© 1997 American Heart Association, Inc.
Articles |
Cellular Dynamics in Early Atherosclerotic Lesion Progression in White Carneau Pigeons
Spatial and Temporal Analysis of Monocyte and Smooth Muscle Invasion of the Intima
From the Department of Pathology, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC.
Correspondence to W. Gray Jerome, PhD, Department of Pathology, Bowman Gray School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157-1092.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract The early stages of atherosclerosis are characterized by macrophage invasion of the intimal space and proliferation of smooth muscle cells (SMCs). In this study, we characterize the timing and location of each of these processes. Dividing cells of the developing lesions were labeled with [3H]thymidine at different times during lesion progression. Comparable stages of lesion development were analyzed, with the only variable being the length of time between labeling and necropsy. In this way, the mechanism by which lesion composition changes could be inferred from changes in the distribution of label over time. When cells were labeled just before necropsy, serial sections showed monocyte influx predominantly occurring at lesion edges and resulting in a labeling index of 27%. Near the lesion edge, medial SMCs were not stimulated to proliferate. In contrast, in more central areas of the lesion, monocyte influx and intimal SMC proliferation were minimal. Medial SMC proliferation in central regions, however, was high, with a labeling index of 20% in the upper media. Over time, proliferating medial SMCs migrated into the intimal lesion but did not migrate laterally to any great extent. This migration pattern maintained precise subdomains during lesion development, with each domain representing a different stage of lesion development. The edge of lesions thus represents very early events, central regions indicate later events, and intermediate regions provide information about stages between these extremes.
Key Words: atherosclerosis • atherogenesis • smooth muscle • macrophages
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Atherosclerotic lesion development is complex, involving interactions of cells in the artery wall with each other, extracellular tissue, and blood components. Both human and animal studies have broadly defined the cellular events that constitute lesion development, and it has been shown that most atherosclerotic plaques begin as fatty streaks.1 2 The earliest detectable form of fatty streak consists predominantly of lipid-engorged macrophage foam cells.2 3 The formation of these macrophage foam cells begins with the invasion of blood monocytes into the artery wall.4 5 Within the artery these monocytes differentiate into macrophages and accumulate lipid, taking on the foamy appearance characteristic of lesion cells.6 Some but not all fatty streaks progress, becoming more fibrous until a complicated occlusive plaque has evolved.7 8 9 The impetus for initiating the transition from fatty streak to complicated plaque is not known, but the hallmark features of the transition are the progressive increase in the number of smooth muscle cells (SMCs) within the arterial intima and a resultant accumulation of extracellular fibrous connective tissue.9 The SMCs originate in the medial layers underlying the lesion10 or in preexisting intimal "cushions."11
The dynamics of SMC migration into the arterial intima has been the focus of considerable research. Balloon injury to arteries results in an orderly progression of events that involves modulation of SMCs to a proliferative phenotype, followed by migration of these cells from the media into the intima.12 13 Studies of SMC proliferation in atherosclerotic arteries are generally consistent with data from balloon-injured arteries.14 Rosenfeld and Ross,15 who studied the timing of proliferation in rabbit atherosclerosis, suggested that SMC proliferation is minimal in early lesions but becomes more pronounced as lesions progress.15 The stimuli for SMC proliferation and migration in atherosclerosis are not well understood but undoubtedly involve multiple factors,16 including many macrophage products.17
The present study extends previous investigations of atherosclerosis in pigeons by exploring the dynamics of cell proliferation and migration during the early stages of lesion transition. The precise timing of lesion progression and the lack of preexisting intimal cell masses in pigeon arteries provided a unique model for exploring early cellular events in atherogenesis.3 18 The studies presented herein document that although monocyte adherence and migration into the lesion occur preferentially at lesion margins, SMC proliferation is found only in the more central regions of the lesion. On the basis of the predictable timing of lesion progression, we conclude that the stimulus for SMC proliferation occurs only after the macrophage–foam cell component of the lesion is well established.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
General Procedures
The animals in this study were randomly bred White Carneau (WC) pigeons from a closed colony at the Pigeon Resource at Bowman Gray School of Medicine. The birds were housed in flight pens, and all veterinary care and use followed the guidelines approved by the institutional animal use committee. The program is certified under provisions of the American Association for Accreditation of Laboratory Animal Care. WC pigeons were chosen for these studies because of the unique predictability of the site and timing for initiation and progression of atherosclerosis in this species.3 18
Five pigeons were assigned to each of six groups. Three groups represented controls and received a normal, cholesterol-free, pigeon pellet diet. The other three groups were designated experimental. At 8 weeks of age, the experimental animals were switched from a control to an atherogenic diet composed of pigeon pellets supplemented with 0.4% cholesterol and 10% lard. This diet was selected to produce lesions with nascent intimal SMC involvement after 20 weeks. At intervals, blood samples were taken from both experimental and control pigeons for plasma cholesterol analysis by the AutoAnalyzer II method.19
To label cells in the S phase of the cell cycle, animals were injected
with [3H]thymidine over a 6-day period. The labeling
regimen is diagrammed in Fig 1
and consisted of daily
injections of 150 µCi [3H]thymidine into the axillary
vein. This protocol had proven effective in previous
studies.20 21 The daily injection schedule was used to
maximize labeling and minimize stress to the pigeons. One control and
one diet group, designated the 1-week group, received injections during
the week before necropsy. Animals in these two groups received the
final injection the day before necropsy. The other diet and control
groups received [3H]thymidine injections that were begun
either 6 or 11 weeks before necropsy and were completed either 5 or 10
weeks before the animals were euthanized. The 1-week group was studied
to assess labeling efficiency at the time of necropsy, whereas the 6-
and 11-week groups provided information on cell changes between the
time of labeling and necropsy.
|
Blood samples were taken from two experimental and two control pigeons at selective intervals after the [3H]thymidine injection regimen. These samples were used to assess how long the labeled leukocytes remained in the circulation. Blood smears were made on glass slides, stained with Wright's stain, coated with Kodak NTB-2 photographic emulsion, and exposed in light-sealed boxes at 4°C using standard autoradiographic techniques.22 After a 4-week exposure, the slides were photographically developed and the labeling index determined as the number of labeled leukocytes divided by the total number of leukocytes in 10 arbitrarily chosen fields.
Histology
At necropsy, all pigeons were given heparin (300 U) and sodium
pentobarbital (0.45 mL) by intravenous injection. After
anesthesia was induced, the chest cavity was opened and the
animals exsanguinated by cardiac puncture. The vascular system was
flushed via intraventricular pressure perfusion
with 4% paraformaldehyde/0.5%
glutaraldehyde in 0.1 mol/L sodium phosphate buffer at
pH 7.2. Perfusions were performed at normal pigeon body temperature
(41°C). After fixation, the thoracic aorta was excised and a section
of lower thoracic aorta (from a point just caudal to the origin of the
celiac artery to a point 6 mm cranial to the celiac artery) was
removed and placed in fresh fixative overnight. This area
represents the area of highest predilection for
atherosclerosis, and disease progression at this site
has been well characterized.3 23 Importantly, the diet
period used in these studies generally produced a single, large lesion
of predictable size.
The next day the aortas were washed, dehydrated, and embedded in paraffin. The entire 6-mm embedded piece of aorta was serially cross-sectioned beginning at the cranial edge of the specimen. The serial sections were mounted in sequence on glass slides (20 sections per slide) and the slides numbered consecutively. After the slides were cleared and lightly stained with hematoxylin and eosin, they were coated with Kodak NTB-2 photographic emulsion and exposed for autoradiography.20 21 Blank glass slides were also dipped in the photographic emulsion as controls. Immediately after they were coated, two blank slides were developed to determine background levels of exposure. Another two blank slides were exposed to light and developed to determine that the emulsion was sound and that it had uniformly coated the slide. The remaining blank slides were stored in a manner identical to that of the slides with the arterial sections. Just before the arterial slides were developed, blank slides (unexposed and exposed to light) were developed as described above to serve as additional controls on emulsion integrity.
In addition to those pigeons included in the autoradiography study, two pigeons were maintained on the 0.4% cholesterol diet for 20 weeks to assess lesion ultrastructure. Arteries from these animals were fixed by vascular perfusion with 0.4% glutaraldehyde in 0.1 mol/L cacodylate buffer. Perfusion fixation and subsequent processing of the arteries followed the procedures that we have described previously.24 In brief, after perfusion fixation, the thoracic aorta was excised and enzyme-cytochemically stained to demonstrate acid phosphatase, with ß-glycerophosphate as the substrate and lead as the capture reagent.25 After enzyme cytochemistry, arterial samples were embedded in epoxy resin and thin sectioned for electron microscopy.
Morphometry
Lesions for quantitative analysis were divided into
three regions as follows. The serial sections were viewed in sequence
until a section that contained a lesion was encountered. This and
subsequent sections that encompassed the first 80 µm of lesion
(16 sections) were designated the lesion edge. Based on previous
calculations, this distance represents the approximate length
of lesion growth in 1 week under the dietary conditions employed
(unpublished observations). The 80 sections directly behind the lesion
edge represent the intermediate region. Together the edge and
intermediate regions constituted 
6 weeks of lesion growth. The
remaining lesion to a point through the midline was classified as the
middle lesion. The designated regions were chosen to encompass lesion
growth during the preceding week, 6 weeks, and 11 weeks, respectively.
This timing corresponds to the various labeling protocols employed. The
lateral 40 µm of lesion was also considered edge and was not
included in the analysis.
The composition of the different regions as well as nonlesion regions was assessed by using standard point-counting stereology.26 Occasional section defects precluded use of rigorously randomized sections. Instead, six sections from the edge region (first 16 sections) were chosen for analysis in an attempt to distribute the analyses across the entire region, with at least one intervening section between each analyzed section. Eight sections were analyzed for nonlesion, intermediate, and middle regions of each lesion. Again, the sections were chosen to encompass the entire region and provide even spacing between analyzed sections.
Regions were analyzed at 1000x on an Olympus BH-2 light microscope equipped with a special ocular grid composed of intersecting horizontal and vertical lines of regular, defined spacing. The intersections of grid lines were used as sampling points. In this way the volume fraction of lesion occupied by macrophage foam cell, SMC, and extracellular material was determined. In addition to volume fraction analysis, we also determined the average number of cellular profiles in a given area of section. Five sections were chosen arbitrarily for these analyses. The number of cells in the section was determined and the ocular grid used to compute the area of lesion represented in the section. Because endothelial cells (ECs) formed a monolayer covering the lesion, the number of cell profiles per length rather than area was determined. This provided a more representative analysis of changes in EC type. The number of ECs per unit length was determined for the same sections that were used for area analysis. The number of adherent monocyte profiles was also indexed to endothelial length. Statistical analysis of the data from volume fraction and area measurements suggested that our selection criteria for both of these procedures provided a representative sample of lesion composition, with the SE being <10% of the mean.
The labeling index of different cell populations was computed as the
number of [3H]thymidine-labeled cells divided by the
total number of profiles counted. Positive labeling was defined
conservatively. Determining the criteria for positive cells first
involved establishing the "average" label, which was accomplished
by calculating the average nuclear area. A circle of that diameter was
then scanned over areas of acellular sections, and the number of
autoradiographic silver grains within the circle at each
point was determined. After 300 such determinations were made, the
maximum number of grains within a single determination was taken to
represent the maximum background per nuclear area. From 10
slides (1 each from 10 separate arteries), the maximum count was 3
grains. Any cell having twice this background (ie, any cell with 
6
silver grains) was counted as positive. In practice, most positive
cells had 10 or more grains.
Labeling of medial SMCs was not uniform but was significantly higher in the upper media close to the first elastic lamina. Therefore, to provide an accurate representation of changes that occurred in the media, it was subdivided into upper and lower areas. The upper media represented the first seven layers of SMCs and the lower media the remaining SMC layers. In all cases the lower media had a very low labeling index (<1%). The lower-media index was equivalent in control and cholesterol-fed animals and was not significantly different with respect to time of injection. For this reason, only the data for the upper media are reported in this article.
