INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ORBITAL TISSUE is particularly susceptible to the inflammatory manifestations of Graves' disease, termed thyroid-associated ophthalmopathy (TAO). We hypothesize that this propensity is directly related to the orbital fibroblast phenotype (23). In particular, proinflammatory cytokines, such as IL-1, exhibit striking activity in cultured orbital fibroblasts. IL-1 represents a small family of cytokines, including IL-1, IL-1, and IL-1 receptor antagonist (IL-1ra) (8). Occupation of the IL-1 receptor by IL-1ra blocks the binding and action of both IL-1 and IL-1. It is the balance between IL-1 and IL-1ra binding that determines the net biological impact of IL-1 on the target cell.

IL-1 has been shown to regulate the synthesis of prostaglandin E₂ (PGE₂) in a variety of cells, including those from various depots of connective tissue. These include dermal (18) and lung fibroblasts (19), osteoblasts (7), and rheumatoid synoviocytes (6). Prostaglandin-endoperoxide H synthase (PGHS; EC 1.14.99.1) catalyzes the two rate-limiting steps in the synthesis of prostaglandin H₂ (PGH₂) from arachidonic acid. Two isoforms of PGHS have been identified. PGHS-1 is expressed in most tissues and cell types under unprovoked conditions and appears to be responsible for the production of prostanoids involved in physiological functions (28). PGHS-2 becomes induced during inflammation, and its expression is upregulated by various cytokines, growth factors, and serum and is attenuated by glucocorticoids (16). PGHS-2 has been cloned, and the deduced amino acid sequence is 61% identical to PGHS-1. The human PGHS-2 transcript contains at least 22 AUUUA sequence motifs in the 3'-untranslated region (UTR), perhaps accounting for its high degree of instability (15).

We reported previously (31) that leukoregulin, a 50-kDa product of mitogen-activated T lymphocytes, can upregulate dramatically steady-state levels of PGHS-2 mRNA in human orbital fibroblasts. This massive induction of PGHS-2 was associated with a marked increase in PGE₂ production. In contrast, PGHS-2 expression in dermal fibroblasts was induced more modestly by leukoregulin. The rate of PGHS-2 gene transcription is increased modestly, strongly suggesting that transcript stability is being enhanced substantially by the cytokine. CD40 is displayed by orbital fibroblasts (21) and, when ligated with CD154, results in a substantial induction of PGHS-2 and an increase in PGE₂ production (3). Enhanced PGHS-2 mRNA stability probably accounts for this increase in prostanoid generation.

Orbital fibroblasts exhibit a set of phenotypic attributes that distinguish them from cells derived elsewhere (25-27). Activation of these fibroblasts appears to lead to extensive tissue remodeling and accumulation of glycosaminoglycans in TAO (24). In addition, the inflammatory reaction often associated with TAO may result from the local generation by orbital fibroblasts of disease mediators (14). In the current study, we examined the ability of IL-1 to influence PGHS-2 expression and PGE₂ production in these fibroblasts. IL-1 increases substantially the levels of PGHS-2 mRNA and protein. In contrast, PGHS-1 expression in orbital fibroblasts is not influenced by either of these cytokines. The induction of PGHS-2 is blocked by dexamethasone, resulting in decreased PGE₂ synthesis. Both IL-1 gene products can upregulate the expression of endogenous IL-1 and IL-1 far more in orbital than dermal fibroblasts. Exogenous IL-1 can induce IL-1ra, but this is substantially greater in dermal than in orbital fibroblasts. We conclude that a "defective" IL-1ra response to IL-1 may underlie, at least in part, the exaggerated PGHS-2 induction in orbital fibroblasts.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials. Human recombinant IL-1 and IL-1 were purchased from BioSource International (Camarillo, CA). PGHS-1 and PGHS-2 Abs were kindly supplied by Dr. J. MacLouf (CEA, Gif-sur-Yvette, France) or purchased from Cayman Chemical (Ann Arbor, MI). IL-1ra was kindly supplied by Amgen (Boulder, CO). PGE₂ assay kits were purchased from Amersham, and IL-1 and IL-1 ELISA kits were from Immunotech (Westbrook, ME); those for IL-1ra were from R&D Systems (Minneapolis, MN). PGHS-1 and PGHS-2 cDNA plasmids were kindly provided by Drs. D. A. Young and Kerry O'Banion (University of Rochester, Rochester, NY). Anti-IL-1 and IL-1 antibodies were purchased from R&D. SC-58125 was a gift from G. D. Searle (Skokie, IL). Dexamethasone and pyrrolidinedithiocarbamate (PDTC) were purchased from Sigma (St. Louis, MO). NF-B oligonucleotide (5'-AGTTGAGGGGACTTTCCCAGGC-3') and its complement, as well as SP1 oligonucleotide (5'-ATTCGATCGGGGCGGGGGCGATGC-3') and its complement, were obtained from Promega (Madison, WI). A fragment of the 3'-UTR of the human PGHS-2 sequence spanning +1885 to +2395 was generated by PCR using the following primers: 5'-CTAAATACGTAGAACGTTCGACTGAACTG-3' and 5'-GAAATTACTCGAGCTGGTAATGTCTAAT-TTAAATAT-3'.

