Local and systemic glucocorticoid metabolism in inflammatory arthritis

Abstract

Background: Isolated, primary synovial fibroblasts generate active glucocorticoids through expression of 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1). This enzyme produces cortisol from inactive cortisone (and prednisolone from prednisone).

Objective: To determine how intact synovial tissue metabolises glucocorticoids and to identify the local and systemic consequences of this activity by examination of glucocorticoid metabolism in patients with rheumatoid arthritis (RA).

Methods: Synovial tissue was taken from patients with RA during joint replacement surgery. Glucocorticoid metabolism in explants was assessed by thin-layer chromatography and specific enzyme inhibitors. RT-PCR and immunohistochemistry were used to determine expression and distribution of 11beta-HSD enzymes. Systemic glucocorticoid metabolism was examined in patients with RA using gas chromatography/mass spectrometry.

Results: Synovial tissue synthesised cortisol from cortisone, confirming functional 11beta-HSD1 expression. In patients with RA, enzyme activity correlated with donor erythrocyte sedimentation rate (ESR). Synovial tissues could also convert cortisol back to cortisone. Inhibitor studies and immunohistochemistry suggested this was owing to 11beta-HSD2 expression in synovial macrophages, whereas 11beta-HSD1 expression occurred primarily in fibroblasts. Synovial fluids exhibited lower cortisone levels than matched serum samples, indicating net local steroid activation. Urinary analyses indicated high 11beta-HSD1 activity in untreated patients with RA compared with controls and a significant correlation between total body 11beta-HSD1 activity and ESR.

Conclusions: Synovial tissue metabolises glucocorticoids, the predominant effect being glucocorticoid activation, and this increases with inflammation. Endogenous glucocorticoid production in the joint is likely to have an impact on local inflammation and bone integrity.

Figure 1. Glucocorticoid metabolism in synovial explants taken from patients with rheumatoid arthritis (RA) and osteoarthritis (OA). The ability of synovial tissue to interconvert cortisone and cortisol was examined using tissue freshly isolated after joint replacement surgery. Data are steroid generation for each sample adjusted for protein content and assay time and are expressed as mean (SD) of samples from each patient.

Figure 2. Relationship between glucocorticoid activating capacity of synovial explants and the systemic inflammatory response and the effect of a specific inhibitor of 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) on glucocorticoid activation and inactivation. (A) A significant correlation was seen between glucocorticoid activation (oxoreductase activity) and the erythrocyte sedimentation rate (ESR) measured before surgery in patients with rheumatoid arthritis (RA). (B) The effect of PF-877423, a specific bidirectional inhibitor of 11β-HSD1 enzyme activity, on glucocorticoid activation (oxoreductase activity, cortisone to cortisol conversion) and inactivation (dehydrogenase activity, cortisol to cortisone conversion) in synovium from patients with RA or osteoarthritis (OA). The inhibitor reduced the capacity of synovium to activate glucocorticoids but was unable to block the inactivating capacity. In OA tissue inactivation increased with 11β-HSD1 inhibition. These results indicate that the inactivating capacity is not due to 11β-HSD1.

Figure 3. Immunohistological localisation of 11β-hydroxysteroid dehydrogenase (11β-HSD) enzymes in synovium. The expression of 11β-HSD1 and 11β-HSD2 within rheumatoid synovium were examined using specific antibodies. 11β-HSD1 expression colocalised predominantly with a fibroblast marker. Expression of 11β-HSD2 colocalised with expression of CD68.

Figure 4. The functional effect of glucocorticoid activation on interleukin 6 (IL6) synthesis in patients with rheumatoid arthritis (RA) and osteoarthritis (OA). (A) Synovial tissue was incubated for 24 h in the presence or absence of 100 nM cortisone. In both RA and OA synovium cortisone had a significant suppressive effect on IL6 production. (B) A specific 11β-HSD1 inhibitor (PF-877423) was able to block the decrease in IL6 production caused by cortisone treatment.

Figure 5. Systemic measures of local glucocorticoid metabolism in patients with rheumatoid arthritis (RA) and non-inflammatory joint disease. (A, B) Measurement of the balance of glucocorticoid activation/inactivation in non-inflammatory controls and patients with untreated RA. There was a significant increase in the tetrahydrocortisol + allo-tetrahydrocortisol/tetrahydrocortisone (THF+alloTHF)/THE ratio in patients with inflammatory arthritis, whereas the urinary free cortisol/cortisone ratio was unchanged. This indicated that 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1)-mediated glucocorticoid activation was enhanced in patients with RA. (C, D) In patients with RA the relationship between the (THF+alloTHF)/THE ratio, a systemic measure of 11β-HSD1 activity, and inflammatory markers was examined. The erythrocyte sedimentation rate (ESR), but not C-reactive protein (CRP), demonstrated a significant correlation with this measure, suggesting increasing 11β-HSD1-mediated glucocorticoid activation with increasing degrees of inflammation.