Induction of Atherosclerosis by Low-Fat, Semisynthetic Diets in LDL Receptor–Deficient C57BL/6J and FVB/NJ Mice
Comparison of Lesions of the Aortic Root, Brachiocephalic Artery, and Whole Aorta (En Face Measurement)
Objective— A semisynthetic diet with varying amounts of cholesterol was used to achieve hypercholesterolemia and atherosclerosis in LDL receptor–deficient (LDLR−/−) mice. Atherosclerotic lesions were measured as cross-sectional area at the aortic root and brachiocephalic artery and by en face analysis of aortic lesion area in 209 male and female animals on the C57BL/6J (B6.LDLR−/−) and FVB/NJ (FVB.LDLR−/−) backgrounds.
Methods and Results— The semisynthetic diet containing 4.3% fat and 0.00% or 0.02% cholesterol was sufficient to induce hypercholesterolemia (12.6±2.4 mmol/L) and atherosclerosis in B6.LDLR−/− mice at the aortic root (98 980±37 727 μm2) and brachiocephalic artery (12 039±12 750 μm2) but did not produce significant lesions in the aorta measurable by the en face method. Raising dietary cholesterol to 0.15%, 0.30%, or 0.50% more than doubled plasma cholesterol levels (35.9±8.5 mmol/L) and resulted in significant en face lesions. It also led to a significant increase in atherosclerotic lesion area at the aortic root (547 753±182 151 μm2) and brachiocephalic arteries (125 666±59 339 μm2). Although FVB.LDLR−/− mice developed comparable cholesterol levels, they were relatively atherosclerosis resistant and had many-fold smaller lesions.
Conclusions— These results should aid investigations of atherosclerosis in LDLR−/− mice by informing the selection of diet to be used and the location of lesions to be scored.
In the last 10 years, induced mutant mouse models of atherosclerosis have been created and used to study the mechanisms of lesion formation, diet and drug effects on the extent of lesions, and the role of candidate genes in atherosclerosis susceptibility. These studies have generally used apolipoprotein E–deficient and LDL receptor–deficient (LDLR−/−) mice.1,2 In the former, diets have consisted of either chow or Western-type diets,3 whereas in the latter, Western-type4 or high-cholesterol, high-fat, cholate-containing diets have been used to induce hypercholesterolemia and atherosclerosis.5 Each of these diets has drawbacks. The chow diet consists of ≈0.02% cholesterol and 4.5% fat (wt/wt), but the fat content is not defined and can vary, depending on availability and cost factors. A chow diet induces lesions relatively slowly, if at all, in LDLR−/− mice.5,6 The Western-type diet usually consists of 0.15% cholesterol and 20% (wt/wt) fat, and when fed to certain strains of mice, such as C57BL/6J, can induce obesity, with all of the attendant metabolic complications.7 The high-cholesterol, high-fat, cholate-containing diet usually consists of 1.25% cholesterol, 15% fat (wt/wt), and 0.5% cholic acid.8 It can induce an inflammatory response in mice that may complicate atherosclerosis end points.9 These drawbacks make it clear that further work needs to be done on developing better diets for studies of mouse models of atherosclerosis.
In studies of atherosclerosis in the mouse, lesions have traditionally been quantified in oil red O–stained sections through the aortic root by the method of Paigen et al.8 In other studies, investigators have quantified, in en face preparations of the entire aorta, the percentage of surface area occupied by lesions.10 Most recently, it was noted that brachiocephalic artery lesions are larger and more advanced than are those in other areas. Descriptions of this region have been used mainly for qualitative purposes,11,12 but quantification has been done in a few studies.13,14 The literature contains just one study that compared the aortic root cross-sectional lesion area and the en face aortic lesion area for 8 apolipoprotein E−/− and 11 LDLR−/− outbred mice fed Western-type diets.10 Comparisons of lesion development in inbred strains at different sites on different diets are clearly needed.
The current study had 2 major goals. The first was to establish a semisynthetic diet that would achieve hypercholesterolemia and atherosclerosis in LDLR−/− mice without toxicity from very high cholesterol and cholic acid components and without the weight gain induced by high-fat feeding. The other goal was to compare lesion development at the aortic root and brachiocephalic cross-sectional areas and en face aortic lesion area in atherosclerosis-sensitive C57BL/6J and -resistant FVB/NJ mice on the LDLR−/− background. This would allow the determination of regional effects of atherogenic stimuli on lesion development and provide information as to the best method for use in a variety of mouse model studies of genetic and environmental effects on atherosclerosis.
An expanded Methods section is provided in the online supplement (please see http://atvb.ahajournals.org).
Mice and Diets
LDLR −/− mice on the C57BL/6J background (B6.129S7-Ldlrtm1Her, stock No. 002207; henceforth called B6.LDLR−/−) and FVB/NJ mice (stock No. 001800) were purchased from the Jackson Laboratory (Bar Harbor, Me). FVB.LDLR−/− mice were generated in our laboratory by marker-assisted backcrossing15 of the LDLR −/− trait from B6.LDLR−/− to FVB/NJ. Mice were weaned at 28 days of age and fed a semisynthetic, modified AIN76 diet16 containing 0.00%, 0.02%, 0.15%, 0.30%, or 0.50% cholesterol for 16 weeks until sacrifice at 20 weeks of age. Each dietary group included 8 to 14 male and 4 to 9 female B6.LDLR−/− mice and 12 to 15 male and 8 to 15 female FVB.LDLR−/− mice.
Plasma was obtained from fasted animals anesthetized with methoxyflurane at 8, 12, and 16 weeks of age and at 20 weeks, 2 days before sacrifice. Plasma cholesterol, triglyceride, and glucose concentrations were determined enzymatically. Cholesterol distribution among lipoproteins was determined by fast protein liquid chromatography.
Mice were humanely killed at 20 weeks of age. The heart was removed by cutting the ascending aorta halfway between the aortic root and the brachiocephalic artery. The latter was cut at the point where it branches from the aorta and 1 mm distal to its bifurcation into the subclavian and carotid arteries, and the resultant Y-shaped piece was frozen in mounting medium (OCT compound, Tissue-Tek). The remaining aorta was dissected to the iliac bifurcation, opened longitudinally, and fixed between glass slides.
Quantification of Atherosclerosis
To quantify cross-sectional lesion area in the aortic root, formalin-fixed hearts were processed as previously described.8,17 To quantify cross-sectional lesion area in the brachiocephalic artery, the Y-shaped piece of brachiocephalic artery was sectioned distally to proximally at 10-μm thickness, starting from the subclavian and carotid arteries. Atherosclerotic lesions lumenal to the internal elastic lamina were quantified in 6 equidistant (100-μm) oil red O–stained sections 200 to 700 μm from the branching point of the brachiocephalic into the carotid and subclavian arteries. En face lesion area of the aorta was quantified relative to its surface area.
Liver Cholesterol Concentration and Histology
Liver cholesterol concentrations were determined as previously described.18 Liver lipid content was also assessed by histology in oil red O–stained, frozen sections.
Values are given as mean±SD. Statistical analysis was done by t test and ANOVA.
At 4 weeks of age, B6.LDLR−/− mice and FVB.LDLR−/− mice were weaned to a modified AIN76A semisynthetic diet containing 0.00%, 0.02%, 0.15%, 0.30%, or 0.50% cholesterol for 16 weeks. At the end of this time period, B6.LDLR−/− mice weighed less than did FVB.LDLR−/− mice (males: 28.4±2.5 vs 30.5±2.4 g, P<0.001; females: 22.2±1.3 vs 23.9±2.3 g, P<0.001; respectively). The amount of cholesterol in the diet had no effect on body weight in B6.LDLR−/− males, B6.LDLR−/− females, and FVB.LDLR−/− males (ANOVA, P=NS). Although the effect in FVB.LDLR−/− females was statistically significant (ANOVA, P<0.001), this was due to several outliers (Table I in online supplement; please see http://atvb.ahajournals.org).
Plasma cholesterol concentrations were measured after the mice had been on the 0.00%, 0.02%, 0.15%, 0.30%, or 0.50% cholesterol diets for 4, 8, 12, or 16 weeks. As shown in Figure 1, in most cases, there were no significant temporal trends in plasma cholesterol (ANOVA, P=NS). In a few cases, cholesterol levels increased modestly but significantly from 4 to 8 weeks and then remained constant through 16 weeks. It was only in B6.LDLR−/− males fed the 0.50% cholesterol that the cholesterol level increased significantly (48%) between 12 and 16 weeks (Tukey post test, P<0.001). Therefore, in general, cholesterol levels attained a steady state by 4 weeks and at most by 8 weeks and remained at such levels through 16 weeks.
As shown in Figure 1, it is quite remarkable that for each strain and sex, the plasma cholesterol level for the 0.00% and 0.02% cholesterol diets was similar and significantly lower than with the 0.15%, 0.30%, and 0.50% cholesterol diets, which in turn had similar plasma cholesterol levels. When the former and latter groups of mice were compared, B6.LDLR−/− males had cholesterol levels of 12.6±2.4 mmol/L (487±93 mg/dL) and 35.9±8.5 mmol/L (1389±330 mg/dL); B6.LDLR−/− females, 16.4±2.9 mmol/L (633±112 mg/dL) and 32.3±3.1 mmol/L (1250±120 mg/dL); FVB.LDLR−/− males, 14.7±2.2 mmol/L (567±84 mg/dL) and 27.5±3.2 mmol/L (1063±125 mg/dL); and FVB.LDLR−/− females, 10.7±1.6 mmol/L (416±61 mg/dL) and 26.6±2.8 mmol/L (1030±108 mg/dL), respectively. In a comparison of cholesterol levels between strains, the B6.LDLR−/− males fed the 0.00% and 0.02% cholesterol diets had 14% lower levels than did the FVB.LDLR−/− males (P=0.002), and B6.LDLR−/− females had 52% higher levels than did FVB.LDLR−/− females (P<0.001); for the 0.15%, 0.30%, and 0.50% cholesterol diets, B6.LDLR−/− males had 31% higher levels than did FVB.LDLR−/− males (P<0.001), and B6.LDLR−/− females had 21% higher levels than did FVB.LDLR−/− females (P<0.001). It appears that the main difference in plasma cholesterol levels occurred between the low- and high-cholesterol diets, with relatively smaller differences between strains fed the same diet.
The cholesterol distribution between lipoprotein particles was analyzed by fast protein liquid chromatography in samples of pooled plasma after 16 weeks of diet feeding (Figure 2). In each strain, the lipoprotein pattern was similar in the 0.00% and 0.02% cholesterol diets, which in turn differed from the 0.15%, 0.30%, and 0.50% cholesterol diets, which were similar to each other. Mice fed the low-cholesterol diets had lower VLDL, IDL, and LDL and higher HDL cholesterol levels than did mice fed the high-cholesterol diets. On the high-cholesterol diets, B6.LDLR−/− mice had considerably higher VLDL and IDL cholesterol levels and the suggestion of larger LDL particles than did FVB.LDLR−/− mice. On both low- and high-cholesterol diets, B6.LDLR−/− mice had lower HDL cholesterol levels than did the FVB.LDLR−/− mice.
Plasma triglyceride concentrations were also determined after 16 weeks of diet feeding (Table II in online supplement; please see http://atvb.ahajournals.org). Dietary cholesterol had no significant effect on triglyceride concentrations in male mice (ANOVA, P=NS). There were significant effects in female mice (B6.LDLR−/− and FVB.LDLR−/−; ANOVA P=0.005); however, these effects did not show a dose response with dietary cholesterol. In a comparison of triglyceride levels between strains, males showed no difference (B6.LDLR−/−, 3.6±0.8 mmol/L and FVB.LDLR−/−, 3.8±0.9 mmol/L; t test, P=NS), whereas in females, triglyceride values were 13% lower in B6.LDLR−/− than in FVB.LDLR−/− mice (2.9±0.5 and 3.4±0.6 mmol/L, respectively; t test, P<0.001).
The effect of dietary cholesterol on liver cholesterol content was examined in male mice of both strains. As shown in online Figure IA and IB (please see http://atvb.ahajournals.org), for each strain, liver cholesterol contents for the 0.00% and 0.02% cholesterol diets were similar and significantly lower than on the 0.15%, 0.30%, and 0.50% cholesterol diets, which in turn had similar liver cholesterol contents. B6.LDLR−/− mice fed the low-cholesterol diets had total liver cholesterol contents of 4.5±0.9 mg/g, free cholesterol of 3.4±0.5 mg/g, and cholesterol ester of 1.1±1.1 mg/g, whereas on the high-cholesterol diets, the values were 18.1±6.3, 4.3±0.6, and 13.8±6.2 mg/g, respectively (t test; P<0.001, P=0.009, and P<0.001, respectively). FVB.LDLR−/− mice fed the low-cholesterol diets had total liver cholesterol contents of 5.3±1.7 mg/g, free cholesterol of 3.7±1.0 mg/g, and cholesterol ester of 1.6±1.0 mg/g, whereas on the high-cholesterol diets, the values were 34.0±9.7, 4.6±0.5, and 29.4±9.4 mg/g, respectively (t test; P<0.001, P=0.046, and P<0.001, respectively). In a comparison between strains, B6.LDLR−/− and FVB.LDLR−/− mice had comparable liver cholesterol contents when fed the 0.00% and 0.02% cholesterol diets. However, on the 0.15%, 0.30%, and 0.50% cholesterol diets, B6.LDLR−/− mice had 47% lower total liver cholesterol, the same free cholesterol, and 53% lower cholesterol ester contents than did FVB.LDLR−/− mice (P<0.001, P=NS, and P<0.001, respectively). These biochemical findings were corroborated by oil red O-staining of livers from B6.LDLR−/− and FVB.LDLR−/− mice (online Figure IC and ID; please see http://atvb.ahajournals.org). Thus, high-cholesterol diets appeared to increase cholesterol ester content in both strains, with the greater increase in FVB.LDLR−/− mice.
Dietary cholesterol had no significant effect on glucose concentrations in male and female B6.LDLR−/− mice and female FVB.LDLR−/− mice after 16 weeks of diet feeding (online Table III; please see http://atvb.ahajournals.org). There was a significant effect in male FVB.LDLR−/− mice (ANOVA, P=0.003), but there was no dose response. However, the glucose concentrations were significantly higher in B6.LDLR−/− mice than in FVB.LDLR−/− mice (males: 13.2±2.4 vs 9.4±2.7 mmol/L, P<0.001; females: 11.9±2.2 vs 7.9±2.5 mmol/L, P<0.001). Dietary cholesterol had no significant effect on blood pressure (online Table IV; please see http://atvb.ahajournals.org), but male B6.LDLR−/− mice had significantly lower blood pressure than did male FVB.LDLR−/− mice (102±6 vs 109±5 mm Hg, P<0.001).
The effects of dietary cholesterol on atherosclerotic lesion formation were quantified in B6.LDLR−/− and FVB.LDLR−/− mice after 16 weeks of diet feeding (20 weeks of age) as cross-sectional area at the aortic root and brachiocephalic artery and by en face analysis in the whole aorta. Aortic root cross-sectional lesion areas for the 0.00% and 0.02% cholesterol diets were similar and significantly lower than for the 0.15%, 0.30%, and 0.50% cholesterol diets, which in turn were similar to each other (Figure 3). When the former and latter groups of mice were compared, B6.LDLR−/− males had cross-sectional areas of 98 980±37 727 μm2 and 547 753±182 151 μm2; B6.LDLR−/− females, 309 951±72 446 μm2 and 842 214±223 712 μm2; FVB.LDLR−/− males, 14 165±6291 μm2 and 152 664±46 218 μm2; and FVB.LDLR−/− females, 11 904±4832 μm2 and 130 730±39 568 μm2, respectively. In a comparison between strains, 0.00% and 0.02% cholesterol diet–fed B6.LDLR−/− males had lesions 7.0-fold larger than those of FVB.LDLR−/− males (P<0.001), and B6.LDLR−/− females, 26-fold larger lesions than did FVB.LDLR−/− females (P<0.001). In addition, 0.15%, 0.30%, and 0.50% cholesterol diet–fed B6.LDLR−/− males had lesions 3.6-fold larger than those of FVB.LDLR−/− males (P<0.001), and B6.LDLR−/− females, 6.4-fold larger lesions than did FVB.LDLR−/− females (P<0.001). Thus, there were both highly significant diet and strain effects on aortic root cross-sectional lesion areas.
Another site that allows standardized quantification of atherosclerosis is the brachiocephalic artery (Figure 4). By this technique, cross-sectional lesion areas for the 0.00% and 0.02% cholesterol diets were similar and significantly lower than for the 0.15%, 0.30%, and 0.50% cholesterol diets, which in turn were similar to each other (Figure 5). When the former and latter groups were compared, B6.LDLR−/− males had cross-sectional areas of 12 039±12 750 μm2 and 125 666±59 339 μm2; B6.LDLR−/− females, 44 068±27 325 μm2 and 208 406±91 160 μm2; FVB.LDLR−/− males, 250±273 μm2 and 5455±5414 μm2; and FVB.LDLR−/− females, 567±745 μm2 and 3853±4590 μm2, respectively. In a comparison between strains, 0.00% and 0.02% cholesterol diet–fed B6.LDLR−/− males had lesions 48-fold larger than those of FVB.LDLR−/− males (P<0.001), and B6.LDLR−/− females, 78-fold larger lesions than did FVB.LDLR−/− females (P<0.001). In addition, 0.15%, 0.30%, and 0.50% cholesterol diet–fed B6.LDLR−/− males had lesions 23-fold larger than did FVB.LDLR−/− males (P<0.001), and B6.LDLR−/− females, 54-fold larger lesions than those of FVB.LDLR−/− females (P<0.001). Thus, there were both highly significant diet and strain effects on brachiocephalic artery cross-sectional lesion areas. Compared with the aortic root, both the diet and strain effects were larger for the brachiocephalic artery; however, the coefficients of variation were also larger. The distribution of lesions across a 500-μm segment of the brachiocephalic artery was uniform in B6.LDLR−/− mice, whereas lesions in FVB.LDLR−/− mice on the high-cholesterol diets tended to be larger closer to the bifurcation (supplementary online Figure II; please see http://atvb.ahajournals.org).
Atherosclerosis was also measured as the percentage of aortic surface area covered by lesions (en face analysis) in male mice. As shown in Figure 6, for both B6.LDLR−/− and FVB.LDLR−/− mice, in the aortic arch and thoracic aorta, 0.00% and 0.02% cholesterol diets resulted in very significantly lower en face lesion areas compared with 0.15%, 0.30%, and 0.50% cholesterol diets (P<0.001). In B6.LDLR−/− mice in the abdominal aorta, 0.50% cholesterol was required to raise en face lesion area (ANOVA, P<0.001; Tukey post test, 0.50% cholesterol diet vs each of the other diets, P<0.05 or less). On the 0.00% and 0.02% cholesterol diets, mice of both strains had marginal atherosclerosis, with mean en face lesion areas <5%. With regard to comparisons between strains, the B6.LDLR−/− mice fed the high-cholesterol diets had a 4.1-fold larger en face lesion area in the aortic arch than did FVB.LDLR−/− mice (53±11% vs 13±5%, P<0.001). However, both strains, when fed the high-cholesterol diets, did not differ in en face lesion area in the thoracic (16±10% vs 16±8%, P=NS) or abdominal (7±7% vs 9±4%, P=NS) aortic segments.
This study was designed to establish a diet that produces hypercholesterolemia and atherosclerosis in LDLR−/− mice as well as to compare lesion development by quantifying aortic root and brachiocephalic cross-sectional area and en face aortic lesion area. We found that a semisynthetic diet containing 4.3% fat and 0.00% or 0.02% cholesterol was sufficient to induce atherosclerosis in B6.LDLR−/− mice. This was quite apparent by cross-sectional lesion area at both the base of the aorta and the brachiocephalic artery. However, to obtain significant lesions in the aorta measurable by the en face method, it was necessary to raise dietary cholesterol to 0.15% or above and increase plasma cholesterol levels by 2-fold or more. As predicted by studies in FVB.apoE−/− mice,19 the current study found FVB.LDLR−/− mice to be relatively resistant to atherosclerosis. On all diets, B6.LDLR−/− mice had many-fold larger atherosclerotic lesions than did FVB.LDLR−/− mice, measured as cross-sectional lesion area at both the aortic root and the brachiocephalic artery. The fold changes induced by dietary cholesterol and the fold differences between strains were greater for the brachiocephalic artery than for the aortic root; however, the coefficient of variation of measurement was also greater. As in B6.LDLR−/− mice, FVB.LDLR−/− mice required 0.15% or greater cholesterol diets to develop signifi-cant en face lesion areas. By this method of analysis, B6.LDLR−/− mice had more en face lesion area in the aortic arch only but not in the thoracic or abdominal aorta. Thus, depending on the experimental goals, the diet and location of lesion analysis must be chosen carefully.
On a chow diet, LDLR−/− mice have cholesterol levels of 5.2 to 6.5 mmol/L (200 to 250 mg/dL), with non-HDL cholesterol values of ≈2.6 to 3.9 mmol/L (100 to 150 mg/dL).6,20 When LDLR−/− mice were fed this diet, they did not develop significant atherosclerotic lesions. When these mice were fed 7.5% (wt/wt) cocoa butter, 1.25% cholesterol, and 0.5% sodium cholate, their plasma cholesterol levels increased to between 33.7 and 74.8 mmol/L (1304 and 2893 mg/dL, respectively), mainly in the VLDL, IDL, and LDL fractions. This diet produced extensive atherosclerotic lesions both at the aortic root and on the aortic surface.5 Unfortunately, the high-cholesterol, high-fat, cholate diet causes a chronic inflammatory state in mice with expression of oxidative stress genes, colony stimulating factors, heme oxygenase, and nuclear factor-κB activation.9 To circumvent the use of cholic acid, Lichtman et al16 fed a semisynthetic, modified AIN76 diet containing 20% fat (wt/wt) and either 0.5% cholesterol, 1.25% cholesterol, or 1.25% cholesterol plus 0.5% cholic acid. These diets resulted in plasma cholesterol levels of 8.5, 15.4, and 19.7 mmol/L (328, 597, and 761 mg/dL), respectively, and respective total aortic en face lesion areas of 7.0%, 8.3%, and 12.8%. They did not measure cross-sectional lesion area at the aortic root or the brachiocephalic artery. They concluded that cholate was not a necessary component of the high-cholesterol, high-fat diet for the formation of atherosclerosis in LDLR−/− mice.16 Other investigators have circumvented the need for cholate through the use of the Western-type diet, which results in plasma cholesterol levels of ≈26 mmol/L (1000 mg/dL) and causes lesions at the aortic root and the aortic surface.4,10
The current study used the same semisynthetic diet of Lichtman et al, but it contained only 4.3% fat (wt/wt) and 0.00%, 0.02%, 0.15%, 0.30%, or 0.50% cholesterol. The Lichtman study included a fourth diet group exactly comparable to the one in our study fed 0.00% cholesterol. In that group after 12 weeks on the diet, Lichtman et al reported a plasma cholesterol value of 3.2 mmol/L (124 mg/dL, males only) and no significant lesions by the en face method.16 The B6.LDLR−/− male mice in our study at 12 and 16 weeks on the diet had cholesterol levels of ≈11.6 mmol/L (450 mg/dL). The mice reported in the Lichtman study were of mixed genetic background, but it is unlikely that this explains the large difference in plasma cholesterol levels between the studies. However, in agreement with the Lichtman group, our mice that were fed the 0.00% cholesterol diet did not have significant aortic surface lesions. Our mice did have significant cross-sectional lesion areas at both the aortic root and the brachiocephalic artery, a measurement not made in their study. In addition, we found that supplementation with as little as 0.15% cholesterol alone, which raised plasma cholesterol to 32.7±5.7 mmol/L (1264±219 mg/dL) in males and to 32.1±3.2 mmol/L (1244±122 mg/dL) in females, was sufficient to produce significant lesions on the aortic surface while increasing cross-sectional lesion area at both the aortic root and brachiocephalic artery. Although we have not tested diets with cholesterol contents intermediate between 0.02% and 0.15%, one might speculate that these might reveal a more sensitive difference of lesion size between strains and sex at different sites of the vascular bed.
As noted earlier and confirmed in our laboratory, chow-fed B6.LDLR−/− mice have plasma cholesterol values of 5.2 to 6.5 mmol/L (200 to 250 mg/dL) and do not develop significant atherosclerotic lesions.5,6 In the current study, mice fed a semisynthetic diet with comparable fat and no cholesterol added had twice the plasma cholesterol and significant atherosclerotic lesions. This was somewhat surprising and implicates the different dietary sources of macronutrients and micronutrients in the 2 diets as the cause. The AIN76 diet uses casein as the exclusive protein source, which in rodents has been shown to be hypercholesterolemic,21,22 whereas chow uses a variety of protein sources. The AIN76 diet uses soy and cocoa butter as the lipid source, whereas chow uses mainly soy. It is beyond the scope of this study to determine the causative nutrient(s) responsible for the increase in plasma cholesterol and atherosclerosis.
High-fat diets are known to induce obesity in certain strains of mice, including C57BL/6J, when fed for prolonged periods.7 Obesity is accompanied by various metabolic derangements, such as hyperglycemia, insulin resistance, and hypertension, which are risk factors for atherosclerotic disease. Thus, the use of high-fat diets in atherosclerosis studies with B6.LDLR−/− mice could have unintended consequences that might alter the interpretation of the data. In fact, prolonged feeding of the Western-type diet to B6.LDLR−/− mice does result in excessive weight gain.7 This has been less of a problem for a high-cholesterol, high-fat, cholate diet fed to B6.LDLR−/− mice. However, the latter is true probably because weight gain is modulated by the very high amount of cholesterol (compared with chow or Western-type diets) and the presence of cholic acid in the diet. In the current study, the dietary fat content was kept low and the atherogenic stimulus was varied by increasing the amount of cholesterol in the diet. By this method, it was possible to induce even extreme amounts of atherosclerosis without excessive weight gain compared with a chow diet. We did find that liver lipids increased at 0.15% cholesterol in the diet. Thus, for most studies of atherosclerosis progression, we recommend the 0.02% cholesterol diet. However, if it is necessary to induce lesions on the aortic surface, as for regression studies, we recommend the 0.15% cholesterol diet.
With regard to studies that compare atherosclerosis-sensitive and -resistant strains of mice, the biggest differences were observed in cross-sectional lesion area measurements at the aortic root and the brachiocephalic artery on the 0.00% and 0.02% cholesterol diets. At higher dietary cholesterol contents, the fold difference in lesion area was reduced. It was necessary to use 0.15% dietary cholesterol or above to induce lesions on the surface of the aorta, but differences between strains were observable only at the aortic arch, not the thoracic or abdominal aorta. These results strongly suggest that en face measurement of the whole aorta is not the best measurement. However, differences in atherosclerosis susceptibility observed at different locations in the vascular bed might also reflect genetic factors that might differ, depending on the anatomic location. In genetic mapping studies, we cannot say at this point whether the location of lesion assessment influences the loci revealed, but this is certainly a possibility.
In summary, this article presents a model in which a semisynthetic diet with increasing amounts of cholesterol was used to induce atherosclerosis, uncomplicated by changes in nonlipid classic risk factors. Data are also presented in atherosclerosis-sensitive and -resistant strains. Depending on the goals of future research in mouse models of atherosclerosis, this information will be important in rationally designing such studies.
This work was supported by National Institutes of Health grants HL54591-08 and HL70524-02. Dr Daniel Teupser is a fellow of the Emmy Noether-Program of the Deutsche Forschungsgemeinschaft (DFG Te 342/1-1). The authors thank Dr Hayes M. Dansky for the initial backcrossing of the LDLR−/− trait to the FVB background.
- Received May 29, 2003.
- Accepted July 28, 2003.
Knowles JW, Maeda N. Genetic modifiers of atherosclerosis in mice. Arterioscler Thromb Vasc Biol. 2000; 20: 2336–2345.
Palinski W, Tangirala RK, Miller E, Young SG, Witztum JL. Increased autoantibody titers against epitopes of oxidized LDL in LDL receptor–deficient mice with increased atherosclerosis. Arterioscler Thromb Vasc Biol. 1995; 15: 1569–1576.
Ishibashi S, Goldstein JL, Brown MS, Herz J, Burns DK. Massive xanthomatosis and atherosclerosis in cholesterol-fed low density lipoprotein receptor-negative mice. J Clin Invest. 1994; 93: 1885–1893.
Sehayek E, Ono JG, Duncan EM, Batta AK, Salen G, Shefer S, Neguyen LB, Yang K, Lipkin M, Breslow JL. Hyodeoxycholic acid efficiently suppresses atherosclerosis formation and plasma cholesterol levels in mice. J Lipid Res. 2001; 42: 1250–1256.
Merat S, Casanada F, Sutphin M, Palinski W, Reaven PD. Western-type diets induce insulin resistance and hyperinsulinemia in LDL receptor–deficient mice but do not increase aortic atherosclerosis compared with normoinsulinemic mice in which similar plasma cholesterol levels are achieved by a fructose-rich diet. Arterioscler Thromb Vasc Biol. 1999; 19: 1223–1230.
Liao F, Andalibi A, deBeer FC, Fogelman AM, Lusis AJ. Genetic control of inflammatory gene induction and NF-κB-like transcription factor activation in response to an atherogenic diet in mice. J Clin Invest. 1993; 91: 2572–2579.
Tangirala RK, Rubin EM, Palinski W. Quantitation of atherosclerosis in murine models: correlation between lesions in the aortic origin and in the entire aorta, and differences in the extent of lesions between sexes in LDL receptor–deficient and apolipoprotein E-deficient mice. J Lipid Res. 1995; 36: 2320–2328.
Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, Schwartz SM. Advanced atherosclerotic lesions in the innominate artery of the apoE knockout mouse. Arterioscler Thromb Vasc Biol. 2000; 20: 2587–2592.
Bentzon JF, Skovenborg E, Hansen C, Moller J, de Gaulejac NS, Proch J, Falk E. Red wine does not reduce mature atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2001; 103: 1681–1687.
Reardon CA, Blachowicz L, White T, Cabana V, Wang Y, Lukens J, Bluestone J, Getz GS. Effect of immune deficiency on lipoproteins and atherosclerosis in male apolipoprotein E–deficient mice. Arterioscler Thromb Vasc Biol. 2001; 21: 1011–1016.
Lichtman AH, Clinton SK, Iiyama K, Connelly PW, Libby P, Cybulsky MI. Hyperlipidemia and atherosclerotic lesion development in LDL receptor–deficient mice fed defined semipurified diets with and without cholate. Arterioscler Thromb Vasc Biol. 1999; 19: 1938–1944.
Plump AS, Scott CJ, Breslow JL. Human apolipoprotein A-I gene expression increases high density lipoprotein and suppresses atherosclerosis in the apolipoprotein E-deficient mouse. Proc Natl Acad Sci U S A. 1994; 91: 9607–9611.
Sehayek E, Ono JG, Shefer S, Nguyen LB, Wang N, Batta AK, Salen G, Smith JD, Tall AR, Breslow JL. Biliary cholesterol excretion: a novel mechanism that regulates dietary cholesterol absorption. Proc Natl Acad Sci U S A. 1998; 95: 10194–10199.
Dansky HM, Charlton SA, Sikes JL, Heath SC, Simantov R, Levin LF, Shu P, Moore KJ, Breslow JL, Smith JD. Genetic background determines the extent of atherosclerosis in apoE-deficient mice. Arterioscler Thromb Vasc Biol. 1999; 19: 1960–1968.
Ishibashi S, Brown MS, Goldstein JL, Gerard RD, Hammer RE, Herz J. Hypercholesterolemia in low density lipoprotein receptor knockout mice and its reversal by adenovirus-mediated gene delivery. J Clin Invest. 1993; 92: 883–893.
Geelen MJ, van Hoorn D, Beynen AC. Consumption of casein instead of soybean protein produces a transient rise in the concentration of sphingomyelin in VLDL in rats. J Nutr. 1999; 129: 2119–2122.