Definition and importance of postprandial lipaemia
Much of our knowledge about the relationship between lipid, lipoprotein metabolism and the development of atherosclerosis and CVD is based on measurements in the fasting state essentially reflecting endogenous metabolism (Fig. 1). Although such measurements remain the foundation of clinical assessment and an important basis for decisions regarding hypolipidaemic interventions, it should be acknowledged that we spend a considerable amount of time in a non-fasting, postprandial state. Based on typical Western eating patterns, most people consume three or more meals a day, each containing 20–70 g fatReference Cohn, McNamara, Cohn, Ordovas and Schaefer1. Except at breakfast, each of these meals is most likely consumed before plasma triacylglycerol (TAG) levels have returned to baseline from the lipaemic conditions resulting from the previous intake. Thus, people spend the majority of their daytime in a postprandial (fed) state, with a continual fluctuation in the degree of lipaemia throughout the day.
Fig. 1 Human lipoprotein metabolism. Dietary free fatty acids (FFA) are absorbed from the gut and converted to triacylglycerols to be incorporated into chylomicrons in the intestinal epithelial cells. The triglyceride-rich apo B-48-containing chylomicrons enter the plasma via the intestinal lymph. Lipoprotein lipase (LPL) hydrolyses the triacylglycerol in chylomicrons to fatty acids, which are taken up by muscle cells for oxidation or adipocytes for storage. The remaining particles, the chylomicron remnants, are removed from the circulation by the liver through binding of their surface apo E to the LDL receptor or LDL receptor-related protein. VLDL particles are triacylglycerol-rich apo B-100-containing particles, synthesised by the liver. As with chylomicrons, VLDL triacylglycerols are hydrolysed by LPL. VLDL remnants or IDL are taken up by liver receptors via apo E or converted to LDL. Chylomicrons, VLDL and their respective remnants (remnant lipoproteins) are termed triacylglycerol-rich lipoproteins (TRL). Under physiological conditions, insulin, which is raised in the postprandial state, suppresses lipolysis from adipose tissue and hepatic VLDL production, but this insulin action is inappropriate in insulin resistance and type 2 diabetes, resulting in high TRL concentrations. The large amount of TRL and their prolonged residence time in the circulation increase the exchange of esterified cholesterol from HDL and LDL to TRL, and of triacylglycerols to LDL and HDL particles, which is mediated by cholesterol-ester transfer protein. Triacylglycerol enrichment of LDL particles renders them better substrates for hepatic lipase, which hydrolyses triacylglycerols from the core of LDL and turns them into smaller and denser particles. Small, dense LDL are more atherogenic as they readily enter the subendothelial space and become oxidised. Triacylglycerol-enriched HDL particles are smaller and are more rapidly catabolised, which may explain the observed low plasma HDL in insulin resistance and type 2 diabetes.
The postprandial state is a dynamic, non-steady-state condition, with rapid remodelling of lipoproteins compared with the relatively stable fasting condition (Fig. 1). Determination of the postprandial response is complex, and it is therefore more challenging to assess the cardiovascular risk associated with postprandial lipaemia than that during fasting conditions. In spite of this, it is becoming increasingly evident that future efforts to study and treat lipids related to atherogenesis should include postprandial parameters. The aim of this paper is to consider the regulatory pathways of postprandial lipoproteins and the major factors, including nutrition, lifestyle, physiopathology and genetics, that may contribute to interindividual variability in postprandial lipaemia, and thereby susceptibility to atherosclerosis.
Experimental evidence linking postprandial lipaemia with atherosclerosis
The potential atherogenicity of postprandial TAG and TAG-rich lipoprotein (TRL) levels did not gain widespread attention until the idea was put forward in a widely quoted paper by Zilversmit in 1979Reference Zilversmit2, who proposed that the hydrolysis of chylomicron by lipoprotein lipase (LPL) resulted in the subsequent internalisation of cholesterol ester-enriched chylomicron remnants by arterial smooth muscle cells. A confirmation of this hypothesis has been complicated by the multiple factors affecting the postprandial response, the lack of standardised methodology and the considerable heterogeneity between postprandial TRL species. Evidence supporting an association between postprandial lipaemia and atherosclerosis has been provided by clinical trials and mechanistic studies of both the direct and indirect effects of TRL using animal models and cell culture.
Clinical trials
Several clinical studies have shown that a delayed elimination of postprandial TRL is associated with atherosclerosis (Tables 1 and 2). There are also reports of an association between postprandial lipaemic response and subsequent progression of atherosclerosis in patients with pre-existing CHD.
Table 1 Clinical trials summarizing the effect of postprandial lipoprotein metabolism on coronary artery disease (CAD)
Table 2 Clinical trials summarising the effect of postprandial lipoprotein metabolism on carotid artery atherosclerosis
IMT, intima–media thickness.
In men, the presence of CHD is associated with higher postprandial TAG concentrations in plasma compared with healthy controls, even after correction for higher levels of fasting TAG in the CHD groupReference Groot, van Stiphout and Krauss3–Reference Meyer, Westerveld, de Ruyter-Meijstek, van Greevenbroek, Rienks, van Rijn, Erkelens and de Bruin6. The data are less clear for women. One smaller study reported elevated postprandial TAG and apo B-48 concentrations in women with CHDReference Meyer, Westerveld, de Ruyter-Meijstek, van Greevenbroek, Rienks, van Rijn, Erkelens and de Bruin6. However, a larger study showed no significant relationship between prolonged postprandial lipaemia and CHD in middle-aged womenReference Ginsberg, Jones, Blaner, Thomas, Karmally, Fields, Blood and Begg7. In a number of studies, carotid intima–media thickness is used as a surrogate marker for atherosclerosisReference Ryu, Howard, Craven, Bond, Hagaman and Crouse8–Reference Boquist, Ruotolo, Tang, Bjorkegren, Bond, de Faire, Karpe and Hamsten10. Several studies have confirmed a positive association between carotid intima–media thickness and postprandial lipaemiaReference Ryu, Howard, Craven, Bond, Hagaman and Crouse8–Reference Hamsten, Silveira, Boquist, Tang, Bond, de Faire and Bjorkegren11. However, these data do not address the issue of whether prolonged postprandial lipaemia predicts risk of developing CHD or whether the presence of CHD results in a subsequent impairment of postprandial TRL.
In order to address this question, one cross-sectional study examined postprandial TAG levels after the consumption of a high-fat liquid drink in the healthy sons of men with angiographic evidence of severe CHD compared with the sons of control subjects without CHDReference Uiterwaal, Grobbee, Witteman, van Stiphout, Krauss, Havekes, de Bruijn, van Tol and Hofman12. In spite of comparable fasting lipids between groups, the sons of men with CHD had significantly higher plasma TAG levels after 8, 10 and 12 h postprandially, indicative of a delayed clearance of TAG. In another study in the offspring of patients with CHD, young men with (case subjects) or without (control subjects) a paternal history of CHD underwent a postprandial study. Although no difference in postprandial TAG was found in the groups as whole, subgroup analysis revealed an increased postprandial response among individuals with a moderate elevation of fasting TAG levelReference Tiret, Gerdes, Murphy, Dallongeville, Nicaud, O'Reilly, Beisiegel and De Backer13. There is evidence that higher levels of TRL or their remnants predict the progression of disease in subjects with established CHD. In The Montreal Heart Study, undertaken in 335 men and women with moderate-to-extensive CHD, the concentration of hepatic TRL remnants predicted the progression of atherosclerosisReference Phillips, Waters and Havel14. In a summary of clinical studies of postprandial lipaemia and atherosclerosis, KarpeReference Karpe15 suggested that an elevated plasma TAG measured at late postprandial time points after fat intake might reveal a state of fat intolerance linked to an elevated risk of CHD that was under genetic control and could not be detected by a simple measurement of fasting plasma TAG. However, additional studies are needed to determine the effect of specific TRL fractions and the underlying mechanisms for a link between postprandial lipaemia and atherosclerosis.
Mechanistic evidence
The pathogenesis of the relationship between postprandial TRL and CHD remains unclear, but experimental evidence has provided several plausible mechanisms. Atherogenic effects may be mediated directly by TRL particles or components of the particles. A variety of in vitro and clinical studies suggest that postprandial chylomicrons and VLDL are associated with adverse effects on the arterial endothelium (Fig. 2). In cell culture studies, TRL, particularly postprandial remnants, is directly cytotoxic to endothelial cellsReference Speidel, Booyse, Abrams, Moore and Chung16. Clinical evidence also demonstrates that postprandial TRL adversely affects the endothelium by mediating changes in vascular tone. After the consumption of a high-fat meal, a reduction in flow-induced dilation of the brachial artery correlated with postprandial plasma TAG concentration in healthy subjectsReference Vogel, Corretti and Plotnick17. Incubation with remnant lipoproteins, but not VLDL or LDL, induced an elevation in the expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1 and tissue factor in a human umbilical vein endothelial cell model, in part through a redox-sensitive mechanismReference Doi, Kugiyama, Oka, Sugiyama, Ogata, Koide, Nakamura and Yasue18. In addition, indirect mechanisms of TRL atherogenicity may be related to metabolic changes associated with the presence of postprandial TRL. Of particular interest is TRL-mediated modification in LDL composition and size with the generation of small, dense LDLReference Griffin19.
Fig. 2 The effects of postprandial chylomicrons and VLDL on arterial endothelium. VLDL remnants and chylomicron remnants behave in much the same way as LDL. They enter the subendothelial space, where they become modified, and the modified remnants stimulate Chemoattractant protein-1 (MCP-1), promote the differentiation of monocytes into macrophages and are taken up by the macrophages to form foam cells. Like LDL, the remnant lipoproteins are proinflammatory and proatherogenic.
The composition and cholesterol concentration of HDL is inversely related to the magnitude of postprandial lipaemia and the plasma concentration of TAG. Postprandial lipaemia has been shown to be associated with changes in haemostatic variables known to promote the risk of thrombotic eventsReference Miller20. Following the intake of a fat-rich meal, factor VIIc is transiently increased due to an increase in plasma concentration of factor VIIaReference Miller20. A postprandial decline in plasminogen activator inhibitor type-1 activity and an increase in tissue plasminogen activator activity have been observed in various studiesReference Sanders, Oakley, Cooper and Miller21, Reference Tholstrup, Miller, Bysted and Sandstrom22. Finally, postprandial lipaemia is associated with a mild increase in platelet reactivity that increases the expression of cell-surface markers in healthy menReference Broijersen, Karpe, Hamsten, Goodall and Hjemdahl23, Reference Hyson, Paglieroni, Wun and Rutledge24.
Factors affecting the postprandial response
Meal size and composition
Postprandial lipaemia is influenced by the amount and type of dietary fat present in the test meal, as well as other dietary components, including fibre, glucose, starch and alcoholReference Cohen and Berger25–Reference Williams27. The amount and type of dietary fat modulate postprandial lipaemia. The intake of long-chain omega (n)-3 PUFA (predominantly fish oil), results in lower TAG levels and attenuates postprandial lipaemiaReference Tinker, Parks, Behr, Schneeman and Davis28.
Fat
The amount of fat required to result in significant elevations in plasma TAG concentration is in the order of 30–50 g. Some studies have been performed with increasing doses of dietary fatReference Cohen, Noakes and Benade29–Reference Dubois, Beaumier, Juhel, Armand, Portugal, Pauli, Borel, Latge and Lairon32. In these studies, a very low (5 g) or low (15 g) dose of dietary fat does not significantly increase triacylglycerolaemia postprandially; moderate doses (30–50 g) dose-dependently increase postprandial triacylglycerolaemia (from 0·9 to 1·3 mmol/l above baseline, respectively); and finally, very high doses (80 g and above) exaggerate postprandial triacylglycerolaemia but without dose-dependence. On the other hand, consecutive meals containing fat appear to enhance the lipaemiaReference Jackson, Robertson, Fielding, Frayn and Williams33. Most meals contain 20–40 g fat, so that when two or three such meals are eaten consecutively, along with snacks eaten between meals, this pattern of consumption might be expected to maintain circulating TAG well above fasting concentrations for much of the day. Studies that have monitored TAG responses overnight following a fat-containing evening meal have shown values to be elevated for 7–8 h after the meal, only falling towards fasting values between 04·00 and 06·00 hoursReference Williams, Moore, Morgan and Wright34.
Studies have also addressed fatty acid composition. A relatively large number of acute meal studies have attempted to compare the effects on postprandial lipaemia of meals of different fat type (reviewed by Williams, 1998)Reference Williams27, Reference Roche, Zampelas, Jackson, Williams and Gibney35–Reference Mekki, Charbonnier, Borel, Leonardi, Juhel, Portugal and Lairon37. A number of potentially confounding factors, such as amount of fat, type and amount of carbohydrate and physicochemical composition of the mealReference Sakr, Attia, Haourigui, Paul, Soni, Vacher and Girard-Globa38, Reference Armand, Pasquier and Andre39, have differed between many of the studies, which makes comparison difficult, and clear conclusions cannot always be drawn. In this respect, it should be remembered that short or medium dietary fatty acids have a limited effect on postprandial plasma TAG response because they enter the portal route instead of chylomicron secretion into the general circulation. Dairy fats contain significant amounts of short- and medium-chain fatty acids so that studies that have used dairy fats as the source of saturated fatty acids (SFA) generally report, as expected, a lower postprandial TAG response than with other SFA or other types of fat. Taking account of these considerations, most studies have shown that meals enriched with SFA, MUFA or n-6 PUFA do not generally elicit markedly different postprandial lipid responses. However, some studies report exacerbatedReference Thomsen, Rasmussen, Lousen, Holst, Fenselau, Schrezenmeir and Hermansen36 or reducedReference Mekki, Charbonnier, Borel, Leonardi, Juhel, Portugal and Lairon37 responses after an intake of saturated butter fat.
