Reduction of dietary saturated fatty acids correlates with increased ...

European Journal of Clinical Nutrition (2004) 58, 881–887

& 2004 Nature Publishing Group All rights reserved 0954-3007/04 $30.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION Reduction of dietary saturated fatty acids correlates with increased plasma lecithin cholesterol acyltransferase activity in humans AM Be´rard1*, H Dabadie2, A Palos-Pinto1, M-F Dumon1 and M Darmon1 1 Laboratoire de Biochimie et de Biologie Molećulaire, Universite´ Victor Se´galen Bordeaux, Bordeaux, France; and 2Service de Nutrition, Hoˆpital Haut-Le´veˆque, Avenue du Haut-Le´veˆque, Pessac, France

Objective: Increased HDL-cholesterol (HDL-C) concentrations have been associated with lower coronary heart disease risk. On the other hand, dietary fats are known to influence the fatty acid profile of plasma lipids, including phospholipids that are substrates of lecithin cholesterol acyltransferase (LCAT), an important enzyme in HDL metabolism. The purpose of this study was to examine the association between the saturated fatty acid (SFA) intake and LCAT activity. Design: An interventional study was performed in a monk community of 25 men. Setting: A French monk community, South West of France. Subjects and interventions: The basal diet of the study cohort contained SFA in a proportion of 13.5% of their total energy intake (TEI). They were submitted to two experimental isocaloric diets containing either 8.4% of the TEI in SFA (diet A) or 11% (diet B), each lasting 5 weeks. Results: The elevation of SFA in diet B was mainly obtained by decreasing carbohydrates. The only significant difference among total fats between diets A and B was the myristic acid content (0.6 and 1.2% of TEI, respectively). The elevation in SFA in diet B resulted in a significant increase of HDL-C (Po0.04), while plasma apo A-I concentration and LCAT activity both decreased (Po0.02). Conclusion: Altogether, these results are consistent with a negative effect of SFA on reverse cholesterol transport.

European Journal of Clinical Nutrition (2004) 58, 881–887. doi:10.1038/sj.ejcn.1601890 Keywords: lecithin cholesterol acyltransferase (LCAT); reverse cholesterol transport; saturated fatty acids

Introduction Epidemiologic studies carried out at the end of the 20th century have shown that the amount of dietary fat intake is positively correlated with the plasma total cholesterol value and morbidity from coronary heart diseases (CHDs) (Hu et al,

*Correspondence: AM Be´rard, Laboratoire de Biochimie et de Biologie Molećulaire, Universite´ Victor Se´galen Bordeaux 2, Zone Nord–Case 49 146, Rue Leó Saignat, 33076 Bordeaux Cedex, France. E-mail: [email protected] Guarantor: AM Be´rard Contributors: AMB was responsible for lipoprotein metabolism analysis, determination of LCAT and CETP activities, data analysis and preparation of the paper. HD was responsible for the study design and recruiting subjects. APP was responsible for FPLC analysis and CER determination. MFD was responsible for laboratory organization and data collection. MD was responsible for preparation of the paper. Received 17 March 2003; revised 20 August 2003; accepted 10 September 2003

1997). However, a number of studies demonstrated that not only total fat, but also the fatty acid composition of the diet was associated with variations in total cholesterol concentration and development of atherosclerosis (Burr et al, 1989; Grundy, 1994; de Lorgeril et al, 1999; GISSI investigators, 1999). The general picture is that saturated fatty acids (SFAs) elevate total cholesterol concentration and the risk of CHD, while unsaturated fatty acids (UFAs) have the opposite effect. Low-density lipoproteins (LDLs) represent the lipoprotein fraction incriminated in the adverse effects of high total cholesterol concentration, so that presently the concentration of cholesterol carried by LDL is the parameter considered for reducing the CHD risk by the diet and total cholesterol-value lowering drugs. LDL-cholesterol (LDL-C) is also the lipid fraction most affected by the cholesterol amount and fatty acid composition of the diet by mechanisms presently well understood (reviewed in Spector et al, 1980; Dietschy, 1998).

