogenicity of LDL particles [19]. Recently, Luley et al described a small-dense LDLpattern for dyslipidemic patients with ESRD and, in addition, an acute alteration in LDL size during the course of
608
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
hemodialysis [20]. These changes in LDL diameter were pronounced in dyslipidemic patients and accompanied by an increase in total LDL-cholesterol during each hemodialysis session. Consequently, it can be theorized that repeated hemodialysis in itself has an unfavorable effect on the atherogenic profile of LDL. In the present study, we determined the impact of cholesteryl ester transfer protein (CETP) activity and lipolytic activity on changes in LDL composition and size, before and during hemodialysis in normo- and dyslipidemic patients with ESRD. Furthermore, we investigated the functional properties of this lipoprotein species in vitro in relation to the compositional changes caused by hemodialysis. METHODS Study subjects A total of 30 patients with ESRD (17 male and 13 female; from the Department of Internal Medicine, Center for Kidney Disease, University Hospital Magdeburg, Germany) were enrolled in the study. The patients group consisted of 15 hypertriglyceridemic subjects (total triglycerides . 2.2 mmol/liter) and 15 subjects with normal plasma triglycerides. All patients were hemodialyzed against a bicarbonate bath for three to four hours in a treatment three times weekly. Underlying renal disease were: diabetic nephropathy (N 5 7), chronic glomerulonephritis (N 5 7), small kidneys of unknown cause (N 5 9), polycystic kidney disease (N 5 3), pyelonephritis (N 5 3), and Alport syndrome (N 5 1). A total of nine patients were diabetic: three in the NTG group and six in the HTG group (7 patients were on hemodialysis because of diabetic nephropathy, 1 because of polycystic kidney disease and 1 because of pyelonephritis). Two patients in the HTG group and one in the NTG group had a history of renal transplantation with a consecutive transplant dysfunction. In both groups, the prevalence of a history for coronary artery disease (CAD), peripheral vascular disease and for stroke was evaluated: history of CAD in HTG versus NTG patients 6 of 15 versus 4 of 15; for peripheral vascular disease 7 of 15 versus 3 of 15; for stroke 4 of 15 versus 0 of 15. Patients with a history of recent myocardial infarction (within 3 months), unstable angina, severe congestive heart failure, recent stroke and hypertension with systolic RR . 200 mm Hg (and/or diastolic RR . 105) were excluded. In addition, no patient had acute bacterial or viral infection except one patients who had a history of hepatitis C. Since most of our patients suffered from biochemical hyperparathyroidism (serum parathyroid hormone . 65 ng/liter), only patients with a symptomatic manifestation (hyperparathyroid bone disease, electrocardiographic signs of hypercalcemia) were excluded. Exclusion criteria did not allow to study patients with cancer (or a history of cancer). Informed consent was obtained from all patients.
609
Study design Venous blood was drawn after a 12-hour overnight fast to obtain baseline values. Then, a lipid load test was performed (a) because patients normally are advised not to fast before a hemodialysis procedure; (b) to standardize the postprandial lipoprotein metabolism of the patients [21]. Fifteen minutes after the lipid load, the hemodialysis procedure began. Furthermore, two additional blood samples were obtained, one 1.5 hours after the beginning of hemodialysis (peri-dialytically) and one at the end of each hemodialysis session (post-dialytically). The anticoagulant that was employed contained unfractionated heparin during each hemodialysis. All patients were within 2% of their fluid free weight, which was used to calculate the individual heparin doses. The anticoagulant was administered with a single intravenous bolus (60 IU/kg body wt) at the beginning of each hemodialysis session and with a subsequent infusion of 1000 IU/hr. Lipoproteins EDTA-plasma (EDTA 1 mg/liter) for lipoprotein analysis was separated without delay by centrifugation at room temperature and immediately processed or stored at 220°C. For baseline lipoprotein and apoprotein concentrations, very low density lipoprotein (VLDL) and chylomicrons were separated from LDL and HDL by a single centrifugation step over a time period of 18 hours at 100,000 g in a TI 50.4 rotor with a Beckman L-50 ultracentrifuge (all from Beckman, Munich, Germany). The layer of VLDL on the top of the tubes was removed with a Pasteur-pipette. In the infranatent, the LDL fraction was precipitated with phosphotungsten acid/MgCl2 reagent before measurement of HDL composition. The composition of the VLDL-, HDL-fractions and the infranatent were determined by measurement of CH, TG, FC and apoproteins on an automated Hitachi 911 analyzer (Boehringer, Mannheim, Germany). LDL isolation. After separation of chylomicrons by centrifugation at 100,000 g [density (d) , 1.006 g/ml] for 30 minutes, the infranatent was increased to d 5 1.019 mg/liter by the addition of solid potassium bromide (Sigma, Munich, Germany) and then centrifuged at 150,000 g (24 hr at 10°C). The floating intermediate density lipoproteins (IDL) and VLDL were collected. The infranatent was then brought to d 5 1.060 g/liter and centrifuged at 150,000 g (24 hr at 10°C). The floating LDL was obtained and dialyzed against a Tris-EDTA buffer for 24 hours at 8°C and subsequently characterized by their CH, FC, CE, TG and apo B content. Lipoprotein component assays Total apo AI, B, CIII and E were analyzed by immunturbidimetric assays using goat antihuman polyclonal antisera (Greiner, Munich, Germany). For measurement of
610
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
cholesterol, we used the CHOD-PAP method (Chol kit®; Boehringer), for triglycerides and free glycerol the GPOPAP method (Triglyceride GPO-PAP kit®; Boehringer) was used, for free cholesterol the COD-PAP method (free cholesterol C kit®; WAKO, Munich, Germany) and for measurement of free fatty acids the ACS-ACOD method (NEFA C®; WAKO). The CE content was calculated from total CH and total FC concentrations. The hematocrit was measured in each sample and used for corrections of lipoprotein levels. Nondenaturating polyacrylamide gradient gel electrophoresis To estimate LDL particle size, a nondenaturating polyacrylamide 2 to 16% gradient gel electrophoresis (PAGGE) was performed, modified from [22]. Briefly, the stock solutions in the TE-buffer (Tris 210 mM, sodium acid 3 mM, disodium EDTA 3 mM, pH 8.35; all from Sigma) included in following: the high limit solution acrylamide 152.3 g/liter, bis-acrylamide 7.7 g/liter (16% total, 4% crosslinker) and glycerol 260 g/liter (all from Sigma), and the low limit solution acrylamid, 16 g/liter and bis-acrylamide 4 g/liter (2% total, 4% crosslinker). The high and the low limit solutions were made immediately before the gradient was cast and contained 1 volume high or low limit solution, 0.0015 vol freshly prepared ammonium persulfate (100 g/liter; Sigma), and 0.00025 vol 3-dimethylaminopropionitril (TEMED; Sigma). The gradient was generated with a BioRad Econo-pump (2.6 ml/min; BioRad, Heidelberg, Germany), mixed in an external mixing chamber (BioRad), and cast in a gel slab casting apparatus (180 3 160 mm). The gels were prerun in TE buffer at 60 V for 30 minutes. Samples were made dense with glycerol (1:3 vol/vol), and a volume containing 20 ml plasma samples was loaded on the gel. Electrophoresis time was two hours at 60 V, 12 hours at 50 V and six hours at 300 V, with a constant voltage power supply (BioRad) and a constant temperature of 8°C. At the end of electrophoresis, gels were stained with Coomassie brilliant blue G (Sigma) for one hour and destained in a solution of 50% methanol, 10% acetic acid and 40% water. Latex beads 34 nm (Duke Scientific, Palo Alto, CA, USA), thyroglobulin 17 nm and ferritin 12.2 nm (Pharmacia, Freiburg, Germany) were run as standards. The apparent diameters of LDL were determined by using a calibration curve generated from standards according to Williams et al [23]. Reproducibility of the gel analysis of LDL size was evaluated by analyzing the same LDL preparation three times. Coefficient of variation of mean LDL diameters between runs was 5%. Cholesteryl ester transfer protein activity Cholesteryl ester transfer protein (CETP) activity was assessed in a defined system by using a fluorescent cholesteryl linoleate analogue (WAK, Bad Homburg, Germany). Exogenous donor and acceptor particles (human
VLDL) were combined with the CETP source (EDTAplasma) within a total volume of a 10 mM Tris buffer. The donor particle consisted of a nitrobenzooxadiol-fluorophor (NBD), which takes advantage of a self-quenching effect on microemulsion incorporated cholesteryl linoleate. The CETP-facilitated transfer of fluorescent cholesteryl linoleate from the self-quenching donor to the acceptor was determined by an increase in fluorescence intensity (incubation time 3 hr, 37°C; excitation wavelength 465 nm (EX 456 nm) and emission wavelength of 535 nm (EM 535 nm) using an LS50 fluorescence spectrometer (Perkin Elmer, Hannover, Germany). A standard curve to derive the relation between fluorescence intensity and mass transfer of cholesteryl linoleate (pmol/3 hr) was generated by dispersing donor particles in isopropanol in a linear dilution (cholesteryl linoleate concentration was 80.9 mg/ml). Spectrally pure (HPLC grade) isopropanol was used as the solvent to avoid background fluorescence of the blank (isopropanol). From each sample, triplicates were measured and the mean fluorescences were calculated. The intra-assay variance of the mean CETP activities was calculated with 10.8% and the inter-assay variance with 11.6%. The fluorimetric CETP assay was compared to a recently described isotopic CETP assay [24], and a good correlation was found (N 5 25, r 5 0.90, P , 0.001) [25]. Cellular [3H]cholesteryl ester formation in cultured human fibroblasts and murine macrophages Cells. Human skin fibroblasts were cultured from biopsies of the skin from normal adult donors. The cells were plated at 7 3 104 cells/35 mm plastic dish in Dulbecco’s minimal essential medium (DMEM, Sigma), containing sodium hydrogen carbonate 25 mM, HEPES 20 mM, streptomycin 100 mg/liter, penicillin 105 IU/liter (Sigma) and fetal calf serum (FCS) 10% (Biochrom, Munich, Germany). Cultures of fibroblasts in passages 4 to 6 were washed with Dulbecco’s phosphate-buffered saline (PBS, Sigma) and incubated in fresh medium containing lipoproteindeficient serum 10% (LPDS) to up-regulate apoB/E receptor activity before experiments. The P388 murine macrophage-like cell line was purchased from American Type Culture Collection (ATCC). P388 cell were plated at 7 3 104 cells/35 mm dish in RPMI 1640 (Sigma) supplemented with FCS 10%, penicillin, streptomycin and glutamine. The cells were fed every three days and were used for experiments within seven days of plating. Metabolism of LDL in cell culture. For cellular experiments, pooled LDL (d 5 1.06) from six NTG and six HTG subjects, respectively, was prepared from fasting plasma (baseline LDL) and from plasma after heparinization (peridialytically; 1.5 hr after the bolus heparinization at the beginning of hemodialysis). LDL cholesterol uptake by cells was estimated by measurement of the stimulation of [3H]-oleate incorporation into cholesteryl ester [26]. After
