Influence of Diets Rich in Saturated and Omega-6 Polyunsaturated Fatty Acids on the Postprandial Responses of Apolipoproteins B-48, B-100, E, and Lipids in Triglyceride-Rich Lipoproteins
Abstract The effects of diets rich in saturated fatty acids (SFA) (total polyunsaturated fatty acids [PUFA] [g]/total SFA [g][P/S ratio], 0.2) or omega-6 PUFA (P/S ratio, 1.3) on the postprandial response of triglyceride-rich lipoproteins (TRL) was determined in normolipidemic young men. After 15 and 29 days of diet intervention, the postabsorptive concentrations of apolipoprotein (apo) B-48 and apoB-100 were higher in the SFA group than in the PUFA group, but the absolute increase in apoB-48 was similar 3 hours after a challenge meal containing one third of daily energy and returned to postabsorptive values at 6 hours; this response was closely coupled to that of TRL triglycerides. In both groups, the percent increase in TRL apoB-48 and triglycerides was greater after the PUFA meal than after the SFA meal. The concentration of TRL apoB-100 also increased at 3 hours in both diet groups but returned to postabsorptive values at 6 hours only in those fed the PUFA diet; in the SFA group, apoB-100 remained high at 6 hours and fell below postabsorptive values only 9 hours after the meal. This apoB-100 response was affected primarily by the fatty acid composition of the diet and not by that of the challenge meal. The postprandial response of apoB-100 was closely coupled to that of cholesterol and apoE. These observations suggest that in healthy young men, neither the fatty acid composition of the diet nor that of the challenge meal affects the clearance of chylomicron remnants after a fat-containing meal. By contrast, the postprandial accumulation of hepatogenous TRL is prolonged in individuals fed a diet rich in SFA.
Reprint requests to Richard J. Havel, Cardiovascular Research Institute and Department of Medicine, University of California, San Francisco, CA 94143-0130.
- Received July 14, 1995.
- Accepted September 20, 1995.
Numerous studies have shown that dietary fat saturation affects plasma and LDL cholesterol concentrations, measured in the fasting (postabsorptive) state.1 Relatively little is known about the influence of dietary fat saturation on the metabolism of triglycerides and TRL in blood plasma in the fed (postprandial) state. Most individuals spend more than 12 hours daily in the postprandial state; plasma triglycerides increase shortly after the first meal of the day and return to postabsorptive concentrations several hours after the last meal of the day. Postprandial changes in lipoproteins have become increasingly important because accumulating evidence suggests that delayed clearance of postprandial TRL, namely intestine-derived chylomicrons and liver-derived VLDL, increases atherogenic risk.2 3
It has been suggested that individuals consuming a diet rich in SFA have a higher concentration of chylomicrons and their remnants4 and clear these particles more slowly5 than those consuming a diet rich in omega-64 5 or omega-3 PUFA.4 Based on the assumption that vitamin A esters specifically label the core of intestinal lipoproteins, Weintraub et al4 measured retinyl palmitate in Sf>1000 chylomicron and Sf<1000 remnant particles. They reported that the total area below the retinyl palmitate curves (0- to 12-hour interval) for Sf >1000 and Sf<1000 lipoproteins were highest on the SFA diet, lowest on the omega-3 PUFA diet, and intermediate on the omega-6 PUFA diet and concluded that omega-6 and omega-3 PUFA reduce the concentrations of chylomicrons and their remnants. A considerable amount of retinyl esters has been found to associate not only with apoB-48–containing chylomicrons but also with apoB-100–containing VLDL, particularly during later stages of alimentary lipemia (6 to 9 hours after the meal).6 The extent to which the SFA-induced increase in retinyl palmitate in the Sf<1000 fraction4 reflects accumulation of chylomicron remnants is therefore uncertain because this fraction contains not only chylomicron remnants and VLDL but also LDL and HDL, all of which may acquire retinyl esters during postprandial lipemia. Demacker and associates5 approached this problem by measuring the diurnal response of apoB-48 and apoB-100 in TRL (d<1.019 g/mL), estimated from dye absorbance of the proteins separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, in subjects who had consumed a diet rich in SFA or PUFA for 9 days. They reported a greater and more prolonged postprandial increase in apoB-48 in the SFA diet group than in the PUFA diet group and concluded that a diet rich in SFA delays the clearance of chylomicrons and their remnants.
Although the reported studies suggest that dietary fat saturation may affect the metabolism of TRL, data on postprandial changes in the absolute concentrations of apoB-48 and apoB-100, the structural components of chylomicrons and VLDL, are lacking. Furthermore, the procedures generally used to separate TRL have not permitted characterization of chylomicron and VLDL composition. To address these shortcomings, we used a quantitative approach to determine whether the metabolism of chylomicrons and VLDL is altered by a diet rich in SFA compared with a diet rich in omega-6 PUFA. We measured the postabsorptive and postprandial concentrations of apoB-48 and apoB-100, together with other protein and lipid components, in total TRL and a population of particles containing 85% to 90% of total TRL apoB-100 but no apoB-48.7 Because both the fatty acid composition of the background diet and the challenge meal have been suggested to affect the postprandial response,4 we investigated the responses to both an SFA- and a PUFA-rich challenge meal in healthy young men who had consumed an SFA- or PUFA-rich diet for 2 to 4 weeks.