Comparisons of the effect of n-6 PUFA-rich oils with olive oil (rich in oleic acid) or MUFA (rapeseed oil) have shown a lowerReference Cabezas, de Bruin, Jansen, Kock, Kortlandt and Erkelens40 or comparable magnitude of postprandial lipaemiaReference Mekki, Charbonnier, Borel, Leonardi, Juhel, Portugal and Lairon37, Reference Lichtenstein, Ausman, Carrasco, Jenner, Gualtieri, Goldin, Ordovas and Schaefer41, Reference Tholstrup, Sandstrom, Bysted and Holmer42. Eating n-3 PUFA (fish oil) can lower the postprandial TAG response if a sufficient amount is present within the test mealReference Zampelas, Peel, Gould, Wright and Williams43, but the levels used were far greater than those which would be consumed by most populations. Furthermore, several studies have shown that differences in single-meal fatty acid composition exert little or no effect on postprandial changes in plasma lipidsReference Cabezas, de Bruin, Jansen, Kock, Kortlandt and Erkelens40, Reference Jackson, Zampelas, Knapper, Culverwell, Wright, Gould and Williams44–Reference Burdge, Powell and Calder46. The influence of the positional distribution of fatty acids within the dietary TAG moieties has been investigated, some showing some influenceReference Jensen, Christensen and Hoy47 but others no effectReference Zampelas, Peel, Gould, Wright and Williams43 on postprandial lipaemia.
Several studies have found striking findings with regard to the effect of stearic acid-rich fats compared with other SFA on postprandial lipaemia. Two independent studies have found that a stearic acid-rich TAG prepared from a randomised blend of hydrogenated and unhydrogenated high-oleic sunflower oil resulted in decreased postprandial lipaemia compared with unhydrogenated high-oleic acid sunflower oilReference Tholstrup, Miller, Bysted and Sandstrom22, Reference Sanders, de Grassi, Miller and Morrissey48. However, a stearic acid-rich meal using cocoa butter resulted in similar postprandial lipaemia compared with a meal rich in oleate provided by high-oleic sunflowerReference Sanders, Oakley, Cooper and Miller21. Yet in the same study, a synthetic randomised stearic-rich TAG was found to decrease postprandial lipaemia. It was hypothesised that the unique symmetrical TAG structure of cocoa butter, in which almost all of the stearic acid is present as 1,3 di-stearyl, 2 mono-oleylglycerol, was responsible for this difference. In order to test this hypothesis, the postprandial effects of randomised cocoa butter were compared with unrandomised cocoa butterReference Sanders, Berry and Miller49. It currently remains uncertain whether these effects are solely due to TAG structure or are related to changes in the physicochemical properties of the TAG mixture.
Measurement of the postprandial TAG response may provide only a limited evaluation of the true impact of meal fat type on postprandial lipoprotein metabolism.
More recently, studies that have measured particle number (evaluated by apo B-48 response), and which have measured responses in different lipoprotein subfractions, have revealed important differences in lipid, apo, particle size and number in response to meals of different fatty acid composition. The studies showed lipaemic responses to be in the order SFA>MUFA>PUFA. This suggests that findings from studies that have employed plasma TAG analysis alone may have underestimated the impact of meal fatty acid composition on postprandial lipoprotein metabolismReference Jackson, Robertson, Fielding, Frayn and Williams33, Reference Jackson, Robertson, Fielding, Frayn and Williams50, Reference Jackson, Wolstencroft, Bateman, Yaqoob and Williams51. Meals containing olive oil, with oleic acid, result in a greater apo B-48 response compared with palm oil, safflower oil, and a mixture of fish and safflower oil, and it stimulated the formation of both small (S[f] 60–400) and large (S[f]>400) apo B-48-containing lipoproteinsReference Jackson, Robertson, Fielding, Frayn and Williams50. This finding is consistent with data from Caco-2 cell culture studiesReference Black, Roche, Tully and Gibney52, which demonstrated that oleic acid is a potent stimulator of TAG secretion, and consistent with other test-meal studies reporting that meals high in oleic acid-rich oils (e.g. high-oleic sunflower oil) result in a more pronounced and sharper postprandial rise in plasma TAG than -s seen with SFA-rich mealsReference Sanders, de Grassi, Miller and Morrissey48, although the overall TAG response measured as area under the curve does not differ from other fat type meals.
Because it is unclear exactly how postprandial lipaemia impacts on atherosclerosis and CHD risk, the relevance of reported differences in the pattern and timing of the TAG response, as well as in particle number and particle size, elicited by meal fat type, is not yet fully understood. However, it is generally agreed that an elevated TAG response that continues into the late postprandial phase (5–8 h) is unfavourableReference Patsch, Miesenbock, Hopferwieser, Muhlberger, Knapp, Dunn, Gotto and Patsch4; such a response is most commonly observed when non-dairy SFA meals are fedReference Sanders, de Grassi, Miller and Morrissey48, Reference Roche, Zampelas and Knapper53.
The habitual diet of an individual may also influence the postprandial responseReference Williams27, but far fewer data have been published on this aspectReference Williams, Moore, Morgan and Wright34, Reference Roche, Zampelas, Jackson, Williams and Gibney35, Reference Roche, Zampelas and Knapper53–Reference Zampelas, Roche and Knapper55. Background tested diets rich in MUFA or n-6 PUFA tend to lower the postprandial lipid response compared with SFAReference Zampelas, Peel, Gould, Wright and Williams43, Reference Weintraub, Zechner, Brown, Eisenberg and Breslow54, Reference Zampelas, Roche and Knapper55. Compared with an SFA-rich diet, an increase in chronic MUFA intake led to a marked reduction (by 21–54 %) in apo B-48 production following the test meal, but postprandial lipaemia did not differ, which indicated that MUFA diets results in the production of chylomicrons of a larger sizeReference Silva, Kelly, Jones, Smith, Wootton, Miller and Williams56, suggested to be an advantage in the postprandial processing of dietary TAG. However, Rivellese et al. could find no difference in postprandial lipaemia on administration of a diet high in MUFA compared with diets high in SFAReference Rivellese, Maffettone, Vessby, Uusitupa, Hermansen, Berglund, Louheranta, Meyer and Riccardi57. On the other hand, postprandial lipaemia has been shown to be greater on high-oleic acid and trans-18 : 1 diets compared with a high-carbohydrate dietReference Sanders, Oakley, Crook, Cooper and Miller58. Comparisons of the effect of n-6 PUFA-rich oils with olive oil (rich in n-9 MUFA) or MUFA showed lower or comparable magnitude of postprandial lipaemiaReference Mekki, Charbonnier, Borel, Leonardi, Juhel, Portugal and Lairon37, Reference Lichtenstein, Ausman, Carrasco, Jenner, Gualtieri, Goldin, Ordovas and Schaefer41, Reference Tholstrup, Sandstrom, Bysted and Holmer42.
