Lipoprotein Lipase Bound to Apolipoprotein B Lipoproteins Accelerates Clearance of Postprandial Lipoproteins in Humans
Objectives— Experiments in cells and animal models show that lipoprotein lipase (LpL) bound to apolipoprotein (apo)B lipoproteins enhances their uptake by receptor mediated pathways. It is unknown whether this pathway is important in humans.
Methods and Results— ApoB lipoproteins with LpL were isolated from normal subjects after oral fat loading by immunoaffinity chromatography and were further separated into apoB100 and apoB48 lipoproteins. Postprandially, apoB lipoproteins with LpL had significantly greater increases (4- to 10-fold) and faster rates of clearance (5- to 8-fold) percentage-wise than those without LpL. apoB lipoproteins with LpL had enhanced clearance regardless of whether they also contained apoE. LpL was particularly important for the clearance of apoB48 lipoproteins, of which 25% (range, 11% to 31%) could be removed from circulation together with LpL during the postprandial state. apoB lipoproteins with LpL were larger in size and were enriched in triglyceride, cholesterol, and apoE compared with those without LpL. However, neither size nor apoE content explained the faster clearance rates of LpL-containing lipoproteins.
Conclusion— Plasma LpL may act like an apolipoprotein to enhance the clearance of apoB lipoproteins in humans, a mechanism particularly important for intestinal lipoproteins in the postprandial state.
Lipoprotein lipase (LpL) is the key enzyme that hydrolyzes triglyceride (TG) in TG-rich lipoprotein (TRL) such as chylomicrons and very-low-density lipoprotein (VLDL), thereby regulating the uptake of free fatty acids by peripheral tissues, especially adipose tissue, skeletal muscle, and cardiac muscle.1,2
After being synthesized in parenchymal cells of these tissues, LpL is transported to the luminal surface of capillary vascular endothelium and attaches to heparan sulfate side chains of proteoglycans (HSPG), where it binds to TRL and hydrolyzes its TG content.3 Growing evidence suggests that LpL may also be involved in further metabolism of TRL remnants aside from its lipolytic function. A low concentration of LpL has been detected in human blood, and it is mainly associated with apoB lipoproteins.4,5 As early as 1975, Felts et al proposed that LpL could bind to lipoproteins in vivo and might serve as a marker for remnant uptake by the liver.6
In cell culture experiments, LpL is able to enhance the uptake and degradation of chylomicrons, VLDL, and their remnants. Beisiegel et al reported that the addition of LpL increased the cellular binding of chylomicrons to HepG2 cells by up to 30-fold and also increased the cellular binding of β-VLDL to low-density lipoprotein (LDL) receptor-negative fibroblasts up to 40-fold, through either directly binding to LDL-receptor related protein (LRP) or enhancing the binding of apolipoprotein (apo)E to LRP.7 Interestingly, the lipolytic function of LpL is not required in LpL-mediated clearance because the enhanced uptake of LpL-associated lipoproteins is not affected by inhibitors of its lipolytic activity, and a carboxyl-terminal fragment of LpL, which lacks the catalytic domain, is also able to enhance the binding and uptake of lipoproteins by LDL receptor-negative fibroblasts in a manner that mimics the native enzyme.8–10 In addition, Skottova et al discovered that catalytically inactive LpL added to the perfusate significantly increased the clearance of chylomicrons by perfused rat liver.11
However, there is only limited information on the importance of LpL-mediated clearance of apoB lipoproteins in vivo. Transgenic mice that overexpressed catalytically inactive LpL in muscle have decreased plasma TG concentrations,12 and injected human TRL are removed by the liver to a significantly larger extent than in wild-type mice,13 suggesting LpL may be able to enhance the clearance of lipoprotein from circulation in vivo in mice. If this pathway is also present in humans, it may have important physiological implications because plasma LpL was found mainly on triglyceride-rich lipoproteins, including VLDL and chylomicrons.
The current project aims to extend the findings from in vitro experiments and animal studies to humans to determine whether apoB lipoproteins with LpL are cleared from circulation faster than those without LpL, and whether they account for an important amount of postprandial lipoprotein flux.
A detailed description of methods and materials is available online at http://atvb.ahajournals.org.
Nine healthy volunteers (6 men and 3 women) between 25 and 38 years of age completed the study. Blood was first drawn after an overnight fast (baseline). Participants then ingested a fat load in the form of a milkshake (1339 kcal with 70% of energy from fat) and had blood drawn every hour between 3 and 8 hours after oral fat-loading. Tetrahydrolipstatin (THL), an active inhibitor of LpL, was immediately added to inhibit further hydrolysis occurring in vitro after blood sampling.
