Long Pentraxin PTX3 Upregulates Tissue Factor Expression in Human Endothelial Cells
A Novel Link Between Vascular Inflammation and Clotting Activation
Inflammation is a major contributing factor to atherosclerotic plaque development and ischemic heart disease. PTX3 is a long pentraxin that was recently found to be increased in patients with acute myocardial infarction. Because tissue factor (TF), the in vivo trigger of blood coagulation, plays a dominant role in thrombus formation after plaque rupture, we tested the possibility that PTX3 could modulate TF expression. Human umbilical vein endothelial cells, incubated with endotoxin (lipopolysaccharide) or the inflammatory cytokines interleukin-1β and tumor necrosis factor-α, expressed TF. The presence of PTX3 increased TF activity and antigen severalfold in a dose-dependent fashion. PTX3 exerted its effect at the transcription level, inasmuch as the increased levels of TF mRNA, mediated by the stimuli, were enhanced in its presence. The increase in mRNA determined by PTX3 originated from an enhanced nuclear binding activity of the transacting factor c-Rel/p65, which was mediated by the agonists and measured by electrophoretic mobility shift assay. The mechanism underlying the increased c-Rel/p65 activity resided in an enhanced degradation of the c-Rel/p65 inhibitory protein IκBα. In the area of vascular injury, during the inflammatory response, cell-mediated fibrin deposition takes place. Our results suggest that PTX3, by increasing TF expression, potentially plays a role in thrombogenesis and ischemic vascular disease.
Inflammatory events lead to the activation of coagulation, followed by thrombus formation. Among pentraxins, a family of proteins considered to be markers of the acute phase of inflammation,1 a novel group has recently been characterized. PTX3, the first member of this subfamily, named “long” pentraxins, was cloned as an interleukin (IL)-1–inducible and tumor necrosis factor (TNF)-inducible gene in endothelial cells and fibroblasts, respectively.2,3⇓ PTX3 is structurally related to, although distinct from, classic pentraxins, such as C-reactive protein (CRP). Although a role for PTX3 in the amplification of inflammation4 and in its contribution to the prevention of the onset of autoimmune reactions in inflamed tissues has been proposed,5 much remains to be learned about its biological function. Very recently, PTX3 was found in normal and hypertrophied human cardiomyocytes and was increased in the blood of patients with acute myocardial infarction.6 A concomitant disappearance of PTX3 from the necrotic myocytes suggests a role for PTX3 as an indicator of myocyte irreversible injury in ischemic cardiomyopathy.
Strong evidence assigning a central role for tissue factor (TF) in the thrombosis and inflammation associated with atherosclerosis has also been reported.7,8⇓ TF is found in the atheromatous plaque, and its exposure after plaque disruption may lead to thrombosis, vascular occlusion, and myocardial infarction. Moreover, TF is increased in plasma from patients with acute coronary syndromes.7–9⇓⇓ Among the TF-producing cells that participate in plaque formation, endothelial cells play an important role. Indeed, endothelial cells in culture express TF on their membranes after exposure to appropriate stimuli, such as endotoxin (bacterial lipopolysaccharide [LPS])10 and inflammatory cytokines.11
Therefore, we reasoned that PTX3 could function as a modulator of TF expression. In the present study, we show the novel finding that PTX3 enhances the expression of TF on endothelial cells exposed to different pathophysiological stimuli. The mechanism by which PTX3 exerts its effect on TF activity and antigen was also investigated.
Chemicals and Cell Lines
Medium 199 and DMEM were from Biowhittaker Europe. FCS, penicillin, glutamine, and 0.05% trypsin/0.02% EDTA were purchased from GIBCO-BRL. Serum amyloid P (SAP), CRP, heparin, and collagenase were from Sigma Chemical Co. RPMI 1640 medium, PBS, and sodium carbonate were from Biochrom. LPS (Escherichia coli 055:B5) and gelatin were from DIFCO Labs. Recombinant human IL-1β and recombinant human TNF-α were from Peprotec. Tissue-culture dishes were from Falcon Labware Division, Becton Dickinson Co. Sterile pyrogen-free microtubes were obtained from Sarstedt. Chromogenic limulus amebocyte assay was purchased from BioWhittaker Bioproducts, Inc. Human PTX3 was purified from the supernatant of transfected CHO cells as described.4
Cell Isolation and Culture
Endothelial cells were isolated from human umbilical cord vein by digestion with 0.5% collagenase as previously described.12 For experiments, human umbilical vein endothelial cells (HUVECs) were plated at a density of 8×104 cells per well in gelatin-coated 12-well plates and grown to confluence in medium 199/DMEM (1:1) supplemented with 15% FCS, 2 mmol/L l-glutamine, 100 μg/mL streptomycin, 100 U/mL penicillin, 5 mg/mL endothelial cell growth factor, and 10 mg/mL heparin in a humidified atmosphere of 92.5% air/7.5% CO2 at 37°C. Cultured medium was refreshed every other day. The number of HUVECs at confluence was 2×105 cells per well.
After incubation, cells were disrupted by 3 freeze-thaw cycles before testing for procoagulant activity by a 1-stage clotting time test, as previously described.12 The procoagulant activity generated was due to TF, inasmuch as HUVECs incubated with the inhibitory anti-TF antibody HTF1 generated no activity. Results were expressed in arbitrary units (AU) by comparison with a standard curve obtained by use of a human brain thromboplastin standard that was kindly donated by Dr L. Poller, Manchester, UK. This preparation was assigned a value of 1000 AU for a clotting time of 20 seconds. The standard curve was linear from 1000 to 0.01 AU, corresponding to clotting times of 20 and 511 seconds, respectively.
To determine TF antigen, the cells were harvested and lysed in TBS, pH 8.5, containing 1% Triton X-100 overnight. Cell debris were pelleted by centrifugation at 100 000g for 60 minutes at 4°C. The protein content of the supernatant was determined by BCA protein assay kit (Pierce) and adjusted to 2 mg/mL. Samples were then used for ELISA (Imubind TF kit, American Diagnostica Inc), according to the manufacturer’s instructions.
Northern Blot Analysis
For Northern blot analysis, HUVECs were cultured with the different stimuli in the absence or presence of PTX3. At the end of incubation, HUVECs were extensively washed, and total cellular RNA was isolated by the thiocyanate/cesium chloride method, as described.12 As an internal control, filters were also hybridized and autoradiographed by using a 32P-labeled GAPDH. The extent of hybridization was subject to quantitative analysis by Instant Imager (Packard). The quantification for the blots was calculated as the value obtained for TF probing of that blot divided by the same value of its GADPH reprobing and was expressed as a normalized ratio.
Electrophoretic Mobility Shift Assay
To determine the effect of PTX3 on c-Rel/p65 nuclear translocation, HUVECs were cultured with the stimuli for 30 minutes in the presence or absence of PTX3. Nuclear extracts from 2 to 3×106 HUVECs were prepared, and the levels of c-Rel/p65 were monitored by electrophoretic mobility shift assay as previously described.13 A specific κB-like probe for TF was used (5′-GTC CCG GAG TTT CCT ACC GGG-3′, Labtek S.R.L.). Generally, protein concentrations, as determined by Bradford, ranged between 25 and 50 μg.
After experimental treatment, cytoplasmic extracts were prepared. The amount of protein was measured by Bradford assay. The extracts (100 μg total proteins) were then separated by SDS-PAGE and transferred onto nitrocellulose membrane in 25 mmol/L Tris/192 mmol/L glycine/20% methanol for 3 hours at 80 V and 4°C as described previously.14 The blots were blocked for 1 hour in 20 mmol/L Tris (pH 7.6)/137 mmol/L NaCl/0.1% Tween-20 (TBST) containing 5% dry milk. The blots were washed in TBST and incubated with primary antibody for IκBα (Santa Cruz Biotechnology) for 1 hour at room temperature with shaking. After washing with TBST, the blots were incubated with a secondary antibody, and proteins were visualized by chemiluminescence (Amersham).
The results are given as mean±SEM. Statistical analysis was performed by using 1-way ANOVA.
Effect of PTX3 on HUVEC TF Activity and Antigen
When PTX3 (5 μg/mL) was incubated with HUVECs, no procoagulant activity could be observed. By contrast, when HUVECs were stimulated with different concentrations of the classic TF inducers (LPS, IL-1β, or TNF-α), the presence of PTX3 enhanced TF activity (Figure 1). The enhancement could be observed when the induction of TF by the different agonists was maximal and submaximal. To further investigate the effect of PTX3, concentrations of 1 μg/mL for LPS, 10 ng/mL for IL-1β, and 10 ng/mL for TNF-α were selected. PTX3 exerted its effect in a dose-dependent fashion, reaching a plateau between 5 and 10 μg/mL (Figure 2). Further experiments were performed with 5 μg/mL PTX3.
The enhancement of TF activity by PTX3 was accompanied by an increase in detectable TF antigen by ELISA. Little TF protein was detected in resting HUVECs, and no modification could be induced by PTX3 alone (84.4±18.4 versus 89.9±21.1 pg/mL, respectively; n=3). TF antigen was increased in HUVECs exposed to LPS, IL-1β, and TNF-α (1167±379, 1783±409, and 1302±144 pg/mL, respectively; n=3). In the presence of 5 μg/mL PTX3, the TF antigen level of LPS-stimulated, IL-1β–stimulated, or TNF-α–stimulated HUVECs was enhanced to 1657±501, 2586±49, and 2669±3 pg/mL, respectively (n=3).
Because PTX3 belongs to the family of pentraxins that includes classic members such as CRP and SAP, we investigated whether these 2 acute-phase reactants were also modulators of HUVEC TF expression. Neither protein could induce TF activity either in the presence or in the absence of LPS, IL-1β, or TNF-α (not shown).
Regulation by PTX3 of TF mRNA Levels in LPS-Stimulated, IL-1β–Stimulated, or TNF-α–Stimulated HUVECs
To examine the effect of PTX3 on TF mRNA, HUVECs were incubated with or without LPS, IL-1β, or TNF-α in the presence and absence of PTX3 for 1 hour, after which the monolayer was extensively washed, and mRNA was extracted and studied by Northern blot analysis. No mRNA TF product could be detected in control cells. In contrast, the exposure of HUVECs to the different stimuli resulted in the induction of a major 2.3-kb TF mRNA, consistent with the mature TF message (Figure 3). A minor TF transcript of 3.4-kb was also detected. When PTX3, which had no effect when incubated alone with the cells, was present together with the different stimuli, a marked increase in TF mRNA band could be clearly observed. Analysis of GAPDH mRNA showed similar mRNA levels in control and treated cells, indicating that the amount of mRNA loaded was comparable among the experimental groups.
Effect of PTX3 on Activation of c-Rel/p65 Heterodimers in Stimulated HUVECs
To investigate the mechanism by which PTX3 increases TF mRNA synthesis in stimulated HUVECs, we examined c-Rel/p65 heterodimers in HUVECs exposed to the stimulating agents in the absence or in the presence of PTX3. On incubation of the HUVECs with LPS, IL-1β, or TNF-α, a shift in c-Rel/p65 was observed (Figure 4). The presence of PTX3 during the incubation, although not having any effect by itself, determined an increased nuclear accumulation of the transacting factor in all instances.
Effect of PTX3 on Degradation of IκBα in Stimulated HUVECs
Translocation of nuclear factor (NF)-κB to the nucleus requires phosphorylation and subsequent degradation of its inhibitory protein, ΙκβBα. We investigated the effect of PTX3 on the degradation of IκBα in HUVECs stimulated with LPS, IL-1β, or TNF-α. As expected, LPS, IL-1β, and TNF-α treatment resulted in rapid degradation of IκBα (Figure 5). As seen in Figure 3, PTX3, which by itself had no effect, greatly enhanced the proteolytic degradation of IκBα in stimulated HUVECs.
Evidence of the contribution of inflammation to the pathogenesis of ischemic heart disease has been repeatedly provided, and markers of inflammation may predict the risk of coronary heart disease.15 However, the intermediate pathways remain to be clarified. Among the possible mediators, it has been suggested that pentraxins play a role in the pathogenesis of cardiovascular events.6 The presence of SAP and CRP in aortic atherosclerotic lesions has been reported.16,17⇓ In addition, CRP was found to be elevated in plasma during unstable coronary syndromes15 and during episodes of myocardial ischemia6 and is considered to be a major mediator of ischemic myocardial injury.18
PTX3 is a newly discovered marker of the acute-phase response, which belongs to the family of pentraxins, which are structurally related to, but distinct from, the classic members of the same family, SAP and CRP.4 PTX3 was cloned as an IL-1–inducible and TNF-inducible gene in endothelial cells and fibroblasts, respectively.2,3⇓ In contrast to the classic pentraxins, PTX3 is made in minimal amount in the liver, whereas its concentration is higher in muscular tissue, including the heart.19 PTX3 is mostly generated by endothelial cells and mononuclear phagocytes on stimulation with LPS and inflammatory cytokines, such as IL-1β and TNF-α, but not IL-6.2,20⇓
PTX3 production is induced in models of systemic or localized infection, including sepsis, and is detected in vivo at high levels in the hearts of mice injected with bacterial LPS.20 More recently, PTX3 was found to be elevated in critically ill patients, with a gradient from systemic inflammatory response syndrome to septic shock.21 It was also found to be expressed in normal and hypertrophied human cardiomyocytes and to be increased in the blood of patients with acute myocardial infarction.6 Because the increase of PTX3 was concomitant with the disappearance of the molecule from dying or necrotic myocytes, it has been suggested that the necrotic cells could be the source of PTX3. Because its levels were not correlated with those of CRP and SAP, PTX3 may represent a novel and independent indicator of inflammatory components in ischemic heart disease.
Inflammatory processes at the vascular level may be a major cause of endothelial dysfunction.22 During this process, molecules such as cytokines, chemokines, and growth factors, among which several are inducers of monocyte and endothelial cell TF, are generated. The balance between the anticoagulant and procoagulant properties of the endothelial cells participating in the lesion then shifts toward the latter, and the generation of thrombin and local deposition of fibrin are likely to occur. Evidence of the involvement of endothelial cell TF in inflammatory diseases has been provided.23–26⇓⇓⇓ Immunocytochemical analysis of baboons treated with Escherichia coli revealed the presence of TF antigen on endothelial cells, although it was limited to the splenic microvasculature.23 TF mRNA was detected in the endothelial cells lining small and medium blood vessels in an animal model of cardiac xenograft during the onset of acute vascular rejection.24 Moreover, TF antigen has been detected on the endothelial cells overlying the atherosclerotic plaque.25,26⇓
These observations and the presence of PTX3 during the early phases of inflammation offer the biological plausibility for a role of PTX3 in the modulation of endothelial cell procoagulant activity. Indeed, in the present study, we have demonstrated that PTX3 amplifies TF expression in HUVECs exposed to other inflammatory mediators, ie, LPS, IL-1β, and TNF-α.
To assess the specificity of PTX3, we investigated whether the classic pentraxins were active in our system. Our results show that neither CRP nor SAP could modulate TF expression in resting or activated HUVECs, indicating that the effect of PTX3 is specific. CRP has been previously shown to modulate TF expression in peripheral blood monocytes.27 The activity apparently did not require the presence of other exogenously added stimuli, indicating that CRP was an inducer rather than an enhancer of TF generation. Similar to our findings, CRP, tested on HUVECs, did not induce TF synthesis. It was speculated that the differences between monocytes and endothelial cells could reside in the absence of the CRP receptor on cultured endothelial cells. So far, no data are available describing a cell membrane receptor for PTX3, although pentraxin classic ligands are phosphorylcholine and phosphorylethanolamine.28 Recently, it has been reported that PTX3 binds to human leukemia Jurkat T cells.5 The binding, dose dependent and saturable, was observed exclusively on dying cells and, to a lesser extent, on necrotic cells. We do not know whether PTX3 exerts its action on TF synthesis by inducing an intracellular signaling generated by a hypothetical binding of the molecule to the cell membrane. Because inflammatory agents were used in our experimental conditions and because cell death commonly occurs during inflammatory reactions, it is tempting to speculate that our perturbed cells could be in some way associated to PTX3 ligand–presenting dying cells.
The effect of PTX3 is exerted at the level of mRNA as well as the level of protein synthesis. TF gene expression is regulated principally at the level of transcription.29 It is known that in unstimulated endothelial cells, components of the NF-κB transacting factor family are retained in the cytoplasm by the binding of the inhibitor IκBα. On the binding of LPS, IL-1β, or TNF-α to their respective receptors on the cell membrane, IκBα is phosphorylated and degraded, and the heterodimer c-Rel/p65, which belongs to the family of transacting factors NF-κB, migrates to the nucleus, where it binds to a putative κB site in the TF promoter.30 The binding induces TF gene transcriptional activation. This is also the case in our experimental conditions: the present results show that the mechanism by which PTX3 is modulating TF synthesis resides in IκBα phosphorylation and degradation and that it increased the migration of cRel/p65 into the nucleus. The observed enhancement of NF-κB translocation by PTX3 is consistent with the recent finding that activation of NF-κB takes place in patients with unstable angina.31 Moreover, NF-κB has been shown to play a role in ischemia/reperfusion injury32 and in atherosclerotic vessels during accelerated atherosclerosis in rabbits.33
Although the upstream signals for ΙκBα degradation differ according to which agonist/receptor interaction takes place, LPS, IL-1β, and TNF-α intracellular signaling pathways, which are generated on binding to their respective receptors, converge into a step in which protein kinase complexes called ΙκB kinases are involved.34 These kinases, which are then responsible for phosphorylating ΙκBα, are thought to be activated through phosphorylation by an increasing newly discovered number of upstream kinases.35 Attempts to further characterize the involvement of PTX3 in this complex pathway are currently under investigation.
The physiopathological significance of the present study remains to be established. Indeed, although the in vivo expression of TF on endothelial cells is still a controversial issue, an increasing number of studies, possibly because of better sensitivity in detection assays, provide evidence of in vivo expression of TF on endothelial cells.23–26,36⇓⇓⇓⇓ On the other hand, the use of HUVECs to extrapolate the changes that occur during arterial injury may also be questioned for obvious reasons. However, although this could represent a limitation of the present study, a similar behavior has been reported for endothelial cells of arterial and venous origin regarding their capacity of generating TF.11 With these limitations in mind, it is tempting to speculate that PTX3, which was proposed to be an early indicator of acute myocardial infarction,6 might also be involved in the myocardial damage due to vascular ischemia via enhancement of TF.
Inflammation and coagulation are indeed closely interlinked. During the onset of inflammation, molecules inducing the procoagulant signal responsible for the local fibrin deposition are generated. Endothelial cells, at sites of vascular injury and inflammation, actively participate in this general scheme expressing TF on their membrane. PTX3, once generated by or in the proximity of the perturbed endothelial cells, potentiates the expression of TF, which is required for thrombogenesis and vascular ischemia to occur.
This work was supported in part by the Italian National Research Council (Convenzione CNR–Consorzio Mario Negri Sud). The authors thank their colleagues at the Department of Vascular Medicine and Pharmacology for critical reading of this manuscript and fruitful discussions, the assistance of Adriana D’Alessandro, Daniela Sciarra, and Alexandra Cianci for technical help, and the G.A. Pfeiffer Memorial Library staff for their help in the preparation of the manuscript.
Received November 26, 2001; revision accepted January 2, 2002.
- ↵Breviario F, d’Aniello EM, Golay J, Peri G, Bottazzi B, Bairoch A, Saccone S, Marzella R, Predazzi V, Rocchi M. Interleukin-1-inducible genes in endothelial cells: cloning of a new gene related to C-reactive protein and serum amyloid P component. J Biol Chem. 1992; 267: 22190–22197.
- ↵Lee GW, Goodman AR, Lee TH, Vilcek J. Relationship of TSG-14 protein to the pentraxin family of major acute phase proteins. J Immunol. 1994; 153: 3700–3707.
- ↵Bottazzi B, Vouret-Craviari V, Bastone A, De Gioa L, Matteucci C, Peri G, Spreafico F, Pausa M, D’Ettorre C, Gianazza E, et al. Multimer formation and ligand recognition by the long pentraxin PTX3: similarities and differences with the short pentraxins C-reactive protein and serum amyloid P component. J Biol Chem. 1997; 272: 32817–32823.
- ↵Rovere P, Peri G, Fazzini F, Bottazzi B, Doni A, Bondanza A, Zimmermann VS, Garlanda C, Fascio U, Sabbadini MG, et al. The long pentraxin PTX3 binds to apoptotic cells and regulates their clearance by antigen-presenting dendritic cells. Blood. 2000; 96: 4300–4306.
- ↵Peri G, Introna M, Corradi D, Iacuitti G, Signorini S, Avanzini F, Pizzetti F, Maggioni AP, Moccetti T, Metra M, et al. PTX3, a prototypical long pentraxin, is an early indicator of acute myocardial infarction in humans. Circulation. 2000; 102: 636–641.
- ↵Lorenzet R, Napoleone E, Celi A, Pellegrini G, Di Santo A. Cell-cell interaction and tissue factor expression. Blood Coagul Fibrinolysis. 1998; 9 (suppl 1): S49–S59.
- ↵Colucci M, Balconi G, Lorenzet R, Pietra A, Locati D, Donati MB, Semeraro N. Cultured human endothelial cells generate tissue factor in response to endotoxin. J Clin Invest. 1983; 71: 1893–1896.
- ↵Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS, Gimbrone MA Jr. 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.
- ↵Napoleone E, Di Santo A, Lorenzet R. Monocytes upregulate endothelial cell expression of tissue factor: a role for cell-cell contact and cross-talk. Blood. 1997; 89: 541–549.
- ↵Napoleone E, Di Santo A, Camera M, Tremoli E, Lorenzet R. Angiotensin-converting enzyme inhibitors downregulate tissue factor synthesis in monocytes. Circ Res. 2000; 86: 139–143.
- ↵Pierce JW, Schoenleber R, Jesmok G, Best J, Moore SA, Collins T, Gerritsen ME. Novel inhibitors of cytokine-induced Ικβα phosphorylation and endothelial cell adhesion molecule expression show anti-inflammatory effects in vivo. J Biol Chem. 1997; 272: 21096–21103.
- ↵Li XA, Hatanaka K, Ishibashi-Ueda H, Yutani C, Yamamoto A. Characterization of serum amyloid P component from human aortic atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 1995; 15: 252–257.
- ↵Introna M, Alles VV, Castellano M, Picardi G, De Gioia L, Bottazzi B, Peri G, Breviario F, Salmona M, De Gregorio L, et al. Cloning of mouse PTX3, a new member of the pentraxin gene family expressed at extrahepatic sites. Blood. 1996; 87: 1862–1872.
- ↵Alles VV, Bottazzi B, Peri G, Golay J, Introna M, Mantovani A. Inducible expression of PTX3, a new member of the pentraxin family, in human mononuclear phagocytes. Blood. 1994; 84: 3483–3493.
- ↵Cermak J, Key NS, Bach RR, Balla J, Jacob HS, Vercellotti GM. C-reactive protein induces human peripheral blood monocytes to synthesize tissue factor. Blood. 1993; 82: 513–520.
- ↵Oeth P, Parry GCN, Mackman N. Regulation of the tissue factor gene in human monocytic cells: role of AP-1, NF-κB/Rel, and Sp1 proteins in uninduced and lipopolysaccharide-induced expression. Arterioscler Thromb Vasc Biol. 1997; 17: 365–374.
- ↵Read MA, Whitkey MZ, Williams AJ, Collins T. NF-kappa B and I kappa Bα: an inducible regulatory system in endothelial activation. J Exp Med. 1994; 179: 503–512.
- ↵Ritchie ME. Nuclear factor-κB is selectively and markedly activated in humans with unstable angina pectoris. Circulation. 1998; 98: 1707–1713.
- ↵Sawa Y, Morishita R, Suzuki K, Kagisaki K, Kaneda Y, Maeda K, Kadoba K, Matsuda H. A novel strategy for myocardial protection using in vivo transfection of cis element “decoy” against 78 Fκβ binding site: evidence for a role of 78 Fκβ in ischemia-reperfusion injury. Circulation. 1997; 96 (suppl II): II-280–II-285.
- ↵Hernández-Presa M, Bustos C, Ortego M, Tunon J, Renedo G, Ruiz-Ortega M, Egido J. Angiotensin-converting enzyme inhibition prevents arterial nuclear factor-κB activation, monocyte chemoattractant protein-1 expression, and macrophage infiltration in a rabbit model of early accelerated atherosclerosis. Circulation. 1997; 95: 1532–1541.
- ↵May MJ, Ghosh S. ΙκB kinases: kinsmen with different crafts. Science. 1999; 284: 271–273.
- ↵Oitzinger W, Hofer-Warbinek R, Schmid JA, Koshelnick Y, Binder BR, de Martin R. Adenovirus-mediated expression of a mutant IkappaB kinase 2 inhibits the response of endothelial cells to inflammatory stimuli. Blood. 2001; 97: 1611–1617.