LCAT activity and saturated fatty acids AM Be´rard et al

882 Epidemiologic studies have also shown that high-density lipoprotein (HDL)-cholesterol (HDL-C) concentrations, and even more the HDL-C to total cholesterol ratio represent an important protective factor against CHD. This can be explained by the major role of HDL in reverse cholesterol transport. The effects of UFA vs SFA on HDL are not as simple to interpret as those on LDL. Monounsaturated fatty acids (MUFAs) reduce LDL-C without altering HDL-C concentrations (Schaefer et al, 1981; Grundy & Denke, 1990; Riccardi & Rivellese, 1993). Replacing SFA by polyunsaturated fatty acids (PUFAs) results in a decrease in HDL-C that is much less important than the decrease in total cholesterol and LDL-C (Hodson et al, 2001), explaining why UFA are protective albeit their negative effect on absolute HDL-C concentrations. There may be other reasons why the HDL-C value is not always predictive of an increased reverse cholesterol transport. For example, it has been demonstrated that alterations in HDL composition may involve dysfunctional HDL leading to increased atherosclerosis despite the presence of high plasma HDL concentrations (Be´rard et al, 1997). On the contrary, n-3 PUFA, although decreasing HDL concentrations, increase the fractional catabolic rate of HDL particles, thus stimulating reverse cholesterol transport (Le Morvan et al, 2002). Thus, other parameters than the HDL-C concentration must be considered to evaluate beneficial vs adverse effects of fatty acid intake on HDL metabolism. Among those, we wished to address the enzyme lecithin cholesterol acyltransferase (LCAT, EC 2.3.1.43). LCAT is a glycoprotein that transesterifies the sn-2 or less frequently the sn-1 fatty acid of a phosphatidyl-choline molecule to the 3-b-hydroxyl group of cholesterol, resulting in the formation of cholesterol ester and lysophosphatidyl-choline. Moreover, the fatty acid profile of phospholipids—the substrates of LCAT—is known to be influenced by dietary fats (Baudet et al, 1984; Cox et al, 1995; Gillotte et al, 1998). LCAT, being the major enzyme in plasma responsible for esterification of cholesterol present in circulating lipoproteins (Glomset, 1968), and being associated primarily with HDL (Francone et al, 1989), plays a major role in HDL metabolism. By promoting esterification of free cholesterol to cholesteryl ester, which is then packaged into the core of HDL, LCAT helps to maintain a concentration gradient favorable for the diffusion of free cholesterol from peripheral tissues to HDL. Thus, together with cholesteryl ester transfer protein (CETP) and apo A-I, LCAT appears to be necessary for reverse cholesterol transport (Miller et al, 1985). This pathway, known to allow for recycling of cholesterol from peripheral tissues to liver and for biliary excretion, is probably one of the major defense mechanisms against the development of atherosclerosis. The purpose of this investigation was to examine, in a community of monks, the effects of low increased dietary SFA intake on LCAT and CETP activities as well as on lipid and lipoprotein parameters. European Journal of Clinical Nutrition

Subjects and methods Subjects and experimental design A total of 25 men were enrolled in the study. They were part of a religious community with regular lifestyle and dietary habits. All subjects had normal hepatic, renal and thyroid functions. None of the study subjects were taking medications known to affect plasma lipoproteins. All of them gave their informed consent before their inclusion in the study. The protocol was approved by the Human Research Committee of the ‘Centre Hospitalier Universitaire’ of Bordeaux (France). Subjects were requested to maintain their normal patterns of activity, and not to add or to remove any food to their proposed meals during the two experimental diets. Study participants were assigned to two consecutive 5 week experimental isocaloric diets separated by a ‘wash-out’ period of 5 weeks of basal diet. The first diet (A) was close to the current recommendations in SFA content (a mean of 8.4% of the total energy intake (TEI)) and enriched in oleic acid (a mean of 14.4% compared to 11.3% of TEI in the basal diet) (Krauss et al, 2000). The second diet (B) differed from diet A by a higher dietary SFA content compensated by a reduction in carbohydrates obtained by using full-fat milk and decreasing bread and sugar intakes. Diet B contained an amount of SFA closer to the usual diet of the subjects (a mean of 11.0% of the TEI), but was still enriched in oleic acid (a mean of 15.8% of TEI). SFA in diet B were increased with respect to diet A in the following proportions: myristic acid 100%, palmitic acid 43% and stearic acid 35%. The two-fold increase in myristic acid intake was reached by replacing 500 ml/day of low-fat milk by the same quantity of full fat milk. Strict logistics in food preparation and consumption were used in the community. Energy and nutrient intakes were calculated using tables of food composition for the raw weights of foodstuffs. The two experimental diets had the same energy and cholesterol contents, and contained similar amounts of most essential nutrients (except for fat-soluble vitamins). A daily food questionnaire was also done for each monk. All data were analyzed with Bilnut 3 software.

Anthropometric measurements The weight of each subject was measured with a batteryoperated scale accurate to 0.5 kg and height was measured with use of a portable stadiometer. Body mass index (BMI) was calculated as weight (in kg) divided by height squared (in m).

Blood sampling Blood samples in 0.1% EDTA were obtained from each subject by venipuncture, after a 12 h overnight fasting period. Samples were taken before interventional diets and after the 5 weekperiod of diets A and B. Plasma was separated by centrifugation (1200 g, 20 min, þ 41C) and kept in cryovials at 801C until analysis.


883 Lipid, lipoprotein and apolipoprotein analysis Plasma total cholesterol (TC) and triacylglycerols (TG) were quantitated on an automated chemistry analyzer (Paramax, Dade-Berhring, Paris La De´fense, France) using enzymatic methods. HDL-C was measured by an enzymatic method on the supernatant obtained after selective precipitation of apolipoprotein B-containing lipoproteins with phosphotungstic acid in the presence of magnesium ions, and centrifugation. LDL-C was calculated by using the following formula: LDL-cholesterol¼TC(TG/2.2 þ HDL-C). Apolipoproteins A-I, A-II and B were determined with a nephelometer analyzer (Dade-Behring).

Gel filtration chromatography After centrifugation of pooled plasmas at 2500 g for 20 min at 41C, 500 ml of clear supernatant was applied to a fast protein liquid chromatography (FPLC) system with two Superose 6 HR 10/30 columns connected in series (Pharmacia LKB, Orsay, France). Lipoproteins were eluted at 0.3 ml/ min with phosphate buffer saline, pH¼7.4, containing 1 mmol/l EDTA and 0.02% sodium azide. After the first 10 ml were eluted, 60 fractions (0.51 ml each) were collected. Total cholesterol was quantified in each fraction in order to establish an FPLC profile. FPLC fractions corresponding to HDL were pooled and lyophilized in order to measure free cholesterol and phospholipid concentrations by using commercial enzymatic kits (WAK-Chemie, Steinbach, Germany and bioMe´rieux, Marcy l’Etoile, France). Cholesteryl ester values were estimated as the difference between total cholesterol and free cholesterol concentrations. Since the FPLC were performed with pools, we have no direct information on the interindividual variability of the profiles. However, the determination of plasma total cholesterol and HDL-C levels (and non-HDL-C concentrations) showed interindividual variability.

LCAT activity and cholesterol esterification rate (CER) Plasma LCAT activity was determined using 4 ml of human plasma and an artificial HDL-like proteoliposome substrate as previously described (Chen & Albers, 1982). Stable proteoliposomes were synthesized at 371C by incubating for 30 min apo A-I (Sigma, Saint-Quentin Fallavier, France), [14C]cholesterol (NEN, Paris, France) and egg phosphatidylcholine (Sigma) at molar ratios of 0.8:12.5:250. LCAT activity was then determined from the rate of formation of [14C]cholesteryl ester. Plasma CER was determined as described previously (Stokke & Norum, 1971), by quantitation of the radioactivity of esterified cholesterol before and after incubation of 100 ml [14C]cholesterol-labeled autologous plasma at 371C for 6 h. The plasma CER represents the total LCAT activity and cholesterol esterification present in a plasma sample. In both cases, the molar rate of LCAT esterification was calculated by multiplying the percentage of cholesterol

esters formed per hour by the concentration of total cholesterol either in the proteoliposomes or in the plasma sample.

CETP activity Plasma CETP activity was determined by assessing the transfer of 14C-labeled HDL3 cholesteryl ester to the do1.060 g/ml lipoproteins as described previously (Albers et al, 1984). This was performed after incubation for 18 h at 371C with or without the addition of 10 ml of plasma as a source of CETP. The HDL3 and the do1.060 g/ml lipoproteins were separated by heparin–MnCl2 precipitation, and the radioactivity in the supernatant (HDL3) was then determined. Transfer activity was expressed as the percentage transfer/volume/incubation time and calculated as ((buffer control c.p.m.—test sample c.p.m.)/buffer control c.p.m.) 100.

