A 1,3-diacylglycerol-rich oil induces less atherosclerosis and lowers plasma cholesterol in diabetic apoE-deficient mice

Atherosclerosis 193 (2007) 55–61

A 1,3-diacylglycerol-rich oil induces less atherosclerosis and lowers plasma cholesterol in diabetic apoE-deficient mice Akihiko Fujii b,∗ , Terri J. Allen a , Paul J. Nestel a b

a Baker Heart Research Institute, Melbourne, Australia Biological Science Laboratories, Kao Corporation, 2606 Akabane, Ichikai-machi, Haga-gun, Tochigi 312-3497, Japan

Received 26 December 2005; received in revised form 31 July 2006; accepted 11 August 2006 Available online 22 September 2006

Abstract Objective: Recent studies have demonstrated that 1,3-diacylglycerol (1,3-DAG) has several metabolic advantages over triacylglycerol (TAG) in humans and in animal models despite both oils having a similar fatty acid composition. In our current study, we have examined the effects of long-term feeding of a 1,3-DAG-rich oil on the dyslipidemia and atherosclerosis in the experimental model of the diabetic apolipoprotein E (apoE)-deficient mouse that develops accelerated atherosclerosis. Methods and results: Diets containing 1,3-DAG-rich oil or TAG oil were administered to control non-diabetic apoE-dificient and diabetic apoE-deficient mice for 20 weeks. In diabetic apoE-deficient mice, 1,3-DAG reduced the extent of atherosclerotic lesions in the aortic arch and thoracic aorta by 37 and 44%, respectively, compared to TAG. Further, in diabetic apoE-deficient mice, plasma total cholesterol and triglyceride levels were significantly lower in the 1,3-DAG-fed group than in the TAG-fed group. This occurred partially through an apparent reduction in the size of triglyceride-rich lipoproteins but not apparently by reducing the number of lipoprotein particles. By contrast the control non-diabetic apoE-deficient mice showed no differential responses to the type of oil at least over 20 weeks. Conclusions: We have demonstrated that dietary 1,3-DAG-rich oil reduced atherosclerosis in diabetic apoE-deficient mice, and was associated with reduction in plasma cholesterol especially within larger triglyceride-rich lipoproteins. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Diacylglycerol; Diabetic apoE-deficient mice; Dyslipidemia; Atherosclerosis; Triglyceride-rich lipoprotein

1. Introduction Dietary diacylglycerol (DAG), which comprised mainly the 1,3-isoform, occurs naturally in various edible oils as a minor component [1,2]. Consequently, 1,3-DAG is widely consumed. Most ingested triacylglycerol is however hydrolysed by 1,3-specific lipase to 1,2 or 2,3-DAG and fatty acids in the intestinal lumen [3]. These DAG are then hydrolysed to 2-monoacylglycerol (MAG) and fatty acid [3]. Subsequently, these substances are absorbed into the intestinal epithelium and rapidly resynthesised into triacylglycerol by the 2-MAG pathway in the intestinal mucosal cells [3]. In contrast, 1,3-DAG is hydrolysed to 1(3)-MAG and fatty acid in the intestinal lumen [4]. 1(3)-MAG is assumed to be absorbed ∗

Corresponding author. Tel.: +81 285 68 7440; fax: +81 285 68 7495. E-mail address: [email protected] (A. Fujii).

0021-9150/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2006.08.024

into the intestinal epithelium or further hydrolysed to release the remaining fatty acid and glycerol [4]. Consequently, 1,3DAG does not appear to be resynthesized to triacylglycerol to the same extent as 2-MAG. Thus, 1,3-DAG probably has a different metabolic pathway compared to that of DAGs derived from triacylglycerol. Recent studies have demonstrated that dietary 1,3-DAGrich oil has several advantages over dietary triacylglycerol (TAG oil)1 in humans and in animal models despite both oils having a similar fatty acid composition. Rats consuming dietary 1,3-DAG oil showed significantly lower serum triacylglycerol levels than after being fed TAG oil [5]. Clinical studies have shown that long-term dietary 1,3-DAG oil reduced serum triacylglycerol levels in type II diabetic 1 Although 1,3-DAG oil contains other lipids (see Table 1), it will be referred to in this term. The triacylglycerol oil will be referred to as TAG.

