Male Sprague-Dawley (SD) rats (8 weeks old, 280 to 300 g) were obtained from BioLASCO Taiwan (Taipei, Taiwan). All experiments were approved by the local Institutional Review Board and performed in adherence to the National Institutes of Health Guidelines for the treatment of laboratory animals. The synovium of knee joints was aseptically removed from normal SD rats, cut into small fragments and incubated with antimicrobial solution (500 IU/ml penicillin/streptomycin; Gibco Invitrogen, Burlington, Ontario, Canada) for 1 h, washed with sterile phosphate-buffered saline (PBS) before digestion with 3 mg/ml collagenase type H (Sigma, St Louis, MO, USA) at 37°C for 12 h. The resultant cell suspension was centrifuged at 2,500 rpm for 10 minutes following which the supernatant was discarded and the pellet resuspended in PBS. After further centrifugation at 1,000 rpm for 10 minutes, cells were resuspended and seeded in 20 ml of Ham's F12 medium (Sigma) containing 10% fetal bovine serum (Gibco Invitrogen) and 100 IU/ml penicillin/streptomycin (Gibco Invitrogen). The synovial cells were then cultured in a humidified 5% CO₂ atmosphere at 37°C until confluent, detached with 0.05% trypsin/ethylenediaminetetraacetic acid (EDTA) (Gibco Invitrogen) and seeded at a density of 2 × 10⁵ cells/dish in 60 mm tissue culture dishes (Orange scientific, Braine-l'Alleud, Belgium) for further experimental procedures.

C. albicans (ATCC 90028) was grown on Sabouraud dextrose agar (BD Microbiology System, Sparks, MD, USA) at 25°C. After a 16-h culture, colonies were suspended in PBS (Gibco Invitrogen) and prepared to the desired density of 1 × 10³ to 1 × 10⁷ yeasts/ml.

Dishes of synovial fibroblasts were placed in serum-free media (3 ml) overnight and then treated with either 200 μL PBS or 200 μL suspension of C. albicans (2 × 10² to 2 × 10⁶ yeasts/dish) in 5% CO₂ atmosphere at 37°C for 6 or 12 h. In some experiments synovial fibroblasts were pre-incubated with U0126 (Cell Signaling Technology, Beverly, MA, USA), a mitogen-activated protein kinase (MEK)1/2 inhibitor, at a concentration of 20 μM for 2 h; laminarin (Sigma) a β-glucan receptor blocking agent and specific inhibitor of dectin-1 activity at a concentration of 10 mg/ml for 1 h. MG-132 (Calbiochem, San Diego, CA, USA) as a NFκB inhibitor was co-incubated with synovial fibroblasts at a concentration of 35 μM. For the trans-well experiments (Transwell, Corning Incorporated, Corning, NY, USA), synovial fibroblasts were seeded in the upper chamber and C. albicans were plated in the lower chamber overnight, and then interacted for 12 h. In controls C. albicans were omitted from the lower chamber.

After a 12-h co-culture of synovial fibroblasts and C. albicans, cells and fungi on dish were washed with ice-cold PBS twice and then fixed using 2 ml of a 1:1 methanol/acetone mixture per dish for 5 minutes at -20°C. Cells were then stained by immunocytochemistry. Immunodetection for COX-2 was performed with a standard avidin-biotin-peroxidase complex detection kit (DakoCytomation, Glostrup, Denmark). Dishes were washed twice with PBS and blocked by incubation with 200 μL 1% non-immune horse serum (Vector Laboratories, Burlingame, CA, USA) in 1% bovine serum albumin (BSA) in antibody diluent (DakoCytomation) for 30 minutes at room temperature. The solution was poured off and the cells incubated sequentially with anti-COX-2 epitope specific antibody (1:200) (Lab Vision Corporation, Cheshire, UK) or anti-phospho-ERK1/2 (1:100) (Cell Signaling) for 60 minutes, biotinylated secondary antibody (1:200) for 45 minutes, and horseradish peroxidase (HRP)-conjugated streptavidin for 20 minutes. Between each incubation cells were washed with Tris-buffered saline/Tween (TBST) (12.5 mM Tris/HCl, pH 7.6, 137 mM NaCl, 0.1% Tween 20) three times. The chromogen 3-amino-9-ethylcarbazole (AEC) was then added for 15 minutes and finally counterstained with Mayer's hematoxylin. The cells were mounted with a coverslip and visualized under light microscopy.

