![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
9) Depresses HDL and Increases CETP Activity in Cebus Monkeys1,2,3
Foster Biomedical Research Laboratory, Brandeis University, Waltham, MA 02254
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
LITERATURE CITED
ABSTRACT 
The question whether dietary trans fatty acids affect lipoprotein metabolism similarly to specific saturated fatty acids was investigated in 11 normolipemic cebus monkeys by exchanging 5% dietary energy (%en) between elaidic (t-C18:1
9) and palmitic acid (16:0) in two test diets (30%en fat + 100 mg cholesterol /1000 kcal diet) conforming to the American Heart Association (AHA) Step 1 guidelines. These were compared with a normal control diet rich in saturated fat and cholesterol (38%en fat + 180 mg cholesterol /1000 kcal diet). The control diet was fed initially for 14 wk, followed by each of the the two test diets in a crossover design. Plasma lipid concentrations were determined four times between the 6th and 14th wk. Turnover studies (using 125I-HDL and 131I-LDL) were conducted after 9 wk in each dietary period. Relative to the control diet, both test diets significantly reduced plasma total cholesterol (TC), HDL cholesterol (HDL-C) and VLDL plus LDL cholesterol (LDL-C) concentrations; triglyceride (TG) concentrations tended to be lower. However, the trans diet resulted in a significantly greater reduction in HDL-C than the palmitate diet (124 ± 17, 117 ± 18 and 106 ± 13 mg/dL for the control, palmitate and trans diets, respectively). The palmitate diet significantly decreased the TC/HDL-C ratio by 11% when compared with the control diet (1.68 ± 0.17 vs. 1.89 ± 0.30), whereas the trans diet had no effect (1.81 ± 0.20 vs. 1.89 ± 0.30). Kinetic studies revealed that, relative to the control diet, both test diets significantly lowered the LDL apolipoprotein B (apoB) pool size, principally reflecting an increase in the LDL apoB fractional catabolic rate (FCR) related to the reduced cholesterol intake. Between the two test diets, no significant differences in LDL kinetic parameters were observed. Both test diets significantly decreased HDL apoA1 concentrations in comparison with the control diet, which was partly explained by an increase in the fractional catabolic rate of HDL. Of the two test diets, the trans diet was associated with a 9.5% greater HDL FCR than the palmitate diet (P < 0.08) and a significant increase in plasma cholesteryl ester transfer protein (CETP) activity (% transfer 114 ± 7 vs. 91 ± 7; P < 0.03). Thus, palmitic acid- and elaidic acid-rich diets produced identical effects on LDL metabolism in normocholesterolemic cebus monkeys fed diets with low levels of cholesterol, whereas elaidic acid depressed HDL-C, attributable to both increased CETP activity and HDL clearance.
INTRODUCTION
In recent years, the effects of dietary trans fatty acids produced by the hydrogenation of liquid vegetable oils on plasma lipid metabolism have become the subject of much scrutiny and debate (Khosia and Hayes 1996, Kris-Etherton 1995
, Willett and Ascherio 1994, Willett et al. 1993). Although recent clinical studies have documented the cholesterolemic effects of trans isomers relative to their cis counterparts (Judd et al. 1994, Lichtenstein et al. 1993, Mensink and Katan 1990, Nestel et al. 1992a and 1992b, Zock and Katan 1992), detailed information on the response following direct exchange between trans fatty acids and saturated fatty acids (SFA)6 remains sparse. The latter exchange is an important consideration, because current dietary guidelines recommend replacing SFA with monounsaturated fatty acids (MUFA), although the guidelines do not specifically distinguish between cis or trans MUFA. Additionally, because several studies suggest that all SFA are not equally cholesterolemic (Hayes and Khosla 1992, Hayes et al. 1991, Hegsted et al. 1965, Mensink and Katan 1992, Pronczuk et al. 1994, Sundram et al. 1994, Zock et al. 1994), it is important to establish how dietary trans fatty acids affect plasma lipid metabolism in comparison with specific SFA. Accordingly, the current study was conducted to ascertain the effects of replacing 6% dietary energy (%en) from palmitic acid (the most abundant SFA) with 6%en from elaidic acid (t-C18:19), as part of an American Heart Association (AHA) Step 1 diet in normocholesterolemic cebus monkeys that have been shown to be highly responsive to changes in dietary fatty acids (Khosla and Hayes 1992). Diets were formulated by using different blends of vegetable oils such that other key dietary fatty acids (notably oleic, linoleic and myristic acid) were held constant between the two test diets.