It is well documented that diets rich in long-chain n-3 PUFA can lower the postprandial TAG response as long as a high intake (2·7–4 g/d) are givenReference Williams, Moore, Morgan and Wright34, but some opposite effects have also been foundReference Roche and Gibney59. In several studies, LPL activity is increased by supplementation with 3–4 g/d long-chain n-3 PUFAReference Rivellese, Maffettone, Vessby, Uusitupa, Hermansen, Berglund, Louheranta, Meyer and Riccardi57, Reference Khan, Minihane, Talmud, Wright, Murphy, Williams and Griffin60, Reference Park and Harris61. In contrast, the consumption of a diet rich in α-linolenic acid (18 : 3n-3) containing an intake of between 4·5 and 9·5 g/d taken as margarine for 6 months had no effect on postprandial lipaemiaReference Finnegan, Minihane, Leigh-Firbank, Kew, Meijer, Muggli, Calder and Williams62. There is abundant evidence indicating that the reduction in postprandial lipaemia following n-3 PUFA supplementation is due to a decrease in chylomicronReference Harris and Muzio63, Reference Westphal, Orth, Ambrosch, Osmundsen and Luley64 and VLDL-TAGReference Westphal, Orth, Ambrosch, Osmundsen and Luley64–Reference Nozaki, Garg, Vega and Grundy66 synthesis/secretion. On the other hand, there is also limited evidence supporting the hypothesis that the reduction in postprandial lipaemia following n-3 PUFA supplementation is due to an increased rate of TAG clearance associated with increased endogenous (measured in non-heparinised plasma) LPL activityReference Park and Harris61, Reference Harris, Lu, Rambjor, Walen, Ontko, Cheng and Windsor67. A logical conclusion from the above studies would be that both a decrease in chylomicron and VLDL-TAG secretion/synthesis, along with an increased clearance rate, was responsible for the postprandial reductions in lipaemia following n-3 PUFA supplementation.
Overall, studies that have evaluated impact of habitual fat type on postprandial response to acute fat ingestion have shown that, in terms of postprandial TAG response, effects are in the order SFA>MUFA = n-6 PUFA>n-3 PUFA.
Carbohydrate
Clinical studies support the concept that diets rich in highly digestible carbohydrate can lead to elevation in fasting plasma TAG as a result of hepatic VLDL and chylomicron remnants accumulation due to altered lipoprotein secretion and/or clearance, as reviewedReference Parks, Krauss, Christiansen, Neese and Hellerstein68, Reference Roche69. Also, several studies have shown that the amount or nature of carbohydrate in a meal alter postprandial lipid metabolism. Data obtained after the addition of glucose (50 g, 100 g) to fatty test meals have not shown consistent findings in healthy subjectsReference Cohen and Berger25, whereas the addition of sucrose or fructose has consistently been shown to increase postprandial triacylglycerolaemiaReference Grant, Marais and Dhansay70. In healthy subjects, physiological ranges of postprandial hyperglycaemia and hyperinsulinaemia as generated by starchy foods (white bread, pasta, beans) do not induce noticeable alterations in the overall postprandial TAG responseReference Harbis, Defoort and Narbonne71. Furthermore, the data obtained in this study showed that portal and peripheral hyperinsulinism (modulated using different test meals) delays and exacerbates the postprandial accumulation of intestinally derived chylomicrons in plasmaReference Harbis, Defoort and Narbonne71. Moreover, in subjects with insulin resistance, the ingestion of a high-glycaemic index mixed meal, compared with a low-glycaemic index one, increases the postprandial rise in insulinaemia and the accumulation of apo B-100- and apo B-48-containing TRL in those subjects, thus increasing postprandial triacylglycerolaemia as well as modifying the kinetics of peak occurrenceReference Harbis, Perdreau and Vincent-Baudry72. Adding various digestible carbohydrates to a test meal can elicit a biphasic response of postprandial lipaemiaReference Harbis, Perdreau and Vincent-Baudry72. This indicates clearly that the amount as well as the nature of carbohydrate can influence postprandial lipid responses.
Fibre
The addition of some dietary fibre sources into mixed test mealsReference Cara, Dubois, Borel, Armand, Senft, Portugal, Pauli, Bernard and Lairon73, Reference Lia, Andersson, Mekki, Juhel, Senft and Lairon74 at the level of 4–10 g/ meal can moderately reduce postprandial triacylglycerolaemia (by 17–24 %) or chylomicron lipids as generated by a mixed meal. Sources of soluble viscous fibre (e.g. oat bran) or with hypotriacylglycerolaemic properties (e.g. wheat germ) were shown to display a delay in the micellar lipid solubilisation process and a consequent reduction (by 37 %) in the secretion of chylomicrons into the circulationReference Lia, Andersson, Mekki, Juhel, Senft and Lairon74. These data suggest that soluble fibre reduces the rate of digestion of dietary fats and thereby attenuates the postprandial lipaemic response.
Protein
Very little information is available so far regarding the influence of the amount or nature of dietary proteins on postprandial lipid responses. There is, however, evidence indicating that a diet of 20 g soy protein isolate for 3 weeks reduces baseline plasma remnant-like particle-cholesterol levels by 9·8 %Reference Higashi, Abata, Iwamoto, Ogura, Yamashita, Ishikawa, Ohslzu and Nakamura75. Recent studies have shown that postprandial lipaemia can be acutely mitigated when proteins are added to the fatty mealReference Westphal, Taneva, Kastner, Martens-Lobenhoffer, Bode-Boger, Kropf, Dierkes and Luley76. By contrast, a low-protein diet exacerbates the postprandial chylomicron concentration in moderately dyslipidaemic subjects in comparison to a lean red meat protein-enriched dietReference Mamo, James, Soares, Griffiths, Purcell and Schwenke77. Casein added to a fatty meal markedly lowers NEFA in the postprandial and postabsorption phases, and also moderately reduces (by 20 %) postprandial lipaemiaReference Westphal, Orth, Ambrosch, Osmundsen and Luley64.
Lifestyle conditions
Physical activity
An acute bout of aerobic exercise has been shown to significantly reduce postprandial lipaemia by 24–35 %Reference Hardman and Aldred78–Reference Thomas, Horner, Langdon, Zhang, Krul, Sun and Cox82 and to significantly increase LPL activityReference Sady, Thompson, Cullinane, Kantor, Domagala and Herbert83–Reference Zhang, Smith, Langdon, Messimer, Sun, Cox, James-Kracke and Thomas85. The mitigation of the lipaemic response to a meal high in fat and carbohydrate is related to the intensity and/or energy expenditure of the preceding exerciseReference Tsetsonis and Hardman80. Physical activity within the 24 h preceding a high-fat meal greatly improves the rate at which lipids are removed from the circulation. In a meta-analysis of data from interventions involving exercise, it was estimated that there was a reduction of 0·5 sd in the postprandial TAG response in groups that had taken exercise before ingesting a mealReference Petitt and Cureton86. Furthermore, the postprandial response to an oral fat load is lower, and the clearance rates of TRL are higher, in endurance-trained individuals compared with untrained control subjects, although this may not be applicable to moderate exerciseReference Gill, Mees, Frayn and Hardman87. In a recent article, the combination of exercise and n-3 PUFA supplementation reduced postprandial lipaemia response, measured as the incremental area under the postprandial curve of TAG, to a greater degree in recreationally active males when compared with the two treatments individuallyReference Smith, Sun, Donahue and Thomas88.
Smoking
Axelson et al. Reference Axelsen, Eliasson, Joheim, Lenner, Taskinen and Smith89 showed a 50 % greater TAG postprandial increase in habitual smokers without changes in fasting TAG. Mero et al. Reference Mero, Syvanne, Eliasson, Smith and Taskinen90 showed that smoking raised retinyl esters and apo B-48 (by 170 %), but not apo B-100. Data obtained in a large sample of men and women support the interpretation of Axelson et al. that smoking affects postprandial TAG metabolism primarily by raising lipoproteins of intestinal origin because cigarette smokers had substantially greater postprandial retinyl palmitate and apo B-48 (by 114–259 %) responses than did non-smokers, when adjusted for fasting triacylglycerolsReference Sharrett, Heiss, Chambless, Boerwinkle, Coady, Folsom and Patsch91.
Alcohol
The effect of alcohol on postprandial lipids has drawn continued attention over the past 10 years. Clearly, ethanol consumed with a meal elevates total plasma and VLDL-TAG. In a recent studyReference Fielding, Reid, Grady, Humphreys, Evans and Frayn92, the addition of 47·5 g alcohol to a high-fat meal (54 % of energy) was associated with an approximately 60 % increase in the peak plasma TAG concentration compared with a meal consumed without alcohol. The authors attributed this increase to a stimulation of large VLDL secretion. Ethanol has also been shown to increase fatty acid synthesisReference Siler, Neese, Parks and Hellerstein93 and also to reduce TAG clearance from the plasmaReference Pownall, Ballantyne, Kimball, Simpson, Yeshurun and Gotto94.
Physiological factors
Age
In general, tolerance to oral fat intake decreases with ageReference Cohn, McNamara, Cohn, Ordovas and Schaefer1. Information on postprandial lipaemia in children is sparse, but, interestingly, there has been shown to be a significant difference in postprandial response between children and their mothers in spite of similar baseline TAG levelsReference Couch, Isasi and Karmally95. There also appears to be an age-related change in postprandial lipaemia and LPL activityReference Jackson, Knapper-Francis, Morgan, Webb, Zampelas and Williams96, which may in part be attributable to weight gain.
Gender and menopausal status
A number of studies have demonstrated significant differences between fasting and postprandial TAG levels in men and women, with higher levels in menReference Cohn, McNamara, Cohn, Ordovas and Schaefer1. It is noteworthy that, for a given meal, the postprandial lipid response is lower in women than men, due to a higher clearance capacity caused by an increase in LPL activity.
Additional evidence for the presence of exaggerated postprandial lipaemia in postmenopausal women has been reported after adjusting by fasting TAGReference van Beek, de Ruijter-Heijstek, Erkelens and de Bruin97. There are several possible mechanisms that might promote the uptake of fat in women. Oestradiol probably promotes a rapid clearance of chylomicron remnants through its effects on the LDL receptor, but it may also promote more effective fatty acid trapping by subcutaneous adipose stores. It is noteworthy that, for a given meal, the postprandial lipid response is lower in women than men, due to a higher clearance capacity caused by an increase in LPL activity. On the other hand, hormone replacement therapy is associated with an increase in TAG in parallel with a decrease in remnant lipoprotein-cholesterol levelsReference Ossewaarde, Dallinga-Thie, Bots, van der Schouw, Rabelink, Grobbee and Westerveld98. These results suggest that oestrogen might induce a shift in the distribution pattern of TRL, with a decrease of the more atherogenic fractions.
Pathological conditions
Obesity
Obesity, especially central adiposity, is frequently associated with several metabolic abnormalities, including hypertriacylglycerolaemia and hyperinsulinaemia, which would predict an exaggerated postprandial lipid response. However, even in the absence of these associated conditions, obese individuals may have up to three times higher postprandial TAG levels than non-obese control patientsReference Lewis, O'Meara, Soltys, Blackman, Iverius, Druetzler, Getz and Polonsky99–Reference Mekki, Christofilis and Charbonnier102, and an abnormal postprandial lipid pattern is a trait of abdominal obesity even without fasting hypertriacylglycerolaemiaReference Mekki, Christofilis and Charbonnier102. In a postprandial study of non-obese and obese subjects, Goldberg et al. Reference Goldberg, Vanni-Reyes, Ramakrishnan, Holleran and Ginsberg103 reported a significant correlation between LPL activity and the postprandial TAG response only among the non-obese subjects. Obesity is associated with an exaggerated postprandial lipaemia, but a 10 kg weight decrease led to a reduction in chylomicron size but not in the number of particlesReference James, Watts, Barrett, Smith, Pal, Chan and Mamo104.
Hypertriacylglycerolaemia
Subjects with fasting hypertriacylglycerolaemia are known to display exaggerated and prolonged postprandial lipid responses, with a fourfold increase in the half-life of circulating TRL, particularly those of intestinal origin, possibly due to a reduction in LPL activity. Elevated fasting TAG, by enhancing circulating VLDL particle concentration and thus promoting abnormal TRL accumulation postprandially, is the most important and common condition affecting postprandial lipaemia.
Insulin resistance
The insulin-resistant state is associated with a cluster of abnormalities in glucose and lipid homeostasis, including elevated levels of plasma fasting TAG, low plasma concentrations of HDL-cholesterol and an increased prevalence of small, dense LDLReference Reaven105. Metabolic defects include impaired NEFA metabolism, a saturation of the removal of TRL remnants and an increased hepatic secretion of VLDL particlesReference Taskinen106. In several studies, the degree of insulin sensitivity was a determinant of the postprandial lipaemic response among healthy adults independently of body mass index, waist-to-hip ratio, blood glucose level and the insulin effect on fasting plasma TAGReference Jeppesen, Hollenbeck, Zhou, Coulston, Jones, Chen and Reaven107, Reference Boquist, Hamsten, Karpe and Ruotolo108. In women and men, fasting insulin levels have been correlated with the degree of postprandial lipaemiaReference Couillard, Bergeron, Prud'homme, Bergeron, Tremblay, Bouchard, Mauriege and Despres100, Reference Mekki, Christofilis and Charbonnier102. The mechanisms are not entirely understood but are probably due to an aberrant insulin-mediated suppression of hepatic VLDL production and fatty acid release from adipose tissueReference Karpe15.
Type 2 diabetes mellitus
Type 2 diabetes mellitus is associated with a marked increase in risk of CVD. A characteristic clinical feature of individuals with diabetes is the prevalence of a dyslipidaemia with elevated plasma levels of TAG, small, dense LDL particles and a low plasma HDL-cholesterol concentrationsReference Haffner109. Both fasting and postprandial plasma remnant lipoprotein-cholesterol levels were elevatedReference Haffner109–Reference Hirano, Yoshino, Kashiwazaki and Adachi111. Interestingly, the impact of type 2 diabetes mellitus on lipoprotein phenotype and on risk of CHD is enhanced in women compared with men. Thus, women with type 2 diabetes mellitus have a higher proportion of small, dense LDL present that is dependent on the plasma TAG tertileReference Howard, Robbins and Sievers112, Reference Guerin, Le Goff, Lassel, Van Tol, Steiner and Chapman113, and they have relatively higher plasma remnant lipoprotein-cholesterol levels than menReference Schaefer, McNamara, Shah, Nakajima, Cupples, Ordovas and Wilson114. Both parameters significantly contribute to the atherogenic lipoprotein phenotype seen in patients with type 2 diabetes mellitus. In addition, the premenopausal advantage in clearance of dietary lipid is not seen in premenopausal diabetic womenReference Masding, Stears, Burdge, Wootton and Sandeman115, and the oestrogen-associated advantage in the clearance of dietary lipid observed in non-diabetic postmenopausal women is not seen in postmenopausal diabetic womenReference Masding, Stears, Burdge, Wootton and Sandeman116.