Sequential immuno-affinity chromatography, developed from a procedure previously described and validated,14,15 was used to separate lipoproteins according to their contents of LpL, apoB, or apoE. Plasma samples were first applied to Sepharose 4B resin coupled with a monoclonal antibody against LpL (5D2 MAb4). The fraction without LpL (LpL unbound, or LpL−) was collected by gravity flow, and the fraction with LpL (LpL bound, or LpL+) was collected by briefly incubating the resin with 3 mol/L NaSCN solution. LpL+ and LpL− fractions were then immediately applied to anti-apoB100 or anti-apoE immuno-affinity chromatography to further separate apoB100+/− and apoE+/− lipoprotein fractions. Anti-apoB100 resin was made from a monoclonal antibody that recognizes apoB100 but not apoB48, and anti-apoE resin was made from an affinity purified polyclonal antibody. Efficiency of binding for anti-LpL, anti-apoB100 and anti-apoE resin was all excellent (>95%). Western blotting against LpL, apoB100, and apoE did not detect their presence in respective unbound fractions. Recovery rates of lipids (triglyceride and cholesterol) and apolipoproteins (apoB, apoE, apoCI, apoCII, and apoCIII) were >90% for anti-LpL columns and were >80% for anti-apoB100 and anti-apoE columns. Total recovery rate of lipids and apolipoproteins after sequential columns (anti-LpL followed by anti-apoB100 or anti-apoE) were between 65% and 85%. The recovery rate of LpL dimers was between 55% and 80% for the anti-LpL column.
In certain experiments, fractions from immuno-affinity chromatography were further separated by ultracentrifugation into TRL (density <1.025 g/mL) and into LDL (1.025 g/mL < density <1.063 g/mL).
Concentrations of apolipoproteins were determined by sandwich enzyme-linked immunosorbent assay (ELISA) using affinity purified antibodies. LpL dimer concentrations were measured by a monoclonal antibody based ELISA.4 All ELISAs were highly sensitive with signal-to-background absorbance ratios of 40 and higher. Intra-assay coefficients of variances for lipid and apolipoprotein measurements were between 2% and 6%, and interassay coefficients of variances were between 4% and 10%.
Transmission electron microscopy experiments were performed on 3 subjects’ fasting plasma to evaluate particle size distribution of apoB lipoproteins with and without LpL.
All data presented here have been corrected for loss during immuno-affinity chromatography and/or ultracentrifugation. Postprandial clearance rates (mg/dL per hour), ie, the change in concentration per hour of a lipid or apolipoprotein after its postprandial peak, were calculated by fitting data in the SAS software. Percentage clearance rates (%/h) were calculated by dividing postprandial clearance rates of the lipid or apolipoprotein by respective postprandial peak concentrations. The difference between fractions with and without LpL was analyzed by paired t tests. P<0.05 was considered as statistically significant.
Participants of this study are all healthy, nonsmoking, nondiabetic, and not taking medications that may affect lipids, with mean age 30 years and mean fasting total plasma cholesterol 175 mg/dL and fasting TG 108 mg/dL (Table).
Electron microscopy results showed that apoB lipoproteins with LpL comprised a wide spectrum of particle sizes and that the majority was big triglyceride-rich particles. Particle size distributions of LpL+ and LpL− lipoproteins were studied in 3 subjects’ fasting plasma, and in subfractions of apoB lipoproteins in 2 of them. On average, the diameter of the apoB lipoproteins with LpL was 47±17 nm, which was >70% larger than those without LpL, 27±7 nm, (P<0.01). This is also the case in light VLDL (LpL+ versus LpL−: 62±11 versus 45±7 nm), dense VLDL (LpL+ versus LpL−: 43±10 versus 30±4 nm), and intermediate-density lipoprotein (IDL) (LpL+ versus LpL−: 33±5 versus 24±4 nm) fractions (all P<0.05; Figure 1). Thus, the average diameter of LpL bound lipoproteins was 40% to 75% larger than apoB lipoproteins without LpL. Assuming a spherical shape of apoB lipoproteins, this corresponds to a 100% to 200% larger surface area and 170% to 440% bigger volume of LpL+ compared with LpL− particles. Fast protein liquid and density gradient ultracentrifugation conducted on one subject’s plasma also confirmed a significant percentage of LpL bound lipoproteins was VLDL-sized particles and the majority of triglyceride in LpL+ lipoproteins was eluted with VLDL (Figure I, available online at http://atvb.ahajournals.org). TRL with LpL were enriched in apoE compared with apoB lipoproteins without LpL (14 versus 1.7 molecules/particle; P<0.01) but were similar in contents of apoC (apoCI 11 versus 7; apoCII 18 versus 22; apoCIII 17 versus 22 molecules/particle for LpL+ versus LpL−; P=NS).
