Quantification of Postprandial Triglyceride-Rich Lipoproteins in Healthy Men by Retinyl Ester Labeling and Simultaneous Measurement of Apolipoproteins B-48 and B-100
Abstract The metabolism of chylomicrons, very-low-density lipoprotein (VLDL), and their remnants in the postprandial state was studied in normolipidemic healthy men by measuring apoB-48 and apoB-100 and retinyl palmitate (RP) in fractions of triglyceride-rich lipoproteins after a mixed meal type of oral fat load supplemented with vitamin A. ApoB-48 was present at low concentrations in the fasting plasma samples in most subjects and increased in response to the test meal in Svedberg’s flotation rate (Sf) >20 lipoprotein fractions. Concomitantly, the level of Sf 60 to 400 apoB-100 (large VLDL) had doubled at 3 hours and returned to baseline at 9 hours. The number of apoB-48–containing lipoprotein particles did not exceed 20% of the total number of apoB-containing lipoproteins contained in Sf 12 to 400 fractions at any time point after fat intake. The peak plasma level of RP was delayed compared with the peak plasma concentration of apoB-48, suggesting that retinyl ester labeling of chylomicrons is questionable as a means of quantifying postprandial triglyceride-rich lipoproteins of intestinal origin. Approximately 2000 and 4000 RP molecules were carried in each chylomicron particle in the 3- and 6-hour samples, respectively, in contrast to the remnant fractions in which 100 to 600 RP molecules were found for each lipoprotein particle. The limited RP exchange between lipoprotein particles indicates that the smaller intestinal lipoproteins do no originate primarily from larger Sf >400 chylomicron particles but instead are secreted directly into the Sf 20 to 400 fraction and subsequently converted to smaller chylomicron remnants. This notion was further substantiated by injection of RP-labeled postprandial plasma into fasting subjects with subsequent tracing of RP in fractions of triglyceride-rich lipoproteins.
- Received June 22, 1994.
- Accepted November 14, 1994.
Triglycerides are carried in plasma lipoproteins to provide energy for peripheral tissues. In essence, lipoprotein lipase (LPL) hydrolyzes core triglycerides to free fatty acids, either for immediate use in muscles or for storage in adipose tissue. However, the other side of the coin may be accumulation of potentially atherogenic remnant lipoprotein particles.1 Very-low-density lipoprotein (VLDL) synthesized in the human liver has apoB-100 as the major protein component in contrast to chylomicrons secreted from the intestine after fat intake, which have apoB-48 as their structural protein. The two apoB-containing lipoprotein species share and compete for the same lipolytic pathway,2 but the intestinal triglycerides seem to be the favored substrate for LPL. ApoB-48 in chylomicrons and their remnants are found at a very low plasma concentration both in the fasting and the postprandial state, whereas VLDL apoB-100 increases significantly after fat intake, presumably because of delayed hydrolysis. In fact, a major proportion of the lipoproteinemia found after fat intake is accounted for by VLDL.3 4 However, it has recently been demonstrated that an increase in VLDL triglycerides accounts for only approximately 20% of postprandial triglycerides.5
Ingestion of vitamin A with fat gives rise to retinyl ester labeling of chylomicrons.6 Supplementation of the test meal with vitamin A has therefore been commonly used as a means of quantifying lipoproteins of intestinal origin in the postprandial state.7 8 9 10 11 12 13 14 15 16 17 However, the accuracy of using vitamin A in this context has been questioned.18 Retinyl palmitate (RP) appeared in presumably endogenous lipoproteins at late time points, and the RP peak was delayed compared with both the plasma triglyceride and the apoB-48 peaks seen after fat intake.18
Furthermore, Cohn et al5 recently showed that up to 25% of the postprandial plasma RP is bound to apoB-100–containing triglyceride-rich lipoproteins. This RP fraction appeared at late time points, which suggests that it derives from the transfer of core lipids between lipoproteins. No data obtained by simultaneous quantification of apoB-48 and RP after fat intake in subfractions of triglyceride-rich lipoproteins are available yet, and results from specific quantification of chylomicron remnants and VLDL of varying particle size are sparse.4 Against this background, we wanted to investigate the relative contribution of hepatic and intestinal triglyceride-rich lipoproteins to alimentary lipoproteinemia in a large group of healthy men by measuring apoB-48, apoB-100, and RP in isolated fractions of triglyceride-rich lipoproteins after a mixed meal type of oral fat load. It was hypothesized that a considerable heterogeneity also would exist in the response of triglyceride-rich lipoproteins to fat ingestion among normolipidemic subjects. As a secondary issue, we wanted to determine both apoB-48 and RP in fractions of triglyceride-rich lipoproteins at various time points after intake of the test meal to allow for conclusions regarding the precursor-product relations of chylomicron remnants of various particle sizes. This also provided an opportunity to reassess the usefulness of retinyl ester labeling in studies of the metabolism of chylomicrons and their remnants.