Statistics
Data from each animal were pooled and the mean used to
represent that animal. Data were further grouped according to
treatment (identified in the tables and figures), and then the mean,
SD, and SEM were determined. Comparisons between groups were made with
Duncan's new multiple-range test27 after ANOVA. Whenever
percentage data were used, they were first transformed using the arcsin
function.28 Two means were considered significantly
different if the probability of a type I error was <.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Die-away of Labeled Cells From Plasma
The loss of labeled cells from the circulation after completion of 1 week of [3H]thymidine injections was rapid and approximated first-order decay (Fig 2
). Because no
differences were seen between diet and control animals, the data from
these two groups were combined. Immediately after 1 week of injections,
52±4% of circulating leukocytes contained radioactive thymidine. The
half-life of labeled cells in the plasma, however, was only 5.6 days,
and by 3 weeks labeled cells accounted for <2% of the total
leukocytes in circulation. Thus, animals injected 6 or 11 weeks before
necropsy had virtually no labeled cells in the circulation at the time
of euthanasia.
|
Plasma Cholesterol Concentrations
Pigeons that ate the cholesterol-free pellet diet
maintained an average cholesterol concentration of 5.4
mmol/L. Addition of 0.4% cholesterol to the diet, however,
produced a rapid rise in average plasma cholesterol levels,
to 20.3 mmol/L within 2 weeks (Fig 3). Continuation
of the diet beyond 2 weeks produced continued increases in plasma
cholesterol levels. After 20 weeks of the diet, the average
cholesterol concentration in plasma was 34 mmol/L. For
simplicity, Fig 3 presents the average cholesterol
concentration of all pigeons that were fed cholesterol,
regardless of when they received thymidine injections. Separate
analyses of the individual injection groups, however, showed no
significant (P<.05) differences in plasma
cholesterol concentrations between groups. For instance,
after 10 weeks (the time at which the first group of animals finished
receiving thymidine injections), the average plasma
cholesterol concentrations were 23.9±4.0, 24.4±1.8, and
26.5±5.8 mmol/L for the cholesterol-fed groups that
received [3H]thymidine 11, 6, and 1 week before necropsy.
At the time of necropsy, these values were 31.4±5.0, 37.6±10.4, and
33.0±8.5 mmol/L, respectively, for the three groups.
|
Lesion Structure
Cholesterol feeding induced small atherosclerotic
lesions in the aorta just proximal to the celiac bifurcation. In each
case, the principal lesion was 3 mm long (cranial to caudal).
The similarity of lesions in this area is consistent with our
previous observations24 and facilitated quantitative
comparisons among the different groups.
Although these lesions occupied a larger volume, the lesion characteristics were similar to those of early lesions that we had described previously.24 29 The lesions consisted primarily of macrophage foam cells with some SMCs at the base of the intima. Unlike that in the cholesterol-fed animals, intimal accumulation of cells or extracellular material was not seen in animals that received the normal control diet.
Fig 4 is a representative
autoradiographic section through the intermediate region of
a lesion from a pigeon that received the
cholesterol-containing diet. As illustrated in the figure,
large foam cells with rounded nuclei were the predominant cell type in
this lesion. In previous studies we identified these cells as
macrophage foam cells.3 24 In early lesions in
pigeons, lymphocytes and heterophils (the pigeon equivalent of a
neutrophil) are not found. However, intimal SMCs, recognized by their
narrow shape and elongate nuclei, were located at the base of the
lesion close to the first elastic lamina. Many macrophage foam
cell nuclei contained radioactivity, indicating that these cells were
derived from those that had incorporated thymidine 6 weeks before
necropsy. In contrast, labeling was not found in the intimal or medial
SMCs, indicating that these cells were quiescent at the time of
labeling.
|
Transmission electron microscopic images of the intermediate region
confirmed the macrophage character of the foam cells and
documented that a significant amount of lipid in macrophage
foam cells was contained within large, swollen lysosomes (Fig 5). The lysosomal character of lipid accumulation was
confirmed by the presence of acid phosphatase. As we demonstrated
previously for younger lesions, macrophage foam cells at the
lesion edge also contained significant amounts of lipid, but this lipid
was found almost exclusively in cytoplasmic droplets rather than within
lysosomes (data not shown).
|
Transmission electron microscopy also confirmed the localization of
intimal SMCs at the base of the lesion, near the first elastic lamina
(Fig 5
). Notably, the intimal SMCs lacked significant lipid deposits.
This lack of SMC lipid accumulation has been described as a
consistent feature of early lesions in
pigeons.24
Lesion Composition
Consistent with the similarity in plasma
cholesterol concentrations between groups, the lesions
formed in all three cholesterol-fed groups were similar
with respect to size, appearance, and composition. This was to be
expected, since all lesions were of equivalent age. Lesion composition
was analyzed from light microscopic preparations in two ways.
First, the volume of lesion occupied by macrophage foam cells,
SMCs, and extracellular material was determined. As shown in the
Table, distally from the lesion edge toward the middle
of the lesion, the composition of the lesions changed. Specifically,
SMC and extracellular material made up increasing percentages of the
lesion volume. Consistent with our previous
reports,3 macrophage foam cells constituted >90%
of the volume of the lesion edge but only 68% to 69% of the volume of
more central areas of the lesion.
|
Despite the differences among various intralesion regions, significant
(P<.05) differences were not detected in lesion composition
between animals that received injections at different times during
lesion progression. Thus, at the lesion edge, macrophage foam
cells accounted for 91% to 92% of the lesion volume in all three
injection groups. Extracellular material accounted for the remaining
volume. Similarly, the volume composition in the middle of lesions was
constant among the three groups (Table). This finding confirmed that
differences in thymidine regimen did not affect lesion progression. In
contrast to those from cholesterol-fed animals, arteries
from pigeons that received the cholesterol-free diet did
not contain cells within the subendothelial space and
extracellular material made up <0.1% of the intimal volume.
In addition to measuring volume components, we estimated the number of
cells per unit length (endothelial cells) or area
(macrophage foam cells and SMCs) (Fig 6). As
expected, this alternative measure confirmed the lack of significant
differences (P<.05) in lesion composition between groups
that were injected 11 weeks, 6 weeks, or 1 week before necropsy. The
numerical measurements, however, also indicated that the number of SMCs
in the upper media remained constant at 0.58 cells per 10
µm2. The lower media also maintained a constant volume
and number of cells, and media volume did not differ under
atherosclerotic lesions. Thus, even if medial SMCs migrated to the
intima, they were replaced by new SMCs in the media.
|
Presence of Nuclear Profile in Sections
Since the ability to determine a labeled cell was based on the
presence of a nucleus in the section, we computed the percentage of
cells with nuclei in a series of 10 sections through the middle of the
lesions. Seventy-eight percent of foam cells had their nuclei included
in the section. Nuclei for adherent monocytes, ECs, and SMCs were seen
in 90%, 92%, and 93% of cells, respectively.
EC Labeling Indices
The rates for cell turnover and migration into lesions were
inferred from differences in the labeling patterns of cells in animals
that were injected 1 week before necropsy compared with those animals
that received injections either 6 or 11 weeks before necropsy. Fig 7 summarizes the labeling index of ECs.
Consistent with our previous observations21 that
when EC turnover was determined by thymidine administration just before
necropsy, the labeling index of cells over the middle of the lesions
was significantly (P<.05) lower than that over the lesion
edges. Mean labeling was also significantly less over intermediate
areas of the lesion than at the edge. In contrast, when animals were
injected 6 weeks before necropsy, the labeling index of ECs was highest
over intermediate areas. As discussed later, this difference was
consistent with the hypothesis that the edge of the lesion at
the time of labeling matured into the intermediate region by the time
of necropsy and that the region maintained its high labeling index.
Similarly, lesion edge at necropsy would have been a nonlesion region
at the time of [3H]thymidine injection and so lacked the
radiolabel. When 11 weeks elapsed between injection and necropsy, the
ECs were not labeled, suggesting that the turnover rate of ECs was
sufficient to dilute the "label" after 11 weeks to a point below
our minimum criterion for counting the cell as positively labeled. It
should be pointed out that some ECs in the 11-week sample were labeled
with [3H]thymidine, but these cells generally had <3
silver grains and therefore were not considered positive. In nonlesion
areas of all samples and control arteries, endothelial
labeling was <1% of cells.
|
Monocyte/Macrophage Labeling Indices
Illustrated in Fig 8 is the labeling summary of
monocytes that adhered to the endothelial surface and
of macrophage foam cells within different regions of the
lesion. When animals were injected with thymidine during the week
before necropsy, 50% of adherent cells were labeled. This was
similar to the proportion of labeled cells in the circulation (see Fig 2) and suggests that labeled cells were as likely to adhere as
unlabeled cells. However, even though the percentage of labeled,
adherent cells was equivalent for all regions, the total number of
adherent cells was not the same over all areas. The lesion edge had six
times as many adherent cells as the middle region, while the
intermediate region exhibited intermediate levels of adherence (1.5
times as many adherent cells as the middle region).
|
This increased adherence to lesion edges was consistent among
the injection groups. However, accompanying the rapid loss of labeled
cells from the plasma, the labeling indices for adherent cells in
animals injected 6 or 11 weeks before necropsy were significantly
(P<.05) lower (Fig 8). In fact, at 11 weeks no adherent
cells were labeled.
Consistent with the extensive cell adherence at lesion edges
and the high labeling indices for animals injected just before
necropsy, 28% of macrophage foam cells within the lesion edge
were labeled (Fig 8). By comparison, the intermediate and middle
regions showed statistically (P<.05) lower labeling
indices: 5% and 3%, respectively. This suggests that the edge had a
higher rate of monocyte influx than did the more distal regions. When
animals were given [3H]thymidine 6 or 11 weeks before
necropsy, however, the area of highest foam cell labeling was shifted
progressively away from the edge (Fig 8). In the 6-week group, there
was no labeling of intimal foam cells at the edge, whereas 24% of
cells in the intermediate region were labeled. The 24% labeling was
equivalent (P<.05) to that at the edge in the 1-week group.
The most straightforward explanation is that as the lesion
progressively expands, the new lesion edge is formed from a previous
nonlesion area. Therefore, by the time of necropsy, the labeled lesion
edge had become part of the intermediate area. This supposition is
further strengthened by analysis of animals that were injected
11 weeks before necropsy. In these animals, the area of highest
labeling was shifted even farther back, to the middle region. In
addition, the edge and intermediate regions did not exhibit labeling,
suggesting they were nonlesion areas at the time of labeling. The high
labeling in the middle region was statistically (P<.05)
different from that in the other regions.
SMC Labeling Indices
Fig 8 also summarizes the labeling pattern for SMCs in the intima
and upper regions of the underlying media. Importantly, there is very
little labeling within the intimas of animals that were injected just
before necropsy. This indicates that intimal SMCs at this stage of
atherogenesis were not a rapidly dividing population of cells. In
contrast, medial SMCs had a high degree of labeling, with about 1 in 5
cells in the upper media being labeled. However, these high labeling
indices were restricted to areas that underlay the intermediate and
middle regions of the lesion. Labeling in these two areas was not
statistically different. In contrast, normal areas of the artery and
the lesion edge were not labeled. Control arteries also had minimal SMC
proliferation, with <1 in 100 cells in the upper media with
[3H]thymidine. In comparison, when WC pigeons were
injected 6 weeks before necropsy, medial SMCs that underlay the
intermediate areas lacked labeling; thus, only areas beneath the middle
parts of the lesion contained labeled SMCs. Consistent with our
other observations, this lack of labeling under the edge and
intermediate regions is what one would expect after 6 weeks of lesion
progression, since (due to lesion progression) the lesion edge and
intermediate regions would have been derived from areas that had not
been labeled at the time of injection. In this way the medial areas
that originally underlay the intermediate areas at the time of labeling
would now be positioned beneath more central regions.
Lesion progression would also account for the lack of intimal SMC labeling in intermediate regions at 6 weeks. Since at the time of labeling the current intermediate region was the lesion edge, all SMCs that entered during the 5 weeks of progression would have come from an unlabeled population of cells. Most noteworthy, however, is the observation that in the 5 weeks between the time of labeling and necropsy, the middle part of the 6-week lesions had gained significant numbers of labeled SMCs. Since the 1-week pigeons lacked turnover of intimal SMCs, we can conclude that intimal labeling is the result of SMC migration from the media to the intima. When the 6-week animals were compared with the 1-week animals, there was a 40% decrease in medial labeling and a 13-fold increase in intimal labeling in the middle area. These changes were statistically significant (P<.05).
Continued growth and transformation of the lesion were seen in animals injected 11 weeks before necropsy. In the middle part of the lesion, there was a continued, statistically important decrease in medial SMC labeling and an increase in labeled intimal SMCs. This change was consistent with continued migration of SMCs from the media and to the intima. However, the SMC labeling was limited to the middle areas. This suggested very little lateral migration of SMCs, either within the media or within the intimal lesions.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The dynamics of lesion progression have been the subject of several studies, and a general description of lesion development has been published.2 However, the subtle details and timing of changes in lesion morphology are still not well understood, particularly for the earliest stages of atherogenesis. The plaques formed in pigeons are very similar to those found in humans,18 thus making the WC pigeon a useful model. In addition, the predictability of lesion initiation and progression at the celiac bifurcation in aortas from WC pigeons makes this strain a particularly useful animal model for quantitatively investigating early atherosclerosis development.24 The remarkably similar size and composition of all lesions analyzed, coupled with our staggered-injection protocol, allowed us to explore and confirm subtle changes in labeling patterns. Previously, we have exploited the predictability of pigeon lesion formation to show that invasion of the intimal space by monocytes precedes SMC involvement.3 A similar procession of events has been inferred from studies in humans30 and a variety of animal models, such as rabbits31 and nonhuman primates.1 In the study reported here, we extend previous observations by investigating the dynamics of cell turnover and migration at the time that SMCs begin to invade the intima.
Since the intimal space in the region of study is normally devoid of cells, the influx of cells can be easily quantified. Changes in labeling pattern additionally provide information about cell turnover and lesion remodeling. Our data confirm earlier observations from both our laboratory and others, which have shown that events that occur at the developing edge of the lesion are very different from those in more central regions.3 15 21 32 In fact, in pigeons (the only species for which a large number of equivalent nascent lesions have been quantified), the edge of the maturing lesion is compositionally similar to a newly initiated lesion.3 On this basis, we hypothesized that the edge of mature lesions could be used to investigate properties of nascent lesions.
The sequential analyses from the current study add credence to this hypothesis and extend our earlier observations by showing that the region between the edge and middle has characteristics intermediate to the two extremes. The most obvious interpretation of the lesion changes is that edges represent newly initiated areas of atherosclerosis. As the lesion expands laterally, these edges pass through the intermediate stage, finally gaining the characteristics of a more mature lesion. Thus, lesion development can be recapitulated by comparing the edge with the intermediate and central areas. Although this is true for the relatively early lesions studied here, it remains to be seen whether very complex lesions that undergo multiple events (such as rupture, thrombosis, and revascularization) will be amenable to this sort of analysis.
The equivalence of lesion edges and lesion initiation is also suggested by maximal monocyte adherence over the lesion edge. This is similar to what we have shown for nascent lesions.3 21 Of course, there was also adherence over the more central regions, but the levels were much reduced compared with the edge. From the high labeling of macrophage foam cells at the edge in 1-week animals, it is clear that edge adherence was accompanied by infiltration of monocytes into the intima. In contrast, there appeared to be only minimal influx of monocytes in regions away from the edge. This is best seen by comparing the edge in 1-week animals with the intermediate region in 6-week animals. Both sites had equivalent labeling indices. However, by 6 weeks circulating monocytes were no longer labeled. Thus, since the intermediate region after 6 weeks occupied the space that was the lesion edge during labeling, equivalence of the label suggests that either entrance of new (unlabeled) monocytes decreased during the 6 weeks or that substantial replication of labeled intimal macrophages occurred and offset the influx of unlabeled monocytes. Although there is evidence that intimal macrophages can divide, the rates for such division are moderate.15 33 34 In addition, when edge are compared with intermediate regions, the slight increase in intimal height (data not shown) could be accounted for by SMC influx, which argues against significant addition of new monocytes. Thus, although specific testing is required, the most probable explanation is that monocyte migration into the lesion was minimal in areas away from the lesion edge.
There was a diminution of foam cell labeling in the middle area of the 11-week group compared with the intermediate area of animals labeled 6 weeks before necropsy. Although this could be due to an influx of unlabeled cells, such an interpretation is inconsistent with the comparison of 1- and 6-week animals. If labeled cells had preferentially died or emigrated, there would also have been a decrease in the percent of labeled cells. We have no evidence, however, that labeled cells are more prone to cell death or migration than are unlabeled cells. The most probable explanation for the decreased labeling in the middle of the lesion is simply that there were preexisting middle areas of the lesion before labeling. Thus, the percent of labeled cells was diluted by preexisting unlabeled foam cells.
In contrast to monocyte influx, which was most active at the edge, SMC proliferation was confined to the intermediate and middle lesion areas. The analysis of SMC proliferation and migration provided the most novel and clinically important observations of this study, since SMC invasion marks the beginning of lesion transition from the fatty streak to the fibrofatty lesion. It is significant that we did not observe SMC proliferation under the lesion edge or in nonlesion regions. We interpret this observation as an indication that stimulation of SMC proliferation does not occur during lesion initiation. In fact, by comparing labeling at each time point, it would appear that at least 6 weeks of lesion development are required before significant medial SMC proliferation occurs. Moreover, the lack of labeling in intimal SMCs in animals injected 1 week before euthanasia suggests that intimal SMCs did not divide rapidly even in regions where medial SMC division was extensive. From this observation we can infer that the increases in intimal SMC labeling in animals that were injected 6 or 11 weeks before euthanasia were the result of an influx of labeled cells from the media.
The timing of SMC proliferation in atherosclerosis has been somewhat controversial and may depend on the method of lesion initiation. Studies by Rosenfeld and Ross15 of diet-induced atherosclerosis in rabbits suggest that SMC proliferation is minimal in early lesions but becomes more pronounced only as lesions progress. Paradoxically, Gordon and coworkers33 reported low proliferation rates in advanced human atherosclerotic plaques. An explanation for this discrepancy may lie in the observation that SMC proliferation occurs rapidly after initiation of a stimulus but then decreases even with continued stimulation.13 14 Our data confirm that although SMC proliferation occurs early in lesion development, it is not the earliest event.
