The Journal of Nutrition Vol. 128 No. 10 October 1998, pp. 1620-1630

Variations in Dietary Fat and Cholesterol Intakes Modify Antioxidant Status of SHR and WKY Rats^1,2

Yvonne V. Yuan³, David D. Kitts⁴, and David V. Godin^*

Department of Food Science, Faculty of Agriculture, and ^* Department of Pharmacology and Therapeutics, Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada

	ABSTRACT

Abstract Introduction Methods Results Discussion References

The effects of varying dietary fat saturation [butter (B), beef tallow (BT)] or polyunsaturation [(n-6) soybean oil (SBO), (n-3) menhaden oil (MO)] and cholesterol content (0.05 and 0.5 g/100 g) on systolic blood pressure (SBP), plasma lipids and tissue antioxidant status were investigated in 14-wk-old spontaneously hypertensive rats (SHR) and normotensive Wistar Kyoto (WKY) rats. Varying dietary fat composition for 9 wk had no influence on SBP in either SHR or WKY rats. Rats fed MO diets exhibited smaller (P < 0.05) body weight gains, lower (P < 0.05) feed efficiency ratios and lower (P < 0.05) plasma cholesterol concentrations than those fed the B, BT and SBO diets. Significant (P < 0.05) interactions for animal strain × cholesterol intake and animal strain × fat source were noted for serum cholesterol concentrations. SHR exhibited higher (P < 0.05) RBC and liver catalase (CAT), and heart and liver superoxide dismutase (SOD) activities similar to those of WKY rats. The lower (P <0.01) RBC, heart and liver glutathione peroxidase (GSH-Px) activities observed in SHR coincided with higher (P <0.01) glutathione reductase (GSSG-Red), compared with WKY rats. Dietary cholesterol intake had no effect on RBC, heart and liver total sulfhydryl concentration or GSH-Px activities, but increased (P <0.001) liver GSSG-Red. Feeding MO resulted in lower (P <0.001) RBC and heart GSH-Px activities. In contrast, feeding B and BT resulted in lower GSH-Px in liver. The significant (P < 0.01) animal strain × fat source interaction obtained for liver GSH-Px activity indicated that SHR responded differently to polyunsaturated fatty acid feeding than their WKY counterparts. Diet-induced changes in tissue antioxidant status were tissue specific and did not affect the development of hypertension in SHR.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The role of dietary fat intake in modulating hypertension and hyperlipoproteinemias has been investigated in human (Bairati et al. 1992, Bonanome and Grundy 1988, Mattson and Grundy 1987) and animal studies (De Schrijver et al. 1992, Morgenson and Box 1982). Diets high in monounsaturated fatty acids or marine oil long-chain polyunsaturated fatty acids (PUFA)⁵ have been found to be relatively hypocholesterolemic or hypotriacylglycerolemic, respectively (Mattson and Grundy 1985), although in some cases also effectively lowering blood pressure (Bairati et al. 1992). In animal studies, dietary (n-6) and (n-3) PUFA have been shown to decrease blood pressure (Karanja et al. 1989, Smith et al. 1993). Conversely, other investigators have reported that both saturated and polyunsaturated fats can induce hypertension in rats, possibly by altering insulin sensitivity (Kaufman et al. 1994). The role of dietary cholesterol intake and its interaction with dietary fat source on circulating plasma lipid levels have also been investigated (Smit et al. 1994) and have been shown to have no effect on hypertension in the stroke-prone spontaneously hypertensive rat (SHR) animal model (Mori et al. 1993).

Oxidative injury is common in several diseases including hypertension. In both animal and human studies, systolic and diastolic blood pressure correlated inversely with tissue superoxide dismutase (SOD) activity and glutathione (GSH) concentration (Giugliano et al. 1995, Kimoto et al. 1995). Previous work from our laboratories has also shown differences in several tissue antioxidant enzyme activities of the SHR relative to its normotensive counterpart, the WKY rat (Kitts et al. 1998, Yuan et al. 1996). Although many studies have examined the effect of altering diet-derived nonenzymatic antioxidants on antioxidant defenses against various disorders (Halliwell 1996), less attention has been given to the evaluation of the role of specific dietary lipids in modulating antioxidant enzyme status. Reactive oxygen species (ROS), as well as products of lipid peroxidation, are known to affect antioxidant enzyme activity, as evidenced, for example, by the inhibition of both SOD and glutathione peroxidase (GSH-Px) in response to elevated concentrations of H₂O₂ (Remacle et al. 1992). In addition, controversy exists whether the intake of highly unsaturated fatty acids always increases the potential for in vivo peroxidation of plasma and membrane constituent lipid (Alexander-North et al. 1994, Calviello et al. 1997, Nalbone et al. 1989) and urinary excretion of lipid oxidation products (L'Abbé et al. 1991). Oxygen free radicals have also been implicated in the development of atherosclerosis (Prasad et al. 1994) and associated increased demand on endogenous antioxidants (Javouhey-Donzel et al. 1993).

The purpose of this study was to examine the effect of varying dietary fat saturation [e.g., saturated, (n-6) and (n-3) PUFA] and cholesterol intake on plasma lipids, antioxidant enzyme status and on systolic blood pressure in both normotensive and hypertensive rats.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals.  Sixty-four 5-wk-old male spontaneously hypertensive rats (SHR) and sixty-four normotensive Wistar Kyoto (WKY) rats (Charles River, Montreal, Canada) were randomly assigned to one of eight dietary treatment groups differing in dietary fat source (i.e., butter, beef tallow, soybean oil or menhaden oil) and cholesterol level [i.e., low cholesterol (0.5 g/kg diet) and high cholesterol (5 g/kg diet)].