Based on recent findings that plasma and LDL-cholesterol responses to saturated fat and cholesterol feeding may differ between Caucasians and non-Caucasians,8 our sample population was selected only among whites. Because hormonal fluctuations throughout the menstrual cycle affect lipoprotein metabolism and could confound postprandial lipoprotein responses, we did not include women in the study. Thirty-two healthy, nonmoking men (aged 25 to 40 years) were recruited mostly among employees and students from the University of California at San Francisco. We excluded men with plasma cholesterol values above the 95th percentile, HDL cholesterol concentrations below the 10th percentile, plasma triglycerides above the 90th percentile or below the 10th percentile for age,9 and a body mass index (kilograms per meter squared) above the 75th percentile for age.10 Men with a history of cardiovascular disease, diabetes, thyroid disorders, or renal disease were also excluded. Before acceptance, subjects were interviewed by a dietitian and were asked to complete a 24-hour dietary recall, a food frequency questionnaire, and an exercise questionnaire. Individuals with excessive alcohol intake, unusual dietary habits (eg, vegetarians), or irregular exercise habits were excluded. All procedures used during the study were approved by the University of California Committee of Human Research, and subjects gave their written consent to participate.
Energy requirements for each subject were initially estimated from the Harris-Benedict equation11 and a detailed physical activity questionnaire. Caloric adjustments were made, when necessary, to ensure that each participant maintained his weight within 3 lb of his starting weight. Because fluctuations in weight greatly affect triglyceride metabolism, body weight was monitored daily, and subjects were instructed not to deviate from their physical exercise habits, which were monitored weekly.
For the 4-week duration of the study, subjects ate all and only the food provided by the General Clinical Research Center. Subjects were required to consume at least two of three meals per day at the research unit and under the supervision of a dietitian. On request, the third meal was packaged to take home. The SFA- and PUFA-rich diets, which contained 200 mg cholesterol/1000 kcal, were modified from those used in an earlier study8 and consisted of natural foods, prepared daily in weighed, individual portions. To provide variety, 3-day rotating menus for both diets were developed with the use of computer-analyzed recipes designed for the study. To estimate the nutrient composition of the research diets, we used a nutrient calculation system customized for the General Clinical Research Center (Diet Planner, version 2.09) and based on the US Department of Agriculture Handbook 8,12 expanded to include the nutrient composition of chemically analyzed fats and food ingredients commonly used in our study. To determine the chemical composition of the research diets (Table 1⇓), food composites were prepared for each day of the rotating menus and were sent to Hazleton Laboratories America for analysis of energy, protein, carbohydrate, total fat, individual fatty acids, and cholesterol. Trans-fatty acids were analyzed by the Prince Edward Island Food Technology Center.
On entry into the study, subjects were randomly allocated to either the SFA-rich or the PUFA-rich background diet, which they consumed for 29 days (Fig 1⇓). To assess diet-induced changes in plasma lipids and lipoproteins over time, blood was collected from subjects fasted overnight at baseline (day 1) and on days 8, 15, 22, and 29.
Two postprandial studies were conducted during the 4-week diet intervention. Blood was sampled in the postabsorptive state (0-hour sample, obtained at 6 am) and 3, 6, 9, and 12 hours after subjects ate a challenge meal. For the first postprandial study (day 15), aimed at determining the effect of dietary fat saturation in the background diet on the responses of postprandial TRL (“diet effect”), all subjects ate a challenge meal similar in fatty acid composition to their respective background diet. For the second postprandial study (day 29), one half of the subjects on a given background diet were randomly allocated to consume a PUFA-rich challenge meal, and the other half consumed an SFA-rich challenge meal. This allowed us to determine (1) whether the fatty acid composition of the challenge meal itself (“meal effect”) affects the postprandial response and (2) whether the postprandial responses of subjects given the same challenge meal on days 15 and 29 are reproducible. The PUFA-rich challenge meal (P/S ratio, 1.2) and SFA-rich meal (P/S ratio, 0.2) provided one third of the daily energy (39% from fat, 46% from carbohydrates, 15% from protein, and approximately 250 mg of cholesterol/1000 calories). The SFA challenge meal consisted of a soup, crackers, cheese sandwich, applesauce, cookies, and milk, and the PUFA meal consisted of a tuna sandwich, peanut butter and jelly sandwich, custard, fruit juice, and milk. The tuna was packed in water and contained very small amounts of omega-3 fatty acids (PUFA meal, 0.81 g/1000 kcal; SFA meal, 0.43 g/1000 kcal). Subjects ate the challenge meal within 15 to 20 minutes and were not given any other food until the 12-hour postprandial sample of blood was collected. Water was permitted as desired.
Separation and Analysis of Plasma Lipoproteins
Blood samples in the postabsorptive state and for the postprandial studies were drawn into tubes containing disodium EDTA (0.05%) and benzamidine (0.03%) to prevent scission of apoB-100.13 The samples were kept on ice until plasma was separated by centrifugation (720g) at 4°C for 30 minutes. Total TRL were isolated by ultracentrifugation (d=1.006 g/mL)14 of 6 mL of plasma in a 50.3 Beckman rotor at 93 000g (average) at 12°C for 18 hours. TRL were collected by tube slicing, and the TRL infranatant was collected and brought to a final volume of 5 mL with 0.15 mol/L NaCl. A fraction containing VLDL+IDL was isolated from the initial plasma sample by ultracentrifugation at d=1.019 g/mL under the same conditions.
In a subset of individuals fed the SFA- and PUFA-rich diets, plasma collected during the postprandial studies was subjected to immunoaffinity chromatography on a column containing a monoclonal antibody to apoB-100 (JIH, Japan Immunoresearch Laboratories). Thereby, chylomicron-derived particles and a minor portion of apoB-100–containing TRL were separated from the bulk of TRL particles containing apoB-100 that bind to monoclonal antibody JIH.7 The unbound material collected after immunoaffinity chromatography was subjected to ultracentrifugation at d=1.006 g/mL as described above, and the unbound TRL were collected by tube slicing.
Plasma and lipoprotein total cholesterol, free cholesterol, and triglyceride concentrations were determined by enzymatic assays with a Cobas Mira analyzer (Roche Laboratories).15 16 Reagents for total cholesterol and triglyceride assays were from Roche Diagnostics; reagents for free cholesterol assays were from Wako Chemicals. An ABA-100 chromatic analyzer (Abbott Laboratories) was used to measure cholesterol in TRL from hypertriglyceridemic postprandial samples (obtained 3, 6, and 9 hours after the challenge meal) because the turbidity of these samples affected the linearity of the Roche-Diagnostics cholesterol assay. HDL lipids were measured in the supernatant obtained after precipitation of apoB-containing lipoproteins with dextran sulfate and magnesium chloride.17 All lipid analyses were done in duplicate and were routinely calibrated against a pooled serum sample. LDL lipid concentrations were calculated as lipid concentrations in the TRL infranatant minus lipid concentrations in HDL. IDL lipids in postabsorptive samples were calculated as lipid concentrations in the d=1.019 g/mL supernatant minus lipid concentrations in TRL.
ApoB-100, apoB-48, and apoE in total TRL and unbound TRL were quantified by densitometric scanning of Coomassie blue–stained apolipoproteins separated by sodium dodecyl sulfate–polyacrylamide slab gel electrophoresis.18 Briefly, to separate apolipoproteins, duplicate samples of delipidated TRL were subjected to electrophoresis in 3% to 10% linear gradient polyacrylamide slab minigels. After electrophoresis, gels were stained overnight in 0.25% Coomassie brilliant blue R-250, destained for 7 to 8 hours, and then dried. For apolipoprotein quantification, the gels were scanned with a laser densitometer, and the intensity of dye uptake of the apolipoprotein bands was measured in “volume units.” Volume units were translated into actual apolipoprotein mass based on standard curves constructed for pure apoB-100, apoB-48, and apoE. An example of a gel showing changes in these apolipoproteins before and after a challenge meal is shown in Fig 2⇓.
Differences between the postabsorptive concentrations of plasma lipids and lipoproteins in the SFA and PUFA diet groups were analyzed by the nonparametric Mann-Whitney U test. Data collected from the postprandial studies were subjected to two-way ANOVA for repeated measures to test the overall significance of postprandial measurements over time (time effect) and between diet groups (diet effect or challenge meal effect). When the overall F statistic was significant (P<.05), specific comparisons within a given group were evaluated by the nonparametric Wilcoxon matched-pairs test that accounts for skewed distributions. Associations between lipoprotein variables were examined by calculating the Pearson correlation coefficient. Values in the text, tables, and figures are expressed in milligrams per deciliter and as mean±SD. Differences were considered significant at P<.05.
Characteristics of Subjects at Entry Into Study
Of the 32 subjects who completed the diet study, 66% had an apoE3/3 phenotype, 28% had an apoE4/3 phenotype, and 6% had an apoE2/3 phenotype, a distribution close to Hardy-Weinberg equilibrium. Because our preliminary analysis of the data revealed that inclusion of individuals with an apoE4/3 and apoE2/3 phenotype was a confounder in interpreting the influence of dietary fat saturation on TRL responses, only data obtained from individuals with an apoE3/3 phenotype (13 in the SFA diet group, 8 in the PUFA diet group) are presented here.
On entry into the study, the young men allocated to the SFA and PUFA diets had similar anthropometric characteristics and postabsorptive plasma lipoproteins (Table 2⇓). In the group as a whole, mean body mass index (24.2±1.5), plasma cholesterol (171.0±26.1 mg/dL), and triglyceride concentrations (92.0±23.1 mg/dL) were all below the 50th percentile for age. These low plasma lipid concentrations at baseline may in part be explained by the subjects’ usual dietary habits. Based on a 24-hour recall, the average consumption of subjects from each group was approximately 30% of energy from fat (11% monounsaturated fatty acids, 11% SFA, and 8% PUFA) and approximately 105 mg of cholesterol/1000 kcal.
Diet-Induced Changes in Postabsorptive Concentrations of Plasma Lipids and Lipoproteins
The influence of dietary fat saturation in the background diets was as expected. In the group fed the SFA-rich diet, plasma cholesterol concentrations were 15% higher than their entry values after only 8 days (+24.6 mg/dL, P=.004) and remained elevated throughout the rest of the experimental period (Table 3⇓). This increase was primarily in LDL (P=.002). The concentrations of plasma and LDL cholesterol decreased only slightly in the PUFA group and were 7% lower on day 29 than the corresponding values at entry (P=.02 and P=.09, respectively).
Whereas the SFA-rich diet did not affect plasma triglycerides, in individuals fed the PUFA diet for 15 days plasma triglycerides decreased significantly from entry values (−18.4 mg/dL, P=.02) and remained lower thereafter. These differences are important because the magnitude of the postprandial response to a challenge meal is a function of the postabsorptive triglyceride concentration.4 21
Influence of Diets Rich in SFA or PUFA on Postprandial Lipid and Apolipoprotein Responses in Total TRL
Although the postabsorptive concentrations of TRL apoB-48 were 40% lower in the PUFA group than in the SFA group (P=.07) (Fig 3⇓, left panel), the postprandial pattern was similar, with apoB-48 increasing twofold to threefold 3 hours after the meal and returning to near postabsorptive concentrations at 6 hours in both groups. The postprandial response of apoB-48 was closely coupled to that of TRL triglycerides, which also increased twofold to threefold 3 hours after the meal. Unlike apoB-48, TRL triglycerides remained slightly higher than baseline at 6 hours (P<.03), but both components fell below postabsorptive values at 9 hours in both groups. Expressed as percent change from postabsorptive values (Fig 3⇓, right panel), the postprandial increments in TRL triglycerides and apoB-48 at 3 hours were greater in individuals fed the PUFA-rich diet and challenge meal than in those fed the SFA-rich diet and challenge meal (P=.016 and P=.089, respectively). This greater response in the PUFA diet group did not affect the rate of fall of triglycerides and apoB-48 concentrations to or below postabsorptive values.
Whereas fat saturation in the background diet did not affect the apparent rate of clearance of chylomicron remnants, it significantly affected the postprandial response of apoB-100–containing TRL. Concurrent with the increment in apoB-48 and triglycerides, the concentrations of apoB-100 and cholesterol in TRL increased significantly 3 hours after the meal in both diet groups but returned to near postabsorptive values at 6 hours only in individuals fed the PUFA-rich background diet (Fig 3⇑, left panel). In contrast, in those fed the SFA-rich diet the concentrations of apoB-100 and cholesterol were still significantly higher than baseline at 6 hours (P=.027 and P=.01, respectively) and fell below postabsorptive concentrations only 9 hours after the meal, evidently reflecting the accumulation of hepatogenous TRL in individuals fed a diet rich in SFA. In contrast to TRL triglycerides and apoB-48, the percent increase in TRL cholesterol and apoB-100 at 3 hours was similar in the two groups (Fig 3⇑, right panel).
Concentrations of total apoE in plasma were not significantly different in the postabsorptive state in individuals fed the SFA (4.88 mg/dL) and PUFA (4.33 mg/dL) diets and did not change postprandially (data not shown). In both diet groups TRL apoE concentrations increased twofold to threefold 3 hours after the meal (Fig 3⇑). This response was accompanied by a significant and transitory decrease in HDL apoE (calculated as the difference between plasma and TRL apoE) at 3 hours (P=.02; data not shown). As with TRL apoB-100, TRL apoE concentrations 6 hours after the meal were significantly higher than postabsorptive values only in the SFA diet group (P=.04).
The average concentrations of all of the TRL components over the 12-hour postprandial period were higher in the SFA diet group than in the PUFA group: 0.28 and 0.19 mg/dL for apoB-48, P=.08; 11.90 and 6.90 mg/dL for cholesterol, P=.06; 4.99 and 3.18 mg/dL for apoB-100, P=.035; and 0.98 and 0.63 mg/dL for apoE, P=.08.
Influence of Challenge Meals Rich in SFA or PUFA on Postprandial Lipid and Apolipoprotein Responses in Total TRL
Our study design allowed us to determine whether the fatty acid composition of the challenge meal itself affected the postprandial response in individuals fed the SFA diet (Fig 4⇓) or the PUFA diet (Fig 5⇓). In contrast to an earlier report,4 individuals given a PUFA-rich challenge meal had a significantly greater percent increase in TRL triglycerides and a greater percent increase in apoB-48 at 3 hours than those given an SFA challenge meal. This was true for both diet groups (Figs 4⇓ and 5⇓, right panels). These different responses to the PUFA and SFA challenge meals were independent of postabsorptive concentrations of TRL triglycerides and apoB-48, which were identical on days 15 and 29 in individuals fed the SFA background diet (Fig 4⇓, left panel). Although the increases in TRL cholesterol, apoB-100, and apoE at 3 hours showed the same pattern, none of these differences was statistically significant.
Despite the greater postprandial responses of TRL triglycerides and apoB-48 3 hours after the PUFA meal, values still returned to postabsorptive concentrations at 6 hours, as they did after the SFA meal (Figs 4⇑ and 5⇑). This suggests that the rate of clearance of chylomicron remnants was not affected by the fatty acid composition of the challenge meal. Similarly, although the 3-hour increments in TRL apoB-100, cholesterol, and apoE tended to be higher after the PUFA challenge meal than after the SFA meal, the overall response induced by the background diets was not altered. Whether subjects ate a PUFA- or SFA-rich challenge meal, the concentrations of TRL cholesterol, apoB-100, and apoE still returned to postabsorptive values at 6 hours in the PUFA diet group but not in the SFA diet group. Because the fatty acid composition of the challenge meal did not affect the 6-hour postprandial responses, meal data from postprandial studies conducted on day 29 were pooled and analyzed for the total SFA group (n=13) and PUFA group (n=8). As with our findings on day 15, the concentrations of TRL cholesterol, apoB-100, and apoE at 6 hours were significantly higher than postabsorptive values in the SFA diet group (P=.02, .005, and .02, respectively) but not in the PUFA diet group. These consistent findings indicate that the postprandial increase in the concentration of hepatogenous TRL in the plasma is materially influenced by the fatty acid composition of the background diet but not that of the challenge meal.
Postprandial Changes in Total, Free, and Esterified Cholesterol in Plasma, LDL, and HDL
Postprandial increases in triglycerides in total TRL were accompanied by a concurrent redistribution of lipids within plasma, HDL, and LDL. Because the pattern of these changes was similar in both diet groups, the data were pooled and analyzed together (Fig 6⇓). Total cholesterol in plasma decreased slightly at 3 hours and increased significantly above postabsorptive values 12 hours after the challenge meal. These changes were reflected by a decrease in cholesterol in LDL and HDL at 3 hours that exceeded its increase in TRL and, conversely, an increase in LDL and HDL cholesterol above postabsorptive concentrations at 12 hours that exceeded the fall below baseline values in TRL cholesterol. Plasma free cholesterol increased only marginally 12 hours after the meal (+0.77 mg/dL, P=.14). In TRL, the transitory increase in free cholesterol at 3 hours and the decrease below postabsorptive values at 12 hours was accompanied by a proportional fall below postabsorptive concentrations in free cholesterol in LDL and HDL at 3 hours and a rise above postabsorptive values at 12 hours. In contrast, plasma esterified cholesterol fell significantly below postabsorptive concentrations at 3 hours and rose above postabsorptive values at 12 hours. These changes were reflected by a significant decrease in esterified cholesterol in LDL and HDL that exceeded the increase in TRL esterified cholesterol at 3 hours. Conversely, 12 hours after the meal the increase above postabsorptive values in esterified cholesterol in LDL and HDL exceeded its fall below baseline in TRL.
Postprandial Changes in ApoB-48–Containing and ApoB-100–Containing TRL Separated by Immunoaffinity Chromatography
In a subset of the postprandial studies carried out on day 15, apoB-48–containing TRL and 10% to 15% of particles containing apoB-100 were separated from the bulk of TRL by immunoaffinity chromatography.7 22 Bound TRL particles (calculated as the difference between total TRL and unbound TRL) contain the remaining 85% to 90% of TRL apoB-100.
In the postabsorptive state, unbound TRL from individuals in the SFA (n=10) and PUFA diet groups (n=5) contained, respectively, 39% and 43% of total TRL triglycerides, 34% and 28% of TRL cholesterol, 14% and 11% of TRL apoB-100, and 25% and 22% of TRL apoE. None of these differences was statistically significant. Postprandially, lipids and apolipoproteins in both unbound and bound TRL increased significantly. Because the pattern of changes in these particles was similar in both diet groups, data were pooled for analysis. Particles containing apoB-48 and a small fraction of apoB-100 accounted for most of the increase in total TRL lipids 3 hours after the meal, whereas bound TRL apoB-100 accounted for the accumulation of total TRL components after 6 hours (Fig 7⇓). Although significant postprandial changes occurred in both unbound and bound TRL, 60% to 70% of the increase in total TRL triglycerides, cholesterol, and apoE at 3 hours was in the unbound fraction. Six hours after the meal, concentrations of unbound lipids and apolipoproteins were below postabsorptive values, but bound TRL accumulated in the plasma 6 hours, as well as 3 hours, postprandially. These observations indicate that both apoB-48– and apoB-100–containing TRL contributed to postprandial lipemia, but unbound particles containing apoB-48 accounted for most of the early increases in TRL lipids, and bound TRL containing the bulk of apoB-100 accounted for the accumulation of TRL cholesterol, apoB, and apoE during later stages of postprandial lipemia.
In separate analyses, we found that in the PUFA diet group given the PUFA meal the concentrations of unbound TRL components returned to postabsorptive values at 6 hours, whereas in the SFA group given the SFA meal the concentrations of unbound triglycerides, apoB-100, and apoE tended to remain higher than baseline values (P=.07, .05, and .07, respectively) and fell below postabsorptive values only 9 hours after the meal (data not shown).
Correlations Between Postabsorptive and Postprandial Lipoprotein Components
The metabolic pathways of chylomicrons and VLDL are interrelated and closely associated with the metabolism of HDL. The postabsorptive concentration of TRL triglycerides was well correlated with that of TRL apoB-48, apoB-100, apoE, and cholesterol and inversely correlated with that of HDL cholesterol (Table 4⇓). Postabsorptive concentrations of TRL apoB-100 were also inversely related to HDL cholesterol and directly related to TRL apoB-48. Although the postabsorptive concentration of TRL triglycerides exceeded 60 mg/dL in only 3 of the 21 individuals studied, the postabsorptive TRL triglyceride concentration predicted the 3-hour increment in TRL triglycerides; the relationship was significant only in the SFA diet group. No significant relationships were found between postabsorptive TRL triglycerides and 3-hour increments in TRL apoB-48, apoB-100, and cholesterol.
The postprandial increment in TRL triglycerides at 3 hours was well correlated with that of TRL apoB-48, apoE, and cholesterol in both diet groups but was not related to the 3-hour increment in TRL apoB-100 in the SFA diet group. Similarly, the 3-hour increment in TRL apoB-48 was correlated with that of apoB-100 in the PUFA group but not in the SFA group. The postprandial increment in TRL apoB-100 at 3 hours was correlated with that of TRL cholesterol and apoE, but in the SFA diet group these relationships were stronger for the 6-hour increment from baseline. Overall, correlations were stronger in the SFA diet group possibly because there was a limited representation of higher lipid and apolipoprotein values in the PUFA diet group.
In the present experiments and consistent with previous data,23 24 we found that postabsorptive concentrations of plasma and TRL triglycerides were 30% and 40%, respectively, lower in young men consuming a diet rich in PUFA than in men consuming a diet rich in SFA. This triglyceride-lowering effect in the PUFA diet group was accompanied by reduced concentrations of TRL apoB-48, apoB-100, apoE, and cholesterol. The postabsorptive concentration of triglycerides in plasma is an important determinant of the magnitude and duration of the postprandial triglyceride response.25 Therefore, the postprandial response would be expected to be greater in men consuming the diet rich in SFA, as reported by Weintraub et al4 in men given 50 g of fat per square meter of body surface. To the contrary, we found a comparable absolute increase and a significantly greater percent increase in TRL triglycerides at 3 hours in men consuming the PUFA-rich diet and challenge meal. This increased triglyceride response was not significantly related to the postabsorptive concentration of triglycerides (r2=0.34, P=.11) and appeared to be a function of the fatty acid composition of the challenge meal. Thus, as with the PUFA diet group, the percent increase in TRL triglycerides was also greater in the SFA diet group consuming the PUFA meal than in those consuming the SFA meal. This greater response may reflect a more rapid absorption of PUFA. In rats, more rapid esterification of linoleate in the intestinal mucosa has been shown to result in a faster absorption of this fatty acid, relative to palmitate.26 Whether such results apply to humans consuming whole foods, as opposed to meals containing individual fatty acids, is unclear. Regardless of the mechanism, the greater response to the PUFA meal did not affect the rate of return of TRL triglycerides to postabsorptive values that occurred at 6 hours after both meals.
