Dietary manipulation and social isolation alter disease progression in a murine model of coronary heart disease

Abstract

Background: Mice with a deficiency in the HDL receptor SR-BI and low expression of a modified apolipoprotein E gene (SR-BI KO/ApoeR61(h/h)) called 'HypoE' when fed an atherogenic, 'Paigen' diet develop occlusive, atherosclerotic coronary arterial disease (CHD), myocardial infarctions (MI), and heart dysfunction and die prematurely (50% mortality ~40 days after initiation of this diet). Because few murine models share with HypoE mice these cardinal, human-like, features of CHD, HypoE mice represent a novel, small animal, diet-inducible and genetically tractable model for CHD. To better describe the properties of this model, we have explored the effects of varying the composition and timing of administration of atherogenic diets, as well as social isolation vs. group housing, on these animals.

Methodology/principal findings: HypoE mice were maintained on a standard lab chow diet (control) until two months of age. Subsequently they received one of three atherogenic diets (Paigen, Paigen without cholate, Western) or control diet for varying times and were housed in groups or singly, and we determined the plasma cholesterol levels, extent of cardiomegaly and/or survival. The rate of disease progression could be reduced by lowering the severity of the atherogenic diet and accelerated by social isolation. Disease could be induced by Paigen diets either containing or free of cholate. We also established conditions under which CHD could be initiated by an atherogenic diet and then subsequently, by replacing this diet with standard lab chow, hypercholesterolemia could be reduced and progression to early death prevented.

Conclusions/significance: HypoE mice provide a powerful, surgery-free, diet-'titratable' small animal model that can be used to study the onset of recovery from occlusive, atherosclerotic CHD and heart failure due to MI. HypoE mice can be used for the analysis of the effects of environment (diet, social isolation) on a variety of features of cardiovascular disease.

Figure 1. Effects of atherogenic diets on lipoprotein cholesterol profiles of HypoE mice.

HypoE mice housed in groups (including one or more SRBI^+/−ApoeR61^h/h littermates per group, 4–5/cage) were fed Western (open/white symbols, n = 5), Paigen NC (gray symbols, n = 5) or Paigen (black symbols, n = 3) diets for one month starting at two months of age and then plasma was harvested and subjected to FPLC size fractionation. The data shown include only the HypoE mice. A. Averaged plasma lipoprotein total cholesterol (TC) profiles (mg/dL plasma). Brackets indicate the approximate elution positions of VLDL, intermediate-density lipoproteins (IDL)/LDL, and HDL. (Inset) Expanded scale for the IDL/LDL and HDL regions of the profiles. B. For each profile from individual mice, total cholesterol levels in the indicated pooled fractions corresponding to VLDL-, IDL/LDL- or HDL-size particles were summed and averages were calculated. Data are represented as mean ± SD. Statistically significant differences were determined by Kruskal-Wallis tests followed by the Dunn’s multiple comparison post-test. *p<0.05.

Figure 2. Effects of atherogenic diets on survival (A) and heart-to-body weight ratios (B) of HypoE mice.

Mice were housed in groups with mixed genotypes (HypoE and one or more SRBI^+/−ApoeR61^h/h littermates per group, 4–5/cage) and, beginning at two months of age were switched from a normal chow diet to the indicated atherogenic diets. The data shown include only the HypoE mice. A Kaplan-Meier survival curves. Mice were fed either Paigen (gray solid line, n = 15), Paigen NC (gray dotted line, n = 14), or Western (black dashed line, n = 15) diets. B Heart-to-body weight ratios. For the indicated times (1–3 months), mice were fed the Paigen (n = 14), chow (n = 8), Western (n = 5 (1 month), n = 4 (3 months)), or Paigen NC (n = 6 (1 month), n = 4 (2 months)) diets. Gray and white bars represent those populations in which approximately 50% or 100% of the animals survived after the indicated times of feeding. Data are means ± SD. Statistically significant differences were determined by Kruskal-Wallis tests followed by the Dunn’s multiple comparison post-test. *p<0.05.

Figure 3. Population density (A) and mixed genotype housing (B) effects on Paigen diet-fed HypoE mouse survival.

Mice were housed at the indicated population densities (1, 2–3 or 4–5 (group)/cage) with either uniform (HypoE) or mixed genotypes (HypoE and one or more SRBI^+/−ApoeR61^h/h littermates per group). Beginning at two months of age the animals were switched from a normal chow diet to the standard Paigen diet. The data shown include only the HypoE mice. A Kaplan-Meier survival curves for HypoE mice housed singly (1 mouse/cage, black line), or uniform genotype (HypoE only) groups of 2–3 (light gray dashed line), or 4–5 (light gray solid line) mice per cage with the indicated median survival times and number of animals per group (n). A notch in the 4–5 mice per cage line indicates a censored animal that was euthanized due to severe ulcerative dermatitis at 35 days after initiating Paigen diet feeding. B Kaplan-Meier survival curves for HypoE mice housed singly (black line) or in uniform genotype (HypoE only, light gray line) or mixed genotype (HypoE and SRBI^+/−ApoeR61^h/h E, gray line) group housing (4–5 mice/cage).

Figure 4. Effects of population density and atherogenic diets on the survival of Paigen diet-fed HypoE mice.

HypoE mice were housed singly (1/cage, black lines) or in mixed genotype groups (HypoE and one or more SRBI^+/−ApoeR61^h/h littermates per group, 4–5/cage, gray lines). Beginning at two months of age the animals were switched from a normal chow diet to either the standard Paigen (solid lines) or Paigen NC (dashed lines) diets. The Kaplan-Meier survival curves shown include only the HypoE mice. A notch in the singly housed HypoE mice fed the Paigen NC diet indicates a censored animal, which was euthanized due to severe ulcerative dermatitis at 57 days after initiating Paigen NC diet feeding.

Figure 5. Effects of population density and sex on plasma cholesterol levels of Paigen diet-fed HypoE mice.

HypoE mice were housed singly (1/cage, black bars; males, n = 16, females; n = 13) or in mixed genotype groups (HypoE and one or more SRBI^+/−ApoeR61^h/h littermates per group, 4–5/cage, gray bars; males, n = 19, females; n = 24). Beginning at two months of age the animals were switched from a normal chow diet to the Paigen diet. After 19 days of Paigen diet feeding, plasma was harvested and total cholesterol (TC), unesterified cholesterol (UC) and UC/TC ratios were determined from the HypoE mice. Data are means ± SD. Statistically significant differences were determined by unpaired Student’s t or Mann-Whitney test.

Figure 6. Population density and sex effects on plasma corticosterone and oxytocin in Paigen diet-fed HypoE mice.

HypoE mice were housed in mixed genotype groups (HypoE and one or more SRBI^+/−ApoeR61^h/h littermates per group, 4–5/cage, gray bars) or singly (1/cage, black bars). Beginning at two months of age the animals were switched from a normal chow diet to the Paigen diet. After 19 days of Paigen diet feeding, the mice were weighed, plasma and hearts were harvested, heart weights measured and plasma levels of corticosterone and oxytocin were determined. The values presented are from the HypoE mice only; they do not include the companion (group housing) SR-BI^+/−ApoeR61^h/h mice. A Plasma corticosterone levels: group housed males, n = 11; singly housed males, n = 14; group housed females, n = 16; singly housed females, n = 12. B Plasma oxytocin levels: group housed males, n = 8; singly housed males, n = 8; group housed females, n = 12; singly housed females, n = 11. Data are means ± SD. Statistically significant differences were determined by unpaired t-test. C Correlation of oxytocin levels and heart-to-body weight ratios for all HypoE mice (n = 39). Gray and black circles indicate group and single housing, respectively. Statistics were evaluated using Spearman’s rank correlation.

Figure 7. Effects of shifting from Paigen to chow diet on survival (A) and plasma cholesterol (B).

HypoE mice were fed a normal chow diet) from weaning to 2 months of age (39 days). Beginning at two months of age the mice were housed singly (1/cage, both males and females were included) and some of the animals were continued to be fed the chow diet (panel B, white circles) whereas all of the others where switched from a normal chow diet to the Paigen diet (initiation of Paigen diet). The animals were maintained continuously throughout the rest of the experiment on the Paigen diet (“Cont.”, black lines) or after the indicated periods of Paigen diet feeding (10 days, red; 12 days, blue; 14 days, green) the mice were returned to a chow diet for the remainder of the experiment (colored arrows), as indicated in the schematic inset in panel A. A Kaplan-Meier survival curves. Log-rank test was performed to compare survival curves. B Time course of changes in plasma total cholesterol levels. Blood samples were taken serially at the indicated times (see Methods) from subgroups of animals different from those used for the survival curves in panel A (Cont.;n = 9, 10 days; n = 6, 12 days; n = 9, No Paigen(chow); n = 6). The data are presented as mean values of TC in each subgroup at the indicated times. In those cases in which animals died during the period of serial sampling, the data obtained from those mice while alive were incorporated into the calculation of the mean values.