Fatty acid determination Lipid extraction from plasmas was done with hexane–isopropanol, and classes of lipids were separated by thin-layer chromatography (Merck-Cle´venot, Nogent-sur-Marne, France) in diethyl ether, petroleum ether, acetic acid (270: 30:3, by vol), associated with iodine detection Phospholipids and cholesteryl esters were extracted with hexane-isopropanol. Isopropylic ester preparation of fatty acids and their separation for further analysis by a silica capillary column (CP Sil 88 column; Chrompack, Les Ulis, France) with a Carlo Erba model 5160 gas chromatograph equipped with a flame ionization detector (Carlo Erba, Massy, France) were done as described previously (Peuchant et al, 1989). Fatty acids were identified by retention times obtained from various standards (Sigma) and chromatographed under the same conditions.

Statistical analysis Results are expressed as mean7s.d. Differences between values in the study cohort fed different diets were analyzed by a paired Student’s t-test in which individual subjects served as their own controls. Statistical significance was defined as Po0.05.

Results Subjects characteristics The study cohort consisted of 25 men living together in a monk community. This confers an homogeneity advantageous for the study design. Mean age, BMI, lipid and lipoproptein parameters of the study population are shown in Table 1. European Journal of Clinical Nutrition


884 Table 1 Characteristics of the studied monk community

Table 2

Characteristics of the two experimental diets

N¼25

Mean7s.d.

Range

N¼25

Age (y) Weight (kg) Body mass index (kg/m2) Total cholesterol (mmol/l) Triacylglycerols (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l)

62.8710.8 72.278.8 24.673.3 5.5870.70 1.0470.34 1.1770.16 3.9470.67

35–79 57.5–87.5 17.8–31.8 3.95–7.01 0.43–2.11 0.81–1.84 2.49–5.17

Total energy intake (kcal/day) Proteins (g/day) Carbohydrates (g/day) Total fats (g/day) PUFA (g/day) MUFA (g/day) SFA (g/day) PUFA/SFA (ratio) Fibers (g/day) *

Dietary analysis Table 2 reports the characteristics of diets A and B. The two experimental isocaloric diets were normocaloric with an adequate energy contribution from protein (a mean of 14% of TEI). Daily cholesterol intake was E200 mg. Compared to diet A, diet B contained more SFA (11 vs 8.4% of TEI) without modifying notably PUFA and MUFA, and altogether total fats increased from 31 to 35% of TEI, while carbohydrates decreased from 55 to 51% of TEI. Considering the total fat intake, the only significant difference between diets A and B was the SFA intake, mainly myristic acid. Indeed, the myristic acid intake increased two-fold, from 1.5770.29 g/ day (a mean of 0.6% of TEI) to 2.9270.54 g/day (a mean of 1.2% of TEI); this increase was significant (Po0.023) while other SFA, such as palmitic acid and stearic acid increased slightly and nonsignificantly (12.1570.52 vs 8.6570.76 g/day, P¼0.122 and 5.6270.52 vs 4.2370.35 g/day, P¼0.059, respectively).

European Journal of Clinical Nutrition

Diet B

2202785 76.671.9 293.46711.55 73.2772.08 19.971.47 3471.14 19.470.55 1 21.772.00

21987128 75.473.28 269.48720.35* 83.1873.03* 20.171.98 37.171.76 26.170.67* 0.75* 20.271.00

Po0.02.

Table 3 Lipid, lipoprotein, apolipoprotein concentrations LCAT and CETP activities in the study cohort (n¼25) before and after 5 weeks on diet A or B N¼25

Diet A

Total cholesterol (mmol/l) Triacylglycerols (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Apo A-I (g/l) Apo A-II (g/l) Apo B (g/l) Cholesteryl ester/total cholesterol (%) LCAT activity (nmol/ml/h) CER (nmol/ml/h) CETP activity (%) Myristic acid intake Myristic acid into phospholipids *

Po0.0005; **Po0.006; wPo0.02;

VLDL

2.50

IDL

Diet B

4.9870.61 0.9270.29 1.1870.19 3.3970.58 1.6070.15 0.5070.05 1.0170.20

4.9670.69 0.8470.24* 1.2570.24z 3.3470.55 1.4970.20w 0.4870.04 0.9070.16*

8671 74720 3673 2777

8174z 56713w 2973ww 20713

1.5770.29 0.9070.02

2.9270.54ww 1.1070.02**

ww

Po0.03; zPo0.04, compared to diet A.