The concentration of TRL apoB-48 increased and peaked 3 hours after the meal and returned to postabsorptive values at 6 hours in both the SFA and PUFA diet groups consuming either an SFA- or PUFA-rich meal. The postprandial increase in TRL apoB-48 at 3 hours presumably reflects primarily the rate of absorption of dietary triglycerides, whereas the return to postabsorptive concentrations is mainly influenced by the rate of clearance of chylomicron remnants from the blood. As with TRL triglycerides, the percent increase in TRL apoB-48 was greater after the PUFA-rich meal than after the SFA-rich meal, but this greater increment did not affect the return to postabsorptive values that occurred at 6 hours in all cases. Taken together, our data suggest that neither the fatty acid composition of the background diet nor that of the challenge meal materially affected the clearance of chylomicron remnant particles postprandially.
Individuals consuming the diet rich in SFA had higher average postprandial concentrations of TRL apoB-48 over 12 hours (0.29 mg/dL) than those consuming a diet rich in PUFA (0.19 mg/dL; P=.08). This is in agreement with the conclusions of Weintraub and associates.4 They found higher concentrations of retinyl esters in postprandial Sf>1000 chylomicron and Sf<1000 chylomicron remnant particles and concluded that SFA increase the concentration of postprandial TRL relative to PUFA. The postprandial pattern of increase and return to postabsorptive values of retinyl esters in Sf<1000, however, was found to be the same as the postprandial pattern in Sf>1000 in both the SFA and PUFA diet groups, from which they concluded that dietary fat saturation does not significantly affect chylomicron remnant clearance. However, chylomicron triglycerides are normally hydrolyzed by lipoprotein lipase within 5 to 10 minutes,27 whereas the removal of chylomicron remnants is saturated for 6 to 7 hours after ingestion of 70 g of cream fat in healthy individuals.28 Therefore, we have considered all particles containing apoB-48 postprandially to be chylomicron remnants. Our findings of higher average concentrations of apoB-48 in the SFA diet group than in the PUFA diet group are also consistent with those of Demacker and associates,5 who found that the 24-hour area under the curve for TRL apoB-48 was 43% higher in their SFA diet group than in their PUFA diet group. However, we found a 40% higher postabsorptive concentration of apoB-48 in subjects consuming the SFA diet, whereas Demacker et al found the postabsorptive concentration of apoB-48 to be similar in both diet groups. Consequently, the apparent concentration of apoB-48 during the rest of the day and night was higher in their SFA diet group, and they concluded that the clearance of chylomicrons was accelerated by a PUFA-rich diet. The basis for the discrepancy with our findings for TRL apoB-48 in the postabsorptive state is unclear but could in part be methodological. Whereas we used a slab gel electrophoretic method that permits accurate quantification of apolipoproteins in TRL samples containing very small amounts of protein, Demacker and associates used a tube gel electrophoretic method to measure the dye absorbance of apoB-48; the validity of this method has recently been questioned.18 Accurate measurement of apoB-48 is particularly critical in the postabsorptive state, during which apoB-48 constitutes only 2% to 5% of the total apoB mass in TRL.18
Although feeding a diet rich in SFA did not affect the postprandial response of apoB-48, it significantly increased the postprandial response of apoB-100, the structural component of hepatogenous TRL. As with apoB-48, the postabsorptive concentration of TRL apoB-100 was 34% higher in individuals consuming a diet rich in SFA (P=.14), and they also accumulated more apoB-100 in TRL postprandially. TRL apoB-100 increased 1.4-fold 3 hours after the meal and returned to postabsorptive values at 6 hours in the PUFA diet group but only at 9 hours in the SFA diet group. The 3-hour postprandial increase in TRL apoB-100 observed here, which is in agreement with earlier reports,6 22 29 has been attributed to preferential hydrolysis of chylomicron remnants by lipoprotein lipase.22 25 29 This is consistent with our finding in the PUFA diet group that the 3-hour increment in TRL apoB-100 was correlated with that of TRL apoB-48, indicating that the accumulation of hepatogenous TRL in this group was closely coupled to that of chylomicron remnants.
The absence of a correlation between the 3-hour increment in TRL apoB-100 and apoB-48 in the SFA diet group suggests that other mechanisms may contribute to the SFA-induced accumulation of hepatogenous TRL 3 and 6 hours after the meal. In individuals fed a diet rich in SFA, the postabsorptive concentration of plasma and LDL cholesterol increased significantly on day 8 and remained higher than baseline (day 1) for the rest of the study. The hypercholesterolemic effect of a high-cholesterol diet rich in SFA has been extensively documented1 8 30 and has been attributed to suppression of LDL receptors,31 which reduces the clearance of VLDL remnants as well as LDL.32 That postprandial chylomicron remnant clearance was apparently unaffected by SFA feeding presumably reflects the participation of other receptors in the hepatic uptake of chylomicron remnants,33 particularly when the LDL receptor is downregulated.34 Indeed, whereas the processing of most VLDL particles involves mainly interaction of these particles with the LDL receptor, the processing of large VLDL as well as postprandial chylomicrons involves other receptors.34 35 The removal by hepatic receptors of the small chylomicron remnants produced in the postabsorptive state may be more dependent on the LDL receptor than large chylomicron remnants produced after fat ingestion.
The fatty acid composition of the challenge meal did not alter the postprandial response of TRL apoB-100. In both the SFA and PUFA diet groups, individuals given a PUFA challenge meal tended to have a higher percent increase in TRL apoB-100 at 3 hours, but regardless of the composition of the challenge meal individuals fed a diet rich in SFA accumulated hepatogenous TRL at 6 hours, whereas those fed a PUFA diet did not. The prolonged postprandial accumulation of hepatogenous TRL thus appears to be affected primarily by the composition of the background diet.
As evidenced by the significant correlations between the 3-hour increment in TRL triglycerides and apoB-48, the postprandial pattern for TRL triglycerides was remarkably parallel to that of apoB-48 in both diet groups. We found that 70% of the increase in total TRL triglycerides 3 hours after the challenge meal was contained in a TRL fraction that contained all of the apoB-48 but less than 15% of the apoB-100. At the same time, TRL particles containing apoB-100 accounted for 70% of the increment in TRL particle number. Similar observations have recently been reported by Schneeman et al22 and Cohn et al,6 who found that 50% to 80% of the increase in TRL triglycerides at 3 hours is associated with particles containing apoB-48. The 30% contribution of the TRL fraction that contained only apoB-100 to the increase in total TRL triglycerides at 3 hours was also significant and consistent with concomitant changes in other components within this fraction (Fig 7⇑).
The unbound fraction containing apoB-48 also accounted for most of the increment in TRL cholesterol at 3 hours. The concomitant decrease in these components in LDL and HDL suggests that unbound TRL acquired cholesterol from higher-density lipoproteins.36 37 An increase in cholesterol in apoB-48 particles has also been reported by Cohn et al6 but not by Schneeman et al,22 who found that more than 90% of the increase in TRL cholesterol was associated with the bulk of TRL apoB-100. Although unbound TRL accounted for most of the 3-hour increment in triglycerides and cholesterol, as well as apoE, these components had returned to near postabsorptive values 6 hours after the meal.
Whereas the early postprandial changes in total TRL triglycerides, cholesterol, and apoE were closely coupled to that of apoB-48, the later changes in TRL cholesterol and apoE were more closely related to that of TRL apoB-100, particularly in individuals fed the SFA diet. In this group, the 6-hour increment in total TRL apoB-100 was highly correlated with that of total TRL cholesterol and apoE. Likewise, in the bound TRL fraction that lacked apoB-48, increments in cholesterol, triglycerides, and apoE at 6 hours were comparable to that of apoB-100. Because the postprandial accumulation of apoB-100 is prolonged in individuals consuming a diet rich in SFA, the increase in the diurnal concentration of apoB-100 associated with that diet is more pronounced than the increase in the postabsorptive concentration. Indeed, over the 12-hour postprandial period studied, the average concentration of apoB-100 was 57% higher in individuals consuming a diet rich in SFA than in those consuming a diet rich in PUFA.
As reported by others,38 39 40 41 the concentration of free and esterified cholesterol in plasma and their distribution among lipoproteins was altered in the postprandial state. These changes were unrelated to the composition of the diet or challenge meal. The postprandial changes in plasma were most pronounced for esterified cholesterol, which fell significantly at 3 hours, reflecting primarily a reduction in LDL and, to a lesser extent, HDL esterified cholesterol. Concurrently, esterified cholesterol increased significantly in TRL and accounted for half of the increase in TRL total cholesterol at 3 hours. These changes may reflect an augmented transfer of esterified cholesterol from higher-density lipoproteins to TRL postprandially.36 37 The rise in plasma and LDL esterified cholesterol above postabsorptive values observed at 12 hours may reflect downregulation of LDL receptors consequent to uptake of TRL cholesterol by the liver.
Postprandial changes in free cholesterol were similar to those of esterified cholesterol but were less pronounced. These changes presumably reflect passive exchange between lipoproteins and may reflect the increased lipoprotein surface area in TRL postprandially. Dubois and associates,40 41 who systematically studied the distribution of free and esterified cholesterol among lipoproteins in young men given challenge meals with varying amounts of fat and cholesterol, also reported a significant and transitory reduction of esterified cholesterol in serum, LDL, and to a lesser extent HDL after a test meal containing 42 g of fat40 41 and 280 mg cholesterol.41 In contrast with our findings, however, they found a significant increment in free cholesterol in serum at 3 hours that was reflected primarily in LDL but also in HDL. The reason for this discrepancy is unclear.
In summary, we have found no evidence that the relative amounts of SFA and omega-6 PUFA in the diet or a challenge meal affect the clearance of chylomicron remnants postprandially in healthy young men. In contrast, the postprandial accumulation of hepatogenous TRL is more pronounced in individuals fed a diet rich in SFA relative to those fed a diet rich in PUFA. The increased concentration of hepatogenous TRL is affected primarily by the fatty acid composition of the background diet but not by that of a single challenge meal. Given the potential atherogenicity of TRL remnants2 42 and the prolonged elevation of hepatogenous TRL that we observed in normolipidemic young men consuming ordinary meals containing amounts of fat typically consumed by Americans, it will be important to determine how dietary fat saturation affects postprandial lipemia in more diverse population groups, including men and women of different ages, as well as individuals with abnormalities in lipoprotein metabolism.
Selected Abbreviations and Acronyms
|P/S ratio||=||total PUFA (g)/total SFA (g)|
|PUFA||=||polyunsaturated fatty acids|
|Sf||=||negative sedimentation svedberg unit|
|SFA||=||saturated fatty acids|
This study was supported by a grant from the National Institutes of Health (HL-14237) and a General Clinical Research Center grant MO1 RR00079-33. N. Bergeron was the recipient of research fellowships from the Heart and Stroke Foundation of Canada, Québec Affiliate (HSFC-R5); the Fonds pour la Formation de Chercheurs et l’Aide à la Recherche, Québec (B3-001); and the Fonds de la Recherche en Santé du Québec (PR-5), Québec. We gratefully acknowledge the expert assistance of K. Todd in the formulation of the research diets. We also thank P. Frost for help in evaluation of subjects, M. Schloetter for help in recruitment and nutritional assessment, V. Weinberg for valuable statistical advice, and the clinical nursing staff of the General Clinical Research Center for their assistance throughout the study.
Grundy SM, Denke MA. Dietary influences on serum lipids and lipoproteins. J Lipid Res.. 1990;31:1149-1172.
Phillips N, Waters D, Havel RJ. Plasma lipoproteins and progression of coronary artery disease evaluated by angiography and clinical events. Circulation. 1993;88:2762-2770.
Weintraub MS, Zecher 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;88:1884-1893.
