Inflammation, Vol. 23, No. 1, 1999

INTERLEUKIN-4 AND INTERFERON-yINHIBIT PROSTAGLANDIN PRODUCTION

BY INTERLEUKIN-1B-STIMULATED HUMANPERIODONTAL LIGAMENT FIBROBLASTS

K. NOGUCHI, 1 M. SHITASHIGE, 2 H. WATANABE, 1S. MUROTA, 2 and I. ISHIKAWA1

1Department of Periodontology,2Department of Physiological Chemistry

Graduate School, Faculty of Dentistry, Tokyo Medical and Dental University1-5-45, Yushima, Bunkyo-ku, Tokyo 113-8549, Japan

Abstract—The purpose of the present study was to investigate the involvement ofcyclooxygease-1 (COX-1) and cyclooxygenase-2 (COX-2) in prostaglandin (PG)production by human periodontal ligament (PDL) fibroblasts stimulated with aproinflammatory cytokine, inerleukin-1B (IL-1B), and to examine the effect ofinterleukin-4 (IL-4), a Th2 cytokine, and interferon-y (IFN-y ), a Th1 cytokine, on PGproduction by the cells. IL-1B-stimulated PDL fibroblasts produced prostaglandin E2

(PGE2) in a time-dependent manner. Indomelhacin, a non-selective COX-l/COX-2inhibitor, and NS-398, a selective COX-2 inhibitor, completely inhibited PGE2production by IL-1B-stimulated cells. Northern blot analysis showed that COX-2mRNA was detected in IL-1B-stimulated PDL cells, although not detected inunstimulated cells, while expression of COX-1 mRNA was in the same extent inboth the cells. Dexamethasone inhibited COX-2 mRNA expression, COX activityand PGE2 production in IL-1B-slimulated cells. IL-4 and IFN-y suppressed PGE2production by IL-1B-stimulated PDL fibroblasts, but COX activity enhanced by IL-1Btreatment was significantly inhibited by IL-4, not by IFN-y. Northern blot analysisshowed that IL-4 depressed COX-2 mRNA expression with no effect on COX-1mRNA expression. On the other hand, IFN-y had no effect on expression of COX-1and -2 mRNA. These data suggest that COX-2 is primarily responsible for PGE2production by IL-1B-stimulated human PDL fibroblasts and that IL-4 inhibited PGE2

production by IL-1B-stimulated PDL fibroblasts through down-regulation of COX-2expression, while IFN-y suppressed the PGE2 production with no effect on COX-2expression.

INTRODUCTION

Prostaglandins (PGs) play an important role in regulating diverse cellular func-tions in physiological and pathological conditions (1). Two major steps are

1

0360-3997/99/0200-0001 $16. 00/0 ©1999 Plenum Publishing Corporation

2 Noguchi et al.

essential for PG biosynthesis. One step is the release of arachidonic acidfrom membrane phospholipids by phospholipases and the other is the conver-sion of arachidonic acid to PGH2 by cyclooxygenase (COX), also known asprostaglandin endoperoxide synthase (2). Recently, it has been demonstratedthat there are two isozymes of COX, which are called cyclooxygenase-1 (COX-1)and cyclooxygenase-2 (COX-2). The two isozymes are encoded by differentgenes (3, 4). Whereas COX-1 is constitutively expressed in many tissues (5),COX-2 is induced following the activation of cells by a variety of proinflam-matory agents such as cytokines, mitogens and lipopolysaccharides (LPS) (6-8)and recognized as an immediate-early gene (9).

Many studies have demonstrated that PGs are involved in the pathogene-sis of periodontal disease (10). It has been reported that the amounts of PGE2,prostacyclin and thromboxane in inflamed gingival tissues are enhanced, com-pared with those in normal tissues (11). Nonsteroidal anti-inflammatory drugs(NSAIDs) which inhibit PG synthesis have been shown to be able to preventperiodontal destruction (12, 13). In particular, PGE2 has been shown to be apotent stimulator of bone resorption (14, 15) and to be associated with attach-ment loss (10, 16). Thus, the mechanism of the regulation of PGE2 produc-tion in periodontal tissue needs to be recognized. Cavanaugh et al. (17) havedemonstrated that COX-2 protein is expressed in mononuclear inflammatorycells, endothelial cells, epithelial cells and gingival fibroblasts of inflamed humangingival tissue. In vitro studies have shown that human gingival fibroblasts stim-ulated with interleukin-1B (IL-1B) and LPS produce PGE2 via de novo synthe-sis of COX-2 (18, 19). Periodontal ligament (PDL) fibroblasts derive from toothsupporting tissue and have the characteristics different from gingival fibroblasts,in that PDL fibroblasts have osteoblast-like properties such as high levels ofalkaline phosphatase activity and cAMP enhancement by PTH stimulation (20,21). PDL fibroblasts have been demonstrated to produce PGE2 in response toproinflammatory cytokines such as IL-1 and tumor necrosis factor (TNF) (22,23). It is very likely that PDL fibroblasts are also involved in PGE2 productionin periodontal tissue. However, the mechanism of PG production in PDL cellsstimulated with IL-1B is poorly understood.

Interleukin-4 (IL-4), a Th2 cytokine, is a 20 kDa glycoprotein and has anti-inflammatory actions, because it can inhibit the production of proinflammatorycytokines such as IL-1 and TNF and metalloproteinases by monocytes (24, 25)and can suppress osteoclast formation (26). IL-4 can inhibit PG production viadown-regulation of COX-2 expression in human monocytes and neutrophils (24,27, 28), but it can enhance COX-2 expression and PG production in humanamnion cells (29). Thus, the effect of IL-4 on PG production may be dependenton cell types or stimuli. Interferon-y (IFN-y), a Th1 cytokine, primes humanmonocytes/macrophages to secondary stimuli such as LPS to produce proin-flammatory cytokines (30, 31). But, IFN-7 suppresses PG production by human

COX-2 in IL-1B-Stimulated Human PDL Fibroblasts 3

monocytes/macrophages in response to LPS, IL-1 or Con-A (32, 33). Recently,the involvement of IL-4 and IFN-y in the pathology of periodontal disease hasbeen demonstrated (34-36). The roles of IL-4 and IFN-y, however, are not clear.In the present study, we investigated the involvement of COX-1 and COX-2 inPGE2 production by IL-1B-stimulated human PDL fibroblasts and the effect ofIL-4 and IFN-7 on PGE2 production by the cells.

MATERIALS AND METHODS

Culture of Human PDL Fibroblasts. Human PDL fibroblasts were established from explants ofperiodontal ligaments obtained from premolars or third molars extracted from periodontally healthysubjects. Informed consent was obtained from all the subjects. The cells were grown in a-minimumessential medium (a-MEM) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml)and streptomycin (100 ug/ml) and incubated at 37°C in a humidified atmosphere of 5% CO2 in air.Cells used for the experiments were between the 6th and the 15th passages.

Determination of PGE2 Accumulation. PDL fibroblasts were replated at 5 x 104 cells/ml in96-well plates (200 ul of cell suspension/well) in a-MEM with 10% FBS and cultured until con-fluent. The cells were washed with PBS and rendered quiescent by culturing in a-MEM with 0. 5%FBS for 24 h. Then, the cells were treated with PBS or 2 ng/ml of recombinant IL-1B (Sigma, St.Louis, Missouri) in the presence or absence of indomethacin (Sigma), NS-398 (Cayman Chemical,Ann Arbor, Michigan), dexamethasone (Sigma), IL-4 (R&D system, Minneapolis, Minnesota) orIFN-7 (R&D) for indicated periods. IL-4 and IFN-7 were treated 3 h prior to IL-1B addition. Themedia were collected and stored at -20°C for PGE2 assay.

RNA Preparation and Northern Blot Analysis. After PDL fibroblasts were cultured in 100mm2 plastic plates in a-MEM with 10% FBS until confluent, the cells were rendered quiescentin a-MEM with 0. 5% FBS for 24 h. Then the cell were treated with 2 ng/ml IL-1B in the pres-ence or absence of 1 uM of dexamethasone, 100 ng/ml of IL-4 and 100 ng/ml of IFN-y for 12 h.The medium was removed and total cellular RNA was isolated. In brief, the cells were lysed andRNA was extracted by ISOGEN (Nippon Gene, Toyama, Japan). The RNA was precipated withisopropanol and washed with 80% ethanol. After quantitation at 260 nm, the purified RNA wasdenatured with glyoxal and dimethyl sulfoxide and 10 ug of total RNA was separated on a 1. 0%agrose gel, transferred to nylon membrane filters (Amersham Corp., Arlington Heights, Illinois) andhybridized with [32P]-labeled cDNA probes for COX-1 or COX-2 (Cayman). All the membraneswere reprobed with human B-actin cDNA (Funakoshi, Tokyo, Japan). Filters were prehybridized at60°C for 3 h in a mixture of 0. 5% sodium dodecyl sulfate (SDS), 5 x SSC (1 x SSC = 150 mMsodium chloride and 15 mM sodium citrate), 100 ug/ml denatured salmon sperm DNA and 5 xDenhardt's solution at 60°C. Filters were next hybridized in 0. 5% SDS, 5 x SSC, 0. 01 M EDTA(pH 8. 0), and 5 x Denhardt's solution at 60°C with the probes. After 20 h, the filters were washedtwice in 2 x SSC and once in 0. 5% SDS solution for 2 h at 60°C. The hybridized bands werevisualized with the BAS-2500 Bio-image analyzer (Fuji Photo Film Co., Tokyo, Japan).

PGE2 Assay. The amounts of PGE2 in the conditioned media collected from control or IL-1B-stimulated cells were determined by a commercially available enzyme-linked immunosorbent assay(ELISA) kit (Amersham).

Assay of COX Activity. PDL fibroblasts were rendered quiescent in a-MEM with 0. 5% FBSand then stimulated with 2 ng/ml IL-1B in the presence or absence of 1 uM of dexamethasone, 100ng/ml of IL-4 and 100 ng/ml of IFN-y for 24 h. The media were removed after the incubation and

4 Noguchi et al.

the cells were washed with a-MEM three times. Then the cells were incubated in fresh mediumcontaining 15 uM of arachidonic acid (Sigma) for 15 min to determine COX activity. The levels ofPGE2 in the media were measured by ELISA.

Statistics. Statistical significance was determined by analysis of variance using the StatView4. 0 program on a Macintosh computer. Scheffe's F test was used in the post hoc comparison ofspecific groups.

RESULTS

The effect of IL-1B on PGE2 production by human PDL fibroblasts wasstudied. Figure 1 showed that PDL fibroblasts stimulated with 2 ng/ml of IL-1Bgenerated PGE2 in a time-dependent manner. To investigate whether the PGE2production was dependent on COX-1 or COX-2, the effect of indomethacin,a non-selective COX-l/COX-2 inhibitor, and NS-398, known to be a specificCOX-2 inhibitor (37), on PGE2 production was evaluated. As shown in Figure2, 1 uM of indomethacin completely inhibited PGE2 production by IL-1B-stimu-lated fibroblasts. NS-398 inhibited PGE2 production by the cells in a dose-depen-dent manner and 100 nM of NS-398 completely inhibited PGE2 production.Expression of mRNA of COX-1 and COX-2 in IL-l/3-stimulated fibroblasts was

Fig. 1. Time course of the effect of IL-1B on PGE2 production by PDL fibroblasts. After PDLfibroblasts were rendered quiescent in a-MEM with 0. 5% FBS, the cells were stimulated with2 ng/ml of IL-1B or PBS for indicated periods. The levels of PGE2 in the culture media weremeasured by enzyme-linked immunosorbent assay and the data are depicted as the mean ± SD (N= 4). The data are representative of three separate experiments.

COX-2 in IL-1B-Stimulated Human PDL Fibroblasts

Fig. 2. Effect of indomethacin and NS-398 on PGE2 production by IL-1B-stimulated PDL fibro-blasts. PDL fibroblasts were stimulated with PBS or 2 ng/ml IL-1B in the presence or absence of 1uM of indomethacin or various doses of NS-398 for 24 h. The levels of PGE2 in the culture mediawere determined by enzyme-linked immunosorbent assay. The data are shown as the mean ± SD (N= 4). The data are representative of three separate experiments. *: significantly different from IL-1B(P < 0. 005) **: significantly different from IL-1B (P < 0. 0001).

studied by Northern blot analysis. Expression of COX-1 mRNA was detectedat almost the same extent in unstimulated cells and in IL-1B-stimulated cells,12 h after stimulation. As regard to COX-2 mRNA, its expression was detectedin IL-1B-stimulated cells, but could not be detected in unstimulated cell (Figure3A). Furthermore, dexamethasone inhibited COX-2 mRNA expression, withoutaffecting COX-1 expression, and suppressed COX activity and PGE2 productionin IL-1B-stimulated fibroblasts (Figure 3A, 3B and 3C). From these data, it wasdemonstrated that PGE2 production by IL-1B-stimulated PDL fibroblasts wasprimarily due to COX-2, which was de novo synthesized.

The effect of IL-4 and IFN-y on PGE2 production by IL-1B-stimulated PDLfibroblasts was investigated. IL-4 alone had no effect on PGE2 production, but,as shown in Figure 4A, IL-4 inhibited PGE2 production by IL-1B-stimulatedPDL fibroblasts in a dose-dependent manner. 100 ng/ml of IFN-y significantlyinhibited PGE2 production by IL-1B-stimulated PDL cells (Figure 4B). To exam-ine the mechanism of the inhibitory effect of IL-4 and IFN-y, the effect of IL-4and IFN-y on COX activity in IL-1B-stimulated PDL fibroblasts was examined.COX activity enhanced by treatment with 2 ng/ml of IL-1B was depressed byIL-4 treatment, but IFN-y had no significant effect on COX activity (Figure 5).Furthermore, we evaluated the effect of IFN-y and IL-4 on expression of COX-1 and -2 mRNA. As shown in Figure 6, COX-2 mRNA expression induced by

5

Noguchi et al.

Fig. 3. Effect of dexamethasone of expression of COX-1 and 2- mRNA (A), PGE2 production (B)and COX activity (C) in IL-1B-stimulated PDL fibroblasts. PDL fibroblasts were rendered quiescentin a-MEM with 0. 5% FBS and then stimulated with 2 ng/ml IL-1B in the presence or absence of1 uM of dexamethasone. For analysis of expression of COX-1 and -2 mRNA (A), the total RNAwas extracted 12 h after IL-1B stimulation and Northern blotting was performed as described inthe Materials and Methods. For measurement of PGE2 production (B), the culture media were col-lected 24 h after IL-1B stimulation and the levels of PGE2 in the culture media were determinedby enzyme-linked immunosorbent assay. For measurement of COX activity (C), the cells 24 h afterIL-1B stimulation were washed, followed by addition of 15 uM of arachidonic acid and after 15 minincubation, the levels of PGE2 in the culture media were determined by enzyme-linked immunosor-bent assay. The data are shown as the mean ± SD (N = 4). *: significantly different from IL-1B(P < 0. 0001). The data are representative of three separate experiments.

6

COX-2 in IL-1B-Stiraulated Human PDL Fibroblasts

Fig. 4. Effect of IL-4 (A) and IFN-y (B) on PGE2 production by IL-1B-stimulated PDL fibroblasts.PDL fibroblasts were stimulated with PBS or 2 ng/ml IL-1B in the presence or absence of variousdoses of IL-4 (A) or 100 ng/ml of IFN-y (B) for 24 h. The levels of PGE2 in the culture mediawere determined by enzyme-linked immunosorbent assay. The data are shown as the mean ± SD(N = 4). The data are representative of three separate experiments. *: significantly different fromIL-1B (P < 0. 005). **: significantly different from IL-1B (P < 0. 0001).

7

Noguchi et al.

Fig. 5. Effect of IL-4 and IFN-y on COX activity in IL-1B-stimulated PDL fibro-blasts. PDL fibroblasts were stimulated with PBS or 2 ng/ml IL-1B in the presence or absenceof 100 ng/ml of IL-4 or 100 ng/ml of IFN-y for 24 h. The cells were washed and, then, 15 uM ofarachidonic acid was added. After 15 min incubation, the levels of PGE2 in the culture media weredetermined by enzyme-linked immunosorbent assay. The data are shown as the mean ± SD (N =4). The data are representative of three separate experiments. *: significantly different from IL-1B(P< 0. 0001).

IL-1B was diminished by 100 ng/ml of IL-4, while 100 ng/ml of IFN-y couldnot depress COX-2 mRNA expression. Neither IL-4 nor IFN-y affected COX-1mRNA expression.

DISCUSSION

In the present study, we demonstrated that IL-1B-stimulated human PDLfibroblasts produced PGE2 through COX-2, but not by COX-1. IL-1B-stimulatedPDL fibroblasts produced PGE2 in a time-dependent manner, as shown in Figure1. Recent studies have shown that several different types of human cells such asmonocytes, endothelial cells and skin, lung and gingival fibroblasts induce COX-2 expression to produce PGs in response to IL-1B (6, 7, 18, 38, 39). NS-398, aselective inhibitor of COX-2, depressed PGE2 production in a dose-dependentmanner, as shown in Figure 2. COX-2 mRNA was detected in IL-1B-stimulatedPDL fibroblasts, while it was not detected in unstimulated cells (Figure 3A).There was no difference of COX-1 mRNA expression between in unstimulatedcells and in IL-1B-stimulated cells. Dexamethasone has been demonstrated to

8

Fig. 6. Effect of IL-4 and IFN-y on expression of mRNA of COX-1 and -2 in IL-1B-stimulatedPDL fibroblasts. PDL fibroblasts were stimulated with PBS or 2 ng/ml IL-1B in the presence orabsence of 100 ng/ml of IL-4 or 100 ng/ml of IFN-y for 12 h. The total RNA was extracted andNorthern blot analysis was performed as described in the Materials and Methods.

suppress COX-2 expression at transcriptional and/or translational Levels withoutaffecting COX-1 expression (40-42). Treatment of IL-1B-stimulated PDL fibro-blasts with dexamethasone depressed COX-2 mRNA expression, with no effecton COX-1 mRNA expression, and inhibited COX activity and PGE2 production(Figure 3A, 3B and 3C). These results demonstrated that COX-2 was induced inresponse to IL-1B in human PDL fibroblasts and that the induced COX-2 was pri-marily involved in PGE2 production. It has been suggested that in murine fibro-blasts COX-1 cannot metabolize arachidonic acid released in response to mito-gen stimulation, but can convert exogenous arachidonic acid to PGs (43). More-over, it has been demonstrated that subcellular compartmentalization of COX-1and COX-2 is different in murine 3T3 cells and human and bovine endothelialcells (44), and the possibility has been indicated that COX-1 and COX-2 func-tion independently by coupling to different stimuli-initiated pathways to PGD2

generation from endogenously derived arachidonic acid in mouse bone marrow-derived mast cells (45). Recently, two distinct pathways of PG production inmurine 3T3 cells have been suggested; an intracellular pathway to utilize arachi-donate released by activated intracellular phospholipase(s), which is metabo-lized only by COX-2, and a transcellular pathway to utilize arachidonate mobi-lized by 14 kDa secretory phospholipase A2, which is available to COX-1 (46).Thus, it is likely that in human PDL fibroblasts, arachidonic acid endogenouslyreleased in response to IL-1B is exclusively metabolized not by COX-1 but byCOX-2.

Although several reports have showed that IL-4 mRNA could not bedetected in human inflamed gingival tissue (36, 47), immunohistochemical stud-ies have demonstrated the presence of IL-4 protein in human inflamed gingi-val tissue (34). Memory T cells from the peripheral blood of adult periodon-titis patients with high anti-Porphyromonas gingivalis titers stimulated in vitrowith Porphyromonas gingivalis have been shown to produce higher amounts

COX-2 in IL-1B-Stimulated Human PDL Fibroblasts 9

10 Noguchi et al.

of IL-4 than do cells from healthy subjects (48). IFN-7 mRNA is detected inhuman inflamed gingival tissue (36, 47). Thus, it is likely that IL-4 and IFN-7are involved in the pathology of periodontal disease. The roles of IL-4 and IFN-7, however, are not clear. IL-4 can inhibit the production of proinflammatorycytokines such as IL-1, TNF and IL-8 and suppress bone resorption (24-26).IL-4 also depresses PG production through down-regulation of COX-2 induc-tion in human monocytes and neutrophils (24, 27, 28). On the other hand, IFN-7 primes human monocytes/macrophages to secondary stimuli such as LPS toproduce proinflammatory cytokines (30, 31). IFN-7 suppresses PG productionby human monocytes/macrophages in response to LPS, IL-1 or Con-A (32, 33).As shown in Figure 4A and 4B, our data showed that IL-4 and IFN-7 inhib-ited PGE2 production by IL-1B-stimulated PDL fibroblasts. The many biologicalactivities of PGE2 include basodilation, inhibition of lymphocyte proliferation,angiogenesis, and bone resorption (49). It is likely that IL-4 plays a protectiverole to periodontal tissue by inhibiting production of proinflammatory cytokinesand PGE2. IFN-7 also may have anti-inflammatory effects.

The mechanism by which IL-4 inhibited PGE2 production by IL-1B-stimu-lated PDL fibroblasts was due to down-regulation of COX-2 expression, whileIFN-7 suppressed PGE2 production with no effect on COX-2 expression. It hasbeen demonstrated that inhibition of PGE2 production by IL-4 in human lungfibroblasts stimulated with IL-1B is not dependent on down-regulation of COX-2 expression (39) and that IL-4 suppresses PGEa production via down-regula-tion of COX-2, in human blood monocytes and neutrophils (27, 28). Thus, theresponse of human cells to IL-4 may be dependent on cell types or stimuli. In ourstudy, the mechanism of inhibition of COX-2 mRNA expression was not exam-ined. Niiro et al. (27) have demonstrated that in human monocytes, IL-4 reg-ulates COX-2 mRNA expression at the transcriptional and post-transcriptionallevel. It has been shown that IFN-7 increases PG production by induction ofCOX-2 via epidermal growth factor receptor in human bronchial epithelial cells(50) and that IFN-y down-regulates LPS-induced PG production in rat microglialcells by inhibiting COX-2 expression (51). In the present study, IFN-7 inhibitedPGE2 production in IL-1B-stimulated human PDL fibroblasts, without suppres-sion of COX-2 mRNA expression, as shown in Figure 6. It has been reportedthat IFN-y inhibits PG production by depressing phospholipase activity in humanmonocytes (33). Thus, the mechanism by which IFN-7 abrogated PGE2 produc-tion in IL-1B-stimulated human PDL fibroblasts may be due to suppression ofphospholipase activity. Further investigation is necessary to clarify it.

In conclusion, our data suggest that COX-2 is primarily responsible forPGE2 production by IL-1B-stimulated human PDL fibroblasts and that IL-4inhibits PGE2 production by IL-1B-stimulated PDL fibroblasts through down-regulation of COX-2 expression, while IFN-7 suppresses the PGE2 productionwith no effect on COX-2 expression.

COX-2 in IL-1B-Stimulated Human PDL Fibroblasts 11

Acknowledgments—This study was supported by a Grant-in-aid for Scientic Research from the Min-istry of Education of Japan (No, 09671945).

REFERENCES

1. DEWITT, D. L. 1991. Prostaglandin endoperoxide synthease: regulation of enzyme expression.Biochimica. Biophisica. Acta. 1083: 121-134.

2. SMITH, W. L. 1992. Prostanoid biosynthesis and mechanisms of action. Am. J. Physiol. 263F181-F191.

3. KREAMER, S. A., E. A. MEADA, and D. L. DEWITT. 1992. Prostaglandin endoperoxide synthasegene structure: Identification of the transcriptional start site and 5'-flanking regulatory sequences.Arch. Biochem. Biophy. 293: 391-400.

4. KUJUBU, D. A., and H. R. HERSHMAN. 1992. Dexamelhasone inhibits mitogen induction of theTIS10 prostaglandin synthease/cyclooxygenase gene. J. Biol. Chem. 267: 7991-7994.

5. O'NEILL, G. P., and A. W. FORD-HUTCHINSON. 1993. Expression of mRNA for cyclooxygenase-1 and cyclooxygenase-2 in human tissues. FEBS Letters 330: 156-160.

6. HLA, T., and K. NEILSON. 1992. Human cyclooxygenase-2 cDNA. Proc. Natl. Acad. Sci. U. S. A.89: 7384-7388.

7. JONES, D. A., D. P. CARLTON, T. M. MCINTYRE, G. A. ZIMMERMAN, and S. M. PRESCOTT. 1993.Molecular cloning of human prostaglandin endoperoxide synthease type II and demonstrationof expression in response to cytokines. J. Biol. Chem. 268: 9049-9054.

8. LEE, S. H., E. SOYOOLA, P. CHANMUGAM, S. HART, W. SUN, H. ZHONO, S. Liou, D.SIMMON, and D. HWANG. 1992. Selective expression of mitogen-inducible cyclooxygenase inmacrophages stimulated with lipopolysaccharide. J. Biol. Chem. 267: 25934-25938.

9. MAIER, J. A. M., T. HLA, and T. MACIAG. 1990. Cyclooxygenase is an immediate-early geneinduced by interleukin-l in human endothelial cells. J. Biol. Chem. 265: 10805-10808.

10. OFFENBACHER, S., P. A. HEASMAN, and J. G. COLLINS. 1993. Modulation of host PGE2 secre-tion as a determinat of periodonatal disease expression. J. Periodontol. 64: 432-444.

11. OHM, K., H. -K., ALBERS, and B. P. LISBOA. 1984. Measurement of eight prostaglandins inhuman gingival and periodontai disease using high pressure liquid chromatography and radioim-munoassay. J. Periodont. Res. 19: 501-511.

12. WAITE, I. M., C. A. SAXON, A. YOUNG, B. J. WAGG, and M. CORBETT. 1981. The peri-odontai status of subjects receiving non-steroidal anti-inflammatory drugs. J. Periodont. Res.16: 100-108.

13. WILLIAMS, R. C., M. K. JEFFCOAT, T. H. HOWELL, A. ROLLA, D. STUBBS, K. W. TEOH, M. S.REDDY, and P. GOLDHABER. 1989. Altering the progression of human alveolar bone loss withthe non-steroidal anti-inflammatory drug flurbiprofen. J. Periodontol. 60: 485-490.

14. KLEIN, D. C., and J. G. RAISZ. 1970. Prostaglandins: stimulation of bone resorption in tissueculture. Endocrinology 86: 1436-1440.

15. SHELLING, S. H., H. J. WOLFE, and A. H. TASHIJIAN, JR. 1980. Role of the osteoclast inprostaglandin E2 stimulated bone resorption. Lab. Invest. 42: 290-295.

16. OFFENBACHER, S., B. M. ODLE, R. C. GRAY, and T. E. VAN DYKE. 1984. The use of crevicularfluid prostaglandin E levels as a measure of the periodontai disease status of adult and juvenileperiodontitis patients. J. Periodont. Res. 19: 1-13.

17. CAVANAUGH, JR., P. F., J. S. MCDONALD, L. PAYELIC, R. J. LIMARDI, J. L. GLICKMAN, andZ. P. PAVELIC. 1995. Immunohistochemical localization of prostaglandin H synthase isozymeproteins in the gingival tissue of patients with periodontitis. Inflammopharmacology. 3: 109-119.

18. YUCEL-LlNDBERG, T., H. AHOLA, S. NILSSON, J. CARLSTEDT-DUKE, and T. MODEER. 1995.

12 Noguchi et al.

Interleukin-1B induces expression of cyclooxygenase-2 mRNA in human gingival fibroblasts.Inflammation 195: 49-56.

19. NOGUCHI, K., M. SHITASHIGE, M. YANAI, I. MORITA, T. NISHIHARA, S. MUROTA, and I.ISHIKAWA. 1996. Prostaglandin production via induction cyclooxygenase-2 by human gingivalfibroblasts stimulated with lipopolysaccharides. Inflammation 20: 555-568.

20. SOMERMAN, M. J., S. Y. ARCHER, G. R. IMM, and R. A. FOSTER. 1988. A comparative studyof human periodontal ligament cells and gingival fibroblasts in vitro. J. Dent. Res. 67: 66-70.

21. NOJIMA, N., M. KOBAYASHI, M. SHINONOME, N. TAKAHASHI, T. SUDA, and K. HASEGAWA.

1990. Fibroblastic cells derived from bovine periodontal ligaments have the phenotypes ofosteoblasts. J. Periodont. Res. 25: 179-185.

22. RICHARDS, D., and B. RUTHERFORD. 1988. The effects of interleukin 1 on collagenolytic activityand prostaglandin-E secretion by human periodontal-ligament and gingival fibroblast. Archs.Oral Biol. 33: 237-243.

23. SAITO, S., P. NGAN, M. SAITO, R. LANESE, J. SHANFELD, and Z. DAVIDOVITCH. 1989. Interactiveeffects between cytokines on PGE production by human periodontal ligament fibroblasts in vitro.J. Dem. Res. 69: 1456-1462.

24. HART, P., G. F. VITTI, D. R. BURGESS, G. A. WHITTY, D. S., PICCOLI and J. A. HAMILTON.1989. Potential anti-inflammatory effects of interleukin-4: suppression of human monocytetumor necrosis factor alpha, interleukin-I and prostaglandin E2. Proc. Natl. Acad. Sci. U. S. A.86: 3803-3807.

25. CORCHORAN, M. L., W. G. STETLER-STEVENSON, P. D. BROWN, and L. M. WAHL. 1992.Interleukin-4 inhibition of prostaglandin E2 synthesis blocks interstitial collagenase and 92-kDatype IV collagenase/gelatinase production by human monocytes. J. Biol. Chem. 267: 515-519.

26. SHIOI A., N. LAURENT, B. GALVE-DE ROCHEMONTEIX, C. BAUMBERGER, J. M. DAYER, and H.G. WELGUS. 1992. Interleukin-4 inhibits murine osteoclast formation in vivo. J. Cell. Biochem.47: 272-277.

27. NIIRO, H., T. OTSUKA, T. TANABE, S. HARA, S. KUGA, N. YOSHIAKI, Y. TANAKA, H.NAKASHIMA, S. KITAJIMA, M. ABE, and Y. NIHO. 1995. Inhibition of interleukin-10 of induciblecyclooxygenase expression in lipopolysaccharide-stimulated monocytes: its underlying mecha-nism in comparison with interleukin-4. Blood 85: 3736-3745.

28. NIIRO, H., T. OTSUKA, K. IZUHARA, K. YAMAOKA, K. OSHIMA, T. TANABE, S. HARA, Y.NEMOTO, Y. TANAKA, H. NAKASHIMA, and Y. NIHO. 1997. Regulation by interleukin-10 andinterleukin-4 of cyclooxygenase-2 expression in human neutrophils. Blood 89: 1621-1628.

29. ADAMSON, S., S. S. EDWIN, S. LAMARCHE, and M. D. MITCHELL. 1994. Actions of interleukin-4 on prostaglandin biosynthesis by human amnion cells. Prostagl. Leukotriens. Fatty Acids50: 133-135.

30. BURCHETT, S. K., W. M. WEAVER, J. A. WESTALL, A. LARSEN, S. KRONHEIM, and C. B.WILSON. 1988. Regulation of tumor necrosis factor/cachectin and IL-1 secretion in humanmononuclear phagocytes. J. Immunol. 140: 3473-3481.

31. SONE, S., A. OKUBO, and T. OGATA. 1989. Normal human alveolar macrophages have moreability than blood monocytes to produce cell-associated interleukin-la. Am. J. Respir. Cell Mol.Biol. 1: 507-515.

32. BROWNING, J., and A. RIBOLINI. 1987. Interferon blocks interleukin 1-induced prostaglandinrelease from human peripheral monocytes. J. Immunol. 138: 2857-2863.

33. WAHL, L. M., and M. CORCORAN. 1993. Regulation of monocyte/macrophage metalloproteinaseproduction by cytokines. J. Periodontol. 64: 467-473.

34. YAMAZAKI, K., T. NAKAJIMA, E. GEMMELL, B. POLAK, G. J. SEYMOUR, and K. HARA.1994. IL-4 and IL-6-producing cells in human periodontal disease tissue. J. Oral Pathol. Med.23: 347-353.

35. KABASHIMA, H., K. NAGATA, I. HASHIGUCHI, Y. TORIYA, T. IIJIMA, K. MAKI, and K. MAEDA.

COX-2 in IL-1B-Stimulated Human PDL Fibroblasts 13

1996. Interleukin-1 receptor antagonist and interleukin-4 in gingival crevicular fluid of patientswith inflammatory disease. J. Oral Pathol. Med. 25: 449-455.

36. FUJIHASHI, K., M. YAMAMOTO, T. HIROI, T. V. BAMBERG, J. R. MCGHEE, and H. KIYONO. 1996.Selected Th1 and Th2 cytokine mRNA expression by CD4+ T cells isolated from imflamedhuman gingival tissues. Clin Exp. Immunol. 103: 422-28.

37. FUTAKI, N., S. TAKAHASHI, M. YOKOYAMA, I. ARAI, S. HIGUCHI, and S. OTOMO. 1994.New anti-inflammatory agent, selectively inhibits prostaglandin G/H synthase/cyclooxygenase(COX-2) activity in vitro. Prostaglandins 47: 55-59.

38. WARNOCK, L. J., and G. W. HUNNINGHAKE. 1995. Multiple second messenger pathways reg-ulates IL-1B-induced expression of PGHS-2 mRNA in normal human skin fibroblasts. J. Cell.Physiol. 163: 172-178.

39. ENDO, T., F. OGUSHI, S. SONE, T. OGURA, Y. TAKETANI, Y. HAYASHI, N. UEDA, and S.YAMAMOTO. 1995. Induction of cyclooxygenase-2 in responsible for interleukin-1B-dependentprostaglandin E2 synthesis by human lung fibroblasts. Am. J. Respir. Cell Mol. Biol. 12: 358-365.

40. MASFERRER, J. L., S. T. REDDY, B. S. ZWEIFEL, K. SEIBERT, P. NEEDLEMAN, R. S.GILBERT, and H. R. HERSHMAN. 1994. In vivo glucocorticoids regulate cyclooxygenase-2but not cyclooxygenase-1 in peritoneal macrophages. J. Pharmacol. Experiment. Therapeutics.270: 1340-1344.

41. O'BANION, M. K., H. B. SADOWSKI, V. D. WINN, D. A. YOUNG. 1995. A serum- andglucocorticoid-regulated 4-kilobase mRNA encodes a cyclooxygenase-related protein. J. Biol.Chem. 266: 23261-23267.

42. O'BANION, M. K., V. D. WINN, and D. A. YOUNG. 1992. cDNA cloning and functional activ-ity of a glucocorticoid-regulated inflammatory cyclooxygenase. Proc. Natl. Acad. Sci. U. S. A.89: 4888-4892.

43. REDDY, S. X, and H. R. HERSHMAN. 1994. Ligand-induced prostaglandin synthesis requiresexpression of the TIS10/PGS-2 prostaglandin synthase gene in murine fibroblasts andmacrophages. J. Biol. Chem. 269: 15473-15480.

44. MORITA, I., M. SCHINDLER, M. K. REGIER, J. C. OTTO, T. HORI, D. L. DEWlTT, and W. L.SMITH. 1995. Differential intracellular locations for prostaglandin endoperoxide H synthease-1and -2. J. Biol. Chem. 270: 10902-10908.

45. MURAKAMI, M., R. MATSUMOTO, K. F. AUSTEN, and J. P. ARMS. 1994. Prostaglandin endo-peroxide synthase-1 and -2 couple to different transmembrane stimuli to generate prostaglandinD2 in mouse bone marrow-derived mast cells. J. Biol. Chem. 269: 22269-22275.

46. REDDY, S. T., and H. R. HERSHMAN. 1996. Transcellular prostaglandin production followingmast cell activation is mediated by proximal secretory phospholipase A2 and distal prostaglandinsynthase 1. J. Biol. Chem. 271: 186-191.

47. PRABHU, A., B. S. MICHALOWICZ, and A. MATHUR. 1996. Detection of local and systemiccytokines in adult periodontitis. J. Periodontol. 67: 515-522.

48. AOYAGI, X, M. SUGAWARA-AOYAGI, K. YAMAZAKI, and K. HARA. 1995. Interleukin-4 (IL-4) and IL-6-producing memory T cells in peripheral blood and gingival tissues in periodontitispatients with high serum antibody titers to Porphyromonas gingivalis. Oral Microbiol. Immunol.10: 304-310.

49. PAGE, R. C. 1991. The role of inflammatory mediators in the pathogenesis of periodontal dis-ease. J. Periodont. Res. 26: 230-242.

50. ASANO, K., H. NAKAMURA, C. M. LILLY, M. KLAGSBRUN, and J. M. DRAZEN. 1997. Interferon-y induces prostaglandin G/H synthase-2 through an autocrine loop via the epidermal growthfactor receptor in human bronchial epithelial cells. J. Clin. Invest. 99: 1057-1063.

51. MlNGHETTI, L., E. POLAZZI, A. NICOLINI, C. CREM1NON, and G. LEVI. 1996. Interferon-yand nitric oxide down-regulate lipopolysaccharide-induced prostanoid production in culturedrat microglial cells by inhibiting cyclooxygenase-2 expression. J. Neurochem. 66: 1963-1970.