Mast Cells in Rupture-Prone Areas of Human Coronary Atheromas Produce and Store TNF-α
Background Mast cells, a cell type involved in inflammatory reactions, are present in coronary atheromas and localize to the erosion or rupture site of atheromas in myocardial infarction. Here we report the presence of TNF-α, a proinflammatory cytokine, in mast cells of human coronary atheromas.
Methods and Results From samples of 37 coronary arteries from subjects autopsied for medicolegal reasons, sections of the bifurcation area of the left coronary artery were stained immunohistochemically for mast cells and TNF-α. In addition, macrophages, T lymphocytes, smooth muscle cells, and endothelial cells were investigated for their content of TNF-α. In normal intimas and fatty streaks, none of the cell types studied were TNF-α–positive. In 14 of the 24 atheromas found, TNF-α–positive cells were present. Of the total number of mast cells, 23% stained for TNF-α; of the macrophages, 1.3%; and of the smooth muscle cells, 0.4%. The majority (55%) of TNF-α–positive mast cells in the atheromas were located in the shoulder region and the remaining 35% in the cap and 10% in the core regions. Immunoelectron microscopy showed that the TNF-α in mast cells resided within their cytoplasmic secretory granules, demonstrating that these cells contain stores of TNF-α that will be released on degranulation.
Conclusions This study demonstrates the presence of mast cells with TNF-α–containing secretory granules, particularly in the shoulder region of human coronary atheromas. By releasing their TNF-α, mast cells may play an active role in the inflammatory reactions of these rupture-prone areas of atheromas.
Activated mast cells are present in human coronary atheromas. They accumulate especially in the shoulder region of atheromas, the predilection sites of atheromatous rupture.1 Indeed, infiltrates of activated mast cells are present at the sites of coronary atheromatous erosion or rupture in myocardial infarction.2 It was recently shown that the sites of atheromatous rupture contain a strong inflammatory component.3 TNF-α, an important mediator of inflammatory reactions,4 5 has been found in several cell types (SMCs, macrophages, and endothelial cells) present in human atheromas,6 7 and this cytokine is also found in skin mast cells8 and in lung mast cells.9 10 Indeed, mast cells have been found to be unique in that they both synthesize and store TNF-α,11 and they often participate in inflammatory reactions.12 For these reasons, we searched for evidence of the presence of TNF-α in the mast cells in human coronary atheromas.
Autopsy Material and Tissue Preparation
The autopsy series comprised 37 subjects, their ages ranging from 13 to 73 years. The causes of death were cardiovascular disease (n=12) and violent deaths (n=19), ie, accidents, homicides, suicides, and poisonings (n=6). From each cadaver, the unopened left coronary artery was removed from the heart, and three successive tubelike specimens were cut from the bifurcation area of the artery. For light microscopy, the specimens were fixed in Carnoy's fluid (60% ethanol, 30% chloroform, and 10% glacial acetic acid) for 24 hours and embedded in paraffin. The mean interval between death and the start of fixation was 13 hours (range, 2 to 24 hours). Light microscopy was used to evaluate the atherosclerotic involvement of the coronary arteries. For this purpose, the sections (2 to 4 μm) were stained with hematoxylin-eosin or elastica–van Gieson's stain (Weigert's hematoxylin, metanil yellow, acid fuchsin, and picric acid). The normal intima demonstrated variable thickness and moderate overall cellularity. Fatty streaks were recognizable as subendothelial and/or deep intimal collections of foam cells (containing numerous vacuoles). Atheromas appeared as intimal areas in which, in addition to foam cells, there were accumulations of extracellular lipid (large nonstaining areas interspersed with typical cholesterol clefts). These areas contained very few cells or were acellular, a sign of cell loss (necrotic lipid core), and covered by fibrotic tissue of various thicknesses (cap). Using these criteria, we found areas in which the intima appeared normal in all 37 subjects; 30 subjects also had fatty streaks, and 24 subjects, in addition to fatty streaks, had atheromas. In the atheroma, the various cell types to be studied were counted separately in the cap, core, and shoulder regions. Fig 1A⇓ shows a light microscopic view of a segment of a ring from the bifurcation area of the left coronary artery displaying an atheroma. Also shown are the various regions of the atheroma (cap, core, and shoulder) (Fig 1B⇓). For immunoelectron microscopy, specimens were fixed for 3 hours in 3% paraformaldehyde, dehydrated, and embedded in LR white resin. Sections (1 μm) of these specimens were stained with toluidine blue and observed with light microscopy for orientation of the site and for evaluation of the atherosclerotic involvement of each specimen.