Cohn JS, Johnson EJ, Millar JS, Cohn SD, Milne RW, Marcel YL, Russell RM, Schaefer EJ. Contribution of apoB-48 and apoB-100 triglyceride-rich lipoproteins (TRL) to postprandial increases in plasma concentration of TRL triglycerides and retinyl esters. J Lipid Res. 1993;34:2033-2040.
Campos E, Nakajima K, Tanaka A, Havel RJ. Properties of an apolipoprotein E-enriched fraction of triglyceride-rich lipoproteins isolated from human blood plasma with a monoclonal antibody to apolipoprotein B-100. J Lipid Res. 1992;33:369-380.
Fielding CJ, Havel RJ, Todd KM, Yeo KE, Schloetter MC, Weinberg V, Frost PH. Effects of dietary cholesterol and fat saturation on plasma lipoproteins in an ethnically diverse population of healthy young men. J Clin Invest.. 1995;95:611-618.
Lipid Research Clinics. Population Studies Data Book. Washington, DC: National Institutes of Health; 1980;1:136. US Dept of Health and Human Services publication 80-1527.
Najjar MF, Rowland M. Anthropometric Reference Data and Prevalence of Overweight: United States, 1976-1980. Hyattsville, Md: US Dept of Health and Human Services, National Center for Health and Statistics; 1987. US Dept of Health and Human Services publication 87-1688.
Harris JA, Benedict FG. A Biometric Study of Basal Metabolism in Man. Washington, DC: Carnegie Institution; 1919.
USDA Nutrient Database for Standard Reference, Agriculture Handbook No. 8. Hyattsville, Md: Human Nutrition Information Service.
Cardin AD, Witt KR, Chao J, Margolius HS, Donaldson VH, Jackson RL. Degradation of apolipoprotein B-100 of human plasma low density lipoproteins by tissue and plasma kallikreins. J Biol Chem. 1984;259:8522-8528.
Havel RJ, Eder H, Bragdon J. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345-1353.
Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem.. 1974;20:470-475.
Bucolo G, David H. Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem. 1973;19:476-482.
Warnick GR, Benderson J, Albers JJ. Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem. 1982;28:1379-1388.
Kotite L, Bergeron N, Havel RJ. Quantification of apolipoproteins B100, B48 and E in human triglyceride-rich lipoproteins. J Lipid Res. 1995;36:890-900.
Havel RJ, Kotite L, Vigne J-L, Kane JP, Tun P, Phillips N, Chen GC. Radioimmunoassay of human arginine-rich apolipoprotein, apoprotein E. J Clin Invest. 1980;66:1351-1362.
Pagnan A, Havel RJ, Kane JP, Kotite L. Characterization of human very low density lipoproteins containing two electrophoretic populations: double prebeta lipoproteinemia and primary dysbetalipoproteinemia. J Lipid Res. 1977;18:613-622.
Schneeman BO, Kotite L, Todd K, Havel RJ. Relationship between the responses of triglyceride-rich lipoproteins in blood plasma containing apolipoproteins B-48 and B-100 to a fat-containing meal in normolipidemic humans. Proc Natl Acad Sci U S A. 1993;90:2069-2073.
Mattson FH, Grundy SM. Comparison of effects of dietary saturated, monounsaturated, and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J Lipid Res. 1985;26:194-202.
Berr F. Characterization of chylomicron remnant clearance by retinyl palmitate label in normal humans. J Lipid Res. 1992;33:915-930.
Karpe F, Steiner G, Olivecrona T, Carlson LA, Hamsten A. Metabolism of triglyceride-rich lipoproteins during alimentary lipemia. J Clin Invest. 1993;91:748-758.
Hayes K, Khosla P. Dietary fatty acid thresholds and cholesterolemia. FASEB J. 1992;6:2600-2607.
Spady DK, Dietschy JM. Dietary saturated triacylglycerols suppress hepatic low density lipoprotein receptors in the hamster. Proc Natl Acad Sci U S A. 1985;82:4526-4530.
Havel RJ. Triglyceride-rich lipoproteins and atherosclerosis: new perspectives. Am J Clin Nutr. 1994;59:795-799.
Havel RJ. Chylomicron remnants: hepatic receptors and metabolism. Curr Opin Lipidol. In press.
Havel RJ, Yamada N, Shames DM. Watanabe heritable hyperlipidemic rabbit. Arteriosclerosis. 1989;9(suppl I):I-33-I-38.
Castro GR, Fielding CJ. Effects of postprandial lipemia on plasma cholesterol metabolism. J Clin Invest. 1985;75:874-882.
Tall A, Sammett D, Granot E. Mechanisms of enhanced cholesteryl ester transfer from high density lipoproteins to apolipoprotein B-containing lipoproteins during alimentary lipemia. J Clin Invest. 1986;77:1163-1172.
Cohn JS, McNamara JR, Cohn SD, Ordovas SD, Schaefer EJ. Postprandial plasma lipoprotein changes in human subjects of different ages. J Lipid Res. 1988;29:469-479.
Dullaart RPF, Groener EM, van Wijk H, Sluiter WJ, Erkelens DW. Alimentary lipemia-induced redistribution of cholesteryl ester between lipoproteins: studies in normolipidemic, combined hyperlipidemic, and hypercholesterolemic men. Arteriosclerosis. 1989;9:614-622.
Dubois C, Armand M, Azaisbraesco V, Portugal H, Pauli AM, Bernard PM, Latge C, Lafont H, Borel F, Lairon D. Effects of moderate amounts of emulsified dietary fat on postprandial lipemia and lipoproteins in normolipidemic adults. Am J Clin Nutr. 1994;60:374-382.
Dubois C, Armand M, Mekki A, Portugal H, Pauli AM, Bernard PM, Lafont H, Lairon D. Effects of increasing amounts of dietary cholesterol on postprandial lipemia and lipoproteins in human subjects. J Lipid Res. 1994;35:1993-2007.