Cell culture. Fibroblast cultures were initiated from tissue explants obtained during orbital surgery. Dermal fibroblast cultures were obtained from punch biopsies of normal-appearing skin. The Institutional Review Boards of Albany Medical College and Harbor-UCLA Medical Center have approved these activities. Procedures for the cell culture have been described in detail previously (22). Briefly, tissue specimens were minced and covered with Eagles' MEM (GIBCO) containing 10% fetal bovine serum (FBS), L-glutamine, penicillin/streptomycin, and nystatin. When fibroblast confluence was reached, the explant was removed and monolayers were disrupted by treatment with trypsin-EDTA. All experiments were performed with fibroblasts within 12 passages of culture initiation. Seven fibroblast strains, each from a different donor, including four from the orbit and three from skin, were analyzed.

Western blot analysis. Levels of cyclooxygenase proteins were determined by Western immunoblot analysis using Abs directed against PGHS-1 and PGHS-2, as previously described (31). Confluent 60-mm plate cultures were shifted to medium containing 1% FBS for 48 h. Cells were then treated with IL-1 (10 ng/ml) or the test compounds indicated, usually for 16 h. Cell layers were washed three times with phosphate-buffered saline (PBS) and taken up in lysis buffer {20 mM Tris · HCl, pH 7.5, 15 mM 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid (CHAPS), 1 mM EDTA, 10 µM phenylmethylsulfonyl fluoride (PMSF), and 10 U/ml soybean trypsin inhibitor}. Cellular protein (20 µg) was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene difluoride membrane (Bio-Rad). After incubation with primary and then secondary peroxidase-labeled antibodies (10 µg/ml), signals were detected with enhanced chemiluminescence (ECL; Amersham). The resulting bands were scanned densitometrically with a BioImage densitometer (Milligen).

Isolation and quantification of mRNAs in human fibroblasts. Fibroblasts were cultivated to confluence in 100-mm plates, shifted to medium containing 1% FBS for 16 h, and then treated with IL-1 (10 ng/ml) for the times indicated. Cellular RNA was extracted by the method of Chomczynski and Sacchi (4) with the use of guanidium isothiocyanate (Ultraspec RNA isolation systems; Biotecx, Houston, TX), precipitated from the aqueous phase by the addition of isopropanol, washed with 75% ethanol, and solubilized in diethyl pyrocarbonate-treated water. Equal amounts of RNA (10 µg) were electrophoresed in 1% agarose formaldehyde gels and transferred to Zetaprobe (Bio-Rad). [³²P]dCTP random-primed (Bio-Rad) PGHS probes were hybridized in 5× SSC, 5× Denhardt's solution, 50% formamide, 50 mM phosphate buffer (pH 6.5), 1% SDS, and 0.25 mg/ml sheared, denatured salmon sperm DNA at 48°C overnight. Membranes were washed under high-stringency conditions and exposed to X-OMAT AR film (Kodak) at 70°C. To normalize the amount of RNA transferred, membranes were stripped and rehybridized with a human GAPDH cDNA probe. Radioactive DNA/RNA hybrids were quantified by subjecting autoradiographs to densitometric analysis.

PGE₂ assay. Fibroblasts were cultured to confluence in 24-well plates covered with medium containing 10% FBS. Monolayers were then shifted to medium with 1% FBS for 16 h. IL-1 without or with other test compounds was added at the times indicated. Before assay, the medium was removed and replaced with 150 µl of PBS in the presence of the treatment compounds for the final 30 min of incubation. PBS was collected and subjected to radioimmunoassay as previously described (31) to determine PGE₂ release from cultured cells.

Assay of IL-1ra proteins. Fibroblasts were cultured in 24-well plates to confluence in medium supplemented with 10% FBS. Cultures were shifted to 1% FBS for the final 16 h of incubation. Test compounds were added at the times indicated. After treatment, the monolayers were washed extensively with PBS and the cells were taken up in lysis buffer. Both medium (200 µl) and cellular protein (10 µg) were subjected to an ELISA assay (R&D).

Isolation of nuclear proteins and EMSA assays. Nuclear proteins were prepared essentially as described by Andrews and Faller (1). Confluent fibroblasts, grown in 100-mm-diameter dishes, were treated with IL-1 (10 ng/ml) for up to 2.0 h. Cells were then washed and scraped in PBS, and then microcentrifuged. The cell pellet was suspended in 0.5 ml of buffer A [10 mM HEPES (pH 7.8), 15 mM KCl, 2 mM MgCl₂, 0.1 mM EDTA, 1 mM dithiothreitol (DTT), and 1 mM PMSF] and centrifuged at 750 g for 5 min. The pellet was resuspended in 200 µl of buffer A and incubated at 4°C for 10 min. Nonidet P-40 was added to a final concentration of 0.5%, and the suspension was centrifuged at 14,000 g for 15 min. The resultant nuclear pellet was suspended in 15 µl of 20 mM HEPES (pH 7.9), 1.5 mM MgCl₂, 0.5 mM DTT, 0.42 M NaCl, 0.2 mM EDTA, 25% glycerol, and 0.5 mM PMSF, incubated for 15 min at 4°C, and centrifuged at 14,000 g for 10 min. The supernatant (15 µl) was diluted with 35 µl of 20 mM HEPES (pH 7.9), 50 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, and 0.5 mM PMSF and stored at 80°C.

Transfection of fibroblasts with gene promoter/reporter constructs and IL-1ra cDNA. Some cultures were transiently transfected with an 1,800-bp fragment of the PGHS-2 promoter, kindly supplied by Dr. Stephen Prescott (University of Utah) fused to a luciferase reporter gene. Others were transfected with a fragment of the PGHS-2 3'-UTR spanning nt +1885 to +2395 and fused to a CAT reporter. Control cultures were transfected with an unrelated yeast gene sequence fused to the same reporter. Fibroblasts 60%-80% confluent in 60-mm-diameter plates were transfected with 2 µg of plasmid DNA with LipofectAMINE PLUS reagent and allowed to incubate for 5 h. Medium containing 10% FBS was then added. CAT assays were performed on cultures after 48-h incubations with or without the test agents indicated by using a kit purchased from Promega.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IL-1 enhances PGE₂ synthesis in human orbital fibroblasts, which is associated with a dramatic induction of PGHS-2. Confluent orbital and dermal fibroblasts synthesize low levels of PGE₂ (31). Treatment of orbital fibroblasts with IL-1 (10 ng/ml) increases prostanoid production. Levels are enhanced 50- to 70-fold. In contrast, PGE₂ synthesis is upregulated more modestly (10-fold) in dermal fibroblasts treated under identical culture conditions (Fig. 1A). We examined seven different strains of fibroblasts, each from a different donor. These included individuals with Graves' disease and those without any evidence of autoimmune disease. Four of the strains were from the orbit, whereas three derived from the skin. From these studies, it appears that orbital fibroblasts, regardless of whether they are from patients with Graves' disease or from normal connective tissue, exhibit substantially greater increases in PGE₂ synthesis when activated by IL-1 than do their dermal counterparts.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   A: effects of IL-1 and IL-1 on levels of PGE2 production in human orbital and dermal fibroblasts (B). These effects can be attenuated by dexamethasone (Dex) and pyrrolidinedithiocarbamate (PDTC), an inhibitor of NF-B activity. Confluent fibroblasts in 24-well plates were treated without (control) or with IL-1 (10 ng/ml) or IL-1 (10 ng/ml), in the absence or presence of Dex (10 nM) or PDTC (100 µM), for 16 h. Medium was replaced with PBS for the final 30 min of incubation. The PBS (100 µl) was subjected to PGE2 analysis. Values are means ± SE of triplicate determinations from representative experiments.

PGHS-2 protein and steady-state mRNA levels are increased by IL-1 in fibroblasts. The effect of IL-1 on PGHS-2 expression in orbital fibroblasts was determined by incubating cultures in the absence or presence of IL-1 (10 ng/ml) for the times indicated in Fig. 2. Cellular proteins and RNA were harvested and subjected to Western and Northern blot analysis, respectively, as indicated in Fig. 2. PGHS-2 protein was undetectable at time 0 and peaked at 12 h, when levels were at least 10-fold above controls. In contrast, PGHS-1 protein levels were uninfluenced by IL-1 (data not shown). A concentration curve was generated using graded doses of IL-1 (Fig. 3). The data from that study demonstrate that the induction of PGHS-2 protein is near maximal at an IL-1 concentration of ~1 ng/ml in orbital fibroblasts. PGHS-2 induction was far less substantial in dermal cultures than that found in orbital fibroblasts (Fig. 2). The protein was readily detectable only at the 16-h time point.

View larger version (45K): [in this window] [in a new window]   Fig. 2.   IL-1 upregulates the expression of prostaglandin-endoperoxide H synthase-2 (PGHS-2) mRNA (A) and protein (B) in orbital and dermal fibroblasts. Confluent cultures of orbital fibroblasts from a patient with severe thyroid-associated ophthalmopathy (TAO) and abdominal wall dermal fibroblasts were treated with IL-1 (10 ng/ml) for the times indicated. Monolayers were processed as described in METHODS. Cellular RNA was electrophoresed and subjected to Northern blot hybridization with a probe generated from PGHS-2 cDNA and rehybridized with a GAPDH probe. For Western blot analysis, 15 µg of cellular protein were subjected to SDS-PAGE. Separated proteins were transferred to polyvinylidene difluoride membrane and immunoblotted with an anti-human PGHS-2 antibody (10 µg/ml).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Induction by IL-1 of PGHS-2 protein in orbital fibroblasts is concentration dependent. Confluent cultures were treated with the graded concentrations of IL-1 indicated. Cell lysates were then subjected to Western blot analysis for PGHS-2 protein. Data are shown as relative density of protein expressed.

IL-1 activates NF-B in orbital fibroblasts. The effects of IL-1 on PGE₂ production and PGHS-2 expression in orbital fibroblast cultures could be attenuated by treatment with PDTC, a relatively specific inhibitor of NF-B (30). The human PGHS-2 promoter contains two NF-B binding sites at 214 to 204 and 447 to 437, and it has been shown previously that IL-1 can activate NF-B in other cell types. Therefore, the impact of IL-1 on NF-B binding activity in orbital fibroblasts was determined. Cultures were treated with nothing or IL-1 (10 ng/ml), and nuclear proteins were harvested after 2 h. There was negligible NF-B binding activity in the control nuclear extract (Fig. 4A), but a substantial increase occurred after addition of the cytokine. Binding could be quenched with excess unlabeled NF-B but not by the SP1 consensus sequence oligonucleotide, indicating that the binding of the nuclear proteins is specific to NF-B (Fig. 4B).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   NF-B binding activity is increased by IL-1 in orbital fibroblasts. A: confluent orbital fibroblasts were treated with nothing (control) or IL-1 (10 ng/ml) for 2 h. Nuclear proteins were harvested and subjected to EMSA as described in METHODS. B: EMSA competition assay showing binding buffer (without nuclear extract) and nuclear extract in the absence or presence of unlabeled NF-B or SP1 oligonucleotides, as indicated. C: supershift assay. EMSA was performed without (left lane) antibodies directed against NF-B or with antibodies against p50 (middle lane) or p65 (right lane).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   NF-B binding activity in IL-1-treated orbital and dermal fibroblasts. Cultures were treated with IL-1 (10 ng/ml) for 2 h. Nuclear extracts were then processed as described in METHODS.

IL-1 increases steady-state PGHS-2 mRNA levels primarily through stabilization of the transcript. IL-1 influences target gene transcription and is also known to influence the stability of mature transcripts encoding inflammation-related proteins. Its impact on levels of PGHS-2 expression in orbital fibroblasts is considerable, and thus the effect of IL-1 on PGHS-2 promoter activity was assessed. As Fig. 6 indicates, the cytokine fractionally increased the activity of an 1,800-bp fragment of the human PGHS-2 promoter fused to a luciferase reporter and transiently transfected into orbital and dermal fibroblasts. The magnitude of increase in promoter activity was modest in both cultures (2- to 3-fold), suggesting that the levels of PGHS-2 gene transcription may not differ in IL-1-treated orbital and dermal fibroblasts.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effect of IL-1 on PGHS-2 gene promoter activity in orbital (A) and dermal (B) fibroblasts. Fibroblasts were transiently transfected with the +1840 to +123 bp fragment of the human PGHS-2 promoter fused to a luciferase (Luc) reporter gene. Cells were treated for 3 h with IL-1 (10 ng/ml). Luciferase activity was determined as described in METHODS. Data are expressed as means ± SE of triplicate cultures.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Effect of IL-1 on PGHS-2 mRNA stability and 3'-untranslated region (UTR) of PGHS-2. A: orbital fibroblast cultures were pretreated with cycloheximide (10 µg/ml) for 3 h, and then all were shifted to fresh medium containing 5,6-dichlorobenzimidazole (50 µM) without (control) or with IL-1, and monolayers were harvested at the times indicated. Cellular RNA was subjected to Northern blot hybridization with a 32P-labeled PGHS-2 cDNA probe and subsequently with GAPDH, the later signal used to normalize the RNA transfer. B: orbital fibroblasts were transiently transfected with SV40 CAT, an unrelated yeast gene control sequence, or a fragment (+1885/+2395) of the PGHS-2 3'-UTR fused to a CAT reporter gene. Cultures were treated with nothing or IL-1 for 3 h. CAT activity was determined as described in METHODS. Data are expressed as means ± SE of triplicate cultures.

Exogenous IL-1 upregulates IL-1ra expression in orbital and dermal fibroblasts. We have shown that treatment of orbital fibroblasts with exogenous IL-1 results in an upregulation of PGHS-2 expression that is far greater than that in dermal cultures. We next determined whether IL-1 is inducing the expression of other members of its cytokine family in fibroblasts, as has been shown in other cell types (9). Treatment of orbital fibroblasts with either IL-1 (10 ng/ml) or IL-1 (10 ng/ml) resulted in substantial induction of IL-1 (30- to 60-fold above control) (Fig. 8A). The peak levels for IL-1 expression were 408 ± 45 and 211 ± 35 pg/10 µg protein when cells were treated with IL-1 and IL-1 for 12 h, respectively. IL-1 was also dramatically induced in these cells. The levels were undetectable under control conditions but increased to 97.5 ± 8 and 116 ± 3.5 pg/10 µg protein after 12 h of IL-1 and IL-1 treatment, respectively.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Effect of IL-1 and IL-1 on IL-1 expression (A) and the effect of IL-1 on IL-1 receptor antagonist (IL-1ra) protein levels (B) in fibroblasts. Confluent cultures in 24-well plastic plates were incubated without (control) or with IL-1 or IL-1 (10 ng/ml) for 12 h in A or for the times indicated in B. The cellular proteins were harvested, and 10 µg of protein were subjected to the respective ELISA assays. Results are expressed as means ± SE of triplicate determinations.

Overexpressing IL-1ra in orbital fibroblasts or interrupting its expression in dermal fibroblasts alters PGHS-2 induction by IL-1. To determine whether differential induction by IL-1 of PGHS-2 expression in orbital and dermal fibroblasts was related to IL-1ra expression, levels of the cytokine antagonist were altered and the impact on enzyme induction assessed. As the data in Fig. 9A indicate, treating dermal fibroblasts with IL-1 in addition to neutralizing anti-IL-1ra antibodies could enhance the induction of PGHS-2. Interrupting IL-1ra expression in dermal cultures using an IL-1ra anti-sense oligonucleotide could produce an equivalent upregulation in PGHS-2 induction with IL-1 (Fig. 9B). Orbital fibroblasts were then transiently transfected with IL-1ra. As the Western blot in Fig. 9C, left, indicates, levels of IL-1ra protein were increased dramatically after the transfection. When these cells were then challenged with IL-1 (10 ng/ml), the PGHS-2 induction was substantially attenuated (Fig. 9C, right) and the levels were similar to those in dermal fibroblasts. Thus it would appear that the relative level of IL-1ra protein expressed by fibroblasts after cytokine activation represents an important determinant of PGHS-2 inducibility in human fibroblasts.

View larger version (24K): [in this window] [in a new window]   Fig. 9.   Effect of interrupting or augmenting IL-1ra activity in orbital and dermal fibroblasts. A: orbital and dermal fibroblast monolayers were treated without (control) or with IL-1 without or with neutralizing anti-IL-1ra antibodies (Ab) for 16 h. Cultures were solubilized and subjected to Western blot analysis for PGHS-2 protein expression. B: orbital and dermal fibroblasts were treated with an antisense oligonucleotide for 5 h. Fibroblasts were then treated without (control) or with IL-1 for 16 h, and PGHS-2 protein expression was determined by Western blot analysis. C: orbital fibroblasts were transiently transfected with an empty vector or with one containing IL-1ra cDNA. IL-1ra protein was assayed (left) to confirm expression, and then Western blot analysis of PGHS-2 protein was performed on the transfected cells after a 16-h treatment without (control) or with IL-1. Data are shown as relative density of protein expressed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Exogenous IL-1 can dramatically enhance PGHS-2 expression and PGE₂ production in orbital fibroblasts. Similar actions of IL-1 on prostanoid biosynthesis have been reported in a wide array of cell types, including primary and established endothelial cells, synovial cells, and astrocytes (28). The induction of PGHS-2 by IL-1 is considerably less robust in dermal fibroblasts. We have reported previously that leukoregulin can also substantially upregulate PGHS-2 expression in orbital fibroblasts (31). Those effects were also anatomic site selective.

A potentially important insight into the basis for differential PGHS-2 inducibility in orbital and dermal fibroblasts relates to the differences in levels of cytokine-inducible IL-1ra protein observed in the two cell types. A major determinant of fibroblast phenotype underlying its potential to participate in inflammation may relate to the usage of the IL-1 family of cytokines and the magnitude of the IL-1ra response to a particular proinflammatory signal.

With regard to PGHS-2, IL-1 has been shown to exert effects on gene transcription and on mRNA stability in endothelial cells in culture (19). Consistent with those findings, we show that induction of PGHS-2 by IL-1 in orbital fibroblasts is a consequence of both a modest increase in gene transcription and enhanced PGHS-2 mRNA stability. Morrison and colleagues (5, 29) demonstrated the critical importance of the 3'-UTR in PGHS-2 mRNA to cytokine regulation of cyclooxygenase expression.

Our finding that PGHS-2 expression in orbital fibroblasts can be dramatically upregulated by IL-1 is of considerable clinical relevance. The cytokine milieu associated with TAO remains poorly defined. IL-1 as well as TNF- and interferon- have been detected with immunohistological techniques in the orbital tissues from patients with TAO (12). Though that study failed to examine IL-1 expression or to establish convincingly the identity of the particular cell type expressing these cytokines, it does suggest that IL-1 may play a role in TAO.

The dramatic increases in PGE₂ production found in cytokine-activated orbital fibroblasts suggest that orbital connective tissue might generate high levels of the prostanoid. PGE₂ exerts important influences on T and B cells (2, 10, 20). For instance, the development of naive T cells is biased from T_H0 to T_H2 cells at the expense of the T_H1 phenotype. Moreover, PGE₂ influences B cell development. Mast cells also react to PGE₂, and the molecule plays a role in their behavior. Thus high levels of prostanoid generation in the orbit might condition the immune responses occurring there (17).

Our current findings advance the understanding of prostanoid generation in fibroblasts and suggest that the expression of IL-1 and IL-1ra in human cultures may provide an important phenotypic signature for potential involvement in inflammation. On the basis of our results to date, it would appear that orbital fibroblasts represent a cell population particularly poised to engage in inflammatory activities. Substantial debate remains concerning the nature of PGE₂ effects on inflammation and immune function (17). Clearly, the realization that orbital fibroblasts produce unusually high levels of this prostanoid when treated with proinflammatory cytokines suggests that the connective tissues from which they derive might be exposed to PGE₂ as a consequence of disease.