MATERIALS AND METHODS
) and were initially fed a purified control diet (Table 1) rich in saturated fat and cholesterol for 14 wk. The control diet approximated the average American diet in terms of the total fat (38%en), polyunsaturated/saturated (P/S) ratio (0.57) as well as dietary cholesterol content (180 mg/1000 kcal diet). Subsequently, monkeys were divided into two groups and fed one of two test diets formulated according to the guidelines for an American Heart Association Step 1 diet, in which total fat and cholesterol were reduced to ~30%en and 100 mg/1000 kcal diet, respectively, and the P/S ratio increased substantially. The only difference between the two test diets was a 6%en exchange between palmitic and elaidic acid, which consequently shifted the P/S ratio from 0.94 to 2.22, respectively. The source of the elaidic acids was a specially prepared partially hydrogenated soybean oil, unusually rich in t-C18:19 (elaidic acid), provided by K. Sundram [Palm Oil Research Institute of Malaysia (PORIM), Kuala Lumpur, Malaysia; Sundram et al. 1997] from the identical lot of industrially prepared oil fed simultaneously to humans. Fatty acid analysis of the hydrogenated fat, provided by PORIM, revealed that the t-C18:19 isomer accounted for ~80% of the total trans isomers, and other unspecified isomers of t-C18:1 and t-C18:2 accounted for the rest. Based on this information, diets were formulated as indicated in Table 1 to achieve the appropriate fatty acid exchanges (Table 2), and the formulations were verified by gas liquid chromatography (GLC) analysis of the diets, as detailed previously (Pronczuk et al. 1991). After 14 wk of consuming the test diets, monkeys were switched to the opposite diet and feeding continued for an additional 14 wk. During all three dietary phases (1 control + 2 test periods), plasma lipid concentrations and lipoprotein metabolism were evaluated.
|
Table 1. Composition of purified diets |
|
Table 2. Percentage of total dietary calories contributed by each fatty1 acid |
, Khosla and Hayes 1992 and 1993, Pronczuk et al. 1991), each monkey was fed a fixed amount of diet each day (~170-220 g/d as a starch-gel) that ensured maintenance of a constant body weight. However, because of the different energy densities of the control and test diets, each animal received a fixed amount of total calories during the entire study (~800 kcal/d for each monkey; range 700-950 kcal). Additionally, the protein and micronutrient content of the control and test diets was adjusted such that all animals also received the same quantity of nutrients during all phases of the study. All procedures and protocols were in accordance with the University's Animal Use and Radiation Safety Committees.
, Schumaker and Puppione 1986).
) as detailed previously (Khosla and Hayes 1992). The isolated LDL and HDL were washed and concentrated by recentrifugation at their appropriate densities. Following dialysis (0.15 mol/L NaCl:1 mmol/L EDTA, pH 7.4), lipoprotein protein concentration was determined using Markwell's modification (Markwell et al. 1978) of the Lowry procedure (Lowry et al. 1951). LDL and HDL were radiolabeled with Na131I and Na125I (Amersham, Chicago, IL) respectively, (to specific activities of 300 and 250 cpm/ng), and the intramolecular distribution of radioactivity was determined (Khosla and Hayes 1992). For the 131I-LDL, the proportion of total radioactivity associated with apolipoprotein B (apoB) was 95 ± 1% whereas the lipid-bound radioactvity was <2%. ApoA1 accounted for 82% of the 125I-HDL radioactivity and >90% of the HDL protein mass, quantified as described previously (Khosla and Hayes 1992).
and 1993). Briefly, animals were injected simultaneously with 185-370 kBg of homologous 131I-LDL and 125I-HDL. A 5-mL blood sample was collected from a femoral vein 10 min postinjection, and additional 1-mL samples obtained periodically up to 8 d. (The amount of blood taken from each monkey over the 8-d period averaged <10% of the estimated blood volume.) LDL and HDL were isolated by sequential ultracentrifugation from the 10-min plasma sample of each monkey (obtained following a 16-h fast) and the LDL apoB and HDL apoA1 concentrations determined (Khosla and Hayes 1992).
Kinetic analyses.
Plasma 131I and 125I radioactivity data for both tracers were bi-exponential; the data were analyzed in accordance with a two-pool model (Matthews 1957), and the FCR calculated (Kushwaha and Hazzard 1977). The assumptions and rationale for using the two-pool model, as well as the calculations for pool sizes (mg/kg body weight) and transport rates (TR) [mg/(kg·d)] for each tracer have been detailed previously (Khosla and Hayes 1992 and 1993). Additionally, the limitations of the analyses in using HDL apoA1 concentrations and the FCR for whole HDL as a measure of the transport rate of HDL apoA1 have also been discussed (Khosla and Hayes 1993).
Plasma cholesteryl ester transfer protein (CETP) activity.
The analysis was conducted essentially as described by Tato et al. (1995) with the following two exceptions: 1) normocholesterolemic human plasma was used as the source of donor and acceptor lipoproteins, and 2) the mixture for separating HDL from non-HDL lipoproteins used 92 mmol/L MnCl2 and 0.22 mg/mL of heparin. CETP activity in monkey plasma samples (50 µL) was determined following incubations with 3H-cholesteryl ester-labeled HDL3 and LDL. Radioactivity transferred from 3H-HDL3 to LDL (measured in the supernatant following precipitation with heparin/MnCl2+) was used to calculate CETP activity (expressed as the percentage of radioactivity transferred from 3H-HDL3 to LDL per 16 h of incubation). CETP measurements (on plasma samples from the two test diets) were conducted in a single assay, using aliquots of frozen plasma (
70oC), collected after 8 wk of diet consumption.
Statistical analyses.
All statistical analyses were performed using a Power Macintosh 6100 computer (Apple Systems, Cupertino, CA) with the Statview 512+ (Brain Power, Calabasca, CA) statistical package. Significant differences were calcualted using repeated measures ANOVA. Results are presented as means ± SD.
RESULTS
|
Table 3. Plasma lipid concentrations1 |
|
Table 4. LDL kinetic parameters1 |
|
Table 5. HDL kinetic parameters1 |
DISCUSSION
, Mensink and Katan 1989, Nestel et al. 1992b, Zock and Katan, 1992). However, data on the relative effects between trans fatty acids, both collectively and as individual trans isomers, and SFA remain unclear. The availability of a partially hydrogenated soybean oil preparation, in which one trans isomer (elaidic acid) predominated, allowed us to directly compare elaidic and palmitic acids, whereas all other fatty acids, specifically, myristic, oleic and linoleic acids were kept constant. The control of "other fatty acids" is an important point because it was recently proposed (Nicolosi and Dietschy 1995) that one of the reasons why trans fatty acids appeared to be cholesterolemic in earlier human studies (Judd et al. 1994, Mensink and Katan 1989, Nestel et al. 1992b, Zock and Katan 1992) was that the trans fatty acid-enriched diets also had depressed oleic or linoleic acid content in comparison with the respective control diet. By equalizing the amount of both oleic and linoleic acid between our test diets, we eliminated this confounding variable. Additionally, by formulating diets according to the AHA guidelines, we also addressed the issue of whether trans fatty acids represent an equitable replacement for SFA in conventional lipid-lowering regimens. Our data clearly show that under these conditions, the primary effect of replacing palmitic acid with elaidic acid in the cebus monkey fed a Step 1 diet is a lowering of HDL-C. A similar effect on HDL-C was observed in the parallel human study in which the same trans fat preparation was fed (Sundram et al. 1997). While it may be argued that the decreased HDL observed during trans fatty acid intake might have been due to the relative absence of palmitic acid, a previous study in these monkeys revealed that plasma lipid levels (including HDL-C) were unaffected when 10%en from palmitic acid was replaced with an equivalent amount of oleic acid (Khosla and Hayes 1993). Thus, the cebus monkey, an animal with ~60% of its TC in the HDL fraction, provided a clear focus on the HDL component of the trans effect. In contrast to the human study (Sundram et al. 1997), a simultaneous increase in LDL-C was not observed in these monkeys, for whom LDL-C represents a much smaller percentage of lipoprotein cholesterol.
, Van Tol et al. 1995), suggesting that feeding trans fatty acids increases cholesteryl ester transfer from HDL to the apoB-containing lipoproteins, thereby lowering HDL-C. The dietary trans fatty acids incorporated into foods of the Australian study (Abbey and Nestel 1994) were reportedly rich (not actually measured) in elaidic acid, whereas a mixture of trans isomers was fed in the Dutch preparation (Van Tol et al. 1995). In addition, our data suggest that part of the decrease in HDL-C induced by trans fatty acids can be attributed to accelerated HDL clearance. A major difference between the current study and the corresponding human study utilizing the same trans fat preparation (Sundram et al. 1997) was the failure to elevate the LDL concentration or to perturb LDL kinetic parameters following the palmitic/elaidic acid exchange, clearly indicating that palmitic and elaidic acid had identical effects on LDL metabolism in the cebus monkeys. LDL clearance was more efficient with the test diets compared with consumption of the control diet, which contained more saturated fat (12:0 + 14:0-rich) and cholesterol (80% more). Recent studies from this laboratory have provided substantial evidence that palmitic acid can appear like oleic acid (or even linoleic acid) in normocholesterolemic animals consuming diets with low levels of cholesterol and myristic acid (Hayes and Khosla 1992, Hayes et al. 1991, Khosla and Hayes 1991, 1992 and 1993, Pronczuk et al. 1991). Based on these cebus monkey data, we would conclude that the primary effect of elaidic acid does not appear to be on LDL metabolism. This conclusion concurs with the hamster data of Woollett et al. (1994), but would seem to contradict the human data showing trans fatty acids increase LDL-C relative to oleic acid (Judd et al. 1994, Mensink and Katan 1989, Nestel et al. 1992b, Sundram et al. 1997, Zock and Katan 1992).
), the Kd for LDL-C uptake via the receptor-dependent process is ~90 mg/dL. Thus receptor-dependent LDL-C uptake is "half-saturated" at a plasma LDL-C concentration of 90 mg/dL. In the various human studies (Judd et al. 1994, Mensink and Katan 1989, Nestel et al. 1992b, Sundram et al. 1997, Zock and Katan 1992), the LDL-C of subjects receiving trans fatty acids was considerably higher than 90 mg/dL. Thus, in these subjects, receptor-dependent uptake was in all likelihood suppressed more than 50%. In such situations, increased cholesteryl ester transfer from HDL to LDL (mediated by CETP), coupled with partially depressed LDL receptor activity, would result in delayed clearance of the expanded LDL pool, thereby resulting in an elevated LDL-C. In contrast, the LDL-C concentration was inherently below 90 mg/dL (~60-70 mg/dL) in our normolipemic cebus monkeys, as well as in the previously mentioned hamsters (Woollett et al. 1994). Because the LDL-C concentration increases significantly only after LDL receptors have been suppressed >50%, the failure of elaidic acid consumption to increase LDL-C under our dietary conditions implies that LDL clearance was not impaired in the two test diets (which had negligible levels of 12:0 + 14:0 and low dietary cholesterol content), even though CETP activity was increased. Accordingly, the effect of dietary trans fatty acids on LDL metabolism may depend, in part, on the LDL receptor activity of the host. In agreement with this concept is a preliminary report (McNamara et al. 1995), which found that replacing 4%en from oleic acid with elaidic acid failed to affect plasma lipid and lipoprotein levels, hepatic apoB/E receptor number, or in vivo LDL clearance when guinea pigs were fed low cholesterol diets (0.04% wt/wt). By contrast, in the presence of 0.25% wt/wt dietary cholesterol (which is sufficient to down-regulate LDL receptors by >50% in guinea pigs), the same fatty acid exchange resulted in increased LDL-C, associated with a further suppression of hepatic apoB/E receptor number and reduced in vivo LDL clearance. Dietary cholesterol also exacerbated the hyperlipemic effect of trans fatty acid intake in gerbils (Dictenberg et al. 1995).
). The effectiveness of these latter diets in improving the plasma lipid profile may therefore depend on reductions in 12:0 + 14:0, dietary cholesterol or both. Concurrent data from rhesus monkeys, fed diets approximating the average American diet or AHA Step 1 diets, rich in either 12:0 + 14:0 or 16:0, are in agreement with this conclusion (Khosla et al., 1997).
8, t-C18:110) affect lipoprotein metabolism in a similar manner has yet to be established.
FOOTNOTES
LITERATURE CITED
This article has been cited by other articles:
|
|
 |
|
  S. K. Cheema, A. Agarwal-Mawal, C. M. Murray, and S. Tucker Lack of stimulation of cholesteryl ester transfer protein by cholesterol in the presence of a high-fat diet J. Lipid Res., November 1, 2005; 46(11): 2356 - 2366. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