Healthy men 35 to 45 years of age were recruited from a population survey including 160 subjects of north European descent. A total of 129 of the 160 men approached agreed to give blood samples for routine determination of fasting plasma lipoproteins. To allow identification of a homogeneous group of normolipidemic subjects, apoE phenotyping19 was performed, and the 90th percentiles for plasma concentrations of VLDL triglycerides and low-density lipoprotein (LDL) cholesterol were used for lipoprotein phenotyping. After subjects homozygous for the ε3 allele with normal fasting plasma VLDL triglycerides and LDL cholesterol (below 1.95 and 4.70 mmol/L, respectively) were selected, a total of 58 subjects remained. Of these, 38 were randomly selected for the present study and asked to come for an oral fat tolerance test. Of the 38, 9 declined participation, and 2 could eventually not participate, which left 27 subjects for the present report.
Oral Fat Load
Participants were admitted early in the morning to the Clinical Research Unit for a mixed meal type of oral fat tolerance test. They had been fasting for 12 hours and were asked to refrain from smoking during the fasting period and from alcohol intake during the preceding 3 days. The protocol for the fat tolerance test was a slight modification of that of Cohn et al.20 An emulsion consisting of soybean oil (Karlshamns Oils & Fats AB, 50 g/m2 body surface area),21 glucose (50 g/m2), egg-white protein (Sigma 0500, Sigma Chemical Co, 25 g/m2), dried egg yolk (Sigma 0625, 6.3 g/m2), and 200 mL water prepared with some lemon flavor (60.2% fat, 13.3% protein, 26.5% carbohydrate by energy) was ingested within 10 minutes between 7 and 7:30 am. Five tablets containing the equivalent of 250 000 IU of vitamin A as RP (Arovit, Hoffman-La Roche) were added to the meal. The subjects were instructed to chew the tablets carefully. All subjects tolerated the test meal. Blood samples were obtained through an in-dwelling catheter inserted into the antecubital vein. A fasting blood sample was taken before the test meal. Subsequent blood samples were drawn hourly, and the last sample was taken 9 hours after ingestion of the emulsion. Samples for lipoprotein fractionation were drawn before and 3, 6, and 9 hours after intake of the test meal. Participants were allowed to be ambulant throughout the test period. Smoking was prohibited. Water but no food was allowed during the test.
Venous blood samples were drawn into precooled sterile tubes (Vacutainer, Becton Dickinson) containing Na2EDTA (final concentration, 4 mmol/L), which were instantly put into ice water and protected from light. Plasma was then recovered within 30 minutes by low-speed centrifugation (1.750g, 20 minutes, 1°C) and kept at this temperature throughout the preparation procedures. Phenylmethylsulfonyl fluoride (10 mmol/L, dissolved in isopropanol) and aprotinin (1400 μg/mL) (Trasylol, Bayer) were immediately added to the isolated plasma before fractionation of triglyceride-rich lipoproteins to final concentrations of 10 μmol/L and 28 μg/mL, respectively.
Triglyceride-rich lipoprotein was subfractionated by cumulative rate ultracentrifugation4 from plasma drawn before ingestion of the test meal and 3, 6, and 9 hours afterward. Plasma was adjusted to a density of 1.10 kg/L with solid NaCl. A density gradient consisting of 4 mL of 1.10 kg/L plasma and 3 mL each of 1.065, 1.020, and 1.006 kg/L NaCl solutions was then formed in Beckman Ultraclear tubes (volume, 13.4 mL) that had been coated with polyvinyl alcohol (BDH Chemicals Ltd). Ultracentrifugation was performed in a Beckman SW40 Ti swinging bucket rotor at 40 000 rpm and 15°C (Beckman XL-70 Optima ultracentrifuge). Consecutive runs calculated to float Svedberg’s flotation rate (Sf) >400 (32 minutes), Sf 60 to 400 (3 hours, 28 minutes), and Sf 20 to 60 (16 hours) particles were made. After each centrifugation, the top 0.5 mL of the gradient containing the respective lipoprotein subclasses was aspirated, and a salt solution with a density of 1.006 kg/L was used to refill the tube before the next run. The Sf 12 to 20 fraction was recovered after the last ultracentrifugal run by slicing the tube 29 mm from the top after the Sf 20 to 60 lipoproteins had been aspirated. All salt solutions used to prepare the density gradients were adjusted to pH 7.4 and contained 0.02% NaAzide and 0.01% EDTA. Densities were verified to the fourth decimal place (Paar densitometer). The isolated fractions were denoted according to Sf nomenclature, despite the fact that plasma lipoproteins were isolated by means of cumulative flotation in a density gradient.