56

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61

patients with hypertriacylglyceridemia and in one patient with lipoprotein lipase deficiency [6,7]. Moreover, the longterm ingestion of 1,3-DAG oil decreased body weight and body fat in humans and in high fat-fed C57Bl/6J mice [8,9]. Thus, although there are several studies showing that 1,3DAG ameliorates obesity and hypertriacylglyceridemia, the effect of this novel DAG on atherosclerosis has not yet been demonstrated. Patients with type 2 diabetes generally develop a more extensive and inflammatory form of atherosclerosis than non-diabetic individuals [10–12]. Atherosclerosis leading to myocardial infarction and stroke is the major serious complication in diabetics [12] although the precise molecular mechanisms underlying the development of a more severe form of atherosclerosis remain uncertain. Postprandial hypertriacylglyceridemia is a recognized risk factor for cardiovascular disease in diabetics in whom the degree of lipemia is greater than in non-diabetic subjects [13–15]. It is likely that reducing hypertriacylglyceridemia may be an important strategy important in combating atherothrombotic vascular disease as suggested by at least one secondary prevention trial [16]. In our current study, we have examined the effects of long-term feeding of 1,3-DAG oil on the dyslipidemia and atherosclerosis in an experimental model of the diabetic apolipoprotein E (apoE)-deficient mouse. ApoE-deficient mice typify a well-established murine model of hypercholesterolemia [17,18] and such mice fed a Western-type diet are highly susceptible to develop atherosclerotic lesions. Additionally, streptozotocin-induced diabetic apoE-deficient mice are highly vulnerable to dyslipidemia and atherosclerosis even without supplemental fat and cholesterol [19,20]. We have compared in such mice the effects of feeding identical amounts of oil of similar fatty acid composition from either TAG or 1,3-DAG. 2. Methods 2.1. Experimental design Six-week-old homozygous ApoE-deficient male mice (back-crossed 20 times from C57BL/6 strain; Animal Resource Centre, Canning Vale, WA, Australia) were housed at the Precinct Animal Centre, Baker Heart Research Institute. All procedures were done according to National Health and Medical Research Council guidelines after written approval was obtained from the Institute’s Animal Ethics Committee. Mice were rendered diabetic by intraperitoneal injection of streptozotocin (MP Biomedicals, Eschwege, Germany) at a dose of 55 mg/kg in citrate buffer for 5 consecutive days. Control mice received citrate buffer alone. Mice that did not become diabetic were discarded. When the streptozotocin-treated mice were confirmed to be diabetic, each group diabetic and non-diabetic was divided into two groups, one receiving the TAG and the other the 1,3-DAG oil incorporated into the feed to provide an additional 15% fat by weight. Thus, four groups were studied. The non-diabetic

Table 1 Composition of the diets (percentage by weight) Ingredients

TAG diet

Triacylglycerol (TAG) oil 1,3-Diacylglycerol (DAG)-rich oila Casein Cellulose Starch Dextrinised starch dl-Methionine AIN-93-G-trace minerals Lime (fine calcium carbonate) Salt (fine sodium chloride) Potassium dihydrogen phosphate Potassium sulphate Potassium citrate AIN-93-G-vitamins Choline chloride 60% (w/w)

15.0 20.0 4.0 43.4 13.2 0.3 0.1 1.3 0.3 0.7 0.2 0.2 1.0 0.3

DAG diet 15.0 20.0 4.0 43.4 13.2 0.3 0.1 1.3 0.3 0.7 0.2 0.2 1.0 0.3

a The DAG-rich oil contains approximately 90% DAG and 10% TAG. The DAG is comprised of 1,3-DAG and 1,2 (or 2,3) in a ratio of 7:3.

mice were in effect a control group for the diabetic mice since the former were not expected to develop significant atherosclerosis within 20 weeks. Diabetic and non-diabetic (control) animals had unrestricted access to the synthetic diets described in Table 1 for 20 weeks using Roden Caffe (Oriental Yeast Co., Tokyo, Japan) to minimize dispersion of diets. TAG and DAG-rich oils include following fatty acids, respectively, 16:0, 5.3 and 3.1%; 18:0, 2.2 and 1.2%; 18:1, 37.6 and 38.8%; 18:2, 46.0 and 47.8%; 18:3, 7.4 and 8.0%; 20:0, 0.5 and 0.2%; 20:1, 0.7 and 0.5%; 22:0, 0.3 and 0.2%; 22:1, 0.1 and 0.1%. The diets containing oils were prepared by Specialty Feeds (WA, Australia), stored at 4 ◦ C before use and changed three times per week. Growth rates, consumption of feed and blood glucose concentration were monitored at regular intervals. Food intake was monitored on a per-cage basis three times per week. At 19 weeks of the study, food intake for 24 h per mouse was measured using a metabolic cage. Body weight was measured weekly. Blood to monitor glucose concentration was collected from saphenous vein monthly. Systolic blood pressure was assessed by a computerized, non-invasive tail cuff system in conscious mice at week of 20 of the study. Mice were habituated to the device before measuring the blood pressure to ensure accurate measurements. At 20 weeks of the study, mice were anesthetized by an intraperitoneal injection of Euthal (10 mg/kg body weight, Delvet Limited, Seven Hills, Australia) and aortas were excised. 2.2. Atherosclerotic lesions The extent of atherosclerosis was determined using an en-face method. Aortas were rapidly removed and fixed in phosphate-buffered 10% formalin after staining with Sudan IV–Herxheimer’s solution. Aortas were then opened longitudinally, divided into arch, thoracic and abdominal segments and pinned out flat on wax blocks. Images of each segment were captured with an Axiocam Camera (Zeiss, Heidelberg,

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61

Germany) mounted on a dissection microscope. The extent of atherosclerotic lesions was analysed with Adobe PhotoShop 7.0 Software and expressed as a percentage area of stained aorta of the total area. 2.3. Plasma lipids At 20 weeks of the study blood samples were taken by cardiac puncture after the mice had been deprived of feed for 3 h (overnight fasts lead to deterioration in diabetic mice). Plasma lipids were determined enzymatically using commercial kits for triacylglycerol (Triglyceride E-Test Wako; Wako), cholesterol (Cholesterol E-Test Wako; Wako) and HDL-cholesterol (HDL-Cholesterol E-Test Wako; Wako). 2.4. Lipoprotein analysis Subfractionation of the plasma lipoproteins was performed as previously described by Ishibashi [21]. In brief, 0.25 ml of plasma was adjusted to density 1.10 g/ml by adding 0.07 g of KBr and 0.5 ml of saline, and mixed with 0.75 ml of d = 1.10 g/ml KBr solution. Gradients consisting of 1.5 ml of d = 1.10 g/ml plasma, 1.2 ml of d = 1.063 g/ml, 1.2 ml of d = 1.019 g/ml and 1.2 ml of d = 1.006 g/ml KBr solutions were formed by being overlaid and centrifuged at 30,000 rpm for 19 min (chylomicrons), at 30,000 rpm for 36 min (VLDLA), at 40,000 rpm for 90 min (VLDL-B), at 40,000 rpm for 95 min (VLDL-C), at 40,000 rpm for 14 h (VLDL-D) at 20 ◦ C. Each fraction (0.5 ml) was taken from the top layer of the tube between each centrifugation, and d = 1.006 g/ml KBr solution was used to refill the tubes before further centrifugation in a Beckman SW41 rotor. The approximate densities of the four VLDL fractions are 0.94, 0.95, 0.96 and 0.98 g/ml, respectively. 2.5. Apolipoprotein analysis ApoB48 and apoB100 in the whole plasma were analysed by Western blot analysis. Plasma pooled from all animals in each group was electrophoresed on 4–15% polyacrylamide gels under reducing condition and transferred to polyvinyli-

57

dene fluoride (PVDF) membrane. The membrane was blocked for 2 h in Tris-buffered saline/Tween buffer (10 mM Tris/HCl, pH 7.4, 150 mM NaCl, 0.1% Tween-20) containing 5% (w/v) non-fat dry milk, and then incubated overnight with rabbit polyclonal antibodies against mouse apoB48/apoB100 (Biodesign International; titer 1:300). After incubation for 1 h with horseradish peroxidase conjugated secondary antibody, the signals were detected using a Chemiluminescence Detection System (Supersignal, Pierce) and Hyperfilm MP (Amersham Pharmacia Biotech). The bands of apoB proteins were quantified using a densitometer and expressed as relative optical density. 2.6. Statistical analysis All data were normally distributed. Student’s t-test was used to define difference between TAG and 1,3-DAG-fed mice. P < 0.05 was accepted as statistically significant. All data are presented as mean ± S.D. 3. Results 3.1. Effects of oils on body weights, blood pressure, plasma glucose and lipids As expected, growth rates as judged by body weights were less in the diabetic than non-diabetic apoE-deficient mice, despite greater consumption of feed (Table 2). Plasma glucose concentration remained consistently elevated in the diabetic mice. However, there were no significant differences in body weight, food intake, and blood pressure and plasma glucose between TAG and 1,3-DAG oil-fed groups at the end of the study (Table 2). In the diabetic apoE-deficient mice, plasma total cholesterol and triacylglycerol levels were significantly lower in the 1,3-DAG-fed group than in the TAG-fed group. On the other hand, in non-diabetic apoEdeficient mice, there were no significant differences in plasma total cholesterol and triacylglycerol levels between TAG and 1,3-DAG-fed groups. HDL-cholesterol in the 1,3-DAG-fed diabetic apoE-deficient mice was not different from that in TAG-fed diabetic apoE-deficient mice.

Table 2 Body weight, blood pressure, glucose and lipid levels in TAG or DAG-fed diabetic and control non-diabetic apoE-deficient mice Diabetic TAG (n = 13) Body weight (g) Food intake (g/day) Systolic blood pressure (mmHg) Blood glucose (mmol/l) Total plasma cholesterol (mmol/l) Plasma triglyceride (mmol/l) HDL-cholesterol (mmol/l)

18.2 4.1 105 31.3 29.2 3.90 0.28

± ± ± ± ± ± ±

3.0 1.0 8 6.9 7.3 3.67 0.11

Non-diabetic DAG (n = 14) 18.9 4.6 107 29.8 21.7 1.79 0.33

± ± ± ± ± ± ±

2.1 0.5 11 5.2 4.6† 0.52* 0.13

TAG (n = 8) 29.0 2.6 113 10.2 8.4 1.13 0.41

± ± ± ± ± ± ±

2.1 0.6 10 1.6 2.7 0.39 0.12

DAG (n = 8) 28.3 2.6 116 10.7 9.0 1.14 0.43

± ± ± ± ± ± ±

1.8 0.4 5 0.9 1.8 0.14 0.07

Blood samples for glucose, total cholesterol, triglyceride and HDL-cholesterol are taken at the end of the study after fasting mice for 3 h. Values are mean ± S.D. * P < 0.05 vs. TAG-fed diabetic group. † P < 0.01 vs. TAG-fed diabetic group.

58

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61

Fig. 1. Extent of atherosclerotic lesions in diabetic apoE-deficient mice at the end of the study. Atherosclerotic lesions were assessed by staining with Sudan VI–Herxheimer’s solution in each segment and expressed as percentage surface area showing atherosclerosis relative to the total area. Values are mean ± S.D. * P < 0.05 vs. TAG-fed diabetic group.

3.2. Effect of oils on atherosclerotic lesion The extent of atherosclerosis was estimated by measuring the lipid-stained lesions in each segment of aortic arch, thoracic aorta and abdominal aorta. As expected, in nondiabetic apoE-deficient mice, most of the aortas were free of detectable atherosclerotic lesions in both TAG and 1,3-DAGfed groups. In diabetic apoE-deficient mice, atherosclerosis in 1,3-DAG-fed mice was significantly less than in those fed TAG in aortic arch and thoracic aorta by 37 and 44%, respectively, from initial 29.1 and 4.0% involvement (Fig. 1).

Fig. 3. (A and B) Effects of 1,3-diacylglycerol on the distribution of lipoproteins. Equal volumes of plasma from diabetic or control non-diabetic mice fed with TAG or 1,3-DAG diet were pooled, and 250 ␮l was used for the ultracentrifugation. Pooled plasma samples were obtained from TAG-fed diabetic group (n = 13), 1,3-DAG-fed diabetic group (n = 14), TAG-fed control group (n = 8) and 1,3-DAG-fed control group (n = 8).

3.4. Effect of oils on plasma triacylglycerol-rich lipoproteins 3.3. Correlation between atherosclerosis and plasma lipids In diabetic apoE-deficient mice, a significant correlation (R = 0.49, P < 0.05) was detected between extent of atherosclerosis in aortic arch and plasma cholesterol concentration (Fig. 2A). However, there was no significant correlation between extent of lesions in aortic arch and plasma triacylglycerol concentrations (Fig. 2B).

To study the effect of 1,3-DAG on the size distribution of lipids among plasma lipoproteins, separation of plasma lipoproteins was performed by sequential ultracentrifugation. In diabetic apoE-deficient mice, cholesterol levels in chylomicrons, VLDL-A, VLDL-B, VLDL-C, VLDL-D and IDL/LDL in the 1,3-DAG-fed group were lower than in the TAG-fed group by 46, 45, 42, 29, 27 and 38%, respectively (Fig. 3A). Thus, 1,3-DAG led to lesser cholesterol

Fig. 2. Linear regression analysis between extent of atherosclerosis of aortic arch and plasma total cholesterol (A) or triacylglycerol (B). The slope of the regression line in (A) but not in (B) was significant (P < 0.05), as shown by analysis of Pearson product moment correlation.

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61

Fig. 4. Western blot analysis of apoB in the plasma. (A) Equal volumes of plasma from diabetic or control non-diabetic mice fed with TAG or 1,3-DAG were pooled, and 0.125 ␮l was used for Western blot analysis. Pooled plasma samples were obtained from TAG-fed diabetic group (n = 13), 1,3-DAG-fed diabetic group (n = 14), TAG-fed control group (n = 8) and 1,3-DAG-fed control group (n = 8). (B) Relative optical densities of apoB48 and apoB100 bands in each group.

concentrations especially in larger lipoproteins. Similarly 1,3-DAG also led to lesser triacylglycerol concentrations in larger lipoproteins (Fig. 3B). There appeared to be a shift to smaller lipoproteins. 3.5. Effect of oils on plasma apolipoprotein B To determine whether 1,3-DAG affects the number of lipoproteins particles, apoB48 and apoB100 levels were determined by Western blotting (Fig. 4). The apoB concentration reflects the total number of VLDL, IDL and LDL because each of these lipoproteins contains one apoB molecule. The concentration of apoB48 in diabetic apoE-deficient mice was greater than that in non-diabetic apoE-deficient mice as shown in Fig. 4. Concentrations of apoB100 appeared not to differ. Concentrations of apoB48 appeared to begreater with DAG than TAG in the diabetic mice. However, since total cholesterol and triglyceride concentrations were lower in plasma, despite an increase in the number of apoB48 particles it seems likely that feeding the mice DAG reduced the content of lipids in these particles. There appeared to be no differences in apoB100 concentrations between TAG-fed and the 1,3-DAG-fed group.