Genetic background
The effect of several polymorphisms on postprandial lipoprotein metabolism have been studied. However, in the majority of studies described in this section, single-nucleotide polymorphisms have been studied, and few studies have performed a more comprehensive analysis involving haplotypes and multiple genes. A summary with the more recent studies is shown in Table 3.
Table 3 Recent genetic association studies on postprandial lipoprotein response
Apo polymorphisms
Apo A-I is the main HDL protein and plays a crucial role in lipid metabolism. It is an in vivo activator of the enzyme lecithin-cholesterol acyltransferaseReference Fielding, Shore and Fielding117 and an essential element of reverse cholesterol transportReference Reichl and Miller118. These facts may be relevant to postprandial metabolism. Calabresi et al. Reference Calabresi, Cassinotti, Gianfranceschi, Safa, Murakami, Sirtori and Franceschini119 showed that carriers of the rare apo A-I Milano mutation have a threefold higher greater postprandial lipaemia but that after correction for the different baseline TAG levels, it was similar to that of control subjects. In another study, carriers of the A allele in the promoter region of apo A-I ( − 76 base pairs G/A genotype), which occurs at a frequency of 0·15–0·20 in white populations, have a greater postprandial increase in large TRL (35 %) and a smaller decrease in LDL-cholesterol (10 %) and apo B (8 %) after the consumption of a fatty meal than do those with the G/G genotypeReference Marin, Lopez-Miranda, Gomez, Paz, Perez-Martinez, Fuentes, Jimenez-Pereperez, Ordovas and Perez-Jimenez120. The different postprandial responses observed could be due to changes in lipid absorption and/or clearance of TRL particles of intestinal origin, as indicated by the greater increase in apo B-48 (twofold) and large postprandial TRL-TAG concentrations, independently of baseline TAG plasma levels.
Apo A-IV influences dietary fat absorption and chylomicron synthesisReference Ordovas, Cassidy, Civeira, Bisgaier and Schaefer121, modulates the activation of LPL by apo C-IIReference Goldberg, Scheraldi, Yacoub, Saxena and Bisgaier122 and activates lecithin-cholesterol acyltransferaseReference Steinmetz and Utermann123. The most common variant detected are the Gln360His and Thr347Ser polymorphismsReference Menzel, Sigurdsson, Boerwinkle, Schrangl-Will, Dieplinger and Utermann124, Reference de Knijff, Johansen, Rosseneu, Frants, Jespersen and Havekes125. Subjects with the His360 allele, which occurs at a frequency of 0·08–0·10 in white populations, had a higher postprandial increase in small TRL-cholesterol, small TRL-TAG (P < 0·01) and large TRL-TAG than 360Gln/Gln subjectsReference Ostos, Lopez-Miranda, Marin, Castro, Gomez, Paz, Jimenez Pereperez, Ordovas and Perez-Jimenez126, probably due to a delayed hepatic clearance of chylomicron remnants. The Thr347Ser polymorphism, which occurs at a frequency of 0·18–0·22 in white populations, also modulates the postprandial lipaemic response, so that carriers of the Ser347 allele presented a lower postprandial response ( − 26 %) in the TAG levels of chylomicron remnant particles associated with a higher postprandial response in the plasma levels of apo A-IV of chylomicrons (70 %) than did those homozygous for the Thr347 alleleReference Ostos, Lopez-Miranda, Ordovas, Marin, Blanco, Castro, Lopez-Segura, Jimenez-Pereperez and Perez-Jimenez127.
Apo A-V plays an important role in lipid metabolism by modulating hepatic VLDL synthesis and/or secretion, as well as TRL catabolism at the level of LPLReference Weinberg, Cook, Beckstead, Martin, Gallagher, Shelness and Ryan128. Associations between T-1131C and Ser19⇛Trp polymorphisms and TAG concentrations have been found in different population samplesReference Ribalta, Figuera, Fernandez-Ballart, Vilella, Castro Cabezas, Masana and Joven129, Reference Vrablik, Horinek, Ceska, Adamkova, Poledne and Hubacek130. In addition, The C allele of the T-1131C polymorphism, which occurs at a frequency of 0·20–0·25 in white populations, was found to be associated with higher concentrations of plasma TAGReference Masana, Ribalta, Salazar, Fernandez-Ballart, Joven and Cabezas131 and higher postprandial TAG (+30 %)Reference Jang, Kim, Kim, Lee, Cho, Ordovas and Lee132, Reference Moreno, Perez-Jimenez and Marin133. This indicates that the effect of this polymorphism on postprandial lipoprotein response may be mediated, at less in part, by its effect on fasting TAG levels.
Apo B is required for the assembly and secretion of chylomicrons in the small intestine and VLDL in the liver, and also acts as the ligand for the recognition of LDL by LDL receptor. Since apo B is the main protein of LDL and a major component of VLDL, it is to be expected that genetic variations at this locus could influence plasma cholesterol and/or TAG levels in both the fasting and postprandial states. The XbaI polymorphism, a silent mutation (ACC⇛ACT) in exon 26Reference Carlsson, Darnfors, Olofsson and Bjursell134, was related to the interindividual variability observed during postprandial lipaemia. Thus, the frequent X allele is associated with a significantly increased postprandial response of retinyl palmitate (50 %) in all TRL fractions, independently of baseline TAG levelsReference Lopez-Miranda, Ordovas and Ostos135. This mutation does not lead to an amino acid change at the affected codon and cannot have a direct functional effect. Moreover, the D allele at the three-codon (leucine–alanine–leucine) I/D polymorphism within the apo B signal peptideReference Boerwinkle and Chan136 was associated with a reduced postprandial lipid response in comparison with that of individuals homozygous for the I allele, thus suggesting that this signal peptide mutation may affect apo B secretion during the postprandial state. More recently, the association between postprandial NEFA concentrations and TRL has been reported to be influenced by this common deletion polymorphismReference Byrne, Wareham, Mistry, Phillips, Martensz, Halsall, Talmud, Humphries and Hales137, which is also involved in the postprandial responseReference Regis-Bailly, Fournier, Steinmetz, Gueguen, Siest and Visvikis138.
Apo C-I is a constituent of TRL and has been shown to displace apo E from TAG-rich emulsions and interfere with their hepatic clearance. Apo C-I also interferes with the binding of VLDL to the LDL-related protein receptorReference Weisgraber, Mahley, Kowal, Herz, Goldstein and Brown139 and to LDL receptorsReference Sehayek and Eisenberg140. The presence of the apo C-I 317-321ins allele, which occurs at a frequency of 0·30 in white populations, has been shown in vitro to increase the expression of apo C-I by 50 %Reference Jong, Hofker and Havekes141. Thus, a direct inhibitory mechanism would most likely explain the high levels (40 %) of remnant lipoprotein-TAG and remnant lipoprotein-C observed in apo C-I 317-321ins/ins subjectsReference Waterworth, Hubacek, Pitha, Kovar, Poledne, Humphries and Talmud142. This effect appeared to be recessive, with no obvious effect in heterozygous carriers.
Plasma apo C-III inhibits LPL and the binding of apo E-containing lipoproteins to its receptors. Five polymorphisms ( − 641C/A, − 630G/A, − 625T/deletion, − 482C/T, − 455T/C) have been identified in the promoter region of this gene, all of which are in linkage disequilibrium with the SstI site in the 3′ untranslated region, distinguishing the S1 and S2 alleles. Recently, the raising effect of the − 482C/T variant on plasma remnant particles has been shown to be confined to homozygous carriers of the − 482T allele rather than Sst I polymorphic siteReference Waterworth, Hubacek, Pitha, Kovar, Poledne, Humphries and Talmud142. It should be noted that a second variant, − 455T/C, which was not evaluated in that study, is also present in the insulin response element, and would also be likely to show an association with remnant lipoprotein-TAG as it is in strong linkage disequilibrium with the − 482C/T variant. In another study, homozygosity for the G allele at the apo C-III T2854G polymorphism were associated with an increase in the postprandial TAG responseReference Woo and Kang143. The GG homozygotes had 21 % higher TAG area under the curve than the T/T homozygotes, and a 22 % higher TAG value than T/G heterozygotes.
Apo E is a structural component of several lipoproteins and serves as a ligand for the LDL receptor and the LDL receptor-related proteinReference Weisgraber, Mahley, Kowal, Herz, Goldstein and Brown139, Reference Beisiegel, Weber, Ihrke, Herz and Stanley144. Therefore, apo E plays an important role in postprandial lipid metabolism by promoting the efficient uptake of TRL from the circulationReference Gylling, Hallikainen, Pihlajamaki, Agren, Laakso, Rajaratnam, Rauramaa and Miettinen145. However, such functions are not uniformly effective because apo E is present in the population in three main isoforms (E2, E3, E4), which determine apo E concentrations and differ in their affinity to bind to the specific receptorsReference Mahley146, Reference Mahley, Palaoglu and Atak147. In fact, apo E isoforms are important determinants of postprandial lipaemia. It has been demonstrated that apo E-2 homozygous subjects have a delayed postprandial clearance due to the lowest affinity for TRL remnant receptor(s). Compared with apo E-3 homozygous patients, apo E-4 carriers tend to have an enhanced clearance of remnantsReference Weintraub, Eisenberg and Breslow148. However, several studies have found enhanced and/or prolonged postprandial lipid and apo responses in apo E-4 carriersReference Dart, Sherrard and Simpson149, Reference Dallongeville, Tiret and Visvikis150. Patients with the metabolic syndrome who do not have the E-3/3 genotype have a greater risk (odds ratio 6·2, CI 1·41–16·08) of hyperuricaemia and postprandial hypertriacylglycerolaemia after a fat overloadReference Cardona, Morcillo, Gonzalo-Marin and Tinahones151.
On the other hand, a polymorphism in the proximal promoter region of the apo E gene was recently described at position − 219G/TReference Mui, Briggs, Chung, Wallace, Gomez-Isla, Rebeck and Hyman152–Reference Boisfer, Lambert and Atger154, which is associated with an increased risk of myocardial infarctionReference Boisfer, Lambert and Atger154 and CHDReference Viitanen, Pihlajamaki, Miettinen, Karkkainen, Vauhkonen, Halonen, Kareinen, Lehto and Laakso155. The − 219T allele was associated with decreased transcriptional activityReference Artiga, Bullido, Sastre, Recuero, Garcia, Aldudo, Vazquez and Valdivieso153, Reference Boisfer, Lambert and Atger154, decreased plasma apo E concentration both in the fasting and the postprandial stateReference Boisfer, Lambert and Atger154, Reference Moreno, Lopez-Miranda and Marin156 and a prolonged and enhanced postprandial lipaemic response (50 % increase for T/T homozygotes and 15 % for T/G heterozygotes)Reference Moreno, Lopez-Miranda and Marin156.
Transport proteins
The intestinal fatty acid-binding protein-2 is located in the intestine and involved in long-chain fatty acid transport and metabolismReference Matarese, Stone, Waggoner and Bernlohr157. A common alanine for threonine substitution at the FABP2 codon 54 (the A54T polymorphism), which occurs at a frequency of 0·28 in white populations, has been associated with hypertriacylglycerolaemia, obesity, hyperinsulinaemia and insulin resistanceReference Baier, Sacchettini and Knowler158–Reference Yamada, Yuan, Ishiyama, Koyama, Ichikawa, Koyanagi, Koyama and Nonaka160. The T54 allele is associated with a 41 % increased postprandial lipaemia in obeseReference Agren, Valve, Vidgren, Laakso and Uusitupa161 and an 80 % increase in diabeticReference Georgopoulos, Aras and Tsai162 subjects. However, not all studies have supported the associations with postprandial lipaemiaReference Tahvanainen, Molin, Vainio, Tiret, Nicaud, Farinaro, Masana and Ehnholm163. It has been proposed that this association might depend on the type of fat ingested. Thus, in a recent study in which subjects were given three oral fat-tolerance tests (butter, safflower oil, olive oil), the T54 group showed increased chylomicron cholesterol levels only after the olive oil-containing testReference Dworatzek, Hegele and Wolever164.
The fatty acid transport proteins have been implicated in the facilitated cellular uptake of NEFA, thus having the potential to regulate local and systemic NEFA concentrations and metabolism. Hypothesising that genetic variation within the fatty acid transport protein genes may affect postprandial metabolism, the G/A substitution at position 48 in intron 8 of the fatty acid transport-1 gene was studied. Although fasting plasma TAG concentrations were no different, male A/A individuals had significantly higher postprandial TAG concentrations and ratio of VLDL1 (Sf 60–400 apo B-100) to VLDL2 (Sf 20–60 apo B-100) compared with male individuals with the G/A and G/G variationsReference Gertow, Skoglund-Andersson, Eriksson, Boquist, Orth-Gomer, Schenck-Gustafsson, Hamsten and Fisher165.
Enzymes and receptor polymorphisms
The LPL gene is the obvious candidate for studies of postprandial lipaemia, in as much as it codes for the single protein hydrolysing TAG from chylomicrons and large VLDL. It also enhances the binding of apo E-containing lipoproteins to the LDL receptor-related protein, thus affecting the catabolism of chylomicron remnantsReference Beisiegel, Weber and Bengtsson-Olivecrona166. Talmud et al. Reference Talmud, Hall, Holleran, Ramakrishnan, Ginsberg and Humphries167 have studied the interaction between the functional variants involving the LPL-93T/G promoter polymorphism and the LPL D9N substitution, which were identified with a combined population frequency of 3–6 %. Carriers of the haplotype constituting the rare LPL-93G variant (presumably higher transcriptional activity) and the common LPL9N variant (presumably secretion-defective LPL protein) exhibited higher plasma TAG levels after a meal than did carriers of other haplotypesReference Talmud, Hall, Holleran, Ramakrishnan, Ginsberg and Humphries167.
The LPL A291S residue variant affects the specific activity of the enzyme and has a carrier frequency of 4–6 %Reference Fisher, Humphries and Talmud168. Two studies show that carriers of this variant have a significantly higher (41 % higher area under the curve) postprandial triacylglycerolaemiaReference Gerdes, Fisher, Nicaud, Boer, Humphries, Talmud and Faergeman169, Reference Mero, Suurinkeroinen, Syvanne, Knudsen, Yki-Jarvinen and Taskinen170. In a recent study, the association between LPL HindIII (H1/H2) and Ser447-stop (S447X) polymorphisms and postprandial lipaemia was analysed. Thus, carriers of the H1X447 genotypes presented a lower postprandial lipaemic response (42 % lower area under the curve) than subjects with the H2S447 genotype (homozygote for the H2 allele of the LPL HindIII polymorphism and S447 allele), independently of baseline TAG levelsReference Lopez-Miranda, Cruz, Gomez, Marin, Paz, Perez-Martinez, Fuentes, Ordovas and Perez-Jimenez171.
Hepatic lipase has been implicated in the removal of remnant lipoproteins. The promoter of the hepatic lipase gene contains several single-nucleotide polymorphismsReference van't Hooft, Lundahl, Ragogna, Karpe, Olivecrona and Hamsten172. The rare variant of the − 480C/T (also called − 514C/T) polymorphism, present in 0·15–0·21 of the white population, has been associated with lower hepatic lipase activity. Jansen et al. observed that this polymorphism did not seem to affect total postprandial triacylglycerolaemia but did affect the retention of a specific lipoprotein subspecies in the postprandial state, the LpCIII:B particles, which are likely to reflect remnant lipoproteinsReference Jansen, Chu, Ehnholm, Dallongeville, Nicaud and Talmud173. However, in a recent study, subjects homozygous for the T allele showed a lower postprandial response of TRL particles (47 % lower area under the curve) with a decrease in both total TAG and small and large TRL-TAG postprandial responsesReference Gomez, Miranda, Marin, Bellido, Moreno, Moreno, Perez-Martinez and Perez-Jimenez174.
Microsomal triglyceride transfer protein plays a role in the formation of VLDL in the liver and of chylomicrons in the intestine by transferring core lipids to the apo B molecule. Common polymorphisms have been described at position − 493G/T, − 400A/T and − 164T/C in the promoter region for the microsomal triglyceride transfer protein. Homozygous carriers of the rare MTP-493T variant, which is associated with higher transcriptional activity of the gene in vitro Reference Karpe, Lundahl, Ehrenborg, Eriksson and Hamsten175, showed a markedly elevated accumulation of small apo B-48-containing lipoproteins in the postprandial state in healthy subjects and individuals with type 2 diabetesReference Karpe, Lundahl, Ehrenborg, Eriksson and Hamsten175, Reference Lundahl, Hamsten and Karpe176. The − 400 A/T substitution gave very similar lipoprotein results, but there was significant linkage disequilibrium between the two polymorphisms.
Scavenger receptor class B type I is one of the intestinal proteins involved in the absorption of dietary cholesterol and triacylglycerols, suggesting that it may also play a role in postprandial responsesReference Hauser, Dyer and Nandy177, Reference Bietrix, Yan and Nauze178. Thus, the presence of the 2 allele at the scavenger receptor class B type I polymorphism in exon 1 was associated with a faster clearance of small TRL, probably related to a more rapid hepatic uptakeReference Perez-Martinez, Lopez-Miranda and Ordovas179.
Conclusion
As reviewed, postprandial lipid and lipoprotein metabolism is modulated by background dietary pattern as well as meal composition and also by several lifestyle conditions (physical activity, smoking, alcohol consumption), physiological factors (age, gender, menopausal status) and pathological conditions (hypertriacylglycerolaemia, diabetes mellitus, insulin resistance, central obesity). Although these above-mentioned factors do influence postprandial lipid response and metabolism, the weight of their respective effect is variable, as illustrated in Table 4. The most important ones appear to be the amount of meal fat and other components (carbohydrate, protein, alcohol, fibre), physical exercise, tobacco use, gender, pre-existing hypertriacylglycerolaemia, obesity and insulin resistance/type 2 diabetes.
Table 4 Factors affecting postprandial lipid metabolism
+++, ++,+, very important, important or moderate increase; − − , − , important or moderate reduction; No, no noticeable change.
The postprandial lipid response has been shown to be modified by polymorphisms within the genes for apo A-I, E, B, C-I, C-III, A-IV and A-V, LPL, hepatic lipase, fatty acid-binding protein-2, the fatty acid transport proteins, microsomal triglyceride transfer protein and scavenger receptor class B type I. Nevertheless, most previous and current studies have been conducted using the simplest scenarios, that is, one single dietary component, one single nucleotide polymorphism and one single risk factor. We have to evolve toward more realistic situations involving interactions between multiple genes, dietary components and risk factorsReference Corella, Qi, Tai, Deurenberg-Yap, Tan, Chew and Ordovas180. This will require large genetic epidemiological studies and intervention studies involving groups of individuals selected for specific genotype combinations and phenotypic characteristics and subjected to controlled dietary intervention protocols in order to establish the specific gene–diet interactions. Such kind of studies are being conducted through European consortia such as the LIPGENE project (www.lipgene.tcd.ie).
Nutrigenetics examines the effect of genetic variation on the interaction between diet and disease as several risk factors. This includes identifying and characterising gene variants and factors associated with or responsible for differential responses to nutrients or the postprandial response. One of the goal of nutrigenetics is to generate recommendations regarding the risks and benefits of specific diets or dietary components to the individual. It has been also termed ‘personalised nutrition’ or ‘individualised nutrition’.
Intervention and observational studies that attempt to examine gene–diet interactions need to include repeated sampling and measurement to provide an accurate measure of the phenotypes. To elucidate gene–environment interactions, and specifically gene–diet interactions, we need population sizes several orders of magnitude larger than those currently used for common multifactorial diseases. This will require the creation of international consortiums built along the models of the EPIC study or the Human Genome Project. Complex phenotype and genotype interactions require an analysis of their combined effects. The information will need to be incorporated into predictive models that can be used clinically to improve disease assessment and prevention. This will be probably happen within the umbrella of bioinformatics or computational biology.
Acknowledgements
This work was commissioned by the Nutritional Value of Food Task Force of the European branch of the International Life Sciences Institute (ILSI Europe). Industry members of this task force are Nestlé, Danone, Südzucker and Danisco. For further information about ILSI Europe, email info@ilsieurope.be or call +32 27710014. The opinions expressed in this article are those of the authors and do not necessarily represent the views of ILSI Europe.