Diets.  The composition of the purified diets used in this study varied in lipid source, and cholesterol and cholic acid content (Table 1). A basal diet containing 30 g canola oil (Neptune Food Services, Richmond, Canada) per kg diet provided an adequate supply of essential fatty acids, and was formulated with thorough mixing of ingredients before the addition of experimental dietary fat sources and sterols. Dietary fat sources consisted of nonsalted butter (B; Dairyworld Foods, Burnaby, Canada), beef tallow (BT; Cargill Foods, High River, Canada), soybean oil (SBO; Bioforce Canada, Burnaby, Canada) or menhaden oil (MO; Zapata Haynie, Reedville, VA) with either a low (0.0295 mg/kJ) or high (0.290 mg/kJ) level of cholesterol (Table 1). No additional antioxidants were added to the dietary formulations, with the exception of the vitamin E that was present as a component of the vitamin mixture (see legend of Table 1). Both the butter and beef tallow were liquefied using mild heat treatment for a short time (45-50°C for 10-15 min) to facilitate the uniform distribution of cholesterol and cholic acid (2:1 ratio, the latter added to ensure maximal absorption of dietary cholesterol) into diets. No heat treatment was necessary for the soybean or menhaden oils. Sterols were added slowly to the liquefied experimental fat sources and thoroughly mixed to ensure uniform incorporation. Dietary fats containing sterols at the levels shown in Table 1 were slowly added to the basal diet during reblending and mixed in completely using a Hobart mixer with aluminum bowl over a period of ~20-25 min. After mixing, individual diets were stored in double, dark plastic bags at 15°C throughout the experimental period. A sample of each experimental diet was taken for analysis of fatty acid, gross energy and dry matter content.

Experimental fats were added to the basal diet at a level of 50 g/kg diet to make a final calculated total fat content of 80 g/kg diet. This level of fat in the diets meets the nutritional recommendations for the laboratory rat (50-100g/kg diet) as outlined by the Canadian Council on Animal Care (1984). The amount of butter added to experimental diets was calculated to take into account the amount of moisture naturally present in processed butter (~20% moisture). The level of 0.5 g cholesterol/kg diet was chosen to equalize all low cholesterol diets for the sterol concentration naturally present in butter (~0.2 g/100 g) and beef tallow (~0.1 g/100 g). Dietary cholesterol content was equalized between diets through the addition of crystalline cholesterol into the liquefied fat sources as detailed above. Analysis of the diets for cholesterol confirmed the calculated uniform content of sterol in both individual low cholesterol and high cholesterol diets. Experimental diets were isonitrogenous and contained a comparable level of energy (e.g., 16.84-17.67 MJ/kg; Table 1). Energy content of all experimental diets was determined using a bomb calorimeter and was corrected for the dry weight of the diet.

Dietary lipid analysis.  The fatty acid content of the experimental diets and cholesterol content of butter, tallow and menhaden sources (see legend of Table 1) were determined by gas chromatography. For fatty acids, diet samples were extracted with Folch's reagent (Folch et al. 1957), methylated with BF₃ and analyzed for component fatty acids using a Varian Model 3700 GC equipped with a 60 m × 0.53 mm i.d. column coated with 0.25 µm Supelcowax 10 (Supelco; Bellefonte, PA) according to the method of Nwokolo and Kitts (1988). The internal standard included in these fatty acid analyses was 17:0 (Supelco). Cholesterol in fat and oil samples was quantitated by the method of Yuan et al. (1997).

Animal feeding.  Animals were trained to meal-feed from 0900 to 1500 h after 1 wk of free access to diets (Kitts et al. 1992). This meal-feeding schedule ensured a similar postprandial time period in animals before blood pressure measurements were performed. All diets were replaced daily to minimize autoxidation of dietary lipids during the 6-h feeding period. Systolic blood pressure recordings were taken using an indirect tail-cuff method (Harvard Apparatus, South Natick, MA) between 1500 and 1700 h (Kitts et al. 1992).

After 9 wk of consuming the experimental diets, rats were killed by exsanguination under halothane anesthesia at 1300 h. Rats were killed in a fed state to minimize changes to antioxidant status due to possible stresses of food deprivation. Blood was collected into heparinized tubes for separation of plasma by centrifugation at 1000 × g for 5 min, 4°C. Heart and liver tissues were collected into chilled 50 mmol/L Tris 0.1 mmol/L EDTA (pH 7.6) homogenizing buffer. Aliquots of plasma were analyzed for both free and total cholesterol (Siedel et al. 1983) using commercial kits from Boehringer Mannheim (Laval, Canada). Red blood cells were washed twice with isotonic (0.15 mol/L) saline; hemolysates were then prepared by diluting RBC 1:10 with double-distilled H₂O and lysed three times in dry ice/acetone to ensure complete cell disruption. Heart and liver tissues were blotted dry, weighed and prepared as 1.0 g tissue/10 mL buffer homogenates in fresh, chilled 50 mmol/L Tris 0.1 mmol/L EDTA (pH 7.6) homogenizing buffer using a Polytron homogenizer (Brinkmann Instruments, Westbury, NY) at 25% maximum speed for 30 s (2 × 15 s). Tissue cytosolic fractions used in the enzymatic assays were prepared from homogenates by centrifugation at 105,000 × g, 15 min, at 4°C using a Beckman L2-65 ultracentrifuge with an sw40Ti rotor (Beckman, Montreal, Canada). Tissue cytosolic fractions were assayed for protein content according to Bradford (1976). Enzyme activity determinations were carried out using a Perkin-Elmer model Lambda 6B spectrophotometer (Perkin-Elmer, Norwalk, CT) with temperature control set for 25°C.

Analytical methods.  Red blood cell, heart and liver tissue catalase (CAT; EC.1.11.1.6), glutathione peroxidase (GSH-Px; EC.1.11.1.9), glutathione reductase (GSSG-Red; EC.1.6.4.2) and superoxide dismutase (SOD; EC.1.15.1.1) activities were determined as described by Yuan et al. (1996). RBC hemolysates and tissue cytosolic fractions were assayed for hemoglobin (Hb) according to the method of Drabkin and Austin (1935). Because RBC contain various amounts of antioxidant enzyme activities, it was necessary to correct tissue cytosolic enzyme activity values for any contribution due to the presence of contaminating RBC. This was achieved by determining the Hb content of tissue homogenate cytosolic fractions and RBC fractions and then subtracting the enzyme activity associated with the amount of contaminating RBC in the tissue cytosolic fraction from the enzyme activity determined for each tissue homogenate cytosolic preparation. Catalase activity of RBC was expressed as k/mg Hb and that of tissues as k/mg protein, where k is the first-order rate constant (s¹). Glutathione peroxidase and glutathione reductase activities of RBC were both expressed as nmol NADPH/(min·mg Hb) and those of tissues as nmol NADPH/(min·mg protein). Superoxide dismutase (SOD) activity of RBC was expressed as units of SOD activity/mg Hb and that of tissues as units of SOD activity/mg protein. One unit of SOD activity is defined as the amount of enzyme activity that causes 50% inhibition of nitroblue tetrazolium reduction.

Tissue and RBC sulfhydryl group content, which was used as a measure of glutathione (GSH), was determined as described by Yuan et al. (1996). Briefly, tissue trichloroacetic acid (TCA)-soluble sulfhydryl groups were reacted with 5,5'-dithiobis-(2-nitrobenzoic acid) (DTNB; Sigma Chemical, St. Louis, MO) in 0.1 mol/L phosphate buffer, pH 8.0, and the absorbance read at 412 nm. RBC acid-soluble sulfhydryl groups were determined using 50 µL packed RBC that had been lysed with cold double-distilled H₂O, followed by the addition of cold 0.31 mol/L TCA/1.0 mmol/L Na₂-EDTA. The supernatant was reacted with DTNB as above.

Statistics.  All data are expressed as mean ± SEM. Analysis of variance (MANOVA; SPSS/Win, Chicago, IL) was used to test for differences among experimental treatments and possible interactions between animal strain, dietary cholesterol level and dietary fat source. If treatment differences and corresponding interactions did exist, the source was identified by the Student-Newman-Keuls multiple range test at a P < 0.05 significance level.

	RESULTS

Abstract Introduction Methods Results Discussion References

Fatty acid content of purified diets.  Fatty acids with chain lengths <14 were not measured in these analyses. The absence of this information is important only for the butter diet, because butter contains ~15% of total fatty acids as 4:0-12:0. The content of 14:0 was greater in diets containing butter or menhaden oil than in diets with beef tallow or soybean oil (see legend of Table 1). B and BT diets contained similar amounts of 16:0 and 18:0, which were greater than those found in the MO or SBO diets. BT and SBO diets contained similar amounts of 18:1(n-9) which were greater than those found in B and MO diets. SBO diets contained the greatest quantities of 18:2(n-6) compared with all other fat diets. The content of 18:3(n-3) was greater in SBO diets and to a lesser degree in MO compared with the B and BT diets. The MO diets were unique in the content of eicosapentaenoic [20:5(n-3); EPA] and docosahexaenoic [22:6(n-3); DHA] acids. On the basis of the fatty acid analyses performed, the calculated polyunsaturated to saturated fatty acid ratios for the B and BT diets were 0.39 and 0.28, respectively, whereas that of the SBO diets was 2.3 and the MO diets was 1.4. The ratio of (n-6) to (n-3) fatty acids was 4.4 and 2.8 for the B and BT diets, respectively, and 3.8 and 0.74 for the SBO and MO diets, respectively.

View larger version (52K): [in this window] [in a new window]   Fig 1. Feed efficiency ratio (FER) of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) rats fed two levels of cholesterol (0.05% or 0.5%) in butter (B), beef tallow (T), soybean (S) and menhaden oil (M) diets. FER is the ratio of grams body weight gained per gram diet consumed over the 9-wk experimental period. Values are means ± SEM (n = 8). x,yDenote a dietary fat treatment effect at P < 0.001; a,bdenote a rat strain difference at P < 0.001.

View larger version (38K): [in this window] [in a new window]   Fig 2. Plasma free and total cholesterol concentrations of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) rats fed two levels of cholesterol (0.05% or 0.5%) in butter (B), beef tallow (T), soybean (S) and menhaden oil (M) diets. (Panel A): free cholesterol of SHR; (panel B): free cholesterol of WKY; (panel C): total cholesterol of SHR; (panel D): total cholesterol of WKY. There was a significant rat strain difference at P < 0.001 for both free and total cholesterol. Values are means ± SEM (n = 8). x,yDenote a dietary fat treatment effect at P < 0.001; a,bdenote the cholesterol treatment effect at P < 0.001.

Rat growth.  Initial body weights of SHR (156 ± 2 g) and WKY rats (150 ± 2 g) did not differ. Body weight gain during the experimental period was not influenced by animal strain or dietary cholesterol level; however, both SHR and WKY rats fed MO diets gained less weight (P < 0.001) than rats fed other dietary fat sources. The feed efficiency ratio (FER) of SHR and WKY rats was influenced by strain and dietary fat source (Fig. 1), as demonstrated by the interaction recorded (P = 0.023). SHR had lower (P < 0.001) FER than WKY rats. Rats of both strains fed MO diets had reduced (P < 0.001) FER compared with rats fed other dietary fat sources. Liver weights of SHR were greater (P < 0.05) than those of WKY rats in those fed both the low cholesterol diets (SHR range 12.1 ± 0.1 vs. WKY 10.1 ± 0.1 g) and the high cholesterol diets (SHR range 19.0 ± 0.1 vs. WKY 15.4 ± 0.1 g). Dietary fat had no effect on liver weight of either SHR or WKY rats.

Systolic blood pressure of SHR and WKY rats.  Systolic blood pressure measured at 13 (data not shown) and 14 wk of age confirmed that SHR (treatment mean range: 174 ± 4 to 197 ± 10 mm Hg) were hypertensive (P < 0.001) compared with WKY (treatment mean range: 125 ± 6 to 147 ± 7 mm Hg) rats. Systolic blood pressure of rats was not affected by dietary fat source or cholesterol level.