4. Discussion In the current study, we have demonstrated in diabetic apoE-deficient mice that the consumption of 1,3-

59

diacylglycerol oil led to less aortic atherosclerosis as well as reduced lipoprotein cholesterol compared with dietary triacylglycerol of similar fatty acid composition. This occurred partially through a probable reduction in the size of triacylglycerol-rich lipoproteins. By contrast, in apoE deficient but non-diabetic mice that served as a control group, feeding DAG or TAG did not result in quantifiable atherosclerosis at least over 20 weeks. The catabolism of the DAG oil which consists mainly of the 1,3-isoform differs from that of TAG. However, the oil contained also other DAG isomers (1,2-DAG and 2,3-DAG) that are rarely present in the diet and are formed during the catabolism of triacylglycerol. Previous research has shown that this difference may be ascribed to the structural difference between 1,3-DAG and the major diacylglycerol species derived from the catabolism of TAG [4,5,22]. Our data are consistent with other findings that a 1,3-DAG-rich high fat diet decreased fasting plasma cholesterol and triacylglycerol in mice deficient in both apoE and LDL receptors compared to a TAG-rich high fat diet [23]. Further, Ijiri et al. [23] showed a significantly lower predisposition to thrombogenesis to DAG than to TAG feeding in mice deficient in both the apoE and LDL Receptor genes. Our data suggested that reduction in plasma cholesterol but not that in triacylglycerol was associated with less aortic atherosclerosis. The mechanism underlying the reduction in plasma cholesterol is unclear although it is worth noting that 1,3-DAG reduced cholesterol in the diabetic apoE-deficient mice, but not in non-diabetic apoE-deficient mice. Much of plasma cholesterol in apoE-deficient mice is contained in the VLDL and chylomicron remnant fractions. The present data suggest that 1,3-DAG shifted the distribution of cholesterol-rich VLDL particles to smaller species while not affecting the number of particles. Although this conclusion is based on pooled samples, the differences between samples from the TAG and DAG-fed mice in both cholesterol and triacylglycerol content across the entire lipoprotein spectrum strongly supports the validity of that conclusion. The reduction in atherosclerosis occurred despite an apparent increase in apoB48 particles that are transporters of triglyceride-rich lipoproteins from both intestine and liver in rodents; however, as seems probable, these particles carried less lipid after DAG feeding than similar particles following TAG. We postulate that the likely mechanism for our observation relating to the difference in dyslipidemia between the two oil-fed mice reflects partial inhibition by 1,3-DAG with the assembly of VLDL in the liver or intestine of diabetic apoE-deficient mice. Microsomal triglyceride transfer protein (MTP) is a key regulator in the assembly and secretion of VLDL from the liver and intestine and previous studies with rats have demonstrated that dietary 1,3-DAG reduced MTP activity in the liver [24]. Moreover, MTP inhibition reduced plasma VLDL-cholesterol more than plasma LDLcholesterol [25]. Murase et al. have shown that dietary 1,3DAG affected the expression of several genes associated with lipid metabolism such as acyl-CoA oxidase (ACO), liver-fatty

60

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61

acid binding protein (L-FABP), medium-chain acyl-CoA dehydrogenase (MCAD), uncoupling protein 2 (UCP2) and fatty acid transporter (FAT) in the intestine leading to the increased ␤-oxidation of fatty acid [9]. This potentially reduces the pool of hepatic fatty acid available for triacylglycerol synthesises. The changes in gene expressions reflect in part activation of PPAR␣ attributed to increased fatty acids in the intestine resulting from the particular catabolism of 1,3-DAG [9]. Increased concentrations of plasma triacylglycerol, cholesterol and VLDL/LDL have been reported in PPAR␣-deficient mice [26], consistent with the findings with 1,3-DAG feeding that led to the falls in VLDL-cholesterol. However, there are no direct comparisons with other diabetic apoE-deficient mice which in the present study have resulted in much higher concentration of plasma total and VLDL-cholesterol than was observed in non-diabetic apoEdeficient mice. Catabolism of triacylglycerol-rich lipoproteins is retarded with insulin deficiency, allowing enrichment with cholesterol and this mechanism might have contributed to the different degrees of dyslipdemia between diabetic and non-diabetic apoE-deficient mice. The salient features that distinguish the type of atherosclerosis in the diabetic apoE-deficient mouse from that in the non-diabetic apoE-deficient mouse have been described by Candido et al. [20]. The induction of diabetes leads to a much greater infiltration with macrophages and monocytes, to increased proliferating cell nuclear antigen, ␣-smooth muscle actin, collagen content, and expression of connective tissue growth factor and VCAM-1. In some respects this reflects the more aggressive form of atherosclerosis that has also been observed in diabetic subjects in whom the plaques contain large numbers of macrophages, inflammatory cells and smooth muscle cells [11]. Although similar measurements were not made in the present study, it is possible that 1,3-DAG oil modified some of these atherogenic processes. Dyslipidemia is common in patients with type 1 and type 2 diabetes [27,28] and is considered to be a key determinant of accelerated atherosclerosis in diabetes [29–31]. One of the features of diabetic dyslipidemia is altered metabolism of triacylglycerol-rich lipoproteins, comprising increased hepatic production of VLDL and impaired clearance of VLDL and chylomicrons [32]. While hypercholesterolemia is not a pre-requisite for diabetic dyslipidemia in humans, increased numbers of atherogenic remnants and small dense LDL are a feature of severe diabetes [32]. The minor degree of atherosclerosis in the non-diabetic apoE-deficient mice is consistent with the conclusion reached by Meir and Leitersdorf [33] that 20 weeks is insufficient time for substantial development of aortic atherosclerosis in the absence of a saturated-fat high-cholesterol diet. In conclusion, we have demonstrated that dietary 1,3DAG oil reduced atherosclerosis in diabetic apoE-deficient mice that was associated with reduction in plasma cholesterol especially within remnants of triacylglycerol-rich lipoproteins. This is the first study to show that dietary DAG-rich oil ameliorates not only dyslipidemia but also atherosclerosis

associated with diabetes. Although our data are not directly applicable to humans, 1,3-DAG may be a preferable dietary oil to TAG. Acknowledgements Kao Corporation (Tokyo, Japan) provided the test oils and partly funded the project. We acknowledge the assistance of Philip Koh and Sandra Miljavec. TJA is supported by a fellowship from Diabetes Australia and the National Health & Medical Research Council of Australia. References [1] Abdel-Nabey AA, Shehata AAY, Ragab MH, Rossell JB. Glycerides of cottonseed oils from Egyptian and other varieties. Riv Ita Sostanze Grasse 1992;69:443–7. [2] D’alonzo RP, Kozarek WJ, Wade RL. Glyceride composition of processed fats and oils as determined by glass capillary gas chromatography. J Am Oil Chem Soc 1982;59:292–5. [3] Mattson FH, Volpenhein RA. The digestion and absorption of triacylglycerol. J Biol Chem 1966;43:286–9. [4] Kondo H, Hase T, Murase T, Tokimitsu I. Digestion and assimilation features of dietary DAG in the rat small intestine. Lipids 2003;38:25–30. [5] Hara K, Onizawa K, Honda H, et al. Dietary diacylglycerol-dependent reduction in serum triacylglycerol concentration in rats. Ann Nutr Metab 1993;37:185–91. [6] Yamamoto K, Asakawa H, Tokunaga K, et al. Long-term ingestion of dietary diacylglycerol lowers serum triacylglycerol in type II diabetic patients with hypertriglyceridemia. J Nutr 2001;131:3204–7. [7] Yamamoto K, Asakawa H, Tokunaga K, et al. Effects of diacylglycerol administration on serum triacylglycerol in a patient homozygous for complete lipoprotein lipase deletion. Metabolism 2005;54:67– 71. [8] Nagao T, Watanabe H, Goto N, et al. Dietary diacylglycerol suppresses accumulation of body fat compared to triacylglycerol in men in a double-blind controlled trial. J Nutr 2000;130:792–7. [9] Murase T, Aoki M, Wakisaka T, Hase T, Tokimitsu I. Anti-obesity effect of dietary diacylglycerol in C57BL/6J mice: dietary diacylglycerol stimulates intestinal lipid metabolism. J Lipid Res 2002;43:1312–9. [10] Bierman EL. George Lyman Duff Memorial Lecture. Atherogenesis in diabetes. Arterioscler Thromb 1992;12:647–56. [11] Burke AP, Kolodgie FD, Zieske A, et al. Morphologic findings of coronary atherosclerotic plaques in diabetics: a postmortem study. Arterioscler Thromb Vasc Biol 2004;24:1266–71. [12] Rajaram V, Pandhya S, Patel S, et al. Role of surrogate markers in assessing patients with diabetes mellitus and the metabolic syndrome and in evaluating lipid-lowering therapy. Am J Cardiol 2004;93:32C–48C. [13] Ginsberg HN, Illingworth DR. Postprandial dyslipidemia: an atherogenic disorder common in patients with diabetes mellitus. Am J Cardiol 2001;88:9H–15H. [14] Assmann G, Schulte H, von Eckardstein A. Hypertriglyceridemia and elevated lipoprotein(a) are risk factors for major coronary events in middle-aged men. Am J Cardiol 1996;77:1179–84. [15] Jeppesen J, Hein HO, Suadicani P, Gyntelberg F. Triglyceride concentration and ischemic heart disease: an eight-year follow-up in the Copenhagen male study. Circulation 1998;97:1029–36. [16] Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of high-density lipoprotein cholesterol. Veterans affairs high-density lipoprotein cholesterol intervention trial study group. N Engl J Med 1999;341:410–8.