Determination of ApoB-48 and ApoB-100
Samples containing isolated lipoprotein fractions were delipidated in a methanol/diethyl ether solvent system. A volume of 50 to 750 μL of the sample was injected into 4.0 mL methanol and 4.0 mL ice-cold diethyl ether was added. The tube was centrifuged for 30 minutes at 4000g at 1°C. After solvent removal, another 4.0 mL of diethyl ether was added, and the tube was centrifuged for 20 minutes. The ether was then removed, and the protein pellet was dissolved in 100 to 500 μL of 0.15 mol/L sodium phosphate, 12.5% glycerol, 2% sodium dodecyl sulfate (SDS), 5% mercaptoethanol, and 0.001% bromophenol blue, pH 6.8, at room temperature for 30 minutes and denatured at 80°C for 10 minutes. Most samples were then frozen at −20°C until their content of apoB-48 and apoB-100 was determined by analytical SDS–polyacrylamide gel electrophoresis (SDS-PAGE). Electrophoresis was performed with a vertical Hoefer Mighty Small II electrophoresis apparatus connected to an EPS 400/500 (Pharmacia) power supply and 3% to 20% gradient polyacrylamide gels. The upper and lower electrophoresis buffers contained 25 mmol/L Tris, 192 mmol/L Glycin, and 0.2% SDS adjusted to pH 8.5. ApoB-100 derived from LDL was used as a reference protein and for standard-curve dilutions. The apoB-100 preparation was obtained from LDL (1.030<d<1.040 kg/L) isolated from fasting plasma samples by the rate ultracentrifugation procedure described above. The total protein content was then determined22 with addition of SDS (final concentration, 1%) to the reagent mixture to reduce turbidity. The isolated LDL was delipidated and thereafter treated exactly as the samples containing triglyceride-rich lipoproteins. When restricted amounts of apoB are applied on the gel, there is no difference in chromogenicity between apoB-100 and apoB-48.23 A separate protein standard containing isolated apoB-48 was therefore not used to determine chylomicrons and chylomicron remnants. A dilution curve ranging from 0.10 to 2.0 μg of apoB-100 was applied to six lanes of the gel. When two gels were run simultaneously and thereafter fixed, stained, and destained in the same bowl, a standard curve was applied to only one gel. Electrophoresis was first run at 60 V for 20 minutes and then at 100 V for 2 hours. Gels were fixed in 12% trichloroacetic acid for at least 30 minutes and stained in 0.2% Coomassie G-250 (Serva)/40% methanol/10% acetic acid for at least 4 hours in a glass Petri dish. Destaining was done in 12% methanol/7% acetic acid with four changes of destainer for 24 hours. Gels were scanned with a laser scanner (Ultroscan XL, Pharmacia-LKB) connected to a personal computer equipped with software providing automatic integration of areas under scanning curves (Gelscan XL, Pharmacia). Background intensity was withdrawn by scanning of an empty lane. Bands with an obvious optical density greater than the highest standard point were not evaluated. With the present methodology, the limit for detection of apoB-48 and apoB-100 is between 0.01 and 0.02 mg/L plasma concentration. The coefficient of variation for the SDS-PAGE step was 4.4% for apoB-48 and 4.2% for apoB-100. If the error introduced during ultracentrifugation was included, the coefficient of variation ranged from 3.1% for components present at a high concentration to 14% for components present at a low concentration (>0.3 mg/L). The SDS-PAGE method for determining apoB-48 and apoB-100 and its evaluation have been described in detail elsewhere.23
Plasma and fractions of triglyceride-rich lipoproteins were protected from light throughout the preparation procedure. Volumes ranging from 100 to 250 μL plasma and triglyceride-rich lipoprotein fractions were extracted with 4 mL methanol and 4 mL hexane. The upper phase was removed, and the solvent was evaporated with a gentle stream of N2. The lipid was dissolved in 200 μL methanol/chloroform (3:1) and injected into a high-performance liquid chromatography system (System Gold, Beckman). RP was eluted isocratically with methanol, and peaks were recorded by UV absorbance at 326 nm. Peaks were integrated with the System Gold software, and concentration was determined with a standard curve of five dilutions (7.6 to 38 nmol) of highly purified RP (Sigma R-3375). The mean recovery of RP in fractions compared with plasma was 89±15% at 3 hours and 86±17% at 6 hours. On analysis of the same sample in triplicate, the coefficients of variation were 1.8% for a plasma with a high (6-hour sample) and 5.8% for a plasma sample with a low (9-hour sample) RP content.
Turnover of RP-Labeled Lipoproteins
To study the turnover of RP-labeled triglyceride-rich lipoproteins, a technique developed by Berr and Kern24 25 was used with minor modifications. Five healthy men ingested the mixed meal supplemented with vitamin A. Plasmapheresis was performed 4 hours later to isolate approximately 0.5 L of postprandial plasma containing RP-labeled triglyceride-rich lipoproteins. The plasma bag was stored at 4°C on a slowly shaking tray and reinjected intravenously 24 hours after plasmapheresis. Subjects were instructed to ingest a light breakfast at 7 am, and the infusion took place at 11 to 11:30 am that day. The infusion was completed within 4 to 5 minutes. Blood samples were taken before (three samples during 7.5 minutes) and 1, 2.5, 5, 7.5, 10, 15, 20, 30, 45, 60, 90, 120, 180, 240, and 360 minutes after the infusion was completed. Plasma was isolated from the blood samples and the plasma bags, and the concentration of RP in Sf >400, Sf 60 to 400, and Sf 20 to 60 lipoprotein fractions was determined as previously described. Table 1⇓ lists the characteristics of the five subjects undergoing plasmapheresis and reinjection of RP-labeled postprandial plasma.
Major Plasma Lipoproteins
The major fasting plasma lipoproteins (VLDL, LDL, and high-density lipoprotein [HDL]) were determined by a combination of preparative ultracentrifugation and precipitation of apoB-containing lipoproteins followed by lipid analysis.26 Total cholesterol27 and triglycerides28 were determined in triplicate in plasma and in the lipoprotein fractions. Lipids were first extracted with chloroform/methanol.29 Cholesterol and triglycerides were then determined on an Ultrolab (LKB).
Conventional methods were used to calculate means and SDs. SD is given in the text; SEM is given in the figures. Interassay and intra-assay coefficients of variation were calculated according to standard procedures.30 The coefficient of variation was given as the SD in percent. Associations among lipoprotein parameters were determined by calculation of Spearman rank correlation coefficients.
Study protocol was approved by the local ethics committee at Karolinska Hospital. All subjects gave informed consent.
At the time of the oral fat load, a total of 24 men had fasting plasma levels of VLDL triglycerides and LDL cholesterol below the 90th percentiles of their age segment of the population. Three subjects had mild to moderate hypertriglyceridemia (VLDL triglycerides of 2.10, 2.81, and 4.95 mmol/L, respectively) and were analyzed separately. Four of the 24 normotriglyceridemic and 1 of the 3 hypertriglyceridemic men were smokers. Body mass index was below 28.0 kg/m2 in all subjects (mean±SD, 24.1±2.2 and 24.6±2.7 kg/m2 in the normotriglyceridemic and hypertriglyceridemic subjects, respectively). Table 2⇓ gives the fasting plasma concentrations of cholesterol and triglycerides in the major lipoprotein fractions. The hypertriglyceridemic subjects had LDL cholesterol levels within the same range as the normotriglyceridemic subjects (2.71 to 3.84 mmol/L), whereas their HDL cholesterol levels were lower (0.66 to 1.17 mmol/L).
Plasma triglycerides increased from 0.99±0.35 to 2.31±1.11 mmol/L at 4 hours and returned to baseline 8 hours after the oral fat intake in the normotriglyceridemic subjects (Fig 1⇓). The peak value for the three hypertriglyceridemic subjects was attained at 4 hours and returned to baseline within 9 hours (Fig 1⇓).
Fasting Plasma ApoB-48 and ApoB-100 Concentrations
ApoB-48 was present at low concentrations in fasting plasma samples in most subjects. In the Sf >400 fraction, only 2 of the 24 subjects had levels above the detection limit, whereas the corresponding figures for the Sf 60 to 400, Sf 20 to 60, and Sf 12 to 20 fractions were 19, 21, and 7 of 24, respectively (Table 3⇓). It is notable that apoB-100 was present at a low concentration in the Sf >400 fraction in 12 of 24 normolipidemic and in all 3 hypertriglyceridemic subjects. There was a 20- to 50-fold range of apoB-100 concentrations in the three Sf 12 to 400 fractions in the normotriglyceridemic subjects. Plasma levels ranged between 0.8 and 39.4 mg/L in the Sf 60 to 400 fraction. The corresponding figures for the Sf 20 to 60 and Sf 12 to 20 fractions were 4.1 to 85.6 and 5.5 to 129.8 mg/L, respectively. The triglyceride concentrations were strongly correlated with the respective apoB-100 levels in the Sf 60 to 400 (r=.78, P<.001) and the Sf 20 to 60 (r=.60, P<.01) fractions but not in the Sf 12 to 20 fraction (r=.04, NS). In contrast, there was no statistically significant correlation between the triglyceride concentration and the apoB-48 content in any of these fractions (r=.38, r=.06, and r=.10, respectively).
Postprandial ApoB-48 and ApoB-100 Concentrations
ApoB-48 increased in response to the test meal in the Sf >20 fractions, whereas the low initial level was unchanged in the Sf 12 to 20 fraction (Fig 2⇓, top). The most prominent increase was seen in the Sf 60 to 400 fraction, with an elevation from 0.2±0.3 to 1.0±0.7 mg/L at 3 hours. ApoB-100 remained very low in the Sf >400 fraction after fat intake, whereas it approximately doubled between 0 and 3 hours in the Sf 60 to 400 fraction (Fig 2⇓, bottom). At 6 hours, the Sf 60 to 400 apoB-100 level was still elevated but had returned to baseline values at 9 hours. In the Sf 20 to 60 fraction, no significant change was seen in response to the oral fat load. In contrast, a 20% decrease of apoB-100 was seen at 3 hours in the Sf 12 to 20 fraction.
In hypertriglyceridemic subjects, the increase of apoB-48 in the Sf >400 fraction was larger than for normotriglyceridemic men (1.0±0.3 versus 0.2±0.2 mg/L at 3 hours). In addition, the mean fasting level of apoB-48 in the Sf 60 to 400 fraction of the hypertriglyceridemic subjects was higher than the 3-hour postprandial level of the normotriglyceridemic men (1.4±1.4 versus 1.0±0.7 mg/L) (Fig 3⇓). It increased to 3.2 mg/L at 3 hours and to 4.0±2.0 mg/L at 6 hours and declined to 2.2±2.7 mg/L at 9 hours. The Sf 20 to 60 apoB-48 concentrations were high in fasting plasma, and a heterogeneous pattern was seen in the postprandial state. Sf 60 to 400 apoB-100 increased considerably in two of the three hypertriglyceridemic subjects at 3 hours (Fig 3⇓). In absolute terms, the increments were far more prominent than for normotriglyceridemic subjects. Of note, the 9-hour level of apoB-100 in the Sf 60 to 400 fraction of the hypertriglyceridemic subjects was well below the fasting level (51.9±33.5 versus 83.2±47.5 mg/L). Similar to what was noted among the normolipidemic participants, no major change was seen in the apoB-100 concentration in the Sf 20 to 60 fraction in two of the three hypertriglyceridemic subjects (Fig 3⇓), and the apoB-100 concentration of the Sf 12 to 20 fraction had decreased by 20% at 3 hours after fat intake (from 71.7±16.2 to 55.5±23.3 mg/L) and returned to baseline at 9 hours.
There was considerable heterogeneity in the responses to fat intake of plasma triglycerides, apoB-48, and apoB-100 in subfractions of triglyceride-rich lipoproteins, despite the fact that healthy normolipidemic subjects with an apo ε3/ε3 phenotype were studied. The median relative increase of whole plasma triglycerides between 0 and 3 hours was 117% (range, 20% to 316%); the corresponding figure for Sf 60 to 400 apoB-100 was 81% (range, −20 to 1912%). Interestingly, the relative increase of Sf 60 to 400 apoB-48 was 79% (3% to 168%). This indicates that the postprandial Sf 60 to 400 apoB-100 concentration was the most heterogeneous variable.
Body mass index did not correlate significantly with the increase in apoB-100 in Sf 20 to 60 and Sf 60 to 400 fractions between 0 and 3 hours or to the corresponding increase between 0 and 6 hours (data not shown).
Relations Between Fasting and Postprandial ApoB-48 and ApoB-100 Concentrations
Spearman rank correlation coefficients were calculated to examine the relations between fasting plasma apoB-48 and apoB-100 in Sf 60 to 400 and Sf 20 to 60 fractions and the corresponding measurements at 3, 6, and 9 hours (Table 4⇓). Strong relations were found between fasting and postprandial levels of apoB-100 in the Sf 60 to 400 and Sf 20 to 60 fractions, whereas considerably weaker or even nonsignificant relations were seen between apoB-48 measurements in the same fractions. This implies that fasting VLDL concentrations are good predictors of postprandial levels in contrast to fasting chylomicron remnant concentrations. A weak correlation was also found between the fasting levels of apoB-48 and apoB-100 in the Sf 60 to 400 fraction (r=.48, P<.05). The corresponding correlation for the Sf 20 to 60 fraction was nonsignificant (r=.32).
The postprandial plasma concentration of RP reflected the levels in the Sf >400 and Sf 60 to 400 lipoprotein fractions because the amount of RP in the Sf 20 to 60 and Sf 12 to 20 fractions were very low or even below the limit of detection. The peak RP levels in both Sf >400 and Sf 60 to 400 fractions were found in the 6-hour sample (Fig 4⇓). Of note, the very low concentrations of RP in the Sf 20 to 60 fraction still increased at 9 hours. In accordance with the findings of Krasinski et al,18 the amount of RP in non–triglyceride-rich lipoproteins (difference between the plasma RP level and RP contained in Sf >12 lipoprotein fractions) increased with time. At 3 hours, 13±3% of the plasma RP could not be retrieved in the Sf >12 lipoprotein fractions. The corresponding values for 6 and 9 hours were 15±5% and 17±4%, respectively. Fig 5⇓ shows a comparison between the total incremental apoB-48 in the Sf >12 fraction and the plasma concentration of RP. The peaks do not seem to coincide. The number of RP molecules per apoB-48–containing lipoprotein particle in the Sf >400, Sf 60 to 400, and Sf 20 to 60 fractions at 3, 6, and 9 hours was calculated by dividing the RP concentration by the incremental apoB-48 (260 kD) expressed in molar terms (Table 5⇓). All fractions and time points were not included for all subjects because measurable levels of both apoB-48 and RP were required. The number of RP molecules in an average chylomicron (Sf >400 lipoprotein particle) was 2000 at 3 hours and increased to 4000 at 6 hours. In the Sf 60 to 400 fraction containing large chylomicron remnants and VLDL particles, the number of RP molecules per apoB-48–containing lipoprotein particle increased from 300 to 600 between 3 and 6 hours. In the Sf 20 to 60 fraction, the number of RP molecules per apoB-48–containing particle was of the same order of magnitude or slightly lower compared with the Sf 60 to 400 fraction. The corollary of the findings with respect to RP is that Sf 20 to 400 chylomicron remnants do not originate primarily from larger Sf >400 chylomicron particles but instead are secreted directly into the Sf 20 to 400 fraction and subsequently are converted to smaller chylomicron remnants by endothelial lipases.
Turnover of RP-Labeled Postprandial Triglyceride-Rich Lipoproteins
Fig 6⇓ shows the disappearance of RP from the Sf >400, Sf 60 to 400, and Sf 20 to 60 fractions. The total amount of reinjected RP and the distribution of RP between the lipoprotein fractions contained in the plasma differed markedly among the five subjects studied (Table 1⇑). The distributions of RP between subfractions of triglyceride-rich lipoproteins isolated from plasma samples immediately after completed plasmapheresis and after the overnight storage of the plasma were almost identical, which indicates that the chylomicron and chylomicron remnant particles remained intact and did not aggregate during storage. Furthermore, the total RP concentration in plasma was not affected by storage (Table 1⇑). The decay of RP in the Sf >400 fraction was very rapid in all subjects. The t1/2s of the rapid component of the RP decay in the Sf >400 fraction were 11.1, 4.9, 10.4, 6.8, and 9.8 minutes for subjects 1 through 5, respectively, with the assumption that the first six time points exhibited first-order kinetics. The initial increase of RP in the Sf 60 to 400 fraction was minimal, if present at all, suggesting that very small amounts of intestinal lipoproteins entered this fraction from the Sf >400 fraction. The reinjected amount of RP-labeled lipoproteins contained in the Sf 20 to 60 fraction was low, and no increase in RP was seen in this fraction after the infusion. Instead, a slow, almost linear decline of RP was observed in this fraction in the five subjects. This pattern indicates that only a limited amount of intestinal lipoproteins in the Sf 60 to 400 range are converted to smaller remnant particles as a consequence of lipolysis.
The present study dealt with the metabolism of postprandial triglyceride-rich lipoproteins in healthy men after intake of a standardized oral fat load. Very low concentrations of apoB-48–containing lipoproteins were found in the fasting state, and the increases of chylomicrons and their remnants after fat intake were small in absolute terms. In contrast, a substantial increase of large VLDL (Sf 60 to 400 apoB-100) was observed, whereas the plasma concentration of small VLDL (Sf 20 to 60 apoB-100) was not affected by the oral fat load. We previously described a similar response to fat intake in a study comprising selected normotriglyceridemic and hypertriglyceridemic young male postinfarction patients and a small group of normolipidemic control men.4
The prominent postprandial increase of large VLDL is likely to result from competition with chylomicrons for a common lipolytic pathway.2 4 This observation was also made in a recent study by Schneeman et al3 using a somewhat different methodology, including separation of postprandial triglyceride-rich lipoproteins on an immunoaffinity column selecting apoB-100–containing lipoprotein particles. These authors demonstrated that 80% of the increase in postprandial triglyceride-rich lipoprotein particles was accounted for by VLDL. The present data confirm this figure. Furthermore, our results show that the number of apoB-48–containing lipoprotein particles does not exceed 20% of the total number of apoB-containing lipoproteins in Sf <400 subfractions at any time point after fat intake. These figures for the relation between apoB-100– and apoB-48–containing lipoproteins and those of Schneeman et al should be compared with the findings of Cohn et al,5 who showed that 80% of the postprandial lipemia was accounted for by apoB-48–containing lipoproteins. Considering the massive triglyceride load of each chylomicron particle, the relations seem to be consistent. Accordingly, the fact that the contribution of chylomicron remnant particles to alimentary lipoproteinemia is limited in relation to VLDL must be taken into consideration when an evaluation of the association between postprandial lipoproteins and the development of coronary atherosclerosis is attempted.
The plasma levels of apoB-100 and in particular apoB-48 in the various lipoprotein fractions were somewhat lower in the present study than in our previous study of a small group of healthy men.4 This may have several explanations. First, the previously described control subjects were about 10 years older. Second, an improved method for quantifying apoB-48 and apoB-100 was used in the present study, which was validated against amino acid analysis of the respective proteins.23 The previous method was based on the relative chromogenicities of apoB-48 and apoB-100 in fractions of triglyceride-rich lipoproteins on SDS-PAGE rod gels. This latter procedure may have inherent methodological problems if the gels are overloaded.23
The fasting and postprandial plasma levels of apoB-48 and apoB-100 in Sf >400, Sf 60 to 400, and Sf 20 to 60 lipoprotein fractions showed a considerable heterogeneity despite the fact that nonobese healthy normolipidemic subjects who were homozygous for the ε3 allele were selected. The postprandial 3-hour response of large VLDL (Sf 60 to 400 apoB-100) to fat intake was extremely variable, ranging from a small decrease to a 20-fold increase. This was in contrast to whole plasma triglycerides and chylomicron remnants, the increases of which ranged from 20% to 300%.
Despite the selection procedure, 3 of the 27 subjects had elevated fasting plasma triglycerides at the time of the oral fat tolerance test. In our recent studies of hypertriglyceridemic individuals with manifest coronary atherosclerosis and a previous myocardial infarction, a pattern of delayed clearance of both postprandial whole-plasma triglycerides and large VLDL was evident.4 It was therefore of interest to note that the three hypertriglyceridemic subjects in the present study could clear elevated postprandial lipoproteins as efficiently as normotriglyceridemic men. Further studies are needed to establish whether there is a difference in composition and metabolism of postprandial triglyceride-rich lipoproteins between hypertriglyceridemic subjects with and without manifest coronary atherosclerosis.