Plasma lipid profiles.  Both plasma free (FC) and total cholesterol (TC) concentrations were significantly (P < 0.001) affected by strain, dietary cholesterol intake level and dietary fat source (Figs. 2A-D). SHR had lower (P < 0.001) plasma FC and TC (Figs. 2A,B) concentrations than WKY rats (Figs. 2C,D). As expected, rats fed high cholesterol diets exhibited greater (P < 0.001) levels of plasma FC and TC, compared with rats fed low cholesterol diets, thus explaining the significant (P <0.01) interaction between rat strain and cholesterol intake for both free and total cholesterol. A significant interaction between rat strain and fat source (P = 0.023) was traced to rats fed MO diets; they had lower (P < 0.001) levels of plasma FC compared with other dietary fat sources, for both dietary cholesterol intake levels (Figs. 2A,C). Dietary fat source did not affect plasma TC concentrations of SHR rats or plasma TC levels of WKY rats fed low cholesterol diets. WKY rats fed high cholesterol diets displayed significantly (P < 0.01) higher plasma TC in the BT-fed group and significantly (P < 0.01) lower plasma TC in the MO-fed group, a finding that contributed to a significant interaction (P < 0.001) between cholesterol intake and fat source for TC.

Tissue antioxidant status. 

Red blood cell (RBC) antioxidant enzymes.  RBC activities of CAT were higher (P < 0.001) in SHR compared with WKY rats, but were not influenced by either dietary fat source or cholesterol (Table 2). SOD activities were lower (P < 0.001) in SHR relative to WKY rats and were affected by both dietary fat source [e.g., SBO-fed rats had higher (P = 0.029] SOD activities than MO-fed rats) and the cholesterol intake [SOD activity was higher (P = 0.044) in rats fed high cholesterol diets]. RBC GSH-Px activity was significantly affected by both rat strain and dietary fat source, but not by the level of dietary cholesterol intake. Specifically, GSH-Px activity was greater (P = 0.002) in SHR relative to WKY rats. In addition, rats fed MO diets exhibited lower (P < 0.001) activities of RBC GSH-Px relative to rats fed other dietary fats sources, whereas rats fed BT had the highest RBC GSH-Px activity.

Heart antioxidant enzymes.  Heart tissue had negligible levels of CAT after correction for enzyme activity attributed to contaminating RBC (data not shown). SOD activity was significantly influenced by dietary fat source but not by rat strain or dietary cholesterol intake level (Table 3). Specifically, rats fed B and SBO diets exhibited greater (P = 0.014) heart SOD activities than those fed MO diets. Activities of GSH-Px and GSSG-Red were also significantly influenced by dietary fat source and rat strain, but not by cholesterol intake level. SHR had lower (P = 0.024) GSH-Px activities and higher (P = 0.014) GSSG-Red activities than WKY rats. GSH-Px activities were also lower (P = 0.001) in rats fed MO diets relative to other dietary fat sources. Also, B- and BT-fed rats had greater (P = 0.001) heart GSSG-Red activities than those fed MO diets (Table 5).

Liver antioxidant enzymes.  Liver CAT activity was significantly influenced by rat strain and dietary cholesterol intake level, but was not affected by dietary fat source (Table 4). SHR had greater (P < 0.001) activities of liver CAT than WKY rats. Rats fed high cholesterol diets generally exhibited reduced (P = 0.045) liver CAT activities compared with rats fed low cholesterol diets. In contrast, SOD activities were not different between SHR and WKY rats; however, SOD activity was affected by both dietary fat source and cholesterol intake. Generally, SOD activities were lower (P = 0.04) in rats consuming high cholesterol diets relative to rats fed low cholesterol diets and lower (P < 0.001) in rats fed MO diets compared with those fed diets containing other fat sources. Liver GSH-Px and GSSG-Red activities were both significantly influenced by rat strain, dietary fat source and cholesterol intake level (Table 5). SHR had lower (P < 0.001) liver GSH-Px activities compared with WKY rats. High cholesterol diets generally resulted in lower (P = 0.022) GSH-Px activities and greater (P < 0.001) GSSG-Red activities in the liver. GSH-Px activities were greater (P = 0.001) in WKY rats fed SBO diets than in MO-fed rats. Activity of liver GSSG-Red was greater (P < 0.001) in SHR compared with WKY rats. Liver GSSG-Red activities were greater (P = 0.049) in rats fed SBO and MO diets compared with B- and BT-fed rats.

Tissue and RBC total sulfhydryl concentration.  SHR exhibited significantly higher (P < 0.01) total sulfhydryl concentrations in heart (Fig. 3 A,B), but not liver (Fig. 4 A,B) or RBC (Fig. 5 A,B) compared with WKY rats. Dietary cholesterol intake and fat source had no effect on heart, liver or RBC total sulfhydryl concentration.

View larger version (43K): [in this window] [in a new window]   Fig 3. Heart total sulfhydryl (total thiols) concentrations of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) rats fed two levels of cholesterol [0.05% (solid bars) or 0.5% (Hatched bars)] in butter (B), beef tallow (T), soybean (S) and menhaden oil (M) diets. (Panel A): total thiols of SHR; (panel B): total thiols of WKY. Values are means ± SEM (n = 8). There was a significant rat strain difference at P < 0.001 for heart GSH.

View larger version (43K): [in this window] [in a new window]   Fig 4. Liver total sulfhydryl (total thiols) concentrations of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) rats fed two levels of cholesterol (0.05% or 0.5%) in butter (B), beef tallow (T), soybean (S) and menhaden oil (M) diets. (Panel A): liver total thiols of SHR; (panel B): liver total thiols of WKY. Values are means ± SEM (n = 8).

View larger version (47K): [in this window] [in a new window]   Fig 5. Red blood cell total sulfhydryl (total thiols) concentrations of spontaneously hypertensive (SHR) and Wistar-Kyoto (WKY) rats fed two levels of cholesterol (0.05% or 0.5%) in butter (B), beef tallow (T), soybean (S) and menhaden oil (M) diets. (Panel A): Red blood cell total thiols of SHR; (panel B) red blood cell total thiols of WKY. Values are means ± SEM (n = 8). x,yDenote a dietary fat treatment difference at P = 0.025 for RBC in WKY rats.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The presence of cholesterol in the diet can enhance the cholesterolemic response to saturated fatty acids such as 14:0 and 16:0 (Hayes et al. 1995). Long-chain (n-3) PUFA (e.g., EPA and DHA) derived from marine oils reduce plasma cholesterol levels and have beneficial effects on hypertension (Smith et al. 1993) and blood clotting time (Smit et al. 1994). However, the increased incorporation of PUFA from marine or vegetable oil dietary sources into plasma lipoproteins and cell membrane phospholipids has been shown to increase both lipoprotein (Nardini et al. 1995) and tissue susceptibility to lipid peroxidation (De Schrijver et al. 1992, L'Abbé et al. 1991, Skúladóttir et al. 1994). This investigation describes the effects of feeding diets varying in the degree of saturation or PUFA unsaturation [e.g., (n-6) or (n-3)] of dietary fat source and cholesterol intake on plasma lipids, systolic blood pressure and antioxidant status in SHR and WKY rats.

Feed efficiency.  The uniformly reduced FER in SHR compared with WKY rats fed experimental diets is consistent with previous studies with these strains of rats (Kitts et al. 1992) and may reflect the reported insulin resistance of the SHR (Hulman et al. 1993), given that similar reductions in energy accretion and feed efficiency have been previously reported in diabetic animal models (Wohaieb and Godin 1987). Furthermore, the decrease in body weight gained and FER of both SHR and WKY rats fed the MO diets is consistent with previous reports that have shown a reduction in energy gain and energy efficiency in rats fed diets enriched in fish oil (Su and Jones 1993).

Systolic blood pressure.  The genetic predisposition of SHR to hypertension observed in rats at 13 and 14 wk of age has been reported in our laboratory (Kitts et al. 1992, Yuan et al. 1996). The fact that dietary fat was ineffective in modulating hypertension in the SHR, however, differs from some studies that have reported a specific hypotensive effect of dietary fish oil and butter fat compared with corn oil in older (26 wk) SHR (Karanja et al. 1989). It should be noted that these workers did not begin to observe a significant hypotensive effect of feeding 18% menhaden oil or butter fat diets until rats were at least 16 wk of age (Karanja et al. 1989). Our findings parallel those of McGregor et al. (1981), who reported that systolic blood pressure of SHR fed an (n-6) PUFA diet for 16 wk was not reduced compared with rats fed a highly saturated fat diet. Other workers have demonstrated that both saturated fat (e.g., lard) and polyunsaturated fat (e.g., corn oil) sources can induce hypertension in normal rats over a 10-wk feeding period (Kaufman et al. 1994). Finally, other workers have reported that stroke-prone SHR fed diets containing 20% milk fat exhibited a reduced incidence of cerebrovascular disease, but without a reduction in blood pressure (Ikeda et al. 1987). The short-chain saturated fatty acid content and the relatively high level of 16:1 unique to butter fat was suggested to have a role in the effects of butter fat diets on stroke in hypertensive rats (Ikeda et al 1987, Karanja et al 1989). The inconsistencies among various studies examining the effect of dietary lipids on hypertension in the SHR may be related to such variables as the age of the rats, the duration of feeding of experimental diets and the absolute amount or proportions of fatty acids in the fat sources fed to the rats.

Plasma and tissue lipids.  The lower concentrations of plasma cholesterol observed in SHR compared with WKY rats fed formulated purified diets agree with previous studies in rats fed a commercial nonpurified diet (Kitts et al. 1998, Yuan et al. 1996) and possibly reflect specific differences in cholesterol metabolism between these two inbred strains. The interaction noted between rat strain and MO feeding for plasma cholesterol levels supports the finding of others that although the hypotriacylglycerolemic effect of dietary fish oil is well recognized, the effect on plasma cholesterol concentrations can be variable (De Schirijver et al. 1992, Ikeda et al. 1994).

Animal strain differences in tissue antioxidant status.  The insulin resistance reported in SHR (Hulman et al. 1993) may be a key underlying mechanism for the observed differences in tissue antioxidant status between SHR and WKY rats because other studies with diabetic rats have reported alterations in tissue antioxidant enzyme activity and susceptibility to in vitro oxidation (Wohaieb and Godin 1987). More recently, hypertension in the SHR was found to be associated with alterations in heart tissue and RBC antioxidant status relative to normotensive controls [i.e., increased susceptibility of heart tissue to in vitro lipid oxidation and increased RBC and liver CAT activities in SHR compared with WKY (Yuan et al. 1996)]. Similar differences in antioxidant activity in hepatic tissue have also been observed in SHR and WKY rats (Kitts et al. 1998). An alternative explanation for the differences in antioxidant status between these two rat strains may be related to specific differences in fatty acid -oxidation or lipid metabolism in general. Previous workers have reported that increased peroxisomal fatty acid -oxidation and concomitant hypolipoproteinemia in normal rats are associated with up-regulation of liver CAT activities (Demoz et al. 1992). Plasma lipid levels have been noted to be reduced in SHR compared with WKY rats when sampled in the food-deprived state (McGregor et al. 1981, Mori et al. 1993, Kitts et al. 1998, Yuan et al. 1996). This could not be completely confirmed in this study due to the limitations of the study design in which the primary focus was placed on measuring the antioxidant status of the rats; thus, food was not withheld before killing in order to limit any stresses before removal of tissues for antioxidant determinations. Taken together, these results suggest that an important underlying component of the differences in antioxidant status between SHR and WKY rats is related to the metabolism of lipids and the concurrent effects on oxidative balance in key tissues involved in lipid metabolism.

Dietary fat and cholesterol effects on antioxidant status.  Regardless of their susceptibility to the development of hypertension, SHR and WKY rats exhibited similar RBC antioxidant enzyme responses to the feeding of saturated or PUFA [e.g., (n-6) and (n-3)] diets. Our observation of lower RBC GSH-Px activities in both SHR and WKY rats fed MO diets compared with rats fed saturated fats seems to contradict the positive correlation reported between the DHA concentration in RBC membranes and erythrocyte lipid peroxidation (Clemens and Waller 1987). However, this response to MO feeding corresponded to a similarly lower RBC SOD activity, which could indicate an O₂-induced inhibition of GSH-Px activity due to reduced SOD activity and no change in CAT activity. Previous studies designed to examine the antioxidant status of rats fed diets varying in PUFA composition have reported an inverse correlation between the ratio of tissue SOD/GSH-Px activities with urinary TBARS, suggesting that the balance of these two enzymes is a factor in tissue susceptibility to lipid oxidation (L'Abbé et al. 1991). Because of parallel reductions in activities of both RBC SOD and GSH-Px activities in our study, the ratio of enzyme activities was not altered between rats fed MO, SBO or saturated fat diets. This situation was changed in rats fed high cholesterol diets, as evidenced by the fact that both hypercholesterolemic MO- and SBO-fed rats exhibited enhanced RBC SOD activities, whereas RBC GSH concentrations, and GSH-Px and CAT activities were not similarly affected. Elevated plasma cholesterol levels have been shown to result in reduced RBC peroxidative stress (Bereza et al. 1985), presumably by causing physical alterations in membrane fluidity and reduced susceptibility to oxidation.

In this study, the observation that heart GSH-Px and GSSG-Red activities were similarly reduced in both SHR and WKY rats fed MO diets may explain, in part, the lower tissue sulfhydryl concentrations observed. GSSG-Red activity serves to regenerate reduced GSH from its oxidized form (GSSG) by the action of GSH-Px. The expected effect of replacement of membrane 20:4(n-6) with (n-3) PUFA from the MO diet to reduce cyclooxygenase and lipooxygenase activities (Kinsella 1988), for which 20:4(n-6) is a substrate, would reduce the load of ROS (e.g., superoxide radicals) produced from these biosynthetic pathways. The alteration of the cyclooxygense pathway and arachidonate metabolism in vitamin E-deficient rats has also been shown to alter vascular function in response to increased lipid peroxidation (Davidge et al. 1993). The fact that both SHR and WKY fed MO diets exhibited lower SOD activities suggests further that parallel functional relationships exist between antioxidant enzymes, enabling a balance in cellular protection against lipid peroxidation. The stimulus for antioxidant enzymes to work in concert may exist initially in combination with the overall metabolic activity of the tissue. For example, using the formula for specific metabolic rate, Godin and Garnett (1992) demonstrated a positive correlation between this parameter and heart SOD activity. The reduced body weight gain and FER values of rats fed MO diets in this study correspond to the reduced heart SOD activity, also reported by others in rats fed energy-restricted diets (Xia et al. 1995) and supports the positive correlation between SOD activity and specific metabolic rate calculations in other animal species (Godin and Garnett 1992, Tolmassoff et al. 1980). The fact that high cholesterol diets had no detectable effect on heart GSH-Px, GSSG-Red and SOD activities indicates that diet-induced alterations in membrane cholesterol/phospholipid composition were not as important in modulating oxidative stress in heart tissue as in RBC.

Marine-oil (n-3) PUFA, EPA and DHA have been reported to induce hepatic peroxisomal -oxidation enzymes and up-regulate hepatic antioxidant enzymes and GSH levels (Demoz et al. 1992). The up-regulation of hepatic antioxidant defenses and reduced lipid peroxide levels in EPA-fed mice reported by other investigators was attributed to an increase in the requirement for redox H₂O₂-metabolizing enzymes as a result of H₂O₂ production by oxidase enzymes. For example, CAT, the primary peroxisomal enzyme responsible for H₂O₂ metabolism, was increased in EPA-fed animals, as were hepatic GSH-Px and GSSG-Red activities (Demoz et al. 1992). In our study, although MO-fed rats did not exhibit enhanced hepatic SOD or CAT activities, the fact that the GSH redox enzymes GSH-Px and GSSG-Red were both elevated indicates that functionally coupled antioxidant enzyme pairs such as CAT and SOD, and GSH-Px and GSSG-Red are independently regulated by dietary lipid intake. One explanation for the differences between our findings and those of other studies may be the use of dietary fat blends compared with the single purified fatty acids EPA (Demoz et al. 1992) and DHA (Willumsen et al. 1993) used by other workers. For example, EPA and DHA from the MO diets would be expected to exert similar, but potentially less dramatic effects on fatty acid metabolism and antioxidant defenses in vivo because of the modulating presence of the other dietary long-chain fatty acids. Thus, despite the similarity of the peroxisomal CAT activities in all of the dietary treatment groups, mitochondrial GSH-Px (an alternative H₂O₂-metabolizing enzyme) was indeed up-regulated along with the GSH recycling enzyme, GSSG-Red in MO-fed animals. The significant interactions noted for rat strain and dietary fat source for GSH-Px indicate that SHR may have a relatively reduced capacity to detoxify ROS via the GSH-Px pathway compared with WKY rats. Cholesterol feeding has been previously reported to decrease hepatic GSH-Px activity in rats and to reduce hepatic microsomal NADPH-dependent lipid peroxidation (Tsai et al. 1977). However, differences in plasma cholesterol due to either endogenous metabolic differences between SHR and WKY rats and/or an effect of increased cholesterol intake were found to have only a minimal effect on hepatic GSH redox enzymes and total sulfhydryl concentrations in this study.

In conclusion, genetic (e.g., endogenous) factors were shown to have a strong influence on hypertension, plasma cholesterol concentration and tissue-specific antioxidant enzyme activities, and were only partially modified by exogenous (e.g., dietary saturated and PUFA-containing fats and cholesterol) factors. Although modulation of hypertension did not occur with various dietary fat sources and cholesterol intakes, specific changes in plasma cholesterol concentrations, and RBC, heart and liver antioxidant enzyme activities were observed. Antioxidant enzyme patterns in experimental rats were tissue specific, denoting a similar complexity of endogenous antioxidant defenses in the control of diet-induced changes associated with antioxidant status that has been reported in other animal models (Yuan et al. 1997). In particular, coincident changes in coupled GSH-Px/GSSG-Red and SOD/CAT activities were found to be most influenced by the genetic makeup of rats compared with dietary lipid intake. Further studies are required to confirm the significance of the in vitro evaluation of antioxidant enzyme activities reported here for changes in target organ lipid peroxidation and genetic predisposition to hypertension.

	FOOTNOTES

Manuscript received 17 February 1998. Initial reviews completed 13 March 1998. Revision accepted 9 June 1998.

	ACKNOWLEDGMENT

The authors thank Lana Fukumoto for assistance with the plasma lipid analyses.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Alexander-North L. S., North J. A., Kiminyo K. P., Buettner G. R., Spector A. A. Polyunsaturated fatty acids increase lipid radical formation induced by oxidant stress in endothelial cells. J. Lipid Res. 1994; 35:1773-1785[Abstract]
• Bairati I., Roy L., Meyer F. Effects of a fish oil supplement on blood pressure and serum lipids in patients treated for coronary artery disease. Can. J. Cardiol. 1992; 8:41-46[Medline]
• Bereza U. L., Brewer G. J., Hill G. M. Effect of dietary cholesterol on erythrocyte peroxidant stress in vitro and in vivo. Biochim. Biophys. Acta 1985; 835:434-440[Medline]
• Bonanome A., Grundy S. M. Effect of dietary stearic acid on plasma cholesterol and lipoprotein levels. N. Engl. J. Med. 1988; 318:1244-1248[Abstract]
• Bradford M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976; 72:248-254[Medline]
• Calviello G., Palozza P., Franceschelli P., Bartoli G. M. Low-dose eicosapentaenoic or docosahexaenoic acid administration modifies fatty acid composition and does not affect susceptibility to oxidative stress in rat erythrocyte and tissues. Lipids 1997; 32:1075-1083[Medline]
• Canadian Council on Animal Care (1984) Guide to the Care and Use of Experimental Animals, vol. 2, p. 180. CCAC, Ottawa, Canada.
• Clemens M. R., Waller H. D. Lipid peroxidation in erythrocytes. Chem. Phys. Lipids. 1987; 45:251-268[Medline]
• Davidge S. T., Hubel C. A., McLaughlin M. K. Cyclooxygenase-dependent vascoconstrictor alters vascular function in the vitamin E-deprived rat. Circ. Res. 1993; 73:79-88[Abstract]
• De Schrijver R., Vermeulen D., Daems V. Dose-response relationships between dietary (n-3) fatty acids and plasma and tissue lipids, steroid excretion and urinary malondialdehyde in rats. J. Nutr. 1992; 122:1979-1987
• Demoz A., Willumsen N., Berge R. K. Eicosapentaenoic acid at hypotriglyceridemic dose enhances the hepatic antioxidant defense in mice. Lipids 1992; 27:968-971[Medline]
• Drabkin D. L., Austin J. H. Spectrophotometric studies. II. Preparation from washed blood cells: nitric oxide hemoglobin and sulfhemoglobin. J. Biol. Chem. 1935; 112:51-65[Free Full Text]
• Folch J., Lees M., Sloane-Stanley G. H. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 1957; 226:497-509[Free Full Text]
• Giugliano D., Ceriello A., Paolisso G. Diabetes mellitus, hypertension and cardiovascular disease: which role for oxidative stress? Metabolism 1995; 3:363-368
• Godin, D. V. & Garnett, M. E. (1992) Species-related variations in tissue antioxidant status. I. Differences in antioxidant enzyme profiles. Comp. Biochem. Physiol. 103B: 737-742.
• Halliwell B. Antioxidants in human health and disease. Annu. Rev. Nutr. 1996; 16:33-50[Medline]
• Handler J. A., Thurman R. G. Catalase-dependent ethanol oxidation in perfused rat liver: requirement for fatty-acid-stimulated H₂O₂ production by peroxisomes. J. Biochem. 1988; 176:477-484
• Hayes K. C., Pronczuk A., Khosla P. . A rationale for plasma cholesterol modulation by dietary fatty acids: modulation of human response in animals. J. Nutr. Biochem. 1995; 59:39-56
• Hulman S., Falkner B., Freyvogel N. Insulin resistance in the conscious spontaneously hypertensive rat: euglycemic hyperinsulinemic clamp study. Metabolism 1993; 42:14-18[Medline]
• Ikeda K., Mochizuki S., Nara Y., Horie R., Yamori Y. Effect of milk protein and fat intake on blood pressure and the incidence of cerebrovascular diseases in stroke-prone spontaneously hypertensive rats (SHRSP). J. Nutr. Sci. Vitaminol. 1987; 33:31-36
• Javouhey-Donzel A., Guenot L., Maupoil V., Rochette L., Rocquelin G. Rat vitamin E status and heart lipid peroxidation: effect of dietary -linolenic acid and marine n-3 fatty acids. Lipids 1993; 28:651-655[Medline]
• Karanja N., Phanouvong T., McCarron D. A. Blood pressure in spontaneously hypertensive rats fed butterfat, corn oil, or fish oil. Hypertension 1989; 14:674-679[Abstract/Free Full Text]
• Kaufman L. N., Peterson M. M., Smith S. M. Hypertensive effect of polyunsaturated dietary fat. Metabolism 1994; 43:1-3[Medline]
• Kimoto S., Nishida S., Funaskaka K., Nakano T., Teramoto K., Tomura T. T. Regional distribution of superoxide dismutase in the brain and myocardium of the stroke-prone spontaneously hypertensive rat. Clin. Exp. Pharmacol. Physiol. 1995; 22:S160-S161[Medline]
• Kinsella, J. E. (1988) Food lipids and fatty acids: importance in food quality, nutrition, and health. Food Technol. 124-145.
• Kitts, D. D., Yuan, Y. V., & Godin, D. V. 1998. Plasma and lipoprotein lipid composition and hepatic antioxidant status in spontaneously hypertensive and normotensive rats. Can. J. Physiol. Pharmacol. 72: 202-209.
• Kitts D. D., Yuan Y. V., Nagasawa T., Moriyama Y. Effect of casein, casein phosphopeptides and calcium intake on ileal ⁴⁵Ca disappearance and temporal systolic blood pressure in spontaneously hypertensive rats. Br. J. Nutr. 1992; 68:765-781[Medline]
• L'Abbé M. R., Trick K. D., Beare-Rogers J. L. Dietary (n-3) fatty acids affect rat heart, liver and aorta protective enzyme activities and lipid peroxidation. J. Nutr. 1991; 121:1331-1340
• Mattson F. H., Grundy S. M. Comparison of effects of dietary saturated, monounsaturated, and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J. Lipid Res. 1985; 26:194-202[Abstract]
• McGregor L., Morazain R., Renaud S. Platelet functions and fatty acid composition of platelet phospholipids in spontaneously hypertensive rats fed saturated or polyunsaturated fats. Atherosclerosis 1981; 38:129-136[Medline]
• Morgenson G. J., Box B. M. Physiological effects of varying dietary linoleic acid in spontaneously hypertensive rats. Ann. Nutr. Metab. 1982; 26:232-239[Medline]
• Mori H., Ishiguro K., Okuyama H. Hypertension in rats does not potentiate hypercholesterolemia and aortic cholesterol deposition induction by a hypercholesterolemic diet. Lipids 1993; 28:109-113[Medline]
• Nalbone G., Leonardi J., Termine E., Portugal H., Lechene P., Pauli A.-M., Lafont H. Effects of fish oil, corn oil and lard diets on lipid peroxidation status and glutathione peroxidase activities in rat heart. Lipids 1989; 24:179-186[Medline]
• Nardini M., D'Aquino M., Tomassi G., Gentili V., DiFelice M., Scaccini C. Dietary fish oil enhances plasma LDL oxidation modification in rats. J. Nutr. Biochem. 1995; 6:478-480
• Nwokolo E., Kitts D. D. Growth parameters and plasma and tissue fatty acid profiles in rats fed rubber seed oil. Food Chem. 1988; 30:219-229
• Prasad K., Kalra J., Lee P. Oxygen free radicals as a mechanism of hypocholesterolemic atherosclerosis: effects of probucol. Int. J. Angiol. 1994; 3:100-112
• Remacle J., Lambert D., Raes M., Pigeolet E., Michiels C., Toussaint O. Importance of various antioxidant enzymes for cell stability: confrontation between theoretical and experimental data. Biochem. J. 1992; 286:41-46
• Siedel J., Hagele E. O., Ziegenhorn J., Wahlefeld A. W. Reagent for the enzymatic determination of serum cholesterol with improved lipolytic efficiency. Clin. Chem. 1983; 29:1075-1080[Abstract/Free Full Text]
• Skúladóttir G. V., Shi-Hua D., Brodie A. E., Reed D. J., Wander R. C. Effects of dietary oils and methyl ethyl ketone peroxide on in vivo lipid peroxidation and antioxidants in rat heart and liver. Lipids 1994; 29:351-357[Medline]
• Smit M. J., Verkade H. J., Havinga R., Vonk R. J., Scherphof G. L., In't Veld, G., Kuipers F. Dietary fish oil potentiates bile acid-induced cholesterol secretion into bile in rats. J. Lipid Res. 1994; 35:301-310[Abstract]
• Smith J. M., Kopp S. J., Paulson D. J., Daar J. T. Effects of dietary fish oil on cardiovascular responsiveness to adrenergic agonists in spontaneously hypertensive rats. Can. J. Physiol. Pharmacol. 1993; 71:432-438[Medline]
• Su W., Jones P. J. H. Dietary fatty acid composition influences energy accretion in rats. J. Nutr. 1993; 123:2109-2114
• Tolmasoff J. M., Ono T., Cutler R. G. Superoxide dismutase: correlation with life-span and specific metabolic rate in primate species. Proc. Natl. Acad. Sci. U.S.A. 1980; 77:2777-2781[Abstract/Free Full Text]
• Tsai A. C., Thie G. M., Lin C. R. Effect of cholesterol feeding on tissue lipid peroxidation, glutathione peroxidase activity and liver microsomal functions in rats and guinea pigs. J. Nutr. 1977; 107:310-319
• Willumsen N., Hexeberg S., Skorve J., Lundquist M., Berge R. K. Docosahexaenoic acid shows no triglyceride-lowering effects but increases the peroxisomal fatty acid oxidation in liver of rats. J. Lipid Res. 1993; 34:13-22[Abstract]
• Wohaieb S. A., Godin D. V. Alterations in free radical tissue-defense mechanisms in streptozocin-induced diabetes in rat: effects of insulin treatment. Diabetes 1987; 36:1014-1018[Abstract]
• Xia E., Rao G., Van Remmen H., Heydari A. R., Richardson A. Activities of antioxidant enzymes in various tissues of male Fischer 344 rats are altered by food restriction. J. Nutr. 1995; 125:195-201
• Yuan Y. V., Kitts D. D., Godin D. V. Heart and red blood cell antioxidant status and plasma lipid levels in the spontaneously hypertensive and normotensive Wistar-Kyoto rat. Can. J. Physiol. Pharmacol. 1996; 74:290-297[Medline]
• Yuan Y. V., Kitts D. D., Godin D. V. . Influence of dietary cholesterol and fat source on atherosclerosis in the Japanese quail (Coturnix japonica). Br. J. Nutr. 1997; 78:993-1014[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences