Ž .Immunopharmacology 40 1998 209–217

The effects of nonsteroidal anti-inflammatory drugs on immunefunctions of human peripheral blood mononuclear cells

Koji Tanaka, Hirokazu Tanaka, Yoshiko Kanemoto, Izumi Tsuboi )

DiÕision of Immunopharmacology, Hikone Research Laboratory, Maruho Co. Ltd., 2763 Takamiya, Hikone, Shiga 522-0201, Japan

Received 29 December 1997; revised 8 February 1998; accepted 28 July 1998

Abstract

Ž .It is generally accepted that the nonsteroidal anti-inflammatory drugs NSAIDs exhibit anti-inflammatory effectsŽ .primarily through inhibition of prostaglandin PG synthesis. However, effects of NSAIDs on immune responses are not

Ž w Ž . xfully understood. This study investigated effects of indomethacin and a new NSAID d-2- 4- 3-methyl-2-thienyl phenyl pro-.pionic acid, termed as M-5011 in this study on cytokine production, lymphocyte proliferation, activities of natural killer

Ž . Ž . Ž .NK and lymphokine activated killer LAK cells and secretion of immunoglobulin Ig . Both indomethacin and M-5011Ž .augmented interleukin IL -2 production, whereas they suppressed IL-6 production both at the protein and mRNA levels.

Ž .These two NSAIDs augmented proliferation of phytohemagglutinin PHA -stimulated PBMC and enhanced NK and LAKcell activities. In contrast, indomethacin was more potent than M-5011 in inhibition of both PG synthesis and Ig secretions

Ž .by pokeweed mitogen PWM -stimulated PBMC. These results suggest that these two NSAIDs equally augment cell-media-ted immunity, whereas indomethacin was more potent than M-5011 in the inhibition of humoral immunity and PG synthesis.q 1998 Elsevier Science B.V. All rights reserved.

Keywords: NSAIDs; PBMC; PGE2

1. Introduction

It is generally accepted that NSAIDs exhibit anti-inflammatory effects mainly through inhibition of

Ž .PG synthesis Vane, 1971 . The mechanisms of

Abbreviations: NSAIDs, nonsteroidal anti-inflammatory drugs;PBMC, peripheral blood mononuclear cells; PG, prostaglandin;NK, natural killer; LAK, lymphokine activated killer; PHA, phy-tohemagglutinin; PWM, pokeweed mitogen; LPS, lipopolysaccha-

w Ž . xride; M-5011, d-2- 4- 3-methyl-2-thienyl phenyl propionic acid;Ig, immunoglobulin; LU, lytic unit; IFN-g, interferon-gamma; IL,interleukin; RT–PCR, reverse transcriptase–polymerase chain re-action; CM, culture medium

) Corresponding author. Tel.: q81-749-23-5545; Fax: q81-749-22-5859

anti-inflammatory effects are not fully explained onlyŽby the inhibition of PG synthesis Abramson and

Weissmann, 1989; Flescher et al., 1991; Hanif et al.,.1996 . Previously, we have shown that indomethacin

regulated the production and expression of cytokinesŽ .such as tumor necrosis factor TNF , IFN-g, IL-2,

ŽIL-4, IL-6 on stimulated mononuclear cells Tsuboi.et al., 1995 . PGE also inhibited the production of2

Th1 cytokines: IFN-g and IL-2, whereas PGE had2

no effect on production of Th2 cytokines: IL-4 andŽIL-5 Novak and Rothenberg, 1990; Betz and Fox,.1991 . The relationship between NSAIDs and PGE2

in immune responses needs to be more explored.wRecently, we developed a new compound, d-2- 4-

Ž . x Ž3-methyl-2-thienyl phenyl propionic acid termed as

0162-3109r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved.Ž .PII: S0162-3109 98 00046-0

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217210

.M-5011 . We reported that the effects of M-5011 oncytokine production was significantly stronger thanthat of indomethacin, whereas it was opposite on

Ž .PGE production Kanemoto et al., 1998 . Because2

of M-5011 inhibited less PG synthesis, M-5011 haveanti-nociceptive, anti-pyretic and anti-inflammatoryeffects, yet be substantially less ulcerogenic than a

Žrange of indomethacin Murakami et al., 1996; Kido.et al., 1998 . Therefore, this study has investigated

the effects of the two NSAIDs, indomethacin andM-5011, on lymphocyte proliferation, cytotoxicityactivity, Ig production and cytokine expressions in

Ž .human peripheral blood mononuclear cells PBMC .

2. Materials and method

2.1. Culture medium

ŽCM was RPMI-1640 medium Nikken Bio Medi-.cal Laboratory supplemented with 10% heat-in-

Ž .activated fetal calf serum Gibco Laboratory , 10mM HEPES, 100 Urml penicillin, 100 mgrmlstreptomycin and 50 mM 2-mercaptoethanol.

2.2. Drugs

Indomethacin was purchased from Sigma and M-Ž5011 originated from Maruho Co. Ltd. Murakami et

.al., 1996 . These drugs were dissolved in dimethylŽ .sulfoxide Nacalai Tesque and diluted to working

concentrations in CM.

2.3. PBMC preparation and cell culture

PBMC from healthy adult donors were isolatedfrom heparinized venous blood on Ficoll–PaqueŽ .Pharmacia LKB Biotechnology density gradientcentrifugation. The cells were recovered from inter-face and washed in PBS buffer. Then the cells wereresuspended with CM.

Ž 6To measure IL-2 production, PBMC 1=10.cellsrml were distributed into wells of 48-well flat

bottom culture plates and stimulated with 10 mgrmlŽ .of PHA Difco Laboratory in the absence or pres-

Ž .ence of test compound for 24 h 378C, 5% CO . To2

measure PGE synthesis and IL-6 production, PBMC2Ž 6 .1=10 cellsrml were distributed into wells of

Ž .48-well flat bottom culture plates Corning andŽ .stimulated with 50 ngrml of LPS Difco Laboratory

in the absence or presence of test compound for 24 hŽ .378C, 5% CO . The evaluation of IL-6 or IL-22

Ž 6 .mRNA on PBMC 1=10 cellsrml carried outincubating for 1 or 12 h, respectively.

Ž 6 .PBMC 1=10 cellsrml were also cultured inŽ .wells of 24-well plate Corning for 3 days with 100

Ž .Urml of rIL-2 Genzyme and various concentra-tions of NSAIDs. Then, the cells as LAK cells andthe supernatants separated by centrifugation, thesecells were used.

Ž 6To measure Ig production, PBMC 1 = 10.cellsrml were incubated with 12.5 mgrml of PWM

Ž .Sigma for 7 days in wells of 48-well plate. Thecells were removed by centrifugation, and the cell-free supernatants were collected.

The supernatants describing above were separatedby centrifugation from CM, and were stored aty808C until the cytokine production was analyzed.

2.4. PGE and cytokine measurements2

PGE and IFN-g were assayed by ELISA kit2Ž .Amersham and Otsuka Pharmaceutical , respec-tively, according to the manufacturer’s instructions.The lower limit of sensitivity in the PGE and IFN-g2

were 0.2 ngrml and 15.6 pgrml, respectively.IL-2 and IL-6 bioactivity in PBMC supernatants

were measured by a cell proliferation assay using theŽIL-2 dependent T-cell line, CTLL-2 Gillis et al.,

.1978 , and the IL-6 dependent hybridoma, MH60PŽ . 4BSF-2 Matsuda et al., 1988 . Very briefly, 1=10

CTLL-2 and MH60PBSF-2 cells were plated intoeach wells of 96-well microtiter plate contain CM. Adiluted PBMC supernatant was added. Cells were

Ž .incubated for 2 days 378C, 5% CO . The following2Ž . Ž . Žmorning 2- 4-lodophenyl -3- 4-nitrophenyl -5- 2,4-.disulfophenyl -2 H-tetrazolium, monosodium salt

Ž . ŽWST-1: Wako for IL-2 bioactivity and 3- 4,5-di-.methylthiazol-2-yl -2,5-diphenyl tetrasolium bromide

Ž .MTT: Chemicon International for IL-6 bioactivitywere added, and determined at 450r650 or 570r650

Žnm, respectively, on a microplate reader Molecular.Devices . These cytokines bioassay in PBMC super-

natants estimated by references to the standard curvethat was developed using recombinant human IL-2Ž . Ž .Genzyme and IL-6 Wako , and the lower limit of

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217 211

sensitivity in the IL-2 and IL-6 were 0.195 and 0.004Urml, respectively. Each sample was assayed induplicate.

2.5. Expression of the cytokine mRNA

The cytokine mRNA expression on lymphocyteswas examined by RT–PCR as described previouslyŽ .Becker et al., 1991 . Total cellular RNA was iso-

Žlated from the PBMC using the RNAzole B Bio-.tecx Laboratories according to the manufacturer’s

instructions. For semi-quantitative analysis of rela-tive levels of cytokine mRNA, we prepared for

Ž .titration a half dilution of the reverse transcribedcDNA. Briefly, 21 ml of cDNA was prepared from 2mg of total RNA in PBMC using reverse transcrip-

Ž .tase Life Technologies . A total of 1 ml of eachdiluted cDNA was amplified in a final volume of 20

Žml containing 1= PCR buffer 1 mM Tris–HCl, pH7.5, 30 mM KCl, 0.1 mM dithiothreitol, 10 mM

.EDTA, 50 mgrml BSA, 0.5% glycerol , 200 mMdNTP, 1.5 mM MgCl , 0.2 mM each of the appro-2

priate sense and antisense primers, and 1 U TaqŽ .DNA polymerase Promega . The mixture was am-

Žplified in a GeneAmp PCR System 9600 Perkin-.Elmer Cetus . The sequences for specific primers for

IL-2, IL-6 and b-actin were described previouslyŽ .Tsuboi et al., 1995 . These primer sets yield PCRproducts of 457, 627 and 838 bp for IL-2, IL-6 andb-actin, respectively. The first PCR cycle consistedof denaturing at 948C for 2 min, followed by denatu-ration at 948C for 15 s and primer annealing andextension at 608C for 30 s. The PCR reaction con-sisted of 30 cycles for IL-2, 26 cycles for IL-6 and23 cycles for b-actin. Finally, the last cycle was at728C for 7 min.

2.6. Cell proliferation assay

Ž 6 .PBMC 1=10 cellsrml were distributed intowells of 96-well microtiter plates with flat-bottomed

Ž .wells Corning in a total volume 200 ml at theŽ .triplicate culture with PHA, anti-CD3 mAb NU-T3

Ž .or rIL-2 Genzyme in the absence or presence of thevarious concentrations of NSAIDs. The cells were

w3 xharvested after 3 days of incubation with H -Ž .thymidine DupontrNEN Research Products pulsed

Ž .at 37 kBqrwell 1 mCirwell during the last 4 h of

incubation. Then the cells were harvested on theŽgrass filter by 96-well harvester M&S Instruments

.Trading , and each grass filter was dissolved byŽ . w3 xAquasol-2 Dupont . H -thymidine incorporation

into the cells was measured by liquid scintillationŽ .analyzer Aloka .

2.7. Target cells and cytotoxic assay

The NK-sensitive human erythroleukemia cellline, K562 cells were used as targets for cytotoxicanalysis. NSAIDs were added while measuring NKactivity of PBMC and inducing LAK cells fromPBMC. NK and LAK cytotoxicity was measured by

51 Ž 6a Cr-release assay. Briefly, K562 cells 5=10. Ž .cells were labeled with 9.25 MBq 250 mCi of

51 Ž .sodium Cr chromate Amersham International for1 h at 378C and washed three times before use. K562

Ž 4 .target cells 1=10 cellsrwell were mixed withvarious numbers of effector cells: PBMC as NK cellsand PBMC cultured with IL-2 as LAK cells, inround-bottom microtiter plates containing 200 ml ofCM for 4 h. The cytotoxicity activity was defined by

Ž .the following formula: % cytotoxicity LU swŽ . Žexperimental releaseyspontaneous release r max-

.ximal releaseyspontaneous release =100. Experi-mental release was defined as the counts of 51Crreleased from target cells caused by effector cellsinduced with the test compounds. Maximal releasewas the 51Cr release from target cells lysed by 2%NP-40 or 10% Triton-100 detergent. Spontaneousrelease was the counts of 51Cr from target cellsincubated with CM alone. The results were ex-pressed at LU r106 effector cells. The LU was the30

number of effector cells required to cause 30% lysisof target cells estimated by the effectors to target

Ž .ratio Ibayashi et al., 1987 .

2.8. Measurements of immunoglobulin

The Ig productions were measured using a modi-Ž .fied ELISA Shingu et al., 1992 . In briefly, ELISA

Ž .microtiter plates flat-bottom were coated overnightŽ X.at 48C with goat F ab 2 fragment to human IgG

Ž X. Ž X. Ž .F ab 2 or IgM F ab 2 Cappel . The plates werewashed three times with 300 mlrwell of PBS con-

Ž .taining 0.05% Tween 20 PBS–Tween . A dilutedPBMC supernatants with CM was added and incu-

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217212

Fig. 1. Effects of NSAIDs on PGE synthesis in LPS-stimulated2

PBMC. PBMC were isolated from three healthy donor wereincubated with LPS and indomethacin or M-5011 for 24 h. PGE2

levels were measured by ELISA. Each point represents the meansŽ ."S.D. ns3 . The control was added DMSO as solvent instead

of NSAIDs. The statistical significance were tested by the multi-ple comparison, Scheffe. )) p-0.01 vs. indomethacin.

bated for 1 h at room temperature. The plates werewashed three times with PBS–Tween, then peroxi-

Ž .dase-mouse anti-human IgG or IgM Zymed inPBS–Tween was incubated for 1 h. The plates were

Table 1Effects of M-5011 on cytokine production in stimulated PBMC

Ž .Concentration Cytokine production UrmlŽ .mM IL-2 IL-6

0 3.6"1.0 78.5"8.11 5.1"1.8 66.2"5.9

10 7.5"0.6) 51.6"0.8))

100 9.2"0.2)) 58.3"3.9)

PBMC were isolated from healthy donor. To IL-2 analysis, thePBMC were cultured with M-5011 for 24 h in the presence of10mgrml PHA. The measurements of IL-2 levels in the super-natants carried out by bioassay using CTLL-2 cells. To IL-6analysis, the PBMC were cultured with M-5011 for 24 h in thepresence of 50 ngrml LPS. The measurements of IL-6 levels inthe supernatants carried out by bioassay using MH-60PBSF-2cells. Each value represents the means"S.D. of three samples.The control was added DMSO as solvent instead of M-5011.) p-0.05 and )) p-0.01 vs. control.

Fig. 2. Effects of M-5011 on mRNA expression of cytokines. Foranalysis of IL-2 mRNA level, total RNA were extracted fromPBMC stimulated with 10 mgrml PHA in the presence of 50 mM

Ž .of M-5011 after 12 h A . For analysis of IL-6 mRNA level, totalRNA extracted from PBMC stimulated with 50 ngrml LPS in the

Ž .presence of 50 mM of M-5011 after 1 h B .

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217 213

washed three times with PBS–Tween again, and 100X X Ž .mgrml 3,3 ,5,5 -tetramethylbenbenzidine Wako —

Ž . Ž .0.003% H O Wako added 100 mlrwell . After2 2

10 min, the reaction of each well stopped with 100Ž .ml of 5 N H SO Nacalai Tesque , the optical2 4

density was determined at 450r650 nm on a mi-Ž .croplate reader Molecular Devices . The Ig content

in PBMC supernatants was estimated by referencesto the standard curve that was developed by using

Ž .purified human IgG or IgM Cappel .

2.9. Statistic analysis

Data are expressed as the mean"S.D. and statis-tically evaluated using a multiple range test ofScheffe.

3. Results

3.1. Effect of NSAIDs on PGE synthesis2

PBMC were incubated with LPS and NSAIDs for24 h followed by harvesting the supernatant to mea-sure the PGE level. In the absence of NSAIDs,2

LPS-stimulated PBMC produced 2895 pgrml ofPGE . The PGE synthesis was inhibited in dose-de-2 2

pendent manner, when PBMC were cultured in theŽ .presence of either indomethacin or M-5011 Fig. 1 .

The value of IC for inhibition of PGE synthesis50 2

by indomethacin was about 30-fold lower than M-Ž5011 on LPS-stimulated PBMC indomethacin: 0.039

.mM vs. M-5011: 1.3 mM .

Table 2w3 xEffects of NSAIDs on H -thymidine incorporation into PBMC stimulated with optimum- or sub-optimum-concentrations of PHA,

anti-CD3 mAb or IL-2

Stimulant Concentration of Indomethacin M-5011Ž .NSAIDs mM

Ž .PHA 10 mgrml 0 52 299"6041 51 117"321 54 377"2333

aa10 61 807"514)) 67 100"2083))

a100 69 214"1197)) 72 264"933))

Ž .PHA 1 mgrml 0 2081"921 3045"71 3143"105)

10 4045"160)) 4185"98))

100 5115"117)) 5429"100))

Ž .Anti-CD3 1 mgrml 0 15 742"7921 18 473"406)) 19 333"690))

10 20 866"228)) 21 670"571100 22 690"342)) 23 582"1029))

Ž .Anti-CD3 0.1 mgrml 0 8131"2011 9041"114)) 10 473"516

10 12 401"714)) 13 187"661))

100 15 122"293)) 15 984"589))

Ž .IL-2 1000 Urml 0 8260"1281 9095"24) 9352"224)

10 11 050"197)) 11 783"377))

100 13 010"434)) 13 453"435))

Ž .IL-2 100 Urml 0 2762"1151 3077"216 3322"171)

10 4313"182)) 4569"140))

100 6309"115)) 6563"196))

PBMC were isolated from healthy donor. The PBMC were incubated with indicating stimuli and M-5011 or indomethacin for 3 days. Each5 Ž .value of tritium–thymidine incorporation represents the means"S.D. at dpm into 1=10 cellsrwell PBMC during last 4 h ns3 . The

control was added DMSO as solvent instead of NSAIDs. The statistical significance were tested by the multiple comparison, sheffe.´) p-0.05 and )) p-0.01 vs. control. ap-0.05 and aap-0.01 vs. indomethacin.

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217214

3.2. Effects of NSAIDs on production of IL-2 andIL-6

We investigated the effects of NSAIDs on thecytokines production in PBMC. M-5011 up-regu-lated IL-2 production in a dose-dependent manner on

Ž .PHA-stimulated PBMC Table 1 . On the other hand,M-5011 down-regulated IL-6 production on LPS-

Ž .stimulated PBMC Table 1 . The effects of indo-methacin was similar to that of M-5011, although it

Ž .was weaker than that of M-5011 data not shown .To explore further the mechanisms by which M-5011predominantly regulates cytokines production, PBMCwere stimulated with PHA for 12 h to investigateIL-2 expression or with LPS for 1 h to investigateIL-6 expression in the presence or absence of M-5011. The treatment of cells with M-5011 results inapproximately four-fold increase of IL-2 mRNA ex-

Ž .pression levels Fig. 2A , whereas it induced approx-imately two- to four-fold decrease of IL-6 mRNA

Ž .expression levels Fig. 2B . The expression of b-actinmRNA was not affected by M-5011. These datademonstrated that M-5011 could regulate the cy-tokine production both at the protein and mRNAlevels.

3.3. Effect of NSAIDs on lymphocyte proliferation

To assess the effects of NSAIDs on lymphocyteproliferation, PBMC were stimulated with PHA inthe absence or presence of different concentrations

Ž .of NSAIDs Table 2 . Both indomethacin and M-Ž .5011 effectively augmented the PHA 10 mgrml -

induced lymphocyte proliferation in a dose-depen-dent manner. The dose-dependent augmentation oflymphocyte proliferation by NSAIDs was also ob-

Žserved with a sub-optimal concentration of PHA 1. Ž .mgrml and the significant augmentation p-0.05

Žwas observed with at least 1 mM of NSAIDs Table.2 . The effect of M-5011 was stronger than that of

Žindomethacin p-0.01 at 1 mM, p-0.05 at 10.mM . PBMC were also stimulated with anti-CD3 or

Ž .IL-2. NSAIDs significantly p- at least 0.05 aug-mented the lymphocyte proliferation in response toanti-CD3 or IL-2 in a dose-dependent mannerŽ .Table 2 .

Fig. 3. Effects of NSAIDs on cytotoxicity in LAK cell activity toŽ 6 .K562 cells. The cytotoxic activities LU 30%r10 cells were

estimated by 4 h 51Cr-release assay during 4 h. LAK cells wereinduced from PBMC incubated with 100 Urml IL-2 for 3 days.Ž . Ž .A Effects of 50 mM of NSAIDs. B Dose-dependency of

Ž .M-5011. These figures represent the means"S.D. ns3 . ) p-

0.05 and )) p-0.01 vs. control.

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217 215

3.4. Effects of NSAIDs on cytotoxicity

We studied the effects of NSAIDs on NK activityand on IL-2-induced LAK activity. NK activity was

Žaugmented by addition of M-5011 5.6"0.6, 8.8"

1.0, or 10.4"0.4 LU r106 effector cells without,30.with 1 or 10 mM of M-5011, respectively . NKŽactivity was also augmented by indomethacin 5.6"

Ž .Fig. 4. Effects of NSAIDs on the production of IgG A and IgMŽ .B by PWM-stimulated PBMC for 7 days. PBMC were incubatedwith 12.5 mgrml of PWM in the presence or absence of M-5011for 7 days. The supernatants were collected, the Ig levels in theculture supernatants of PWM-stimulated PBMC were measured

Ž .by ELISA. Each point represents the means"S.D. ns3 . )) p-0.01 vs. control. ap-0.05 and aap-0.01 vs. indomethacin.

0.6, 8.4"0.3, or 8.3"1.0 LU r106 effector cells30

without, with 1 or 10 mM of indomethacin, respec-.tively . The augmentation was statistically significant

Ž .in all cases p-0.01 . The treatment of PBMC with50 mM of NSAIDs and 100 Urml of IL-2 resultedin an enhancement of LAK activity as compared to

Ž .IL-2 alone Fig. 3A . This enhancement on the in-duction of LAK activity by M-5011 was dependent

Ž .on doses used for the study Fig. 3B .Because generation of LAK cell with IL-2 may

involve a collaboration between IL-2 and IFN-gŽ .Ellis et al., 1989 , we investigated whether NSAIDsincrease IFN-g production by LAK cells. PBMCwere cultured with IL-2 for 3 days in the absence orpresence of M-5011 followed by harvesting culturesupernatants for measurement of IFN-g. M-5011 in-

Žcreased IFN-g productions by LAK cells 24.7"1.6,75.0"2.9, 57.6"3.3, or 142.2"13.3 pgrml of

Ž .IFN-g without, with 1 mM p-0.01 , 10 mMŽ . Ž .p-0.01 or 100 mM p-0.01 of M-5011, re-

.spectively . The similar augmentation was also ob-Ž .served by indomethacin data not shown .

3.5. Effects of NSAIDs on productions of IgG andIgM

The effects of NSAIDs on the production of IgGand IgM by PBMC stimulated with 12.5 mgrmlPWM for 7 days were investigated. Both M-5011and indomethacin suppressed IgG and IgM secretionin a dose-dependent manner with more potent sup-

Ž .pression by indomethacin Fig. 4 .

4. Discussion

We studied in this manuscript the effects ofNSAIDs on immune responses using indomethacinand M-5011. The cell-mediated immunity involvesthe destruction of infected cells by NK cells, cyto-toxic T-cells and activated macrophages. Th1 cells

Ž .and cytokines TNF, IL-2 and IFN-g produced byTh1 cells are essential for activation of cell-mediatedimmunity. On the other hand, the humoral immunitylargely depends upon the production of antibody byB-cells activated by Th2 cells producing IL-4, IL-5and IL-6.

We previously showed that indomethacin up-regu-lated production of Th1 type’s cytokine: IL-2, IFN-g

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217216

and TNF, whereas down-regulated production of Th2Žtypes: IL-4 and IL-6, respectively Tsuboi et al.,

.1995 . We demonstrated that M-5011 also up-regu-lated IL-2 and IFN-g productions, whereas down-regulated IL-6 production by PBMC. The effects ofNSAIDs to inhibit the IL-6 production was also

Ž .observed in PHA-stimulated PBMC data not shown .Thus these effects of NSAIDs on cytokines produc-tion may not be related to affecting cells. PGE2

inhibited the production of Th1 cytokines: IFN-gand IL-2, whereas PGE had no effect on production2

Žof Th2 cytokines: IL-4 and IL-5 Novak and Rothen-.berg, 1990; Betz and Fox, 1991 . The effect of

indomethacin on PG synthesis were more potent thanthose of M-5011, whereas the effects of M-5011 oncytokine production were more potent than those ofindomethacin. These results suggest that the regula-tory effects on cytokine productions of NSAIDs arenot always dependent on the inhibition of PGE2

synthesis.Two NSAIDs, indomethacin and M-5011 aug-

mented the lymphocyte proliferation in a similardose-dependent manner. PGE suppresses T-cell pro-2

Žliferation and IL-2 production Chouaib and Fradelizi,.1982; Rappaport and Dodge, 1982; Brune, 1988 ,

Žand inhibit IL-2 receptor expression Vercammen.and Ceuppens, 1987 . The effect of indomethacin on

PG synthesis were more potent than those of M-5011,whereas M-5011 have the similar effects comparewith indomethacin on lymphocyte proliferation.These effect of NSAIDs may be not always depen-dent on the inhibition of PGE synthesis.2

M-5011 indicated the potent enhancement of cyto-toxicity on NK and LAK cells. It is known that thesecytotoxic activity were enhanced by IL-2 and IFN-g.M-5011 up-regulated the these cytokines on stimu-

Žlated PBMC. While PGE impairs NK activity Goto2.et al., 1983 , M-5011 that inhibited the PGE syn-2

thesis enhanced the cytotoxicity. We suggested thatthe enhanced NK and LAK cell activity by NSAIDscould be in part due to increased IL-2 and IFN-gproduction.

The two NSAIDs also suppressed the Ig produc-tions by PWM-stimulated PBMC. These effects ofNSAIDs on Ig production were similar to the effectson PGE synthesis. Namely, indomethacin was more2

potent than M-5011 in inhibition of Ig production.The B-cell activity and Ig production were impaired

by PG, and proliferation of some B-lymphomas wereŽinhibited by PGE Phipps et al., 1989; Roper et al.,

.1994 . While PG inhibited the Ig production, NSAIDssimultaneously inhibited both PGE synthesis and Ig2

production. These results may indicate that the ef-fects of Ig regulation on NSAIDs are PG-indepen-dent. We reported that NSAIDs had the potency toup-regulate IL-2 production and down-regulate IL-6

Ž .production on T-cell subsets Tsuboi et al., 1995 .IL-2 stimulate certain subpopulations of B-cells to

Ž .produce antibody Agematsu et al., 1995 . On thecontrary, IL-6 is a key Th2-type cytokine for B-cell

Žactivation to exert humoral immunity Muraguchi et.al., 1988 . NSAIDs may inhibit Ig productions by

mediating cytokine productions.Two NSAIDs, indomethacin and M-5011, have

different potencies to inhibit PGE synthesis. How-2

ever these NSAIDs enhanced the cell-mediated re-sponses such as lymphocyte proliferation and cyto-toxic activity in the similar dose-dependent manner.Recently, we reported that down-regulation of IL-6production by NSAIDs may be partly independent of

Žinhibition of PGE production Kanemoto et al.,2.1998 . These results suggest that the NSAIDs-media-

ted immune regulation is not entirely through theirability to inhibit PGE synthesis, but rather is through2

their direct effects on immune competent cells. Thusthese matter may contribute the facts that M-5011had more potent anti-inflammatory effects with lowulcerogenic activity compared with classicalNSAIDs.

Acknowledgements

I wish to express my thanks to Dr. K. ItohŽKurume University School of Medicine, Fukuoka,

.Japan for numerous helpful suggestions during thepreparation of this work, Ms. F. Maruyama and K.

Ž .Imamura Maruho Co. Ltd., Hikone, Japan for theirhelp in carrying out this study.

References

Abramson, S.B., Weissmann, G., 1989. The mechanisms of actionof nonsteroidal anti-inflammatory drugs. Arthritis Rheum. 32,1–9.

Agematsu, K., Kobota, T., Yang, F.C., Nakazawa, T., Fukushima,

( )K. Tanaka et al.r Immunopharmacology 40 1998 209–217 217

K., Kitahara, M., Mori, T., Sugita, K., Morimoto, C.,Komiyama, A., 1995. CD27rCD70 interaction directly drivesB-cell IgG and IgM synthesis. Eur. J. Immunol. 25, 2825–2829.

ŽBecker, S., Quay, J., Soukup, J., 1991. Cytokine tumor necrosis.factor, IL-6, and IL-8 production by respiratory syncytial

virus-infected human alveolar macrophages. J. Immunol. 147,4307–4312.

Betz, M., Fox, B.S., 1991. Prostaglandin E inhibits production of2

Th1 lymphokines but not of Th2 lymphokines. J. Immunol.146, 108–113.

Brune, K., 1988. Nonsteroidal anti-inflammatory drugs: Effects onŽ .lymphocyte function. In: Bray, M.A., Morely, J. Eds. , The

Pharmacology of Lymphocytes. Springer-Verlag, New York,NY, p. 577.

Chouaib, S., Fradelizi, D., 1982. The mechanism of inhibition ofhuman IL-2 production. J. Immunol. 129, 2463–2468.

Ellis, T.M., McKenzie, R.S., Simms, P.E., Helfrich, B.A., 1989.Induction of human lymphokine-activated killer cells by IFN-aand IFN-g. J. Immunol. 143, 4282–4286.

Flescher, E., Fossum, D., Gray, P.J., Fernandes, G., Harper,M.J.K., Talal, N., 1991. Aspirin-like drugs prime humanT-cells: modulation of intracellular calcium concentrations. J.Immunol. 146, 2553–2559.

Gillis, S., Ferm, M.M., Ou, W., Smith, K.A., 1978. T cell growthfactor: parameters of production and a quantitative microassayfor activity. J. Immunol. 120, 2027.

Goto, T., Herberman, R.B., Maluish, A., Strong, D.M., 1983.Cyclic AMP as a mediator of prostaglandin E-induced sup-pression of human natural killer cell activity. J. Immunol. 130,1350–1355.

Hanif, R., Pittas, A., Feng, Y., Koutsos, M.I., Qiao, L., Staiano-Coico, L., Shiff, S.I., Rigas, B., 1996. Effects of nonsteroidalanti-inflammatory drugs on proliferation and on induction ofapoptosis in colon cancer cells by a prostaglandin-independentpathway. Biochem. Pharmacol. 52, 237–245.

Ibayashi, Y., Dave, S., Hoon, B., Golub, S.H., 1987. The regula-tory effect of adherent cells on lymphokine activated killercells. Cell. Immunol. 110, 365–378.

Kanemoto, Y., Tanaka, H., Tanaka, K., Imamura, K., Maruyama,F., Tsuboi, I., 1998. Effects of NSAIDs on interleukin-6production. Jpn. J. Inflamm. 18, 11–17.

Kido, H., Murakami, N., Ito, A., Kimura, K., Kodera, N., Doi, T.,

Naruse, T., 1998. Anti-inflammatory, analgesic and anti-pyreticw Ž . xeffects of d-2- 4- 3-methyl-2-thienyl phenyl propionic acid

Ž .M-5011 , a new nonsteroidal anti-inflammatory drug, in ratsand guinea pigs. Jpn. J. Pharmacol. 76, 75–86.

Matsuda, T., Hirano, T., Kishimoto, T., 1988. Establishment of anŽ .interleukin-6 IL-6 rB-cell stimulatory factor 2-dependent cell

line and preparation of anti-IL-6 monoclonal antibodies. Eur.J. Immunol. 18, 951–956.

Muraguchi, A., Hirano, T., Tang, B., Matsuda, T., Horii, Y.,Nakajima, K., Kishimoto, T., 1988. The essential role of

Ž .B-cell stimulatory factor 2 BSF-2rIL-6 for the terminaldifferentiation of B-cells. J. Exp. Med. 167, 332–344.

Murakami, N., Takakase, H., Tomita, T., Iwata, K., Naruse, T.,1996. Anti-nociceptive activity of a novel nonsteroidal anti-in-

Ž .flammatory drug M-5011 with low ulcerogenic effects inmice. Jpn. J. Pharmacol. 72, 29–37.

Novak, T.J., Rothenberg, E.V., 1990. cAMP inhibits induction ofinterleukin-2 but not of interleukin-4 in T-cells. Proc. Natl.Acad. Sci. USA 87, 9353–9357.

Phipps, R.P., Lee, D., Schad, V., Warner, G.L., 1989. E-seriesprostaglandins are potent growth inhibitors for some B-lymphomas. Eur. J. Immunol. 19, 995–1001.

Rappaport, R.S., Dodge, G.R., 1982. Prostaglandin E inhibits theproduction of human interleukin-2. J. Exp. Med. 155, 943–948.

Roper, R.L., Ludlow, J.W., Phipps, R.P., 1994. Prostaglandin E2

inhibits B-lymphocyte activation by a cAMP-dependent mech-anism: PGE-inducible regulatory proteins. Cell. Immunol. 154,296–308.

Shingu, M., Wada, T., Ezaki, I., Nobunaga, M., 1992. Pharmaco-logic modulation by alminoprofen of immunoglobulin andrheumatoid factor production by PWM-stimulated normal hu-man mononuclear cells. Jpn. J. Inflamm. 12, 327–331.

Tsuboi, I., Tanaka, H., Nakao, M., Shichijo, S., Ito, K., 1995.Nonsteroidal anti-inflammatory drugs differentially regulatecytokine production in human lymphocytes: up-regulation ofTNF, IFN-g and IL-2, in contrast to down-regulation of IL-6production. Cytokine 7, 372–379.

Vane, J.R., 1971. Inhibition of prostaglandin synthesis as a mech-anism of action for aspirin-like drugs. Nature 231, 232–235.

Vercammen, C., Ceuppens, J.L., 1987. Prostaglandin E inhibits2

human T-cell proliferation after crosslinking of the CD3–Ticomplex by directly affecting T-cells at an early step of theactivation process. Cell. Immunol. 104, 24–36.