Total RNA was isolated from cells after a 12-h co-culture of synovial fibroblasts and C. albicans using easy-BLUE Total RNA Extraction Kit (iNtRON Biotechnology, Gyeonggi-do, Korea). For first strand cDNA synthesis, 3 μg of total RNA was used in a single-round RT reaction (total volume 20 μl), containing 0.75 μg oligo(dT)₁₄ primer, 1 mM deoxynucleosides (dNTPs), 1 × first strand buffer, 0.4 mM dithiothreitol (DTT), 40 units RNaseOut recombinant ribonuclease inhibitor, and 200 units of superscript II reverse transcriptase (Gibco Invitrogen). The reverse transcription reaction was performed at 42°C for 2 h, followed by 95°C for 5 minutes. PCR was run using 0.9 μl of the reverse transcription reaction mixture as template, 0.4 mM of gene specific primers, 1 × PCR buffer, 0.25 mM dNTPs, and 1.5 units of Taq DNA polymerase (BioMan, Taipei, Taiwan). The amplification was carried out at 94°C for 1 minute, then for 30 cycles at 94°C for 1 minute, 56°C for 1 minute, and 72°C for 1 minute followed by a final extension at 72°C for 10 minutes. All PCR products were size-fractionated by a 1.5% agarose gel electrophoresis, and DNA bands were visualized by staining the gel with 0.1 μg/ml ethidium bromide. The bands were analyzed using gel documentation system (Bio-Profil, Bio-1D version 99; Viogene, Sunnyvale, CA, USA). The values were expressed as ratio of the band intensity of the target gene to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and the ratio of the band intensity of COX-2/GAPDH in the control condition was normalized to 1. Variance and P values were analyzed by Alphaimager 1220 V5.5 (Alpha Innotech Corporation, San Leandro, CA, USA). A Student t test was used for statistical comparison between groups. A P value of less than 0.05 was considered statistically significant.

The primers used were as follows: COX-2 5'-GTCTCTCATCTGCAATAATGTG-3' (sense) and 5'-ATCTGTGTGGGTACAAATTTG-3' (antisense) ([GenBank:S67722]; PCR product 801 base pairs (bp)); GAPDH 5'-CCCATCACCATCTTCCAGGAG-3' (sense) and 5'-GTTGTCATGGATGACCTTGGCC-3' (antisense) ([GenBank:X02231]; PCR product 284 bp).

Following C. albicans infection cells for 12 h were immediately washed with ice-cold PBS containing 100 μM Na₃VO₄ (Sigma) and lysed in situ with ice-cold lysis buffer at 4°C for 15 minutes. Lysis buffer contained 1% Igepal (Sigma), 100 μM Na₃VO₄, and a protease inhibitor cocktail tablet (Roche Diagnostics, Mannheim, Germany). Whole cell lysates were collected after centrifugation at 14,500 rpm for 15 minutes. Protein concentration was determined by the Lowry method. Equal amounts of protein (20 μg) were loaded onto 10% SDS polyacrylamide gels and were transferred to polyvinylidene difluoride (PVDF) membranes (Millipore Immobilon-P, Sigma). Membranes were blocked overnight at 4°C with 2% BSA in TBST. After washing three times with TBST, blots were incubated for 1 h at room temperature with primary antibody (anti-COX-2, 1/1,000 dilution; anti-total ERK1/2, 1/2,000 dilution; anti-phospho-ERK1/2, 1/2,000 dilution) diluted with 2% BSA in TBST. After washing six times with TBST, the blots were then incubated with HRP-labeled secondary antibody (1/1,000 dilution) for 1 h at room temperature. Membranes were rewashed extensively and binding was detected using Enhanced Chemiluminescense western blotting detection system (Amersham Pharmacia Biotech, Piscataway, NJ, USA), according to the manufacturer's instructions. Anti-ERK1/2 and phospho-ERK1/2 antibodies were from Cell Signaling Technology. Mouse monoclonal antibody tubulin Ab-4 (primary antibody, 1/5,000 dilution; secondary antibody, 1/20,000 dilution) (Lab Vision) served as internal control. The band was semiquantified by densitometry using systems as described above.

Cells were infected with 2 × 10⁵ C. albicans at 37°C for 6 h. Nuclear and cytoplasmic extracts of synovial fibroblasts were prepared using NE-PER nuclear and cytoplasmic extraction reagents according to the manufacturer's protocols (Pierce, Rockford, IL, USA). A non-radioactive EMSA was performed using an EMSA kit according to the manufacturer's instructions (Panomics, Redwood City, CA, USA). Nuclear protein (8 μg) was used to bind biotinylated oligonucleotides containing the NFκB binding site for 30 minutes at room temperature. The blank control was nuclear extracts being replaced with water. A competition/cold control was set up by adding non-biotin-labeled cold probes to the reaction. Samples were separated in a non-denaturing polyacrylamide gel (6%, with 2.5% glycerol) and blotted on a Biodyne B Pre-cut Modified Nylon membrane (Pierce). The biotin was labeled with alkaline phosphatase-conjugated streptavidin and alkaline phosphatase was detected with Enhanced Chemiluminescense western blotting detection system (Amersham). The band was semiquantified by densitometry using systems as described above.

Cells were infected with 2 × 10⁵ C. albicans in the presence or absence of U0126 (20 μM, pre-incubation for 2 h) at 37°C for 12 h. The culture supernatant was harvested, and PGE₂, IL1β, and TNFα concentrations were measured by ELISA (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions.

The effect of C. albicans on COX-2 expression by synovial fibroblasts was assessed at the molecular and protein level. Extraction of total RNA from synovial fibroblasts was performed after 12-h co-culture of synovial fibroblasts with different seeding densities of C. albicans and COX-2 induction examined by RT-PCR. Addition of C. albicans to synovial fibroblasts increased COX-2 expression in a dose dependent manner. A significant increase in COX-2 expression over basal conditions was seen at a dose of 2 × 10⁴ yeasts/dish (2.03 ± 0.74-fold increase, P = 0.0185) with no further increase when higher numbers of yeast were added (Figure 1a). The expression of COX-2 protein showed a similar pattern to that of mRNA expression (Figure 1b).

To ascertain whether COX-2 induction was mediated by production of a soluble mediator in the system culture medium was collected from co-cultures of synovial fibroblasts and C. albicans and added directly to non-infected synovial fibroblasts. No change in COX-2 expression was seen. The levels of IL1β and TNFα production were also undetectable (data not shown).

COX-2 expression by proinflammatory cytokines is associated with ERK1/2 and NFκB activation. To establish if similar events were occurring with C. albicans infection of synovial fibroblasts a series of experiments were undertaken to identify whether either ERK1/2 or NFκB were activated under the experimental conditions that result in increased COX-2 expression. The results are shown in Figure 2. Co-incubation of synovial fibroblasts resulted in ERK1/2 activation in a dose dependent manner. Significant levels of ERK1/2 phosphorylation were identified with the addition of C. albicans at doses of 2 × 10⁴ yeasts/dish and above (Figure 2a). Following co-culture of synovial fibroblasts with C. albicans at 2 × 10⁵ yeasts/dish for 6 h, NFκB electrophoretic-mobility shift showed activation of NFκB (Figure 2b).

We next examined whether COX-2 expression was regulated by ERK1/2 activation. Synovial fibroblasts were pretreated with U0126, a MEK1/2 inhibitor, at a concentration of 20 μM for 2 h before addition of C. albicans (2 × 10⁵ yeasts/dish for 12 h) (Figure 3a). C. albicans increased ERK1/2 phosphorylation and COX-2 expression in the absence but not the presence of U0126. U0126 by itself had no effect on COX-2 expression or ERK1/2 phosphorylation. MG132 as an NFκB inhibitor suppressed the COX-2 expression (Figure 3b). Immunohistochemistry (Figure 4) demonstrates increased phospho-ERK1/2 and COX-2 expression in synovial fibroblasts to which C. albicans are adherent. However, the cells without C. albicans attachment demonstrated only very weak positivity. In the presence of U0126 no expression of phospho-ERK1/2 or COX-2 is demonstrable in the infected synovial fibroblasts.

To assess whether increased expression of COX-2 was associated with changes in prostaglandin production levels of PGE₂ released into the media was measured (Figure 5). In the presence of C. albicans infection PGE₂ release into the media was significantly increased over basal levels. This effect of C. albicans was suppressed by the addition of U0126.

To assess whether COX-2 induction was dependent on interactions with the dectin-1 receptor synovial fibroblasts were infected with C. albicans in the presence of laminarin (Figure 6). Laminarin had no effect on levels of synovial fibroblast COX-2 mRNA in the absence of C. albicans. Infection of synovial fibroblasts with C. albicans resulted in a 2 ± 0.3-fold increase in COX-2 gene expression (P < 0.05). In the presence of laminarin there was a lower, 1.6 + 0.3-fold, but significant (P < 0.05) increase in COX-2 gene expression when synovial fibroblasts were infected with C. albicans. This 20% decrease in COX-2 gene expression by laminarin was statistically significantly (P < 0.05, synovial fibroblasts + C. albicans vs synovial fibroblasts + C. albicans + laminarin) (Figure 6a). COX-2 gene expression was significantly upregulated (1.71 ± 0.3-fold increase, P < 0.05) (Figure 6b) when synovial fibroblasts and C. albicans were co-cultured in different trans-well chambers. This indicates that direct contact may have only a minor contribution to the elevation of COX-2 gene expression seen when synovial fibroblasts are infected with C. albicans.