A. Fujii et al. / Atherosclerosis 193 (2007) 55–61 [17] Zhang SH, Reddick RL, Piedrahita JA, Maeda N. Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 1992;258:468–71. [18] Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. ApoEdeficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb 1994;14:133–40. [19] Park L, Raman KG, Lee KJ, et al. Suppression of accelerated diabetic atherosclerosis by the soluble receptor for advanced glycation endproducts. Nat Med 1998;4:1025–31. [20] Candido R, Jandeleit-Dahm KA, Cao Z, et al. Prevention of accelerated atherosclerosis by angiotensin-converting enzyme inhibition in diabetic apolipoprotein E-deficient mice. Circulation 2002;106:246–53. [21] Ishibashi S, Perrey S, Chen Z, et al. Role of the low density lipoprotein (LDL) receptor pathway in the metabolism of chylomicron remnants. A quantitative study in knockout mice lacking the LDL receptor, apolipoprotein E, or both. J Biol Chem 1996;271:22422–7. [22] Mori Y, Nakagiri H, Kondo H, et al. Dietary diacylglycerol reduces postprandial hyperlipidemia and ameliorates glucose intolerance in Otsuka Long-Evans Tokushima Fatty (OLETF) rats. Nutrition 2005;21:933–9. [23] Ijiri Y, Naemura A, Yamashita T, et al. Dietary diacylglycerol extenuates arterial thrombosis in apoE and LDLR deficient mice. Thromb Res 2006;117:411–7. [24] Taguchi H, Omachi T, Nagao T, et al. Dietary diacylglycerol suppresses high fat diet-induced hepatic fat accumulation and microsomal triacylglycerol transfer protein activity in rats. J Nutr Biochem 2002;13:678–83.

61

[25] Chandler CE, Wilder DE, Pettini JL, et al. CP-346086: an MTP inhibitor that lowers plasma cholesterol and triglycerides in experimental animals and in humans. J Lipid Res 2003;44:1887–901. [26] Costet P, Legendre C, More J, et al. Peroxisome proliferatoractivated receptor alpha-isoform deficiency leads to progressive dyslipidemia with sexually dimorphic obesity and steatosis. J Biol Chem 1998;273:29577–88. [27] Dullaart RP. Plasma lipoprotein abnormalities in type 1 (insulindependent) diabetes mellitus. Neth J Med 1995;46:44–54. [28] Haffner SM. Management of dyslipidemia in adults with diabetes. Diab Care 1998;21:160–78. [29] Pedersen TR. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–9. [30] Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and recurrent events trial investigators. N Engl J Med 1996;335:1001–9. [31] Mazzone T. Strategies in ongoing clinical trials to reduce cardiovascular disease in patients with diabetes mellitus and insulin resistance. Am J Cardiol 2004;93:27C–31C. [32] Tomkin GH, Owens D. Abnormalities in apoB-containing lipoproteins in diabetes and atherosclerosis. Diab Metab Res Rev 2001;17:27–43. [33] Meir KS, Leitersdorf E. Atherosclerosis in the apolipoprotein Edeficient mouse. A decade of progress. Arterioscler Thromb Vasc Biol 2004;24:1006–14.