Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2008 Mar;151(3):519–527. doi: 10.1111/j.1365-2249.2007.03584.x

Adhesion of Epstein–Barr virus-positive natural killer cell lines to cultured endothelial cells stimulated with inflammatory cytokines

H Kanno *, D Watabe *, N Shimizu †, T Sawai *
PMCID: PMC2276960  PMID: 18190605

Abstract

Chronic active Epstein–Barr virus (EBV) infection (CAEBV) is characterized by chronic recurrent infectious mononucleosis-like symptoms. Approximately one-fourth of CAEBV patients develop vascular lesions with infiltration of EBV-positive lymphoid cells. Furthermore, EBV-positive natural killer (NK)/T cell lymphomas often exhibit angiocentric or angiodestructive lesions. These suggest an affinity of EBV-positive NK/T cells to vascular components. In this study, we evaluated the expression of adhesion molecules and cytokines in EBV-positive NK lymphoma cell lines, SNK1 and SNK6, and examined the role of cytokines in the interaction between NK cell lines and endothelial cells. SNKs expressed intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) at much higher levels than those in EBV-negative T cell lines. SNKs produced the larger amount of tumour necrosis factor (TNF)-α, which caused increased expression of ICAM-1 and VCAM-1 in cultured human endothelial cells, than that from EBV-negative T cell lines. Furthermore, SNKs exhibited increased adhesion to cultured endothelial cells stimulated with TNF-α or interleukin (IL)-1β, and the pretreatment of cytokine-stimulated endothelial cells with anti-VCAM-1-antibodies reduced cell adhesion. These indicate that the up-regulated expression of VCAM-1 on cytokine-stimulated endothelial cells would be important for the adhesion of EBV-positive NK cells and might initiate the vascular lesions.

Keywords: adhesion molecule, endothelial cell, Epstein–Barr virus, NK/T cell lymphoma, vascular cell adhesion molecule-1

Introduction

Epstein–Barr virus (EBV) is a ubiquitous virus in humans. Most individuals are infected with EBV by early adulthood. Primary EBV infection is usually asymptomatic, but sometimes results in infectious mononucleosis, which is basically self-limited due to the development of EBV-specific immunity. However, some individuals in Asia, mainly in Japan, are reported to develop chronic infection with EBV. Chronic active EBV infection (CAEBV) is characterized by chronic recurrent infectious mononucleosis-like symptoms over a long period of time and by an unusual pattern of anti-EBV-antibodies [1–3]. CAEBV is a disease with a high mortality with life-threatening complications, such as virus-associated haemophagocytic syndrome (VAHS), due probably to cytokinenaemia induced by EBV-infected lymphoid cells, EBV-positive lymphoid neoplasia mainly in T cell and natural killer (NK) cell lineage and cardiovascular diseases and large-vessel arteritis with infiltration of EBV-positive lymphoid cells [1–5].

Approximately one-fourth of CAEBV patients develop inflammatory vascular lesions, which would be fatal [1]. Nakagawa et al. reported the histopathology of coronary disease developing in a CAEBV patient [4]. They observed marked lymphoid infiltration in the subepicardial adipose tissue, mild perivascular lymphoid infiltration, fibrinoid necrosis of some coronary arteries and coronary aneurysms [4]. In their case, the infiltrating lymphoid cells exhibited T cell markers such as CD3 and CD45RO. On the other hand, Murakami et al. reported large-vessel arteritis involving coronary arteries, aorta and its major branches in a CAEBV patient [5]. Furthermore, affinity of EBV-infected lymphoid cells to blood vessels has been reported previously in EBV-positive NK/T cell lymphomas [6, 7]. The terms ‘angiocentric’ or ‘angiodestructive pattern’ are often used to describe these lesions of the lymphoma tissue [6]. These suggest an affinity of EBV-positive NK/T cells to vascular components, irrespective of neoplastic transformation.

Interestingly, some CAEBV patients show skin symptoms such as hypersensitivity to mosquito bite (HMB) and hydroa vacciniforme-like eruptions [8]. We have reported previously the vasculitis lesion induced by mosquito bite with the infiltration of non-neoplastic EBV-positive cells and the subsequent development of NK/T cell lymphoma with angiodestruction [9]. In the mosquito bite lesion, the expression of EBV genes and deposition of immunoglobulins (Igs) were not detected in the affected arterial wall. Furthermore, mosquito-specific IgE antibodies were not detected in the patient [9], suggesting that the development of vasculitis at the sites of mosquito bites is not mediated by some antigen-specific immunological mechanisms. Also, the nose for NK cell lymphomas and insect bite for vasculitis are the favourite sites of inflammation for these lesions. This suggests that a non-specific inflammatory response promotes the gathering of EBV-positive NK cells along the vascular structure.

The interaction between inflammatory cells and vascular structure must be initiated with leucocyte adhesion to endothelial cells. As for NK cells, lymphocyte function-associated antigen-1 (LFA-1; αLβ2-integrin) [10] and very late T cell antigen-4 (VLA-4; α4β1-integrin) [11] are expressed on NK cells and counter-receptors of intercellular adhesion molecule-1 (ICAM-1) [12] and vascular cell adhesion molecule-1 (VCAM-1) [13], respectively, which are expressed constitutively on endothelial cells and further induced strongly by inflammatory cytokines, such as interleukin (IL)-1β, tumour necrosis factor (TNF)-α and interferon (IFN)-γ[14–16]. Furthermore, EBV-infected NK/T cells produce several cytokines which activate macrophages and would be the cause of subsequent VAHS [17–19]. These cytokines would also activate endothelial cells to promote cell adhesion [14–16]. Thus, in the site of inflammation, EBV-positive NK cells might adhere easily to endothelial cells expressing ICAM-1 and VCAM-1 strongly, and could initiate the vascular lesions.

In the current study, we evaluate the expression of adhesion molecules and cytokines in EBV-positive NK lymphoma cell lines, and examine the role of inflammatory cytokines in the interaction between lymphoma cell lines and endothelial cells.

Materials and methods

Cells

SNK1 [20] and SNK6 [21] were established from the peripheral blood and tumour tissue specimens of nasal NK/T cell lymphoma cases, respectively, and were grown in RPMI-1640+7 medium (Nikken Bio-Medical Laboratories, Kyoto, Japan) supplemented with 10% heat-inactivated human AB serum (Nabi, Boca Raton, FL, USA) and 700 U/ml of recombinant human IL-2. Human T cell lines, Jurkat and Molt14, were maintained in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum (Cansera International Inc., Etoblcoke, Ontario, Canada). Human coronary arterial endothelial cells (HCAEC) and human skin microvascular endothelial cells from adult (HMVEC-Ad) were obtained from Cambrex (Walkersville, MD, USA), and were cultured in the growth medium supplied specifically by the manufacturer.

RNA extraction and real-time reverse transcriptase–polymerase chain reaction (RT–PCR) analyses

Total RNA was extracted from frozen tissues and cell lines using Trizol reagent (Invitrogen, Carlsbad, CA, USA), quantified by measuring optical density (OD)260, and ethanol precipitated at −80°C until use. For the reverse transcription of RNA samples, 1 μg of total RNA was converted to cDNA with TaqMan reverse transcription reagents (Applied Biosystems, Foster City, CA, USA) in a 50-μl reaction volume in the presence of random hexamers according to the manufacturer's protocol. For real-time PCR analyses, a 2-μl aliquot of the cDNA sample was diluted into 25 μl of a solution containing TaqMan universal PCR master mix and each predeveloped TaqMan assay reagent (#NM 000576 for human IL-1β, #NM 000594 for human TNF-α, #NM 000619 for human IFN-γ, #NM 000201 for human ICAM-1, #NM 001078 for human VCAM-1, #NM 000885 for human integrin α4, #NM 002209 for human integrin αL and #NM 002046 for human glyceraldehyde-3-phosphate dehydrogenase (GAPDH); Applied Biosystems). Real-time PCR amplification was performed using the ABI PRISM 7700 sequence detection system (Applied Biosystems), according to the manufacturer's protocol.

Immunoblot analyses

Cells were lysed and cell lysates were separated and transferred to membranes, as described previously [22]. After incubation in 5% non-fat dried milk in Tris-buffered saline (TBS), pH 7·6, containing 0·1% Tween 20 (TBST) at room temperature for 3 h, membranes were incubated at 4°C overnight with antibody solutions diluted with 5% bovine serum albumin (Fraction V; Sigma-Aldrich, St Louis, MO, USA) in TBST. Antibodies used were mouse monoclonal antibodies to ICAM-1 (Santa Cruz Biotech, Inc., Santa Cruz, CA, USA), VCAM-1 (Santa Cruz Biotech, Inc.) and β-actin (Sigma-Aldrich), rabbit polyclonal antibody to integrin α4 (Santa Cruz Biotech, Inc.) and goat polyclonal antibody to integrin αL (Santa Cruz Biotech, Inc.). After washing with TBST, membranes were incubated at room temperature for 1 h with horseradish peroxidase (HRP)-linked F(ab′)2 fragment of sheep anti-mouse Igs (Amersham Biosciences, Amersham, Bucks, UK), with HRP-linked F(ab′)2 fragment of donkey anti-rabbit Igs (Amersham Biosciences) or with HRP-labelled protein G (Sigma-Aldrich). After washing with TBST and subsequently with TBS, signals were generated with ECL Plus detection reagents (Amersham Biosciences) by chemiluminescence and exposed at room temperature, according to the protocols suggested by the manufacturer.

Recombinant cytokines

Recombinant human IL-1β (rhIL-1β; 1 × 107 U/mg protein) and recombinant human TNF-α (rhTNFα; 2 × 107 U/mg protein) were purchased from PeproTechEC (London, UK). Human IFN-γ was kindly provided by Ohtsuka Pharmaceutical Co. (Tokyo, Japan).

Measurement of cytokines

The amount of human IL-1β, TNF-α and IFN-γ in the culture supernatant of cell lines was quantified by enzyme-linked immunosorbent assay (ELISA) using human IL-1β, TNF-α and IFN-γ ELISA kits (R&D Systems, Inc., Minneapolis, MN, USA), respectively.

Endothelial cell monolayer adhesion assay

Human endothelial cells were seeded in six-well tissue culture plates. After the endothelial cells reached subconfluency, they were stimulated with various cytokines for the indicated time-periods. Then, where appropriate, endothelial cells were incubated with antibodies diluted with growth medium at 37°C for 60 min. The antibodies used in this study were mouse monoclonal antibodies with neutralizing activity to human ICAM-1 (clone W-CAM-1; Laboratory Vision Corp., Fremont, CA, USA) [23, 24] and to human VCAM-1 (clone 4B2; R&D Systems, Inc.). Mouse monoclonal antibody to keyhole limpet haemocyanin (KLH; R&D Systems, Inc.) was used as control. Human NK cell lines were harvested and resuspended to a density of 1 × 106 cells/ml, and labelled with 20 μg/ml of calcein AM (Molecular Probes, Eugene, OR, USA) in the growth medium for NK cell lines at 37°C for 45 min [25]. After washing once with Dulbecco's phosphate-buffered saline, human NK cell lines were resuspended if necessary to a density of 1·5 × 106 cells/ml and incubated with antibodies diluted with RPMI-1640+7 medium at 37°C for 60 min. In addition to the antibodies used in the treatment of endothelial cells, mouse monoclonal antibodies with neutralizing activity to human integrin αL (α chain of human LFA-1) (clone TS1/22; Endogen, Rockford, IL, USA) [26] and to human integrin α4 (clone 2B4; R&D Systems) [26] were used at this step. After washing once with RPMI-1674+7 medium, human NK cell lines were resuspended to a density of 1·2 × 105 cells/ml in the growth medium. After aspiration of the medium with or without antibodies from the culture of endothelial monolayer, 2 ml of cell suspensions of NK cell lines was added to each well and incubated at 37°C for 1 h, as described previously [27]. The wells were washed to remove unattached cells, and attached cells were visualized using an inverted fluorescence microscope (Olympus, Tokyo, Japan), as described previously [27]. Fluorescence was stored as digital images, and the number of adhering cells in two of 4× objective fields was counted using the Image Processor for Analytical Pathology (IPAP; Sumitomo-Kagaku Technoservice Inc., Takarazuka, Japan) system.

Statistical analysis

Statistical significance was determined by the two-tailed Student's t-test. P < 0·05 was considered statistically significant.

Results

Expression of adhesion molecules in lymphoid cell lines

The expression of adhesion molecules, LFA-1 (αLβ2-integrin) and VLA-4 (α4β1-integrin), which work as the major adhesion molecules for attachment to endothelial cells, was examined at both mRNA and protein levels. The expression of ICAM-1 and VCAM-1, which work mainly on the endothelial side at the lymphoid cell attachment to endothelial cells, was also examined.

Using real time RT–PCR analyses, mRNA expression of integrin αL in SNKs was higher than that in EBV-negative T cell lines, Jurkat and Molt14 (Fig. 1a). The expression of integrin α4 mRNA in SNKs, on the other hand, was lower than that in EBV-negative T cell lines (Fig. 1a). Surprisingly, the expression of ICAM-1 and VCAM-1 in SNKs was much higher than that in EBV-negative T cell lines, more than 104-fold in VCAM-1 mRNA expression (Fig. 1a).

Fig. 1.

Fig. 1

Open in a new tab

Expression of adhesion molecules in Epstein–Barr virus (EBV)-positive and -negative lymphoid cell lines. (a) Real-time reverse transcriptase–polymerase chain reaction (RT–PCR) analyses for adhesion molecules. Bars indicate the relative expression level of mRNA compared to that in Molt14 cells, which are standardized with the expression of mRNA of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Closed bar, SNK1 (K1); hatched bar, SNK6 (K6); dotted bar, Jurkat (J); open bar, Molt14 (M). (b) Western blot for adhesion molecules; 500 μl of cell lysates were prepared from 5 × 106 cells, and a 10-μl (for β-actin) or 30-μl (for adhesion molecules) of aliquot of each sample was applied to each wall. After electrophoresis in 8% (β-actin) or 12% (adhesion molecules) polyacrylamide gels, proteins were transferred to a polyvinylidene difluoride membrane, incubated with the antibodies, and visualized as described in Materials and methods.

Using Western blot analyses, higher expression of ICAM-1 and VCAM-1 proteins was observed in SNKs than that in Jurkat and Molt14, which might be less than sensitivity of the current analyses (Fig. 1b). Conversely, the difference in integrin α4 protein expression between SNKs and EBV-negative T cell lines was not obvious (Fig. 1b). Western blot analysis for integrin αL did not produce any significant signals (results not shown), probably indicating the weak protein expression in the cells, which might be less than sensitivity of the current analysis. We also tried flow cytometric analyses for the expression of LFA-1. However, the higher binding of isotype-matched negative control antibodies interfered with the evaluation of protein expression (data not shown).

Expression of cytokines in lymphoid cell lines

The expression of cytokines from lymphoid cells, which might affect the expression of adhesion molecules in endothelial cells, was examined at both mRNA and protein levels.

Using real time RT–PCR analyses, mRNA expression of TNF-α and IFN-γ in SNKs was much higher than that in EBV-negative T cell lines, Jurkat and Molt14 (Fig. 2a). The expression of IL-1β mRNA was detected only in SNK1, thus comparison of expression levels was not possible (results not shown). The expression of IL-1β mRNA in SNK1 cells was confirmed by conventional RT–PCR and agarose gel electrophoresis (data not shown).

Fig. 2.

Fig. 2

Open in a new tab

Expression of cytokines in Epstein–Barr virus (EBV)-positive and -negative lymphoid cell lines. (a) Real-time reverse transcriptase–polymerase chain reaction (RT–PCR) analyses for cytokines. Bars indicate the relative expression level of mRNA compared to that in Molt14 cells, which are standardized with the expression of mRNA of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). (b) Concentration of tumour necrosis factor (TNF)-α and interferon (IFN)-γ proteins in the culture of cell lines. After the 3-day culture of cell lines, the supernatants were harvested by centrifugation and stored at −80°C. The concentrations of TNF-α and IFN-γ were measured by enzyme-linked immunosorbent assay. The bars represent the mean values of duplicate from three independent supernatant preparations plus/minus standard deviation. Closed bar, SNK1; hatched bar, SNK6; dotted bar, Jurkat; open bar, Molt14.

Using ELISA, TNF-α and IFN-γ protein was detected in 3-day culture supernatants of SNKs (Fig. 2b). Conversely, culture supernatant of EBV-negative T cell lines contained only IFN-γ protein at a lower level than that of SNKs, and TNF-α secreted from EBV-negative T cell lines was less than the sensitivity of the assay (3 pg/ml) (Fig. 2b). IL-1β protein was not detected in the culture supernatants of any cell lines, although SNK1 exhibited IL-1β mRNA.

Adhesion of SNK cell lines to cytokine-activated endothelial cells

We next examined the in vitro adhesion phenomenon between endothelial cells stimulated with cytokines [26, 28–30] and NK cell lines.

TNF-α-stimulated HCAEC adhered to many round cells, SNK1 and SNK6, as shown in Fig. 3a and c, respectively. On the other hand, non-stimulated HCAEC adhered to only a few SNK cells (Fig. 3b and d). To quantify the adhesion phenomenon with the clear distinction between SNK cells and endothelial cells, we next performed cell adhesion experiments using the labelled SNK cells with the fluorescence dye, calcein, for easier discrimination from endothelial cells. As shown in Fig. 3e, SNKs adhered more frequently to IL-1β- or TNF-α-pretreated endothelial cells; the difference was significant (cytokine-treated endothelial cells versus non-treated endothelial cells: P < 0·05), except for the combination of IL-1β-stimulated HMVEC-Ad and SNK1. On the other hand, pretreatment of endothelial cells with IFN-γ enhanced cell adhesion significantly only in the combination of HCAEC and SNK6. In the adhesion of SNKs, HCAEC showed a better response to treatment with IL-1β and TNF-α than HMVEC-Ad, especially in combination with SNK6 (Fig. 3e).

Fig. 3.

Fig. 3

Open in a new tab

Adhesion between cytokine-treated endothelial cells and SNKs. (a–d) Phase-contrast microphotographs of adhesion phenomenon between human coronary arterial endothelial cells (HCAEC) and SNKs. Subconfluent culture of HCAEC in six-well plates were incubated with or without 1000 U/ml of recombinant human tumour necrosis factor (rhTNF)-α for 72 h, then the suspension of SNKs were added to the wells and incubated as described in Materials and methods. TNF-α-stimulated HCAEC culture (a, c) attaches much more round cells than non-stimulated HCAEC culture (b, d). (a) TNF-α-stimulated HCAEC and SNK1. (b) Non-treated HCAEC and SNK1. (c) TNF-α-stimulated HCAEC and SNK6. (d) Non-treated HCAEC and SNK6. (e) Effects of cytokine treatment of endothelial cells to the attachment of SNKs. Subconfluent culture of HCAEC and human skin microvascular endothelial cells from adult (HMVEC-Ad) in six-well plates were incubated for 72 h with recombinant human interleukin (rhIL)-1β (100 U/ml), rhTNF-α (1000 U/ml), human interferon (IFN)-γ (1000 U/ml) or none, then the suspensions of SNK1 (K1) or SNK6 (K6) labelled with calcein were added to the wells and incubated as described in Materials and methods. The relative numbers of adherent cells compared to those of non-stimulated endothelial cells were calculated. The data present the mean plus/minus standard deviation of three independent experiments; *P < 0·05 (cytokine-treated endothelial cells versus non-treated endothelial cells). Closed bar, IL-1β, 100 U/ml; hatched bar, TNF-α, 1000 U/ml; dotted bar, IFN-γ, 1000 U/ml; open bar, no stimulation.

Modulation of the expression of adhesion molecules on endothelial cells with cytokines

The change of expression of adhesion molecules, ICAM-1 and VCAM-1, in the cytokine-treated endothelial cells, which would be responsible for the adhesion of SNKs, was examined at both mRNA and protein levels.

Using real time RT–PCR analyses, mRNA expression of ICAM-1 and VCAM-1 in both endothelial cells stimulated with IL-1-β or TNF-α was increased greatly compared to that in non-stimulated cells. Pretreatment with IFN-γ, on the other hand, increased slightly the expression of adhesion molecules (Fig. 4a).

Fig. 4.

Fig. 4

Open in a new tab

Expression of adhesion molecules in endothelial cells stimulated with cytokines. (a) Real-time reverse transcriptase–polymerase chain reaction (RT–PCR) analyses for adhesion molecules. Subconfluent cultures of human coronary arterial endothelial cells (HCAEC) and human skin microvascular endothelial cells from adult (HMVEC-Ad) were incubated for 24 h with recombinant human interleukin (rhIL)-1β (100 U/ml), recombinant human tumour necrosis factor (rhTNF)-α (1000 U/ml), human interferon (IFN)-γ (1000 U/ml) or none. Cells were harvested, and the expression of mRNA of intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1) was examined as described in Materials and methods. Bars indicate the relative expression level of mRNA compared to that in the non-stimulated endothelial cultures, which are standardized with the expression of mRNA of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Closed bar, IL-1β, 100 U/ml; hatched bar, TNF-α, 1000 U/ml; dotted bar, IFN-γ, 1000 U/ml; open bar, no supplements. (b) Western blot for adhesion molecules. Subconfluent cultures of HCAEC and HMVEC-Ad were incubated for 72 h with rhIL-1β (100 U/ml), rhTNF-α (1000 U/ml), human IFN-γ (1000 U/ml) or none. Cells were harvested, 500 μl of cell lysates were prepared from 5 × 106 cells and a 10-μl (for β-actin) or 30-μl (for adhesion molecules) of aliquot of each sample was applied to each wall. After electrophoresis in 8% (β-actin) or 12% (adhesion molecules) polyacrylamide gels, proteins were transferred to a polyvinylidene difluoride membrane, incubated with the antibodies, and visualized as described in Materials and methods.

Using Western blot analyses, higher protein expression of ICAM-1 and VCAM-1 was observed in both endothelial cells stimulated with IL-1β or TNF-α (Fig. 4b), compatible with the results in real-time RT–PCR analyses. Pretreatment with IFN-γ, on the other hand, also slightly increased the expression of adhesion molecules at protein level, except for VCAM-1 in HMVEC-Ad (Fig. 4b).

Inhibition of cell adhesion between SNK cell lines and cytokine-activated endothelial cells with anti-adhesion molecule-antibodies

To identify the adhesion molecules responsible for the increased adhesion between SNK cells and IL-1β or TNF-α-stimulated endothelial cells (Fig. 3e), we further performed the inhibition assays using anti-adhesion molecule-antibodies in SNKs adhesion to HCAEC, which exhibited a better response to cytokine treatment in the adhesion of SNKs than HMVEC-Ad.

Pretreatment of IL-1β-activated HCAEC, but not TNF-α-activated HCAEC, with anti-VCAM-1-antibodies reduced significantly the number of adherent SNKs compared with control antibodies (Fig. 5a). This indicates that VCAM-1 induced on IL-1β-stimulated endothelial cells is the adhesion molecule responsible for endothelial cell–SNK interactions. Pretreatment with anti-ICAM-1-antibodies, on the other hand, significantly enhanced the cell adhesion phenomenon between cytokine-activated HCAEC and SNKs except for the combination of IL-1β-stimulated HCAEC and SNK1 (Fig. 5a).

Fig. 5.

Fig. 5

Open in a new tab

Effects of anti-adhesion molecule-antibodies in the cell adhesion between cytokine-activated human coronary arterial endothelial cells (HCAEC) and SNK cells. Effects of the treatment of cytokine-activated HCAEC with anti-adhesion molecule-antibodies (a) and the treatment of SNK cells with anti-adhesion molecule-antibodies (b). Subconfluent culture of HCAEC in six-well plates were incubated for 72 h with recombinant human interleukin (rhIL)-1β (100 U/ml) or hrTNF-α (1000 U/ml). In (a), HCAEC was incubated subsequently with anti-adhesion molecule-antibodies at 37°C for 60 min. The suspensions of SNK1 (K1) or SNK6 (K6) were labelled with calcein. In (b), SNKs were incubated subsequently with anti-adhesion molecule antibodies at 37°C for 60 min. Then, the labelled SNKs were added to the wells and incubated as described in Materials and methods. The relative numbers of adherent cells compared to those of SNK6 adhering to TNF-α-stimulated HCAEC without antibody were calculated. The data present the mean plus/minus standard deviation of three independent experiments; P < 0·05 (anti-adhesion molecule antibody-treated cells versus control antibody-treated cells with the same cytokine stimulation). I, anti-intercellular adhesion molecule-1 (ICAM-1), 30 μg/ml; V, anti-vascular cell adhesion molecule-1 (VCAM-1), 30 μg/ml; αL, anti-integrin αL, 30 μg/ml; α4, anti-integrin α4,5 μg/ml; C, anti-KLH (control mouse IgG), 30 μg/ml; (–), no antibody.

Conversely, pretreatment of SNK6 with anti-VCAM-1 antibodies before adding to the endothelial monolayers reduced the cell adhesion significantly to HCAEC stimulated with IL-1β, and somewhat reduced the adhesion to HCAEC stimulated with TNF-α, compared with control antibodies (Fig. 5b). Pretreatment of SNK6 with anti-integrin αL and anti-integrin α4 antibodies reduced cell adhesion to HCAEC stimulated with IL-1β and TNF-α, respectively. However, the difference was not significant compared with control antibodies (Fig. 5b). Pretreatment of SNK1 with these three antibodies did not change the cell adhesion to cytokine-activated HCAEC compared with control antibodies (Fig. 5b). Pretreatment of SNKs with anti-ICAM-1 antibodies also enhanced significantly the cell adhesion phenomenon between cytokine-activated HCAEC and SNKs, except the combination of IL-1β-stimulated HCAEC and SNK6 (Fig. 5b).

Discussion

In the current study, we showed clearly that EBV-positive NK cell lines adhered to cultured endothelial cells via cytokine-induced VCAM-1 (Figs 4a and b and 5a). In SNK cells the expression of adhesion molecules which work primarily on the endothelial cell side, ICAM-1 and VCAM-1, were highly enhanced when compared to EBV-negative T cell lines, although the expression of counter-receptors on the lymphoid cell side, LFA-1 and VLA-4, were not greatly different from those in T cell lines (Fig. 1a and b). In fact, cutaneous CD56-positive NK/T cell lymphomas with angiodestructive lesions express ICAM-1 and VCAM-1 more frequently in vivo than those without angiodestruction, as detected by immunohistochemistry [7]. These point out the possible homophilic interaction between VCAM-1 on EBV-positive NK cells and that on cytokine-stimulated endothelial cells in the adhesion phenomenon. However, the inhibitory effects of anti-VCAM-1 antibodies on SNK cells were not as obvious as those on cytokine-activated endothelial cells and, to a certain extent, anti-integrin α4 and anti-integrin αL antibodies also reduced cell adhesion. Thus, the responsible adhesion molecule on SNK cells was still uncertain. Furthermore, the biological role of homophilic interaction of VCAM-1 has not been recognized, although the homophilic interaction is expected at the molecular level as a member of immunoglobulin supergene family. Therefore, the role of homophilic interaction between VCAM-1 is still uncertain in adhesion, and we point out the important role of VCAM-1 only on endothelial cells.

Bender et al. reported that the pretreatment of peripheral blood NK cells with anti-LFA-1 antibodies reduces adhesion to HMVEC derived from the foreskin of newborns, suggesting the involvement of ICAM-1 in NK cell adhesion to endothelial cells [31]. We used HCAEC, not HMVEC derived from adult skin, as cultured endothelial cells because of its higher responsiveness in NK cell adhesion to cytokine treatment (Fig. 3e). Thus, the molecule responsible for NK cell adhesion might be different among the endothelial cells derived from specific organs, i.e. HCAEC from heart coronary arteries and HMVEC from adult skin or newborn foreskin, and these differences might explain the organ specificity of the vasculitis lesions. Furthermore, in our assay system endothelial cells were stimulated with inflammatory cytokines, and EBV-infected NK cell lines instead of peripheral blood NK cells were adhered. Activated conditions of adhesive cells at both endothelial cells and leucocytes would affect the adhesion molecules responsible for the adhesion.

In the adhesion inhibition experiments using anti-adhesion molecule-antibodies, pretreatment of both endothelial cells and NK cell lines with anti-ICAM-1-antobodies enhanced the adhesion phenomenon, rather than showing no inhibitory effects (Fig. 5a and b). In our cell adhesion study, the endothelial monolayers became a little sparse after much longer incubation (6 h) with SNK cells (results not shown), due probably to the killing by SNK cells. Because LFA-1, not VLA-4, is an activating receptor of NK cells [32], SNK cells might be stimulated via the heterophilic interaction between LFA-1 and ICAM-1, and injure the endothelial cells more efficiently. This might lead to the phenomenon of a somewhat sparse endothelial monolayer. Anti-ICAM-1-antibodies, at least on endothelial cells, would inhibit this activation process of SNK cells, preserve the endothelial culture in better conditions, increase the number of adhering cells and compensate the reduced number of adherent cells via the original blocking process between ICAM-1 and LFA-1 (Fig. 5a). On the other hand, VCAM-1 is not included in the activation process of NK cells, and therefore anti-VCAM-1-antibodies would exhibit the original blocking process alone. The mechanisms of adhesion-promoting effects of anti-ICAM-1-antibodies on the SNK cell side (Fig. 5b) are, however, still unknown. In the development of vasculitis lesions, the injury of endothelial cells subsequent to leucocyte adhesion must be important. Our system would also be useful for analysis of the mechanism of endothelial injury by activated leucocytes, and we are now beginning evaluation of endothelial injury by SNK cells.

Human endothelial cells stimulated with TNF-α or IL-1β showed increased adhesion of SNKs (Fig. 3e). Therefore, up-regulated cytokine production would be important for the interaction between NK cells and endothelial cells, which might lead to the development of vasculitis lesions. We shown the up-regulated cytokine expression in SNK cells, such as TNF-α and IFN-γ, in Fig. 2a and b. The EBV genes responsible for the up-regulated cytokine expression are, however, still unknown. Although the function of EBV genes in NK lineage cells is completely unknown among EBV latent infection genes expressed in EBV-positive NK lineage cells, latent membrane protein 1 (LMP1) and EBV-encoded small RNA 1 and 2 (EBERs) are reported to have the function of transcriptional activation of cellular genes in human B-lineage cells: EBV infection of normal B cells (LCL) induces LFA-1 expression mediated by LMP1 [33] and enforced expression of EBERs in EBV-negative Burkitt's lymphoma cell lines induces IL-10 expression [34]. Thus, we are now trying to encourage the stable transformants of EBV-negative T cell lines to express EBERs or LMP1, which might exhibit alterations in the expression of cytokines and adhesion molecules, and furthermore we are currently beginning RNA interference experiments for some EBV genes in EBV-positive NK cell lines in order to identify the EBV genes responsible for these functional alterations.

In this study, we showed the adhesion of EBV-positive NK cells to cultured endothelial cells and clarified the responsible adhesion molecule on endothelial cells and the role of inflammatory cytokines in this phenomenon. Further studies concerning the identification of EBV genes responsible for the up-regulation of cytokines and adhesion molecules and the subsequent injury of endothelial cells would be necessary for delineation of the mechanisms of vasculitis lesions in CAEBV patients.

Acknowledgments

We thank Ms Etsuko Komai, Ms Mayumi Iwasaki-Endo and Mr Yuji Shibata for technical assistance, and Ms Kasumi Fujiwara-Suzuki and Ms Yumi Kanno for secretarial assistance. This work was supported in part by a grant from the Naito Foundation and grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology (17590359 and High-Tech Research Center Project for Private Universities 2004-08), Japan.

References

  • 1.Kimura H, Hoshino Y, Kanegane H, et al. Clinical and virologic characteristics of chronic active Epstein–Barr virus infection. Blood. 2001;98:280–6. doi: 10.1182/blood.v98.2.280. [DOI] [PubMed] [Google Scholar]
  • 2.Kasahara Y, Yachie A, Takei K, et al. Differential cellular targets of Epstein–Barr virus (EBV) infection between acute EBV-associated hemophagocytic lymphohistiocytesis and chronic active EBV infection. Blood. 2001;98:1882–8. doi: 10.1182/blood.v98.6.1882. [DOI] [PubMed] [Google Scholar]
  • 3.Kinura H, Hoshino Y, Hara S, et al. Differences between T-cell-type and natural killer cell-type chronic active Epstein–Barr virus infection. J Infect Dis. 2005;191:531–9. doi: 10.1086/427239. [DOI] [PubMed] [Google Scholar]
  • 4.Nakagawa A, Ito M, Iwaki T, Yatabe Y, Asai J, Hayashi K. Chronic active Epstein–Barr virus infection with giant coronary aneurysms. Am J Clin Pathol. 1996;105:733–6. doi: 10.1093/ajcp/105.6.733. [DOI] [PubMed] [Google Scholar]
  • 5.Murakami K, Ohsawa M, Hu S-X, Kanno H, Aozasa K, Nose M. Large-vessel arteritis associated with chronic active Epstein–Barr virus infection. Arthritis Rheum. 1998;41:369–73. doi: 10.1002/1529-0131(199802)41:2<369::AID-ART22>3.0.CO;2-S. [DOI] [PubMed] [Google Scholar]
  • 6.Jaffe ES, Chan JKC, Su I-J, et al. Report of the workshop on nasal and related extranodal angiocentric T/natural killer cell lymphomas: definitions, differential diagnosis, and epidemiology. Am J Surg Pathol. 1996;20:103–11. doi: 10.1097/00000478-199601000-00012. [DOI] [PubMed] [Google Scholar]
  • 7.Takeshita M, Yamamoto M, Kikuchi M, et al. Angiodestruction and tissue necrosis of skin-involving CD56+ NK/T-cell lymphoma are influenced by expression of cell adhesion molecules and cytotoxic granule and apoptosis-related proteins. Am J Clin Pathol. 2000;113:201–11. doi: 10.1309/BFH5-NCNP-DK3J-DQBH. [DOI] [PubMed] [Google Scholar]
  • 8.Kawa K, Okamura T, Yagi K, Takeuchi M, Nakayama M, Inoue M. Mosquito allergy and Epstein–Barr virus-associated T/natural killer-cell lymphoproliferative disease. Blood. 2001;98:3173–4. doi: 10.1182/blood.v98.10.3173. [DOI] [PubMed] [Google Scholar]
  • 9.Kanno H, Onodera H, Endo M, et al. Vascular lesion in a patient of chronic active Epstein–Barr virus infection with hypersensitivity to mosquito bites: vasculitis induced by mosquito bite with the infiltration of nonneoplastic Epstein–Barr virus-positive cells and subsequent development of natural killer/T-cell lymphoma with angiodestruction. Hum Pathol. 2005;36:212–18. doi: 10.1016/j.humpath.2004.11.005. [DOI] [PubMed] [Google Scholar]
  • 10.Pardi R, Bender JR, Dettori C, Giannazza E, Engleman EG. Heterogeneous distribution and transmembrane signaling properties of lymphocyte function-associated antigen (LFA-1) in human lymphocyte subsets. J Immunol. 1989;143:3157–66. [PubMed] [Google Scholar]
  • 11.Mainiero F, Gismondi A, Milella M, et al. Long term activation of natural killer cells results in modulation of β1-integrin expression and function. J Immunol. 1994;152:446–54. [PubMed] [Google Scholar]
  • 12.Marlin SD, Springer TA. Purified intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen 1 (LFA-1) Cell. 1987;51:813–19. doi: 10.1016/0092-8674(87)90104-8. [DOI] [PubMed] [Google Scholar]
  • 13.Elices MJ, Osborn L, Takada Y, et al. VCAM-1 on activated endothelium interacts with the leukocyte integrin VLA-4 at a site distinct from the VLA-4/fibronectin binding site. Cell. 1990;60:577–84. doi: 10.1016/0092-8674(90)90661-w. [DOI] [PubMed] [Google Scholar]
  • 14.Myers CL, Wertheimer SJ, Schembri-King J, Parks T, Wallace RW. Induction of ICAM-1 by TNF-α, IL-1β, and LPS in human endothelial cells after downregulation of PKC. Am J Physiol. 1992;263:C767–72. doi: 10.1152/ajpcell.1992.263.4.C767. [DOI] [PubMed] [Google Scholar]
  • 15.Swerlick RA, Lee KH, Li L-J, Sepp NT, Caughman SW, Lawley TJ. Regulation of vascular cell adhesion molecule 1 on human dermal microvascular endothelial cells. J Immunol. 1992;149:698–705. [PubMed] [Google Scholar]
  • 16.Zadeh MS, Kolb J-P, Geromin D, et al. Regulation of ICAM-1/CD54 expression on human endothelial cells by hydrogen peroxide involves inducible NO synthase. J Leukoc Biol. 2000;67:327–34. doi: 10.1002/jlb.67.3.327. [DOI] [PubMed] [Google Scholar]
  • 17.Yoshiyama H, Shimizu N, Takada K. Persistent Epstein–Barr virus infection in a human T cell line: unique program of latent virus expression. EMBO J. 1995;14:3706–11. doi: 10.1002/j.1460-2075.1995.tb00040.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Lay J-D, Tsao C-J, Chen J-Y, Kadin ME, Su I-J. Upregulation of tumor necrosis factor-α gene by Epstein–Barr virus and activation of macrophages in Epstein–Barr virus-infected T cells in the pathogenesis of hemophagocytic syndrome. J Clin Invest. 1997;100:1969–79. doi: 10.1172/JCI119728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ohga S, Nomura A, Takada H, et al. Epstein–Barr virus (EBV) load and cytokine gene expression in activated T-cells of chronic active EBV infection. J Infect Dis. 2001;183:1–7. doi: 10.1086/317653. [DOI] [PubMed] [Google Scholar]
  • 20.Nagata H, Numata T, Konno A, et al. Presence of natural killer-cell clones with variable proliferative capacity in chronic active Epstein–Barr virus infection. Pathol Int. 2001;51:778–85. doi: 10.1046/j.1440-1827.2001.01276.x. [DOI] [PubMed] [Google Scholar]
  • 21.Nagata H, Konno A, Kimura N, et al. Characterization of novel natural killer (NK)-cell and γδT-cell lines established from primary lesions of nasal T/NK-cell lymphomas associated with the Epstein–Barr virus. Blood. 2001;97:708–13. doi: 10.1182/blood.v97.3.708. [DOI] [PubMed] [Google Scholar]
  • 22.Yoshimura F, Kanno H, Uzuki M, Tajima K, Shimamura T, Sawai T. Downregulation of inhibitor of apoptosis proteins in apoptotic human chondrocytes treated with tumor necrosis factor-alpha and actinomycin D. Osteoarthritis Cartilage. 2006;14:435–41. doi: 10.1016/j.joca.2005.11.003. [DOI] [PubMed] [Google Scholar]
  • 23.Boyd AW, Wawryk SO, Burns GF, Fecondo JV. Intercellular adhesion molecule 1 (ICAM-1) has a central role in cell–cell contact-mediated immune mechanisms. Proc Natl Acad Sci USA. 1988;85:3095–9. doi: 10.1073/pnas.85.9.3095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Boyd AW, Dunn SM, Fecondo JV, et al. Regulation of expression of a human intercellular adhesion molecule (ICAM-1) during lymphohematopoietic differentiation. Blood. 1989;73:1896–903. [PubMed] [Google Scholar]
  • 25.Werther WA, Gonzalez TN, O'Connor SJ, et al. Humanization of an anti-lymphocyte function-associated antigen (LFA)-1 monoclonal antibody and reengineering of the humanized antibody for binding to rhesus LFA-1. J Immunol. 1996;157:4986–95. [PubMed] [Google Scholar]
  • 26.Lou J, Gasche Y, Zheng L, et al. Interferon-β inhibits activated leukocyte migration through human brain microvascular endothelial cell monolayer. Lab Invest. 1999;79:1015–25. [PubMed] [Google Scholar]
  • 27.Watabe D, Kanno H, Yoshida A, Kurose A, Akasaka T, Sawai T. Adhesion of peripheral blood mononuclear cells and CD4+ T-cells from patients with psoriasis to cultured endothelial cells via the interaction between lymphocyte function-associated antigen type 1 and intercellular adhesion molecule 1. Br J Dermatol. 2007;157:259–65. doi: 10.1111/j.1365-2133.2007.08039.x. [DOI] [PubMed] [Google Scholar]
  • 28.Thornhill MH, Haskard DO. IL-4 regulates endothelial cell activation by IL-1, tumor necrosis factor, or IFN-γ. J Immunol. 1990;145:865–72. [PubMed] [Google Scholar]
  • 29.Doukas J, Pober JS. IFN-γ enhances endothelial activation induced by tumor necrosis factor but not IL-1. J Immunol. 1990;145:1727–33. [PubMed] [Google Scholar]
  • 30.Borghi MO, Panzeri P, Shattock R, Sozzani S, Dobrina A, Meroni PL. Interaction between chronically HIV-infected promonocytic cells end human umbilical vein endothelial cells: role of proinflammatory cytokines and chemokines in viral expression modulation. Clin Exp Immunol. 2000;120:93–100. doi: 10.1046/j.1365-2249.2000.01186.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bender JR, Pardi R, Karasek MA, Engleman EG. Phenotype and functional characterization of lymphocytes that bind human microvascular endothelial cells in vitro. Evidence for preferential binding of natural killer cells. J Clin Invest. 1987;79:1679–88. doi: 10.1172/JCI113007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Barber DF, Faure M, Long EO. LFA-1 contributes an early signal for NK cell cytotoxicity. J Immunol. 2004;173:3653–9. doi: 10.4049/jimmunol.173.6.3653. [DOI] [PubMed] [Google Scholar]
  • 33.Wang D, Liebowitz D, Wang F, et al. Epstein–Barr virus latent infection membrane protein alters the human B-lymphocyte phenotype: deletion of the amino terminus abolishes activity. J Virol. 1988;62:4173–84. doi: 10.1128/jvi.62.11.4173-4184.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kitagawa N, Goto M, Kurozumi K, et al. Epstein–Barr virus-encoded poly(A)-RNA supports Burkitt's lymphoma growth through interleukin-10 induction. EMBO J. 2000;19:6742–50. doi: 10.1093/emboj/19.24.6742. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES