313
Characterization of Sites With Elevated LDL
Permeability at Intercostal, Celiac, and Iliac
Branches of the Normal Rabbit Aorta
Robert A. Herrmann, Richard A. Malinauskas, George A. Truskey
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Abstract En face autoradiography of the endothelium was
used to quantify the distribution, area, and permeability of
sites with enhanced permeability to !25I-low-density lipoprotein (125I-LDL) around the intercostal and celiac arteries and
at the iliac bifurcation of normal rabbit aortas. The density of
such sites was highest in the upper thoracic aorta and around
the celiac and superior mesenteric branches and was lowest in
the lower abdominal aorta. Permeable sites occurred more
frequently distal to the intercostal branch orifices and both
lateral and distal to the orifice at the celiac branch. At the
intercostal branch orifices, these sites were larger, with a lower
permeability and higher frequency than those away from the
branch. At the celiac flow divider, sites of elevated autoradiographic grain density were more permeable and larger than at
other locations in the abdominal aorta. Mean regional permeabilities were obtained by weighted area averages of low- and
high-permeability sites. Mean regional permeabilities around
the intercostal branches were 1.5 times higher than values
away from the intercostal branches. Within 0.25 and 1 mm
away from the celiac flow divider, mean regional permeability
was 3.1 and 1.3 times higher, respectively, than those away
from the flow divider. Few sites of elevated permeability were
found distal to the aortoiliac bifurcation, and the permeabilities at the medial and lateral walls of the iliac arteries were not
different. Mitotic cells were associated with 13 ±8% of all sites
with elevated permeability to 125I-LDL. The frequency of
mitotic endothelial cells was not increased at branch sites,
suggesting that mechanisms other than cell replication were
responsible for increased LDL permeability in the rabbit.
These results suggest that the permeability and frequency of
occurrence of sites with elevated permeability around the
celiac and intercostal branches may influence the distribution
and severity of early lesions in rabbits fed a hypercholesterolemic diet. {Arterioscler Thromb. 1994;14:313-323.)
Key Words • LDL • permeability • endothelium •
rabbits • arteries
A therosclerotic lesions occur preferentially at arte/ \
rial branches and along curved arteries such as
J. JL. the aortic arch and coronary arteries. At lesionprone sites in normocholesterolemic animals, endothelial permeability to macromolecules is increased.1-2 In
the rabbit aorta, foci of enhanced horseradish peroxidase (HRP) permeability were associated with sites of
increased 125I-low-density lipoprotein (125I-LDL) permeability3 and were preferentially distributed in the arch
and at arterial branch sites.4 Correspondingly, sites of
increased concentrations of radioiodinated albumin and
cholesterol in pig and dog aortas correlated with regions
prone to subsequent lesion formation.5'6 Radiolabeled
LDL concentrations and degradation rates at lesionprone sites around arterial branches in normocholesterolemic rabbits and pigeons are greater than at lesionresistant sites away from branches.7'8 However, with
cholesterol feeding, radiolabeled LDL concentrations
at lesion-prone sites increased further9 without significant changes in permeability.2
The mechanisms responsible for altered endothelial
permeability to macromolecules are poorly understood,
although sites with elevated LDL and albumin permeability in the rat aorta are associated with mitotic and
dying cells.1011 Theoretical calculations indicate that in
the absence of significant cellular metabolism and matrix binding, the steady-state concentration of LDL can
double if only 0.2% to 1% of the endothelium exhibits
increased permeability.12'13 Since long-term accumulation of LDL within arteries is significantly affected by
alterations in the permeability of a small percentage of
endothelial cells, high-resolution distributions of endothelial permeability to macromolecules are needed.
To characterize the distribution of LDL permeability
sites, we developed a method of quantitative autoradiography of en face endothelial preparations14 that
provides spatial resolution on the order of single cells.
Radioactive standards placed over endothelial preparations exhibited a linear relation between grain density
and sample activity, although the background signal
over endothelial preparations was higher than over
planar glass. Errors in the estimated permeability arising from nonuniforrnity in tissue thickness were within
experimental limits of variability. Initial results with this
technique provided estimates for values of local endothelial permeability and diffusivity of 125I-LDL in the
intima.
In this study we used quantitative autoradiography of
en face endothelial preparations to determine the spatial distribution, area, and permeability of sites with
elevated 125I-LDL permeability along the normocholesterolemic rabbit aorta, as well as the association between sites with elevated permeability and cell mitosis.
Detailed distributions were obtained around intercostal
and celiac branches and along the iliac arteries distal to
the aortoiliac bifurcation. The intercostal and celiac
Received October 29,1992; revision accepted October 13,1993.
From the Department of Biomedical Engineering, Duke University, Durham, NC.
Correspondence to George A. Truskey, PhD, Department of
Biomedical Engineering, Duke University, Box 90281, Durham,
NC 27708-0281.
314
Arteriosclerosis and Thrombosis
Vol 14, No 2 February 1994
branch regions were examined because they exhibit
different rates of lesion growth in the hypercholesterolemic rabbit.15 The iliac arteries were studied because of
the reported association between long-term radiolabeled-LDL accumulation and local arterial fluid
dynamics.16
Methods
Isolation of LDL
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Fresh human plasma from healthy donors was mixed with
phenylmethylsulfonyl fluoride to a final concentration of 2
mmol/L. LDL was isolated by conventional precipitation techniques17 as modified by Truskey et al.14 After isolation, LDL
was dialyzed at 4°C for 24 hours against three changes of 1 L
each of 0.15 mmol/L NaCl and 1 mmol/L EDTA, pH 7.4
(solution A). After dialysis the LDL solution was filtered
through a 0.22-/im filter (Millipore) and stored at 4°C. Purity
of the LDL solution was assessed by Ouchterlony immunodiffusion.18 Protein content was determined by the method of
Lowry et al19 with a mean value of 4.8 mg LDL protein/mL
(range, 2.1 to 6.9 mg/mL). Endotoxin levels were 0.2 ng/mL as
assessed by a standard Limulus amebocyte lysate assay (Lineberger Cancer Research Center, University of North Carolina
at Chapel Hill), which are below the concentrations reported
to cause alterations to the rabbit endothelium in vitro20 or
febrile response in rabbits in vivo.21 LDL was free of oxidation
products on the basis of the thiobarbituric acid-reactive
substances assay22 and fluorescence emission at 430 nm.23
Additional characterizations of the preparation have been
described in Reference 14.
Iodination of Proteins
LDL and bovine serum albumin (BSA, Sigma Chemical Co)
were iodinated with [125I]NaI (DuPont/NEN Research Products) using a modification of the iodine monochloride method.24 Unbound 125I was removed by passing the sample over a
PD-10 column (G-25M, Pharmacia Fine Chemicals) and then
by dialysis against solution A at 4°C. Greater than 95% of the
radioactivity was precipitable in 10% trichloroacetic acid.
Specific activities ranged from 131 to 369 cpm/ng LDL protein.
125
I-LDL was filtered through a 0.22-/xm filter before use.
Animal Experiments
All animal procedures were reviewed and approved by the
Duke University Institutional Animal Care and Use Committee. Four animals received injections of 125I-LDL and the fifth
served as a source of nonradioactive tissue. Over a 20-second
period, 3 mL 125I-LDL (0.46 to 1.1 mCi) was injected through
a catheter into the marginal ear vein of male New Zealand
White rabbits weighing 2 to 3 kg. The vein was flushed with 3
to 5 mL sterile isotonic saline. After 5 minutes a blood sample
was drawn for subsequent determination of trichloroacetic
acid-precipitable 125I-LDL. Eight minutes after injection, 1400
U heparin (sodium salt, grade 1, Sigma) in 2 mL saline was
injected through the catheter and flushed with saline. Ten
minutes after 125I-LDL injection, the animal was killed with 311
mg/kg sodium pentobarbital solution (Anthony Products Co).
The thoracic cavity was quickly opened, and the heart and
aortic arch were exposed. A 1/8-in. catheter with a 3/16-in.
tubing tip was inserted through an incision in the left ventricle
and into the aortic arch. A 60-mL bolus of saline at 38°C was
injected over approximately 10 seconds. Glutaraldehyde (2.5%
solution in 0.1 mol/L sodium cacodylate, pH 7.4, at 38°C) was
perfused at 100 mm Hg as the inferior vena cava was opened.
Blood and fixative were collected by aspiration. The catheter
in the aortic arch was secured with woven cotton tape.
Pressure perfusion continued for 30 to 45 minutes. After
fixation the entire aorta and iliac arteries were removed,
placed in fresh fixative, and stored at 4°C.
Tissue Preparation
Connective tissue and adherent blood were removed from
the adventitial surface under a dissecting microscope. The
aorta was opened along the ventral surface, and the celiac and
superior mesenteric arteries were divided in half. Each renal
artery was either removed and opened separately or cut and
opened along the cranial edge where it joined the aorta.
Within 1 cm proximal to the iliac bifurcation, the aortic cutting
plane was rotated 90° to produce a dorsal and ventral section
of each iliac artery. Hautchen preparations were obtained by
using a modification14 of the procedure of Hirsch et al.25 The
aorta was rinsed for at least 1 hour in running tap water. The
tissue was then cut into 10 to 14 sections, each of which was
counted for 5 minutes in a gamma counter (Multi-Prias,
Packard Instrument Co). Individual section activities were
typically 200 to 300 cpm above background (100 cpm), yielding
counting errors between 6.5% and 8.8%. Sections were flattened between glass slides, dehydrated in a graded series of
ethanol-water mixtures, and stored in fresh ethanol overnight.
Each section was removed from ethanol, blotted dry, and
pressed, luminal side down, onto clear double-sided tape on a
glass slide. Care was taken to remove all air bubbles beneath
the tissue. The adventitial surface was covered with 10%
glycerol for 20 minutes. The slide was chilled on a block of ice
for several minutes, and fine forceps were used to separate the
media from the adherent endothelial layer. Slides were air
dried in dust-free containers. After preparation of the
Hautchen layers, medial sections were counted again in a
gamma counter.
The fractional activity in the Hautchen layer, R, equals the
ratio of the radioactivities in the Hautchen preparation and
the entire tissue section^ Whole-sample tissue 125I-LDL concentrations in arteries, C,, were calculated by dividing counts
per minute per gram wet weight by the fixed tissue density of
1.09 g/cm3.26 Normalized tissue concentrations are the ratio of
tissue concentration and plasma concentration measured 5
minutes after injection of 125I-LDL, CPo. Whole-tissue permeabilities (ke, cm/s) were calculated as in Reference 14:
(1)
where L is tissue thickness and t is elapsed time in seconds
(600 seconds). A tissue thickness of 0.015 cm was used on the
basis of rabbit relaxed intimal-medial thickness,
0.0251 ±0.0031 cm, as reported by Bratzler et al,27 and the
measured ratio of rabbit intimal-medial thickness of the relaxed aorta to that of in vivo conditions, 1.72±0.15.28
Autoradiography
Uniformly labeled standards of 125I-BSA were prepared
according to the procedures of Schnitzer et al.29 Hautchen
preparations and calibration slides were dipped in nuclear
emulsion (Kodak NTB-2, Eastman Kodak Co, Inc) at 40°C and
dried vertically on filter paper at room temperature and 80%
relative humidity for 1 hour. 1429 Controls consisted of clean
blank slides and slides with a Hautchen preparation of nonradioactive tissue. Slides were packaged in light-tight microscope
slide boxes containing anhydrous calcium sulfate, sealed with
black tape, and exposed for 1 to 3 months in a refrigerator kept
at 4°C. Autoradiographs were developed by immersion in 72.5
g/L Kodak Dektol for 2 minutes with no agitation, in cold
distilled water for 15 seconds, in Kodak Regular Fix for 5
minutes, and in cold distilled water for 15 minutes. Slides were
air dried horizontally at room temperature in a dust-free
environment. To visualize endothelial cell nuclei, the slides
were stained with Gill's Hematoxylin No. 3 (Sigma) for 1
minute and rinsed with tap water. After drying, the slides were
coverslipped with a few drops of Permount (Fisher Scientific).
Herrmann et al LDL Permeability Around Arterial Branches
Image Analysis and Quantitation
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Autoradiographs were examined by dark-field microscopy
with a x40 oil-immersion objective (numerical aperture, 0.5)
and a xlO ocular lens. Video images (Series 70, Dage-MTI)
were digitized by a Macintosh IIx computer (Apple Computer
Co) equipped with an 8-bit frame grabber (Neotech Group
Ltd) and an image accelerator board (DSP 5600). Image
processing and analysis were performed with ULTIMAGE
(GFTS Inc) and IMAGE 1.43 (National Institutes of Health)
software.
The concentration and permeability over a specified region
of tissue were determined by using a relative gray value (RGV)
as follows. At the beginning of each session, a reference slide
was examined and the brightness of the microscope lamp and
image-capture software adjusted to ensure that grain-free
regions always had the same gray value. To account for
variations in light intensity due to nonuniform hematoxylin
staining of the Hautchen preparations, gray values in grainfree regions were set equal to the gray values of grain-free
regions of nonradioactive Hautchen preparations. Sites with
elevated permeability were outlined and the mean gray value
was determined. An adjacent low-permeability region of equal
area was analyzed in an identical manner. Away from the
branches, high- and low-permeability sites were examined in
either 33 or 36 1-mm2 areas uniformly distributed along each
thoracic aorta. All high-permeability sites within each 1-mm2
region were analyzed. Results were reported in terms of the
RGV, which equals the mean gray value of a region of interest
divided by the total number of gray values.14
For each experiment, activities of calibration standards were
corrected to the day of the experiment, t,, with the following
formula:
where Cg(t2) (cpm//wn2) is the activity before correction, t,-t2
is the time in days to be corrected, and A is the decay constant
for 125I (A=ln2/60 days). RGV wasfitto the following equation:
(3)
RGV=b+mCg(t,)
in which they intercept b is unitless and the slope m is in units
of 1000 £i.m2/cpm. To correct for higher background levels over
nonradioactive tissue, the intrinsic background level of nonradioactive tissue, be, was determined from slides of nonradioactive Hautchen preparations that were processed along with
the radioactive tissue. Details of the validation of the quantitative analysis of en face autoradiography have been described
in Reference 14.
