Articles |
From the Monash University Department of Medicine, Monash Medical Centre, Clayton, Victoria, Australia.
Correspondence to P.G. Tipping, Monash University Department of Medicine, Monash Medical Centre, Clayton, Victoria 3168, Australia.
| Abstract |
|---|
|
|
|---|
Key Words: T lymphocyte atherosclerosis CD4 CD5 macrophage
| Introduction |
|---|
|
|
|---|
, have been
detected, suggesting a functional role for these cells in the
development of disease.4 There are fewer studies of the
participation of T cells in early atherosclerotic lesions. A detailed
study of early human lesions suggests that CD8-positive T cells are
more common in early lesions3 ; however, in this study
there were technical difficulties with demonstrating some T cell
antigens. Cholesterol-fed rabbits develop atherosclerotic lesions with features similar to those of early human plaques.5 6 7 This model allows systematic investigation of the earliest events in atherogenesis. By use of a monoclonal antibody (L11/135) directed against the rabbit CD43 antigen,8 the participation of T cells has been previously suggested in this model.9 This antigen is expressed on T cells, B cells, and granulocytes and also weakly on monocytes,10 and thus the precise identity of these CD43-positive cells warrants further clarification.
Recently, monoclonal antibodies to rabbit T cell antigens (CD4 and CD5) have become available that are more specific for T cells and allow T helper cells to be identified in rabbit tissues. The availability of these antibodies provides a more precise means of identifying T cells in atherosclerosis and allows definition of the participation of T helper cells in the early phase of development of plaques. In this study, these antibodies have been used to characterize the development of the T cell infiltrate and the association of T cells with cellular proliferation within plaques.
| Methods |
|---|
|
|
|---|
Measurement of Serum Cholesterol
Blood samples were collected from the central ear artery of
rabbits at 0, 2, 4, 8, and 12 weeks on the cholesterol
diet. Serum total cholesterol levels were measured by
cholesterol esterase/cholesterol oxidase
enzymatic assay with Trace Liquid Cholesterol reagents
(Trace Scientific).
Quantitation of Atheroma
Fixed aortas were stained with Sudan IV (Sigma), mounted, and
analyzed with OLYMPUS CUE 2 IMAGE
ANALYZER software (Galai Production Ltd). The
software calculates the stained surface area and the total aortic
surface area by use of two preset light intensity thresholds. The
surface area covered by atherosclerotic plaques was expressed as a
percentage of the total surface area of the aortic regions. The
cross-sectional area of plaques was measured by image
analysis on ethanol-fixed tissue sections taken from the
aortic arch and stained with hematoxylin and eosin (BDH Chemicals
Ltd).
Quantitation of T Cells and Macrophages in
Plaques
Frozen 4-µm sections of aortic tissue were fixed in ethanol
and stained with hematoxylin for estimation of total cellularity and
with monoclonal antibodies to rabbit T cells, macrophages, and
smooth muscle cells for definition of cell populations in plaques.
Cellularity of plaques was assessed by manual counting of the number of
cells in a total plaque area of 0.2 mm2 with a graticule.
Because of their small cross-sectional area, all cells in 2-week
plaques were counted.
T cells were identified by use of L11/135 (anti-rabbit
CD43),11 KEN4 (anti-rabbit CD4), and KEN5
(anti-rabbit CD5).12 Macrophages were
identified by use of RAM11 (anti-rabbit
macrophage)5 and smooth muscle cells with HHF35
(anti-smooth muscle
-actin).13 14 A
streptavidin/biotin alkaline phosphataselinked detection system
(Dako Quickstaining Kit, Alkaline Phosphatase system 40, Dako Corp) was
used with these antibodies and sections were counterstained with
hematoxylin. CD5- and CD4-positive T cells were counted in three aortic
arch cross-sections from each rabbit. The density of the T cell
infiltrate was calculated by use of the cross-sectional area of the
lesion measured by computerized image analysis.