Four hours after oral fat-loading, the concentration of TG in TRL with LpL increased &5-fold, from 0.9 to 5.1 mg/dL (P<0.05), whereas its apoB concentration increased & 3-fold, from 0.01 to 0.03 mg/dL (P<0.05) (Figure 2). Corresponding postprandial increments percentage-wise in TRL without LpL were significantly less, for TG from 80 mg/dL to 180 mg/dL, and for apoB from 10 mg/dL to 13 mg/dL. The concentrations of TG and apoB in LDL with LpL increased 4-fold and 5-fold after fat ingestion, respectively. In contrast, the concentrations of TG and apoB in LDL without LpL changed little. Interestingly, LDL with LpL exhibited a postprandial response pattern more like TRL than typical LDL. In fact, LDL with LpL had significantly greater percentage increases in TG and apoB than TRL without LpL (TG: 4-fold versus 2-fold; apoB: 5-fold versus 30%; both P<0.05).
Two hours after the concentrations of apoB and TG in TRL with LpL peaked, they returned to around the fasting level (Figure 2). During these 2 hours, 23% of the peak level of apoB and 75% of the peak level of TG in TRL with LpL disappeared from circulation. In contrast, apoB and TG concentrations in TRL without LpL decreased much more slowly. Only 5% and 28% of peak levels of apoB and TG disappeared per hour and it took 3 to 5 hours after the peaks for the concentrations to reach near-fasting levels (P<0.05 between LpL+ and LpL−). ApoB in LDL with LpL also disappeared rapidly from circulation, with 75% of peak concentration disappearing within 1 hour, whereas apoB in LDL without LpL changed little.
LpL was found on both apoB48 and apoB100 lipoproteins, and both had significantly bigger postprandial increments percentage-wise than their LpL free counterparts (Figure 3). The concentration of apoB48 lipoproteins with LpL increased 10-fold, and TG 6-fold postprandially, significantly higher than the respective 2.4-fold and 2-fold increments of apoB and TG in apoB48 lipoproteins without LpL (P<0.05 for both). apoB100 lipoproteins with LpL also significantly increased their apoB and TG concentrations more than those without LpL (LpL+ versus LpL−: apoB100, 4.5-fold versus 16%; TG, 4-fold versus 2-fold; P<0.05).
The bulk of apoB48 lipoproteins with LpL (79%) disappeared from circulation within one hour after their concentration peaked postprandially (Figure 3). In contrast, apoB48 lipoproteins without LpL were cleared much more slowly: three hours after their peak, the plasma concentration declined only 25%. The percentage clearance rate was 9-times greater for apoB48 lipoproteins with LpL than those without LpL (P<0.05). Similar to apoB48 lipoproteins, most of apoB100 lipoproteins with LpL (71%) disappeared from circulation within 1 hour after the peak, whereas the concentration of apoB100 lipoproteins without LpL changed little during the 8-hour postprandial period.
ApoB48 lipoproteins with LpL had a bigger postprandial response after fat ingestion than apoB100 lipoproteins with LpL. In the fasting state, the concentration of apoB48 lipoprotein with LpL (expressed in molar units: 0.12 nmol/L), was only approximately half of apoB100 lipoprotein with LpL (0.22 nmol/L). However, postprandially their molar concentrations became similar at their respective peak levels (1.05 nmol/L for apoB48 versus 1.13 nmol/L for apoB100), reflecting a bigger postprandial percentage increment of apoB48 than apoB100. In the fasting state, <1% of apoB48 lipoproteins had LpL, but it increased to >5% (range, 2% to 9%) in the postprandial state.
Electron microscopy experiments showed that on average apoB lipoproteins with LpL were bigger than those without LpL, and this disparity in particle size might contribute to the faster clearance of the former. Thus, we further isolated TRL into light VLDL, dense VLDL, and IDL (Figure 1), and studied the postprandial response between LpL+ and LpL− lipoproteins of similar size distributions. This would help address the potential confounding effect of particle size on postprandial response. Results showed that concentrations of LpL− light VLDL increased &44% (from 1.01 to 1.45 mg/dL) after oral fat-loading, which was significantly lower than the 3- and 5-fold increase observed in LpL+ TRL and LDL (Figure 4A and 4B). After its concentration peaked at 5 hours, LpL− light VLDL was removed at a percentage clearance rate of 9%/h, which was significantly lower than the respective 23%/h and 20%/h percentage clearances of LpL+ TRL and LDL. We further separated LpL− light VLDL into apoB100 and apoB48 lipoproteins. Results also showed that concentrations of apoB48 and apoB100 in LpL− light VLDL increased &90% (from 0.084 to 0.16 mg/dL) and 50% (from 0.96 to 1.46 mg/dL) postprandially, which were also significantly lower than the 10- and 5-fold increase observed in LpL+ apoB48 and apoB100 lipoproteins (Figure 4C and 4D). ApoB48 in LpL− light VLDL was gradually cleared after its concentration peaked at 3 hours with a percentage clearance rate of 10%/h, which was significantly slower than apoB48 in LpL+ apoB lipoproteins (79% peak concentration disappeared within 1 hour). Thirty-three percent apoB100 in LpL− light VLDL was cleared in 1 hour after it peaked at 5 hours after oral fat loading, a percentage clearance rate less than half of apoB100 lipoproteins with LpL (71%/h). Thus, lipoprotein size, per se, did not account for the fast metabolism of apoB lipoproteins with LpL.
ApoB lipoproteins with LpL have high contents of apoE. ApoE has a well-established role in enhancing lipoprotein removal through receptor-mediated pathways. Therefore, apoE could also be a confounding factor when we evaluate the independent contribution of LpL to the enhanced clearance of apoB lipoproteins from circulation. To distinguish the effect of LpL from that of apoE, we separated lipoproteins with and without LpL further according to their apoE content with anti-apoE immuno-affinity chromatography. We found that regardless of apoE, TRL with LpL decreased significantly faster than those without LpL. TRL that contained both LpL and apoE was removed &2.5-times as fast as TRL with apoE and not LpL (19%/h versus 7.5%/h; P<0.05) (Figure 5, upper panel). However, TRL that contained LpL and not apoE was cleared &2.3-times as fast as that of TRL with neither LpL nor apoE (27%/h versus 12%/h; P<0.05) (Figure 5, lower panel). Overall, LpL enhanced the clearance of triglyceride-rich lipoproteins by 2- to 3-fold, whether or not apoE coexisted with it.
It has been suggested that the binding between plasma LpL and apoB lipoproteins might not be as strong as other apolipoproteins such as apoE, mainly because of the observation that LpL dissociates from lipoproteins during ultracentrifugation in a high-gravity, high-salt environment.16 This raised the question whether a circulating LpL+ lipoprotein could become an LpL− lipoprotein by transferring LpL to a circulating LpL− lipoprotein. To investigate this, we labeled all protein contents of LpL+ lipoproteins with biotin, and incubated the biotin labeled LpL+ fraction with biotin free LpL− fraction at room temperature for 30 minutes. Then, the mixture was separated again into LpL+ and LpL− with anti-LpL immuno-affinity chromatography. If significant transfer of LpL from LpL+ lipoproteins to LpL− lipoprotein did happen during the incubation period creating new LpL− apoB lipoprotein particles, biotin-labeled apoB would then appear in the LpL− fraction after incubation caused by the loss of LpL. However, Western blotting against biotin with avidin–alkaline phosphatase conjugates did not show a biotin labeled apoB band in the LpL− lane (Figure II, available online at http://atvb.ahajournals.org), suggesting the transfer of LpL from LpL+ to LpL− is negligible under these conditions.
Lipids in the postprandial state may be more closely related to risk for coronary heart diseases than levels in the fasting state, and impaired clearance of postprandial lipoproteins may be an underlying risk factor for coronary heart diseases.17 As a pivotal regulator of postprandial lipid metabolism, LpL could accelerate TRL clearance by receptor mediated pathways in addition to its role as a lipase. The relevance of this apolipoprotein-like function in an in vivo metabolic setting has not been studied before and is the goal of this study. Our results strongly suggest that postprandially, apoB lipoproteins with LpL have significantly faster rates of clearance than those without LpL, regardless of the apoE content or particle size.
