Little is known concerning the possible contribution of T helper 2 (Th2)-type cytokines to the recruitment of neutrophils into the lung tissue. In the present study, endothelial cells from equine pulmonary arteries were cultured in the presence of recombinant equine (re) IL-4 and reIL-5, and the cytokine mRNA expression of molecules implicated in the chemotaxis and migration of neutrophils was studied using real-time RT-PCR. The functional response of reIL-4-induced endothelial cell stimulation on neutrophil migration was also studied using a chemotaxis chamber. ReIL-4 either increased the expression of CXCL-8, E-selectin, vascular endothelial growth factor (VEGF), and inducible nitric oxide synthase (iNOS), or potentiated the coeffects of lipopolysaccharide (LPS) and tumor necrosis factor-α (TNF-α) on CXCL-8. Supernatants collected from cultured endothelial cells stimulated with reIL-4 significantly promoted neutrophil migration in a dose-dependent manner. Dexamethasone (DXM) decreased the expression of CXCL-8, VEGF, and iNOS induced by reIL-4, while 1400W dihydrochloride (1400W), a selective inhibitor of iNOS, decreased the expression of E-selectin, VEGF, and iNOS. DXM and 1400W attenuated the mRNA expression of E-selectin and iNOS induced by the costimulation of reIL-4, reTNF-α, and LPS. Neither equine nor human recombinant IL-5 influenced the mRNA expression of CXCL-8, E-selectin, or VEGF. These findings suggest that Th2-type cytokines may contribute to pulmonary neutrophilia during allergic inflammation by the increased expression of neutrophil chemokines and adhesion molecules by endothelial cells. DXM and the iNOS inhibitors may decrease pulmonary neutrophilia due, in part, to a direct inhibition of some of these factors.
- neutrophil chemokines
migration of neutrophils from the vascular compartment to inflamed tissue is a multistep process involving cytokines and chemokines released from numerous cells, including activated leukocytes and endothelial cells. The arrest of circulating neutrophils at a site of inflammation is a prerequisite early event for the cell migration. It requires the sequential exposure to different adhesion molecules, such as E-selectin, and the expression of intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1, which are involved in the rolling and firm adhesion of neutrophils to the endothelium. Neutrophils then migrate into the inflammatory sites through transendothelial diapedesis and chemotaxis. CXCL-8, a member of the C-X-C family of chemokines, is the most potent neutrophil chemotactic cytokine known to date. CXCL-8 not only upregulates the expression of β2-integrins, it induces degranulation and promotes the attachment and transendothelial migration of neutrophils (27) and is also angiogenic (33). The expression of CXCL-8 may be upregulated by nitric oxide (NO), a volatile free radical, through a cGMP-independent pathway (13). VEGF, a multifunctional angiogenic regulator that has the ability to generate tissue edema through increased vascular permeability (12, 55), also stimulates endothelial synthesis of CXCL-8 in humans and induces a transient upregulation of human umbilical vein endothelial cell (HUVEC) expression of ICAM-1 and VCAM-1 (38, 66) and, therefore, could contribute to endothelial cell promotion of tissue neutrophilia.
Allergen exposure induces trafficking of newly produced neutrophils into the airways in both experimental and natural allergic lung diseases. Furthermore, neutrophils may facilitate the migration of eosinophils into the airways, possibly aggravating the asthmatic response (1). However, the mechanisms responsible for the chemotaxis and activation of neutrophils in these processes remain poorly defined; a number of pathways have been shown to possibly contribute to neutrophilia in allergic inflammation (15, 54, 59). There is also evidence that T helper 2 (Th2)-type cytokines, which are considered central to allergic inflammation, may contribute to neutrophil activation and recruitment to lung tissue. IL-4 in humans and IL-5 in Brown Norway rats cause a neutrophilia and accelerate the maturation of myelocytes to become neutrophils (5, 49). IL-4 also stimulates neutrophil phagocytic response (5), enhances neutrophil-mediated bactericidal activity, induces lysozyme release, and enhances the respiratory burst (7). Conversely, IL-4 has been shown to have anti-inflammatory effects on neutrophils by the inhibition of proinflammatory cytokines, including CXCL-8 (64) and IL-1β (43), and by increasing IL-1α production (44). T lymphocytes may also indirectly contribute to the chemotaxis of neutrophils into lung tissue by the release of IL-17 (17, 48), a cytokine that selectively activates and recruits neutrophils through the release of inflammatory mediators, including CXCL-8 and MIP-2 (26).
Whereas the endothelium presents a critical barrier between blood and surrounding tissues and regulates leukocytes trafficking, there is limited and conflicting information on the role of Th2-type cytokines on the release of neutrophil chemotactic factors by vascular endothelial cells. IL-4 has been reported to either suppress the expression of E-selectin and CXCL-8 by endothelial cells (11, 57) or to increase CXCL-8 production after LPS activation (16). Little is known on the effects of IL-5 on neutrophil recruitment. However, the finding that it induces the expression of P-selectin by endothelial cells in nasal polyps (40) suggests that IL-5 may also contribute to neutrophilic inflammation.
Endothelial cells exhibit highly specialized functions in different vascular sites. Most of the above studies have used immortalized HUVECs, for which phenotypes and physiological responses may differ from those of mature individuals. Thus it may be necessary to use primary endothelial cells derived from the lung of adult subjects to investigate the role of the endothelial cell adhesion molecule in the pathogenesis of inflammatory conditions within this organ. To evaluate the possible contribution of Th2-type cytokines to the lung neutrophilia, we stimulated pulmonary artery endothelial cells harvested from healthy adult equine subjects. This species was chosen because horses have a pulmonary anatomy and blood vascularization that closely resemble that of humans (47, 56) and they are affected with a condition of the lungs that shows many similarities with chronic asthma (18, 36). We studied the expression of selected chemokines and adhesion molecules after stimulation with IL-4 and IL-5 alone, and with LPS and TNF-α, to determine synergistic or inhibitory effects. We also examined the reversibility of this response by glucocorticoids or through the inhibition of inducible NO synthase (iNOS) production.
MATERIALS AND METHODS
Endothelial Cell Culture
Equine artery endothelial cells were isolated and cultured as described by MacEachern et al. (42). Briefly, pulmonary arteries were removed from adult horses at a slaughterhouse. Each end of the vessel was occluded with bowel clamps, and 75% ethanol was applied to the external wall of the artery. A sterile catheter (Angiocath; Becton Dickinson Infusion Therapy Systems Sandy, UT) inserted through the wall of the artery was used to wash the lumen of the artery three times with sterile phosphate-buffered saline (PBS) pH 7.4 solution, supplemented with penicillin/streptomycin (GIBCO), 200 U/ml and 200 μg/ml, respectively. Twenty milliliters of sterile 0.1% type II collagenase (Invitrogen, CA) was then infused into the vessel, incubated for 30 min at room temperature, and flushed with 20 ml of sterile PBS. The cell suspension was centrifuged at 390 g for 10 min, and endothelial cells were resuspended in Dulbecco's modified Eagle's medium (DMEM, Invitrogen) containing 10% fetal calf serum and penicillin/streptomycin (200 U/ml and 200 μg/ml, respectively). The cells were cultured in sterile T-75 culture flasks placed in a humidified incubator (37°C in 95% air, 5% CO2) and allowed to adhere overnight. The medium was replaced every 48 h thereafter until confluence was attained. Endothelial cells were then removed using trypsin-EDTA (Invitrogen), washed once with PBS, transferred to sterile six-well plates, and allowed to form a confluent monolayer. In all experiments, cells were used at passages 3–5.
Factor VIII Immunofluorescence Staining
The endothelial cells were grown on 22-mm2 glass coverslips in sterile six-well plates. The coverslips were rinsed with PBS three times and then fixed in 4% paraformaldehyde for 30 min at room temperature. Approximately 200 μl of a solution containing the primary antibody (rabbit anti-human von Willebrand factor, Sigma) diluted in PBS (1:200) or PBS alone (control) was placed on each coverslip and incubated for ∼24 h at 4°C. The coverslips were rinsed with PBS three times, the secondary antibody was added (FITC-conjugated goat anti-rabbit Ig, Sigma) at a 1:100 dilution on both coverslips, and they were incubated for 24 h at 4°C. The coverslips were then rinsed with PBS and allowed to dry. Two drops of mounting agent were placed on the slides. The cells were viewed and photographed with a Nikon Eclipse E800 ultraviolet microscope.
Groups and Cytokine Induction
Added directly to the medium were the following: 1, 10, and 100 ng/ml recombinant equine (re) IL-4 (R&D Systems); 10 ng/ml reIL-4 plus 10−5 M dexamethasone (DXM; Vetoquinol, CA); 10 ng/ml reIL-4 plus 10 μg/ml 1400W dihydrochloride (1400W; Sigma, Germany); 1, 10, and 100 lab units of reIL-5 (kindly provided by Dr. Falko Steinbach, Institute for Zoo and Wildlife Research, Berlin, Germany); 10 and 100 ng/ml recombinant human (rh) IL-5 (Becton Dickinson); 2 ng/ml reTNF-α (R&D Systems) plus 2 ng/ml LPS (Sigma, CA); 2 ng/ml reTNF-α plus 2 ng/ml LPS (TL) plus 10−5 M DXM; 10 μg/ml TL plus 1400W; 10 ng/ml TL plus reIL-4 (LPIL-4); 10−5 M LPIL-4 plus DXM; 10 μg/ml LPIL-4 plus 1400W; or PBS (negative control). Incubation time and concentration of drugs were chosen based on preliminary experiments to stimulate the mRNA expression of genes of interest without saturating the response or altering endothelial cell viability (data not shown).
Endothelial cells were homogenized in TRIzol reagent (Invitrogen Life Technologies), and extraction of RNA was performed according to the instructions of the manufacturer. At the end of the procedure, the RNA pellet was briefly dried in air and dissolved in RNase-free water. The concentration and the purity of RNA were determined by spectrophotometry.
Five hundred nanograms of RNA was reverse-transcripted using oligo(dT) (Invitrogen) and reverse transcriptase (AMV, Roche) according to the manufacturer's instructions. Real-time PCR reactions were performed in a 20-μl mixture containing 1 μl of the cDNA preparation, 10 μl of SYBR Green (Qiagen), and 5 μM each primer in a Rotor-Gene system (Corbett Research). The sequences of the primers used for the amplifications are shown in Table 1. All sets of primers span at least two intron-exon boundaries to allow for the discrimination of amplified genomic DNA. The absence of nonspecific products was confirmed by the analysis of the melting point curves and by electrophoresis in 1.5% agarose gels. Ubiquitin (UBQ), GAPDH, and β2-microglobulin (β2M) served as an internal standard of mRNA expression.
mRNA quantification was achieved by constructing standard curves using cDNA of known concentrations. The standard curves consisted of serial dilutions of the PCR amplicon corresponding to the gene of interest; the amplicons were previously gel-extracted (QIAquick PCR Purification Kit, Qiagen) and quantified using a spectrophotometer (expressed as ng/μl). Four standard samples were included in each run. All concentrations of target gene cDNA were calculated relatively to their respective standard curves.
Neutrophil Migration Assay
Supernatant (30.5 μl) derived from endothelial cells stimulated with reIL-4 at 1, 10, and 100 ng/ml were added in triplicate to the lower wells of a 48-well chemotaxis chamber (Neuro Probe, CA). Similarly, wells containing PBS or 10−7 M leukotriene B4 (LTB4; Cayman Chemical, CA) served as negative and positive controls, respectively. The wells were then overlaid with an 8-μm pore size cellulose nitrate filter. Equine neutrophils (55 μl, 1.5 × 105 cells/ml), isolated from equine peripheral blood by density gradient centrifugation using lymphocyte poly (Cedarlane, CA) and suspended in RPMI 1640 (GIBCO), were placed in the upper wells. After 1 h of incubation (37°C, 5% CO2), filters were removed, fixed with mercuric chloride, and stained with hematoxylin and chromotrope 2R. Chemotactic activity was determined by counting the number of cells at the lower side of the filter using a light microscope and an eyepiece graticule to facilitate the counting. Five distinct fields were counted at high magnification (×400), and the mean values of the triplicate samples were reported.
Data were expressed as means ± SE. Data were statistically analyzed using nonparametric Friedman and Wilcoxon tests. Differences between groups were considered significant when P values were ≤0.05.
Factor VIII Expression on Endothelial Cells
Von Willebrand factor (factor VIII) is a glycoprotein synthesized by endothelial cells, which is not found in cultured smooth muscle or fibroblasts and is therefore considered a marker of endothelial cells. Figure 1 shows that endothelial cells stained positive for von Willebrand factor. The cells had a cobblestone appearance, were homogeneous, closely packed, and polygonal in shape with green fluorescence staining in their cytoplasm. Only weak fluorescence was detectable in the control cells.
Cytokine mRNA Expression
Effects of reIL-4.
UBQ was found to be a suitable housekeeping gene when studying the reIL-4-, reIL-5-, and rhIL-5-induced gene expression by equine pulmonary artery endothelial cells (Fig. 2, A–D). ReIL-4 increased in a dose-dependent manner the expression of CXCL-8 mRNA (Fig. 3A). Similarly, supernatants collected from cultured endothelial cells containing reIL-4 at 1, 10, and 100 ng/ml, respectively, significantly promoted neutrophil migration (Fig. 3B), whereas reIL-4 alone did not stimulate neutrophil chemotaxis (data not shown). ReIL-4 also enhanced the expression of E-selectin (Fig. 4B), VEGF (Fig. 4C), and iNOS (Fig. 4D), whereas CXCL-8, VEGF, and iNOS expression was significantly inhibited by DXM and almost reached significance for E-selectin (P = 0.07). The iNOS inhibitor 1400W significantly attenuated the mRNA expression of E-selectin, VEGF, and iNOS (Fig. 4, B–D).
Effects of TNF-α and LPS costimulation.
TNF-α and LPS significantly influenced the expression of UBQ and β2M (Fig. 5), and significance was almost reached for GADPH (P = 0.07). Thus values for the expression of genes of interest are expressed as concentration. TNF-α and LPS costimulation increased the expression of CXCL-8, E-selectin, and iNOS but did not influence VEGF; DXM inhibited the expression of CXCL-8 and iNOS induced by TNF-α and LPS stimulation, whereas 1400W inhibited only iNOS mRNA expression (Fig. 6).
Effects of reIL-4, TNF-α, and LPS costimulation.
ReIL-4 enhanced further the TNF-α and LPS expression of CXCL-8, while it almost reached significance for E-selectin (P = 0.07) but not for iNOS. DXM and 1400W attenuated the expression of E-selectin and iNOS, while the CXCL-8 mRNA was inhibited by DXM only.
Effects of reIL-5.
Neither reIL-5 nor rhIL-5 significantly increased the expression of cytokines nor growth factors by equine pulmonary endothelial cells (Fig. 7).
Neutrophils are recruited in experimentally induced allergic lungs (34, 54) and in natural pulmonary conditions such as human asthma (41, 54) and equine heaves (20). We chose to study heaves as this condition of horses shares many similarities with human asthma, including airway smooth muscle remodeling (24), hyperresponsiveness, and reversible airway obstruction. Furthermore, the airway neutrophilic inflammation present in heaves is associated with a predominant Th2-type cytokine response (36). Currently, little is know concerning the pathways leading to airway neutrophilia following the inhalation of allergens by susceptible individuals. We report here on the possible contribution of the Th2-type cytokines, IL-4 and IL-5, to airway neutrophilia using primary pulmonary artery endothelial cell culture. In the present study, reIL-4, but not reIL-5 or rhIL-5, induced a selective increase in the mRNA expression of CXCL-8, E-selectin, and VEGF, key elements for the trafficking of neutrophils to the inflamed tissue. We were unable to confirm these results at the protein level due to the lack of specific antibodies against these targets, a common limitation when studying this equine model. Nevertheless, the mRNA signal for those genes usually leads to protein synthesis (9, 28, 31), and, using Boyden chambers, reIL-4 stimulation of endothelial cells was shown to induce the release of chemotactic factors for neutrophils. ReIL-4 also enhanced the CXCL-8 expression by endothelial cells costimulated with TNF-α, a proinflammatory cytokine present in the allergic lungs (62), and LPS, a bacterial wall component present in house dust. We also determined that DXM has selective inhibitory effects on the IL-4-, LPS-, and TNF-α-stimulated endothelial expression of chemokines and adhesions molecules. Lastly, the upregulation of E-selectin and VEGF by endothelial cells was at least partially mediated by iNOS.
In the present report, IL-4 increased the CXCL-8 mRNA expression by equine pulmonary artery endothelial cells in a dose-dependent manner; moreover, the expression of CXCL-8 was further increased when IL-4 was coincubated with TNF-α and LPS, two proinflammatory mediators commonly present in the allergic lungs. These results are in contrast with some earlier reports indicating that IL-4 did not influence or even decrease the expression of CXCL-8 and E-selectin by HUVEC or human lung microvascular endothelial cells (3, 4, 11, 58). Interestingly, the response to LPS and TNF-α here was similar to those observed using HUVEC cells (14). The cell type studied may contribute to these differences, as we used primary cell cultures harvested from healthy adult animals, whereas previous studies have used primarily immortalized umbilical cells. The origin of the cells (lungs vs. umbilicus) and age (neonates vs. adults) of the cell donors may explain some of these findings. Also, even within the pulmonary vasculature, endothelial cells from various tissues possessed significant morphological distinctions, both in circulation and culture (32, 50). Thus the response of pulmonary artery endothelial cells, such as in the present study, may differ from that of pulmonary veins or capillary endothelial cells. However, this remains to be ascertained, as the overall gene expression pattern of lung macro- and microvascular endothelial cells after stimulation with IFN-γ, TNF-α, and IL-4 was reported not to be significantly different (52). Species differences may also contribute to these apparent discrepancies, considering that neutrophils predominate in the airways of horses following both experimental (8) and natural (20) allergen challenge. However, airway neutrophilia is also present in human asthma (23), especially in patient with severe asthma (51), nocturnal asthma (45), and steroid-dependent asthma (63). Conversely, neither reIL-5 nor rhIL-5 altered the mRNA expression of CXCL-8, E-selectin, and VEGF production by equine endothelial cells. These findings are in agreement and extend previous reports suggesting that IL-5 may not contribute to the neutrophil recruitment and adhesion to pulmonary endothelium (60, 61).
The role of NO in leukocyte recruitment and migration remains controversial (2, 25). Inducible NOS has long been considered responsible for the excessive levels of NO found in chronic pathogenic inflammatory lesions. For these reasons, we postulated that selective blockade of the iNOS pathway, without impinging on endothelial NOS- or neuronal NOS-dependent events, would provide an ideal strategy to disrupt only the presumptive harmful NO being produced in the vessels. We believed that the iNOS inhibitor 1400W would be a more effective anti-inflammatory agent than NG-monomethyl-l-arginine (l-NMMA), which acts as a competitive inhibitor of both constitutive and inducible isoforms of NOS. In the present study, 1400W had no effect on the expression of CXCL-8 by endothelial cells stimulated with IL-4 alone or in combination with TNF-α and LPS. These results suggest that the IL-4-induced CXCL-8 expression by endothelial cells does not depend on an iNOS-dependent pathway. Interestingly, and in agreement with a previous report (21), we also found that 1400W decreased the E-selectin mRNA expression by endothelial cells stimulated with IL-4, indicating that NOS may nevertheless contribute to neutrophil migration.
VEGF plays a role in vascular growth and remodeling (39, 46). VEGF has also been shown to be capable of promoting airway neutrophilia by stimulating endothelial synthesis of CXCL-8, upregulating endothelial expression of cell-adhesion molecules, and increasing the adhesion and transendothelial migration of neutrophil (38). VEGF modulated the airway hyperresponsiveness (37) and extracellular matrix (10) and thus may contribute to the airway remodeling. Our results show that IL-4, but not IL-5, could promote neutrophil migration and airway remodeling indirectly through the modulation of the secretion of CXCL-8, E-selectin, and VEGF by endothelial cells. Conversely, neither TNF-α nor LPS influenced the expression of VEGF. These results contrast with those originating from other cell populations. For instance, VEGF gene expression was shown to be regulated by LPS in macrophages (19), and TNF-α is a potent activator of VEGF synthesis and secretion by neutrophils (22, 53). However, the different VEGF isoforms have distinctive biological effects and pattern of expression in specific cell population (29, 30). Therefore, the lack of a regulation of the equine VEGF164 in the present study does not totally rule out an effect of LPS and TNF-α on VEGF in equine endothelial cells.
Glucocorticoids are the drugs most commonly used for the treatment of asthma. DXM, a potent glucocorticoid, had been shown to inhibit the CXCL-8 production in response to TNF-α by human coronary artery endothelial cells, human pulmonary artery cells, and HUVEC (35). Furthermore, DXM inhibited the upregulation of E-selectin expression in cultured human nasal microvascular endothelial cells stimulated by IL-1β and TNF-α (65). Our findings, that DXM inhibited the effects of IL-4, LPS, and TNF-α on CXCL-8 and also E-selectin, suggest that glucocorticoids may decrease pulmonary neutrophilia by a direct effect on the pulmonary endothelium.
This study was supported financially by the National Research Council Canada.
We gratefully acknowledge Isabelle Daneau from the Centre de recherche en reproduction animale, Université de Montréal, for skillful assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society