LDL

HDL2

+

HDL3

2.00 total cholesterol (mmol/L)

Comparison of the effects of diet B vs diet A on lipid, lipoprotein, apolipoprotein, LCAT and CETP analysis As diet B differed from diet A by a reduction in carbohydrates in order to get a higher dietary SFA content with a two-fold increase in myristic acid content, plasma phospholipid analysis showed a significant increase in myristic acid content (1.0370.20 vs 0.9070.17%, Po0.006) without any alteration of palmitic acid (26.1671.71 vs 26.3171.40%) and stearic acid (10.4071.69 vs 10.5571.37%) content. A comparison of lipid and lipoprotein parameters after 5 weeks of either diet A or B shows that diet B (Table 3) provoked a significant decrease in triacylglycerol (Po0.0005) and apo B concentrations (Po0.0005) without modifying LDL-C concentrations, suggesting a decrease in VLDL particle levels. Moreover, it induced an elevation of HDL-C (Po0.04) leading to a slight decrease in total cholesterol/HDL-C ratio (4.1770.77 vs 4.4170.93, Po0.04). It is worth noting that the increased HDL-C in diet B was accompanied by a decreased apo A-I, while apo A-II concentrations did not change. Figure 1 illustrates the FPLC profile of fasting plasma from study subjects on diet A or B. As the IDL fraction did not change, the variation in VLDL-cholesterol concentration could be estimated by calculating the difference between total cholesterol and (LDL-cholesterol þ HDL-C): 0.4170.13

Diet A

diet A diet B 1.50

1.00

0.50

0.00 10

15

20

25 30 elution volume (ml)

35

40

Figure 1 The cholesterol distribution in the plasma lipoproteins from study subjects on diet A (dotted lines) or B (solid lines). The elution positions of VLDL, IDL/LDL, HDL2/HDL3 lipoproteins are indicated.

and 0.3770.11 mmol/l in diets A and B, respectively (Po0.03). In agreement with these data and prior plasma lipid and lipoprotein determinations, FPLC analysis revealed


885 a decrease in VLDL-cholesterol and an increase in HDL-C when monks were feeding diet B. Moreover, when all lipid levels were compared, diet B induced marked alterations in lipid composition of HDL particles. The increase in total cholesterol in HDL, affecting all HDL subfractions, was due to an increase in cholesteryl ester and free cholesterol HDL content E15 and 40% higher, respectively, leading to a lower cholesteryl ester/total cholesterol ratio, suggesting a diminution in cholesterol esterification reaction rate. In addition, triglycerides were unchanged while phospholipids were 2.3fold increased so that the total cholesterol/phospholipids ratio was 48% lower (2.30 vs 4.45). Eventually, in order to better determine the mechanism(s) responsible(s) for the elevated HDL-C concentrations noted in diet B, we assessed plasma LCAT and CETP activities (Table 3). The LCAT activity was determined in two ways: (1) by measuring serum cholesterol esterification rates, which are an estimate of LCAT action on endogenous lipoproteins, and (2) by measuring serum LCAT activity levels with excess exogenous substrates, an estimate of LCAT mass. Both determination demonstrated a marked decrease of the amount in diet B compared to diet A: E20 and 25%, respectively. CETP activity was lower after diet B than after diet A and this decrease was rather important even not significant (E25%).

Discussion Our study addressed the question whether low variations in amounts of SFA in the diets could impair the progression of risk factors for atherosclerosis, and thus contribute to reduce the risk of cardiovascular diseases. This question is of practical importance since the SFA content of diet B is close to dietary contents observed in epidemiological studies, while SFA content in diet A is equal to the present recommendations (Krauss et al, 2000). Since myristic acid was reported to be in human subjects, the SFA inducing the most important rise in plasma total cholesterol, especially through an increase of LDL-C levels (Kris-Etherton & Dietschy, 1997), we specifically raised the myristic acid content component of the saturated fatty acids in the study diet. Nevertheless, in most previous studies, myristic acid represented a very high percentage of the total dietary energy and replaced PUFA. On the opposite, PUFA were not different in diets A and B, and the increase in MUFA in diet B was not significant. Moreover, since diets A and B were isocaloric a nd brought identical energy intake from proteins, the increased SFA in diet B led to more elevated total fat and to lesser carbohydrate intakes than in diet A. When compared to diet A, diet B reduced non-HDL lipoproteins. This reduction probably interested VLDL since LDL lipoproteins were equivalent in diets A and B, and both triacylglycerol and apo B concentrations were reduced in diet B. This hypothesis was confirmed by the FPLC analysis showing a marked decrease in VLDL particles. The reduction

