Modes of Defining Atherosclerosis in Mouse Models: Relative
Merits and Evolving Standards
Alan Daugherty, Hong Lu, Deborah A. Howatt, and Debra L. Rateri
Mouse models have become the most common model for defining mechanisms of atherosclerotic disease.
Many genetic manipulations have enabled the development of atherosclerosis in mice due to either
endogenous or diet-induced hypercholesterolemia. This availability of lesion-susceptible mice has facili-
tated many studies using pharmacological and genetic approaches. Unfortunately, this expansive literature
on mouse atherosclerosis has generated many contradictions on the role of specific pathways. A contri-
butor to these inconsistencies may be the multiple modes in which atherosclerosis is evaluated. Also, for
each specific technique, there are no consistent standards applied to the measurements. This chapter will
discuss the imaging, biochemical, and compositional modes of evaluating atherosclerosis with suggestions
for standard execution of these techniques.
Key words: Atherosclerosis, mouse, imaging, cholesterol.
Atherosclerotic diseases, such as ischemic heart disease and stroke,
are the most common causes of morbidity and mortality in devel-
oped countries. There are some well-validated pharmacological
approaches for attenuating atherosclerotic diseases, of which the
most uniformly successful focus on dyslipidemias and the renin–
angiotensin system (1). However, there are still considerable
unmet needs for improved therapies in patients with atherosclero-
tic diseases. Therefore, the identification and validation of new
targets to prevent atherosclerosis remain a high priority. A com-
mon step in identifying and validating new therapeutic targets and
strategies involves the use of animal models of atherosclerosis. The
K. DiPetrillo (ed.), Cardiovascular Genomics, Methods in Molecular Biology 573,
DOI 10.1007/978-1-60761-247-6_1, ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009
most common species used in contemporary studies is mouse, in
which atherosclerosis can be induced by dietary and genetic
Human atherosclerosis is a chronic disease that progresses
over decades in a clinically silent process. The evolution of lesions
over these decades involves complex compositional changes in the
relative presence of different cell types. Atherosclerotic diseases
most frequently become a clinical imperative when lesions evolve
to structures that promote acute thrombosis resulting in restricted
blood flow to vital organs, such as the heart and brain. Many
mouse models of atherosclerosis recapitulate facets of the forma-
tive phase of human disease, although there is less agreement that
lesions in mouse models develop to the stage of generating ather-
osclerosis-associated thrombosis (3, 4).
The most common mode of assessing the clinical presence of
atherosclerosis is by procedures such as angiography and ultra-
sound (external and intravascular) in which the primary endpoint
atherosclerosis studies. While size continues to be an important
parameter for assessing atherosclerosis severity, it is becoming
increasingly clear that the common clinical manifestation of ather-
osclerosis occurs due to acute thrombosis, which is determined by
lesion composition (5). As a consequence, there has been an
increased emphasis on compositional analysis of lesions from
mouse models of atherosclerosis.
There is now a vast literature that has used mouse models to
determine atherogenic mechanisms and the effects of interven-
tions that promote or inhibit the development of disease. As can
be seen in Fig. 1.1, there was a small number of studies using
mouse models of atherosclerosis prior to the generation of
Number of Publications per Year
Fig. 1.1. The number of hits per year in PubMed using the search terms ‘‘mouse’’ and
2 Daugherty et al.
genetically engineered mice to enhance lesion susceptibility. Since
the availability of these mice, primarily apolipoprotein (apoE) ?/
? (6, 7) and low-density lipoprotein (LDL) receptor ?/? (8)
mice, in the early 1990s, there has been a precipitous increase in
publications. Unfortunately, this large literature is replete with
contradictory studies, particularly in studies using compound-
deficient mice in which a gene of interest is manipulated in an
atherosclerosis-susceptible background. For example, genetic
manipulations of class A scavenger receptors have shown this
protein to promote, retard, or have no effect on atherosclerosis
(9–12). There are also many examples in which the same genetic
manipulation has generated different results (13, 14).
There are several potential reasons for the contradictory litera-
ture. One is likely to be the different methods of quantifying
atherosclerosis. Table 1.1 provides selected references that have
detailed protocols for measuring mouse atherosclerosis. Even
within a specific method of lesion analysis, there are several
approaches in its execution that could lead to different results.
The rigor in the application of standards will differ between for-
ums. For example, the data needed to assist a pharmaceutical
company in deciding to advance a potential anti-atherosclerotic
compound to a human study may be far beyond the reasonable
scope of data needed to justify a more incremental advance that is
common in publications. The major emphasis of this chapter is to
provide criteria for critically evaluating results from mouse models
References that provide detailed protocols for evaluation
of mouse atherosclerosis experimental designs and
Experimental design Daugherty and Rateri (20)
Aortic root analysis Paigan et al. (15)
Purcell-Huynh et al. (18)
Daugherty and Whitman (16)
Baglione and Smith (17)
En face analysisTangirala et al. (19)
Daugherty and Whitman (16)
Biochemical analysisDaugherty and Whitman (16)
Compositional analysisLu et al. (26)
Modes of Defining Atherosclerosis in Mouse Models3
2.1. Ex Vivo Imaging
The most common mode of assessing lesion size in mice is by
removing aortas (see Note 1) and using imaging software to define
the size of lesions in either tissue sections of aortic roots or the
intimal surface of filleted aortas. Fortunately, both of these mea-
surements may be performed on the same mouse, although this
has not been common practice.
2.1.1. Aortic Root
The analysis of atherosclerosis in the aortic root was the first
described process of measuring lesions in mice and remains as
one of the most commonly used (15). The aortic root is the
site of the greatest proclivity to developing lesions in mice and
aortic root analysis is commonly the only method amenable to
studies in which profound hyperlipidemia was not achieved to
promote robust atherogenic responses, such as studies in
which modified diets are fed to C57BL/6 mice that are wild
type for genes such as LDL receptors and apoE. Detailed
protocols fortissue sectioning
described previously (16, 17). However, there are many issues
regarding the execution of these methods that may have pro-
found impact on the interpretation of results.
1. What region of the aortic root should lesions be measured?
a specific region (15). Lesion size was quantified in the aortic root
region that was proximal to the valve cusps where the aorta has a
roundedappearance. No lesionswere measured in the aorticsinus,
as defined by the presence of aortic valves. In many subsequent
publications measuring atherosclerosis in mice, it became com-
mon to include the aortic sinus in the measurement of athero-
sclerosis. An informal assessment of the literature shows that most
publications quantify lesions in the aortic sinus only.
Although the original description of measuring lesions in the
aortic root did not include the sinus, it is not clear whether there is
any adverse consequence of measuring atherosclerosis in different
regions of the aortic root. However, the size of lesions, as defined
by the area on tissue sections, may vary greatly along the length of
the aorta (18). Therefore, the selection of the region for acquiring
tissue sections will have a major impact on the measurements and
requires a reliable landmark aortic structure. The most common
the myocardium. In the description of the original method and in
our own experience, this tends to be a variable landmark. Instead,
the disappearance of the valve cusps provides a more reproducible
4 Daugherty et al.
To overcome the concern of obtaining aortic sections from
different regions, serial sections should be obtained throughout
lesions in aortic roots. The length of this can vary depending on
the severity of the disease. In addition to overcoming the potential
error of selecting different regions, this also permits the analysis of
whether lesions have changes in thickness or length.
Overall, tissue sectioning throughout lesions in the aortic root
overcomes potential errors due to sampling bias. It also provides
an indication of whether the lesions are expanding in thickness
and/or laterally. The most time-consuming task in aortic root
analysis is usually the mounting of the tissue and cutting until
the valves become visible. The visualization of the valves is the
indication that subsequent sections need to be acquired for ather-
osclerosis analysis. Relative to mounting and initial cutting, it is
not an onerous task to acquire sections throughout lesions and
measure their area.
2. How many tissue sections should be used to measure atherosclerosis?
As noted above, there may be considerable differences in
lesion areas in tissue sections from different aortic root regions.
The combination of this variance (see Note 2) and the imprecision
of defining the specific location for acquiring a tissue section high-
lights the importance of measuring lesion size in multiple tissue
sections. However, it isnot advisable to perform measurements on
a randomly selected number of tissue sections. Instead, it is advi-
sable to measure lesions in aortic root tissue sections that are
spaced 80–100 mm apart, with the number of sections used to
measure lesion size dictated by the distance they extend through-
out the aortic root.
3. What parameter should be used to define lesion ‘‘area’’?
Oil Red O staining of neutral lipids is frequently used to
facilitate identification of lesions. This staining is particularly help-
ful when there is a preponderance of small lesions adjacent to valve
cusps. In thisregion, the angle of the elastinfibers may hamperthe
of Oil Red O staining seldom covers the entire lesion area in a
section. The major reasons for this include the presence of non-
lipid-engorgedsmooth musclecells, extracellularmatrix,and non-
neutral lipids, such as unesterified cholesterol. Consequently, neu-
tral lipid staining routinely underestimates the size of lesions. Of
greater concern is that change in composition can affect neutral
lipid staining independent of size. Therefore, the analysis of lesion
size usually requires the manualmanipulation of image software to
outline lesions defined by the internal elastic lamina and the lumi-
nal boundary. In advanced lesions in mice, there have been a
being fragmented and atherosclerotic lesions extending into the
Modes of Defining Atherosclerosis in Mouse Models5
media. This scenario would call for some arbitrary decision on
defining the lesion boundary. However, encroachment of lesions
into the media is a relatively unusual phenomenon in mice.
4. Representation of lesion size
Image analysis software permits the use of calibrations to
generate the absolute area of atherosclerotic lesions. These are
represented either as mean lesion area per region or as a mean of
multipleregions.Lesionsize hasalsobeenrepresented asapercent
of the lumen area. However, since lumen area may change during
the atherosclerotic process, the representation of lesion areas in
absolute units is preferable.
2.1.2. En Face Analysis
of Aortic Intima
En face analysis of mouse atherosclerosis is a scaled-down version
of an approach that has been used in many large animal species.
Since the initial description of the en face analysis of lesions in
aortic intima, its use has gradually increased (19). A detailed pro-
tocol for the analysis has been described previously (16, 20).
Briefly,theaortaisdissectedfree,cuttoexpose theintimal surface,
pinned, and digital images are acquired. The analysis of these
images usually involves tracing the outline of the aorta to deter-
mine intimal area followed by determination of surface area of the
lesion. Data are expressed either as lesion area or as a percent of
1. Standardization of area of analysis
The original description of en face analysis of atherosclerosis in
the aorta emerges from the ventral aspect of the ventricle to the
iliac bifurcation (19). In this study, lesions were measured in mice
fed with high-fat diets for prolonged intervals that had extensive
lesions throughout the aorta. Although prolonged high-fat feed-
ing or protracted aging can lead to lesions throughout the aorta,
lesions are most prevalent in the arch in the earlier stages of the
disease (21). As a consequence, many studies only measure lesions
in restricted areas of the aortic intima. Therefore, the area of lesion
analysis must be strictly defined. Fortunately, unlike analysis of
lesions in the aortic root, the ability to reference landmarks to
identify and quantify the region of analysis is much easier in aortic
intima that has been opened and pinned.
However, there are facets of area for standardization that need
to be considered when analyzing lesions in the aortic arch. Com-
mon to all forms of en face analysis, there needs to be a clear
definition of the proximal and distal ends of the tissue. In our
own studies, the arch area is defined by the ascending aorta imme-
distal to the branch of the subclavian artery. There is also a need to
standardize the inclusion of the three arterial branches off the
6 Daugherty et al.
contains large lesions for which inconsistent inclusion may influ-
ence data. In our own studies, we analyze the most proximal 1 mm
of the innominate and left carotid arteries.
Another issueisstandardizingthedissection oftheaorticarch.
The mode used in the original description of Tangirala et al. (19)
made a longitudinal cut through the inner curvature of the aorta.
The second cut was a longitudinal cut through the outer curva-
ture, followed by mid-line cuts through the three branches off the
aorta. This mode of cutting the tissue permits it to be pinned flat
and avoids distortion of areas that can occur if the tissue is only cut
longitudinally through the inner curvature.
2. Use of neutral lipid stains
Lesions have been visualized and quantified by the en face
approach with or without lipid staining. The combination of the
transparency of mouse aortas and the opacity of lesions negate the
need to use stain to visualize lesions when they are sufficiently
advanced. However, neutral lipid stains assist in visualizing small
lesions, especially if they are in mice with more opaque aortas, such
as occurs during angiotensin II infusion (22). As mentioned for
analysis of aortic root, neutral lipid is not present throughout all
forms of atherosclerosis, so there could be a discrepancy between
area of stained and unstained lesions. Lipid staining also has the
will be intensely stained and may be confused with intimal staining of
lesions by inexperienced operators. Finally, lipid staining requires the
use of organic solvents, which negates the ability to perform tissue
sterol analysis as a complimentary approach to lesion size evaluation.
3. Multiple operators to analyze lesions
The definition of lesion boundaries may frequently be an
arbitrary decision, particularly in lesions that are small. Therefore,
it isadvisable to haveat leasttwo operators to determine the image
area of lesions with an adjudication process for the operators to
compare their data. It is best to do this while viewing the original
aortic tissue under a microscope, since camera images may gener-
ate artifacts, such as glare, that may be mistaken for a lesion.
4. Description of lesions to consider thickness
As noted above, atherosclerotic lesions are formed initially in
the aortic arch, with the inner curvature being a preferred site.
However, continued lesion formation may be disproportionately
represented by increased thickness compared to increased area.
Therefore, operators should consider apparent thickness during
their visual inspection of lesions. If the analysis shows no change in
secondary analysis for lesion quantification would be justified.
Modes of Defining Atherosclerosis in Mouse Models7
For most mouse studies, the measurement of aortic sterols
may be considered as a complimentary approach (discussed in
Analysis in Other Vascular
1. Innominate artery
The original detailed description of atherosclerotic lesions in
the innominate arteries of hypercholesterolemic mice demon-
strated more advanced compositional characteristics compared to
other regions (23). Like analysis in the aortic root, tissue section-
ing in this region has the advantage that it may be orientated so
that tissue sections may be acquired perpendicular to the plane of
the media. This feature reduces artifactual distortion of lesion size
that may be generated by cutting lesions in different planes. Rela-
tively few publications have measured lesion size in this region
(24). The challenges to using this region include difficulty in
establishing landmarks and unusually small length of lesions. The
latter concern may be overcome by acquiring and measuring a
sufficient number of sections along the length of lesions. Unlike
the extensive literature on lesions in aortic roots, there is insuffi-
cient literature to judge whether 80–100-mm interval for tissue
sections is appropriate for the innominate region.
2. Longitudinal sections of aortic arch
Although there are a few publications that have analyzed
lesions in longitudinal sections of the aortic arch, the primary
concern is the ability to obtain tissue sections that do not distort
the size of lesions due to the plane in which they are cut. For
example, in the aortic arch, only sections that are acquired at the
center of the arch will provide authentic lesion thickness and area
measurements. To satisfy this requirement, the tissue region being
cut would have to be flat. Also, only the sections in the mid-point
of the arch may be used for lesion measurements. Given the
difficulty of satisfying these criteria, other approaches to athero-
sclerotic lesion size analyses are preferred.
Although not used in many publications, measurement of aortic
sterol content has been one approach used to quantify mouse
atherosclerotic lesions (25). This approach is only valid for
lesions that are predominantly filled with unesterified and ester-
ified cholesterol, which is the case for most mouse atherosclero-
tic lesions. The detailed approach to the analysis has been
presented in a previous edition of Methods in Molecular Medi-
1. What is the optimal mode of normalizing the sterol content of
The main issue is the mode used to normalize the data. The
only factor that is likely to cause confusion is tissue wet weight.
Since the tissue is so small (only 10 mg if the total aorta is used),
8Daugherty et al.
the wet weight of the tissue may be influenced greatly by either
residual fluid or excessive removal of fluid. Although sterol con-
tent may be normalized to parameters such as protein and dry
weight, it is preferable to use intimal area of the identical aortic
areas that are used to measure lesion size by en face ex vivo
Compositional analysis in the context of this chapter considers
the use of histological and immunostaining techniques in tissue
sections of atherosclerosis. The potential issues of acquiring
immunostaining that is restricted to the development of chromo-
gen only at loci of authentic antibody–antigen interaction have
been discussed in a previous issue of Methods in Molecular
The determination of mouse atherosclerotic lesion compo-
sition is relatively new and is frequently not described in much
detail in publications. As an overview, it generally involves the
demarcation of a lesion as an area of interest followed by some
form of color segmentation analysis in which a threshold is set
for a specific color or hue to determine the relative area of
1. How many sections should be quantified?
As described for measurement of lesion area in the aortic root,
subject to sampling error. For example, in the case of a classic
fibrolipid lesion, compositional analysis would detect a relatively
high smooth muscle cell content at lesion edges with a progressive
pragmatic issue is the number of sections needed to provide
authentic compositional data that encompass the heterogeneity
of lesions versus the time needed to perform this analysis. No
studies have been published to provide an indication of the var-
iance of composition in mouse lesions, as performed in this region
to determine lesion size (18). In the absence of data, it is difficult
to suggest standards. However, it is clear that lesion composition
frequently differs along the length of a lesion, which necessitates
that compositional analysis be performed on multiple tissue sec-
2. How to define area of staining?
All histological and immunostaining techniques result in a
range of color intensities that usually require an arbitrary deci-
sion to determine the border of the staining. Since there is
seldom a situation in which a definite threshold exists for
defining staining for a specific component, it is advisable to
have at least two operators agree on the designation of the
Modes of Defining Atherosclerosis in Mouse Models9
3. What is the meaning of area of a specific cell defined?
Many cell markers are used to define the area of a specific cell
that occupies a lesion in a tissue section. However, this interpreta-
tion will only be accurate if the antigen being immunostained is
For example, CD68 is a lysosomal protein that is commonly used
to determine ‘‘macrophage area.’’ Immunostaining of very thin
tissue sections would restrict the positively reacted area to just this
intracellular organelle. However, much of the compositional ana-
lysis is performed on frozen tissues that are cut on a cryostat with a
thickness of 8–10 mm. Since macrophages are ?7 mm in diameter,
the two-dimensional view of the tissue section from the micro-
the entire cells. This impression would be enhanced by the use of
many chromogens that smear slightly beyond the region of pri-
mary antibody interactions. However, macrophages within ather-
osclerotic lesions commonly become lipid engorged and may
hypertrophy to 50 mm or more. This hypertrophy is the result of
the intracellular deposition of lipids in droplets. Therefore, much
of the area of a macrophage will be devoid of CD68. Conse-
quently, the area of CD68 immunostaining will differ greatly
from the actual macrophage area. Since it is unlikely that cell-
specific markers will be found that cover the entire area of specific
cell types, especially in hypertrophied states, the interpretation
should be restricted to the area of chromogen development by
the specific protein being detected rather than inferring an area
covered by a cell type.
