Proliferating Arterial Smooth Muscle Cells After Balloon Injury Express TNF-α but Not Interleukin-1 or Basic Fibroblast Growth Factor
Abstract We have recently reported that balloon withdrawal injury to rabbit abdominal aortas induces sustained activation indicated by the expression of certain adhesion molecules such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 in regenerating endothelial cells and/or proliferating smooth muscle cells (SMCs). Local cytokine signaling may contribute to ongoing modulation of cellular functions and proliferation of intimal SMCs after acute vascular injury. We therefore studied the expression of tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), proinflammatory and SMC growth–promoting cytokines, and basic fibroblast growth factor (bFGF) in SMCs of rabbit aorta at 2 (n=4), 5 (n=4), and 10 days (n=6) after balloon injury. All animals were given bromodeoxyuridine (BrdU, 10 mg/kg per day) continuously to label proliferating SMCs. Frozen cross sections of injured vessels at each time point after balloon injury were examined by immunoperoxidase staining with monoclonal antibodies. As early as 2 days after injury, before intimal thickening begins, foci of medial SMCs expressed TNF-α, but not all TNF-α–positive medial SMCs had incorporated BrdU, suggesting that TNF-α expression by medial SMCs may precede their proliferation. At 5 days, TNF-α–bearing and BrdU-labeled medial SMCs increased in number. At 10 days after injury, when uniform intimal thickening occurred, almost all neointimal SMCs and foci of medial SMCs labeled with BrdU. Most of the BrdU-positive (proliferating) SMCs expressed immunoreactive TNF-α. Reverse transcription polymerase chain reaction showed increased TNF-α mRNA at 10 days after ballooning in the injured portion of the aorta. In contrast, regions of SMC proliferation showed inconsistent IL-1β expression, and bFGF, abundant in normal rabbit arteries, was not detected in areas of SMC replication. These data indicate that replication of arterial SMCs after balloon injury occurs in regions of TNF-α but not IL-1β expression and correlates inversely with the presence of bFGF. These results indicate that SMC-derived TNF-α serves as a marker of modulated SMC phenotype after acute vascular injury and may contribute to local cellular activation and proliferation of SMCs at sites of arterial injury.
- Received March 25, 1995.
- Accepted September 11, 1995.
The signals that stimulate SMC activation and proliferation in the intima of injured blood vessels remain uncertain. We have recently presented evidence for SMC activation in injured arteries sustained for periods as long as 30 days after injury.1 Excellent experimental evidence demonstrates that bFGF mediates the first wave of medial SMC proliferation after balloon injury to the previously normal rat carotid artery.2 3 4 Yet bFGF alone does not fully account for the latter stages of SMC activation including ongoing proliferation in the expanding intima.5 Our previous work has implicated cytokines, protein mediators of inflammation and immunity, as potential activators of SMCs in vascular pathology.6 7 8 The cytokine known as TNF has particular interest in this regard. TNF can augment SMC expression of intercellular adhesion molecule-1, one of the indexes of sustained smooth muscle activation that we have previously characterized.1 9 In addition, TNF acts on SMCs by inducing prostanoid synthesis and the expression of genes encoding other cytokines.10 Exposure to TNF can also stimulate proliferation of SMCs under some circumstances,11 12 perhaps by induction of autocrine growth factor expression.12 13 14 15 Additionally, TNF regulates endothelial functions of potential importance in vascular pathology, including the expression of endothelial-leukocyte adhesion molecules and regulation of multiple aspects of the anticoagulant and fibrinolytic capacities of the endothelium.16 17 18 19 20 21
Originally, the macrophage was considered the principal cell type capable of producing TNF-α. Some years ago, our laboratory characterized the ability of SMCs to express the gene encoding TNF-α.10 We demonstrated this capacity at the level of mRNA induction, de novo protein synthesis, and elaboration of biologically active TNF by these human cells. Barath and colleagues22 23 demonstrated that advanced human atherosclerotic lesions contain TNF, a further indication of the involvement of this cytokine in intimal pathology.22 23 Because of these manifold effects of TNF on functions of vascular cells including those expressed after balloon injury, the present study explored the possibility that arterial balloon injury induced the expression of this cytokine. This study also evaluated in parallel bFGF and IL-1, two other protein mediators that might participate in vascular SMC growth control.
Animals and Surgical Procedures
New Zealand White Pasteurella-free rabbits weighing approximately 3 kg were studied. The rabbits were fed normal rabbit chow. The animals were anesthetized with ketamine 35 mg/kg and xylazine 7 mg/kg IM. After systemic heparinization (1 mg/kg), a Fogarty 3F balloon catheter was introduced into the abdominal aorta through the superficial femoral artery to 10 cm proximal to the arteriotomy. Then the balloon was inflated with 0.1 mL saline and gently withdrawn to the distal aorta. This balloon injury to the 5-cm segment of the infrarenal abdominal aorta was repeated three times. All procedures used standard aseptic technique, and animals were carefully monitored by clinical examination during recovery from anesthesia.
BrdU Labeling of Cells
BrdU, a thymidine analogue that is incorporated into the DNA of S-phase proliferating cells, was continuously infused subcutaneously through a miniaturized osmotic pump (Model 2ML-1, Alza Corp) implanted in an anterior lower abdominal subcutaneous pocket during the same anesthetic procedure as balloon injury. BrdU solutions were prepared in sterile Na2CO3/NaHCO3 buffer (0.5 mol/L, pH 9.8). The concentration of BrdU was adjusted to 250 mg/mL to release 10 mg/kg per day of the labeling agent.
Harvest and Preparation of Tissue
At 2 days (n=4), 5 days (n=4), and 10 days (n=6) after surgery, animals were placed in a restraining cage and Evans blue (20 mg · mL saline−1 · kg−1) was injected via an ear vein 20 minutes before euthanasia by infusion of pentobarbital (120 mg/kg IV). Evans blue staining clearly defined the injured area, and the infrarenal aorta was immediately excised and gently trimmed in Dulbecco’s PBS. We prepared cross sections from the center of the injured aorta as defined by Evans blue staining for immunohistochemical evaluation by snap-freezing in isopentane cooled in liquid nitrogen.
The following mAb were used for immunohistochemical analysis: mouse mAb anti-human TNF-α (a mouse IgG1, purchased from UBI) and IL-1β (a mouse IgG1 purchased from Genzyme) antibody were used to recognize the respective rabbit cytokines. An anti-human bFGF antibody (a mouse IgG1 purchased from UBI) was used to visualize the rabbit protein. mAb HHF-35 (mouse IgG1, purchased from Enzo Diagnostics Inc) recognizes muscle-specific actin.24 We detected BrdU in cells undergoing proliferation by using mAb against BrdU (mouse IgG1, purchased from DAKO Corp).
Serial cryostat sections (6 μm) were cut, air dried onto poly-l-lysine–coated slides, and fixed in acetone at −20°C for 5 minutes. Sections were preincubated with PBS containing 0.3% hydrogen peroxide to reduce endogenous peroxidase activity. The sections then were incubated with primary antibodies and diluted in PBS with 10% horse serum at room temperature for 60 minutes. After three washes in Tris-buffered saline containing 2% horse serum, species-appropriate biotinylated secondary antibodies were applied followed by avidin-biotin horseradish peroxidase complex (Vectastain ABC kit, PK 6100, Vector Laboratories). Antibody binding was visualized with 3-amino-9-ethylcarbazole (Sigma Chemical Co). For BrdU labeling, sections were incubated in 2 mol/L HCl at 37°C for 10 minutes to denature DNA before the addition of the primary antibody. Sections were counterstained with Gill’s hematoxylin. Omission of primary antibodies and the staining with type- and class-matched irrelevant immunoglobulin served as negative control for each antibody.
For evaluation of the kinetics of SMC proliferation after balloon withdrawal injury in this model, all BrdU-labeled cells were counted separately in the intima and media of cross sections from each animal. Results are expressed as mean±SD. ANOVA was used to determine the significance of differences by comparing different time points. Differences were considered significant when P<.05.
Double Immunohistochemical Analysis
To evaluate distribution of positive staining for both BrdU and TNF-α in the injured area at 2 and 5 days after balloon injury, we performed double immunohistochemistry. After staining for TNF-α using the peroxidase ABC method described above (yielding a red peroxidase reaction product), sections were washed in PBS for 5 minutes, and avidin-biotin complex remaining from the first step was blocked by incubating sections with an excess of avidin and biotin (Avidin-Biotin blocking kit, Vector Laboratories). After the application of the primary antibody against BrdU at 4°C overnight, sections were incubated with biotinylated horse anti-goat antibody for 45 minutes at room temperature. Sections were then incubated in alkaline phosphatase–avidin-biotin complex (Vectastain ABC kit, AK-5000, Vector Laboratories) and visualized using fast blue (Sigma Chemical Co), an alkaline phosphate substrate that produced a blue product. Sections were counterstained with methyl green. Double-stained cells showed red (TNF-α) and blue (BrdU) staining.
Immunohistochemical Staining of TNF-α in Cultured Rabbit Aortic SMCs Stimulated With Bacterial Lipopolysaccharide
We previously reported that bacterial LPS treatment increases TNF-α expression in rabbit aorta.24 Therefore, we explored the ability of the mAb against human TNF-α to recognize rabbit TNF-α protein by study of LPS-treated SMCs cultured from rabbit aorta by standard explant outgrowth techniques. Cultured SMCs were treated with LPS (10 ng/mL) for 24 hours. After stimulation, SMC monolayers were rinsed twice with PBS and immediately fixed for 1 minute in acetone at −20°C. SMCs with or without LPS treatment were stained immunohistochemically as described above.
PCR Analysis of TNF-α mRNA in Uninjured and Injured Rabbit Aortic Tissue
Five additional New Zealand White rabbits were studied to evaluate the mRNA levels of TNF-α in the uninjured and injured aorta. The whole infrarenal abdominal aorta was injured by using the same procedure described above. After injection of a lethal dose of sodium pentobarbital (120 mg/kg IV) 10 days after balloon injury, the uninjured thoracic aorta and injured infrarenal abdominal aorta were removed from each rabbit, rinsed in sterile saline, trimmed of adventitial tissue, and immediately frozen in liquid nitrogen. RNA samples prepared from the spleen of LPS-treated animals served as a positive control for the PCR analysis. Specimens of aorta and spleen were homogenized in guanidinium isothiocyanate, and total RNA was isolated by differential centrifugation. One microgram of RNA was reverse-transcribed, and cDNA was analyzed by PCR at 35 cycles, a level of amplification previously determined to lie within the exponential range of expansion with these primers and rabbit cDNA. Details of the PCR technique and primers used to detect a rabbit TNF-α message have been previously described.24 25 Seven microliters of the PCR product was mixed with 3 μL of running buffer (orange G dye in 50% Ficoll). This mixture was electrophoresed on a 3% agarose gel containing ethidium bromide and visualized by ultraviolet light.
Expression of TNF-α in Cultured Rabbit SMCs Stimulated With Bacterial LPS
SMCs cultured from rabbit aorta did not stain with the anti-human TNF antibody. However, after exposure to LPS, a prototypic TNF-inducer, rabbit aortic SMCs stained in a fine granular pattern with the anti-TNF antibody (Fig 1⇓). Sections incubated with class- and type-matched immunoglobulin did not stain at all (not shown). These results indicate that this antibody recognizes rabbit TNF-α protein.
Kinetics of SMC Proliferation at 2, 5, and 10 Days After Balloon Injury
At 2 days after balloon injury, the intima had not begun to thicken, but within the tunica media of injured regions some SMCs (96±33 per cross section, mean±SD, n=4) had proliferated as indicated by BrdU staining (Fig 2⇓, Table⇓). Five days after balloon withdrawal, the intima of injured zones usually contained one to three layers of SMCs, many of which had proliferated (Fig 2⇓, Table⇓). At this time, BrdU labeling of medial SMCs had increased almost sevenfold. The extent of medial SMC proliferation varied both from animal to animal (as indicated by the wide standard deviation; see the Table⇓) and within ballooned areas within a given animal. This inhomogeneity in localization of BrdU+ SMCs within the media may result from uneven distribution of the mechanical injury to the aortic wall produced by the withdrawal of the inflated balloon catheter. At 10 days after balloon injury, the expanding intima consisted of 5 to 15 layers of SMCs, of which many had proliferated during the 10-day labeling period with BrdU (1061±469 per cross section, n=6, Table⇓).
Expression of TNF-α in Medial SMCs 2 and 5 Days After Balloon Injury
At 2 days after injury, SMC proliferation detected by BrdU staining was limited to the portion of the media just below the internal elastic lamina. The area of TNF-α expression coincided with this area (Fig 2⇑). Double immunohistochemical staining revealed many medial SMCs that had not divided (BrdU−), surrounded by SMCs that had proliferated during the labeling period (BrdU+) (Fig 3⇓). This result could indicate that expression of TNF-α precedes SMC proliferation, as many of the SMCs in this region will go on to divide during the subsequent 3 days (Table⇑). At 5 days after injury, many medial SMCs had incorporated BrdU, although the extent of medial SMC proliferation varied among regions of the injured media (Fig 3⇓). As noted above, this irregular localization of BrdU+ SMCs in the media may result from inhomogeneous or noncircumferential distribution of mechanical injury to the aortic wall produced by balloon withdrawal. The proliferating medial SMCs clearly colocalized with immunoreactive TNF-α (Fig 3⇓) although some BrdU−/TNF+ SMCs localized within and surrounding dense foci of BrdU+ SMCs. At 5 days after balloon injury, a population of medial SMCs surrounding dense foci of BrdU+ SMCs expressed TNF-α (Fig 3⇓).
Expression of TNF-α, IL-1β, and bFGF in Neointimal and Medial SMCs 10 Days After Balloon Injury
Normal portions of the aorta proximal to the site of injury showed no immunoreactive TNF-α, abundant bFGF associated with medial SMCs as previously reported in rat and human arteries,5 26 and some IL-1β localized to the endothelium (Fig 4⇓). In the injured zone 10 days after balloon injury, the neointima consisted of 5 to 15 layers of SMCs, most of which had undergone division during 10 days after injury as determined by BrdU incorporation (Fig 4⇓). Most of these neointimal SMCs expressed TNF-α. Medial SMCs incorporated BrdU and those SMCs that had proliferated (BrdU+) also stained for TNF-α (Fig 4⇓). In contrast, staining for bFGF in the neointima was much lower than in the underlying media of the injured segment or of the uninjured portion of the same vessel (Fig 4⇓), as previously seen in the rat.5 Neither the media nor intima showed appreciable IL-1β staining (Fig 4⇓). The antibody used in this study does recognize this cytokine in SMCs and endothelial cells in the aortas of rabbits fed an atherogenic diet,27 and preabsorption of this antibody with recombinant human IL-1β reduced this immunoreactivity (data not shown).
Increased TNF-α mRNA in the Injured Aorta
We analyzed RNA extracted from a rabbit aorta 10 days after balloon injury for TNF-α message by reverse transcriptase-PCR analysis. The uninjured aorta contained little or no TNF message detectable by this technique. However, the signal corresponding to the PCR product from the injured zone of the aorta formed an intense ethidium bromide–stained band after 35 cycles of PCR amplification (Fig 5⇓). This increase of TNF-α message in the injured portion of aorta was consistent in five of five rabbits studied. This PCR product comigrated with that amplified from RNA from the spleen of an LPS-treated rabbit used as positive control and corresponded to the expected product based on the primers chosen and the sequence of rabbit TNF-α.
The data presented here establish that balloon injury leads to augmented TNF-α mRNA and protein expression by arterial SMCs. McKeehan and colleagues11 and our laboratory12 have presented evidence that exposure to TNF stimulates SMCs to divide. Thus, locally produced TNF might contribute to SMC proliferation after arterial injury. However, TNF-α itself may not suffice to initiate mitogenesis, as other signals or comitogens might be required for a TNF-producing SMC to undergo division. For example, TNF can regulate the expression of FGF, platelet-derived growth factor, and heparin-binding epidermal growth factor–like molecule.13 14 15
Barath et al22 23 reported that neointimal cells (primarily SMCs) in human atheromatous plaque expressed TNF-α and TNF mRNA, whereas the medial SMCs do not. Neointimal SMCs in atheromatous plaque may represent a modulated phenotype of SMC.28 29 We observed here expression of TNF-α not only in neointimal SMCs but also in proliferating medial SMCs and some BrdU− medial SMCs surrounding dense areas of BrdU+ SMCs. Some of these TNF-α+/BrdU− SMCs may later proliferate or migrate into the intima. Thus, SMC-derived TNF-α may furnish one signal that contributes to phenotypic modulation and proliferation of SMCs after balloon injury. Our data and those from Reidy’s laboratory obtained in the rat5 show much lower levels of immunoreactive bFGF in the neointima after injury than in normal arterial media, suggesting that this growth factor does not contribute to ongoing proliferation of neointimal SMCs. The weak and inconsistent localization of IL-1β in the injured artery does not support a role for this potential growth stimulus in this process either.30
Although proof of a growth-promoting effect of TNF in vivo is lacking, our data establish SMC TNF expression as a novel marker for SMC activation in response to injury. Smooth muscle–derived TNF, if biologically available to neighboring cells, might also enhance nonmitogenic aspects of SMC activation by paracrine or autocrine effects. For example, TNF can increase intercellular adhesion molecule-1 expression by vascular SMCs.9 The expression of this adhesion molecule by SMCs in human atheroma31 32 and in hyperplastic lesions produced by experimental balloon injury1 indicates that augmented intercellular adhesion molecule-1 expression constitutes a marker of SMC activation of considerable in vivo relevance. In addition, TNF, perhaps in conjunction with other cytokines such as interferon-γ, may promote the expression of the inducible form of NO synthase. This enzyme catalyzes the production of the multipotent vascular mediator NO. Recent evidence indicates that balloon injury can induce the expression of an activity that resembles inducible NO synthase in the rat carotid artery.30 Hence, TNF may function as an inducer of this important enzyme as well.
Certain strategies for interfering with TNF action in vivo are becoming available.33 For example, soluble TNF receptors, anti-TNF antibodies, and other similar agents might be used to test critically the function of TNF induced in SMCs by balloon injury. Clausell and colleagues33 have used this approach to demonstrate possible involvement of TNF-α in intimal thickening in the coronary arteries of transplanted hearts. To our knowledge, none have yet characterized the pharmacokinetics of any of these reagents in rabbits. Furthermore, rabbits promptly mount a humoral immune response to foreign proteins. As antigen-antibody complexes can readily enhance intimal lesion formation in this species, administration of foreign proteins may yield unclear results.34 Additionally, achievable local concentrations of exogenously administered neutralizing agents may not prove sufficient to inhibit intracrine or paracrine signaling loops mediated by contact via cell surface–associated forms of TNF. Our present results, however, encourage the exploration of these or other strategies to elucidate the in vivo roles of TNF in vascular pathology. Such modalities might also provide avenues for local therapy in the future. The data presented here provide new evidence for sustained activation of SMCs after balloon injury in vivo and furnish further support for the concept that ongoing inflammatory processes participate in the vascular response to injury.
Selected Abbreviations and Acronyms
|bFGF||=||basic fibroblast growth factor|
|PCR||=||polymerase chain reaction|
|SMC(s)||=||smooth muscle cell(s)|
|TNF||=||tumor necrosis factor|
This work was supported by a grant from the National Heart, Lung, and Blood Institute, PO-1-HL-48743 (incorporating HL-47840).
Tanaka H, Sukhova GK, Swanson SJ, Clinton SK, Ganz P, Cybulsky MI, Libby P. Sustained activation of vascular cells and leukocytes in the rabbit aorta after balloon injury. Circulation. 1993;88:1788-1803.
Lindner V, Majack RA, Reidy MA. Basic fibroblast growth factor stimulates endothelial regrowth and proliferation in denuded arteries. J Clin Invest. 1990;85:2004-2008.
Lindner V, Lappi DA, Baird A, Majack RA, Reidy MA. Role of basic fibroblast growth factor in vascular lesion formation. Circ Res. 1991;68:106-113.
Lindner V, Reidy MA. 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.
Libby P, Goldberg AL. Leupeptin, a protease inhibitor, decreases protein degradation in normal and diseased muscles. Science. 1978;199:534-536.
Libby P, Schwartz D, Brogi E, Tanaka H, Clinton SK. A cascade model for restenosis: a special case of atherosclerosis progression. Circulation. 1992;86(suppl III):III-47-III-52.
Couffinhal T, Duplaa C, Labat L, Lamaziere JM, Moreau C, Printseva O, Bonnet J. Tumor necrosis factor-alpha stimulates ICAM-1 expression in human vascular smooth muscle cells. Arterioscler Thromb. 1993;13:407-414.
Warner SJC, Libby P. Human vascular smooth muscle cells: target for and source of tumor necrosis factor. J Immunol. 1989;142:100-109.
Vilcek J, Palombella VJ, Henriksen-DeStefano D, Swenson C, Feinman R, Hirai M, Tsujimoto M. Fibroblast growth enhancing activity of tumor necrosis factor and its relationship to other polypeptide growth factors. J Exp Med. 1986;163:632-643.
Hajjar KA, Hajjar DP, Silverstein RL, Nachman RL. Tumor necrosis factor-mediated release of platelet-derived growth factor from cultured endothelial cells. J Exp Med. 1987;166:235-245.
Yoshizumi M, Kourembanas S, Temizer DH, Cambria RP, Quertermous T, Lee ME. Tumor necrosis factor increases transcription of the heparin-binding epidermal growth factor-like growth factor gene in vascular endothelial cells. J Biol Chem. 1992;267:9467-9469.
Nawroth PP, Bank I, Handley D, Cassimeris J, Chess L, Stern D. Tumor necrosis factor/cachectin interacts with endothelial cell receptors to induce release of interleukin 1. J Exp Med. 1986;163:1363-1375.
Pober JS, Gimbrone MAJ, Lapierre LA, Mendrick DL, Fiers W, Rothlein R, Springer TA. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol. 1986;137:1893-1896.
Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS, Gimbrone MAJ. Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: characterization and comparison with the actions of interleukin 1. Proc Natl Acad Sci U S A. 1986;83:4533-4537.
Cybulsky MI, McComb DJ, Movat HZ. Neutrophil leukocyte emigration induced by endotoxin: mediator roles of interleukin 1 and tumor necrosis factor a1. J Immunol. 1988;140:3144-3149.
Carlos TM, Schwartz BR, Kovach NL, Yee E, Rosa M, Osborn L, Chi-Rosso G, Newman B, Lobb R, Rosso M, Harlan J. Vascular cell adhesion molecule-1 mediates lymphocyte adherence to cytokine-activated cultured human endothelial cells. Blood. 1990;76:965-970.
Fleet JC, Clinton SK, Salomon RN, Loppnow H, Libby P. Atherogenic diets enhance endotoxin-stimulated interleukin-1 and tumor necrosis factor gene expression in rabbit aortae. J Nutr. 1992;122:294-305.
Salomon RN, Underwood R, Doyle MV, Wang A, Libby P. Increased apolipoprotein E and c-fms gene expression without elevated interleukin 1 or 6 levels indicate selective activation of macrophage functions in advanced human atheroma. Proc Natl Acad Sci U S A. 1992;89:2814-2818.
Brogi E, Winkles J, Underwood R, Clinton S, Alberts G, Libby P. Distinct patterns of expression of fibroblast growth factors and their receptors in human atheroma and non-atherosclerotic arteries: association of acidic FGF with plaque microvessels and macrophages. J Clin Invest. 1993;92:2408-2418.
Tanaka H, Sukhova G, Libby P. Interaction of the allogeneic state and hypercholesterolemia in arterial lesion formation in experimental cardiac allografts. Arterioscler Thromb. 1994;14:734-745.
Campbell GR, Campbell JH, Ang AH, Campbell IL, Horrigan S, Manderson JA, Mosse PRL, Newman RE. Phenotypic changes in smooth muscle cells of human atherosclerotic plaques. In: Glagov S, Newman WPI, Schaffer SA, eds. Pathobiology of the Human Atherosclerotic Plaque. New York, NY: Springer-Verlag; 1989:69-92.
Schwartz SM, Heimark RL, Majesky MW. Developmental mechanisms underlying pathology of arteries. Physiol Rev. 1990;70:1177-1209.
Joly GA, Schini VB, Vanhoutte PM. Balloon injury and interleukin-1 beta induce nitric oxide synthase activity in rat carotid arteries. Circ Res. 1992;71:331-338.
Clausell N, Molossi S, Sett S, Rabinovitch M. In vivo blockade of tumor necrosis factor-alpha in cholesterol-fed rabbits after cardiac transplant inhibits acute coronary artery neointimal formation. Circulation. 1994;89:2768-2779.