of VLDL lipoproteins might be in part a consequence of the reduction of the carbohydrate intake. Nevertheless, the reduction in VLDL-triacylglycerides visualized on FPLC (data not shown) was associated with that of VLDL-cholesterol. This might be explained by a decrease in CETP activity in monks on diet B. Nevertheless, although we found a 25% decrease in CETP activity in diet B, these data were not significant and further studies with more subjects will be required to ascertain this point. There is some discrepancy between our results and those obtained by Jansen et al (2000). These authors demonstrated that diets enriched in SFA at the expense of carbohydrates show higher CETP activity. The fact that they increased palmitic and stearic acid dietary contents could explain the difference between their results and ours. As previously reported (Mensink & Katan, 1992), diets enriched in SFA at the expense of carbohydrates show not only lower plasma triacylglycerol and VLDL concentrations, but also higher HDL-C. Indeed, in our study, diet B led to a significant increase of HDL-C, accompanied by a significant decrease in apo A-I. The enrichment in total cholesterol was due to an increase in free cholesterol rather than in cholesteryl ester so that the cholesteryl ester/free cholesterol ratio was 30% lower in diet B. Altogether these results suggest a diminution of cholesterol esterification reaction rate. Interestingly, LCAT activity evaluated by measuring serum LCAT activity levels with excess exogenous substrates, that is, an estimate of LCAT mass (Romijn et al, 1998), was significantly reduced in diet B. This report is to our knowledge the first one showing in man that a reduction in dietary SFA increases LCAT activity. Although some studies replacing unsaturated by saturated fatty acids showed an increased LCAT activity either in rats or in man (Miller et al, 1975; Takatori et al, 1976; Larking & Sutherland, 1977), Thornburg et al (1995) demonstrated a diminution in LCAT activity in monkey which is in agreement with our results. Moreover, a study performed in the rat (Romijn et al, 1998) showed that the cholesterol esterification rate was worse when linoleic acid was replaced by palmitic acid, but these authors did not find a decreased LCAT activity after using proteoliposomes. They concluded that this absence of correlation was due to changes in the composition of endogenous lipoprotein substrates induced by the diet, compared to the proteoliposomes used in the in vitro assay. Although changes in dietary fats are known to influence the fatty acid composition of LCAT substrates (Baudet et al, 1984; Cox et al, 1995; Gillotte et al, 1998), the enhanced activity that we observed on diet A must have another explanation. Actually, the assay was performed in vitro with artificial substrates and the observed increased activity must be secondary either to an increased mass of the enzyme in plasma or to a better environment of the enzyme (ie the presence of activators) in the HDL particles. We then determined LCAT activity by measuring serum cholesterol esterification rates—which is an estimate of LCAT action—on endogenous lipoproteins, and we also found a significant increase in CER in study subjects fed diet A. European Journal of Clinical Nutrition


886 Indeed, this positive effect on LCAT activity of diets reduced in SFA can be considered as beneficial in terms of reverse cholesterol transport because this enzyme steepens the concentration gradient between peripheral cholesterol and HDL. It is thus possible that the increase of LCAT activity described here participates in the overall healthier status of diets of low SFA/UFA presently recommended. Actually, the American Heart Association (AHA) dietary guidelines (Krauss et al, 2000) advocate a population-wide saturated fat intake of 8–10% of energy that represents a reasonable population target. Nevertheless, the AHA has not incorporated information about individual SFA dietary recommendations because there are no sufficient data in humans. In our study, the major and significant change in SFA content in diet B compared to diet A was the myristic acid intake: 1.2% of the TEI vs 0.6%. That led to a significant increase in HDL-C already described elsewhere (Temme et al, 1997). The elevated HDL-C value was accompanied by a significant decrease in apo A-I and LCAT, as well as a 25% diminution in CETP activity. Moreover, further characterization of the plasma lipids into HDL isolated by FPLC demonstrated alterations in HDL composition, such as a 57% increase in phospholipid HDL content in diet B. In fact, the increase in phospholipids may reflect a requirement for additional phospholipids to be added to the surface layer of the lipoprotein to accomodate an expanding cholesteryl ester core. Taken together, these results are consistent with the idea that a higher SFA intake might induce an increase in the residence time of HDL particles. Consequently, HDL might be less capable to return cholesteryl esters to liver. Such an effect has been described in hamsters (Loison et al, 2002). The authors observed a negative correlation between the HDL-C concentration and the hepatic mass of scavenger receptor B1 (SR-B1), an HDL receptor, when animals were fed diets with increasing contents of myristic acid (from 0.6 up to 2.4% TEI). Moreover, they showed a lower cholesterol 7ahydroxylase activity that signed a reduced activity of the biliary pathway, the main route for cholesterol excretion. The results of the present study, performed in humans, show that increasing dietary SFA—mainly myristic acid— might result in a decrease of reverse cholesterol transport, an important pathway known to modulate the cardiovascular risk. These observations are in agreement with reports showing that reverse cholesterol transport might happen to be impaired despite of high plasma HDL-C (Be´rard et al, 1997), and provide further evidence that SFA with a carbon chain length of C12:0–C16:0 might be submitted to individual dietary recommendations. This last point needs further human studies.