Regardless of timing, our data suggest that there is very little lateral movement of SMCs, because as the intermediate area matured, the label remained in this maturing area and did not spread into the new intermediate region. In comparison, the new intermediate region in 6-week animals had equivalent numbers of SMCs, but these apparently were derived from unlabeled medial SMCs directly below this region rather than from labeled SMCs beneath the new midlesion. This suggests that the factors that stimulate proliferation and migration are localized to a small area and do not diffuse laterally to any great extent.
The discrete localization of SMC proliferation also suggests that the factors involved in stimulating proliferation were not distributed uniformly throughout the lesion. Rather, they appear to remain localized within relatively small areas. In the past decade, many factors have been implicated as SMC mitogens and/or chemoattractants.16 The most studied of these factors have been growth factors, principally fibroblast growth factor and platelet-derived growth factor,35 36 and cytokines, such as interleukin-1.37 On this basis, a central role for macrophage foam cells has been posited, since all the putative stimulatory agents are secreted either by macrophages or by cells that are influenced by macrophages. In addition, lipid-laden macrophages have been shown to stimulate SMC proliferation.38 In our study, the most obvious structural difference in areas of SMC proliferation was the pronounced lysosomal lipid accumulation in macrophage foam cells. This lysosomal accumulation corresponded spatially with areas of high SMC proliferation, and we had previously shown a temporal association between these two phenomena.24 One interpretation of these data, which warrant further investigation, is that lysosomal lipid accumulation enhances the ability of macrophages to stimulate SMC proliferation. Other investigators have also suggested that sequestering of lipid within a lysosomal pool could promote further disease progression.39 40
The stimulation of SMC mitosis appeared to be confined to areas proximal to the intima, since deeper layers of the media showed no such cell proliferation. The narrow localization of stimulatory activity is consistent with the short-lived activity of most cytokines and growth factors.41 However, we cannot rule out the possibility that liberation of stored lipid or trapped lipoproteins could also directly or indirectly act to stimulate SMC proliferation.42 43
The labeling indices of foam cells at the lesion edge in 1-week animals and the intermediate region of 6-week animals are high compared with some autoradiography studies of other species. Likewise, SMC labeling in the middle regions of 11-week animals is somewhat elevated compared with previous reports. We do not believe that this indicates that pigeon lesions have accelerated levels of monocyte influx or SMC proliferation. Instead, the increased labeling is probably related to two factors that differentiate our study from others: (1) the early stages of lesion development analyzed herein and (2) subdivision of the lesion into discrete regions. Although not directly quantified, labeling appeared higher in superficial regions and near the lesion edge in other studies.32 44 Cavallero and Turolla44 have also documented that early lesions have higher labeling than do more advanced plaques. If averaged over the entire lesion (rather than subdivided into edge, intermediate, and middle regions), the labeling indices of the lesions studied here would average 6.4%. This level of labeling is similar to that seen in other studies.32 44 45
One of the most puzzling observations of this study is the presence of labeled monocytes on the surface of the lesion 6 weeks after injection. Although the labeling indices were quite low (<1%), this finding is still surprising, given that we found no labeled cells in the circulation at this time. We advance three possible explanations for this phenomenon. First, there could be a small subset of labeled cells that persisted in the circulation but were not detected. Alternatively, some cells may have remained adherent to the surface for some weeks. Both of these explanations, however, seem unlikely. A more plausible explanation is suggested by the work of Gerrity,46 who documented eggression of macrophage foam cells from plaques. In Gerrity's and other more recent studies, some eggressing foam cells can be seen extruding through the endothelium, with part of the cell lying on the endothelial surface.46 47 In pigeons, macrophage eggression from lesions does not appear to occur as readily as in other species. However, Landers and coworkers48 have shown that some macrophage foam cells can bulge out between ECs into the arterial lumen. These may represent eggressing macrophages or foam cells caught between the intimal space and the arterial lumen. Importantly, the lipid-filled areas of these cells were within the plaque, but most often the nuclear portion of the cytoplasm was located within the arterial lumen. In our study, the nuclear portion of the cell, which generally lacked lipid, would not be distinguishable from adherent monocytes and if labeled would be counted as a labeled monocyte. This may explain the few labeled cells detected in the 6-week group.
Two issues warrant discussion with respect to interpretation of the
labeling pattern. First, recent reports have documented both
necrosis49 and apoptosis50 51 52 in
atherosclerotic lesions. Such cell death could decrease labeling within
the intima. Although the current study does not directly address this
question, in this and other studies of early pigeon lesions we have not
found morphological evidence of cell death. In contrast, in later
lesion stages, we have seen apoptotic cells. This observation
is consistent with other studies that have shown the greatest
amount of cell death in older, more advanced lesions.50 51 52
A second issue involves incorporation of [3H]thymidine
into nuclei due to DNA repair. This would create false-positive
staining. However, only limited amounts of thymidine are incorporated
during DNA repair. To avoid including false-positives, we set our
labeling criteria high (6 silver grains). For this reason, our
labeling indices probably underestimate cell turnover, since some
dividing cells probably lacked enough label to be included. Thus, for
both cases (repair and cell death), the errors introduced would only
subtly alter labeling indices in these early lesions. Since the
difference in labeling between regions was generally large, neither
instance would alter the conclusions drawn in this study.
In summary, our investigations suggest that at least for the early stages of lesion development and transition studied here, one can infer disease progression from serial sections through different regions of lesion. The studies show that the distinct morphological domains within the atherosclerotic lesion are the result of different aspects of lesion progression. In particular, SMC proliferation is localized both temporally and spatially to discrete regions of the developing plaque, and by inference, the factors that control SMC proliferation would be equally confined. In contrast, macrophage invasion of the intima occurs primarily at the edge of lesions. It will be of interest of course to see whether these aspects of lesion development maintain their spatial and temporal association as lesion progression continues to more complex stages.
|
Acknowledgments |
|---|
The studies were supported by Specialized Center of Research in Atherosclerosis grant HL14164, American Heart Association Grant 900709, and NIH Grant RO1 HL49148-01A2 (to W.G.J.). The authors thank the staff of the Lipid Lab of Bowman Gray School of Medicine for the cholesterol analyses, the staff of Micromed (Ken Grant, Stephanie Evans, and Paula Moore) and Cindy Cooke for technical assistance, and Bobbie Lindsay for preparation of this manuscript.
Received December 29, 1995; accepted June 4, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Faggiotto A, Ross R, Harker L. Studies of hypercholesterolemia in the nonhuman primate, I: changes that lead to fatty streak formation. Arteriosclerosis.. 1984;4:323-340.
2.
Stary H, Chandler A, Glagov S, Guyton J, Insull WJ,
Rosenfeld M, Schaffer S, Schwartz C, Wagner W, Wissler R. A
definition of initial, fatty streak, and intermediate lesions of
atherosclerosis: a report from the Committee on
Vascular Lesions of the Council on
Arteriosclerosis, American Heart
Association. Arterioscler Thromb.. 1994;14:840-856.
3. Jerome WG, Lewis JC. Early atherogenesis in White Carneau pigeons, II: Ultrastructural and cytochemical observations. Am J Pathol.. 1985;119:210-222. [Abstract]
4. Gerrity RG. The role of the monocyte in atherogenesis, I: transition of blood-borne monocytes into foam cells in fatty lesions. Am J Pathol.. 1981;103:181-190. [Abstract]
5. Lewis JC, Taylor RG, Jerome WG. Foam cell characteristics in coronary arteries and aortas of White Carneau pigeons with moderate hypercholesterolemia. Ann N Y Acad Sci.. 1985;454:91-100. [Medline] [Order article via Infotrieve]
6. Fowler S, Shio H, Haley NJ. Investigation of macrophage-like properties of aortic cell populations. Lab Invest.. 1979;41:372-378. [Medline] [Order article via Infotrieve]
7. Stary H. Changes in components and structure of atherosclerotic lesions developing from childhood to middle age in coronary arteries. Basic Res Cardiol. 1994;89(suppl 1):17-32.
8.
Cornhill J, Barrett W, Herderick E, Mahley R, Fry D.
Topographic study of sudanophilic lesions in
cholesterol-fed minipigs by image analysis.
Arteriosclerosis.. 1985;5:415-426.
9. Bocan T, Schifani T, Guyton J. Ultrastructure of the human aortic fibrolipid lesion: formation of the atherosclerotic lipid-rich core. Am J Pathol.. 1986;123:413-424.[Abstract]
10. Webster W, Bishop S, Geer J. Experimental aortic intimal thickening, I: morphology and source of intimal cells. Am J Pathol.. 1974;76:245-260. [Medline] [Order article via Infotrieve]
11. Thomas W, Reiner J, Florentin R, Scott R. Population dynamics of arterial cells during atherogenesis, VIII: separation of the roles of injury and growth stimulation in early aortic atherogenesis in swine originating in pre-existing intimal smooth muscle cell masses. Exp Mol Pathol.. 1979;31:124-144. [Medline] [Order article via Infotrieve]
12.
Chamley-Campbell J, Campbell G, Ross R.
Phenotype-dependent response of cultured aortic smooth
muscle to serum mitogens. J Cell Biol.. 1981;89:379-383.
13.
Clowes A, Schwartz S. Significance of quiescent
smooth muscle migration in injured rat carotid artery.
Circ Res.. 1985;56:139-145.
14. Kim DN, Schmee J, Ho H, Thomas WA. The turning off of excessive cell replicative activity in advanced atherosclerotic lesions of swine by a regression diet. Atherosclerosis.. 1988;71:131-142. [Medline] [Order article via Infotrieve]
15.
Rosenfeld M, Ross R. Macrophage and
smooth muscle cell proliferation in atherosclerotic lesions of WHHL and
comparably hypercholesterolemic fat-fed
rabbits. Arteriosclerosis.. 1990;10:680-687.
16.
Thyberg J, Hedin U, Sjolund M, Palmberg L, Bottger B.
Regulation of differential properties and proliferation of
arterial smooth muscle cells.
Arteriosclerosis.. 1990;10:966-990.
17. Ross R. Cell biology of atherosclerosis. Annu Rev Physiol.. 1995;57:791-804. [Medline] [Order article via Infotrieve]
18. Prichard RW, Clarkson TB, Goodman HO, Lofland HB. Aortic atherosclerosis in pigeons and its complications. AMA Arch Pathol.. 1964;77:244-257.
19. Rush RL, Leon L, Turrell J. Automated simultaneous cholesterol and triglyceride determination on the AutoAnalyzer II instrument. In: Barton EC, DeCross MJ, Erdich MM, eds: Technicon International Congress. Mt Kisco, NY: Futura Publishing Co; 1970:503-507.
20. Jerome WG, Lewis JC, Taylor RG, White MS. Concurrent endothelial cell turnover and leukocyte margination in early atherosclerosis. Scanning Elect Microsc. 1983:1453-1459.
21. Taylor RG, Lewis JC. Endothelial cell proliferation and monocyte adhesion to atherosclerotic lesions of White Carneau pigeons. Am J Pathol.. 1986;125:152-160. [Abstract]
22. Williams MA. Autoradiography and immunocytochemistry. In: Glauert AM, ed. Practical Methods in Electron Microscopy. Amsterdam, The Netherlands: Elsevier Biomedical Press; 1977.
23. Jerome WG, Lewis JC. Early atherogenesis in White Carneau pigeons, I: leukocyte margination and endothelial alterations at the celiac bifurcation. Am J Pathol.. 1984;116:56-68. [Abstract]
24. Jerome WG, Lewis JC. Early atherogenesis in White Carneau pigeons, III: lipid accumulation in nascent foam cells. Am J Pathol.. 1987;128:253-264. [Abstract]
25. Gomori G. Microscopic Histochemistry: Principles and Practice. Chicago, Ill: University of Chicago Press; 1952.
26. Weibel ER. Stereologic principles for morphometry in electron microscopic cytology. Int Rev Cytol.. 1969;26:235-302. [Medline] [Order article via Infotrieve]
27. Duncan DB. Multiple range and multiple F tests. Biometrics. 1955;11:1-42.
28. Zar JH. Biostatistical Analysis. Prentice Hall Inc; 1984.
29. Jerome WG, Lewis JC. Early atherogenesis in White Carneau pigeons: effect of a short-term regression diet. Exp Mol Pathol.. 1990;53:223-238. [Medline] [Order article via Infotrieve]
30. Stary H. Structure and ultrastructure of the coronary artery intima in children and young adults up to age 29. Atherosclerosis.. 1983;6:82-86.