4. Inferential interpretation of compositional analysis
Data from compositional analyses are frequently used as a
basis for using terms such as vulnerable and unstable in lesion
analysis. These terms are highly controversial (3, 4). Part of this
controversy stems from the disagreement on whether mouse
atherosclerotic lesions progress to the stage of plaque rupture.
generally refer to a specific tissue configuration of a macrophage
lipid-laden core encapsulated by smooth muscle cells and extra-
cellular matrix that exhibit thinning at the shoulder regions (27).
However, mouse compositional analysis infrequently considers
the spatial distribution of components. Instead, it more com-
monly determines a ratio of staining of components such as
neutral lipid, extracellular matrix components, macrophages,
and smooth muscle cells, without regard to spatial arrangements.
In the absence of a widely reproducible mouse atherosclerosis
model that exhibits plaque rupture, it would be advisable to
restrict the interpretation to a statement of the data rather than
to include inferential suggestions of plaque vulnerability based
on the data.
10 Daugherty et al.
3. Issues for
The preference for each study to require multiple modes of quan-
tifying atherosclerosis with compositional analysis in multiple vas-
cular beds has to be balanced by the pragmatic aspects of time
consumed to perform these techniques in a high-quality manner.
This section provides some suggestions for guidelines to provide
practical restrictions for evaluating atherosclerosis in mice. The
following suggestions are based on our experiences in designing
1. Should more than one vascular bed be measured by ex vivo
The most critical human diseases that arise from atherosclero-
sis are ischemic heart disease and stroke, due to atherosclerosis in
the coronary and carotid arteries, respectively. It is not clear
whether atherosclerotic lesions develop by identical processes in
these two regions. Although there is some controversy (3, 4),
development of atherosclerosis in mice does not generally lead to
disease consequences that are similar to humans. Certainly, the
regions in which atherosclerosis is usually quantified in mouse
studies (aortic root, aortic intimal, and brachiocephalic artery) do
not have direct correlations to the consequences of the disease in
Given the potential for atherosclerotic lesion development by
differing mechanisms in different vascular beds, a uniform change
in lesion sizes in multiple regions in response to a specific manip-
ulation provides an indication of the potential widespread applic-
ability of the manipulation. However, as described at the outset,
atherosclerosis studies are time-consuming in the generation and
requires measurement of atherosclerosis in more than one vascular
bed. Rather, a rigorously performed study using a single vascular
bed should provide sufficient information to provide insight into
2. Should ex vivo imaging be used in conjunction with another
parameter to indicate atherosclerotic lesion size?
En face analysis of lesions in the aortic intima is a common
mode of analysis for determining the area of atherosclerotic
lesions. A major deficiency of this approach is that it provides
a two-dimensional measurement of a three-dimensional athero-
sclerotic lesion. Therefore, a potentially critical element of
lesion thickness or volume is missed in this assay. If an inter-
vention promotes a large difference in the area of lesions, as
measured by en face analysis, it is hard to conceive a scenario in
Modes of Defining Atherosclerosis in Mouse Models11
which the volume of the lesions would not also be greatly
reduced. However, in the case of lesion areas being the same,
but visual observation indicating a difference in lesion thickness,
some measurement of lesion thickness or volume is warranted.
Lesion thickness may potentially be defined on histological
sections. While this approach may provide authentic informa-
tion, it is also prone to the problems of sampling error as
described for the aortic root. An alternative approach is the
use of aortic sterol content. This has been used in past studies
in which en face lesion measurements failed to discriminate
between groups, but differences were readily apparent using
sterol analysis (28). There are a few studies in which ex vivo
image analysis demonstrated no changes, but visual inspection
indicated thickness and volume changes that would warrant a
combination of en face analysis and tissue sterol content.
3. Should lesion size measurements be combined with compositional
The answer to this question largely relies on the purpose of
the study. If the emphasis of a study is that the mechanism
reduces only size, then the need to perform compositional
analysis is reduced. Compositional analysis is frequently applied
to define a cause of changes in the dimensions of lesions.
