![[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}
|
|
|
|
|
||
Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN 55108
ABSTRACT
INTRODUCTION
METHODS AND MATERIALS
RESULTS
DISCUSSION
FOOTNOTES
LITERATURE CITED
ABSTRACT 
Although there is general agreement that saturated fatty acids elevate plasma cholesterol concentrations, the relative effects of individual fatty acids on cholesterol and bile acid metabolism are less clear. In this study, cholesterol and bile acid responses to diets enriched in different saturated fatty acids were investigated in hamsters. The six diets examined were as follows: 5% fat (g/100 g) enriched in palmitic acid (16:0) with no cholesterol, 5% fat 16:0-enriched, 0.05% cholesterol (wt/wt), and four diets containing 0.05% cholesterol and 15% fat with each diet enriched in lauric (12:0), myristic (14:0), palmitic (16:0), or stearic acid (18:0). Total plasma cholesterol concentration was significantly greater in hamsters fed the 14:0-enriched diet relative to those fed the 18:0-enriched diet (P < 0.05). Both plasma and liver cholesterol concentrations of hamsters fed 18:0 did not differ from those of the group fed no dietary cholesterol. In all instances, differences in total plasma cholesterol were accounted for within the HDL fraction; no significant treatment differences in VLDL or LDL cholesterol were found. Total daily fecal bile acid excretion was higher in hamsters fed the 15% fat 16:0 diet compared with those fed no dietary cholesterol (P < 0.05), but not significantly different from other treatment groups. There was greater deoxycholic acid excretion (P < 0.05) from hamsters fed the 14:0 and 16:0 diets compared with those fed the 18:0-enriched diet. Small intestinal + gallbladder bile acids, an index of pool size, did not differ significantly among the groups. The observed relative hypocholesterolemic effect of stearic acid was not mediated by increased bile acid excretion.
KEY WORDS: cholesterol · bile acids · hamsters · saturated fatty acids · lipoproteinsINTRODUCTION
Although fatty acid chain length is well documented as an important determinant of dietary cholesterolemic response, disagreement persists concerning the relative hypercholesterolemic potency of individual saturated fatty acids. In 1965, Keys et al. interpreted lauric (12:0), myristic (14:0) and palmitic (16:0) acids as exerting similar hypercholesterolemic effects in humans when fed on an equivalent energy basis (%); this hypothesis has since been supported in studies with hamsters (Woollett et al. 1992
). In contrast, Hegsted et al. (1965) believed myristic acid to be the most potently hypercholesterolemic saturated fatty acid, a hypothesis supported in studies of nonhuman primates by Hayes et al. (1991) and more recently by the human study of Zock et al. (1994). Additionally, Hayes and colleagues (Hayes et al. 1991, Khosla and Hayes 1992) have proposed that palmitic acid is not hypercholesterolemic relative to oleic acid under conditions of low dietary cholesterol intake and/or low plasma LDL cholesterol concentrations. There is more consistent evidence supporting hypotheses that lauric acid is less hypercholesterolemic than palmitic acid (Denke and Grundy 1992, Hegsted et al. 1965, McGandy et al. 1970) and that stearic acid (18:0) is hypocholesterolemic relative to myristic or palmitic acid (Bonanome and Grundy 1988, Denke and Grundy 1991, Hegsted et al. 1965, Imaizumi et al. 1993, Keys et al. 1965, Woollett et al. 1992). Medium-chain saturated fatty acids (6:0, 8:0, 10:0) appear to have little or no effect on plasma cholesterol concentrations in either humans (Grande 1962) or hamsters (Woollett et al. 1992), although more recent data from humans are lacking.
Characterization of cholesterolemic effects of specific fatty acids is complicated by the fact that commonly consumed dietary triacylglycerols are composed of fatty acids that may vary considerably in their chain length, degree of unsaturation, isomeric orientation of double bonds and position within the triacylglycerol molecule. Interpretation of results is further obscured because effects can be described only in relative terms, and outcomes may be greatly dependent upon the specific experimental conditions employed. A number of investigators have attempted to circumvent some of these limitations by using interesterification processing to obtain dietary fats with desired fatty acid compositions (Imaizumi et al. 1993, McGandy et al. 1970, Woollett et al. 1992, Zock et al. 1994). Although this approach facilitates optimization of experimental comparisons, it also introduces a potentially confounding factor of altered triacylglycerol structure (Kritchevsky 1988). Disparities in results obtained from naturally occurring vs. semisynthetic fat sources have been documented (McGandy et al. 1970), and the relevance of these observations to free-living humans has been questioned (Kritchevsky 1988). Also, little is known about the mechanisms by which the dietary fatty acid structure elicits metabolic responses within the liver, although previous work has characterized some of the metabolic events associated with cholesterolemic responses to dietary saturated fatty acids (Imaizumi et al. 1993, Khosla and Hayes 1991, Spady and Dietschy 1988, Woollett et al. 1992).