Acknowledgements We acknowledge Mrs RP Lacome`re and M Bernard for collecting dietary information and for their contribution to diet analysis. We thank E Peuchant for her contribution to fatty acid analysis. European Journal of Clinical Nutrition

References Albers JJ, Tollefson JH, Chen CH & Steinmetz A (1984): Isolation and characterization of human plasma lipid transfer proteins. Arteriosclerosis 4, 49–58. Baudet MF, Dachet C, Lasserre M, Esteva O & Jacotot B (1984): Modification in the composition and metabolic properties of human low density and high density lipoproteins by different dietary fats. J. Lipid Res. 25, 456–468. Be´rard AM, Foger B, Remaley A, Shamburek R, Vaisman BL, Talley G, Paigen B, Hoyt Jr RF, Marcovina S, Brewer Jr HB & Santamarina-Fojo S (1997): High plasma HDL concentrations associated with enhanced atherosclerosis in transgenic mice overexpressing lecithin-cholesteryl acyltransferase. Nat. Med. 3, 744–749. Burr ML, Fehily AM, Gilbert JF, Rogers S, Holliday RM, Sweetnam PM, Elwood PC & Deadman NM (1989): Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction: diet and reinfarction trial (DART). Lancet 2, 757–761. Chen CH & Albers JJ (1982): Characterization of proteoliposomes containing apoprotein A-I: a new substrate for the measurement of lecithin: cholesterol acyltransferase activity. J. Lipid Res. 23, 680–691. Cox C, Mann J, Sutherland W, Chisholm A & Skeaff M (1995): Effects of coconut oil, butter, and safflower oil on lipids and lipoproteins in persons with moderately elevated cholesterol levels. J. Lipid Res. 36, 1787–1795. de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J & Mamelle N (1999): Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation 99, 779–785. Dietschy JM (1998): Dietary fatty acids and the regulation of plasma low density lipoprotein cholesterol concentrations. J. Nutr. 128(Suppl), 444S–448S. Francone OL, Gurakar A & Fielding C (1989): Distribution and functions of lecithin:cholesterol acyltransferase and cholesteryl ester transfer protein in plasma lipoproteins. Evidence for a functional unit containing these activities together with apolipoproteins A-I and D that catalyzes the esterification and transfer of cell-derived cholesterol. J. Biol. Chem. 264, 7066–7072. Gillotte KL, Lund-Katz S, de la Llera-Moya M, Parks JS, Rudel LL, Rothblat GH & Phillips MC (1998): Dietary modification of high density lipoprotein phospholipid and influence on cellular cholesterol efflux. J. Lipid Res. 39, 2065–2075. GISSI investigators (1999): Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione TRIAL. Lancet 354, 447–455. Glomset JA (1968): The plasma lecithin:cholesterol acyltransferase reaction. J. Lipid Res. 9, 155–167. Grundy SM (1994): Influence of stearic acid on cholesterol metabolism relative to other long-chain fatty acids. Am. J. Clin. Nutr. 60(Suppl), 986S–990S. Grundy SM & Denke MA (1990): Dietary influences on serum lipids and lipoproteins. J. Lipid Res. 31, 1149–1172. Hodson L, Skeaff CM & Chisholm WA (2001): The effect of replacing dietary saturated fat with polyunsaturated or monounsaturated fat on plasma lipids in free-living young adults. Eur. J. Clin. Nutr. 55, 908–915. Hu FB, Stampfer MJ, Manson JE, Rimm E, Colditz, GA, Rosner BA, Hennekens CH & Willett WC (1997): Dietary fat intake and the risk of coronary heart disease in women. N. Engl. J. Med. 337, 1491–1499. Jansen S, Lopez-Miranda J, Castro P, Lopez-Segura F, Marin C, Ordovas JM, Paz E, Jimenez-Pereperez J, Fuentes F & Perez-Jimenez F (2000): Low-fat and high-monounsaturated fatty acid diets decrease plasma cholesterol ester transfer protein concentrations in young, healthy, normolipemic men. Am. J. Clin. Nutr. 72, 36–41.


887 Krauss RM, Eckel RH, Howard B, Appel LJ, Daniels SR, Deckelbaum RJ, Erdman Jr JW, Kris-Etherton P, Goldberg IJ, Kotchen TA, Lichtenstein AH, Mitch WE, Mullis R, Robinson K, Wylie-Rosett J, St Jeor S, Suttie J, Tribble DL & Bazzarre TL (2000): AHA Dietary Guidelines: revision 2000: a statement for healthcare professionals from the Nutrition Committee of the American Heart Association. Circulation 102, 2284–2299. Kris-Etherton PM & Dietschy J (1997): Design criteria for studies examining individual fatty acid effects on cardiovascular disease risk factors: human and animal studies. Am. J. Clin. Nutr. 65, 1590S–1596S. Larking PW & Sutherland WH (1977): Lecithin: cholesterol acyl transferase activity in the serum of rats fed saturated and unsaturated fats. Atherosclerosis 26, 225–232. Le Morvan V, Dumon MF, Palos-Pinto A & Be´rard AM (2002): n-3 FA increase liver uptake of HDL-cholesterol in mice. Lipids 37, 767–772. Loison C, Mendy F, Serougne C & Lutton C (2002): Dietary myristic acid modifies the HDL-cholesterol concentration and liver scavenger receptor BI expression in the hamster. Br. J. Nutr. 87, 199–210. Mensink RP & Katan MB (1992): Effect of dietary fatty acids on serum lipids and lipoproteins A meta-analysis of 27 trials. Arterioscler. Thromb. 12, 911–919. Miller JP, Chait A & Lewis B (1975): The relationship between dietary fat composition and plasma cholesterol esterification in man. Clin. Sci. Mol. Med. 49, 617–620. Miller NE, La Ville A & Crook D (1985): Direct evidence that reverse cholesterol transport is mediated by high-density lipoprotein in rabbit. Nature 314, 109–111.

Peuchant E, Wolff R, Salles C & Jensen R (1989): One-step extraction of human erythrocyte lipids allowing rapid determination of fatty acid composition. Anal. Biochem. 181, 341–344. Riccardi G & Rivellese AA (1993): An update on monounsaturated fatty acids. Curr. Opin. Lipidol. 4, 13–16. Romijn D, Wiseman SA, Scheek LM, de Fouw NJ & van Tol A (1998): A linoleic acid enriched diet increases serum cholesterol esterification by lecithin:cholesterol acyltransferase in meal-fed rats. Ann. Nutr. Metab. 42, 244–250. Schaefer EJ, Levy RI, Ernst ND, Van Sant FD & Brewer Jr HB (1981): The effects of low cholesterol, high polyunsaturated fat, and low fat diets on plasma lipid and lipoprotein cholesterol levels in normal and hypercholesterolemic subjects. Am. J. Clin. Nutr. 34, 1758–1763. Spector AA, Kaduce TL & Dane RW (1980): Effect of dietary fat saturation on acylcoenzyme A:cholesterol acyltransferase activity of rat liver microsomes. J. Lipid Res. 21, 169–179. Stokke KT & Norum KR (1971): Determination of lecithin: cholesterol acyltransfer in human blood plasma Scand. J. Clin. Lab. Invest. 27, 21–27. Takatori T, Phillips FC, Shimasaki H & Privett OS (1976): Effects of dietary saturated and trans fatty acids on tissue lipid composition and serum LCAT activity in the rat. Lipids 11, 272–280. Temme EH, Mensink RP & Hornstra G (1997): Effects of medium chain fatty acids (MCFA), myristic acid, and oleic acid on serum lipoproteins in healthy subjects. J. Lipid Res. 38, 1746–1754. Thornburg JT, Parks JS & Rudel LL (1995): Dietary fatty acid modification of HDL phospholipid molecular species alters lecithin: cholesterol acyltransferase reactivity in cynomolgus monkeys. J. Lipid Res. 36, 277–289.

European Journal of Clinical Nutrition