Cardiac and vascular phenotypes in the apolipoprotein E-deficient mouse.

Abstract

Cardiovascular death is frequently associated with atherosclerosis, a chronic multifactorial disease and a leading cause of death worldwide. Genetically engineered mouse models have proven useful for the study of the mechanisms underlying cardiovascular diseases. The apolipoprotein E-deficient mouse has been the most widely used animal model of atherosclerosis because it rapidly develops severe hypercholesterolemia and spontaneous atherosclerotic lesions similar to those observed in humans. In this review, we provide an overview of the cardiac and vascular phenotypes and discuss the interplay among nitric oxide, reactive oxygen species, aging and diet in the impairment of cardiovascular function in this mouse model.

Full text

Page 1

REVIEWOpen Access
Cardiac and vascular phenotypes in the
apolipoprotein E-deficient mouse
Elisardo C Vasquez1,2*, Veronica A Peotta3, Agata L Gava1,4, Thiago MC Pereira5and Silvana S Meyrelles1
Abstract
Cardiovascular death is frequently associated with atherosclerosis, a chronic multifactorial disease and a leading
cause of death worldwide. Genetically engineered mouse models have proven useful for the study of the
mechanisms underlying cardiovascular diseases. The apolipoprotein E-deficient mouse has been the most widely
used animal model of atherosclerosis because it rapidly develops severe hypercholesterolemia and spontaneous
atherosclerotic lesions similar to those observed in humans. In this review, we provide an overview of the cardiac
and vascular phenotypes and discuss the interplay among nitric oxide, reactive oxygen species, aging and diet in
the impairment of cardiovascular function in this mouse model.
Keywords: Hypercholesterolemia, atherosclerosis, mouse, cardiac function, vascular senescence
Introduction
Over the last two decades, the mouse has become a very
important experimental animal in studies of physiology
and a growing number of pathophysiologies due to its
easy breeding, short generation time and the availability
of inbred strains. The limitations of the small size of the
mouse for studying cardiac and vascular hemodynamics
and function have been overcome in our and other
laboratories due to advances in surgical techniques, as in
Borst et al. [1], who described detailed procedures for the
analysis of cardiac functional parameters in the mouse.
As a consequence of the progressive advancement of
molecular biology techniques, it is possible to knockout
and restore endogenous genes or add exogenous genes
into the mouse, allowing the development of mouse
models for human diseases. Two decades ago, the first
gene-targeted murine model of atherosclerosis was cre-
ated by the inactivation of the apolipoprotein E (apoE)
gene by homologous recombination [2,3]. Among the
genetically engineered models, the apoE-deficient
(apoE-/-) mouse is considered to be one of the most
relevant models because it develops spontaneous
hypercholesterolemia and arterial lesions similar to
those observed in humans.
In the apoE-/-mouse, most of the plasma cholesterol
(PC) is found in the atherogenic lipoprotein fractions,
namely the very-low-density lipoprotein (VLDL), inter-
mediate-density lipoprotein (IDL) and low-density lipo-
protein (LDL) fractions, and this profile is aggravated by
a Western-type diet. On a chow diet, apoE-/-mice exhi-
bit increased total PC (~8-fold), triglycerides (1.7-fold),
VLDL+IDL (18-fold) and LDL (14-fold) compared to
C57BL/6J (C57) mice. When fed a Western-type diet, a
dramatic increase in the proportions of these lipids is
observed in total PC (~14-fold), particularly in the
VLDL+IDL lipoprotein fraction (~30-fold) [2,4-10].
The combination of the availability of mouse models
of atherosclerosis and the technology to perform hemo-
dynamic measurements in mice has enabled the study of
the effects of hypercholesterolemia and/or atherosclero-
sis on cardiac vascular function. In this review, we focus
on studies from our laboratory and from other investi-
gators regarding the vascular and cardiac phenotypes in
the apoE-/-mouse. We also discuss the underlying
mechanisms that have been shown to contribute to car-
diac and vascular dysfunction in this mouse model.
Vascular phenotypes in apoe-/-mice
Initiation and progression of atherosclerosis
The tendency to consider atherosclerosis as the effect of
dyslipidemia or inflammation alone has been replaced
by a new concept that this disease results from lipid
* Correspondence: evasquez@pq.cnpq.br
1Department of Physiological Sciences, Health Sciences Center, Federal
University of Espirito Santo, Vitoria, ES, Brazil
Full list of author information is available at the end of the article
Vasquez et al. Journal of Biomedical Science 2012, 19:22
http://www.jbiomedsci.com/content/19/1/22
© 2012 Vasquez et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.

Page 2

disorders, enhanced oxidative stress and inflammation
[11-18]. The development of atherosclerosis in condi-
tions of hypercholesterolemia is initiated by the oxida-
tion of lipoproteins such as LDL (oxLDL) in the
subendothelial space and dysfunction of endothelial
cells. In parallel endothelial cells express vascular and
intercellular adhesion molecules (VCAM-1 and ICAM-
1). Consequently, monocytes move into the subendothe-
lial space where they undergo differentiation into
macrophages under the influence of stimulating factors.
These cells express scavenger receptors that recognize
oxLDL and promote their phagocytosis. The degradation
of oxLDL in lysosomes is slow, facilitating the formation
of foam cells. The pivotal stage of atherogenesis seems
to be antigen presentation by macrophages to T lym-
phocytes that recognize oxLDL and locally release
proinflammatory cytokines such as interleukins (IL-1, 6,
12 and 18) and chemokines such as tumor necrosis fac-
tor alpha (TNFa) and monocyte chemoattractant pro-
tein-1 (MCP-1), thereby enhancing the inflammatory
cascades. The accumulation and transformation of
macrophages into large foam cells form fatty streaks
that are characteristics of early atherosclerosis, which
progresses into mature atherosclerotic plaques (fibrous
caps) in the intima.
The mechanisms and progression of atherosclerosis in
the apoE-/-mice have been described before [19-22]. In
this model, atherosclerotic lesions are observed through-
out the macrovasculature, with the most prevalent sites
located in the aortic root, aortic arch, common carotid,
superior mesenteric artery and renal and pulmonary
arteries. Histopathological studies of the progression of
the lesions in the apoE-/-mouse fed a chow diet
revealed that the first signs of lesions appear at 6-8
weeks of age, as indicated by the attachment of mono-
cytes to the endothelial cells and a disruption of the
subendothelial elastic lamina [21,22]. Lesions containing
foam cells and smooth muscle cells are observed at 8-10
weeks and fibrous plaques appear at 15-20 weeks
[21,22]. The acceleration and severity of atherosclerotic
lesions at all stages can be induced by a Western-type
diet.
Hemodynamics: endothelial dysfunction
Recently, we reviewed the potential mechanisms under-
lying endothelial dysfunction in the apoE-/-mouse and
how endothelial dysfunction is influenced by aging, gen-
der and diet [23]. In brief, the conducting vessels of
apoE-/-mice show an impaired endothelial nitric oxide
(NO)-dependent relaxation response to acetylcholine
(ACh), which is associated with plaque formation
[6,24,25] and aggravated by a Western-type diet
[10,26-28] and by aging [24,29]. As illustrated in Figure
1, the main potential mechanism underlying the
endothelial dysfunction of apoE-/-mice may be the
endothelial NO synthase (eNOS) pathway, due to the
uncoupling of eNOS in the endothelium. This uncou-
pling decreases the bioavailability of NO, increases pro-
duction of superoxide anion (•O2-) and generates
peroxynitrite (•ONOO-), which is associated with the
activationof the endothelin-1
[6,28,30-33]. In resistance vessels, which rarely develop
atherosclerotic lesions, impaired endothelial NO-depen-
dent relaxation responses to ACh have also been
observed, primarily in female apoE-/-mice [34,35]. This
effect appears to be mediated by hydrogen peroxide
(H2O2) or an endothelium-derived hyperpolarizing fac-
tor (EDHF)-like principle agent, increased NADPH oxi-
dase-derived •O2-and ET-1.
Local oxidation of circulating lipoproteins and their
incorporation within the vascular endothelium, asso-
ciated with oxidative processes, plays a pivotal role in
the pathogenesis of endothelial dysfunction in the
apoE-/-mouse [6,10]. The endothelial dysfunction pro-
motes the entry and retention of LDL particles in the
artery wall [36,37] enhancing the progression of lesions
and creating a vicious cycle. This idea is corroborated
by the finding that in vitro treatment of aortas with
oxLDL mimicked the endothelial NO dysfunction that
was observed in apoE-/-mice [10]. The finding that gene
transfer of human paraoxonase 1, which destroys lipid
peroxides, into apoE-/-mice decreased the oxLDL con-
tent of the plaques and restored endothelial function
[29] corroborates this point of view. Others have shown
that treatment of apoE-/-mice with large empty phos-
pholipid vesicles, which accelerates the reverse pathway
of lipid transport from peripheral tissues to the liver,
restored endothelium-dependent relaxation, leukocyte
adherence, and endothelial expression of VCAM-1 to
normal or nearly normal levels [38].
Hemodynamics: arterial blood pressure
Genetically engineered mouse models have been used
to investigate hemodynamics in several cardiovascular
diseases, including hypercholesterolemia and athero-
sclerosis. Several studies, in which blood pressure (BP)
was recorded acutely, have shown that BP levels in
young-adult (up to 30-week-old) hypercholesterolemic
apoE-/-mice fed a standard chow diet or a Western-
type diet are similar to those of control mice
[7,31,32,39-45]. Conversely, Yang et al. [46], observed
normal resting values of BP in young apoE-/-mice but
high levels in adult apoE-/-mice, compared with age-
matched control mice. However, continuous and pro-
longed recordings of BP have revealed that apoE-/-
mice exhibit elevated mean 24-hour BP levels, total
abolition of BP circadian cycles and increased BP
variability in the very-low-frequency band, which can
be improved by the restoration of vagal and NOS-
mediated regulation with statins [47].
(ET-1)system
Vasquez et al. Journal of Biomedical Science 2012, 19:22
http://www.jbiomedsci.com/content/19/1/22
Page 2 of 9

Page 3

The contribution of high BP to the development of
atherosclerosis, and vice-versa, is being actively debated
because data involving concurrent measurements of BP
and atherosclerosis in mouse models have revealed an
association in some studies and a lack of association in
others between hypertension and atherosclerosis [48].
One of the first studies showed that subcutaneous infu-
sions of norepinephrine induced increases in BP and in
atherosclerotic lesion size in apoE-/-mice [33]. Consid-
ering that angiotensin II (Ang II) accelerates the devel-
opment of atherosclerosis in both LDL receptor-/-and
apoE-/-mice [33,49,50] and that hypercholesterolemia
has been associated with activation of the renin-angio-
tensin system [51,52], the apoE-/-mouse became an
exciting model for evaluating the hemodynamic effects
of Ang II in hypercholesterolemic mice (see a hypotheti-
cal mechanism in Figure 1). Currently, the suggested
mechanisms by which Ang II promotes atherosclerosis
include increased oxidative stress, the production of
monocyte chemoattractants, and the diapedesis of
monocytes [33]. Previous studies of chronic infusions of
Ang II in apoE-/-mice have led to three conclusions: (a)
Ang II-induced atherosclerosis was not associated with
an increase in BP [53]; (b) Ang II-induced hypertension
was not the direct cause of the profound increase in
atherosclerotic lesions [33] because this increase in BP
was similar in magnitude to that observed in wild-type
control mice [54]; and (c) low doses of Ang II that did
not affect BP increased aortic atherosclerotic lesions
[55,56]. Likewise, it has been shown that pharmacologi-
cal blockade of the Ang II AT1receptor in apoE-/-mice
or genetic disruption of the AT1receptor (apoE-/-/
AT1-/-mice) reduces the inflammatory and athero-
sclerotic lesion process irrespective of BP [32,57]. Addi-
tionally, long-term Ang (1-7) treatment has been shown
to produce vasoprotective and atheroprotective actions
in the apoE-/-mouse without direct effects on BP [45].
Conversely, it has been shown that AT1 receptor
antagonist treatment lowered BP but had no effect on
atherosclerosis [58]. Thus, it appears that the actions of
Ang peptides on the pathogenesis of atherosclerosis and
on BP occur through independent mechanisms. In our
and other laboratories, this issue has been addressed by
activating the renin-angiotensin system through partial
clipping (stenosis) of the left renal artery (known as two
kidney-one clip, or 2K1C, hypertension) in young
Vascular dysfunction
(reduced vasodilation)
NO
availability
Uncoupled
eNOS
? ? ADMA
Renin-angiotensin system
? ?
NADPH
oxidase
PGI2
synthase
? ? BH4
eNOS
activity
? ? •ONOO?
? ? •O2?
? ? Antioxidant gene expression
Aging
Proinflammatory gene expression
Oxidative
damage
to DNA
Vascular
Senescence
PARP
activation
ROS
Hypercholesterolemia
Atherosclerosis
Hypertension
? ?NO
Figure 1 A hypothetical scheme illustrating the role of ROS and decreased NO bioavailability in vascular dysfunction in conditions of
hypercholesterolemia and atherosclerosis in the apoE-/-mouse. Abbreviations: ADMA, asymmetric dimethyl-L-arginine; BH4,
tetrahydrobiopterin; eNOS, endothelial nitric oxide synthase; NADPH, nicotinamide adenine dinucleotide phosphate; NO, nitric oxide; •O2-,
superoxide anion; •ONOO-, peroxynitrite; PARP, poly-ADP-ribose polymerase; PGI2, prostacyclin; ROS, reactive oxygen species.
Vasquez et al. Journal of Biomedical Science 2012, 19:22
http://www.jbiomedsci.com/content/19/1/22
Page 3 of 9

Page 4

apoE-/-mice. In animals with concurrent hypercholes-
terolemia and activation of the endogenous renin-angio-
tensin system, the level of arterial hypertension was
similar to that observed in normocholesterolemic wild-
type C57 mice with high Ang II [41,42]. However, an
increase in the thickness of the aortic wall is noted to
some extent in the hypertensive C57 mouse and espe-
cially in the hypercholesterolemic apoE-/-mouse, and
the concurrence of both factors (hypercholesterolemia
and hypertension) is observed to exert an additive effect
on the aortic wall thickness in the apoE-/-mouse [42].
Therefore, it appears that the development of arterial
hypertension in atherosclerotic apoE-/-mice is asso-
ciated with other concurrent factors, particularly aging,
imbalance in NOS/reactive oxygen species (ROS) pro-
duction and Ang II (see schematic illustration in Figure
1).
Vascular remodeling
The vessel wall of a plaque-prone region expands to
protect the lumen from the expanding plaque (positive
remodeling) until the plaque grows so extensively that
the outer lumen can no longer compensate and the pla-
que begins to encroach upon the lumen (negative remo-
deling) [59,60]. In apoE-/-mice, positive remodeling has
been reported in aortas at both early and late stages of
atherosclerosis [42,59,61-63]. However, this observation
may not apply for all large vessels as, in carotid arteries,
positive remodeling has been reported at the early stages
[60], but negative remodeling has been reported in aged
animals [59,60]. A possible explanation for this discre-
pancy is that an excessive accumulation of lipids along
the internal elastic lamina and within the media results
in the death of medial cells and loss of the normal capa-
city to remodel in the carotid artery [59]. An additional
and reasonable factor influencing or determining this
difference between the carotid artery and the aorta in
aged animals is that the animals were fed a regular diet
in the study reporting positive remodeling [60], whereas
the animals were fed a Western-type diet in the study
reporting negative remodeling [59]. In addition to the
lack of compensatory remodeling in the carotid arteries
[59], we have observed that aged apoE-/-mice show an
inward (negative) remodeling in the coronary arteries
(unpublished data).
Interplay among lipid and lipoprotein oxidation, DNA
damage and inflammation
Oxidative stress and inflammation are intimately linked
to the evolution of atherosclerosis. Oxidative stress ori-
ginated primarily from ROS in mitochondria can
damage cells or components of cells by initiating chemi-
cal chain reactions, such as lipid peroxidation, leading to
the formation of oxLDL and oxidizing proteins or DNA
[64,65]. LDL is oxidized in a slow process upon contact
with ROS in the sub-intimal space by all major cells of
the arterial wall via different mechanisms. ROS induce
the oxidation of cholesterol and polyunsaturated fatty
acids in LDL. Consequently, lipid peroxidation results in
the formation of several products, such as oxysterols
and aldehydes, which react with lysine residues of apoli-
poprotein B (apoB), generating oxLDL [14]. In addition
to their metabolic deviation, oxLDL exhibits a large vari-
ety of biological and atherogenic properties involved in
the activation of inflammatory, mitogenic or proapopto-
tic pathways, contributing to the progression of athero-
sclerosis [14,66].
Another cellular target of ROS is DNA; oxidative
DNA lesions have been described ranging from base
modifications to single- and double-strand breaks [67].
Consequently, excessive oxidative stress-induced DNA
damage, which was originally described as the limited
ability of a cell to divide, also appears to contribute to
the pathogenesis of age-associated vascular disorders
[65,68]. Recent studies in our laboratory [61] revealed
vascular senescence in the aortas of atherosclerotic aged
apoE-/-mice but not in non-atherosclerotic aged wild-
type C57 mice. This result suggests that in the aorta the
occurrence of vascular senescence in aging may be
related to the development of vascular disease, and this
phenomenon may be the cause of the endothelial dys-
function observed in this murine model. Based on the
observation that the induction of senescence in human
aortic endothelial cells by the inhibition of telomere
function results in decreased eNOS activity [69,70], stu-
dies with the atherosclerotic apoE-/-mouse may provide
an opportunity to evaluate and understand the interac-
tion of vascular endothelial dysfunction and vascular
senescence. We postulated that the renin-angiotensin
system plays a dual role, contributing additively with
ROS to DNA damage and additively with aging to arter-
ial hypertension.
Oxidative damage to DNA in atherosclerosis also
modifies other biochemical and regulatory pathways
such as the overexpression of poly(ADP-ribose) poly-
merase (PARP-1) [71,72]. PARP-1 is a nuclear enzyme
that recognizes and is activated by DNA strand breaks
and is thus involved in DNA damage response pathways;
simultaneously, PARP-1 may induce cell dysfunction or
necrosis by depleting cellular energy pools [73]. Pacher
et al. [74] reported that the activation of PARP-1 is
associated with hypertension and aging but not with
atherosclerosis. In contrast, other investigators [71,75]
have shown that functional alterations in the endothe-
lium of the apoE-/-mouse are dependent on the activa-
tion of PARP-1 in endothelial cells. This activation leads
to proinflammatory gene expression and thus contri-
butes to the vascular senescence and endothelial dys-
function that are observed in atherosclerosis. Moreover,
pharmacological blockade and genetic manipulations
Vasquez et al. Journal of Biomedical Science 2012, 19:22
http://www.jbiomedsci.com/content/19/1/22
Page 4 of 9

Page 5

have demonstrated that endogenous PARP-1 is required
for atherogenesis in the apoE-/-mouse and functions by
increasing the expression of adhesion molecules upon
endothelial activation, enhancing inflammation, and
inducing features of plaque vulnerability. Thus, inhibi-
tion of PARP1 may represent a promising therapeutic
target in atherosclerosis [73]. Figure 1 illustrates the
interplay among ROS, oxidative damage to DNA, the
renin-angiotensin system, vascular senescence and ather-
osclerosis in the apoE-/-mouse.
Atherosclerosis is a chronic disease, and whether it is
initiated and/or aggravated by dyslipidemia depends on
the contribution of oxidative stress and proinflammatory
mediators. Thus, the impact of these contributors on
the development of atherosclerotic lesions may reveal
new treatment options, independent of serum lipids
reduction. Recent studies in apoE-/-mice have demon-
strated a reduction in atherosclerosis development using
an immunodeficiency strategy [20], low dose of the phy-
tochemical tetrahydrocannabinol [76], antioxidant N-
Acetylcysteine [77], PARP inhibition [73] or mononuc-
lear cell therapy [63].
In relation to cell therapy, there is consistent evidence
[78] that the atherosclerotic process initiated by
endothelial death results in the subsequent replacement
of the affected cells by endothelial progenitor cells; in
addition, the evidence indicates that the cellular repair
of ongoing vascular injury is mediated by progenitor
cells. Indeed, treatment with spleen-derived mononuc-
lear cells increases vascular NOS activity and restores
endothelium-dependent relaxation in the aortas of
apoE-/-mice [79]. Moreover, mononuclear cell therapy
in apoE-/-mice results in the homing of endothelial pro-
genitor cells, a decrease in oxidative stress and an upre-
gulation of eNOS protein expression [63], which
indicates that cell therapy is a promising tool for the
restoration of endothelial function and the prevention of
atherosclerosis development, independent of a reduction
in serum lipids.
Cardiac phenotypes in apoe-/-mice
Resting HR
The mouse has a high resting heart rate (HR), which is
related to its high metabolism. Hemodynamic measure-
ments have shown that the resting HR in the conscious
wild-type C57 mouse ranges from 520 to 650 bpm
[42,80,81]. Similar values have been observed with acute
measurements in conscious apoE-/-mice [32,41,42,
46,80,82,83]. Interestingly, some studies have shown that
HR is preserved even when the BP is high in apoE-/-
mice fed either a standard chow diet [46] or a Western-
type diet [84]. However, others have evaluated HR by
continuous recording over 24 hours and observed a sig-
nificantly increased mean HR and a total abolition of its
circadian cycles in the apoE-/-mouse compared with
wild-type control mice [47]. Moreover, apoE-/-mice
exhibited decreased HR variability associated with
abnormal neurohumoral control of HR and a defective
parasympathetic drive to the heart, which were further
exacerbated under a high-fat diet regimen [47,71]. It has
also been observed that chronic treatment of apoE-/-
mice with a statin reduced their HRs to control levels
and restored their circadian variations, despite persistent
elevations in PC levels [47]. Although further studies are
needed to clarify the mechanisms involved in the abnor-
mal cardiac rhythm in the apoE-/-mouse, there is evi-
dence that the imbalance in NO/•O2-production may
play a pivotal role [47,71].
Cardiac hypertrophy
Cardiac hypertrophy is also an important parameter to
evaluate because it can be related to cardiac dysfunction
(see the illustration in Figure 2). In studies with apoE-/-
mice in our laboratory, we regularly evaluate cardiac
hypertrophy, which can be determined by measuring the
cardiac weight-to-body weight ratio [42,81]. In apoE-/-
Increased aortic stiffness
Hypertension
Cardiac dysfunction
Young
Aged
Aortic valve damage
Atherosclerosis
Aging
Cardiac
Hypertrophy
Normal
Cardiac
Hypertrophy
Cardiac Dysautonomia
Hypercholesterolemia
Atherosclerosis
Western
diet
Regular
diet
Figure 2 A schematic illustration describing cardiac
phenotypes and hypothetical factors that influence cardiac
dysfunction in the apoE-/-mouse. Differences in line thickness
indicate the relative contribution of that factor.
Vasquez et al. Journal of Biomedical Science 2012, 19:22
http://www.jbiomedsci.com/content/19/1/22
Page 5 of 9