31. Kunz J, Schreiter B, Schubert B, Voss K, Krieg K. Experimentelle Untersuchungen uber die Regeneration der Aortenendothelzellen: automatische und visuelle Auswertung von Autoradiogrammen. Acta Histochemic.. 1978;61:53-63.
32. McMillan G, Stary H. Preliminary experience with mitotic activity of cellular elements in the atherosclerotic plaques of cholesterol-fed rabbits studied by labeling with tritiated thymidine. Ann N Y Acad Sci.. 1968;149:699-709. [Medline] [Order article via Infotrieve]
33.
Gordon D, Reidy M, Benditt E, Schwartz S. Cell
proliferation in human coronary arteries. Proc
Natl Acad Sci U S A.. 1990;87:4600-4604.
34. Spagnoli LG, Orlandi A, Santeusanio G. Foam cells of the rabbit atherosclerotic plaque arrested in metaphase by colchicine show a macrophage phenotype. Atherosclerosis.. 1991;88:87-92. [Medline] [Order article via Infotrieve]
35. Raines EW, Ross R. Smooth muscle cells and the pathogenesis of the lesions of atherosclerosis. Br Heart J.. 1993;69:S30-S37.
36.
Lindner V, Reidy M. Proliferation of smooth
muscle cells after vascular injury is inhibited by an antibody against
basic fibroblast growth factor. Proc Natl Acad Sci
U S A.. 1991;88:3739-3743.
37. Libby P, Galis ZS. Cytokines regulate genes involved in atherogenesis. Ann N Y Acad Sci.. 1995;748:158-168. [Medline] [Order article via Infotrieve]
38. Hayashi K, Nishio E, Nakashima K, Amioka H, Kurokawa J, Kajiyama G. Role of cholesterol-accumulating macrophages on vascular smooth muscle cell proliferation. Clin Biochem.. 1992;25:345-349. [Medline] [Order article via Infotrieve]
39.
Maor I, Mandel H, Aviram M. Macrophage
uptake of oxidized LDL inhibits lysosomal sphingomyelinase, thus
causing the accumulation of unesterified
cholesterol-sphingomyelin-rich particles in the
lysosomes: a possible role for
7-ketocholesterol. Arterioscler Thromb Vasc
Biol.. 1995;15:1378-1387.
40. Tangirala R, Jerome WG, Jones N, Small D, Johnson W, Glick J, Mahlberg F, Rothblat GH. Formation of cholesterol monohydrate crystals in macrophage-derived foam cells. J Lipid Res.. 1994;35:93-104. [Abstract]
41. Vilcek J, Le J. Immunology of cytokines: an introduction. In: Thomson A, ed. The Cytokine Handbook. London, England: Academic Press; 1994:1-20.
42. Loscalzo J. The induction of cellular interactions in atherogenesis and their modulation. Biochem Soc Trans.. 1993;21:656-659.
43.
Nilsson J. Cytokines and smooth muscle
cells in atherosclerosis. Cardiovasc
Res.. 1993;27:1184-1190.
44. Cavallero C, Turolla EGR. Cell proliferation in the atherosclerotic plaques of cholesterol-fed rabbits, 1: colchicine and [3H]thymidine studies. Atherosclerosis. 1990;13:9-20.
45.
Spraragen S, Bond V, Dhal L. Role of hyperplasia
in vascular lesions of cholesterol-fed rabbits studied with
thymidine-3H autoradiography.
Circ Res.. 1962;11:329-336.
46. Gerrity R. The role of the monocyte in atherogenesis, II: migration of foam cells from atherosclerotic lesions. Am J Pathol.. 1981;103:191-200. [Abstract]
47.
Faggiotto A, Ross R. Studies of
hypercholesterolemia in the nonhuman primate,
II: fatty streak conversion to fibrous plaque.
Arteriosclerosis.. 1985;4:341-356.
48. Landers S, Gupta M, Lewis J. Ultrastructural localization of tissue factor on monocyte-derived macrophages and macrophage foam cells associated with atherosclerotic lesions. Virchows Arch.. 1994;425:49-54. [Medline] [Order article via Infotrieve]
49. Heinle H, Kling D, Betz E. Metabolism of fibromuscular and atheromatous plaques in an experimental model: causal mechanisms for the development of intimal necrosis. Curr Topics Pathol.. 1993;87:193-221. [Medline] [Order article via Infotrieve]
50. Han DK, Haudenschild CC, Hong MK, Tinkle BT, Leon MB, Liau G. Evidence for apoptosis in human atherogenesis and in a rat vascular injury model. Am J Pathol.. 1995;147:267-277. [Abstract]
51. Geng YJ, Libby P. Evidence for apoptosis in advanced human atheroma: colocalization with interleukin-1 beta-converting enzyme. Am J Pathol.. 1995;147:251-266. [Abstract]
52.
Isner JM, Kearney M, Bortman S, Passeri J.
Apoptosis in human atherosclerosis and
restenosis. Circulation.. 1995;91:2703-2711.
This article has been cited by other articles:
 |
|
 |
|
  A. Daugherty, N. R. Webb, D. L. Rateri, and V. L. King Thematic review series: The Immune System and Atherogenesis. Cytokine regulation of macrophage functions in atherogenesis J. Lipid Res., September 1, 2005; 46(9): 1812 - 1822. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. G. Yancey and W. G. Jerome Lysosomal cholesterol derived from mildly oxidized low density lipoprotein is resistant to efflux J. Lipid Res., March 1, 2001; 42(3): 317 - 327. [Abstract] [Full Text]
|
|
|
|
|
|
P. G. Yancey and W. G. Jerome Lysosomal sequestration of free and esterified cholesterol from oxidized low density lipoprotein in macrophages of different species J. Lipid Res., July 1, 1998; 39(7): 1349 - 1361. [Abstract] [Full Text]
|
|
|
|
|
|
W. G. Jerome, C. Cash, R. Webber, R. Horton, and P. G. Yancey Lysosomal lipid accumulation from oxidized low density lipoprotein is correlated with hypertrophy of the Golgi apparatus and trans-Golgi network J. Lipid Res., July 1, 1998; 39(7): 1362 - 1371. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||