For immunocytochemistry, fixed serial sections (2 to 4 μm) were dewaxed in xylene and rehydrated in a graded series of ethanol solutions, and endogenous peroxide activity was inhibited by incubation in 0.6% H2O2 in methanol. The sections were then incubated with one of the following: anti-tryptase monoclonal antibodies G3 (1.5 μg/mL) and B7 (4 μg/mL) (kind gifts from Dr L.B. Schwartz, Medical College of Virginia, Richmond) for mast cells13 ; HAM 56, a monoclonal antibody for macrophages (1:50); UCHL 1, a monoclonal antibody for T lymphocytes (1:50); and a polyclonal antibody for von Willebrand factor for endothelial cells (1:2000) (all from Dakopatts); a monoclonal antibody for α-smooth muscle actin for SMCs (1:12 000) (from Sigma Chemical Co); and a monoclonal antibody for TNF-α (1 μg/mL) (from Boehringer Mannheim). Mast cells and SMCs were stained according to the indirect immunoperoxidase method, and T lymphocytes, macrophages, and endothelial cells were stained by the avidin-biotin complex method, as recently described.1 2 TNF-α was stained by a modified avidin-biotin complex method to enhance the reaction, as follows: the sections were first incubated with normal horse serum for 10 minutes and with anti–TNF-α for 75 minutes, then with bridge antibody for 10 minutes, with avidin-biotin complex for 10 minutes, and again with bridge antibody for 10 minutes and with avidin-biotin complex for 10 minutes. The sections were then incubated overnight with anti–TNF-α, 10 minutes with the bridge antibody, 10 minutes with avidin-biotin complex, and 15 minutes in diaminobenzidine-peroxidase substrate solution. Finally, the samples were incubated with avidin-biotin complex for 10 minutes, stained again with diaminobenzidine-peroxidase substrate, and counterstained with Mayer's hematoxylin. All staining steps were carried out in a humidified chamber at 37°C except the overnight TNF-α incubation, which was carried out at room temperature. Indirect evidence that the antibody against TNF-α was recognizing this cytokine (positive controls) was obtained by staining one skin specimen and two colon cancer specimens.6 As expected, HAM 56–positive cells (macrophages) stained positively for TNF-α. Similarly, in skin, several tryptase-positive cells stained positively for TNF-α. Moreover, some basal keratinocytes stained positively, as also previously demonstrated by Walsh et al.8 As negative controls, we used sections from one brain6 stained for TNF-α as described above and coronary sections in which the primary antibody was omitted or replaced with normal serum (irrelevant antibody). In these sections, no positive staining was observed. Immunopositive mast cells, T lymphocytes, macrophages, SMCs, and endothelial cells were counted at ×100 magnification. Magnification ×1000 was used to observe degranulation of mast cells.
The ultramicrotome sections of LR white resin–embedded samples on nickel grids were incubated at room temperature with buffer (PBS containing 3% BSA) for 10 minutes and then overnight with anti–TNF-α (10 μg/mL). After two rinses in the buffer, the sections were incubated for 2 hours in 10 nm gold-labeled anti-mouse IgG+IgM (1:25) (Amersham) and finally stained with uranyl acetate and lead citrate. As negative controls, we used coronary sections in which the primary antibody was omitted or replaced with normal serum (irrelevant antibody). In these sections, no TNF-α staining was observed. The ultramicrotome sections were viewed with a Jeol JEM-1200EX transmission electron microscope at the Department of Electron Microscopy, University of Helsinki.
The proportion of TNF-α–positive cells was analyzed by logistic regression with a cell type (mast cells, macrophages, or SMCs) and a region of an atheroma (shoulder, cap, or core) used as explanatory variables. Endothelial cells and T lymphocytes were excluded from the analysis, because no TNF-α positivity was found in these cells. Differences were considered statistically significant when P<.05.
To demonstrate the presence of mast cells containing TNF-α in coronary atheromas, serial sections of atheromatous coronary specimens from the bifurcation areas of left coronary arteries were stained with monoclonal antibody against mast cell tryptase and against TNF-α. Fig 2A⇓ shows a typical example of mast cells in the shoulder region of a coronary atheroma; the five tryptase-containing mast cells are stained red-brown. When the same area was stained with anti–TNF-α antibody, four cells stained positively (Fig 2B⇓, brown). Comparison of the two adjacent sections revealed that, of the four TNF-α–positive cells, in this particular field, three were mast cells.
We also stained the serial sections with antibodies against macrophages, T lymphocytes, SMCs, and endothelial cells. The frequency of TNF-α positivity in normal and atherosclerotic lesions as a function of histological classification is shown in Table 1⇓. All 37 subjects studied had normal intima, 30 had fatty streaks, and 24 had not only fatty streaks but also atheromas. The total number of mast cells counted in the normal intimas was 106; in the fatty streaks, 350; and in the atheromas, 542. In the areas classified as normal or as fatty streaks, there were no detectable TNF-α–positive cells. In 14 of 24 atheromas, TNF-α–positive cells were visible. TNF-α–positive staining was found in three cell types: mast cells (in 8 atheromas), macrophages (in 8 atheromas), and SMCs (in 7 atheromas). In these samples, T lymphocytes and endothelial cells did not stain for TNF-α. In 4 of 14 atheromas, TNF-α–positive mast cells, macrophages, and SMCs were found. In the remaining 10 of 14 atheromas, only one or two of the above cell types were TNF-α positive (see Table 1⇓).
We next made a separate study of the three regions of atheromas: shoulder, cap, and core (Table 2⇓). Of the total number (n=542) of mast cells found, 23% (n=124) stained for TNF-α, whereas only 1.3% of the macrophages and 0.4% of the SMCs stained for TNF-α. The frequencies of TNF-α–positive mast cells in the various regions of the atheromas were as follows: 31% (68/219) in the shoulder, 26% (43/165) in the cap, and 8% (13/158) in the core region. It appeared that most (55%; 68/124) of the TNF-α–positive mast cells were located in the shoulder region, 35% (43/124) in the cap region, and the remaining 10% (13/124) in the core region. In contrast to the mast cells, the TNF-α–positive macrophages and SMCs were evenly distributed among the three regions of the atheromas.
To search for evidence of TNF-α storage in mast cell granules, we turned to immunoelectron microscopy. Ultramicrotome sections obtained from coronary atheromas were stained with anti–TNF-α, and 10 nm gold IgG+IgM were used as secondary antibodies. Fig 3⇓ shows a typical mast cell in the shoulder region of a coronary atheroma staining positively for TNF-α. TNF-α (gold particles) can be clearly seen within the cytoplasmic secretory granules of mast cells. Furthermore, in agreement with the light-microscopic results, we failed to observe TNF-α positivity in any cells in normal intima or in fatty streaks when examined by immunoelectron microscopic techniques.
The observation of TNF-α storage in mast cell granules is compatible with the view that one mechanism of TNF-α secretion by coronary mast cells is exocytosis of TNF-α–containing granules. Accordingly, we next examined the mast cells at high magnification (×1000) to detect extracellularly located granules. Table 3⇓ shows the results of this study. It appears that in each atheromatous region, a significant proportion of the mast cells had degranulated.
The present study shows that mast cells in human coronary atheromas contain TNF-α. As in earlier studies, we also found mast cells in normal intima and in fatty streaks.1 2 However, in contrast to the mast cells in atheromas, the mast cells in normal intima and in fatty streaks did not contain detectable amounts of TNF-α. Thus, as the atherosclerotic process advances (fatty streak→atheroma), the phenotype of a fraction of the intimal mast cells changes into an inflammatory one, as reflected by their TNF-α positivity. This finding is in accord with the observation that other intimal cell types, such as macrophages and SMCs, also began to express TNF-α as the atherosclerotic process advanced, an observation made by Barath et al.6 7 Similar disease-related upregulation of TNF-α has been observed in lung mast cells in patients suffering from atopic asthma.9
In the atheromatous lesions, the mast cells were shown to store TNF-α in their cytoplasmic secretory granules. Earlier studies with mouse peritoneal mast cells and with human dermal and lung mast cells have demonstrated that mast cells are capable of synthesizing TNF-α for storage in their granules.8 9 10 Thus, in all these experimental systems, IgE-dependent mast cell stimulation with ensuing degranulation resulted in significant increases in mRNA for TNF-α in the degranulated cells. What factors might have triggered the generation of TNF-α stores in mast cells in the atheromas? In other cell types, both IFN-γ and TNF-α have been documented to stimulate TNF-α synthesis.4 14 Recent findings that human coronary and aortic atheromas contain greatly increased (10- to 20-fold) numbers of T lymphocytes,1 2 a known source of IFN-γ,15 16 and macrophages,1 2 a known source of TNF-α,15 point to these two cell types as potential stimulators of TNF-α production in mast cells. In addition, the intimal SMCs, also a source of TNF-α,6 7 could stimulate the mast cells to synthesize TNF-α. Although the percentages of TNF-α–positive macrophages and SMCs were lower than that of mast cells, the absolute numbers of these cells were higher, making them potentially important sources of TNF-α. Because vascular lesions can exhibit enhanced expression of TNF-α,6 cytokine regulation of TNF-α synthesis and storage by mast cells in the atheromatous lesions may bear in vivo relevance.
The observation that TNF-α is localized to the cytoplasmic granules in mast cells demonstrates that these cells are able to store TNF-α and suggests that, once the mast cells have been stimulated to degranulate, TNF-α can be released rapidly in sizable amounts. Indeed, we recently found that the majority of the mast cells in the eroded or ruptured shoulder and cap regions have extracellular granules in their vicinity, a sign of their degranulation.2 In the present study also, signs of mast cell degranulation were found in the atheromas. What factors might trigger mast cells to degranulate? The best-understood immunological stimulus leading to degranulation of human mast cells is their activation on binding of a relevant antigen (allergen) to the IgE molecules present on the mast cells.17 A second, and perhaps even more likely, pathway leading to mast cell degranulation is their stimulation by other immunologically activated cells, such as T lymphocytes18 and macrophages,19 two cell types also present in the shoulder and cap regions of coronary atheromas.2 Moreover, atheromatous plaques activate complement and produce C5a,20 a powerful activator of mast cells.21 Admittedly, we cannot exclude the possibility that some coronary mast cells degranulate after death.
The mast cell–derived TNF-α may have several proinflammatory actions in coronary atheromas. It is known that TNF-α upregulates the leukocyte-endothelial adhesion molecules vascular cell adhesion molecule-1 and intercellular adhesion molecule-1.22 Interestingly, stimulation of human dermal mast cells results in release of TNF-α, which induces expression of endothelial-leukocyte adhesion molecule-1 on dermal endothelium.8 Therefore, the observation that TNF-α is expressed in the subendothelial mast cells of atheromas (ie, in the shoulder and cap regions) suggests that the mast cells are involved in recruitment of macrophages and T lymphocytes into the lesions. Moreover, TNF-α may enhance the growth of T lymphocytes4 and, by this means, may also contribute to the strongly increased (10-fold) number of T lymphocytes in atheromas compared with normal coronary intima.1 2 TNF-α is also known to enhance the number of IL-2 receptors on T lymphocytes and to induce production of IL-1, PGE2, and hydrogen peroxide in macrophages.4 Thus, one important function of the TNF-α–containing mast cells of atheromas could be amplification of the immune response in these lesions.
Finally, an important concept is emerging that increased activity of enzymes required for extracellular matrix digestion (MMPs) are causing coronary atheromas to rupture.23 24 25 The MMPs are synthesized and secreted by SMCs and macrophages of the atherosclerotic lesions.23 24 25 In vitro studies have shown that both TNF-α and IL-1 stimulate SMCs to synthesize de novo 92-kD gelatinase, interstitial collagenase, and stromelysin, the three major MMPs found in atherosclerotic lesions.26 Moreover, we have observed that TNF-α stimulates production of 92-kD gelatinase by human monocytes-macrophages in vitro.27 Thus, activation of TNF-α–containing mast cells in human coronary atheromas, with ensuing degranulation and TNF-α release, could locally weaken the atheroma and lead to atheromatous erosion or rupture.2
In conclusion, the present study reveals that a fraction of the mast cells in human coronary atheromas contain a potent proinflammatory cytokine, TNF-α. This observation strengthens the notion that mast cells play a role in the initiation and maintenance of the inflammatory process found in advanced human coronary atheromas.
Limitations of the Study
In this autopsy study, in which TNF-α–positive mast cells in human coronary atheromas are described for the first time, the specimens were examined 2 to 24 hours after death. These conditions are suboptimal for detection of cytokine-positive cells by immunohistochemistry. Accordingly, the number of atheromas with TNF-α–positive cells is likely to represent an underestimate of the actual proportion of the atheromas in which cells were TNF-α–positive. Despite these problems, comparison between normal intima, fatty streaks, and atheromas was of interest, since it revealed that TNF-α–positive cells are found only in the most advanced lesions, which contain a strong inflammatory component and are prone to rupture.
Selected Abbreviations and Acronyms
|SMC||=||smooth muscle cell|
|TNF||=||tumor necrosis factor|
This study was supported in part by the Aarne Koskelo Foundation. The authors thank the staff of the Department of Forensic Medicine and Electron Microscopy, University of Helsinki, for their excellent technical assistance, and Jari Haukka for advice in statistical analysis of the data.
- Received May 6, 1996.
- Revision received June 27, 1996.
- Accepted July 1, 1996.
- Copyright © 1996 by American Heart Association
Kaartinen M, Penttila¨ A, Kovanen PT. Accumulation of activated mast cells in the shoulder region of human coronary atheroma, the predilection site of atheromatous rupture. Circulation. 1994;90:1669-1678.
Kovanen PT, Kaartinen M, Paavonen T. Infiltrates of activated mast cells at the site of coronary atheromatous erosion or rupture in myocardial infarction. Circulation. 1995;92:1084-1088.
van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994;89:36-44.
Wong GHW, Goeddel DV. Tumor necrosis factor. In: Zembala M, Asherson GL, eds. Human Monocytes. San Diego, Calif: Academic Press; 1989:195-215.
Walsh LJ, Trichieri G, Waldorf HA, Whitaker D, Murphy GF. Human dermal mast cells contain and release tumor necrosis factor α, which induces endothelial leukocyte adhesion molecule 1. Proc Natl Acad Sci U S A. 1991;88:4220-4224.
Bradding P, Roberts JA, Britten KM, Montefort S, Djukanovic R, Mueller R, Heusser CH, Hoearth PH, Holgate ST. Interleukin-4, -5, and -6 and tumor necrosis factor-α in normal and asthmatic airways: evidence for the human mast cell as a source of these cytokines. Am J Respir Cell Mol Biol. 1994;10:471-480.
Ohkavara Y, Yamauchi K, Tanno Y, Tamura G, Ohtari H, Nugura H, Ohkuda K, Takishima T. Human lung mast cells and pulmonary macrophages produce tumor necrosis factor-α in sensitized lung tissue after IgE receptor triggering. Am J Respir Cell Mol Biol. 1992;7:385-392.
Irani A-M, Bradford TR, Kepley CL, Schechter NM, Schwartz LB. Detection of MCT and MCTC types of human mast cells by immunohistochemistry using new monoclonal anti-tryptase and anti-chymase antibodies. J Histochem Cytochem. 1989;37:1509-1515.
Liu MC, Proud D, Lichtenstein LM, MacGlashan DW, Schleimer RP, Adkinson NF, Kagey-Sobotka A, Schulman ES, Plaut M. Human lung macrophage-derived histamine-releasing activity is due to IgE-dependent factors. J Immunol. 1986;136:2588-2595.
Metcalfe DD, Kaliner M, Donlon DA. The mast cell. In: CRC Critical Reviews in Immunology. Boca Raton, Fla: CRC Press; 1981:23-74.
Henney AM, Wakeley PR, Davies MJ, Foster K, Hembry R, Murphy G, Humphries S. Localization of stromelysin gene expression in atherosclerotic plaques by in situ hybridization. Proc Natl Acad Sci U S A. 1991;88:8154-8158.
Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest. 1994;94:2493-2503.
Brown DL, Hibbs MS, Kearney M, Loushin C, Isner JM. Identification of 92-kD gelatinase in human coronary atherosclerotic lesions: association of active enzyme synthesis with unstable angina. Circulation. 1995;91:2125-2131.
Galis ZS, Muszynski M, Sukhova GK, Simon-Morrissey E, Unemori EN, Lark MW, Amento E, Libby P. Cytokine-stimulated human vascular smooth muscle cells synthesize a complement of enzymes required for extracellular matrix digestion. Circ Res. 1994;75:181-189.
Saren P, Welgus HG, Kovanen PT. Tumor necrosis factor-α and interleukin-1β selectively induce expression of 92-kDa gelatinase by human macrophages. J Immunol. 1996;157:4159-4165.