Endothelial Determinants of Dendritic Cell Adhesion and Migration
New Implications for Vascular Diseases
Objective— Atherosclerosis is a chronic disease triggered by endothelial injury and sustained by inflammation. Dendritic cells (DCs) are critical for the cell-mediated arm of an immune response and are known to influence inflammatory immunity. A fundamental aspect of DC function is their capacity to adhere and migrate through vascular endothelial cells (ECs). We investigated the role of endothelial activation and dysregulation of the NO pathway on DC adhesion and migration.
Methods and Results— We discovered that DC adhesion and migration are modulated by changes in endothelial function. DC adhesion and transmigration were markedly increased after exposing ECs to hypoxia, oxidized low density lipoprotein, or tumor necrosis factor-α. Specifically, inhibition of endothelial NO synthase increased DC binding and transmigration. l-Arginine or 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibition partially decreased DC-EC interaction.
Conclusions— The results of this study suggest that the adhesion and migration of DCs are increased by stimuli known to accelerate atherogenesis. Vice versa, augmentation of endothelial NO synthase activity prevents DC adhesion. These findings may provide insight into the inflammatory processes occurring in atherosclerosis. Because DCs control immunity, regulating DC-EC interaction may be relevant to inflammation and atherogenesis.
Atherosclerosis begins as an inflammatory immunological disease.1 Atherosclerotic lesions are the result of a series of highly specific cellular and molecular responses to various endogenous risk factors and antigenic stimuli. The inflammatory response is mediated by functional changes in endothelial cells (ECs), T lymphocytes, monocyte-derived macrophages, and smooth muscle cells.2 Activation of these cells leads to the elaboration of a wide spectrum of inflammatory hydrolases, cytokines, chemokines, adhesion molecules, and growth factors, together with lipid accumulation and proliferation of smooth muscle cells and fibroblasts.1,2⇓ Immunohistochemical studies have revealed mononuclear cell infiltrations and accumulation of activated T cells in early and late atherosclerotic lesions.3 The mechanisms of T-cell activation and antigen acquisition in atherogenesis are incompletely characterized. Recently, T-cell activation has been attributed to dendritic cells (DCs).
DCs are derived from monocyte precursors and are believed to play a central role in controlling immunity by activating T lymphocytes.4 DCs are involved in the initiation and progression of various diseases, including tumors, viral and bacterial infections, autoimmune diseases, and transplant rejection.4 A fundamental aspect of DC function is the capacity of DCs to migrate. They perform ongoing surveillance for incoming antigens, which induce them to stimulate T cells in secondary lymphoid organs.5 To exit from the bloodstream, DCs first need to tether to the ECs. E- and P-selectins mediate this initial process of endothelial DC rolling.6 The initial contact is followed by the development of a more firm adhesion between DCs and ECs, which is postulated to be required for subsequent transmigration. Mediators involved in DC adhesion and endothelial transmigration include CC-chemokines, lymphocyte function–associated antigen [LFA]-1 and CD11b, intercellular adhesion molecule (ICAM)-2, and its ligand the DC-specific C-type lectin DC-SIGN.7–9⇓⇓ Previous studies investigating vascular DCs have demonstrated their occurrence in atherosclerotic lesions as well as in arterial regions that are known to be predisposed for the later development of atherosclerosis, namely, turbulent flow regions.10,11⇓
Endothelium-derived NO inhibits endothelium-leukocyte interaction.12 NO decreases the expression of adhesion molecules such as CD11/CD18, P-selectin, vascular cell adhesion molecule (VCAM)-1, and ICAM-1. Moreover, endothelial NO is a survival factor, protecting the vasculature from apoptosis and oxidative damage.13 Endothelial activation and dysfunction, as well as increased vascular inflammatory stimulation, precede the development of atherosclerosis and transplant vasculopathy.1,14⇓ Although the general principles of leukocyte trafficking are similar, it is becoming increasingly apparent that the underlying cellular and molecular mechanisms are highly dependent on the inflammatory stimulus, the specific leukocyte population, and the tissue involved. We hypothesized that DC adhesion and migration are modulated by endothelial activation. In the present study, we characterize the capacity of blood DCs to adhere to unstimulated and activated/apoptotic macrovascular and microvascular endothelium. We further characterized DC interaction (adhesion and transmigration) with ECs after modulation of the endothelial NO synthase (NOS) pathway. Finally, we describe potential molecular regulators of DC adhesion to activated ECs.
Generation of DCs
Production of Autologous MCM
We adopted techniques from Thurner et al15 to produce monocyte-conditioned medium (MCM). IgG-coated bacteriological plates (85 mm, Falcon) were prepared not later than 30 minutes before use. The plates were coated with D-PBS (10 mL, BioWhittaker) supplemented with IgG for 30 minutes at room temperature. After the plates were coated, they were rinsed twice with D-PBS. Peripheral blood mononuclear cells (PBMCs, 8×107) were plated in RPMI 1640 (BioWhittaker) containing 1% autologous serum for 24 hours (37°C, 5% CO2). Supernatant was harvested and then centrifuged (1350g, 5 minutes, 25°C) and sterilely filtered (0.22 μm, Millipore) to divide into aliquots and stored at −20°C.
Isolation of Monocytes
Peripheral blood was obtained from the Stanford Medical School Blood Center as standard buffy coat preparations from normal donors. The buffy coat (35 mL mixed at 1:1 with PBS) was layered over Ficoll (15 mL) and centrifuged (30 minutes, 1300g, room temperature). PBMCs were harvested from the interface and washed 3 times with PBS containing 1% autologous serum. PBMCs were depleted of T, B, and natural killer cells by means of an immunomagnetic technique (Dynabeads, No. 113.09, Dynal). The kit contains anti-CD2, -CD7, -CD16 (a and b), -CD56, and -CD19 monoclonal antibodies. Dynabeads were washed 4 times with PBS containing 1% autologous serum. PBMCs (107/mL) were incubated with magnetic beads at 4°C for 30 minutes by using the Dynal mixer. Lymphocytes were then depleted by means of the Dynal magnet. Nonadherent cells were harvested and washed once, and depletion was repeated once. The Dynal kit isolates monocytes with >98% viability and >94% yield and purity. Fluorescence-activated cell sorter scan revealed the starting cell population to be CD14+ with concentrations of T cells/natural killer cells of <5%; these cells are referred to below as monocytes.
Generation of Immature and Mature DCs From Monocytes
We adopted techniques from Romani et al16 to generate immature DCs (or precursor DCs, referred to as pDCs; 7 days of culture) and mature DCs (10 days of culture, referred as DCs). Lymphocyte-depleted PBMCs were plated in 6-well culture plates at a density of 2×106 cells per well in complete medium (CM, 3 mL). CM was RPMI 1640 containing 1% autologous serum supplemented with granulocyte-macrophage colony–stimulating factor (800 U/mL, Peprotech) and interleukin-4 (1000 U/mL, Peprotech). Cultures were fed every other day (days 2, 4, and 6) by removing 1.5 mL and adding back 1.5 mL fresh CM. On day 7, buoyant cells were harvested, depleted once again, and split to analyze, or they were taken for experiments as pDCs or transferred to new 6-well plates and further cultured. Seven-day pDCs were kept in culture until day 10 in the presence of MCM and CM at a ratio of 1:3 and were finally harvested and analyzed, or they were used for experiments as mature DCs. DCs and pDCs were labeled with 5-chloromethylfluorescein diacetate (CellTracker Green CMFDA, Molecular Probes) and used for adhesion/transmigration assays not later than 1 hour after labeling.
Human dermal microvascular ECs (HMVECs) and human aortic ECs (HAECs) were purchased from Clonetics and cultured in EC basal medium (Clonetics) supplemented with Bulletkit (Clonetics), including 10% FBS, human epidermal growth factor (10 ng/mL), bovine brain extract (12 μg/mL), hydrocortisone (1.0 μg/mL), gentamycin (50 μg/mL), and amphotericin B (50 ng/mL). ECs were grown at 37°C in a humidified 5% CO2 and split once a week. HAECs and HMVECs were used between passages 3 and 6. In some experiments, ECs were exposed to 2% O2 over 24 hours by using hypoxic chambers (Billups-Rothenberg) and a gas analyzer.
Human ECs exposed to the above conditions and cultured in 8-well chamber slides (LabTekII) were washed with HBSS (Irvine Scientific) containing (in mmol/L) CaCl2 2, MgCl2 2, and HEPES 20. Fluorescence labeling of DCs was achieved by incubating DCs with 1:2000 dilution (in binding buffer to remove cytokines and growth factors) of CMFDA twice for 30 minutes each, followed by washing to remove free dye. Chamber slides were then placed on a rotating platform, and fluorescently labeled DCs (2×105/mL) were incubated with ECs for 4 hours at 37°C under constant flow conditions. Medium was aspirated and replaced with fresh HBSS to remove nonadherent cells. After a second washing, dishes were returned to the rocker platform for an additional 5 minutes. Medium was again aspirated and replaced with HBSS containing 3.7% paraformaldehyde. In some experiments, the cells were incubated with fluorescent annexin V (annexin V, Alexa Fluor 594, Molecular Probes) over 30 minutes before fixation for detection of early apoptosis. After 20 minutes of fixation and washing, adherent cells were immediately quantified by using a confocal microscope. All adhesion experiments were performed with ≥5 different preparations of DCs at least in triplicate. Endothelial adhesion of prestained DCs was analyzed in 6 independent high-power fields for each experiment (see Confocal Microscopy).
For transmigration assays, fibronectin-coated transwell inserts (Millipore) were used to culture endothelial monolayers in 24-well plates. Just before the assay, the monolayers were rinsed with serum-free HBSS, and prelabeled DCs were added to the upper chamber. At the same time, 200 μL serum-free medium, supplemented with 100 ng/mL monocyte chemotactic protein (MCP)-1 (R&D Systems), was added to the lower chamber. Transmigration was stopped after 2, 4, and 8 hours, and transmigrated cells were immediately counted in the lower chamber with the use of a fluorescence multireader (GENios, Tecan).17 Relative fluorescence intensity has been shown to increase linearly with cell number.17 The migration index was calculated on the basis of the ratio between the absolute numbers of migrated cells to the DC-EC binding ratio (DC adhesion and transmigration were determined from the same well). All transmigration experiments were performed with ≥5 different preparations of DCs at least in duplicate.
Imaging was performed at Stanford’s Cell Sciences Imaging Facility on a MultiProbe 2010 laser confocal microscope (Molecular Dynamics). The MultiProbe uses an argon/krypton mixed gas laser with excitation lines of 488, 568, and 647 nm and is built on a Nikon Diaphot 200 inverted microscope. To collect the dual signal, the sample was excited with 488- and 568-nm lines, and the emitted light was passed through a 488/568/647-nm triple dichroic beam splitter. The emitted light corresponding to FITC was collected with a 515- to 545-nm bandpass filter, and Alexa Fluor 594 emission was collected with a 590-nm long-pass filter. A Nikon ×60 (numerical aperture 1.4, Planapo) objective was used. Images of serial optical sections were recorded every 1.5 μm per vertical step along the z-axis. Five sections in each of the 6 random fields were analyzed offline for each experiment in a blinded fashion, and each experiment was performed at least in triplicate. The average number of DCs adhering to ECs was calculated and expressed as the DC/EC ratio. Interobserver and intraobserver variability was ≤4.5%.
The following drugs were used: The substrate for NOS, l-arginine (Sigma), was used to enhance NO synthesis, and the calcium ionophore (A23187, Sigma) was used as a receptor-independent stimulus of NOS. Asymmetric dimethyl l-arginine (ADMA, Sigma), an endogenous NOS inhibitor, was used to block endothelial NOS activity. Cerivastatin, a 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase inhibitor (statin), was kindly provided by Bayer AG (Leverkusen, Germany). Atorvastatin was a gift from U. Laufs (Homburg, Germany). Oxidized LDL (oxLDL) was prepared by incubation of LDL (100 μg/mL) with CuSO4 (10 μg/mL) at 37°C for 24 hours and donated by P. Tsao (Stanford University, Stanford, Calif).
Blocking Monoclonal Antibodies
Endothelial dysfunction may be associated with enhanced expression of the adhesion molecules ICAM-1, VCAM-1, and platelet and endothelial cell adhesion molecule (PECAM)-1.18 Therefore, we were interested in describing their roles as potential regulators of DC interaction with activated ECs by the use of blocking antibodies. Anti-human PECAM-1 (HEC7, 20 μg/mL), VCAM-1 (1G11, 50 μg/mL), and LFA-1α (CD11a, TS1/22, 50 μg/mL) were obtained from Endogen, and anti–ICAM-1 (11C81, 50 μg/mL) was purchased from R&D Systems. All monoclonal antibodies have been previously characterized as specific blocking antibodies in ECs.19,20⇓
Data are expressed as mean±SEM. Comparisons of multiple mean values were by ANOVA, followed by the Student-Newman-Keuls post hoc test for statistical significance. A value of P<0.05 was accepted as statistically significant.
Characterization of Cultured DCs
HLA-DR+, CD14+, CD11c+, and lineage-negative DC precursors were cultured for 7 days in serum-free media supplemented with granulocyte-macrophage colony–stimulating factor and interleukin-4 to give rise to immature and mature DCs. Immature DCs were characterized as CD80+/−, CD1a+, and CD11c+. Further cultivation with MCM until day 11 yielded mature DCs with upregulated CD80, CD83, and CD86 (please see online Figure I, available at http://atvb.ahajournals.org).
Adhesion of DCs to Normoxic HMVEC Monolayers
Table 1 shows adhesion of DCs to normoxic HMVEC monolayers. In vitro adhesion of unstimulated (immature) DCs to monolayers of HMVECs was 30% higher than DC adhesion to macrovascular HAECs. Mature DCs were less adhesive than were immature DCs to either ECs. Results could be confirmed with the use of HAECs (data not shown).
Adhesion of Unstimulated DCs to Preactivated HMVEC Monolayers
Activation of ECs usually occurs in the early stages of atherosclerosis and transplant vasculopathy and is associated with EC dysfunction. To mimic EC activation in vivo, we incubated ECs under hypoxic conditions (2% O2 for 24 hours) or exposed them to tumor necrosis factor (TNF)-α (100 ng/mL) or oxLDL (10 μg/mL).21,22⇓ The concentrations chosen mimic physiological conditions in vivo and are not associated with cell-injurious effects (10-fold higher concentrations resulted in enhanced EC necrosis; data not shown). For all different activation procedures, the adhesion of DCs was significantly enhanced compared with that of resting ECs (Figure 1). In the presence of calcium ionophore (A23187, 1 μmol/L), an agonist that promotes NOS, hypoxia- and oxLDL-mediated DC adhesion was suppressed by 39% (Figure 1).
Because the induction of apoptosis was enhanced in hypoxic ECs (17±5% versus 6±4% in normoxic ECs, P<0.01), we used triple-labeling experiments to characterize the adhesion of DCs on apoptotic ECs. We found that DC binding affinity was significantly enhanced in apoptotic ECs (annexin V positive) compared with nonapoptotic ECs (42±11%, P<0.01). However, the majority of DCs binds to nonapoptotic ECs.
Modulation of DC-EC Binding: Role of NO
To investigate the role of NO in DC-EC binding, ECs were incubated with the endogenous NOS inhibitor ADMA (1 to 10 μmol/L). We detected a time- and concentration-dependent increase of DC binding in ECs exposed to ADMA that was partially inhibited by adding l-arginine (1 mmol/L), the substrate for NOS (Figure 2).
ECs pretreated with the HMG-CoA reductase inhibitors cerivastatin or atorvastatin (0.01 μmol/L) demonstrated significantly decreased DC adhesion (Figure 2) that was partially inhibited by 10 μmol/L ADMA (61±17% inhibition, P<0.01).
Adhesion Molecules Involved in DC-EC Binding
The identification of adhesion molecules involved in the enhanced DC binding after hypoxia was performed by using functional blocking monoclonal antibodies against cell adhesion molecules. Anti–ICAM-1 antibody partially inhibited DC adhesion to normoxic ECs and hypoxic ECs. Anti–VCAM-1 and anti-PECAM antibodies did not affect DC binding to unstimulated ECs, whereas anti–PECAM-1 antibody partially inhibited DC adhesion to hypoxic ECs. Preincubation of DCs with anti–LFA-1 (DC counterreceptor to ICAM-1) inhibited DC adhesion on hypoxic ECs. Combined blockade of LFA-1 and ICAM-1 inhibited DC adhesion between 62% and 84%. Results in Table 2 are shown for HMVECs and could be confirmed by using HAECs (data not shown). Higher concentrations of antibodies (up to 100 μg/mL for each antibody) did not result in significant changes in DC adhesion (data not shown).
Transmigration of DCs Through HMVECs
Endothelial migration of DCs is a crucial step after endothelial adhesion. To uncover factors relevant for DC migration, a migration index was calculated on the basis of the ratio between the absolute numbers of migrated cells to the DC/EC binding ratio. EC activation with TNF-α (100 ng/mL), oxLDL (10 μg/mL), or hypoxia enhanced DC migration by 23±7%, 48±13%, and 19±10%, respectively (P<0.01 versus unstimulated ECs).
DC migration was unaffected by preincubation with blocking ICAM-1 and VCAM-1 antibodies and was slightly decreased by LFA-1 blockade. In contrast, inhibition of PECAM-1 reduced DC migration by 52% (Figure 3a). The combined use of ICAM-1 and PECAM-1 antibodies had no further inhibitory effect on DC migration (Figure 3a). However, the absolute number of transmigrated DCs (net effect of adhesion and migration) was significantly reduced by the combined use of ICAM-1 and PECAM-1 antibodies (−72±14% versus −45±9% in ECs treated with anti–PECAM-1 and −24±4% in ECs treated with anti–ICAM-1, P<0.001).
ADMA (10 μmol/L) time-dependently enhanced DC migration, whereas cerivastatin and atorvastatin significantly blocked DC migration. The latter effect was inhibited in part by 10 μmol/L ADMA (38±20% inhibition, P<0.05). EC preincubation with calcium ionophore (1 μmol/L) significantly suppressed DC migration (Figure 3b). We could partially abrogate the A23187 effects on DC migration by use of coincubation with 100 μmol/L NG-monomethyl-l-arginine (−39±7%, P<0.01) or 10 μmol/L ADMA (−48±15%, P<0.001). Symmetric dimethylarginine, which is not an inhibitor of NOS, had no effect on A23187-mediated blockade of DC migration.
The salient findings of this investigation are as follows: (1) endothelial adhesion and migration of DCs is increased by stimuli known to accelerate atherosclerosis, (2) enhanced endothelial NO activity inhibits DC-EC interaction, and (3) HMG-CoA reductase inhibition reduces endothelial DC transmigration.
Inflammation and Atherosclerosis
T lymphocytes and macrophages are present in atherosclerotic plaques as components of the inflammatory process. Cytokines released by the T lymphocytes are believed to play an important role in monocyte recruitment.1 However, it remains undefined how T lymphocytes are recruited to the lesion. In other inflammatory disorders, DCs are critical for the induction of primary and T-cell–mediated immune responses.4 It is conceivable that immature DCs traverse the vessel wall to identify and capture antigens (eg, viral and bacterial antigens, oxLDL, heat shock proteins 60/65, and advanced glycosylation end products) and subsequently migrate (as mature DCs) to lymphoid organs to stimulate naive T lymphocytes, therefore establishing the adaptive arm of an immune response.
Endothelial Activation and DCs
The present study indicates that DC adhesion and migration can be increased by endothelial activation. ECs exposed to oxLDL, TNF-α, or hypoxia are much more vulnerable to DC attachment and transmigration than are nonactivated ECs. Of interest, EC apoptosis markedly enhanced DC adhesion. These results support the concept that endothelial activation accelerates DC-mediated immune activation, finally leading to enhanced vascular inflammation. In this regard, it has been shown that accumulation of DCs occurs in atherosclerotic areas23 and in vascular regions predisposed to atherosclerosis, ie, those regions exposed to high cyclic strain and low shear stress.11 These vascular regions are also sites of endothelial dysfunction.24
Evidence suggests that NO is an endogenous antiatherogenic molecule.12 Accordingly, we wished to characterize its role in EC-DC interaction. We demonstrated that DC interaction (adhesion and transmigration) with ECs is strongly enhanced after blocking endothelial NOS activity by using the endogenous NOS inhibitor ADMA. Moreover, the precursor of NOS, l-arginine, partially abolished the increased DC adhesion to ECs exposed to ADMA or hypoxia and oxLDL. Calcium ionophore as a receptor-independent stimulus of NOS strongly diminished DC adhesion and transmigration. These results extend previous investigations that revealed that endogenous NO can decrease endothelial monocyte adhesion.25 Specific mechanisms by which NO interferes with DC transmigration might include alterations in adhesion molecule expression, changes in endothelial permeability, and modification of gap junction function. Moreover, data from one of our earlier reports suggest that NO activity may regulate monocyte migration by affecting EC oxidative stress.25 Future studies should focus on the role of oxidative stress and nuclear factor-κB–regulated transcriptional pathways on DC-EC interaction.
HMG-CoA Reductase Inhibition and DC-EC Interaction
Statins have been shown to protect endothelial function,26 and they are powerful drugs for the prevention of cardiovascular diseases. The beneficial effects of statins may be partially independent of their lipid-lowering potential. Statins increase endothelial NO bioactivity, decrease endothelial apoptosis, and inhibit smooth muscle cell proliferation.22,27⇓ In the present study, we demonstrate for the first time that physiological concentrations of statins decrease DC adhesion and transmigration. Statin-mediated effects were at least in part mediated by NO, inasmuch as coincubation with the NOS inhibitor ADMA diminished the effect of statins. Thus, by enhancing NO synthesis, statins may diminish inflammatory activity (eg, DC invasion and T-cell activation) in the phase of early atherosclerosis.
DC Invasion and Adhesion Molecules
To determine which adhesion molecules might be mediating enhanced DC adhesion on activated ECs, functional binding studies were performed. DC adhesion to hypoxic ECs was partially mediated by ICAM-1, whereas monocyte binding was primarily mediated by VCAM-1. Notably, stimulation of NOS by the calcium ionophore reduced ICAM-1 expression on ECs (data not shown).
Unlike adhesion, migration of DCs is not blocked by anti–ICAM-1 antibodies. Our investigation that DC migration is reduced by PECAM-1 antibodies extends previous findings by D’Amico et al8 obtained with human umbilical vein ECs to primary microvascular ECs. Moreover, we demonstrate for the first time that the combined use of anti–ICAM-1 and anti–PECAM-1 antibodies inhibits the absolute number of transmigrated DCs (as a net effect of adhesion and transmigration) by 70%. The scheme in Figure 4 illustrates the role of adhesion molecules ICAM-1, PECAM-1, and VCAM-1 in EC-DC interaction. Other potential adhesion molecules that might be responsible for endothelial adhesion or transmigration of DCs are CD18 and ICAM-2.8,9,28⇓⇓
In summary, DC adhesion and transmigration is enhanced by TNF-α, oxLDL cholesterol, and hypoxia, effects that are partially mediated by the adhesion molecules ICAM-1 (adhesion) and PECAM-1 (transmigration). Stimulation of the endothelium to release NO inhibits adhesion and transmigration of DCs, whereas inhibition of NO synthesis increases DC interaction with the endothelium. Statins decrease DC adhesion and transmigration, an effect mediated in part by NO.
In conclusion, we have characterized important determinants of DC adhesion to ECs. The present study indicates that alterations in endothelial function have a significant impact on DC adherence to the endothelium. Endothelial activation and DC recruitment may play an important role in atherosclerosis and transplant vasculopathy. Delineation of the determinants of DC-EC interaction may lead to new therapeutic avenues. Inhibition of DC-EC interaction may have an application in reducing the progression of cardiovascular disease. Conversely, enhancing EC-DC interaction may be a novel strategy for immunotherapy by initiating T-cell responses against cancer and viral and bacterial infections.29,30⇓
This study was supported in part by grants from the National Institutes of Health (grant CA09151-26 to Drs Schlichting and Engleman) and the National Heart, Lung, and Blood Institute (grant R01 HL-58638 to Dr Cooke). Dr Schlichting was a fellow of the Stanford University Tumor Biology Training Program. Dr Weis was supported by the German Society of Cardiology, Germany. Dr Cooke is an Established Investigator of the American Heart Association.
↵*These authors contributed equally to the present study.
Received August 7, 2002; revision accepted August 23, 2002.
- ↵Libby P. Molecular bases of the acute coronary syndromes. Circulation. 1995; 91: 2844–2850.
- ↵Sallusto F, Lanzavecchia A. Mobilizing dendritic cells for tolerance, priming, and chronic inflammation. J Exp Med. 1999; 189: 611–614.
- ↵Robert C, Fuhlbrigge RC, Kieffer JD, Ayehunie S, Hynes RO, Cheng G, Grabbe S, von Andrian UH, Kupper TS. Interaction of dendritic cells with skin endothelium: a new perspective on immunosurveillance. J Exp Med. 1999; 189: 627–636.
- ↵D’Amico G, Bianchi G, Bernasconi S, Bersani L, Piemonti L, Sozzani S, Mantovani A, Allavena P. Adhesion, transendothelial migration, and reverse transmigration of in vitro cultured dendritic cells. Blood. 1998; 92: 207–214.
- ↵Millonig G, Niederegger H, Rabl W, Hochleitner BW, Hoefer D, Romani N, Wick G. Network of vascular-associated dendritic cells in intima of healthy young individuals. Arterioscler Thromb Vasc Biol. 2001; 21: 503–508.
- ↵Dimmeler S, Zeiher AM. Endothelial cell apoptosis in angiogenesis and vessel regression. Circ Res. 2000; 87: 434–439.
- ↵Lum H, Roebuck KA. Oxidant stress and endothelial cell dysfunction. Am J Physiol. 2001; 280: C719–C741.
- ↵Muller WA, Weigl SA, Deng X, Phillips DM. PECAM-1 is required for transendothelial migration of leukocytes. J Exp Med. 1993; 178: 449–460.
- ↵Meyer DM, Dustin ML, Carron CP. Characterization of intercellular adhesion molecule-1 ectodomain (sICAM-1) as an inhibitor of lymphocyte function-associated molecule-1 interaction with ICAM-1. J Immunol. 1995; 155: 3578–3584.
- ↵Hoffmann J, Haendeler J, Zeiher AM, Dimmeler S. TNFalpha and oxLDL reduce protein S-nitrosylation in endothelial cells. J Biol Chem. 2001; 276: 41383–41387.
- ↵Weis M, Heeschen C, Glassford AJ, Cooke JP. Statins have biphasic effects on angiogenesis. Circulation. 2002; 105: 739–745.
- ↵Bobryshev YV, Lord RS. Mapping of vascular dendritic cells in atherosclerotic arteries suggests their involvement in local immune-inflammatory reactions. Cardiovasc Res. 1998; 37: 799–810.
- ↵Tsao PS, Buitrago R, Chan JR, Cooke JP. Fluid flow inhibits endothelial adhesiveness: nitric oxide and transcriptional regulation of VCAM-1. Circulation. 1996; 94: 1682–1689.
- ↵Koh K. Effects of statins on vascular wall: vasomotor function, inflammation, and plaque stability. Cardiovasc Res. 2000; 47: 648–657.