Effect of Ciprofloxacin-Induced Prostaglandin E2 on Interleukin-18-Treated Monocytes

ABSTRACT Ciprofloxacin, a fluorinated 4-quinolone, is useful for the clinical treatment of infections due to its antibacterial properties and also modulates the immune response of monocytes isolated from human peripheral blood mononuclear cells. In the present study, we found that ciprofloxacin induced the production of prostaglandin E2 in monocytes in a concentration-dependent manner regardless of the presence of interleukin-18 by enhancing the expression of cyclooxygenase-2 protein and that this in turn led to the elevation of intercellular cyclic AMP in monocytes via the stimulation of prostaglandin receptors. The prostaglandin E2 and cyclic AMP production increased by ciprofloxacin was inhibited by indomethacin, a nonselective cyclooxygenase-2 inhibitor, and NS398, a selective cyclooxygenase-2 inhibitor. In addition, ciprofloxacin suppressed the interleukin-18-induced production of tumor necrosis factor alpha, gamma interferon, and interleukin-12 in peripheral blood mononuclear cells by inhibiting the expression of intercellular adhesion molecule 1, B7.1, B7.2, and CD40 on monocytes, and this effect could be reversed by the addition of indomethacin or NS398. These results indicate that ciprofloxacin exerts immunomodulatory activity via the production of prostaglandin E2 and imply therapeutic potential of ciprofloxacin for the treatment of systemic inflammatory responses initiated by interleukin-18.

A major product of cyclooxygenase (COX)-initiated arachidonic acid metabolism, prostaglandin E 2 (PGE 2 ), which is released from antigen-presenting cells, primes naive human T cells and enhances their production of anti-inflammatory cytokines while inhibiting their synthesis of proinflammatory cytokines (6,9). Among the four PGE 2 receptor subtypes, Eprostanoid 1 (EP 1 ), EP 2 , EP 3 , and EP 4 , activation of the EP 2 and EP 4 receptors leads to an increase in cyclic AMP (cAMP) levels and protein kinase A (PKA) activity (3). The stimulation of EP 2 receptors directly inhibits T-cell proliferation, while that of EP 2 and EP 4 receptors regulates antigen-presenting cell functions (11). In a previous study, we found that PGE 2 prevented the IL-18-induced expression of ICAM-1, B7.2, and CD40 on monocytes and the production of IL-12, TNF-␣, and IFN-␥ in human peripheral blood mononuclear cells (PBMC) (20,21).
The effects of fluoroquinolone antibacterial agents on immune modulation have been well documented (16), and fluoroquinolones are known to exert their bactericidal activity by inhibiting bacterial type II topoisomerases (TOPO II), a major component of mitotic chromosomes. Ciprofloxacin (CIP), a fluorinated 4-quinolone, may interact with TOPO II in human T cells, because the quinolone derivative CP-115,953, which displays high specificity against mammalian TOPO II, mimics the inducing effect of CIP on the production of IL-2 (5,17). The synthesis of IL-1␤ and TNF-␣ by lipopolysaccharide-stimulated human monocytes is significantly inhibited by CIP (18). However, little is known about the mechanism responsible for CIP activity, including the regulation of adhesion molecule expression.
In the present study, we found that CIP induces the production of PGE 2 in monocytes through the induction of COX-2 protein. Therefore, we analyzed the effect of CIP-induced PGE 2 production on the expression of ICAM-1, B7.1, B7.2, CD40, and CD40L on monocytes and the production of IL-12, TNF-␣, IFN-␥, and IL-10 in PBMC using COX inhibitors in the presence and absence of IL-18.
Isolation of PBMC and monocytes. Normal human PBMC were collected from human volunteers after obtaining their oral informed consent. Samples of 20 to 50 ml of peripheral blood were withdrawn from a forearm vein, after which the PBMC were isolated, and monocytes isolated from PBMC were separated by counterflow centrifugal elutriation as previously described (20,21). The PBMC and monocytes were then suspended at a final concentration of 10 6 cells/ml in the medium as previously described (20,21).
Flow cytometric analysis. Monocytes at 10 6 cells/ml were incubated with IL-18, CIP, NS398, indomethacin, and H-89 for 24 h at 37°C in a 5% CO 2 -air mixture under different conditions, for which the reagents were added to the medium at the start of incubation. The cells at 5 ϫ 10 5 cells/sample were prepared as previously described (20,21) and analyzed using a FACSCalibur (Becton Dickinson Biosciences, San Jose, CA), after which the data were processed using the CELL QUEST program (BD Biosciences). The results were presented as means Ϯ standard errors of the mean (SEM) for five donors.
ELISAs. PBMC at 10 6 cells/ml used to analyze cytokine production and monocytes at 10 6 cells/ml used to analyze PGE 2 production were incubated for 24 h at 37°C in a 5% CO 2 -air mixture under different conditions, for which the reagents were added to the medium at the start of incubation. After culture, IL-12 (p70), TNF-␣, IFN-␥, IL-10, and PGE 2 proteins in the cell suspensions were prepared as previously described (20,21) and measured using an enzyme-linked immunosorbent assay (ELISA) kit (IL-12 [p70], TNF-␣, IFN-␥, and IL-10 were from R&D Systems, Minneapolis, MN, and PGE 2 was from Cayman Chemical), where the detection limits of the kit for IL-12 (p70), TNF-␣, IFN-␥, IL-10, and PGE 2 were 10 pg/ml. The results were expressed as means Ϯ SEM for five donors.
Measurement of cAMP production in monocytes. Monocytes at 10 6 cells/ml were incubated at 37°C in a 5% CO 2 -air mixture under different conditions. After 24 h, cells at 2 ϫ 10 5 cells/200 l/well were supplemented with trichloroacetic acid to a final concentration of 5% and 3-isobutyl-1-methylxanthine, an inhibitor of phosphodiesterase, at 100 M and frozen at Ϫ80°C. Frozen samples were subsequently sonicated and assayed for cAMP using a cAMP enzyme immunoassay kit (Cayman Chemical) according to the manufacturer's instructions, for which no acetylation procedures were performed. The results were expressed as means Ϯ SEM for five donors.
Statistical analysis. Statistical significance was evaluated using analysis of variance followed by Dunnet's test, where a probability value less than 0.05 was considered to indicate significance.

RESULTS
The effect of CIP on COX-1 and COX-2 protein expression in monocytes. The effect of 100 g/ml CIP on COX-1 and COX-2 protein expression in monocytes in the presence and absence of 100 ng/ml IL-18 was determined by Western blot analysis after 24 h of incubation (Fig. 1). COX-1 and COX-2 expression in monocytes cultured in the medium was marginal, but the addition of CIP in the presence and absence of IL-18 remarkably induced the expression of COX-2 24 h after the start of the incubation.
The effect of CIP on PGE 2 production in monocytes. The effect of 0 to 100 g/ml CIP on PGE 2 production in medium from incubated monocytes in the presence and absence of 100 ng/ml IL-18 was determined by ELISA after 24 h of incubation ( Fig. 2A). IL-18 had no effect on the production of PGE 2 , but production induced by 100 g/ml CIP increased in a timedependent manner and reached a maximum level after 24 h. The CIP concentration directly elicited the production of PGE 2 in both the presence and absence of IL-18. At 100 g/ml, CIP induced 30 nM PGE 2 production irrespective of the presence of IL-18. The 50% effective doses for the effect of CIP on the production of PGE 2 in the presence and absence of IL-18 were 2 and 20 g/ml, respectively.
The effect of indomethacin and NS398 on the CIP-induced production of PGE 2 in monocytes. The effects of different concentrations ranging between 10 Ϫ7 and 10 Ϫ4 M indomethacin, a nonselective COX-2 inhibitor, and NS398, a selective COX-2 inhibitor, on the CIP-enhanced production of PGE 2 in monocytes in the presence and absence of 100 ng/ml IL-18 were determined by ELISA after 24 h of incubation (Fig. 2B). NS398 and indomethacin inhibited the production of PGE 2 irrespective of the presence of IL-18 in a concentration-dependent manner.
The effect of CIP on cAMP production in monocytes. The effect of 100 g/ml CIP and 30 nM PGE 2 on the elevation of intercellular cAMP in monocytes in the presence and absence of 100 ng/ml IL-18 was determined by ELISA (Fig. 3). IL-18 had no effect on the production of cAMP, but CIP and PGE 2 elicited production irrespective of the presence of IL-18. Also, NS398 at 10 Ϫ4 M blocked the inhibitory effect of CIP on the production of cAMP irrespective of the presence of IL-18. The effect of CIP on the expression of ICAM-1, B7.1, B7.2, CD40, and CD40L on monocytes. The effects of 0 to 100 g/ml CIP on the changes in expression of ICAM-1, B7.1, B7.2, CD40, and CD40L on monocytes in the presence and absence of 100 ng/ml IL-18 were determined by flow cytometry after 24 h. In the absence of IL-18, CIP inhibited the expression of ICAM-1, B7.1, B7.2, and CD40 ( Fig. 4A) but had no effect on the expression of CD40L (data not shown). The expression of CD40 (data not shown) in addition to ICAM-1 and B7.2 (9) was up-regulated in a time-dependent manner by IL-18 at 100 ng/ml and reached a maximum after 24 h. IL-18 between 0 and 100 ng/ml elicited ICAM-1, B7.2, and CD40 expression in a concentration-dependent manner (20,21), but the expression of B7.1 was not changed in the presence and absence of IL-18. CIP inhibited ICAM-1, B7.1, B7.2, and CD40 expression in a concentration-dependent manner in the presence of IL-18 ( Fig. 4A) but had no effect on the expression of CD40L (data not shown). The 50% inhibitory concentrations for the inhibitory effect of CIP on the expression of ICAM-1, B7.1, B7.2, and CD40 in the presence of IL-18 were estimated as 2, 3, 2, and 2 g/ml, respectively.
The effect of CIP on cytokine responses in PBMC. The effect of 0 to 100 g/ml CIP on the production of IL-12, IFN-␥, TNF-␣, and IL-10 in PBMC incubated in medium in the presence and absence of IL-18 was determined by ELISA after 24 h (Fig. 5A). In IL-18-treated PBMC, CIP prevented the production of IL-12, IFN-␥, and TNF-␣ but induced IL-10 production. The 50% inhibitory concentrations for the inhibitory effect of CIP on the production of IL-12, IFN-␥, and TNF-␣ in the presence of IL-18 were estimated as 3, 3, and 2 g/ml, respectively. In the absence of IL-18, CIP had no effect on cytokine production in the PBMC medium. CIP-inhibited production of IL-12, IFN-␥, TNF-␣, and IL-10 in PBMC treated with 100 ng/ml IL-18 were determined by ELISA after 24 h of incubation (Fig. 5B). NS398, indomethacin, and H-89 blocked CIP-initiated production of TNF-␣, IL-12, IFN-␥, and IL-10. The rates of TNF-␣ production recovered by indomethacin, NS398, and H-89 at 10 Ϫ4 M were 72, 63, and 61%, respectively. These inhibitors had no effect in the absence of CIP (data not shown).

DISCUSSION
The present study examined for the first time the effects of CIP on the immune response of IL-18-treated monocytes. CIP induced the expression of COX-2 protein in monocytes treated with IL-18 or not treated (Fig. 1). In the absence and presence of IL-18, an unexpectedly large concentration 30 nM PGE 2 was detected in the medium of 100 g/ml CIP-treated monocytes ( Fig. 2A). CIP-initiated endogenous PGE 2 production was inhibited by the nonselective and selective COX-2 inhibitors indomethacin and NS398 (Fig. 2B), respectively, indicating that the increase in endogenous PGE 2 production might have depended on the enhancement of COX-2 expression. CIP as well as exogenous PGE 2 induced the elevation of intercellular cAMP in monocytes irrespective of the presence of IL-18 (Fig.  3). Also, CIP-enhanced cAMP expression was abolished by NS398. These results suggest that endogenously produced PGE 2 and the elevation of cAMP are associated with the CIP-induced enhancement of COX-2 expression.
Recently, we found that PGE 2 prevented IL-18-enhanced ICAM-1, B7.2, and CD40 expression through stimulation of the EP 2 /EP 4 receptor (20). As shown in Fig. 4A, CIP suppressed IL-18-enhanced ICAM-1, B7.2, and CD40 expression on monocytes. Whereas the inhibitory effect of 100 g/ml CIP on the expression of ICAM-1 was 50% (Fig. 4A), that of 30 nM  (20). The inhibitors of COX-2 and PKA partially blocked the effect of CIP on IL-18-initiated adhesion molecule expression (Fig. 4B). Therefore, there might exist endogenous PGE 2 -dependent and -independent pathways associated with the effects of CIP activity on adhesion molecule expression in the presence of IL-18. On the other hand, whereas PGE 2 had no effect on the adhesion molecule expression in the absence of IL-18 (20), CIP inhibited the expression of ICAM-1, B7.1, B7.2, and CD40 (Fig. 4A). The COX-2 inhibitors, but not the PKA inhibitor, abolished the adhesion molecule expression-suppressing effect of CIP in the absence of IL-18 (Fig. 4B), suggesting that the endogenous PGE 2 might not be involved in the effect of CIP in the absence of IL-18. Previously, we reported that the inhibition of ICAM-1, B7.2, and CD40 expression on monocytes contributed to the suppression of IL-18-initiated cytokine production in PBMC (20,21). PGE 2 inhibited the production of IL-12, IFN-␥, and TNF-␣ but induced the production of IL-10 in PBMC treated with IL-18 (20). We found here that CIP mimicked the effect of PGE 2 on IL-18-initiated cytokine production (Fig. 5A). The inhibitors of COX-2 and PKA also blocked the effect of CIP on IL-18-initiated cytokine production (Fig. 5B). However, the rates of cytokine production as well as those of adhesion molecule expression recovered by indomethacin, NS398, and H-89 (10 Ϫ4 M) were similar and reached only between 60 and 70%. Therefore, the suppressive effect of CIP on IL-18-initiated cytokine production might depend on the inhibition of ICAM-1, B7.2, and CD40 expression.
IL-18 plays a role in inflammatory conditions, such as graftversus-host disease (15) and Crohn's disease (14), and antimicrobial chemotherapy targeted against intestinal anaerobic bacteria significantly reduces the severity of the acute stage of these diseases (4,13). CIP significantly ameliorates the severity of graft-versus-host disease and Crohn's disease by reducing the number of intestinal bacteria, some of which induce the lipopolysaccharide-initiated production of TNF-␣ (1,4,19). In a randomized crossover study, the mean maximum concentration of CIP in the serum of normal human volunteers who received a single oral dose of 500 mg for up to 24 h was 2.46 g/ml (7), which is within the range of the concentration noted in the present study. Therefore, the effects of CIP on immune responses may indicate new therapeutic potential for IL-18induced diseases.
TOPO II-targeting drugs are apoptosis-inducing drugs, and both isoforms of TOPO II, alpha and beta, are inhibited by the chemotherapeutic agent etoposide (22). We found that etoposide had no effect on COX expression, adhesion molecule expression, and cytokine production in the absence and presence of IL-18 (data not shown). Cell viabilities of monocytes FIG. 5. The effect of CIP on cytokine response of PBMC. (A) PBMC at 10 6 cells/ml were treated with between 0 and 100 g/ml CIP in the presence and absence of 100 ng/ml IL-18 for 24 h. After the treatment, IL-12, TNF-␣, IFN-␥, and IL-10 production was determined by ELISA. Filled circles and filled squares represent the results obtained with medium and IL-18, respectively. ##, P Ͻ 0.01 compared with the value for IL-18. (B) PBMC (10 6 cells/ml) were incubated with between 0 and 0.1 mM indomethacin, NS398, and H-89 in the presence and absence of 100 ng/ml IL-18 or 100 g/ml CIP for 24 h. Open squares, filled circles, and filled squares represent the results obtained with indomethacin, H-89, and NS398, respectively. ##, P Ͻ 0.01 compared with the value for IL-18 and CIP. The results are the means Ϯ SEM for five donors. When an error bar was within a symbol, the bar was omitted.

3232
TAKAHASHI ET AL. ANTIMICROB. AGENTS CHEMOTHER. and lymphocytes in the presence of CIP and/or IL-18 were almost the same and were estimated to be 90% after 24 h of incubation. Therefore, the effect of CIP might be independent of the inhibition of TOPO II, and the regulation of cytokine production and adhesion molecule expression was not due to a reduction in cell viability. It is still unclear what the primary target or binding site of CIP in monocytes is for regulating immune responses. In conclusion, we found a regulation profile exists for the antimicrobial agent CIP on monocyte responses, as seen through increased PGE 2 production.