Postprandial plasma levels of RP have commonly been used as an indirect measure of chylomicrons and their remnants after intake of a fat meal supplemented with vitamin A. Krasinski et al18 indicated that this approach is invalid at late time points after intake of fat and vitamin A. First, they documented the appearance of RP in lipoprotein species of hepatic origin. Second, simultaneous peaks were noted for plasma triglycerides and apoB-48, whereas the RP peak was delayed. Furthermore, it has recently been shown that a significant proportion of RP (up to 25%) is carried by apoB-100–containing lipoproteins, in particular at late postprandial time points.5 In the present work, the appearance of RP seemed to be delayed compared with the pattern of elevation for chylomicrons and chylomicron remnants. This delay, however, should be interpreted with caution because we are dealing with lipoprotein particles with an estimated half-life of 5 to 15 minutes, and the time interval between samples was 3 hours. However, as a consequence of the delayed appearance of RP, the number of RP molecules per apoB-48–containing lipoprotein particle increased considerably between 3 and 6 hours in all fractions of triglyceride-rich lipoproteins. It should be emphasized that this calculation was based on two assumptions, which may not be entirely correct. We assumed first that the intestine is not producing significant amounts of apoB-100–containing lipoproteins and second that exchange of RP between lipoprotein particles in humans8 15 or mere hydrolysis of RP31 is negligible. Of note, the fraction containing “true chylomicrons” (the Sf >400 fraction) had between 2000 and 4000 RP molecules per chylomicron in our calculations; the fractions containing chylomicron remnants had between 100 and 600 RP molecules per chylomicron remnant particle. One obvious interpretation of this finding is that the chylomicron remnant particles in the Sf 20 to 400 fraction do not originate primarily from larger chylomicrons. Instead, we suggest that large chylomicrons are rapidly and efficiently catabolized before they enter the Sf 20 to 400 fraction, a notion that is compatible with the very low levels of apoB-48 in the Sf >400 fraction, and that the intestine also secretes smaller chylomicron particles (contained in the Sf 20 to 400 fraction) that are converted to even smaller chylomicron remnants by endothelial lipases. This hypothesis was strongly supported by the in vivo turnover studies made with RP-labeled postprandial plasma, which consistently showed that there was no significant entry of RP-labeled lipoproteins either from the Sf >400 fraction into the Sf 60 to 400 fraction or from the Sf 60 to 400 fraction into the Sf 20 to 60 fraction. Indeed, the experiments illustrate the efficiency of the removal mechanisms for chylomicrons and chylomicron remnants from human plasma.32 Furthermore, there was no relation between the fasting and postprandial levels of apoB-48 in the subfractions of triglyceride-rich lipoproteins. This is in contrast to apoB-100, the fasting level of which was a strong predictor of the postprandial response. This finding may indicate that small apoB-48–containing lipoproteins are released continuously in the fasting state, which does not relate to the massive demand for triglyceride transporters from the intestine that arises after fat intake.
Delayed appearance of RP compared with the pattern of increase for plasma triglycerides is not a consistent finding in the literature. Several researchers who have quantified both plasma triglycerides and RP found simultaneous peaks,10 11 12 14 15 whereas a few reported diverging peaks,17 18 findings that in some instances have not been discussed.13 16 The present investigation was not aimed at elucidating the mechanism of the delayed appearance of RP compared with chylomicrons and chylomicron remnants. A specific delay in the removal of RP-enriched apoB-48–containing lipoproteins is unlikely because the late appearance of RP compared with apoB-48 was noted in all fractions, including the one containing the largest particles (Sf >400). However, it is tempting to speculate that the vitamin absorption rate causes this apparent discrepancy. Indeed, polyunsaturated fats have been described to delay the absorption of vitamin A.33 In most previous studies on postprandial lipoprotein metabolism, dairy cream has been used in the fat load, whereas we, together with Krasinski et al,18 used soybean oil. A delayed absorption of vitamin A caused by the large amount of polyunsaturated fats contained in the test meal might explain the late appearance in plasma of RP compared with chylomicrons and facilitated demonstration of a dissociation between plasma levels and time courses for apoB-48 and RP. Consequently, ingestion of vitamin A with fat to produce retinyl ester labeling of chylomicrons seems not to be a valid means of quantifying triglyceride-rich lipoprotein particles of intestinal origin in the postprandial state. However, RP may still be used as a tracer substance for intestinal lipoproteins.
Despite more than 30 years of epidemiological and clinical research, the relations among hypertriglyceridemia, triglyceride-rich lipoproteins, and coronary heart disease remain obscure and much debated.34 From the respective basal plasma levels and magnitudes of responses to fat intake observed for chylomicron remnants and VLDL in the present study, it could be hypothesized that chylomicrons and their remnants are implicated in atherogenesis by impeding the normal LPL-mediated catabolism of VLDL.
This study was supported by grants from the Swedish Medical Research Council (8691), the Swedish Heart-Lung Foundation, the Marianne and Marcus Wallenberg Foundation, the King Gustaf V 80th Birthday Fund, the Professor Nanna Svartź Fund, the Nordic Insulin Foundation, the Swedish Margarine Industry Fund for Research on Nutrition, and the Thuring Foundation. Dr Hamsten is career investigator of the Swedish Heart-Lung Foundation. We are grateful to Karin Danell-Toverud, Ninna Eriksson, and Anita Larsson for expert technical assistance.
Zilversmit DB. Atherosclerosis: a postprandial phenomenon. Circulation. 1979;60:473-485.
Brunzell JD, Hazzard WR, Porte D Jr, Bierman EL. Evidence for a common, saturable, triglyceride removal mechanism for chylomicrons and very low density lipoproteins in man. J Clin Invest. 1973;52:1578-1585.
Schneeman BO, Kotite L, Todd KM, Havel RJ. Relationship between the responses of triglyceride-rich lipoproteins in blood plasma containing apolipoproteins B-48 and B-100 to fat-containing meal in normolipidemic humans. Proc Natl Acad Sci U S A. 1993;90:2069-2073.
Karpe F, Steiner G, Olivecrona T, Carlson LA, Hamsten A. Metabolism of triglyceride-rich lipoproteins during alimentary lipemia. J Clin Invest. 1993;91:748-759.
Cohn JS, Johnson EJ, Millar JS, Cohn SD, Milne RW, Marcel YL, Russel RM, Schaefer EJ. Contribution of apo B-48 and apo B-100 triglyceride-rich lipoproteins (TRL) to postprandial increases in the plasma concentration of TRL triglycerides and retinyl esters. J Lipid Res. 1993;34:2033-2040.
Weintraub MS, Eisenberg S, Breslow JL. Different patterns of postprandial lipoprotein metabolism in normal, type IIa, type III and type IV hyperlipoproteinemic individuals: effects of treatment with cholestyramine and gemfibrozil. J Clin Invest. 1987;79:1110-1119.
Weintraub MS, Eisenberg S, Breslow JL. Dietary fat clearance in normal subjects is regulated by genetic variation in apolipoprotein E. J Clin Invest. 1987;80:1571-1577.
Weintraub MS, Zechner R, Brown A, Eisenberg S, Breslow JL. Dietary polyunsaturated fats of the W-6 and W-3 series reduce postprandial lipoprotein levels: chronic and acute effects of fat saturation on postprandial lipoprotein metabolism. J Clin Invest. 1988;82:1884-1893.
Rubensztein DC, Cohen JC, Berger MC, van der Westhuyzen DR, Coetzee GA, Gevers W. Chylomicron remnant clearance from the plasma is normal in familial hypercholesterolemic homozygotes with defined receptor defects. J Clin Invest. 1990;86:1306-1312.
Sprecher DL, Knauer SL, Black DM, Kaplan LA, Akeson M, Dusing AA, Lattier D, Stein EA, Rymaszewski M, Wiginton DA. Chylomicron-retinyl palmitate clearance in type I hyperlipidemic families. J Clin Invest. 1990;88:985-994.
Ooi TC, Simo IE, Yakichuk JA. Delayed clearance of postprandial chylomicrons and their remnants in the hypoalphalipoproteinemia and mild hypertriglyceridemia syndrome. Arterioscler Thromb. 992;12:1184-1190.
Cabezas MC, de Bruin TWA, Jansen H, Kock LAW, Kortlandt W, Erkelens DW. Impaired chylomicron remnant clearance in familial combined hyperlipidemia. Arterioscler Thromb. 1993;13:804-814.
Cohn JS, McNamara JR, Cohn SD, Ordovas JM, Schaefer EJ. Plasma apolipoprotein changes in the triglyceride-rich lipoprotein fraction of human subjects fed a fat-rich meal. J Lipid Res. 1988;29:925-936.
Diem K, Lentner C, eds. Documenta Geigy. 7th ed. Basel, Switzerland: JR Geigy SA; 1970:537.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951; 193:265-275.
Karpe F, Hamsten A. Determination of apolipoproteins B-48 and B-100 in triglyceride-rich lipoproteins by sodium dodecyl sulfate polyacrylamide gel electrophoresis. J Lipid Res. 1994;35:1311-1317.
Berr F, Kern F. Plasma clearance of chylomicrons labeled with retinyl palmitate in healthy human subjects. J Lipid Res. 1984; 25:805-812.
Berr F. Characteristics of chylomicron remnant clearance by retinyl palmitate label in normal humans. J Lipid Res. 1992;33:915-930.
Carlson K. Lipoprotein fractionation. J Clin Pathol. 1973;5(suppl 26):32-37.
Carlson LA. Determination of serum triglycerides. J Atheroscler Res. 1963;3:334-336.
Snedecor GW, Cochran WG. Statistical Methods. 7th ed. Ames, Iowa: Iowa State University Press; 1980.
Blaner WS, Obunike JC, Kurlandsky SB, Al-Haideri M, Piantedosi R, Deckelbaum RJ, Goldberg IJ. Lipoprotein lipase hydrolysis of retinyl ester: possible implication for retinoid uptake of cells. J Biol Chem. 1994;369:16559-16565.
Patsch J. Postprandial lipaemia. In: Shepherd J, ed. Baillière’s Clinical Endocrinology and Metabolism. Bailliere Tindall, London, England; 1987;1:551-580.
Austin M. Plasma triglyceride and coronary heart disease. Arterioscler Thromb. 1991;11:2-14.