Unfortunately, it is not possible to define whether size changes
are the cause or consequence of compositional changes. There-
fore, unless the specific conclusion of a study requires the
determination of lesion composition, it seems unreasonable
for this analysis to be performed without justification. Based
on these considerations, we provide Table 1.2 as suggestions
for standards to evaluate studies describing effects on athero-
Summary of suggested standards for mouse atherosclerosis analysis
Aortic root1. Section up to 1,000 mm of aortic root depending on lesion
2. Measure lesion area on sections spaced 80–100 mm
throughout lesions in root.
defined by internal elastin lamina and lumen.
4. Represent data as absolute areas.
5. At least two operators agree on lesion designation in images.
1. Standardize intimal area to the same area used for ex vivo
2. At least two operators agree on lesion designation in images.
1. Perform appropriate normalization process.
12Daugherty et al.
1. Atherosclerotic lesions in the intima of arteries in mice are
relatively easily dislodged from the media. This fragility of
attachment is enhanced in mice that have undergone
irradiation and repopulation with bone marrow-derived
stem cells to determine atherogenic mechanisms (29).
The analytical considerations described in this chapter
are prejudiced on the assumption that lesions are not
lost during processing.
2. Despite the uniformity of genetic background and environ-
ment in mouse atherosclerosis studies, it is common that
there is a large variance in these studies. Consequently, there
needs to be sufficient number of mice to obtain statistically
robust results. Also, the variance and distribution of data are
often not compatible with parametric analysis. Therefore,
data frequently need to be analyzed using non-parametric
tests that are usually less sensitive for detecting a statistical
difference between groups.
The authors’ laboratories are supported by the National Institutes
of Health (HL08100 and HL62846).
Table 1.2 (continued)
1. Multiple sections need to be analyzed to account for
potential lesion heterogeneity.
2. At least two operators perform color segmentation analysis.
3. Strict interpretation of stained area to specific entity.
Analysis of multiple vascular beds 1. Preferable for defining the uniformity of the response, but
Lesion size and composition1. The combination is only warranted if required to
substantiate the conclusions of the study.
2. If performed, given the potential for difference effects of
lesion size in different vascular beds, composition analysis
needs to be performed in the same region as size
Modes of Defining Atherosclerosis in Mouse Models 13
1. Rader, DJ, Daugherty, A. (2008) Translat-
ing molecular discoveries into new therapies
for atherosclerosis. Nature 451, 904–913.
2. Daugherty, A. (2002) Mouse models of
atherosclerosis. Am J Med Sci 323, 3–10.
3. Schwartz, SM, Galis, ZS, Rosenfeld, ME,
et al. (2007) Plaque rupture in humans and
mice. Arterioscler Thromb Vasc Biol 27,
4. Jackson, CL, Bennett, MR, Biessen, EA,
et al. (2007) Assessment of unstable athero-
sclerosis in mice. Arterioscler Thromb Vasc
Biol 27, 714–720.
5. Falk, E. (1999) Stable versus unstable ather-
osclerosis: clinical aspects. Am Heart J 138,
6. Piedrahita, JA, Zhang, SH, Hagaman, JR,
et al. (1992) Generation of mice carrying a
mutant apolipoprotein-E gene inactivated
by gene targeting in embryonic stem cells.
Proc Natl Acad Sci USA 89, 4471–4475.
7. Plump, AS, Smith, JD, Hayek, T, et al.
(1992) Severe hypercholesterolemia and
atherosclerosis in apolipoprotein-E-deficient
mice created by homologous recombination
in ES cells. Cell 71, 343–353.
8. Ishibashi, S, Goldstein, JL, Brown, MS,
et al. (1994) Massive xanthomatosis and
atherosclerosis in cholesterol-fed low den-
sity lipoprotein receptor-negative mice. J
Clin Invest 93, 1885–1893.
9. Suzuki, H, Kurihara, Y, Takeya, M,
et al. (1997) A role for macrophage
scavenger receptors in atherosclerosis
and susceptibility to infection. Nature
(2002) Macrophage-specific expression of
class A scavenger receptors in LDL recep-
tor(–/–) mice decreases atherosclerosis and
changes spleen morphology. J Lipid Res 43,
11. Herijgers, N, de Winther, MP, Van Eck, M,
et al. (2000) Effect of human scavenger
receptor classAoverexpression inbonemar-
row-derived cells onlipoproteinmetabolism
and atherosclerosis in low density lipopro-
tein receptor knockout mice. J Lipid Res 41,
12. Daugherty, A, Whitman, SC, Block, AE,
et al. (2000) Polymorphism of class A sca-
venger receptors in C57BL/6 mice. J Lipid
Res 41, 1568–1577.
13. Witztum, JL. (2005) You are right too!
J Clin Invest 115, 2072–2075.
14. Curtiss, LK. (2006) Is two out of three
enough for ABCG1? Arterioscler Thromb
Vasc Biol 26, 2175–2177.
15. Paigen, B, Morrow, A, Holmes, P, et al.
(1987) Quantitative assessment of athero-
sclerotic lesions in mice. Atherosclerosis 68,
16. Daugherty, A, Whitman, SC. (2003) Quan-
tification of atherosclerosis in mice. Methods
Mol Biol 209, 293–309.
17. Baglione, J, Smith, JD. (2006) Quantitative
assay for mouse atherosclerosis in the aortic
root. Methods Mol Med 129, 83–95.
18. Purcell-Huynh, DA, Farese, RV, Johnson,
DF, et al. (1995) Transgenic mice expres-
sing high levels of human apolipoprotein B
develop severe atherosclerotic lesions in
response to a high-fat diet. J Clin Invest
19. Tangirala, RK, Rubin, EM, Palinski, W.
(1995) 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 36, 2320–2328.
20. Daugherty, A, Rateri, DL. (2005) Develop-
ment of experimental designs for athero-
sclerosis studies in mice. Methods 36,
21. Nakashima, Y,Plump,AS,Raines,EW,etal.
(1994) ApoE-deficient mice develop lesions
of all phases of atherosclerosis throughout
the arterial tree. Arterioscler Thromb 14,
22. Daugherty, A, Manning, MW, Cassis, LA.
(2000) Angiotensin II promotes athero-
sclerotic lesions and aneurysms in apolipo-
protein E-deficient mice. J Clin Invest 105,
23. Rosenfeld, ME, Polinsky, P, Virmani, R,
et al. (2000) Advanced atherosclerotic
lesions in the innominate artery of the
ApoE knockout mouse. Arterioscler Thromb
Vasc Biol 20, 2587–2592.
24. Reardon, CA, Blachowicz, L, Lukens, J,
et al. (2003) Genetic background selec-
tively influences innominate artery athero-
sclerosis – Immune system deficiency as a
probe. Arterioscler Thromb Vasc Biol 23,
25. Daugherty, A, Pure, E, Delfel-Butteiger, D,
et al. (1997) The effects of total lymphocyte
deficiency on the extent of atherosclerosis in
14Daugherty et al.
apolipoprotein E –/– mice. J Clin Invest Download full-text
26. Lu, H, Rateri, DL, Daugherty, A. (2007)
Immunostaining of mouse atherosclerotic
lesions. Methods Mol Med 139, 77–94.
27. Falk, E. (2006) Pathogenesis of athero-
sclerosis. J Am Coll Cardiol 47, C7–C12.
28. Daugherty, A, Zweifel, BS, Schonfeld, G.
(1991) The effects of probucol on the pro-
heritable hyperlipidaemic rabbits. Br J Phar-
macol 103, 1013–1018.
29. Fazio, S, Linton, MF. (1996) Murine bone
marrow transplantation as a novel approach
to studying the role of macrophages in lipo-
Trends Cardiovasc Med 6, 58–65.
30. Whitman, SC. (2004) A practical approach
to using mice in atherosclerosis research.
Clin Biochem Rev 25, 81–93.
Modes of Defining Atherosclerosis in Mouse Models15