611
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL Table 1. Anthropometric data and lipoprotein profile of the study groups Normolipidemic group, (NTG), N 5 15
Hyperlipidemic group (HTG), N 5 15
Significance P
Age years BMI kg/m2 Gender male/female Duration of dialysis months (range) Creatinine mmol/liter Urea mmol/liter
49.3 (38 –76) 23.7 (17.8 –29.9) 8/7 34 (15–107) 768 (389 –1373) 19.3 (14.0 –31.8)
51.6 (41–74) 19.4 (18.2–21.5) 9/6 37.5 (25–167) 746 (217–1304) 20.3 (14.9 –28.6)
NS NS NS NS NS
Triglycerides mmol/liter Cholesterol mmol/liter ApoCIII mg/dl ApoE mg/dl ApoAI mg/dl VLDL triglyceride mmol/liter cholesterol mmol/liter LDL cholesterol mmol/liter triglyceride mmol/liter cholesteryl ester mmol/liter unesterified chol. mmol/liter apoB g/liter HDL cholesterol mmol/liter triglyceride mmol/liter cholesteryl ester mmol/liter unesterified cholesterol mmol/liter FFA mmol/liter CETP activity pmol/3 hr
1.54 (0.43–2.04) 5.53 (3.84 – 8.81) 13.3 (9.1–28.7) 5.13 (2.62–9.17) 1.26 (0.89 –2.14)
3.67 (2.20 –7.41) 7.44 (4.32–9.71) 21.7 (13.4 –35.0) 6.80 (3.94 –9.72) 1.14 (0.85–1.57)
, 0.000 , 0.05 , 0.02 , 0.05 NS
1.15 (0.06 –2.18) 1.35 (0.18 –2.34)
2.44 (1.88 –5.98) 2.13 (1.36 – 4.86)
, 0.001 , 0.01
3.37 (2.23– 6.71) 0.39 (0.23– 0.53) 2.42 (1.46 –5.03) 0.96 (0.60 –1.72) 0.85 (0.58 –1.70)
3.16 (1.96 –5.32) 0.64 (0.30 –1.01) 2.29 (1.22–3.95) 0.81 (0.61–1.34) 1.04 (0.65–1.51)
NS , 0.02 NS NS NS
1.24 (0.86 – 4.17) 0.20 (0.13– 0.35) 1.02 (0.69 –3.27) 0.20 (0.03– 0.90) 0.36 (0.12–1.31) 22.0 (9.4 –37.1)
1.04 (0.69 –2.11) 0.25 (0.19 – 0.38) 0.87 (0.58 –1.81) 0.16 (0.08 – 0.35) 0.46 (0.21– 0.77) 24.2 (5.3– 45.6)
NS , 0.03 NS NS NS NS
Values are median (min-max). FFA is free fatty acids.
a 48 hour incubation with medium containing LPDS, cells were washed with PBS and fresh medium containing either LPDS alone or LPDS with LDL. After 24 hours of incubation with lipoproteins, cultures were washed with PBS three times and incubated with [3H]-oleate (0.2 mM, 10 mCi/ml; Amersham, Braunschweig, Germany) in the presence of fatty acid-free albumin (0.07 mM) for two hours at 37°C. The cells were then washed twice with PBS, damaged in an ultrasound bath and incubated with 1 ml of chloroform/methyl alcohol (2:1, vol/vol) to extract lipids. After two more washes with these solvents, the pooled lipid extract was dryed and resolubilized in chloroform/methyl alcohol. The labeled cholesteryl ester was isolated by thin-layer chromatography on silica gel plates using hexane/ diethyl ether/acetic acid (83:16:1). The protein content of cell extracts was measured by a modified method of Lowry et al [27]. All cell experiments were done for three times in triplicates. Statistical analysis The results were presented as medians with minimum and maximum when data were not normally distributed, or as mean 6 SD (intracellular esterification rates). Data were compared using the U-test for interindividual differences and the Wilcoxon test for intraindividual differences. Differences were considered significant at P , 0.05. Coefficients of skewness and kurtosis were calculated to test deviations from a normal distribution. Associations be-
tween lipoprotein parameters were determined by calculation of Pearsons correlation coefficients or by Spearman rank correlation analysis. For all analysis, the SPSS software package (version 6.13) was used. RESULTS Baseline characteristics and lipoprotein profile of the study subjects General characteristics of the subjects, parameters of renal function and results of the biochemical lipoprotein analyses are summarized in Table 1. There were no significant differences between the HTG and NTG groups resulting from sex, age, body mass index (BMI) and parameters of renal function. The lipoprotein profiles of the HTG subjects were characterized by elevated total triglycerides (TG) (P , 0.000), total cholesterol (CH; P , 0.05), VLDL-TG (P , 0.001) and VLDL-CH concentrations (P , 0.01). Within the HTG group, no differences were found between diabetic and non-diabetic patients with regard to the lipoprotein profile (data not shown). ApoCIII and apoE corresponded well to the VLDL-TG and varied significantly between both study groups (P , 0.02 and P , 0.05, respectively). The plasma free fatty acids (FFA) were not significantly different between both groups. Whereas total CH, FC, CE and apo AI in HDL did not differ between both groups (apolipoprotein data not shown), HDL-TG was significantly increased in the HTG group
612
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL Table 2. LDL composition and size in subjects with ESRD
LDL diameter nm CH/apoB ratio CE/apoB ratio FC/apoB ratio TG/apoB ratio
Normotriglyceridemic group (NTG)
Hypertriglyceridemic group (HTG)
Significance P
26.83 (23.91–29.37) 3.77 (2.12– 4.04) 2.97 (2.06 –3.43) 1.08 (0.74 –1.59) 0.40 (0.24 – 0.78)
23.76 (22.31–26.84) 3.16 (1.81–3.89) 2.63 (1.79 –3.22) 0.92 (0.65–1.12) 0.60 (0.25–1.23)
, 0.001 , 0.01 , 0.04 , 0.001 , 0.01
Values are median (min-max). Abbreviations are: LDL, low density lipoprotein; CH, cholesterol; CE, esterfied cholesterol; FC, unesterified cholesterol; TG, triglycerides; apo, apolipoprotein.
