composition in hamsters. MATERIALS AND METHODS. Animals and diets. Male Syrian hamsters weighing 60 g (Charles River Laboratories, Montreal, Quebec) ...
815
Influence of Dietary Fatty Acid Composition on Cholesterol Synthesis and Esterification in Hamsters Peter J.H. Jones*, JuUe E. Ridgen and Alexander P. Benson Division of Human Nutrition, School of Family and Nutritional Sciences, University of British Columbia, Vancouver, British Columbia, V6T IW5 Canada
To investigate the effects of dietary fat quality on synthesis and esterification of cholesterol, Syrian hamsters were fed diets containing corn, olive, coconut or menhaden oils (10% w/w) with added cholesterol (0.1% w/w). After 3 weeks, animals were sacrificed 90 min following IP injection of 3H20. Synthesis of free chw lesterol and movement of free cholesterol into ester pools were measured from 3H-uptake rate in liver and duodenum. Plasma total cholesterol and triglycerides levels were highest in coconut oil-fed animals, whereas hepatic total cholesterol and ester levels were elevated in olive oil-fed animals, as compared with all other groups. No diet-related differences were seen in duodenal cholesterol or total fatty acid content. In duodenum, uptake of 3H per g t i s s u e into cholesterol was greater compared with liver; however, within each tissue, 3H-uptake into cholesterol was similar across groups. Notably, 3H-uptake into cholesterol ester in liver was highest in menhaden oil-fed animals. These data suggest that menhaden fish oil consumption results in enhanced movement of newly synthesized cholesterol into ester as compared with other fat types. Lipids 25, 815-820 (1990).
responsible remain to be established. Several mechanisms have been proposed. Data from animal experiments suggest that co6 PUFA reduce plasma cholesterol concentrations by altering hepatic low density lipoprotein (LDL) metabolism, redistributing cholesterol from blood to tissue pools (11). Alternatively, co6 PUFA may increase hepatic cholesterol esterification, which reduces the free cholesterol pool size and elevates LDL receptor synthesis (12). Conversely, fish oil fatty acids are reported to inhibit release of very low density lipoprotein (VLDL) from liver (13). Such findings emphasize the need to study the influence of qualitative fat intake on cholesterol turnover to better understand the role of fatty acid intake in controlling plasma cholesterol levels. The purpose of the present study was to examine specific effects of feeding common dietary fats on organ cholesterol synthesis and esterification and plasma and organ lipid composition in hamsters.
MATERIALS AND METHODS
Animals and diets. Male Syrian hamsters weighing 60 g (Charles River Laboratories, Montreal, Quebec) were maintained in a windowless room artificially illuminated from 21:00 to 09:00 hr and fed laboratory diet Dietary fatty acid composition has been identified as (Rodent Laboratory Diet 5001, Purina Mills Inc., St. influencing plasma cholesterol levels in humans. Satu- Louis, MO) and water ad libitum for two weeks. The rated dietary fatty acids are categorically most hyper- gross diet composition was (w/w): protein, 23%; fat, cholesterolemic (1-3). Conversely, monounsaturated 4.5%; fiber, 5.8%; ash, 7.3%; with the remainder as fatty acids (MUFA) {3,4) and polyunsaturated fatty carbohydrate, vitamin and mineral mixes and moisacids {PUFA) (1-3) lower plasma cholesterol levels when ture. Cholesterol content of the laboratory diet was substituted for saturated fatty acids, although consid- 0.03%. Animals were weighed, randomly divided into erable variability between individuals has been reported four groups and individually housed in stainless steel, in the actions of these fatty acids (5). For PUFA, lino- wire mesh cages. For three weeks each group was fed leic acid appears to be the major cholesterol lowering one of four diets containing laboratory diet with 10% fatty acid, affecting plasma cholesterol levels in a man- {w/w) of corn (i), olive (ii), coconut (iii) or menhaden fish ner which extends beyond t h a t achieved by simple oil (iv). Added fats were obtained from local sources, replacement of dietary saturated fats. Similarly, die- except for menhaden oil, which was purchased from tary fish oils containing eicosapentaenoic acid (EPA, ICN Biochemicals (Cleveland, OH). Cholesterol levels 20:5oj3) and docosahexaenoic acid (DHA; 22:6w3) ex- of each diet containing added oil were then measured hibit variable effects on plasma cholesterol content and made up to 0.1% (w/w) by the addition of free {6-10), but reduce triglyceride levels (9,10). cholesterol. Diets were prepared every two weeks and Although the influence of dietary fat type on lipid stored at -10~ F a t t y acid composition of each diet levels has been identified, the underlying mechanisms was determined (Table 1) by gas-liquid chromatography IGLC) after extraction with chloroform/methanol (2:1, v/v) and methylation using boron trifluoride {14). On day 21 of study, food cups were removed at the *To whom correspondence should be addressed at the Division beginning of the dark cycle and 3 to 4 hr later animals of Human Nutrition, Schoolof Family and Nutritional Services, received an IP injection of approximately 30 mCi 3H20. 2205 East Mall, University of British Columbia, Vancouver, Ninety minutes after injection animals were weighed British Columbia,V6T lW5 Canada. and sacrificed by heart puncture following light anesAbbreviations: DHA, docosahexaenoic acid; EPA, eicosapen- thesia. Blood was taken for measurement of plasma taenoic acid; GLC, gas-liquid chromatography; HDL, high den- lipid levels and 3H-activity {15). Liver and duodenum sity lipoprotein; LDL, low density lipoprotein; MCFA, medium chain fatty acids, MUFA, monounsaturated fatty acids; PUFA, were removed, flushed with saline solution and immepolyunsaturated fatty acids; VLDL, very low density lipopro- diately frozen in liquid nitrogen. Animals were cared tein. for in accordance with the principles of the "Guide to LIPIDS,VoL 25, No. 12 (1990)
816
P.J.H. JONES ETAL. TABLE 1 Fatty Acid Composition of Diets (wt % of Total Fatty Acids} a
Diet fat consumed Olive Coconut Menhaden Fatty acid off off oil 12:0 -35.0 -14:0 1.2 16.1 7.6 16:0 14.6 13.8 19.8 16:1co7 2.4 1.6 11.6 18:0 5.7 5.7 7.6 18:1co9 56.9 15.0 19.9 18:2co6 13.9 7.3 7.7 20:0 0.9 0.6 1.1 20:1o~9 0.3 0.1 1.0 18:3co3 2.0 1.5 2.0 20:5w3 0.7 0.8 10.5 22:5co3 --1.9 22:6r 5.8 aFatty acids are designated by chain length, number of double bonds and position of the first double bond from the fatty acid methyl terminus. Only major dietary fatty acids are reported. Corn oil -1.2 13.2 1.6 2.8 27.3 46.2 0.9 0.5 2.2 1.1 --
-
-
-
-
RESULTS
-
the Care and Use of Experimental Animals", Vol. 2, 1984.
Measurement of tissue lipid synthesis and esterification. Rates of synthesis and esterification of cholesterol were determined by the uptake rate of 3H from aH20 (16,17). Measurement of synthesis by 3H-uptake has been previously validated {16) and yields a measure of absolute synthetic rate when the specific activity of the substrate water pool is known {17}. The rate of appearance of all-labeled cholesterol as ester was taken to reflect to movement of newly synthesized sterol from the free to ester form. Tissue samples were extracted in duplicate using hexane/chloroform {4:1, v/v} and water after the addition of a 14C-cholesterol internal standard. Hexane/chloroform phases were combined and solvent removed under N2. Free and esterified cholesterol bands were separated from extracts by thin-layer chromatography with double development for 1 hr followed by 45 min using petroleum ether/ diethyl ether/acetic acid (85:15:1.5, v/v/v). Bands were eluted from silica scrapings using chloroform. To remove all-containing fatty acids, cholesterol esters were hydrolyzed with KOH, rechromatographed and the resulting free cholesterol band was scraped from plates and eluted. Samples were aliquoted for analyses of all-uptake and lipid content and composition. Tritium activity was measured by liquid scintillation counting (Isocap, Picker}. Results were expressed as nmol aH incorporated per g tissue per hr. Aliquots for fatty acid analysis were methylated as previously described {14). Analytical methods. Plasma, liver and duodenal cholesterol levels were determined enzymatically (18). High density lipoprotein {HDL) cholesterol was measured using a modified dextran sulfate precipitation method (19). Triglyceride levels in plasma were also determined enzymatically ~20). F a t t y acid composition of dietary fats was analyzed using a gas-liquid chromatograph {GLC) {model 6000, Varian Instruments, Palo Alto, CA) utilizing a 30 m SP-2330 {10%) capillary column and helium carrier gas (2 mL/min). Compositional analysis of tissue fatty acid methyl esters was LIPIDS,Vol. 25, No. 12 (1990)
carried out by GLC {model 5750, Hewlett-Packard, Hewlett-Packard, Norwalk, CT} using a 2.5 m by 3 mm steel column packed with SP-2330 (10% on 100/ 120 Chromosorb) and nitrogen carrier gas (10 mL/min}. Chromatograph peaks were identified by comparison of their retention data with those of authentic fatty acid methyl ester standards ~Supelco Inc., BeUefonte, PA). Peaks were quantified by ratio integration using a 17:0 fatty acid methyl ester standard. Statistical analysis. Data were analyzed using oneway ANOVA {analysis of variance} procedures with Tukey's post hoc tests performed for inter-group comparisons. A value of p < 0.05 was used as a level of statistical significance.
Mean body weight, liver weight, liver weight to body weight ratio and food intake of hamsters fed diets differing in fat composition were similar across all groups. Lipid contents of plasma, liver and duodenum for animals consuming different dietary fats are shown in Table 2. Plasma total cholesterol values were higher in animals fed the coconut oil diet as compared with all other groups. Corn oil-fed animals had lower plasma cholesterol levels in comparison to olive and coconut oil-fed groups. For H D L cholesterol, the group fed olive oil showed higher values compared with that consuming menhaden oil. Plasma triglyceride levels were elevated in coconut fed animals in relation to other all groups. With the exception of greater hepatic total cholesterol and cholesterol ester levels of olive oil-fed animals, no significant diet treatment effect was observed in cholesterol or fatty acid content of liver or duodenum. A white color, likely cholesterol, was noted in livers of olive off-fed animals at sacrifice. No significant differences were observed in hepatic or duodenal fatty acid content across treatment groups. Liver and duodenal total lipid extract fatty acid composition are shown in Tables 3 and 4, respectively, for animals fed the different diets. Overall, liver contained a greater proportion of w3 and fewer monoenoic fatty acids as compared with duodenum, independent of dietary fat intake. Corn off-fed animals showed highest levels of w6 fatty acids in both tissues as compared with groups consuming other diets. Similarly, in each tissue, groups fed menhaden and olive oil diets displayed the highest level of r PUFA and MUFA, respectively. Notably, whereas the duodenum responded to the coconut oil diet by increasing proportions of medium chain saturated fatty acids (MCFA) compared with other groups, liver did not. Also, liver contained a greater proportion of f a t t y acids of chain length greater than 18 carbons, compared with duodenum. Uptake of aH into free cholesterol in liver and duodenum and esterified cholesterol in liver in animals consuming different dietary fats is illustrated in Figure 1. Incorporation rates into free cholesterol were lower in liver as compared with duodenum. In liver, no effect of diet fat type was observed in free cholesterol 3H-uptake. Uptake into cholesterol ester was greater in groups fed menhaden oil over those fed other
817
DIETARY FAT AND CHOLESTEROL SYNTHESIS TABLE 2 Plasma, Liver and Duodenal Lipid Contentsa
Plasma level (rag/100 mL) Cholesterol total HDL Triglyceride Liver (mg/100 g tissue) Cholesterol total Free Esters Fatty acids
Corn oil
Diet fat consumed Olive Coconut oil oil
190.0a 12.0 (n=9) 105.1 8.6 (n=7) 106a 9.5 (n=8)
226.1 b 8.7 (n=9) 134.6a 9.0 (n=7) 95.7a 6.4 (n=9)
378.9c 37.0 (n=9) 130.9 18.8 (n=7) 230.6b 38.2 (n=8)
221.5a,b 22.0 (n=9) 76.6b 8.2 (n=7) 131.8a 22.6 (n=8)
104a 23 (n=9) 60.7 8.0 (n=9) 43.5a 16.1 (n=9) 3292 270 (n=6)
212b 31 (n=8) 76.6 5.8 (n=8) 135.6b 27.6 (n=8) 2969 235 (n=6)
82.5a 14 (n=8) 67.0 6.5 (n=8) 21.4a 6.3 (n=8) 2709 267 (n=6)
78.6a 8.0 (n=9) 64.2 3.1 (n=9) 14.4a 5.7 (n=9) 3363 290 (n=6)
Duodenum (mg/100 g tissue) Cholesterol total
Menhaden oil
49.3 5.8 in=9) 44.9 5.7 (n=9)
58.8 66.5 52.2 8.3 3.8 4.1 (n=8} (n=9) (n=9) Free 51.7 61.9 47.3 5.8 3.7 4.0 (n=8) (n=9) (n=9) Esters 4.4 7.1 4.6 4.9 0.8 2.0 0.8 0.6 (n=9) (n=8) (n=9) (n=9) Fatty acids 4500 7168 7308 4360 215 514 820 294 (n=6) (n=6) (n=6) in=6) a,b, cValues are means +_ SEM. Any two values in the same row followed by different letters differ significantly (p