For each region of interest, the amount of 125I-LDL per unit
area in the Hautchen preparation, M,(LH,t), the mean concentration in the Hautchen layer, (cpm/cm3), and the local permeability, kc, were determined as in Reference 14:
(RGV-be)
V*)
m
M,{LH,t)
(5)
C,(LH,t)=
(6)
M,(LH,t)JRGV-be)
CPoRt ~ mCPoRt
where R is the fractional amount of 125I-LDL in the Hautchen
layer and t is the time that radiolabeled LDL was present in
blood (600 seconds). A mean value of 2.0 fim was used for the
thickness of the Hautchen layer, LH, on the basis of measured
thicknesses at branch and nonbranch regions.
315
To compare autoradiographic results from this study with
whole-sample data from other studies, mean regional permeability, k~t, was calculated using the following formula:
(7)
A,
where L(kaA,) is the sum of products of individual sites with
increased permeability multiplied by their respective area; 2L4,
is the sum of areas of sites with increased permeability; AT is
the area over which mean regional permeability is determined,
and ktL is mean permeability in low-permeability areas. The
two terms in Equation 7 represent the contribution to total
permeability from focal sites of increased permeability and
low-permeability areas.
Mapping Regions of Elevated Permeability
A composite image of each autoradiograph of each
Hautchen preparation was created by pasting together lowpower (x25) images. The planar (x-y) coordinates of tissue
borders and locations of major branches were determined.
Individual sites of elevated permeability were located on the
composite image and the coordinates recorded. Individual
aortic-section maps prepared from Hautchen-layer images
were combined to form composite maps of high-permeability
sites along the entire length of excised aorta.
Around individual vessel branches, a polar-coordinate system15'30 with the origin at the center of the orifice was
superimposed onto the image of the orifice with 0°, proximal;
180°, distal; 90°, to the animal's left; and 270°, to the animal's
right. Angles were grouped into six regions: 330° to 30°
(proximal); 30° to 90°, 90° to 150°, and 150° to 210° (distal);
210° to 270°; and 270° to 330°. For intercostal orifice pairs, the
animal's left branch retained this orientation, whereas a
mirror-image coordinate system was used for the animal's right
intercostal orifice. This designation allowed definition of a
medial line down the aorta and centered between each orifice
pair. To include a uniform area around each orifice, the
examined area was restricted to a circle with a radius equal to
half the distance between the center of the branches. The
polar angle and distance from the center of each orifice to the
center of the site with elevated permeability about the orifice
were recorded.
Polar-coordinate maps of the abdominal aorta near the
celiac branch were constructed by locating and estimating the
perimeter of the celiac orifice. The flow divider was identified
by an abrupt change in orientation of endothelial cell nuclei.
With theflowdivider and the upstream point (where the celiac
branch was connected to the aorta) as guides, a semicircle
representing the orifice edge was drawn for each half of the
celiac branch and the perimeter was measured (see Fig 4).
Sites of low and high permeability within 0.25, 1, and 4 mm of
the branch were located by determining the radius from the
center of the orifice and the angle relative to the direction of
blood flow.
Statistical Analysis
To account for the significant variability between animals,
ANOVA31 was used to detect differences between anatomic
location (proximity to branch or polar angle around branch)
and among individual animals for variables such as wholetissue concentrations, permeabilities of sites with increased
grain density, and areas of highly permeable regions. Unless
otherwise noted, significant interanimal differences for autoradiographic samples were present with P<.001. Means were
calculated as averages among animals. SEM was calculated
from the mean of averaged data for individual animals. A x2
goodness-of-fit test was used to determine whether the sites
with increased permeability around vessel branches were
uniformly distributed. For 2 animals, frequencies of sites with
increased permeability at several angular locations were zero,
316
Arteriosclerosis and Thrombosis
Vol 14, No 2 February 1994
Intercostal A. Pairs
#7 #8
Intercostal A. Pairs
#3 #4 #5 #6
!'. 1 mm
"-••-«•••
••*-V3VSSs.
•.*.•>..•**« • °*tri
Aortic Arch
CeliacA.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
CeliacA.
S. Mesenteric A
Direction of Flow
S. Mesenteric A.
Lumbar A.
FIG 1. Map showing distribution of sites with increased permeability to low-density lipoprotein (LDL) (o) and mitotic cells (+) from the
third intercostal artery (A.) pair to the region just above the iliac bifurcation. S. indicates superior; R., right; and L., left.
which biased the x2 values.31 Consequently, Rayleigh's nonparametric test31 was applied to the individual animal data to
determine whether the sites with increased permeability followed a uniform circular distribution.
Results
Tissue Concentrations and Permeabilities
For j a c h animal, relative whole-tissue concentrations, CJCfo, were determined from 5 thoracic and 5
abdominal aortic sections. A two-way ANOVA showed
significant differences between thoracic and abdominal
whole-sample tissue concentrations (f><.002) without
significant variations among animals (P<3). The mean
C,ICpo value from the 4 animals was 0.72±0.08xl0" 3
(mean±SEM) in the thoracic aorta and 1.6±0.3xl0" 3
in the abdominal aorta. Corresponding whole-tissue
permeabilities were 1.8±0.2xl0~ 8 cm/s for thoracic
sections and 3.9±0.8xl0~ 8 cm/s for abdominal sections.
Distribution of LDL Permeability in the Aorta
Sites with elevated autoradiographic grain density
and mitotic cells from the third intercostal artery pair to
the lower abdominal aorta from 1 animal are shown in
Fig 1. Sites with elevated permeability to LDL occurred
throughout the aorta, with the highest frequency in the
upper thoracic aorta and the lowest in the lower abdominal region. From 1017 sites examined in 4 animals,
areas ranged from 150 to 3030 jam2 (mean, 2420 /xm2).
Large streaks (> 1 mm) of high grain density oriented in
the direction of flow were present along the thoracic
aorta. Some sites were clustered, and not all sites were
associated with branches. Grain densities for low-permeability regions were uniformly higher in some regions
than others, particularly in the arch and upper thoracic
aorta. As shown in Fig 1, most mitotic cells in the
thoracic aorta showed increased permeability to 125ILDL, whereas along the lower abdominal aorta, fewer
mitotic cells were associated with regions of increased
permeability.
Hautchen preparations were previously characterized14 in terms of the thickness of the preparations and
the presence of any overlying internal elastic laminae or
smooth muscle cells. By using the calibrated fine-focus
knob of the microscope to calculate the distance between the center of the nucleus and the plane of the
autoradiographic grains, we obtained additional observations of the thickness of the Hautchen preparations at
and away from branch sites. Hautchen preparations
were 1.8±1.0 ju.m thick at intercostal and celiac
branches and 2.1 ±1.2 j^m thick at nonbranch sites in the
thoracic and abdominal aortas. These values are similar
to those previously obtained14 for nonbranch regions
(ie, 2.3+0.9 ^m). Patches of internal elastic lamina and
smooth muscle cells were infrequent around both
branch and nonbranch sites.
The mean intimal concentration of high-permeability
sites to 125I-LDL was 0.51 ±0.07, with a mean value of
4.4±0.6xl0" 7 cm/s. The mean intimal concentration of
125
I-LDL in 1017 low-permeability regions, which represented 0.2% of the total intimal area, was 0.032±0.015,
with a mean value of 2.6±l.lxlO~ 8 cm/s. From wholesample measurements, the fractional amount of 125ILDL in the Hautchen layer, R, was 38±5%, which
compares favorably with the value obtained in our
previous study.14 Thus, Hautchen preparations consisted of only about 2% of the thickness of the arterial
wall but contained approximately 38% of the activity,
Herrmann et al LDL Permeability Around Arterial Branches
317
TABLE 1. 125I-LDL Permeability, Area, and Frequency of
Occurrence of Sites With Elevated Grain Density Near
and Away From Intercostal Branches
120°
150"
At Branch
Away From Branch
High permeability,
cm/sxiO" 7
3.5+0.7
4.0±0.4(P<.001)*
Low permeability,
cm/sxiO~ 7
Parameter
Flaw
180°
210°
.* / 330°
300°
270°
FIG 2. Polar-coordinate plot of regions with increased permeability surrounding intercostal branch sites in the thoracic aorta
from 4 animals. Concentric circles represent 0.5-mm distances
from the center of the orifice. Innermost circle represents the
mean orifice perimeter (radius, 0.34±0.05 mm). Direction of
blood flow is indicated by the arrow.
240°
0.14±0.02
0.14+0.02
Area, /im 2 xi0~ 3
2.1 ±0.5
2.1 ±0.5
Frequency,
No./mm2
2.2±1.0
0.91 ±0.22 (P<.001)*
LDL indicates low-density lipoprotein. All data are mean±SEM,
n=4 for all data.
*Significance calculated from two-way ANOVA as described in
"Methods."
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
limited to those sites that were less than or equal to the
smallest half distance between a pair of intercostal
arteries (0.7 mm). The highest frequency of sites with
elevated permeability per square millimeter occurred
between 0.5 and 0.6 mm away from the center of the
orifice in the region between 90° and 270°. The proximal
region (330° to 30°) and the lateral region (30° to 90°)
had the lowest frequency of occurrence. The mean
distance of high-permeability sites from the orifice
center was between 0.6 and 0.7 mm and was independent of polar angle.
A two-way ANOVA of the magnitude of high-permeability values around the intercostal arteries failed to
detect significant differences among the six polar-angle
groups (.10<P<.20). The mean value of elevated-permeability sites around intercostal arteries was 3.5±O.8xlO"7
cm/s. For the same data, the area of regions with increased
permeability did not vary significantly among animals
(.10<P<.2) or among polar-angle groups (P>.5). The
mean area around the intercostal arteries was 2100±500
indicating a steep concentration gradient in the arterial
intima.
Sites of Increased Permeability at
Intercostal Branches
Forty-two intercostal branch regions from 4 animals
were examined, yielding a total of 211 sites with elevated permeability surrounding intercostal branches.
The mean number of sites per intercostal orifice was
5.5±0.8, with a range of 1.1 to 9.4 sites per orifice among
the 4 animals. The angular distribution of these regions
is shown in Fig 2. By the Rayleigh nonparametric test,
all 4 animals had distributions of sites around the
intercostal branches that deviated from a uniform distribution, with a mean angle of 167 ±17° (range, 151° to
190°). The distribution of all 211 sites deviated from a
uniform distribution by both the Rayleigh test (P<.001)
and x1 test (P<.005).
For each of the six polar-coordinate regions denned
in "Methods," the frequency of sites with elevated
permeability per square millimeter of surface area was
plotted versus the radial distance from the orifice center
(Fig 3). To make a valid comparison, the data set was
fim2.
The permeability, area, and frequency of sites with
increased grain density at and away from the intercostal
ostia were compared (Table 1). A two-way ANOVA
indicated that the frequency at branch sites was signif-
8
—O—
—•—
—P—
—•—
—&—A—
a
Edge of Orifice
SMO-SO"
30°-90°
90°-150°
150°-210°
210°-270°
270°-330°
A
rence of regions with increased permeability
vs distance from the center of intercostal
branches. Data are plotted for the six angularsegment regions surrounding the orifice. Innermost circle represents the mean orifice
perimeter (radius, 0.34±0.05 mm).
•
0.0
0.1
0.2
03
0.4
0.5
Distance from the center of orifice, mm
0.6
0.7
318
Arteriosclerosis and Thrombosis
Vol 14, No 2 February 1994
1 mm
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
FIG 4. Individual maps of the celiac region of aortas from the 4 rabbits. Dashed lines represent approximate locations of the celiac
branch. Arrows represent the direction of blood flow. Acellular areas are cross-hatched and diagonal lines represent larger streaks of
elevated grain density.
icantly higher than that away from branches (/><.001).
The mean permeability of highly permeable sites surrounding intercostal branches was 13% lower than that
at sites away from such branches. Low-permeability
values at and away from intercostal arteries were not
significantly different (Table 1). Sites of elevated permeability at intercostal branches were not significantly
larger than those away from such branches (P>.5).
Sites of Increased Permeability at the
Celiac Branch
The location of sites with increased permeability near
the celiac branch area for the 4 animals is shown in Fig
4. All animals showed several sites of increased permeability near the flow divider and a region devoid of sites
with increased permeability along the dorsal surface
approximately 4 to 6 mm from the flow divider. Sites
with elevated permeability to LDL around the celiac
artery were replotted in a polar-coordinate system as
described in "Methods" (Fig 5). By the Rayleigh nonparametric test, all 4 animals had distributions of sites
around the celiac branches that deviated from a uniform
distribution. Unlike that for the intercostal arteries, the
mean angle near the celiac branch varied among the 4
animals, with values of 1°, 79°, 187°, and 262°. Such
variation is consistent with a heterogeneous distribution
of sites.31 With data from all 283 sites from the 4
animals, both the Rayleigh and x2 goodness-of-fit tests
120°
210'
300°
FIG 5. Polar-coordinate plot of regions with increased permeability surrounding the celiac branch site. Direction of blood flow
is indicated by the arrow. Concentric circles represent 1.0-mm
distances from the center of the orifice. Innermost circle represents the mean orifice perimeter (radius, 1.5±0.1 mm).
Herrmann et al
LDL Permeability Around Arterial Branches
319
330°-30°
30°-90°
90°-150°
150°-210°
210°-270°
270°-330°
FIG 6. Line plot showing frequency of occurrence of regions with increased permeability vs distance from the center of the
celiac branch. Data are plotted for the six
angular-segment regions surrounding the
orifice.
•a
1
2
3
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Distance from center of orifice, mm
indicated that the distribution of these regions around
the branch was not uniform (/><.001). Most sites with
elevated permeability occurred lateral to the celiac
orifice.
The frequency of sites with elevated permeability per
square millimeter of surface area was plotted versus the
radial distance from the orifice center (Fig 6). The
frequency tended to be highest at the flow divider and
decreased with distance from the orifice edge. Between
330° and 30°, the frequency was relatively low except at
the flow divider. Lateral regions (30° to 150° and 210° to
330°) had a higher frequency per square millimeter than
did the distal and proximal regions that were more than
0.5 mm from the edge of the orifice.
From a two-way ANOVA of high-permeability sites
around the celiac branch, no significant differences in the
magnitude of permeability among the six polar-angle
groups were determined (.20<P<.25). The mean permeability around the celiac branch was 5.7±0.4xl0~7 cm/s.
The largest mean areas occurred within the angular segments of 150° to 210° (distal) and 270° to 330° (medial).
The mean area of regions with increased permeability
surrounding the celiac branches was 2300±400 jam2.
In an additional effort to isolate the effect of location
on permeability in the vicinity of the celiac branch,
regions of increased permeability were divided into two
groups: those within 0.25 mm of either side of the flow
divider and those more than 0.25 mm away from the
flow divider. Regions of increased permeability at the
flow divider included sites in both main and branch
vessels. A two-way ANOVA indicated that permeabilities of sites with increased grain density at the flow
divider were 18% greater than those of similar sites
away from the flow divider (P<.001) (Table 2). Lowpermeability values within 1 mm of the celiac flow
divider were, however, not significantly different from
those at greater distances from the flow dividers (Table
2). The areas of sites with increased permeability varied
significantly between animals (.001<P<.002) and between angular segments (.01<P<.025), with no significant interaction between these two factors. Furthermore, high-permeability sites were 3.3 times larger at
the flow divider than at sites away from the flow divider
(P<.001). The frequency of sites with elevated permeability at the celiac flow divider was 17.1 ±4.6 sites/mm2,
which was significantly greater than that over the entire
celiac branch region (l.l±0.3 sites/mm2; P<.001).
Mean Regional Permeabilities
The mean regional permeability around intercostal
and celiac branch sites was calculated by using Equation
7 to assess the influence of sites with elevated permeability on mean permeabilities in whole-tissue sections
(Table 3). The intercostal branch sites consisted of
tissue from the orifice edge to 0.7 mm from the edge of
the branch. At the celiac branch site, regions within 0.25
and 1 mm of the flow divider were examined.
The mean regional permeability at the intercostal
branch site was 1.5 times larger than that away from the
site (Table 3). Within 0.25 mm of the celiac flow divider,
the mean regional permeability was 3.1 times larger
than the value away from the branch site. When a region
within 1 mm of the flow divider was examined, the mean
regional permeability was 1.3 times greater than the
value away from the flow divider. The larger mean
regional permeability at the celiac flow divider was due
TABLE 2. 12SI-LDL Permeability, Area, and Frequency of
Occurrence of Sites With Elevated Grain Density Near
and Away From the Celiac Branch
Parameter
At Flow Divider Away From Flow Divider
High permeability,
cm/sx10" 7
6.6±0.3
Low permeability,
cm/sx10" 7
0.33±0.04
Area, /i.m2x10~3
7.3 ±0.9
2.3+0.2 (P<.001)*
17.1 ±4.6
1.1+0.3(P<.001)*
Frequency,
No./mm2
5.6+0.4 (P<.001)*
0.37±0.02
LDL indicates low-density lipoprotein. All data are mean±SEM,
n=4for all data.
•Significance calculated from two-way ANOVA as described in
"Methods."
320
Arteriosclerosis and Thrombosis
Vol 14, No 2 February 1994
TABLE 3. Mean Regional Permeabilities for Regions in the Thoracic Aorta at and Away From the
Intercostal Branch Site and at Regions Surrounding the Celiac Branch at and Away From the Celiac
Flow Divider
Weighted Permeabilities, cm/sxio"
Region
Due to Focal Sites of
Increased Permeability
Due to
Low-Permeability Areas
Mean Regional
Permeability,
Re, cm/sxio 8
Intercostal
0.81 ±0.01
1.37±0.15
2.2+0.2
0.070±0.001
1.41 ±0.20
1.5±0.2
8.62±0.23
3.29±0.44
11.9±0.5
<1 mm from flow divider (90°-270°)
1.30±0.01
3.68±0.31
5.0±0.3
Away from flow divider
0.15±0.01
3.73±0.17
3.9±0.2
At branch
Away from branch
Celiac
<0.25 mm from flow divider
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
to the higher permeabilities, frequencies of occurrence,
and areas of sites with increased permeability around
the flow divider.
Permeability Distal to the Aortoiliac Bifurcation
In 3 animals, enough of the iliac arteries was preserved to allow examination of the region around the
iliac bifurcation for focal regions of increased permeability and variations in background permeability level.
In general, this region contained very few focal regions
of increased LDL permeability. In the iliac artery of 1
animal, four regions of increased permeability were
observed. One site was located at the iliac bifurcation,
two surrounded a small branch vessel, and the fourth
was associated with a mitotic cell and not with any
branch or flow divider. In this same animal, two other
mitotic cells were found that were not associated with
regions of increased permeability. A second animal had
a small streak of elevated grain density (about 1 mm2) in
the vicinity of the bifurcation along the lateral side of
the iliac artery. The third animal had no regions of
elevated permeability. The low frequency of sites with
elevated permeability in the iliac arteries was also
observed in 2 animals that were not part of this study.
In a manner analogous to that of Berceli et al,16 mean
low-permeability values in medial and lateral regions of
the iliac arteries that are 0 to 2, 2 to 4, and 4 to 6 mm
past the iliac bifurcation are shown in Fig 7. From a
three-way ANOVA no significant differences were detected among permeabilities in any of the three regions
or between medial and lateral samples, despite the
different hemodynamic environments.16 The mean permeability of low-permeability sites along the first 6 mm
of the iliac artery was 2.7±0.9xl0" 8 cm/s.
Frequency of Mitotic Cells Around Intercostal and
Celiac Orifices
The "cell turnover-leaky junction" hypothesis postulates that mitotic endothelial cells represent one mechanism by which overall permeability of the endothelium
increases.10-11 If this hypothesis is valid, then mitotic
frequency should be correlated with the frequency of
sites with elevated permeability. A two-way ANOVA
indicated significant differences in mitotic frequency
among the 4 animals (.002</'<.004) but no significant
differences in the frequency of mitotic endothelial cells
between the thoracic (0.13 ±0.05 mitotic cells/mm2) and
abdominal (0.10±0.04 mitotic cells/mm 2 ) aortas
(.05<P<.10). Mitotic frequency away from intercostal
branches (0.09±0.04 mitotic cells/mm2) was not significantly different from that around intercostal branches
(0.49±0.23 mitotic cells/mm2; .10</'<.15). At the celiac
flow divider, no mitotic cells were observed in any of the
4 animals.
The percentage of mitotic cells that showed increased
permeability to 125I-LDL was 90±5% in the thoracic
aorta, 71 ±3% in the abdominal aorta, and 78±5%
overall. The percentage of sites with increased permeability to LDL that was associated with mitotic cells,
however, was 15±9% in the thoracic aorta, 12±8% in
the abdominal aorta, and 13±8% overall. One animal
was excluded from analysis because it had only two
mitotic cells in the tissue under study, and neither
exhibited increased permeability to 125I-LDL. Mitotic
cells represented the same percentage of highly permeable cells both at and away from the branch sites.
Although mitotic endothelial cells are generally highly
permeable to 125I-LDL, they account for only a small
percentage of the total number of sites that exhibit
elevated permeability to 125I-LDL.
Discussion
This study provides the first detailed characterization
of sites with elevated permeability to 125I-LDL around
D Medial
£3 Lateral
Ti
0-2 mm
2-4 mm
4-6 mm
Distance Along the Iliac Artery
FIG 7. Bar graph showing background permeabilities in the first
three 2-mm regions along the medial and lateral walls of iliac
arteries. Distance was measured from the bifurcation.
Herrmann et al LDL Permeability Around Arterial Branches
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
arterial branches in the normocholesterolemic rabbit.
Previous studies with rabbits used gamma counting to
examine regional variations in endothelial permeability
to LDL2-3 for samples as small as 1 mm2. These samples
contained several thousand endothelial cells, and the
association of enhanced permeability to LDL with
cellular events could not be assessed. Fluorescently
labeled LDL was used to determine the relationship
between LDL permeability and endothelial cell mitosis
in the rat aorta,1011 but the permeability and distribution of sites with elevated permeability were not reported. Detailed distributions of sites with elevated
permeability to HRP were obtained for major branches
of the aorta.4 Quantitative autoradiography of endothelial preparations provides the sensitivity and spatial
resolution14 needed to determine the arterial permeability to 125I-LDL at a cellular level.
In the current study significant differences were observed in the distribution, area, and permeability of
sites with elevated permeability to 125I-LDL around the
intercostal and celiac ostia. These sites were most
abundant distal to intercostal orifices, with a higher
frequency but lower permeability than nearby nonbranch regions. In contrast, high-permeability sites
were most abundant distal and lateral to the celiac
orifice, with a higher frequency, area, and permeability
than nearby nonbranch regions. The highest frequency
of these sites was along the celiac flow divider.
Along the iliac arteries, incorporation and degradation of double-labeled LDL (125I-tyramine cellobiose
and 131I) have been reported to be greatest where in
vitro flow studies show a region of flow reversal.16
Analysis of the permeability in this region (Fig 7)
showed no changes along medial and lateral walls or
within the first 6 mm from the iliac bifurcation. In fact,
very few sites with elevated permeability were observed
in this region. Together with results from the current
study, we conclude that increased LDL accumulation
along the rabbit iliac artery is not due to increased
permeability. The increased incorporation and degradation of radiolabeled LDL in regions of flow reversal
may represent altered transport properties due to
changes in extracellular matrix composition or internal
elastic lamina structure or altered LDL degradation
rates in the arterial wall.
A possible limitation of this study is the intra-aortic
injection of 60 mL saline over 10 seconds. This technique represents a rapid 30% increase in blood volume,
which could stretch the aortic wall to artifactually
increase the transport of LDL through experimentally
induced gaps of stretched endothelium. If significant,
we would expect changes in the magnitude of lowpermeability values as well as the magnitude and distribution of high-permeability sites. We did not observe
any increase in low-permeability values around either
the intercostal branches or the celiac branch. Furthermore, the frequency of sites with elevated permeability
to LDL around the intercostal and celiac branches is
comparable to distributions of sites with elevated permeability to HRP. 4 The effect of arterial stretch on
permeability should be minimized and highly transient
because the duration of stretch was short relative to the
overall time of incubation with radiolabeled LDL, and
the volume increase would be largely absorbed by the
more compliant venous system. Furthermore, injection
321
of saline would substantially dilute the concentration of
radiolabeled LDL in contact with the arterial wall.
Assuming that focal sites with elevated permeability to
125
I-LDL were associated with only one endothelial cell
and using a mean endothelial cell area of 300 ju,m2,4 we
calculated that the fraction of endothelial cells associated
with increased permeability was 2.7±1.2xlO~4. This value
compares favorably with a value of 1.9 x 10~4 obtained with
HRP that circulated for only 1 minute in the rabbit.4 The
distribution of sites with elevated 125I-LDL and HRP
permeability obtained in the two studies is similar. The
similarity of frequency and distribution of sites with elevated permeability to HRP and 125I-LDL suggests that
altered endothelial permeability is due to nonspecific
transport mechanisms and does not represent an alteration in receptor-mediated LDL transport across the
endothelium. Increased transport through endothelial cell
junctions is one mechanism that may account for such high
permeabilities.1213
The mean regional permeabilities permitted comparison of the current study with studies in which larger
tissue areas were sampled. Mean regional permeability
values around intercostal branches were 1.5 times
higher than those away from intercostal branches.
Within 0.25 and 1 mm of the celiac flow divider, the
mean regional permeability was 3.1 and 1.3 times
higher, respectively, than that away from the flow
divider. These results are qualitatively consistent with
those of Schwenke and Carew,2 who observed a twofold
increase in permeability at the celiac branch but found
no change in permeability values between intercostal
and nonbranch sites in the thoracic aorta. The quantitative differences in permeabilities between the two
studies may reflect the increased sensitivity of the en
face autoradiographic technique in this study or differences in the size of regions examined in the two studies.
Mean regional permeabilities away from branch sites
and whole-sample permeabilities obtained in this study
are less than those obtained by Bratzler et al27 in the
rabbit thoracic aorta but greater than other reported
values for the thoracic and abdominal aortas in the
rabbit2'32 and squirrel monkey.33
Differences in the frequency, area, and permeability
of sites with elevated permeability to LDL around
arterial branches may influence subsequent accumulation of LDL within the arterial wall. The relative
importance of the endothelium as a barrier to macromolecular transport is characterized by the Biot number,34 which is the ratio of the diffusive resistance within
the intima and media (L2/Dep where Deff is the effective
diffusion coefficient) to the resistance of the endothelium (L/ke). Using an intimal-medial thickness of 0.015
cm, an effective diffusion coefficient of 5.4x10"'° cm2/
s,14 and a permeability of 2.6 xlO" 8 cm/s yields a Biot
number of 0.72 for normal endothelium. Thus, the
normal endothelium represents 58% of the mass transfer resistance in the arterial wall, a value which is
consistent with the low values of normalized arterial
concentrations of 125I-LDL obtained during transient
experiments.7-27-33 At high-permeability sites, the mean
Biot number is 12.2 and the endothelium represents
only 7.6% of the mass transfer resistance in the arterial
wall. Given the large value for diffusion time (116
hours), accumulation and metabolism within the arte-
322
Arteriosclerosis and Thrombosis
Vol 14, No 2 February 1994
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
rial wall are thus likely to be limited by the rates of
diffusion and convection in the intima and media.
Theoretical studies by Weinbaum and colleagues12
and Yuan et al13 show that the concentration of macromolecules within the arterial wall is strongly dependent
on the frequency of highly permeable endothelial cells.
These theoretical studies predict that the low frequency
of highly permeable sites away from branch regions (as
observed in this study; Tables 1 and 2) should have little
effect on the steady-state concentration of LDL within
the arterial wall. For frequencies similar to those
around the celiac flow divider, the theoretical studies
predict that steady-state concentrations should increase
between twofold and fourfold.12'13 Even if metabolic
phenomena are included in the analysis, increases in
endothelial permeability should produce increased concentrations of labeled LDL but at levels lower than
predicted by current theory.
The distribution of lesions around intercostal and
celiac branches in rabbits fed a cholesterol1535 or
casein36 diet was determined by the polar-mapping
method. Around intercostal branches, lesions appeared
lateral to the orifice after 15 days of cholesterol feeding.15 For longer periods of a high-cholesterol1535 or
casein36 diet, the longest lesions were distal to the
branch site. At the celiac branch, lesions developed
along the lateral edge of the flow divider15 within 7 days
of the start of cholesterol feeding. For longer-term
feeding, lesion length was largest on either the lateral 1535 or distal36 portion of the branches.
Barakat et al4 have observed a significant correlation
between the frequency of sites with increased HRP
permeability and lesion length15 around intercostal arteries but have not observed any correlation for areas
around the celiac artery or other abdominal branches.
Near the celiac artery the distribution of sites with
elevated permeability to HRP was more uniform than
that observed with 125I-LDL. Sites with elevated permeability to 125I-LDL around intercostal (Fig 2) and celiac
(Fig 5) branches tend to be greater because lesions are
more likely to occur around these branches. Our results
further indicate that differences in the time course of
appearance of lesions around these two branches may
be partially due to differences in the area and permeability of highly permeable sites.
The cell turnover-leaky junction hypothesis10-11 postulates that increased permeability associated with mitotic endothelial cells may represent one mechanism by
which LDL accumulates within the intima during early
lesion formation. Support for this hypothesis is provided
by the high percentage of mitotic and dying cells in the
rat aorta that are permeable to Evan's Blue albumin, 103738 HRP, 10 and LDL.11 Penn et al,39 however,
have shown that increased permeability to HRP associated with endotoxin damage is not temporally correlated with endothelial cell death or proliferation.
In the present study 71% of all mitotic cells in the
rabbit aorta exhibited increased permeability to 125ILDL, but only 13% of sites highly permeable to LDL
were associated with mitotic cells. Although a similar
percentage of mitotic cells in rat thoracic aortas (80%)
exhibited increased permeability to Lucifer yellowlabeled LDL, mitotic cells accounted for 45% of all
permeable sites.11 Interestingly, the mitotic frequency in
the rat aorta (0.034±0.006%, Reference 11) was signif-
icantly higher than the values we observed in the rabbit
(0.004±0.002%, authors' unpublished data). The different contributions of mitotic cells to the total population
of permeable sites for rats and rabbits may simply
reflect the difference in mitotic frequency for the two
species.
If cell mitosis significantly influences endothelial permeability, then mitotic frequency should be greater at
branch sites that exhibit increased frequencies of sites
with elevated permeability to 125I-LDL. We failed to
observe such a correlation near celiac and intercostal
branches. Although our results are consistent with the
hypothesis that increased endothelial permeability occurs as a result of increased transport through endothelial junctions, endothelial cell mitosis is not a significant
mechanism responsible for increased transendothelial
transport of macromolecules.
In conclusion, endothelial preparations were used to
characterize, at a cellular level, the area, distribution,
and magnitude of regions with increased permeability
to 125I-LDL along the rabbit aorta. Results from this
study suggest that permeability and the frequency of
occurrence of sites with elevated permeability near
celiac and intercostal branches may be important in the
initiation of lesions in hypercholesterolemic rabbits.
Acknowledgments
This study was supported by National Institutes of Health
grant HL-41372. The authors thank Mr A. Barakat from the
Massachusetts Institute of Technology for providing a preprint
of Reference 4 and Dr Sheldon Weinbaum from the City
University of New York for helpful conversations.
References
1. Bell FP, Adamson IL, Schwartz CJ. Aortic endothelial permeability to albumin: focal and regional patterns of uptake and
transmural distribution of 13lI-albumin in the young pig. Exp Mol
Pathol. 1974;20:57-68.
2. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in
cholesterol-fed rabbits, II: selective retention of LDL vs selective
increases in LDL permeability in susceptible sites of arteries.
Arteriosclerosis. 1989;9:908-918.
3. Stemerman MB, Morrel EM, Burke KR, Colton CK, Smith KA,
Lees RS. Local variation in arterial wall permeability to low
density lipoprotein in normal rabbit aorta. Arteriosclerosis. I986;6:
64-69.
4. Barakat Al, Uhthoff PAF, Colton CK. Topographical mapping of
sites of enhanced HRP permeability in the normal rabbit aorta.
J Biomech Eng. 1992;114:283-292.
5. Packham MA, Rowsell HC, Jorgensen L, Mustard JF. Localized
protein accumulation in the wall of the aorta. Exp Mol Pathol.
1967;7:214-232.
6. Somer JB, Schwartz CJ. Focal 3H-cholesterol uptake in pig aorta.
Atherosclerosis. 1971;13:293-304.
7. Schwenke DC, Carew TE. Quantification in vivo of increased LDL
content and rate of LDL degradation in normal rabbit aorta
occurring at sites susceptible to early atherosclerotic lesions. Circ
Res. 1988;62:699-710.
8. Schwenke DC, St Clair RW. Accumulation of 125I-tyramine cellobiose-labeled low density lipoprotein is greater in the atherosclerosissusceptible region of White Carneau pigeon aorta and further
enhanced once atherosclerotic lesions develop. Arterioscler Thromb.
1992;12:446-460.
9. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in
cholesterol-fed rabbits, I: focal increases in arterial LDL concentration precede development of fatty streak lesions. Arteriosclerosis. 1989;9:895-907.
10. Chuang PT, Cheng HJ, Lin SJ, Jan KM, Lee MML, Chien S.
Macromolecular transport across arterial and venous endothelium
in rats: studies with Evans blue-albumin and horseradish peroxidase. Arteriosclerosis. 1990;10:188-197.
Herrmann et al
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
11. Lin SJ, Jan KM, Weinbaum S, Chien S. Transendothelial transport
of low density lipoprotein in association with cell mitosis in rat
aorta. Arteriosclerosis. 1989;9:230-236.
12. Weinbaum S, Tzeghai G, Ganatos P, Pfeffer R, Chien S. Effect of
cell turnover and leaky junctions on arterial macromolecular
transport. Am J Physiol. 1985;248:H945-H960.
13. Yuan F, Chien S, Weinbaum S. A new view of convective-diffusive
transport processes in the arterial intima. JBiomech Eng. 1991;113:
314-329.
14. Truskey GA, Roberts WL, Herrmann RA, Malinauskas RA. Measurement of endothelial permeability to 125I-low density lipoproteins in rabbit arteries using en face preparations. Circ Res.
1992;71:883-897.
15. Zeindler CM, Kratky RG, Roach MR. Quantitative measurements
of early atherosclerotic lesions on rabbit aortae from vascular casts.
Atherosclerosis. 1989;76:245-255.
16. Berceli SA, Warty VS, Sheppeck RA, Mandarino WA, Tanksale
SK, Borovetz HS. Hemodynamics and low density lipoprotein
metabolism: rates of low density lipoprotein incorporation and
degradation along medial and lateral walls of the rabbit aorto-iliac
bifurcation. Arteriosclerosis. 1990; 10:688-694.
17. Burstein M, Scholnick HR, Morfin R. Rapid method for the isolation of lipoproteins from human serum by precipitation with
polyanions. / Lipid Res. 1970;ll:583-595.
18. Hatch FT, Lees RS. Practical methods for plasma lipoprotein
analysis. Adv Lipid Res. 1968;6:l-68.
19. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:
265-275.
20. Navab M, Hough GP, van Lenten BJ, Berliner JA, Fogelman AM.
Low density lipoproteins transfer bacterial lipopolysaccharides
across endothelial monolayers in a biologically active form. / Clin
Invest. 1988;81:601-605.
21. Tomasulo PA, Levin J, Murphy PA, Winkelstein JA. Biological
activities of titrated endotoxins: correlation of the Limulus lysate
assay with rabbit pyrogen and complement-activation assays for
endotoxin. J Lab Clin Med. 1977;89:309-315.
22. Morel DW, DiCorleto PE, Chisolm GM. Endothelial and smooth
muscle cells alter low density lipoprotein in vitro by free radical
oxidation. Arteriosclerosis. 1984;4:357-364.
23. Esterbauer H, Jurgens G, Quehenberger O, Roller E. Autoxidation of human low density lipoproteins: loss of polyunsaturated
fatty acids and Vitamin E and generation of aldehydes. J Lipid Res.
1987;28:495-509.
24. Bilheimer DW, Eisenberg S, Levy RI. The metabolism of very low
density lipoproteins, I: preliminary in vitro and in vivo observations. Biochim Biophys Ada. 1972;260:212-221.
LDL Permeability Around Arterial Branches
323
25. Hirsch EZ, Martino W, Orr CH, White H, Chisolm GM. A simple
method for the preparation of en face endothelial (Hautchen)
monolayers from rat and rabbit aortas. Atherosclerosis. 1980;37:
539-548.
26. Fry DL, Mahley RW, Oh SY, Lewis SJ, Plowman F, Swyt CR.
Quantitative microautoradiography of arteries: comparison of
radioactivity to silver. Am J Physiol. 1980;239:H289-H295.
27. Bratzler RL, Chisolm GM, Colton CK, Smith KA, Lees RS. The
distribution of labeled low-density lipoproteins across the rabbit
thoracic aorta in vivo. Atherosclerosis. 1977;28:289-307.
28. Schneiderman G, Pritchard WF, Ramirez CA, Colton CK, Smith
KA, Stemerman MB. Rabbit aortic medial thickness under relaxed
and specified simulated in vivo conditions. Am J Physiol. 1983;245:
H623-H627.
29. Schnitzer JJ, Morrel EM, Colton CK, Smith KA, Stemerman MB.
Absolute quantitative autoradiography of low concentrations of
[125I]-labelled proteins in arterial tissue. J Histochem Cytochem.
1987;35:1439-1450.
30. Cornhill JF, Roach MR. Quantitative method for the evaluation of
atherosclerotic lesions. Atherosclerosis. 1974;20:131-136.
31. Zar JH. Biostatistical Analysis. 2nd ed. Englewood Cliffs, NJ:
Prentice-Hall; 1984.
32. Nielsen LB, Nordestgaard BG, Stender S, Kjeldsen K. Aortic
permeability to LDL as a predictor of aortic cholesterol accumulation in cholesterol-fed rabbits. Arterioscler Thromb. 1992;12:
1402-1409.
33. Tompkins RG. Quantitative analysis of blood vessel permeability
of squirrel monkeys. Am J Physiol. 1991;260:H1194-H1204.
34. Truskey GA, Colton CK, Smith KA. Quantitative analysis of
protein transport in the arterial wall. In: Schwartz CJ, Werthessen
NT, Wolf S, eds. Structure and Function of the Circulation, III. New
York, NY: Plenum Publishing Corp; 1981:287-355.
35. Cornhill JF, Roach MR. A quantitative study of the localization of
atherosclerotic lesions in the rabbit aorta. Atherosclerosis. 1976;23:
489-501.
36. Kratky RG, Ivey I, Rogers KA, Daley S, Roach MR. The distribution of fibro-fatty atherosclerotic lesions in the aortae of caseinand cholesterol-fed rabbits. Atherosclerosis. 1993;9:121-131.
37. Lin SJ, Jan KM, Schuessler G, Weinbaum S, Chien S. Enhanced
macromolecular permeability of aortic endothelial cells in association with mitosis. Atherosclerosis. 1988;73:223-232.
38. Lin SJ, Jan JM, Chien S. Role of dying endothelial cells in transendothelial macromolecular transport. Arteriosclerosis. 1990;10:
703-709.
39. Penn MS, Saidel GM, Chisolm GM. Vascular injury by endotoxin:
changes in macromolecular transport parameters in rat aortas in
vivo. Am J Physiol. 1992;262:H1563-H1571.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Characterization of sites with elevated LDL permeability at intercostal, celiac, and iliac
branches of the normal rabbit aorta.
R A Herrmann, R A Malinauskas and G A Truskey
Arterioscler Thromb Vasc Biol. 1994;14:313-323
doi: 10.1161/01.ATV.14.2.313
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1994 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/14/2/313
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/