(P , 0.03). In addition, LDL from HTG patients were enriched in triglycerides in comparison to LDL from NTG patients (P , 0.05; Table 1) and differed in their triglyceride content per apoB (P , 0.01; Table 2). Furthermore, the apparent diameter of the major plasma LDL subfraction as determined by PAGGE was significantly decreased in the HTG group (P , 0.05; Table 2). According to the classification for LDL-phenotypes by Krauss and Burke [28], 11 of 15 of the HTG patients were found to have a phenotype B-pattern of LDL (,25.5 nm) in contrast to only 4 of 15 NTG participants. The smaller LDL particles were characterized by a decreased CE content (P , 0.04) and FC content (P , 0.001) per apoB (Table 2) and, as mentionend above, by an increased TG proportion (P , 0.01, Table 2). Therefore, LDL diameters correlated inversely with the TG/apoB ratios (r 5 20.42, P , 0.02; Fig. 1A) and positively with the FC content per particle (r 5 0.43; P , 0.03; Fig. 1B). In addition, a significant inverse correlation was found between the CE and TG proportion per apoB (r 5 20.45; P , 0.02; Fig. 1C). Triglyceride depletion of LDL during hemodialysis One and a half hours (1.5 hr) after the bolus heparinization at the beginning of hemodialysis (peri-dialytically), the LDL-TG/apoB ratios decreased significantly in HTG patients [LDL-TG/apoB baseline vs. postheparinic: 0.60 (0.25 to 1.23) versus 0.40 (0.22 to 0.97); P , 0.02], but remained unchanged in the NTG individuals [LDL-TG/apoB baseline vs. postheparinic: 0.40 (0.24 to 0.91) versus 0.39 (0.23 to 0.78); NS]. The changes in the LDL-TG/apoB ratio (DLDL-TG/apoB; calculated as the differences between the baseline LDL-TG/apoB ratios and postheparinic LDLTG/apoB ratios) were positively related (r 5 0.38, P , 0.05; Fig. 2) to the decrease in the LDL diameter (DLDLdiameter; calculated as differences between baseline LDL diameter and postheparinic LDL diameter). Up to the end of the hemodialysis therapy, the LDL-TG/apoB ratios increased in the HTG patients [LDL-TG/apoB at the end of therapy, 0.50 (0.18 to 1.08); versus postheparinic values; P , 0.05], but remained significantly different from baseline (LDL-TG/apoB ratio on baseline vs. end of therapy, P , 0.05). The LDL-CE and -FC, and their ratios per apoB were unaltered in both study groups during the course of hemodialysis (data not shown).
Baseline CETP activities and CETP activation during hemodialysis Baseline CETP activities tended to be increased in the HTG subjects, but this was not statistically significant (Table 1). Shortly after the bolus heparinization at the beginning of the hemodialysis procedure (1.5 hr), CETP values were increased in the respective study groups in comparison to baseline (P , 0.001). Nevertheless, the extent of the peri-dialytical CETP activation was nearly twice as high in the HTG group when compared with the NTG subjects (P , 0.001) (Fig. 3). Furthermore, the scope of the CETP activation (DCETP; calculated as differences between peri-dialytical CETP activities and baseline CETP levels) was related (r 5 0.39, P , 0.04) to the extent of the peri-dialytical FFA increase (DFFA; calculated as differences between postheparinic FFA concentrations and baseline FFA levels). Until the end of hemodialysis, CETP activities decreased in both study groups but remained elevated when compared with baseline levels (P , 0.01, respectively; Fig. 3). Furthermore, a multiple regression analysis was performed to determine a possible effect of CETP activities on the LDL composition and diameter. No correlation was found between CETP and LDL-TG (0.26, NS) or LDL-CE proportion (r 5 20.26, NS), or the LDL diameter (r 5 20.16, NS). Cellular cholesteryl ester formation in fibroblasts and murine macrophages P388 induced by LDL Baseline LDL derived from HTG subjects induced the esterification rate in fibroblasts in a concentration-dependent, saturable manner but to a minor (60%) degree (with 50 mg lipoprotein) compared to baseline LDL levels from NTG individuals (P , 0.05; Fig. 4A). On the other hand, twice as many cholesteryl esters were formed in macrophages after preincubation with baseline LDL (50 mg) from HTG subjects when compared with LDL from NTG individuals (P , 0.04; Fig. 4B). The esterification rates in macrophages were unsaturable up to LDL-protein concentrations of 100 mg (Fig. 4B). When fibroblasts and macrophages were preincubated with LDL preparations derived during hemodialysis from hypertriglyceridemic patients, intracellular [3H]-cholesteryl
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
613
Fig. 1. Inverse correlations between the size of the major low-density lipoprotein (LDL) subfractions (nm) and the triglyceride (TG) content per particle (A; r 5 0.42, P < 0.02) and the relationship between LDL size and unesterified cholesterol (FC) per apoB ratios (B; r 5 0.40, P < 0.04) in normo- and hypertriglyceridemic patients. (C) Inverse correlation between cholesteryl ester content per particle and the triglyceride content in LDL within normo- (NTG; Œ) and hypertriglyceridemic (HTG; M) study participants (r 5 20.37, P , 0.05). 4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™
Fig. 2. Changes in the triglyceride content per apoB during hemodialysis (expressed as DLDL-TG/apoB) are positively related to the changes in the LDL diameter (expressed as DLDL-diameter). Symbols are: (M) hypertriglyceridemic and (Œ) normoglyceridemic subjects. r 5 0.38, P , 0.05.
Fig. 3. CETP activities before, during (1.5 hr after the bolus heparinization at the beginning of hemodialysis) and after hemodialysis therapy in normo- (NTG; Œ) and hypertriglyceridemic (HTG; M) patients. Each time point represents the median values with minimum and maximum. #P , 0.001, NTG vs. HTG group; *P , 0.01 and **P , 0.001 vs. baseline levels within each study group.
614
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
Fig. 4. Cellular [3H]-oleate incorporation into cholesteryl ester in fibroblasts (A) and murine macrophages P388 (B) induced by LDL prepared from NTG and HTG patients before [HTG-LDL (f) and NTG-LDL () and during hemodialysis [HTG-LDL (hemodialysis) (M) and NTG-LDL (hemodialysis) (‚)]. #P , 0.05, between NTG-LDL and the corresponding HTG-LDL concentration; *P , 0.05, between hemodialysis-modified LDL and baseline LDL within each group and LDL concentration. Values are mean 6 SD.
ester formation was significantly increased in comparison to the baseline LDL (P , 0.05, respectively, in Fig. 4 A and B). Nevertheless, esterification rates in fibroblasts did not reach the values obtained with LDL from normolipidemic patients. No differences in the esterification rates were observed when cells were preincubated with normotriglyceridemic LDL derived during hemodialysis or at baseline (Fig. 4). DISCUSSION This study demonstrates that triglyceride-enriched and cholesterol-depleted LDL from hypertriglyceridemic patients with ESRD leads to an increased and unsaturable cholesterol esterification rate in macrophages and a decreased rate in fibroblasts when compared with LDL from normolipidemic patients. In addition, modification of LDL lipids during hemodialysis obviously due to a lipase-mediated triglyceride hydrolysis and not to an increased CETP activity enforces these effects particularly in macrophages. An atherogenic aspect of hypertriglyceridemia in ESRD is the prevalence of small and dense LDL fractions known to be associated with marked changes in the LDL composition [16, 29 –31]. In our hypertriglyceridemic patients, decreasing LDL diameters were related to decreasing amounts of unesterified cholesterol and an exchange between cholesteryl esters and triglycerides, thus proposing a role of lipid transfer in LDL size reduction. Since accelerated CETP activity might explain the exchange between triglycerides and cholesteryl esters in LDL, Deckelbaum and collegues proposed an in vitro model for the size reduction of isolated LDL involving lipid transfer [32].
Indeed, the baseline CETP activity in our hypertriglyceridemic patients tends to be increased but differences did not reach statistical significance. In addition, CETP did not correlate with LDL components or particle diameter. This finding is consistent with results of several other investigators who have also failed to demonstrate a significant correlation between CETP activity and LDL particle diameter [33–35]. In this context, it must be taken into consideration that alterations of the cholesterol ester transfer (CET) are not only determined by CETP levels, but also by the quantity and the quality of the interacting lipoprotein particles. There is evidence that both an increase in triglyceride-rich lipoproteins and compositional changes in VLDL or HDL affected their participation in CET and potentially drives a higher exchange with LDL at the same CETP level [36 –39]. In this context, it is noteworthy that our HTG group had increasing triglycerides in HDL and an increase of cholesterol in VLDL. Both compositional changes might not only be a result of an accelerated cholesterol ester transfer, but might also trigger their process [38, 39]. Furthermore, the increase of cholesterol in the VLDL-fraction additionally suggests a large amount of remnant particles in the plasma of our HTG patients. Although the VLDL-CH levels in our patients are quite high in comparison to other studies, there is evidence that HTG patients with ESRD have a severe defect in clearing cholesterol and triglyceride enriched remnants [9, 14, 40 – 42]. In this context, Weintraub and colleagues described a threefold accumulation of remnants after more than 10 hours postprandially in HTG patients with ESRD in comparison to dyslipidemic patients without renal failure [42].
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
One hypothesis concerning an active role of small LDL particles in the atherogenesis came from observations about their increased susceptibility to in vitro oxidation. The increased susceptibility is obviously related to the LDL composition and might be explained in part by the lower antioxidant content in small dense LDL [42– 44]. Nevertheless, only a few studies investigated the oxidizability of LDL in patients with renal failure, and the results were somewhat contradictory [45– 47]. Another hypothesis about the potential atherogenicity of hypertriglyceridemic LDL might be gained from our cell culture experiments. In fact, triglyceride-enriched LDL enhanced the cholesteryl ester formation in cultured fibroblasts to a lesser degree than LDL from normolipidemic subjects. Since apoB-containing lipoproteins are taken up in fibroblasts by the classical LDL receptor, decreased intracellular cholesteryl ester formation might be caused by a decrease in the receptor affinity. Such a circumstance might be attributed to the higher triglyceride content in LDL altering the conformation of apoB [36, 48, 49], or to an oxidative modified recognition site. On the other hand, the unsaturable esterification rates in murine macrophages– our in vitro model for the nonspecific receptor pathway– demonstrated that LDL from HTG subjects enhanced the uptake more likely than LDL from NTG patients. From all these facts, it could be proposed that a decreased uptake of hypertriglyceridemic LDL by the specific pathway in vivo leads to a prolonged clearance of these particles, followed by an intracellular deposition of LDL-cholesterol in macrophages [50, 51]. In comparison to baseline levels, the lipid profile in patients with ESRD was adversely affected during hemodialysis obviously due to the peri-dialytical anticoagulation [20]. Particularly in hypertriglyceridemic patients, a shortterm decrease in triglyceride rich lipoproteins and a corresponding excess of FFA concentrations illustrated a postheparin activation of the lipolytic system. Besides these effects, an increase in the total LDL cholesterol concentrations and a significant corresponding decrease in the LDL particle diameter have been observed and recently described [20]. The present data suggest that the decrease in LDL diameters during hemodialysis is caused by a substantial reduction of the core triglycerides (Fig. 2). These observations are consistent with findings that lipoprotein lipase (LPL) as well as the hepatic lipase (HL) induce alterations in the LDL composition that are characterized by a reduction in core triglycerides [37, 45, 52], and a conversion towards subpopulations of smaller size [53, 54]. On the other hand, Lagrost, Gambert and Lallement [55] demonstrated in vitro that not lipolysis alone, but sequential effects of lipolysis and lipid transfer reactions induced dramatic changes in the mean size of LDL. In our study, CETP activity increased during hemodialysis with a peak at 1.5 hours after the beginning of anticoagulation and lipid
615
ingestion. By these facts, it is difficult to determine whether the increase in CETP activity is the result of postprandial changes or the effect of hemodialysis procedure. We prefere the latter interpretation, since several studies in patients without ESRD usually demonstrated a postprandial CETP peak in correlation to the remnant peak four to six hours after lipid ingestion [48, 56]. Therefore, our observations of an early CETP peak more likely illustrated an effect of hemodialysis than a “physiological” time course. It might be possible that the increase in CETP activity during the early phase of hemodialysis is triggered by the postheparinic excess of FFA. This interpretation is supported by recent findings that triglyceride-lipolysis in general enhanced the catalytic efficiency of CETP between triglyceride-rich and cholesterol-rich lipoproteins [50, 57, 58]. These observations may be result of an increase in lipid-bound FFA that influenced the binding properties of CETP to lipoproteins, resulting in an accelerated transfer rate. In addition, the increased presence of lipids available for the exchange with cholesterol esters during lipolysis may promote the formation of shuttle complexes containing CETP [57]. However, the physiological significance of the short-term CETP activation during hemodialysis on the LDL size pattern remains uncertain, since the LDL cholesterol content was unaltered and the depletion of triglycerides was due to the hydrolytic activity. From these observations, we propose that compositional changes of LDL during hemodialysis may also affect the properties of the lipoprotein surface, thereby modifying the interaction of the lipoprotein with the receptor pathway. Indeed, LDL derived from HTG patients during hemodialysis led to an accelerated cholesteryl ester formation in both fibroblasts and murine macrophages. These findings are of physiological significance in several respects: since the LDL-receptor in fibroblasts is regulated by the intracellular cholesterol content, an accelerated uptake of LDL usually results in a down-regulation of the receptor expression [51]. An increase in the number of circulating LDL in vivo may be the respective result and contributes to the observed increase of LDL-cholesterol [20] during hemodialysis. Since the scavenger receptor activity in macrophages is not regulated by the cell cholesterol content, increased uptake can lead to an unsaturable intracellular accumulation of cholesterol and foam cell generation [50, 51]. In conclusion, we believe that small and triglycerideenriched LDL particles derived from dyslipidemic patients with ESRD are an inferior substrate for the specific pathway in fibroblasts and a superior substrate for the nonspecific pathway in macrophages. In addition, modifications of the core triglycerides of LDL during hemodialysis may provide an alternative means of causing macrophages to take up and accumulate cholesterol. Since these effects potentially occur in vivo, our data might contribute to the
616
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
understanding of the high prevalence of premature atherosclerosis in hypertriglyceridemic patients with ESRD on the maintenance of hemodialysis. Reprint requests to Dr. Andreas Ambrosch, Institute of Clinical Chemistry, University Hospital of Magdeburg, Leipziger Str. 44, 39120 Magdeburg, Germany. E-mail: [email protected]
APPENDIX Abbreviations used in this article are: apo, apolipoprotein; BMI, body mass index; CE, esterified cholesterol; CETP, cholesteryl ester transfer protein; CH, cholesterol; ESRD, end-stage renal disease; FC, unesterified cholesterol; FFA, free fatty acids; HDL, high density lipoprotein; HL, hepatic lipase; HTG, hyperlipidemic patients; LDL, low density lipoprotein; LPL, lipoprotein lipase; NTG, normolipidemic patients; PAGGE, nondenaturating polyacrylamide gradient gel electrophoresis; TG, triglyceride; VLDL, very low density lipoprotein.
REFERENCES 1. KJELLSTRAND CM, HYLANDER B, COLLINS A: Mortality on dialysis. On the influence of early start, patients characteristics and transplantation and acceptance rates. Am J Kidney Dis 15:483– 490, 1990 2. GOKAL R, KING J, BOGLE S, MARSH F, OLIVER D, JAKUBOWSKI C, HUNT L, BAILLOD R, OGG C, WARD M, WILKINSON R: Outcome in patients on continuous ambulatory peritoneal dialysis and hemodialysis: 4 years analysis of a prospective multicentric study. Lancet 2:1105–1109, 1987 3. UNITED STATES RENAL DATA SYSTEM: Annual data report 1990. Am J Kidney Dis 16(Suppl):S44 –S52, 1990 4. World Health Annual Statistics. Geneva, Switzerland, World Health Organization, 1990, pp 166 –169 5. MA KW, GREEN EL, RAIJ L: Cardiovascular risk factors in the chronic renal failure and hemodialysis population. Am J Kidney Dis 19:505– 513, 1992 6. BRUNZELL JD, ALBERS JJ, HAAS LB, GOLDBERG AP, AGADOA L, SHERRARD DJ: Prevalence of serum lipid abnormalities in chronic hemodialysis. Metabolism 26:903–910, 1977 7. BAGDADE JD, ALBERS JJ: Plasma high density lipoprotein concentrations in chronic hemodialysis and renal transplant patients. N Engl J Med 296:1436 –1449, 1977 8. HAAS LB, WAHL PW, SHERRARD DJ: A longitudinal study of lipid abnormalities in renal failure. Nephron 33:145–159, 1983 9. CAMEJO G, RIERA G, LEE M, LOPEZ F: Lipoprotein structural abnormalities in chronic failure with and without hemodialysis. BBA 47:239 –245, 1988 10. RUBIES-PRAT J, ESPINEL E, JOVEN J, RAS MR, PIRA L: High density lipoprotein subfractions in chronic renal failure. Am J Kidney Dis 9:60 – 65, 1987 11. SENTI M, ROMERO R, PEDRO-BOTET J, PELEGRI A, NOGUES X, RUBIED-PRAT J: Lipoprotein abnormalities in hyperlipidemic and normolipidemic men on hemodialysis with chronic renal failure. Kidney Int 41:1394 –1399, 1992 12. CHAN MK, PERSAUD J, VARGHESE Z, MOORHEAD JF: Pathogenic roles of postheparin lipases in lipid abnormalities in hemodialysis patients. Kidney Int 25:812– 818, 1984 13. LINDHOLM B, NORBECK HE: Serum lipids and lipoproteins during continuous ambulatory peritoneal dialysis. Acta Med Scand 220:143– 151, 1986 14. NESTEL PJ, FIDGE NH, TAN MH: Increased lipoprotein remnant formation in chronic renal failure. N Engl J Med 307:329 –333, 1982 15. RON D, OREN I, AVIRAM M, BELTER DS, BROOK JG: Accumulation of lipoprotein remnants in patients with chronic renal failure. Atherosclerosis 46:67–75, 1983 16. MCNEIL D, LEE P, MURPHY B, BEST J: Low density lipoprotein particle size distribution in end-stage renal disease treated with hemodialysis or peritoneal dialysis. Am J Kidney Dis 27:84 –91, 1996 17. ATTMAN PO, ALAUPOVIC P, GUSTAVSON A: Serum apolipoprotein profile of patients with chronic renal failure. Kidney Int 32:368 –375, 1987
18. BERGESIO F, MONZANI G, CIUTI R, SERRUTO A, BENUCCI A, FRIZZI V, SALVADORI M: Lipids and apolipoprotein changes during the progression of chronic renal failure. Clin Nephrol 38:264 –270, 1992 19. CORESH J, KWITEROVICH PO, SMITH HH, BACHORIC PS: Association of TG concentrations, LDL particle diameter, density and chemical composition with premature coronary heart disease in men and women. J Lipid Res 34:1687–1697, 1993 20. LULEY C, NEUMANN KH, WESTPHAL S, AMBROSCH A: Atherogenic effects of hemodialysis: Increased generation of small and cholesterol poor LDL particles. (abstract) Atherosclerosis 129:141, 1997 21. ORTH M, WAHL S, HANISCH M, FRIEDRICH I, WIELAND H, LULEY C: Clearance of postprandial lipoproteins in normolipidemics: Role of the apolipoprotein E phenotype. BBA 1303:22–30, 1996 22. RAINWATER DL, ANDRES DW, FORD AL, LOWE F, BLANCHE PJ, KRAUSS RM: Production of polyacrylamide gradient gels for electrophoretic resolution of lipoproteins. J Lipid Res 33:1876 –1881, 1992 23. WILLIAMS PT, KRAUSS RM, NICHOLS AV, VRANIZAN KM, WOODS PDS: Identifying the predominant peak diameter of high-density and low-density lipoproteins by electrophoresis. J Lipid Res 31:1131–1139, 1990 24. TATO F, VEGA GL, GRUNDY SM: Determinants of plasma HDLcholesterol in hypertriglyceridemic patients. Arterioscler Thromb Vasc Biol 17:56 – 63, 1997 25. AMBROSCH A, TATO F, LULEY C: Fluorescent determination of CETP activity in plasma. (abstract) Atherosclerosis 134:100, 1997 26. BROWN MS, HO YK, GOLDSTEIN JL: Cholesteryl ester cycle in macrophages. J Biol Chem 255:9344 –9352, 1980 27. LOWRY OH, ROSEBROUGH NJ, FARR AL, RANDALL AJ: Protein measurement with the Folin-phenol reagent. J Biol Chem 193:265– 275, 1951 28. KRAUSS RM, BURKE DJ: Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res 23:236 –244, 1982 29. STEWART MW, LAKER MF, DYER RG, GAME F, MITCHESON J, WINCOUR PH, ALBERT KGM: Lipoprotein compositional abnormalities and insulin resistance in type II diabetic patients with mild hypertlipidemia. Arterioscler Thromb 12:1046 –1052, 1993 30. HOKANSON JE, KRAUSS RM, ALBERS JJ, AUSTIN MA, BRUNZELL JD: LDL physical and chemical properties in familial combined hyperlipidemia. Arterioscler Throm Vasc Biol 15:452– 459, 1995 31. CAPELL WH, ZAMBON A, AUSTIN MA, BRUNZELL JD, HOKANSON JE: Compositional differences of LDL particles in normal subjects with LDL subclass phenotype A and LDL subclass phenotype B. Arterioscler Thromb Vasc Biol 16:1040 –1046, 1996 32. DECKELBAUM RJ, EISENBERG S, FAINARU M, BARENHOLZ Y, OLIVECRANO T: In-vitro production of human low density lipoprotein-like particles: A model for very low density lipoproteoin catabolism. J Biol Chem 254:6079 – 6087, 1979 33. KARPE F, TORNVALL T, OLIVECRANO T, STEINER G, CARSON LA, HAMSTON A: Composition of human low-density lipoprotein: Effect of postprandial triglyceride-rich lipoproteins, lipoproteine lipase, hepatic lipase and cholesterol ester transfer protein. (abstract) Atherosclerosis 98:33, 1993 34. TAN KCB, COOPER MB, LING KLE, GRIFFIN BA, FREEMAN DJ, PACKERD CJ, SHEPHERD J, HALES CN, BATTERIDGE DJ: Fasting and postprandial determinants for the occurrence of small dense LDL species in NIDDM patients with and without hypertriglyceridemia: The involvement of insulin, insulin precursor species and insulin resistance. (abstract) Atherosclerosis 113:275, 1995 35. WATSON TDG, CASLAKE DJ, FREEMAN BA, GRIFFIN BA, HINNIE J, PACKARD CJ, SHEPHERD J: Determinants of LDL subfraction distribution and concentration in young normolipidemic subjects. (abstract) Arterioscler Thromb 14:902, 1994 36. AVIRAM M, BIERMAN EL, CHAIT A: Modification of low density lipoprotein by lipoprotein lipase or hepatic lipase induces enhanced uptake of cholesterol accumulation in cell. J Biol Chem 263:15416 – 15422, 1988 37. DECKELBAUM RJ, EISENBERG S, OSCHRY Y, BUTBUL E, SHARON I, OLIVECRONA T: Reversible modification of human plasma low density lipoproteins towards triglyceride-rich precursors: A mechanismn for losing excess cholesterol esters. J Biol Chem 257:6509 – 6517, 1982 38. BAGDADE JD, RITTER MC, SUBBAIAH PV: Accelerated cholesteryl
Ambrosch et al: Hypertriglyceridemic ESRD patients and LDL
39. 40. 41. 42.
43.
44. 45. 46.
47.
48.
ester transfer in patients with insulin-dependent diabetes mellitus. Eur J Clin Invest 21:161–167, 1991 EISENBERG S: Preferential enrichment of large-sized very low density lipoprotein populations with transferred cholesteryl esters. J Lipid Res 26:487– 494, 1985 WANNER C, FROMMHERZ K, HORL WH: Hyperlipoproteinemia in chronic renal failure: Pathophysiological and therapeutical aspects. Cardiology 78:202–217, 1991 CHEUNG A, WU LL, KABLITZ C, LEYPOLDT K: Atherogenic lipids and lipoproteins in hemodialysis patients. Am J Kidney Dis 22:271–276, 1993 WEINTRAUB M, BURSTEIN A, RASSIN A, LIRON M, RINGEL Y, CABILI S, BLUM M, PEER G, IAINA A: Severe defect in clearing postprandial chylomicron remnants in dialysis patients. Kidney Int 42:1247–1252, 1992 DEGRAAF J, HAK-LEMMERS HL, HECTORS PC, DEMARKER PNM, HENDRIKS JCM, STALENHOEF AFH: Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects. Arterioscler Thromb 11:298 –306, 1991 TRIBBLE DL, HOLL LG, WOOD PD, KRAUSS RM: Variation of oxidative susceptibility among six low density lipoprotein subfractions of varying size and density. Atherosclerosis 93:189 –199, 1992 SUTHERLAND W, WALKER RJ, BALL MJ, STAPLEY S, ROBERTSON MC: Oxidation of low density lipoproteins from patients with renal failure or renal transplants. Kidney Int 48:227–236, 1995 LOUGHREY CM, YOUNG IS, MCENENY J, MCDOWELL FW, MCMASTER C, MCNAMEE PT, TRIMBLE ER: Oxidation of low density lipoprotein in patients on regular hemodialysis. Atherosclerosis 110:185–193, 1994 MAGGI E, BELLAZZI R, FALASCHI F, FRATTONI A, PERANI G, FINARDI G, GAZO A, NAI M, ROMANINI D, BELLOMO G: Enhanced LDL oxidation in uremic patients: An additional mechanism for accelerated atherosclerosis. Kidney Int 45:876 – 8883, 1994 AGNANI G, BARD JM, CANDELIER L, DELATTRE S, FRUCHARD JC,
49. 50. 51. 52. 53. 54.
55. 56.
57.
58.
617
CLAVEY V: Interaction of LpB, Lpb:E, LpB:CIII, LpB:CIII:E lipoproteins with the low density lipoprotein receptor of HeLa cells. Arterioscler Thromb 11:1021–1029, 1991 NICOLL A, LEWIS B: Evaluation of the roles of lipoprotein lipase and hepatic lipase in lipoprotein matabolism: In-vivo and in-vitro studies in men. Eur J Clin Invest 10:487– 495, 1980 ROSS RN: The pathogenesis of atherosclerosis — An update. N Engl J Med 314:488 –500, 1986 BROWN MS, GOLDSTEIN JL: The low density lipoprotein pathway and is relation to atherosclerosis. Annu Rev Biochem 52:223–261, 1983 MUSLINER TA, HERBERT PN, KINGSTON NJ: Lipoprotein substrates of lipoprotein lipase and hepatic triglycerol lipase from human postheparin plasma. BBA 575:277–288, 1979 HOMMA, Y, NAKAYA N, NAKAMURA H, GOTO Y: Increase in the density of lighter low density lipoprotein by hepatic triglyceride lipase. Artery 13:19 –31, 1985 LAGROST L, FLORENTIN E, GUYARD-DANGREMONT V, ATHIAS A, GANDJINI H, LALLEMANT C, GAMBERT P: Evidence for nonesterified fatty acids as modulators of neutral lipid transfer in normolipidemic human plasma. Arterioscler Thromb Vasc Biol 15:1388 –1396, 1995 LAGROST L, GAMBERT P, LALLEMENT C: Combined effects of lipid transfer and lipolysis on gradient gel patterns of human LDL. Arterioscl Thromb 14:1327–1336, 1994 KARPE F, TORNVALL P, OLIVECRONA T, STEINER G, CARLSON LA, HAMSTEN A: Composition of human low density lipoprotein: Effect of postprandial triglyceride-rich lipoproteins, lipoprotein lipase, hepatic lipase and cholesteryl ester transfer protein. Atherosclerosis 98:33– 49, 1993 TALL AR, SAMMET D, VITA G, DECKELBAUM RJ, OLIVECRONA T: Lipoprotein lipase enhances the cholesteryl ester transfer proteinmediated transfer of cholesteryl esters from high density lipoproteins to very low density lipoproteins. J Biol Chem 260:6687– 6697, 1985 SAMMET D, TALL AR: Mechanismns of enhancement of cholesteryl ester transfer protein activity by lipolysis. J Biol Chem 260:6687– 6697