Plasma concentrations of LpL among the 9 participants are similar to what have been observed by others.1,4,18–20 The majority of apoB lipoproteins with LpL are large, indicating high core lipid contents. Although low in concentration, LpL+ lipoproteins carry a disproportionally higher percentage of plasma triglyceride in the postprandial state (average 4.5%; range 1% to 14%). In all types of apoB lipoproteins that we examined, those with LpL show a distinctive pattern of postprandial response with a sharper increase and faster decline than those without LpL, suggesting plasma LpL may be particularly important postprandially. In this regard, apoB48 lipoproteins with LpL have a 2-fold bigger postprandial increment percentage-wise than apoB100 with LpL, suggesting that LpL is more involved with the metabolism of intestinal than hepatic lipoproteins in the fed state. It has been proposed that chylomicrons and VLDL compete for the same hydrolysis pathway,21 and thus it is possible that higher postprandial response of chylomicrons may inhibit the binding of VLDL to LpL. More importantly, this study demonstrates that a physiologically significant percentage of apoB48 lipoproteins could be removed from circulation together with LpL. We estimate that 25% of chylomicrons and their remnants could be cleared by the LpL mechanism. The computation assumes that the production of chylomicron is low after its concentration peaks (4 to 5 hours after fat ingestion) postprandially and that the response of apoB48 to a single fatty meal is a close approximation to a biological tracer. However, although apoB100 lipoproteins with LpL were removed from circulation much faster than those without LpL, the overall effect of LpL on apoB100 clearance may be less important than on apoB48 because LpL is present on a smaller percentage of apoB100 lipoproteins.
The enrichment of apoE in apoB lipoproteins with LpL does not account for their enhanced clearance. We found that the presence of LpL in TRL is associated with fast clearance whether apoE was also present with LpL in the particles. The contents of C apolipoproteins, which could inhibit receptor mediated lipoprotein removal,22 were similar in apoB lipoproteins that do or do not contain LpL. Particle size also did not seem to be a major factor contributing to the faster removal of lipoproteins with LpL. When compared with the light VLDL without LpL, which have a similar particle size distribution, apoB lipoproteins with LpL still have greater postprandial response and faster clearance. Taken together, this study strongly suggests that LpL per se accelerates clearance of apoB lipoproteins in vivo in humans.
Our results also suggest that LpL on an apoB lipoprotein marks it for clearance by hepatic receptors, and that apoB lipoproteins with LpL are remnants of triglyceride-rich lipoproteins having undergone intravascular metabolism. ApoB lipoproteins with LpL are heterogeneous in size with 2 peaks on electron micrographs corresponding to typical chylomicron remnants and VLDL remnants.23 The key events causing the production of these types of remnants may first be lipolysis of nascent triglyceride-rich chylomicrons and VLDL, then uptake of LpL from the capillary endothelium by these lipoproteins, and finally release into the circulation. These events may constitute normal processing of chylomicrons and VLDL to form their specific remnant particles, which then are rapidly cleared by hepatic receptors.
The mechanisms that cause triglyceride-rich postprandial apoB lipoproteins undergoing lipolysis to remove LpL molecules from the capillary endothelium are unknown. This study suggests that enrichment of triglyceride, cholesterol, and apoE are possible contributors. Particle size may also be important. Chylomicrons compete with VLDL for LpL binding and triglyceride hydrolysis in vivo,21 and it has been shown that LpL binds larger particles better than smaller ones in vitro.24 It is unclear how much lipolytic processing occurred before the production of particles with LpL and their release into the circulation. Removal of LpL by a partially metabolized triglyceride-rich lipoprotein may also involve direct interaction between LpL and other lipid or protein components of the lipoprotein, such as apoA-V.25 Nevertheless, once an apoB lipoprotein re-enters circulation with LpL, it is rapidly cleared.
In conclusion, our study extends to humans the findings of cell culture experiments and animal studies, and demonstrates that a significant amount of postprandial apoB lipoproteins, especially chylomicrons, can be removed from circulation with the help of LpL. LpL is not only responsible for hydrolyzing apoB lipoproteins at the endothelium but also able to function like an apolipoprotein by facilitating receptor-mediated uptake during circulation. The apolipoprotein function of LpL may constitute a normal “low-concentration, high speed” pathway for clearance of triglyceride-rich lipoproteins.
This work was supported by grant HL069376 from the National Heart, Lung, and Blood Institute, National Institutes of Health. We gratefully acknowledge the advice of Drs Hannia Campos and Gokhan Hotamisligil. We express thanks to Lisa Martin for her excellent technical assistance with ELISA experiments, Steve Hashimoto for advice on LpL ELISA and assistance with fast protein liquid experiments, and Hoffman La Roche, Switzerland, for providing the tetrahydrolipstatin (Orlistat).
- Received August 4, 2005.
- Accepted December 23, 2005.
Brunzell JD, Deeb SS. Familial lipoprotein lipase deficiency, apo CII deficiency and hepatic lipase deficiency, in The Metabolic and Molecular Basis of Inherited Diseases. In: Sciver C, ed. New York: McGraw-Hill Book Co; 2000: 2789–2816.
Goldberg IJ. Lipoprotein lipase and lipolysis: central roles in lipoprotein metabolism and atherogenesis. J Lipid Res. 1996; 37: 693–707.
Pedersen ME, Cohen M, Schotz MC. Immunocytochemical localization of the functional fraction of lipoprotein lipase in the perfused heart. J Lipid Res. 1983; 24: 512–521.
Zambon A, Schmidt I, Beisiegel U, Brunzell JD. Dimeric lipoprotein lipase is bound to triglyceride-rich plasma lipoproteins. J Lipid Res. 1996; 37: 2394–2404.
Kern PA, Martin RA, Carty J, Goldberg IJ, Ong JM. Identification of lipoprotein lipase immunoreactive protein in pre- and postheparin plasma from normal subjects and patients with type I hyperlipoproteinemia. J Lipid Res. 1990; 31: 17–26.
Beisiegel U, Weber W, Bengtsson-Olivecrona G. Lipoprotein lipase enhances the binding of chylomicrons to low density lipoprotein receptor-related protein. Proc Natl Acad Sci U S A. 1991; 88: 8342–8346.
Chappell DA, Inoue I, Fry GL, Pladet MW, Iverius PH, Lalouel JM, Strickland DK. The carboxy-terminal domain of lipoprotein lipase induces cellular catabolism of normal very low density lipoproteins via the low density lipoprotein receptor-related protein/alpha 2-macroglobulin receptor. Ann N Y Acad Sci. 1994; 737: 434–438.
Chappell DA, Inoue I, Fry GL, Pladet MW, Bowen SL, Iverius PH, Lalouel JM, Strickland DK. Cellular catabolism of normal very low density lipoproteins via the low density lipoprotein receptor-related protein/alpha 2-macroglobulin receptor is induced by the C-terminal domain of lipoprotein lipase. J Biol Chem. 1994; 269: 18001–18006.
Nykjaer A, Nielsen M, Lookene A, Meyer N, Roigaard H, Etzerodt M, Beisiegel U, Olivecrona G, Gliemann J. A carboxyl-terminal fragment of lipoprotein lipase binds to the low density lipoprotein receptor-related protein and inhibits lipase-mediated uptake of lipoprotein in cells. J Biol Chem. 1994; 269: 31747–31755.
Skottova N, Savonen R, Lookene A, Hultin M, Olivecrona G. Lipoprotein lipase enhances removal of chylomicrons and chylomicron remnants by the perfused rat liver. J Lipid Res. 1995; 36: 1334–1344.
Merkel M, Kako Y, Radner H, Cho IS, Ramasamy R, Brunzell JD, Goldberg IJ, Breslow JL. Catalytically inactive lipoprotein lipase expression in muscle of transgenic mice increases very low density lipoprotein uptake: direct evidence that lipoprotein lipase bridging occurs in vivo. Proc Natl Acad Sci U S A. 1998; 95: 13841–13846.
Khoo C, Campos H, Judge H, Sacks FM. Effects of estrogenic oral contraceptives on the lipoprotein B particle system defined by apolipoproteins E and C-III content. J Lipid Res. 1999; 40: 202–212.
Campos H, Perlov D, Khoo C, Sacks FM. Distinct patterns of lipoproteins with apoB defined by presence of apoE or apoC-III in hypercholesterolemia and hypertriglyceridemia. J Lipid Res. 2001; 42: 1239–1249.
Vilella E, Joven J. Lipoprotein lipase binding to plasma lipoproteins. Med Sci Res. 1991; 19: 111–112.
Zilversmit DB. Atherogenesis: a postprandial phenomenon. Circulation. 1979; 60: 473–485.
Karpe F, Olivecrona T, Olivecrona G, Samra JS, Summers LK, Humphreys SM, Frayn KN. Lipoprotein lipase transport in plasma: role of muscle and adipose tissues in regulation of plasma lipoprotein lipase concentrations. J Lipid Res. 1998; 39: 2387–2393.