Assessment of Cellular Proliferation
Cellular proliferation was assessed by use of two separate
markers. For the first, rabbits were injected
intraperitoneally with 50 mg/kg
5-bromo-2'-deoxyuridine (BrdU) (Sigma) 1 hour before they were killed
for labeling of all cells in the S phase of the cell cycle. Aortic
tissue was snap-frozen in liquid nitrogen. Methanol-fixed
4-µm cryostat sections pretreated with 2 mol/L HCl were incubated
with an anti-bromodeoxyuridine monoclonal antibody (BMC9318,
Boehringer Mannheim) at a dilution of 1 in 50. Staining was
detected by use of peroxidase-conjugated sheep anti-mouse
immunoglobulin (Silenus Laboratories) and diaminobenzidine substrate
(Sigma). For the second, ethanol-fixed paraffin tissue was stained
with monoclonal antibody PC10 (Dako Corp), directed against
proliferating cell nuclear antigen (PCNA), expressed on all cells in
the late G1 or S phase of the cell cycle, and detected by use of a
streptavidin/biotin alkaline phosphataselinked system.
Cellular proliferation in aortic plaques was determined in each rabbit by use of both techniques. Labeled cells were counted in three plaque areas, each containing at least 100 cells. The proliferative index was calculated from the percentage total of cells labeled with BrdU or PCNA. The phenotype of proliferating cells was determined by dual labeling with PC10 detected by use of a streptavidin/biotin alkaline phosphataselinked detection system and with RAM11, L11/135, or HHF35 detected by use of peroxidase-conjugated sheep anti-mouse Ig antibodies and diaminobenzidine substrate. An irrelevant isotype-matched monoclonal antibody was used at matched protein concentrations as a negative control in all of the above immunohistochemical procedures.
Experimental Design and Statistical Analysis
Groups of six rabbits were studied after 2, 4, 8, 12, and 16
weeks of cholesterol feeding. Six rabbits fed standard
rabbit chow without cholesterol supplementation were
included as "normal" (control) rabbits. This group was of age and
weight similar to those of the group supplemented with
cholesterol for 8 weeks. Results are expressed as
mean±SEM, and statistical analysis was performed with a
Mann-Whitney U test.
| Results |
|---|
|
|
|---|
|
|
Cellular Infiltration in Plaques
The cellularity of plaques was maximal at 4 weeks (3623±467
cells/mm2). Before this time, plaques contained 1750±296
cells/mm2 (2 weeks), and at subsequent time points the
cellularity of the plaques progressively declined (8 weeks, 2239±271
cells/mm2; 12 weeks, 1535±55 cells/mm2; 16
weeks, 1747±242 cells/mm2; P<.05 for each
group compared with the 4-week group). During this period, the
cross-sectional area of the lesions increased (Table 1
).
RAM11-positive macrophage-derived foam cells were the
predominant cell type in plaques at all times. HHF35-positive smooth
muscle cells were rarely observed in intimal lesions of rabbits fed
cholesterol for 4 weeks but were observed forming smooth
muscle cell "caps" over more advanced lesions at 12 and 16 weeks.
Even in these later lesions exhibiting caps, RAM11-positive
macrophage-derived foam cells constituted more than
95% of all lesion cells.
Analysis of T Cell Infiltration
CD4-, CD5-, and CD43-positive cells were not observed in intimal
regions of aortas from normal rabbits. However, cells expressing these
phenotypic markers were regularly observed in intimal plaques of
cholesterol-fed rabbits (Fig 2
), with
the exception of 2-week plaques. The rate of influx of T cells was
maximal between 2 and 4 weeks. The density of the T cell infiltrate
(number of T cells per unit plaque area) reached its maximum after 4
weeks. At this time five of six rabbits exhibited T cell accumulation
in plaques (Fig 3
), with a mean of 2.67±0.71
CD5-positive T cells per plaque cross-section,
representing 0.31% of the total plaque cells. CD5-positive
cell numbers per plaque cross-section did not increase
significantly after 4 weeks. CD4-positive cells at 4 weeks
represented 89% of the number of CD5-positive cells,
suggesting that the majority of the T cells in plaques are T helper
cells. The anti-CD43 antibody marked a greater number of cells than
either the anti-CD4 or anti-CD5 antibodies, consistent with the
fact that expression of CD43 antigen is not restricted to T cells (Fig 4
).
|
|
|
Cellular Proliferation in Plaques
By use of BrdU incorporation and PCNA expression, similar numbers
of proliferating cells were identified in atherosclerotic plaques at 4
weeks and at subsequent time points. Proliferating cells were not
observed in the aortic intimas of normal rabbits. Cellular
proliferation after 2 weeks of cholesterol feeding was
detected in only one rabbit by use of PCNA expression but not by BrdU
incorporation. The peak proliferative activity (2.05% of plaque cells
expressing PCNA) occurred in the aortic intimal lesions at 4 weeks
(Table 2
). By use of double-labeling
immunohistochemistry with anti-PCNA and RAM11 or HHF35, proliferating
cells were invariably identified as macrophages (Fig 5
). Less than 1% of PCNA-positive cells were smooth
muscle cells, even in more advanced lesions with smooth muscle cell
caps.
|
|
| Discussion |
|---|
|
|
|---|
The participation of T cells in these early atherosclerotic lesions was studied with monoclonal antibodies to rabbit T cells and their subsets. The monoclonal antibodies KEN4 and KEN5 recognize T cell antigens, which are the rabbit equivalents of CD4 and CD5, respectively.12 KEN5 stains 90% of thymocytes and by immunoprecipitation recognizes a single protein of 67 kD, which is identical to the molecular weight of CD5 in humans, mice, rats, sheep, and cattle. KEN4 stains 85% of thymocytes and recognizes a major protein of 50 kD and a minor protein of 42 kD, similar to the bovine pattern of CD4.
By use of these antibodies, T lymphocyte infiltration was observed to be an early event in atherogenesis, with the major influx of T cells occurring between 2 and 4 weeks. The maximum density of CD5-positive lymphocyte infiltration was observed after 4 weeks of cholesterol feeding. Subset analysis with anti-CD4 antibodies demonstrated that the majority of these cells were T helper cells. At later times, despite progressive increase in plaque size, the total numbers of plaque T cells did not increase and the density of the T cell infiltrate decreased.
Hansson et al9 previously used the monoclonal antibody L11/135, an anti-CD43 marker, to quantitate T cells in rabbit atherosclerosis after 6 and 10 weeks of cholesterol feeding. CD43 is expressed on B cells, granulocytes, and weakly on monocytes as well as on T cells10 ; therefore, using this marker alone will lead to overestimation of T cells in lesions where other inflammatory cells are involved. This would explain the higher T cell numbers reported by Hansson et al compared with the present study, in which CD43-positive and CD4- and CD5-negative cells were observed in lesions. It is likely that this population of cells comprises recently infiltrated monocytes or early differentiating macrophages, because significant numbers of granulocytes or B lymphocytes have not been observed in atheromatous plaques.1
Emeson et al3 found T cells present in aortic intimal thickenings of teenagers and young adults who died acutely of trauma, suggesting early T cell involvement in human atherosclerosis. CD8-positive T cells were most consistently demonstrated in these postmortem specimens, whereas studies of more advanced lesions obtained at surgery have demonstrated a predominance of CD4-positive cells.16 Our findings in rabbit atheroma demonstrated similar numbers of CD5- and CD4-positive lymphocytes, suggesting a minor contribution of CD8-positive cells. Although this may represent a species difference between atheroma in humans and rabbits, it may also reflect the difficulty with preservation of some lymphocyte antigens in postmortem tissue, because the quality of CD4 staining in the study by Emeson et al was variable and CD3 antigen could not be detected.
The participation of T helper cells together with macrophages
in early atherosclerotic lesions suggests the possibility that
cell-mediated immunity, akin to delayed-type hypersensitivity,
may participate in atherogenesis. The Th1 subset of helper T cells
plays an important role in directing delayed-type hypersensitivity
responses and is defined by its production of cytokines
including interleukin-2, interferon-
, and tumor necrosis
factorß.17 Th1 cells provide T cell help in the
recruitment, differentiation, and activation of
macrophages on representation of antigen. Th2 cells
produce interleukin-4, interleukin-6, and interleukin-10 and provide T
cell help in antibody production. The demonstration of
interferon-
and the presence of major histocompatibility complex
class II antigen on plaque smooth muscle cells (which is induced by
interferon-
) is consistent with a Th1-type
response.4 18
Proliferating cells were identified by BrdU incorporation and expression of PCNA, and there was a good correlation between results with these two methods (r=.95 by linear regression analysis). BrdU is incorporated into the DNA of replicating cells during the S phase of the cell cycle, whereas expression of PCNA is observed in the late G1 and S phase. Thus, estimates of proliferation made by use of PCNA are often slightly higher than those made with other proliferation detection techniques because of the longer phase of PCNA expression.19 20 The proliferative index of these early rabbit lesions was similar to that reported in early human lesions in which less than 2% of fatty streak cells were PCNA positive.21 The proliferative index in rabbit plaques decreased as lesions became more advanced, reflecting previous findings in human plaques21 and in plaques from cholesterol-fed and Watanabe heritable hyperlipidemic rabbits.22
In the present study, proliferating cells in early plaques were predominantly macrophages, and proliferation of smooth muscle cells was rarely observed. In a previous study in rabbits fed cholesterol for 6 months, the majority of proliferating cells in these advanced lesions were reported to be macrophages.23 In contrast, Rosenfeld and Ross22 studied early, intermediate, and advanced lesions (which were graded according to plaque size) from both cholesterol-fed and Watanabe heritable hyperlipidemic rabbits and found similar rates of smooth muscle cell and macrophage proliferation in advanced lesions. They did not phenotype proliferating cells in early lesions. In human plaques, macrophages have been found to be the predominant proliferative cell type in early lesions,21 24 whereas similar numbers of proliferating macrophages and smooth muscle cells were observed in more advanced human plaques.19
T cell infiltration was demonstrated to be temporally correlated with
macrophage proliferation in plaques. Although monocytes have
been well documented to transverse the endothelium from
the lumen to enter intimal lesions, proliferation may be an additional
mechanism contributing to macrophage accumulation in
atherosclerotic plaques. The association between T cell infiltration
and macrophage proliferation raises the possibility that T cell
products may direct local macrophage proliferation in
plaques. T cellderived cytokines, such as macrophage
colony-stimulating factor, with the potential to induce
macrophage proliferation have been demonstrated in both human
and rabbit atherosclerotic plaques.25 26 27 Other T cell
cytokines such as interferon-
may have an antiproliferative
effect on cells within plaques.28
In summary, T lymphocyte infiltration, predominantly of the T helper cell phenotype, was demonstrated in the earliest stages of atherogenesis in the cholesterol-fed rabbit. Proliferation of macrophage-derived foam cells was also observed at this early stage and was temporally associated with the intensity of T cell infiltration, which may indicate a role for T cells in directing macrophage proliferation. These studies suggest the participation of T helper cells in the early events in atherogenesis.
| Acknowledgments |
|---|
Received November 30, 1994; accepted July 10, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. R. T. Damasceno, H. Goto, F. M. D. Rodrigues, C. T. S. Dias, F. S. Okawabata, D. S. P. Abdalla, and M. Gidlund Soy Protein Isolate Reduces the Oxidizability of LDL and the Generation of Oxidized LDL Autoantibodies in Rabbits with Diet-Induced Atherosclerosis J. Nutr., November 1, 2000; 130(11): 2641 - 2647. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Orlandi, M. Marcellini, and L. G. Spagnoli Aging Influences Development and Progression of Early Aortic Atherosclerotic Lesions in Cholesterol-Fed Rabbits Arterioscler. Thromb. Vasc. Biol., April 1, 2000; 20(4): 1123 - 1136. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Alcouffe, S. Caspar-Bauguil, V. Garcia, R. Salvayre, M. Thomsen, and H. Benoist Oxidized low density lipoproteins induce apoptosis in PHA-activated peripheral blood mononuclear cells and in the Jurkat T-cell line J. Lipid Res., July 1, 1999; 40(7): 1200 - 1210. [Abstract] [Full Text] |
||||
![]() |
H. O. Andersen, B. F. Hansen, P. Holm, S. Stender, and B. G. Nordestgaard Effect of Cyclosporine on Arterial Balloon Injury Lesions in Cholesterol-Clamped Rabbits : T Lymphocyte–Mediated Immune Responses Not Involved in Balloon Injury–Induced Neointimal Proliferation Arterioscler. Thromb. Vasc. Biol., July 1, 1999; 19(7): 1687 - 1694. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Truskey, R. A. Herrmann, J. Kait, and K. M. Barber Focal Increases in Vascular Cell Adhesion Molecule-1 and Intimal Macrophages at Atherosclerosis-Susceptible Sites in the Rabbit Aorta After Short-Term Cholesterol Feeding Arterioscler. Thromb. Vasc. Biol., February 1, 1999; 19(2): 393 - 401. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Roselaar, P. X. Kakkanathu, and A. Daugherty Lymphocyte Populations in Atherosclerotic Lesions of ApoE -/- and LDL Receptor -/- Mice: Decreasing Density With Disease Progression Arterioscler. Thromb. Vasc. Biol., August 1, 1996; 16(8): 1013 - 1018. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |