Selective regulation of intercellular adhesion molecule-1 expression by interleukin-18 and interleukin-12 on human monocytes

Summary

Induction of expression of adhesion molecules is a crucial step in inflammation. The role of interleukin-18 (IL-18) in induction of various adhesion molecules was investigated in freshly isolated peripheral blood mononuclear cells and human monocyte and T-cell lines. IL-18 selectively up-regulated intercellular adhesion molecule-1 (ICAM-1) expression on freshly isolated human monocytes, but not on lymphocytes. The expression of other adhesion molecules was not influenced. Induction of ICAM-1 by IL-18 was dependent on endogenous tumour necrosis factor-α (TNF-α), and IL-12 had an additive effect on that of IL-18. No changes in adhesion molecule expression were observed on the monocytic cell line THP-1 and on the T-cell lines HSB-2 and Jurkat J16. In addition, induction of ICAM-1 on monocytes by lipopolysaccharide was slightly, but significantly, inhibited by blockade of either endogenous IL-18 or TNF-α with IL-18 binding protein or TNF binding protein, respectively. Blocking IL-1 effects with IL-1 receptor antagonist did not influence ICAM-1 levels. In conclusion, IL-18 selectively up-regulates the expression of ICAM-1 on monocytes, and this contributes to the proinflammatory effects of this cytokine.

Abbreviations

IL-18bp

interleukin-18 binding protein

IFN-γ

interferon-γ

TNFbp

tumour necrosis factor binding protein

IL-1Ra

interleukin-1 receptor antagonist

ICAM-1

intercellular adhesion molecule-1

LFA-1

lymphocyte function-associated antigen-1

VCAM-1

vascular cell adhesion molecule-1.

Introduction

Interleukin-18 (IL-18) is a cytokine initially described as a costimulus for interferon-γ (IFN-γ) production.¹ In the context of microbial costimulation, IL-18 is able to induce large amounts of IFN-γ.² It is, however, often overlooked that IL-18 displays a series of proinflammatory effects independent of IFN-γ. IL-18 induces proinflammatory cytokines, such as tumour necrosis factor-α (TNF-α), interleukin-1β (IL-1β), IL-6 and IL-8,^3,4 probably through activation of nuclear factor-κB (NF-κB).⁵ In experimental animals challenged with lipopolysaccharide (LPS), IL-18 contributes to lethality even in the absence of IFN-γ.⁶ Likewise, IL-18 mediates cartilage damage in IFN-γ-deficient mice with collagen arthritis.⁷ These inflammatory effects of IL-18 are not well understood, and it is possible that increased expression of adhesion molecules plays a role.

Adhesion molecule expression is an important step in the induction of inflammation. In response to microbial antigens, host cells are triggered to release cytokines such as IL-1β, TNF-α, and IFN-γ, and these are able to up-regulate expression of various adhesion molecules on monocytes and endothelial cells.⁸ Intercellular adhesion molecule-1 (ICAM-1, CD54) is a transmembrane glycoprotein and member of the immunoglobulin supergene family that is constitutively expressed on a wide variety of cell types.⁹ ICAM-1 is involved in cell–cell interactions and leukocyte extravasation at inflammatory sites by binding to two integrins, CD11a/CD18 (leucocyte function-associated antigen-1; LFA-1) and CD11b/CD18 (Mac-1), both belonging to the β₂ integrin subfamily.^10,11

In the present study, the effect of IL-18 on adhesion molecule expression on monocytes from freshly isolated peripheral blood mononuclear cells (PBMC) was investigated. The role of endogenous IL-18, TNF, and IL-1 in LPS-induced ICAM-1 up-regulation was studied by using specific cytokine inhibitors such as IL-18 binding protein (IL-18bp), TNF binding protein (TNFbp), and IL-1 receptor antagonist (IL-1Ra). We demonstrate that IL-18 selectively up-regulates ICAM-1 expression on human monocytes, whereas IL-18 does not affect any of the other adhesion molecules tested: CD14, LFA-1, Mac-1, P150, very late antigen-4 (VLA-4), VLA-5, ICAM-2, ICAM-3 or activated leukocyte cell adhesion molecule (ALCAM).

Materials and methods

Reagents

Recombinant human IL-18 was kindly provided by Dr M. Kurimoto (Fujisaki Institute, Hayashibara Biochemical Laboratories, Okayama, Japan). Recombinant human IL-12 was purchased from Peprotech (Rocky Hill, NJ). LPS was obtained from Sigma Chemical Co (Escherichia coli O55:B5, St. Louis, MO). IL-18bp was expressed in COS cells as the ‘a’ isoform and purified as a 6 histidine-tagged protein over talon as previously described.¹² TNFbp (p55 TNF soluble receptor¹³) and recombinant human IL-1Ra were a kind gift of Dr Carl Edwards (Amgen, Thousand Oaks, CA). The following primary antibodies were used: anti-CD14 (Pharmingen, San Diego, CA), antibodies to CD11a (L7¹⁴), CD11b (Kim225¹⁵), CD11c (SHCI3¹⁶), CD29 (TS2/16¹⁷), CD49d (HP2/1¹⁸), CD49e (SAM-1¹⁹), ICAM-1 (Rek-1²⁰), ICAM-2 (CBR2/2²¹), and ALCAM (1F2²²). Antibodies to CD18 (AZN-L19) and ICAM-3 (AZN-IC3) were developed at the Department of Tumour Immunology, University Medical Center St. Radboud, Nijmegen, the Netherlands.

Isolation of PBMC, induction of adhesion molecule expression, and stimulation of TNF production

Separation and stimulation of PBMC was performed as described elsewhere,²³ with minor modifications. Briefly, 10 ml EDTA tubes (Terumo, Leuven, Belgium) were filled with venous blood drawn from nine healthy volunteers between 9 and 10 am. The PBMC fraction was obtained by density centrifugation of diluted blood (1 part blood to 1 part pyrogen-free saline) over Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden). PBMC were washed twice in saline and suspended in culture medium (RPMI-1640 Dutch modification, ICN Biomedicals, Aurora, OH) supplemented with human serum 5%, gentamicin 1%, l-glutamine 1%, and pyruvate 1%. The cells were counted in a Coulter counter (Coulter Electronics, Mijdrecht, the Netherlands) and the number was adjusted to 2 × 10⁶ cells/ml and 1 × 10⁶ cells/tube were incubated in 5-ml polypropylene round-bottom tubes (Becton Dickinson, Franklin Lakes, NJ) with IL-18 (1 µg/ml), IL-12 (10 ng/ml), or LPS (1 µg/ml) in the absence or presence of the cytokine inhibitors, IL-18bp (500 ng/ml), TNFbp (10 µg/ml), or IL-1Ra (10 µg/ml), for 24 hr at 37°. In a single experiment PBMC were stimulated with 0, 100 ng/ml, 1 µg/ml, and 10 µg/ml of IL-18 and incubated for 4, 8, 24, and 48 hr. After the incubation period the cells were collected and assayed for adhesion molecule expression. In the supernatants TNF-α concentrations were measured in duplicate by radioimmunoassay (RIA) as previously described.²⁴

Cell lines

THP-1, a monocytic cell line²⁵ and Jurkat J16, a human T-cell leukaemia cell line²⁶ were maintained in RPMI-1640 supplemented with fetal calf serum 10%, gentamicin 1%, l-glutamine 1%, and pyruvate 1%. HSB-2, a human T-cell leukaemia cell line²⁷ was maintained in Iscove's modified Dulbecco's medium (Gibco, Grand Island, NY) supplemented with fetal calf serum 5%, gentamicin 1%, l-glutamine 1% and pyruvate 1%. For assay, the cells were suspended at 2 × 10⁶ cells/ml and 1 × 10⁶ cells/well were incubated in 24-well plates (Costar, Corning, NY) with or without IL-18 (1 µg/ml) for 24 hr at 37°.

Flow cytometry

For flow cytometric analysis 1 × 10⁵ cells/well were plated in a 96-well V-shaped-bottom plate, washed with PBA (PBS containing 1% bovine serum albumin and 0·05% NaN₃) and stained for 30 min at 4° with primary antibodies (5 µg/ml). The cells were washed with PBA and incubated for 30 min at 4° with fluoroscein isothiocyanate-conjugated goat anti-mouse (Fab′)₂ secondary antibody (Zymed, San Francisco, CA). After washing, the cells were analysed on a FACScan (Becton Dickinson, Oxnard, CA). The gates were set to exclude dead cells, and 5000 gated cells were analysed. Monocytes and lymphocytes were gated based on their forward-side-scatter pattern.

Statistical analysis

The experiments were repeated at least two times with three volunteers and the data represent cumulative results. Data were expressed as means ± standard error of the means (SEM). Differences between groups were analysed with the Wilcoxon signed test (double sided).

Results

Effect of IL-18 on the expression of adhesion molecules on human freshly isolated monocytes

The effect of IL-18 (1 µg/ml) on the expression of the following molecules was investigated: CD14, LFA-1, Mac-1, P150, VLA-4, VLA-5, ICAM-1, ICAM-2, ICAM-3 and ALCAM. On human monocytes, ICAM-1 was constitutively expressed. IL-18 induced significant up-regulation of ICAM-1 expression on monocytes (119% increase, P < 0·01) (Fig. 1a). The other molecules tested were not influenced by IL-18. The expression of ICAM-1 and other adhesion molecules on lymphocytes did not change after incubation with IL-18 (Fig. 1b and Table 1).

Figure 1. Effect of IL-18 on ICAM-1 expression on freshly isolated PBMC. PBMC were stimulated with or without IL-18 (1 µg/ml), and ICAM-1 expression on monocytes (a) and lymphocytes (b) was assessed after 24 hr of incubation. Data are representative of four similar experiments.

Table 1.  Effect of IL-18 on expression of adhesion molecules on human monocytes and lymphocytes*

Molecule	Monocytes			Lymphocytes
	Control	IL-18	P-value†	Control	IL-18	P-value†
* PBMC were stimulated with IL-18 (1 µg/ml). The expression of the various adhesion molecules was assessed after 24 hr of incubation. Data represent the mean fluorescence ± SEM of four separate experiments. † NS, not significant.
ICAM-1	309 ± 68	623 ± 89	P < 0·01	15 ± 3	19 ± 3	NS
CD14	810 ± 60	818 ± 96	NS	15 ± 11	10 ± 5	NS
CD18	570 ± 39	681 ± 10	NS	151 ± 17	156 ± 7	NS
CD11a	451 ± 42	474 ± 46	NS	157 ± 22	175 ± 11	NS
CD11b	563 ± 119	581 ± 41	NS	47 ± 26	48 ± 15	NS
CD11c	395 ± 41	526 ± 33	NS	8 ± 1	14 ± 2	NS
CD29	557 ± 76	543 ± 117	NS	80 ± 15	77 ± 12	NS
CD49d	64 ± 3	115 ± 47	NS	43 ± 4	46 ± 4	NS
CD49e	305 ± 47	287 ± 74	NS	24 ± 2	27 ± 5	NS
ICAM-2	89 ± 13	86 ± 20	NS	51 ± 7	50 ± 7	NS
ICAM-3	272 ± 51	253 ± 59	NS	140 ± 22	135 ± 25	NS
ALCAM	126 ± 10	101 ± 24	NS	13 ± 1	11 ± 1	NS

Dose- and time-response of IL-18 on ICAM-1 expression

PBMC were stimulated with various concentrations of IL-18. At different time points, ICAM-1 expression was determined. At a concentration of 1 µg/ml IL-18 a maximum level of ICAM-1 expression on monocytes was reached after 24 hr of incubation (mean fluorescence of 455 ± 26; Fig. 2). Similar kinetics of ICAM-1 expression were observed when PBMC were stimulated with either 100 ng/ml or 10 µg/ml IL-18, with maximal stimulation at 24 hr (mean fluorescence of 430 ± 24, and 436 ± 43, respectively). In subsequent experiments a concentration of 1 µg/ml IL-18 was used with an incubation period of 24 hr.

Figure 2. Time-response of IL-18 on ICAM-1 expression. PBMC were stimulated with (open squares) or without (closed squares) IL-18 (1 µg/ml), and ICAM-1 expression on monocytes was assessed after 4, 8, 24, and 48 hr of incubation. Data represent the mean ± SEM of three donors, expressed as mean fluorescence of ICAM-1.

Induction of ICAM-1 by IL-18 and IL-12

The potential synergistic effect of IL-18 and IL-12 on ICAM-1 expression was investigated by assessing ICAM-1 expression in the presence of either of the two cytokines alone and the combination. As shown in Fig. 3, stimulation of PBMC with IL-18 alone, IL-12 alone, and IL-18 plus IL-12 for 24 hr induced a 119, 208 and 324% increase of ICAM-1 expression on monocytes, respectively. The combination of IL-18 and IL-12 induced significantly more ICAM-1 expression on monocytes compared to either IL-18 or IL-12 alone (P < 0·05).

Figure 3. Induction of ICAM-1 by IL-18 and IL-12. PBMC were stimulated with IL-18 (1 µg/ml) and/or IL-12 (10 ng/ml). ICAM-1 expression was assessed after 24 hr of incubation. Data represent the mean ± SEM of nine donors, expressed as mean fluorescence of ICAM-1. *P < 0·01 versus control. **P < 0·05 versus either cytokine alone.

The role of IL-18 on adhesion molecule expression on THP-1, HSB-2, and Jurkat J16 cells

To investigate if the selective up-regulation of ICAM-1 on human freshly isolated monocytes was also apparent in isolated cell types, we tested the effect of IL-18 on adhesion molecule expression on the monocytic cell line THP-1, and on two T-cell lines, HSB-2 and Jurkat J16. IL-18 did not influence ICAM-1 expression (THP-1, 118% of control; HSB-2, 139% of control; Jurkat J16, 100% of control, all not significant). Moreover, the other adhesion molecules tested did not change after incubation with IL-18 (data not shown).

Induction of ICAM-1 by IL-18 and IL-12 is partially TNF-dependent

It has been shown previously that IL-18 has direct proinflammatory effects independent of IFN-γ³. As shown in Fig. 4(a), both IL-18 and IL-12 were able to induce significant amounts of TNF-α in the supernatant. The combination of IL-18 plus IL-12 significantly augmented the production of TNF-α. To evaluate whether ICAM-1 up-regulation by IL-18 and IL-18 plus IL-12 was dependent on TNF-α production, we used TNFbp to neutralize the activity of endogenously released TNF-α. As shown in Fig. 4(b), TNFbp significantly reduced both the ICAM-1 expression induced by IL-18 (−28%) and by IL-18 plus IL-12 (−36%) (P < 0·05).

Figure 4. Induction of ICAM-1 by IL-18 and IL-12 is TNF-dependent. (a) PBMC were stimulated with IL-18 (1 µg/ml) or IL-12 (10 ng/ml). TNF-α was measured after 24 hr of incubation. Data represent the mean ± SEM of nine donors. **P < 0·01 versus control. *P < 0·05 versus control. (b) PBMC were stimulated with IL-18 (1 µg/ml) or IL-12 (10 ng/ml) in the absence or presence of TNFbp (10 µg/ml). ICAM-1 expression was assessed after 24 hr of incubation. Data represent the mean ± SEM of five donors, expressed as mean fluorescence of ICAM-1. *P < 0·05.

The role of IL-18 in LPS-induced adhesion molecule expression

LPS induced a significant increase in ICAM-1 expression by 172%, and CD14 expression by 25% on human freshly isolated monocytes. The other adhesion molecules tested were not influenced by LPS (Table 2). Expression of ICAM-1 and other adhesion molecules on lymphocytes did not change after stimulation with LPS (Table 2). To investigate the role of endogenous IL-18, TNF and IL-1 in the induction of ICAM-1 by LPS, specific cytokine inhibitors were used. As shown in Fig. 5, neutralization of IL-18 and of TNF by the respective binding proteins slightly, but significantly, decreased LPS-induced ICAM-1 expression by 13 and 20%, respectively. In contrast, blocking IL-1 receptors by IL-1Ra did not influence the LPS-induced ICAM-1 expression. Blocking both endogenous IL-18 and TNF reduced ICAM-1 levels with 23% (P < 0·05). There was no significant additive effect of IL-18 bp and TNFbp.

Table 2.  Effect of LPS on expression of adhesion molecules on human monocytes and lymphocytes*

Molecule	Monocytes			Lymphocytes
	Control	LPS	P-value†	Control	LPS	P-value†
* PBMC were stimulated with LPS (1 µg/ml). The expression of the various adhesion molecules was assessed after 24 hr of incubation. Data represent the mean fluorescence ± SEM of four separate experiments. † NS, not significant.
ICAM-1	315 ± 41	857 ± 60	P < 0·01	11 ± 1	17 ± 2	NS
CD14	870 ± 54	1087 ± 95	P < 0·01	10 ± 5	5 ± 1	NS
CD18	640 ± 88	525 ± 66	NS	152 ± 20	169 ± 20	NS
CD11a	404 ± 52	343 ± 33	NS	175 ± 7	195 ± 32	NS
CD11b	530 ± 90	514 ± 105	NS	50 ± 16	63 ± 19	NS
CD11c	291 ± 56	290 ± 29	NS	7 ± 1	9 ± 1	NS
CD29	703 ± 39	625 ± 78	NS	86 ± 11	91 ± 15	NS
CD49d	50 ± 6	43 ± 4	NS	40 ± 3	39 ± 2	NS
CD49e	337 ± 9	277 ± 32	NS	19 ± 3	18 ± 3	NS
ICAM-2	101 ± 8	109 ± 22	NS	53 ± 2	54 ± 3	NS
ICAM-3	191 ± 28	178 ± 33	NS	106 ± 5	105 ± 7	NS
ALCAM	107 ± 12	82 ± 6	NS	11 ± 1	11 ± 1	NS

Figure 5. Role of endogenous IL-18, TNF, and IL-1 in LPS-induced ICAM-1 production. PBMC were stimulated with LPS (1 µg/ml) in the absence or presence of IL-18bp (500 ng/ml), TNFbp (10 µg/ml) or IL-1Ra (10 µg/ml). ICAM-1 expression was assessed after 24 hr of incubation. Data represent the mean ± SEM of six donors, expressed as mean fluorescence of ICAM-1. *P < 0·05 versus LPS.

Discussion

In this study, investigating the effect of IL-18 on adhesion molecule expression on human monocytes, we found that IL-18 selectively up-regulates ICAM-1 expression, whereas the expression of other adhesion molecules is not affected. IL-12 had an additive effect on ICAM-1 induction by IL-18, and up-regulation of ICAM-1 by IL-18 was partially dependent on initial induction of TNF-α production. The microbial stimulus LPS up-regulated ICAM-1 and CD14 expression, and ICAM-1 induction by LPS was slightly, but significantly, reduced by blocking endogenous IL-18 or TNF-α with their specific inhibitors.

Our finding of increased ICAM-1 expression on monocytes in primary culture exposed to IL-18 does not stand alone. Others have found that IL-18 induces expression of ICAM-1 on glomerular cells in a model of murine glomerulonephritis,²⁸ on endothelial cells and synovial fibroblasts,²⁹ and on a monocytic cell line KG-1.³⁰ Interestingly, lymphocytes do not respond to IL-18 with increased expression of adhesion molecules. In addition, it is remarkable that the adhesion molecule expression on monocytes is selective for ICAM-1, and not for any of the other adhesion molecules. Vascular cell adhesion molecule-1, an adhesion molecule not present on monocytes, has been shown to be induced by IL-18 on hepatic sinusoidal cells,³¹ endothelial cells and synovial fibroblasts.²⁹

We also tested if the selective up-regulation of ICAM-1 on monocytes applied for specific leucocyte cell lines. In contrast to the monocytes in the PBMC culture, none of the adhesion molecules in the monocytic cell line THP-1 was affected by stimulation with IL-18. For the two T-cell lines, HSB-2 and Jurkat J16, results were similar to those with the lymphocytes in the PBMC culture. Apparently, either the leukaemic source of these cell lines or the in vitro culture may have led to specific changes of the reactivity of the cells (e.g. altered IL-18 receptor expression) compared to monocytes from healthy individuals. Another explanation may be that the interplay between different cell types (monocytes, lymphocytes) and the presence of cytokines, for instance TNF-α, in peripheral blood is important for up-regulation of ICAM-1 by IL-18 on monocytes.

Because TNF-α is able to induce ICAM-1,³² we tested whether the up-regulation of ICAM-1 by IL-18 was dependent on intermediate induction of TNF-α. Addition of TNFbp resulted in 28% reduction of IL-18-induced ICAM-1, demonstrating that IL-18-induced ICAM-1 expression is partially dependent on endogenous TNF-α. Others were unable to show a role for endogenous TNF-α in the ICAM-1 up-regulation by IL-18.³³ The reason for this difference is unclear, but may be explained by the use of different reagents to neutralize the effect of TNF. We have used a soluble TNF receptor whereas Yoshida et al. applied an antibody against TNF.³³

It is known that IL-18 and IL-12 exert synergistic effects on IFN-γ production,² and our data suggest similar synergistic activity on the stimulation of TNF. In contrast, the effect of the combination of IL-18 plus IL-12 seems to have an additive effect on expression of ICAM-1, rather than synergistic. We cannot exclude an intermediary role of IFN-γ induction in the expression of ICAM-1. However, Kohka et al. have shown that blocking endogenous IFN-γ does not influence IL-18-induced ICAM-1 up-regulation.³⁰ Moreover, IL-18-induced ICAM-1 expression reached a maximum after 24 hr of incubation, while IL-18-stimulated IFN-γ production is maximal only after 48 hr. This makes it unlikely that endogenous IFN-γ is responsible for the IL-18 induction of ICAM-1.

We also investigated the involvement of endogenous IL-18 in the stimulation of ICAM-1 by bacterial stimuli. LPS, a component of Gram-negative bacteria, is considered to be a central mediator in the pathogenesis of Gram-negative shock. Proinflammatory cytokines such as TNF-α, IL-1β, and IL-18, are released into the circulation in response to LPS, and are thought to be responsible for the deleterious effects of LPS. LPS is known for its up-regulation of ICAM-1.³⁴ In the present study, endogenous IL-18 and TNF-α were involved in LPS-induced ICAM-1 expression, since blocking the effects of these two cytokines with IL-18bp and TNFbp, was able to partially inhibit this effect of LPS. There is no role for endogenous IL-1 in LPS-induced ICAM-1 up-regulation. Although the effect of blocking IL-18 and TNF on LPS-induced ICAM-1 is subtle, our findings are sustained by those of Beck-Schimmer et al. who showed that endogenous TNF plays a role in LPS-induced up-regulation of ICAM-1.³⁵ Others, however, have reported that LPS-induced ICAM-1 is independent of endogenous IL-1 and TNF.^36,37 Given the relatively small contributions of IL-18 and TNF to ICAM-1 up-regulation by LPS, it is tempting to speculate whether this effect of LPS is direct (e.g. through the TLR-4 pathway) or needs other unknown mediators.

Acknowledgments

We thank Marta Sanchez-Hernandez, Erik Bennink, and Roger Sutmuller for help with the experiments.