We attempted here to further define hepatic and plasma responses to saturated fatty acid chain length by employing a combination of naturally occurring triacylglycerol sources. In addition, we examined the effects of dietary cholesterol and dietary fat quantity. Accordingly, we fed diets varying in fatty acid content and profile, examining their effects on plasma and hepatic cholesterol levels as well as bile acid pool size and excretion. Hamsters were used as the animal model because their response to dietary fatty acids is similar in direction and magnitude to that of humans (Spady and Dietschy 1988) and because the metabolic responses to diets containing fatty acids and cholesterol have been well documented (Imaizumi et al. 1993, Woollett et al. 1992).
METHODS AND MATERIALS
|
Table 1. Diet composition1 |
|
Table 2. Fatty acid distribution of the diets |
20°C until analyzed. Food was withheld for 12 h, hamsters were anesthetized with ethyl ether, and blood was collected via cardiac puncture in a syringe containing 1 g EDTA/L blood. Livers were perfused with normal saline, excised, weighed and stored at 20°C until analyzed. Gallbladders and small intestines were excised and combined for each animal, homogenized with double-distilled deionized H2O, freeze-dried and stored at 20°C until analyzed. Whole blood was centrifuged at 1200 × g for 20 min at 4°C and plasma collected. Aliquots were taken for lipoprotein fractionation and the remainder frozen at 20°C until assayed.
Lipids and lipoproteins.
Fatty acid profiles of the dietary fats and oils were determined via gas chromatography (Hewlett-Packard Model 5890A, Wilmington, DE) after saponification, as described by Einig (1987) (Table 2). Total plasma cholesterol concentration was determined using an enzymatic procedure (kit #352-20, Sigma Chemical, St. Louis, MO). The remaining plasma from sets of two hamsters was pooled for lipoprotein separation via sequential ultracentrifugation based on the method of Havel et al. (1955) using a Ti70 rotor (model L2-65B, Beckman Instruments, Spinco Division, Palo Alto, CA) at 15°C. Plasma was separated sequentially into fractions containing VLDL (d < 1.006 kg/L), LDL (1.006 < d < 1.055), and HDL (1.055 < d < 1.225) by successive centrifugation for 24 h at 100,000 × g. Densities of plasma above 1.006 kg/L were adjusted by addition of solid potassium bromide, and plasma was overlayered with a buffer solution of the same density before ultracentrifugation. In each fraction, cholesterol and triglyceride (kit #337, Sigma Chemical) concentrations were enzymatically determined and protein concentrations were measured using the bicinchoninic acid protein assay (Smith et al. 1985). HDL apolipoproteins were qualified by SDS-PAGE using gradient gels (7.5-20% acrylamide) by the procedure of Laemmli (1970). Lipids were extracted from ~1 g of liver by the method of Folch et al. (1957). The lipid extracts were then assayed enzymatically for total and free cholesterol concentrations by the method of Warnick et al. (1982). Esterified cholesterol concentration was calculated by subtracting the free cholesterol from the total cholesterol measured for each animal.
Bile acids.
Bile acids were extracted and partially purified following the method of Locket and Gallaher (1989). Reversed-phase HPLC was used to quantify individual bile acids present in the extracts (Gallaher et al. 1992). Briefly, a stepwise gradient elution system was used, composed of two mobile phases, ammonium dihydrogen phosphate (10 mmol/L, pH 7.8) and acetonitrile. Detection was achieved by use of a second column containing immobilized 3-
-hydroxysteroid dehydrogenase (Sigma Chemical). A buffer containing NAD (0.1 mol/L Tris-HCL, pH 8.5, 2.7 mmol/L EDTA, 0.82 mmol/L dithiothreitol and 0.5 mmol/L NAD) was introduced by means of a tee between the first and second columns at a constant rate of 1 mL/min. NADH produced by the reaction of bile acids and NAD+ with the immobilized enzyme was detected fluorometrically. Peak areas were calculated and bile acids were quantified using detector response factors determined with known standards. Bile acids were extracted and partially purified from small intestine + gallbladder samples following the procedure used for fecal bile acids. Bile acids were deconjugated by overnight incubation at 37°C with choloylglycyl hydrolase (Sigma Chemical) and repurified prior to analysis by HPLC to simplify identification of peaks. Individual bile acids were quantified via HPLC as described above.
Statistics.
One-way ANOVA was used to determine treatment effects (SAS/GLM Version 5.18, SAS Institute, Cary, NC). Differences among means were inspected using Duncan's multiple range test (Duncan 1955) and were considered to be significant at P < 0.05. Values are reported as means ± SEM.
RESULTS
|
Table 3. Body weight, food intake, and fecal weight of hamsters fed diets varying in cholesterol, fat level, and saturated fat type for 6 wk1 |
|
Table 4. Plasma and liver cholesterol concentrations in hamsters fed diets varying in cholesterol, fat level and saturated fat type for 6 wk1 |
-hydroxy-12-keto-5
-cholanoic acid, were significantly lower in the 18:0-enriched diet group compared with the high fat 16:0-enriched diet group, and deoxycholic acid excretion was lower in the 18:0-enriched diet group compared with the high fat 14:0-enriched diet group. Among the other individual bile acids studied, no fatty acid chain length effects were observed.
|
Table 5. Daily fecal bile acid excretion in hamsters fed diets varying in cholesterol, fat level, and saturated fat type for 6 wk1 |
-hydroxy-12-keto-5-cholanoic acid, and 3,7-dihydroxy-12-keto-5-cholanoic acid. Although cholesterol feeding alone had no effect on total bile acid output, it did increase the daily excretion of deoxycholic acid. Similarly, increasing dietary fat quantity increased the daily output of cholic acid, deoxycholic acid and 3-hydroxy-12-keto-5-cholanoic acid.
-hydroxy-12-keto-5-cholanoic acid, 0.15 ± 0.00; 3,7-dihydroxy-12-keto-5-cholanoic acid, 0.64 ± 0.04; sum, 14.82 ± 0.68. However, there was a significant correlation between the group means of total daily fecal bile acid excretion and the sum of bile acids in the small intestines + gallbladders (R2 = 0.78, P < 0.02).
DISCUSSION
, Hegsted et al. 1965, Keys et al. 1965, Woollett et al. 1992), although disagreement exists regarding the extent to which each of these fatty acids is implicated. This issue is important to understand because, among other reasons, it is now possible to modify the abundance of specific dietary fatty acids in the food supply through production and processing technologies. However, obtaining a definitive answer to this question has proven difficult because of a number of factors. First, dietary triacylglycerols are comprised of fatty acids that vary considerably in both chemical structure and position within the triacylglycerol molecule. Resolving independent effects of fatty acid chain length without altering triacylglycerol structure has yet to be accomplished. Of particular concern is the difficulty associated with resolving the independent effects of lauric and myristic acids, because common sources of lauric acid (coconut oil, palm kernel oil and milkfat) also contain appreciable quantities of myristic acid. Several recent investigations (Imaizumi et al. 1993, Woollett et al. 1992, Zock et al. 1994) have attempted to circumvent this problem through the use of interesterification processing to obtain dietary fats with specific fatty acid profiles. Although this approach facilitates comparison of independent fatty acid effects, it also alters triacylglycerol structure. McGandy et al. (1970) indicated that feeding such semisynthetic fat sources might significantly influence the cholesterolemic effects observed for individual fatty acids, whereas Zock et al. (1994) suggested otherwise. Second, results may be greatly dependent upon the specific experimental conditions or species employed. For example, other dietary factors such as availability of cholesterol (Spady and Dietschy 1988) and the abundance of linoleic acid (Khosla and Hayes 1992) influence experimental outcomes. Third, because fatty acid sources must be substituted for one another to avoid confounding effects from other dietary variables, experimental effects can be described only in relative terms. Failure to appreciate this limitation may obscure appropriate interpretation of data. Finally, conclusions regarding independent fatty acid effects often are based upon respective coefficients derived from multiple regression analysis of many feeding experiments (Hegsted et al. 1965, Keys et al. 1965, Khosla and Hayes 1992, Mensink and Katan 1992, Pronczuk et al. 1994). For example, several reports (Hayes et al. 1991, Hegsted et al. 1965, Mensink and Katan 1992) have suggested that myristic acid is several times more potent than palmitic acid in exerting hypercholesterolemic effects when compared on an equivalent energy basis. Such conclusions may be misleading because a high degree of codependence among individual saturated fatty acids is inherent in such studies (Keys et al. 1965).
found that lauric, myristic and palmitic acids were approximately equivalent in their cholesterolemic effects, and that stearic acid was hypocholesterolemic by comparison. Imaizumi et al. (1993) found no significant differences in plasma cholesterol response to lauric, myristic and palmitic acid in hamsters at either 0 or 0.2 g/100 g dietary cholesterol (the latter being four times the concentration used here), but did observe a relative hypocholesterolemic response for stearic acid in both liver and plasma. By contrast, Lindsey et al. (1990) fed hamsters a variety of saturated fatty acid profiles comprised of naturally occurring triacylglycerol sources with no added dietary cholesterol and found that 12:0 + 14:0 was hypocholesterolemic relative to 16:0. The majority of available data suggest that over a fairly wide range of dietary cholesterol concentrations, lauric, myristic and palmitic acids exert similar cholesterolemic effects.
) but agrees with others (Lindsey et al. 1990). This is likely due to differences in the basal diet, because those studies reporting large treatment effects on LDL cholesterol have used a ground nonpurified rodent diet, whereas our study and that of Lindsey et al. (1990) used a semipurified diet. The reasons why different basal diets would lead to differences in response of the LDL fraction are obscure.
, Woollett et al. 1992) and humans (Bonanome and Grundy 1988, Hegsted et al. 1965, Keys et al. 1965). The relative hypocholesterolemic effect of stearic acid could potentially be mediated by an increase in bile acid excretion. However, the results of the present study do not support this possibility. There were no significant differences in bile acid pool size or fecal bile acid excretion between the group fed the stearic acid-enriched diet and the groups fed diets enriched with lauric, myristic or palmitic acid, with the exception of a reduction in deoxycholic acid excretion in the stearic acid groups relative to the other groups. Thus, an enhancement of bile acid excretion does not appear to be an important factor in the hypocholesterolemic effect of stearic acid.
suggests that different long-chain fatty acids alter hepatic LDL receptor activity and thereby influence LDL cholesterol concentration. It was suggested that the type of long-chain fatty acid influences the distribution of intracellular sterol between a putative sterol regulatory pool and the cholesteryl ester pool. However, in the present study, large differences in plasma and liver cholesterol concentrations were found, resulting solely from differences in fatty acid chain length, in the absence of any differences in LDL cholesterol concentration. Thus, changes in LDL receptor activity are unlikely to be involved in the changes in cholesterol metabolism observed. The hypocholesterolemic effect of stearic acid could be mediated by a decrease in cholesterol synthesis, and although rates of cholesterol biosynthesis were not determined in the present experiment, this prospect appears unlikely for at least two reasons. First, under conditions of cholesterol feeding, cholesterol biosynthetic rates are significantly down-regulated (Spady and Dietschy 1988) such that the remaining capacity for additional suppression because of stearic acid would be quite limited. Second, under most circumstances, the activities of hepatic hydroxy methylglutaryl (HMG) CoA reductase and cholesterol 7-hydroxylase vary in parallel (for review, see Vlahcevic et al. 1991). If stearic acid reduces cholesterol synthesis, a decrease in 7-hydroxylase might result, leading to a decrease in the bile acid synthesis rate. The lack of changes in the bile acid pool size or total bile acid excretion in hamsters fed the stearic acid-enriched diet suggests that cholesterol synthesis rates were not altered. More likely explanations for the relative hypocholesterolemic effect of stearic acid arise from data of Imaizumi et al. (1993), who reported greater neutral sterol excretion in hamsters fed fats enriched in stearic acid compared with fats enriched in lauric, myristic or palmitic acid at either 0 or 0.2 g/100 g dietary cholesterol. Fecal neutral sterol excretion rates are influenced primarily by dietary cholesterol concentration, cholesterol absorption efficiency and biliary cholesterol excretion. Stearic acid-rich fats are incompletely absorbed (Imaizumi et al. 1993), which could result in some cholesterol remaining dissolved in the lipid phase of the intestinal contents instead of being solubilized into micelles, a necessary step for absorption. Stearic acid may also influence metabolic events within the hepatocyte so that more cholesterol is directed toward biliary excretion. Either of these possibilities could result in increased neutral sterol excretion.
found no effect of 0.2 g/100 g added cholesterol on daily bile acid excretion. Rats fed diets containing 2 g/100 g cholesterol excreted significantly greater quantities of bile acids and had greater activity of cholesterol 7-hydroxylase activity compared with rats fed a cholesterol-free diet. However, no elevation in 7-hydroxylase activity was found when the cholesterol concentration of the diet was <1 g/100 g (Björkhem et al. 1991). Duane (1994) recently reported a slight but significant increase in fecal bile acid excretion and cholic acid synthesis in men fed a high cholesterol diet relative to a low cholesterol diet. Our results suggest that in hamsters, dietary cholesterol increases hepatic cholesteryl ester concentration and possibly, bile acid excretion.
, Reddy et al. 1974), whereas others found no change (Brussaard et al. 1983, Gallaher and Franz 1990, Gallaher et al. 1992). In the present study, total fecal bile acid excretion did not increase significantly with increasing level of fat; however, excretions of cholic acid, deoxycholic acid and 3-hydroxy-12-keto-5-cholanic acid were significantly increased. Total daily fecal bile acid excretion was significantly increased only when the fat level was increased and cholesterol was added to the diet (5NCh16:0 vs. 15Ch16:0). Chang et al. (1994) apparently found similar results in rats; total fecal bile acid concentration increased as the dietary fat level increased from 5 to 10%. Because the fat source was butter, the cholesterol content of the diet was also increased. Because both diet 15Ch16:0 and butter contain palmitate as the major saturated fatty acid, this suggests that a diet high in palmitate and containing cholesterol is necessary for a substantial increase in fecal bile acid excretion to occur.
reported an increase in the pool size in pigs when the fat level was increased from 2 to 10 g/100 g, but no further increase in pool size when the fat level was increased to 20 g/100 g. In rats, no difference in bile acid pool size was found between those fed diets of 5 and 20 g/100 g fat (Gallaher et al. 1992). Our results showing no difference in pool size between hamsters fed diets of 5 and 15% fat suggest that within the range of dietary fat normally consumed, the bile acid pool size is quite constant in hamsters fed the AIN-76-based purified diet.
FOOTNOTES
Manuscript received 5 September 1996. Initial reviews completed 16 October 1996. Revision accepted 4 February 1997.
LITERATURE CITED
-hydroxylase by dietary cholesterol.
Biochim. Biophys. Acta
1991;
1085:329-335
[Medline]
This article has been cited by other articles:
|
|
 |
|
  H. E. Rasmussen, D. M. Guderian Jr., C. A. Wray, P. H. Dussault, V. L. Schlegel, and T. P. Carr Reduction in Cholesterol Absorption Is Enhanced by Stearate-Enriched Plant Sterol Esters in Hamsters J. Nutr., November 1, 2006; 136(11): 2722 - 2727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Cowles, J.-Y. Lee, D. D. Gallaher, C. L. Stuefer-Powell, and T. P. Carr Dietary Stearic Acid Alters Gallbladder Bile Acid Composition in Hamsters Fed Cereal-Based Diets J. Nutr., October 1, 2002; 132(10): 3119 - 3122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Schneider, R. L. Cowles, C. L. Stuefer-Powell, and T. P. Carr Dietary Stearic Acid Reduces Cholesterol Absorption and Increases Endogenous Cholesterol Excretion in Hamsters Fed Cereal-Based Diets J. Nutr., May 1, 2000; 130(5): 1232 - 1238. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. P. Carr, G. Cai, J.-Y. Lee, and C. L. Schneider Cholesteryl Ester Enrichment of Plasma Low-Density Lipoproteins in Hamsters Fed Cereal-Based Diets Containing Cholesterol Experimental Biology and Medicine, January 1, 2000; 223(1): 96 - 101. [Abstract] [Full Text]
|
|
|
|
|
|
S. K. Cheema and L. B. Agellon Metabolism of Cholesterol Is Altered in the Liver of C3H Mice Fed Fats Enriched with Different C-18 Fatty Acids J. Nutr., September 1, 1999; 129(9): 1718 - 1724. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
