Fractalkine Upregulates Intercellular Adhesion Molecule-1 in Endothelial Cells Through CX3CR1 and the Jak–Stat5 Pathway
Fractalkine (FKN) is a membrane-bound chemokine that can be released by proteolysis to produce soluble FKN (s-FKN). FKN and its receptor, CX3CR1, are believed to be important factors in atherosclerosis and may play a role in acute inflammatory responses. Although FKN is expressed on endothelial cells (ECs), CX3CR1 is reported to reside mainly on certain leukocyte populations. RT-PCR and Western blotting demonstrated that both human coronary artery and umbilical vein ECs expressed CX3CR1 mRNA and protein. Confocal microscopy showed that CX3CR1 was located at the cell membrane and to a lesser extent in the cytoplasm. Following exposure of both types of ECs to hypoxia and reoxygenation, FKN expression increased rapidly and s-FKN was shed into the culture medium. The addition of s-FKN protein to cultured ECs resulted in a dose-dependent increase in intercellular adhesion molecule (ICAM)-1 mRNA. Perfusion of mouse hearts with s-FKN protein increased expression of ICAM-1 protein in vascular endothelium. Transfection of ECs with CX3CR1-interfering RNA to knockdown the receptor resulted in decreased ICAM-1 expression after s-FKN stimulation. In addition, when ECs were stimulated with s-FKN, greater adhesion of human neutrophils to the ECs was observed. This increase was ICAM-1 dependent and was blocked by CX3CR1 knockdown. Following exposure to s-FKN, ECs exhibited increased phosphorylation of Jak2 and Stat5 and the ICAM-1 expression induced by s-FKN was blocked by silencing of Stat5 with small interfering RNA. Vascular ECs express both FKN and its receptor CX3CR1. s-FKN is shed from ECs following hypoxia/reoxygenation and acts through CX3CR1 on ECs to increase ICAM-1 expression and promote neutrophil adhesion through activation of the Jak–Stat5 pathway.
Fractalkine (FKN) is a unique membrane-bound molecule expressed on endothelial cells (ECs), possessing a chemokine domain and an extended mucin-like stalk that allows it to function as both a chemoattractant and an adhesion molecule.1 The chemokine domain contains 2 cysteines separated by 3 other amino acids (CXXXC) and is therefore designated as CX3C ligand-1. Its receptor was identified as a G protein–coupled receptor (GPCR) with high sequence similarity to the genes encoding human chemokine receptors for monocyte attractant protein-1 and macrophage inflammatory protein-1A,2 and has been designated CX3C receptor1 (CX3CR1). CX3CR1 is believed to reside mainly on certain leukocyte populations, including macrophages, lymphocytes, and natural killer cells but has not been definitively described on ECs.3,4
FKN has both membrane-bound and soluble forms. Membrane-bound FKN can act as an adhesion molecule to mediate firm adhesion by binding with its receptor on leukocytes.1,3,4 This membrane-bound form can be cleaved by metalloproteinases5,6 to create circulating soluble (s)-FKN, a potential chemoattractant.5 Plasma s-FKN is increased in patients with certain inflammatory conditions, including coronary artery disease, allergic asthma and rhinitis, and brain inflammation.7–10 Several studies have reported that FKN and CX3CR1 play an important role in vascular inflammation and injury4,11–14 and contribute to atherosclerosis.11,13,14
Although expressed by ECs, it is unknown whether FKN can participate in the activation of ECs or their conversion to a proinflammatory phenotype. FKN-induced endothelial activation could be important not only in atherosclerosis but also in other forms of vascular inflammation, including post–ischemic vascular injury or the acute response to infection. However, for FKN to activate ECs, its receptor, CX3CR1, would also need to be expressed on these cells. This study was therefore performed to define the role of FKN and its receptor in activation of vascular endothelium. We found that (1) ECs express the CX3CR1 receptor; (2) FKN is upregulated and s-FKN is released from ECs by hypoxia/reoxygenation; (3) s-FKN can mediate a proinflammatory phenotype in ECs, with increased intercellular adhesion molecule (ICAM)-1 expression and polymorphonuclear leukocyte (PMN) adhesion; and (4) s-FKN–mediated ICAM-1 upregulation in ECs depends on a specific interaction between FKN and its receptor and downstream activation of the Jak–Stat5 pathway.
Materials and Methods
Human coronary artery ECs (HCAECs), human umbilical vein ECs (HUVECs), and culture media were purchased from Cambrex Bio Science Inc (Walkersville, Md). Anti-human CX3CR1 antibody was purchased from Abcam Inc (Cambridge, Mass). s-FKN protein (recombinant human FKN, comprising the chemokine domain of FKN, amino acids 1 to 76) and rabbit anti-human FKN antibody were bought from Peprotech Inc (Rocky Hill, NJ). Antibodies for Jak2, Stat5, and phosphorylated Jak2 and Stat5 were purchased from Cell Signaling Technology Inc (Boston, Mass). Anti-human Stat5α and Stat5β and ICΑΜ-1 antibodies were from Santa Cruz Biotechnology Inc (Santa Cruz, Calif). Monoclonal anti-human ICAM-1 antibody (used for adhesion blockade, BBA32), goat anti-mouse ICAM-1 antibody, and recombinant mouse sFKN (chemokine domain) were obtained from R&D Systems Inc (Minneapolis, Minn). Fluorescent-labeled rabbit IgG, calcein AM, pertussis toxin (PTX), and PMA (phorbol-12-myristate-13-acetate) were bought from Sigma-Aldrich Inc. Protease inhibitor cocktail (complete, EDTA-free) was from Roche Applied Science (Mannheim, Germany). CX3CR1 RT-PCR was performed using a pair of specific primers (5′-TGGCTGACTGGCAGATCCAGAG-3′ and 5′-TCAGAGAAGGAGCAATGCATCAC-3′) based on published data.2,15 This pair of primers amplifies a 1.2-kb full-length V28 cDNA. Stat5 small interfering (si)RNA16 and CX3CR1 siRNAs (sense, 5′GGAGCAGGCAUGGAAGUGUtt; antisense, 5′ACACUUCCGCCUGCUCCtt) were made by Ambion Inc (Foster City, Calif).
Cell culture, transfections, hypoxia/reoxygenation, ICAM-1 real-time RT-PCR, Western blotting, immunoprecipitation, and electrophoretic mobility-shift assays were performed as described previously.17
EC FKN Cleavage Assay
ECs were cultured and grown to 100% confluence in 6-well plates with EC growth medium (EGM2). Cells were washed 2 times with PBS buffer and replaced with control or ischemic buffer.17 After exposure to hypoxia or normoxia for 2 hours, the cells were replaced with EC basal medium (EBM) without growth factors or serum under normoxic conditions at 37°C. A complete protease inhibitor cocktail, lacking only metalloproteinase inhibitor (EDTA free), was added to the medium according to the instructions of the manufacturer. The media were harvested at various time points, and 5 mmol/L EDTA was added to inhibit metalloproteinase activity. The media were concentrated 20-fold using a Centricon column (10-kDa cutoff units; Millipore, Billerica, Mass) and loaded completely onto a NuPAGE gel (Invitrogen). The presence of released FKN in the conditioned media was demonstrated by Western blotting.
HUVECs were seeded on a glass slide coated with 0.1 mg/mL poly-l-lysine in 6-well plates. For immunostaining, the cells were fixed with methanol, blocked with 10% normal goat serum, and incubated with 1:100 anti-human CX3CR1 antibody. After washing, the cells were incubated with 1:500 secondary antibody (fluorescent-labeled rabbit IgG). Finally, the cells were washed, mounted, and imaged with a Carl Zeiss 510 confocal microscope. Giemsa–Wright staining was used for imaging of isolated PMNs.
Leukocyte Isolation and EC–Leukocyte Adhesion Assay
Human PMNs and monocytes were isolated by Dextron separation and Ficoll-hypaque density gradient centrifugation, labeled with the fluorescent dye calcein AM, and activated by PMA (see the online data supplement at http://circres.ahajournals.org).
HUVECs were seeded into 6 or12 well plates, and transfected with CX3CR1 siRNA or control siRNA for 48 hours. After 3 hours of s-FKN stimulation, 1×106 PMNs per milliliter (final concentration) were loaded onto the HUVECs for one hour. The supernatant was discarded, and adherent cells underwent fluorescence microscopy, with quantification of fluorescein isothiocyanate units. The number of adherent PMNs was also determined by measurement of myeloperoxidase activity.18 To determine whether PMN adhesion in this model was ICAM-1 dependent, after stimulation with 10 nmol/L s-FKN for 3 hours, anti–ICAM-1 antibody (final concentration, 10 μg/mL) was added for 1 hour, the supernatants were discarded, and the ECs were loaded with medium containing PMNs. The remainder of the adhesion assay was completed as described above.
Mouse Heart Langendorff Preparation
C57BL6 mice were purchased from Harland Bioproducts for Science, Inc. (Indianapolis, Ind). The experimental animals were housed in the animal care facility of the Johns Hopkins Medical Institutions, and the protocols were approved by Johns Hopkins Animal Care and Use Committee. Mouse littermates (C57BL6, male, 12 weeks old) were anesthetized with pentobarbital, and hearts were excised and perfused at constant pressure with Krebs–Henseleit buffer. Recombinant mouse s-FKN (50 ng/mL) was administered into the perfusion line for 3 hours, whereas control hearts were perfused with only buffer. At the end of perfusion, hearts were quick frozen in dry ice or fixed with zinc fixative solution (BD Biosciences Pharmingen, San Jose, Calif) for Western blot and immunohistology, respectively (see the online data supplement for details).
All quantitative assays were performed in triplicate, and the results were expressed as means±SD. The Student’s t test was used to determine the significance of differences between groups.
Hypoxia/Reoxygenation Increases Expression and Release of FKN From ECs
Cultured HCAECs and HUVECs were used to determine whether hypoxia/reoxygenation increases FKN expression. ECs were exposed to hypoxia for 2 hours and reoxygenation for up to 120 minutes. Western blots of cell lysates showed an increase in FKN protein expression, beginning within 10 minutes of reoxygenation and continuing for at least 120 minutes (Figure 1a). In addition, FKN protein was shed into the medium (representing s-FKN, based on its lower molecular weight), and the levels increased within 15 minutes of reoxygenation, peaked at 30 minutes, and continued for at least 60 minutes (Figure 1a, bottom).
Exposure to s-FKN Increases ICAM-1 Expression in ECs
Because s-FKN was shed from ECs and increased following hypoxia/reoxygenation, we examined whether s-FKN could stimulate a proinflammatory response in ECs. HUVECs were exposed to increasing concentrations of s-FKN, and the effect on ICAM-1 expression was determined for up to 180 minutes. We found that ICAM-1 mRNA increased in an s-FKN dose-dependent and time-dependent fashion (Figure 1b).
s-FKN Increases ICAM-1 Expression in Ex Vivo Mouse Hearts
In mouse hearts perfused with recombinant mouse s-FKN (soluble chemokine domain), Western blotting demonstrated markedly increased ICAM-1 protein expression compared with control hearts (Figure 2). Furthermore, immunohistology demonstrated that s-FKN–induced ICAM-1 expression was localized in capillary endothelium (Figure 2).
Vascular ECs Express the FKN Receptor CX3CR1
Because s-FKN increased ICAM-1 expression in ECs, we considered that a FKN receptor might exist in these cells. By performing Western blots in lysates from HCAECs and HUVECs and human donor leukocytes, we found that both types of ECs expressed CX3CR1 protein, in addition to the expected FKN protein. In contrast, monocytes and PMNs only expressed the receptor, as reported previously,3,4 but not FKN (Figure 3a). Both types of ECs expressed both 50- and 40-kDa protein bands for CX3CR1, whereas macrophages and PMNs expressed only a 40-kDa protein. The reason for the different molecular-weight bands for the receptor in ECs and leukocytes is unknown but could be related to different protein isoforms expressed19 and/or protein glycosylation.
We amplified the products of reverse transcript cDNA from ECs and leukocytes using a pair of specific primers. An ≈1.2-kb full-length V28 cDNA band spanning exon 2 to 4 was detected in agarose gel in both HUVECs and leukocytes, confirming that ECs expressed CX3CR1 mRNA (Figure 3b).
Using confocal microscopic imaging of immunostained HUVECs, we demonstrated that ECs expressed CX3CR1 and that this receptor was located in the cell membrane and to a lesser extent in the cytoplasm (Figure 3c).
FKN Activates ICAM-1 Through CX3CR1 in ECs and Promotes Neutrophil Adhesion
We next determined whether FKN could activate ICAM-1 in ECs through its receptor CX3CR1 and whether the resulting ICAM-1 upregulation was sufficient to promote increased PMN adhesion. We transfected CX3CR1-interfering RNA or control siRNA into cultured HUVECs and exposed these cells to s-FKN. HUVECs with CX3CR1 knockdown expressed less ICAM-1 mRNA and protein compared with cells transfected with control siRNA (Figure 4a and 4b), indicating that FKN does activate ICAM-1 in ECs through CX3CR1. Adhesion of purified human PMNs to HUVECs, measured by fluorescence intensity and myeloperoxidase activity, was significantly higher in control cells than in cells with CX3CR1 knockdown during s-FKN stimulation (Figure 5a). The increase in PMN adhesion induced by s-FKN was ICAM-1 dependent because it was significantly (P<0.001) inhibited by an ICAM-1–neutralizing antibody (Figure 5b).
FKN Acts Through Its Receptor to Activate the Jak–Stat5 Pathway in ECs
To study the mechanism of FKN-mediated upregulation of ICAM-1, we transfected CX3CR1 siRNA or loaded the G protein inhibitor PTX into cultured HUVECs (CX3CR1 is known to be a GPCR).2 Addition of s-FKN increased phosphorylated Jak2 and Stat5 in control cells but not in cells transfected with CX3CR1 siRNA or loaded with PTX (Figure 6). Jak2 phosphorylation was detected within 7 minutes of s-FKN stimulation and disappeared after 15 minutes. Shortly following Jak2 phosphorylation, an increase in Stat5 phosphorylation occurred and continued for at least 30 minutes. Neither Stat1 nor Stat3 demonstrated an increase in phosphorylation during s-FKN stimulation (data not shown).
Anti-Stat5α or -β antibodies were used to precipitate Stat5 proteins, followed by Western blotting with anti–phosphorylated Stat5 antibody. The results demonstrated that phosphorylated Stat5 following FKN stimulation was the α isoform (Figure 7a). The Western blotting membranes were then stripped and reprobed with Stat5 antibody. The blots showed that both α and β isoforms of Stat5 were present in HUVECs, but only the α isoform was phosphorylated.
The ICAM-1 gene is known to contain a Stat-binding site (GAS element) in its promoter region. Using electrophoretic mobility-shift assay and competition assays, we found that sFKN increased Stat protein binding to a labeled GAS probe. Addition of anti-Stat5 or -Stat5α antibodies resulted in a supershift band (Figure 7b), confirming that the Stat protein responsive to sFKN is Stat5α.
When Stat5-interfering RNA was transfected into HUVECs to knock down Stat5, s-FKN stimulation no longer resulted in an increase in ICAM-1 protein expression. In contrast, HUVECs transfected with control siRNA continued to show an increase in ICAM-1 protein (Figure 8). The results demonstrate that s-FKN upregulates ICAM-1 through Stat5.
FKN is expressed at low levels in normal ECs, but its expression markedly increases in ECs activated by proinflammatory agonists, such as lipopolysaccharide, tumor necrosis factor-α, interleukin-1, and interferon-γ.1,20,21 Tumor necrosis factor-α and interleukin-1 induce expression of FKN through nuclear factor κB signaling, whereas interferon-γ does so through the Jak–Stat1 pathway.21 We show for the first time that hypoxia/reoxygenation of ECs induces rapid FKN expression, with increased FKN protein present within 10 minutes of reoxygenation in both HCAECs and HUVECs (Figure 1a). This increase in FKN expression following hypoxia/reoxygenation is most likely related to the generation of reactive oxygen species, with downstream activation of nuclear factor κB.17,18 Activated nuclear factor κB is well known to regulate many other proinflammatory genes in ECs.17,18
Hypoxia/reoxygenation also resulted in increased shedding of FKN from ECs, with formation of s-FKN, a lower-molecular-weight, soluble form of FKN that circulates in the blood. It has been reported that 2 members of the disintegrin and metalloproteinase (ADAM) family, tumor necrosis factor-α–converting enzyme (TACE or ADAM17) and ADAM10, can cause proteolytic cleavage of membrane bound FKN to s-FKN.5,6 Both of these enzymes are highly expressed in vascular ECs,22,23 and their activities can be increased by reactive oxygen species.24
Although the FKN receptor CX3CR1 was thought to be mainly expressed on certain leukocyte populations,3 CX3CR1 mRNA is strongly expressed in brain, spleen, skeletal muscle, and peripheral blood cells, although less so in the heart.2 One study reported that immunostaining for CX3CR1 could be found throughout the wall of human coronary arteries, including endothelium, intima, and adventitia,25 and the receptor was also shown to be expressed in cultured human coronary artery smooth muscle cells.26 In a very recent report, results from flow cytometry of HUVECs and immunohistology of rat aorta were consistent with expression of CX3CR1 protein in ECs.27 In addition, there are 3 known CX3CR1 isoforms,15 and the gene is driven by 3 different promoters.19 Together, these studies suggest that CX3CR1 may be widely expressed in many different tissues and cell populations.
In our study, we found definitive evidence that ECs, including both arterial and venous ECs, express CX3CR1 mRNA and protein and that the receptor is located primarily on the cell membrane (Figure 3). In contrast to the finding of a single 40-kDa protein band in monocytes and leukocytes, ECs had both 50- and 40-kDa bands, suggesting that ECs might express 2 different CX3CR1 isoforms. Future studies will be needed to determine whether these 2 isoforms perform distinct functions.
FKN and CX3CR1 have been reported to play an important role in atherosclerosis, most likely by promoting inflammation in the vascular wall. High levels of FKN mRNA have been observed in human arteries with advanced atherosclerosis, and FKN is upregulated in atherosclerotic lesions of apolipoprotein (apo)E-deficient (apoE−/−) mice.13 Crossing CX3CR1 knockout mice into an apoE−/− background results in decreased atherosclerotic lesion formation with reduced macrophage accumulation.14 Genetic polymorphisms at amino acids 249 and 280 of human CX3CR1 have been reported to be a risk factor for coronary artery disease. CX3CR1 V249I/T280M heterozygosity is associated with a markedly reduced risk of acute coronary events.11
ICAM-1 is a cell surface glycoprotein that is highly expressed in vascular ECs and plays an important role in mediating neutrophil adherence and tissue injury during processes as diverse as atherosclerosis and ischemia/reperfusion. Many proinflammatory cytokines including tumor necrosis factor-α, interleukin-6, interleukin-1, and interferon-γ can induce ICAM-1 expression.17 We found that s-FKN can also act like these cytokines to upregulate ICAM-1 expression (Figures 1b, 2, and 4⇑⇑). Our results demonstrate that s-FKN can act not only as a leukocyte chemoattractant but also as a ligand to bind to its receptor on ECs to upregulate ICAM-1 and promote a proinflammatory endothelial phenotype (Figure 5). FKN could be shed in high concentration from activated ECs and stimulate ICAM-1 upregulation in the same or other ECs located nearby. In addition, it is possible that because CX3CR1 and FKN are both expressed on the EC cell membrane, membrane-bound FKN, which is on a stalk, could bend and interact with a closely placed CX3CR1 on the same cell. There is precedent for this in the form of protease activated receptors, which can interact with a membrane-bound ligand on the same cell.28
A recent study in rat aorta by Schafer et al reported that FKN did not increase ICAM-1 expression.27 As a possible explanation for this discrepancy, although our study and the study by Schafer et al27 both used human FKN, we used human ECs as the target cell, whereas Schafer et al used rat aortic tissue. Because human FKN shares only 80% homology with rat FKN (GenBank accession nos. 12654650 and 47718005), there may have been an imperfect match between FKN and its receptor in the study by Schafer et al. In addition, whereas we used a recombinant s-FKN containing a single chemokine/receptor binding domain (amino acids 1 to 76), Schafer et al used full-length FKN containing the free transmembrane and cytoplasm domains, and this might have affected exposure of the chemokine domain to its receptor. In addition to the experiments with cultured ECs, we found that s-FKN increased ICAM-1 protein expression in vascular endothelium in the intact perfused mouse heart (Figure 2). For these studies, we used recombinant mouse s-FKN to match the agonist with its target tissue.
Jak–Stat signaling is an important pathway for regulation of ICAM-1.17 Jak–Stat signaling can be activated by cytokines, growth factors, and reactive oxygen species via cytokine and growth factor receptors or other cell surface GPCRs.29 Because CX3CR1 has been identified as a GPCR,2 we thought that it might mediate activation of the Jak–Stat cascade. Our study demonstrated that Stat5, but not Stat1 or Stat3, is activated during s-FKN stimulation and that Stat5 activation can be inhibited by the G protein inhibitor PTX and by CX3CR1 siRNA (Figure 6). Using anti–phosphorylated Stat5 antibody and anti-Stat5α or -β antibodies, we further demonstrated that the activated isoform of Stat5 is Stat5α (Figure 7). Previous studies have reported that ischemia/reperfusion also activates Stat5α but not -β.30,31 Like other Stat proteins, Stat5 can upregulate target gene transcription by binding to GAS elements30–32 in the gene promoter. We demonstrated that following exposure to s-FKN, Stat5α binds to the GAS element of the ICAM-1 promoter (Figure 6) and that Stat5 knockdown decreases ICAM-1 expression (Figure 8). It is likely that Stat5 upregulates ICAM-1 through this mechanism.
We have not defined exactly how CX3CR1 activates the Jak–Stat pathway. GPCRs exist as a 7-membrane-spanning construct, and after chemokine binding, the intracellular loops interact with each other to form functional domains, resulting in selective coupling to a defined G protein.33,34 Rho GTPases have been shown to mediate Jak activation through Rac1 and the production of reactive oxygen species.29 Based on its inactivation by PTX, FKN is likely to act through this general mechanism, leading to activation of Jak2 and Stat5 (Figure 6). Alternatively, activation of GPCRs also brings about the local aggregation of associated Jaks, resulting in their self-activation. Recent studies have demonstrated direct association of GPCRs with tyrosine kinases and Jaks, resulting in transphosphorylation.33 Like most chemokine receptors, CX3CR1 presents a highly conserved DRY box33 in its second intercellular loop, and the tyrosine residue (Y) may relate to Jak2 and Stat protein phosphorylation. Although several docking sites for Jaks in GPCRs have been identified, including 1 proline-rich motif, YIPP, and 3 hydrophobic amino acid LXXXIW and SHSK motifs,33,35 these sequences do not exist in the CX3CR1 intracellular loops. It will be necessary in future studies to determine whether CX3CR1 aggregates directly with Jak2 and to characterize the docking sites involved.
We demonstrated that vascular ECs express both FKN and its receptor CX3CR1. We found for the first time that hypoxia/reoxygenation increases expression of FKN and results in shedding of s-FKN from the EC cell membrane. s-FKN, in turn, can activate the Jak–Stat5 pathway through CX3CR1 to upregulate ICAM-1 and promote neutrophil adhesion.
We gratefully acknowledge the assistance of William M. Baldwin III, MD, PhD, in the preparation and interpretation of the immunohistology images.
Sources of Funding
This study was supported by National Heart, Lung, and Blood Institute Program Project grant HL-65608 (to L.C.B.).
This manuscript was sent to Donald D. Heistad, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Original received May 21, 2007; resubmission received July 30, 2007; revised resubmission received August 28, 2007; accepted September 6, 2007.
Combadiere C, Salzwedel K, Smith ED, Tiffany HL, Berger EA, Murphy PM. Identification of CX3CR1. A chemotactic receptor for the human CX3C chemokine fractalkine and a fusion coreceptor for HIV-1. J Biol Chem. 1998; 273: 23799–23804.
Tsou CL, Haskell CA, Charo IF. Tumor necrosis factor-α-converting enzyme mediates the inducible cleavage of fractalkine. J Biol Chem. 2001; 276: 44622–44626.
Hundhausen C, Misztela D, Berkhout TA, Broadway N, Saftig P, Reiss K, Hartmann D, Fahrenholz F, Postina R, Matthews V, Kallen KJ, Rose-John S, Ludwig A. The disintegrin-like metalloproteinase ADAM10 is involved in constitutive cleavage of CX3CL1 (fractalkine) and regulates CX3CL1-mediated cell-cell adhesion. Blood. 2003; 102: 1186–1195.
Damås JK, Boullier A, Wæhre T, Smith C, Sandberg WJ, Green S, Aukrust P, Quehenberger O. Expression of fractalkine (CX3CL1) and its receptor, CX3CR1, is elevated in coronary artery disease and is reduced during stain therapy. Arterioscler Thromb Vasc Biol. 2005; 25: 2567–2572.
Chapman GA, Moores K, Harrison D, Campbell CA, Stewart BR, Strijbos PJ. Fractalkine cleavage from neuronal membranes represents an acute event in the inflammatory response to excitotoxic brain damage. J Neurosci. 2000; 20: RC87.
McDermott DH, Fong AM, Yang Q, Sechler JM, Cupples LA, Merrell MN, Wilson PW, D’Agostino RB, O’Donnell CJ, Patel DD, Murphy PM. Chemokine receptor mutant CX3CR1–M280 has impaired adhesive function and correlates with protection from cardiovascular disease in humans. J Clin Invest. 2003; 111: 1241–1250.
Combadiere C, Potteaux S, Gao J-L, Esposito B, Casanova S, Lee EJ, Debre P, Tedgui A, Murphy PM, Mallat Z. Decreased atherosclerotic lesion formation in CX3CR1/apolipoprotein E double knockout mice. Circulation. 2003; 107: 1009–1016.
Yang XP; Irani K, Mattagajasingh S, DiPaula A, Khanday F, Ozaki M, Fox-Talbot K, Baldwin WM 3rd, Becker LC. Signal transducer and activator of transcription 3α and specificity protein 1 interact to upregulate intercellular adhesion molecule-1 in ischemic–reperfused myocardium and vascular endothelium. Arterioscler Thromb Vasc Biol. 2005; 25: 1395–1400.
Fan H, Sun B, Gu Q, Lafond-Walker A, Cao S, Becker LC. Oxygen radicals trigger activation of NF-κB and AP-1 and upregulation of ICAM-1 in reperfused canine heart. Am J Physiol Heart Circ Physiol. 2002; 282: 1778–1786.
Garin A, Tarantino N, Faure S, Daoudi M, Lécureuil C, Bourdais A, Debré P, Deterre P Combadiere C. Two novel fully functional isoforms of CX3CR1 are potent HIV coreceptors. J Immunol. 2003; 171: 5305–5312.
Nian M, Lee P, Khaper N, Liu P. Inflammatory cytokines and postmyocardial infarction remodeling. Circ Res. 2004; 94: 1543–1553.
Boulday G, Coupel S, Coulon F, Soulillou JP, Charreau B. Antigraft antibody-mediated expression of metalloproteinases on endothelial cells. Differential expression of TIMP-1 and ADAM-10 depends on antibody specificity and isotype. Circ Res. 2001; 88: 430–437.
Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ, Cerretti DP. A metalloproteinase disintegrin that releases tumour-necrosis factor-α from cells. Nature. 1997; 385: 729–733.
Lucas AD, Bursill C, Guzik TJ, Sadowski J, Channon KM, Greaves DR. Smooth muscle cells in human atherosclerotic plaques express the fractalkine receptor CX3CR1 and undergo chemotaxis to the CX3C chemokine fractalkine (CX3CL1). Circulation. 2003; 108: 2498–2504.
Schafer A, Schulz C, Fraccarollo D, Tas P, Leutke M, Eigenthaler M, Seidl S, Heider P, Ertl G, Massberg S, Bauersachs J. The CX3C chemokine fractalkine induces vascular dysfunction by generation of superoxide anions. Arterioscler Thromb Vasc Biol. 2007; 27: 55–62.
Pelletier S, Duhamel F, Coulombe P, Popoff MR, Meloche S. Rho family GTPase are required for activation of Jak/Stat signaling by G protein-couple receptors. Mol Cell Biol. 2003; 23: 1316–1333.
Mascareno E, El-Shafei M, Maulik N, Sato M, Guo Y, Das DK, Siddiqui MA. JAK/STAT signaling is associated with cardiac dysfunction during ischemia and reperfusion. Circulation. 2001; 104: 325–329.
El-Adawi H, Deng L, Tramontano A, Smith S, Mascareno E, Ganguly K, Castillo, El-Sherif N. The functional role of Jak-Stat pathway in post-infarction remodeling. Cardiovasc Res. 2003; 57: 129–138.
Soldaini E, John S, Moro S, Bollenbacher J, Schindler U, Leonard WJ. DNA binding site selection of dimeric and tetrameric Stat5 proteins reveals a large repertoire of divergent tetrameric Stat5α binding sites. Mol Cell Biol. 2000; 20: 389–401.
Ahr B, Denizot M, Robert-Hebmann V, Brelot A, Biard-Piechaczyk M. Identification of the cytoplasmic domains of CXCR4 involved in Jak2 and STAT3 phosphorylation. J Biol Chem. 2005; 280: 6692–6700.