Differential expression, function and response to inflammatory stimuli of 11beta-hydroxysteroid dehydrogenase type 1 in human fibroblasts: a mechanism for tissue-specific regulation of inflammation

Abstract

Stromal cells such as fibroblasts play an important role in defining tissue-specific responses during the resolution of inflammation. We hypothesized that this involves tissue-specific regulation of glucocorticoids, mediated via differential regulation of the enzyme 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1). Expression, activity and function of 11beta-HSD1 was assessed in matched fibroblasts derived from various tissues (synovium, bone marrow and skin) obtained from patients with rheumatoid arthritis or osteoarthritis. 11beta-HSD1 was expressed in fibroblasts from all tissues but mRNA levels and enzyme activity were higher in synovial fibroblasts (2-fold and 13-fold higher mRNA levels in dermal and synovial fibroblasts, respectively, relative to bone marrow). Expression and activity of the enzyme increased in all fibroblasts following treatment with tumour necrosis factor-alpha or IL-1beta (bone marrow: 8-fold and 37-fold, respectively, compared to vehicle; dermal fibroblasts: 4-fold and 14-fold; synovial fibroblasts: 7-fold and 31-fold; all P < 0.01 compared with vehicle). Treatment with IL-4 or interferon-gamma was without effect, and there was no difference in 11beta-HSD1 expression between fibroblasts (from any site) obtained from patients with rheumatoid arthritis or osteoarthritis. In the presence of 100 nmol/l cortisone, IL-6 production--a characteristic feature of synovial derived fibroblasts--was significantly reduced in synovial but not dermal or bone marrow fibroblasts. This was prevented by co-treatment with an 11beta-HSD inhibitor, emphasizing the potential for autocrine activation of glucocorticoids in synovial fibroblasts. These data indicate that differences in fibroblast-derived glucocorticoid production (via the enzyme 11beta-HSD1) between cells from distinct anatomical locations may play a key role in the predeliction of certain tissues to develop persistent inflammation.

Figure 1

Tissue-specific expression of 11β-HSD1 is an autocrine determinant of cortisol levels in fibroblasts. Results are shown for cultures of dermal (DF), and bone marrow (BF) and synovial fibroblasts (SF). (a) Reductase activity determined by scanning thin layer chromatography (³H-cortisone to ³H-cortisol conversion, 12-hour incubation) following treatment with IL-1 (10 ng/ml, 24 hours) or vehicle. (b) Accumulation of cortisol in cell culture medium (as determined by specific ELISA) from 100 nmol/l cortisone following treatment with TNF-α (10 ng/ml, 24 hours) or vehicle. Values are expressed as mean ± standard deviation of four replicates from a representative culture of RA fibroblasts. Similar values were obtained when the assays were carried out using two other fibroblast cells lines. *P < 0.01, **P < 0.001 versus vehicle-treated cells; ^#P < 0.01 versus equivalent bone marrow fibroblasts; ^&P < 0.01 versus equivalent dermal fibroblasts. ELISA, enzyme-linked immunosorbent assay; IL, interleukin; TNF, tumour necrosis factor; u.d., undetectable.

Figure 2

Autocrine activation of cortisol and the regulation of fibroblast function. Bone marrow, dermal and synovial fibroblasts were treated with vehicle (C), cortisol (F), or cortisone (E; both at 100 nmol/l) in the presence or absence of the 11b-HSD1 inhibitor glycyrrhetinic acid (+ G; 5 μmol/l) for 24 hours. Cells were then assessed expression of IL-6 mRNA and protein. (a) For mRNA analyses, target gene data were normalized for levels of the housekeeping gene 18S rRNA and presented as fold change in expression relative to vehicle-treated fibroblasts. (b) Analysis IL-6 protein secretion in synovial fibroblasts was carried out using a specific ELISA assay and reported as pg IL-6/mg cell protein/24 hours. Values are expressed as mean ± standard deviation of four replicates from a representative culture of rheumatoid arthritis fibroblasts. Similar values were obtained when the assays were carried out using two other fibroblast cells lines. *P < 0.01, **P < 0.001 versus vehicle-treated cells (statistical analysis carried out on unmodified ΔCt values). Ct = the cycle number at which logarithmic PCR plots cross a calculated threshold line; ELISA, enzyme-linked immunosorbent assay; 11β-HSD = 11β-hydroxysteroid dehydrogenase.

Figure 3

Expression of 11β-HSD1 in synovial tissue. Co-localization of 11β-HSD1 with markers of endothelial cells (vWF), fibroblasts (ASO2/CD90), T-cells (CD3), dendritic cells (CD11c) and lining macrophages (lining mac; CD68). Fluorescence immunohistochemistry was carried out using green, red and blue fluorescent tagged antiserum as shown in each panel. In each case the scale bar is 20 μm. A indicates co-expression of 11β-HSD1 and ASO2 (CD90) in fibroblasts; B highlights co-expression of 11β-HSD1 and vWF in endothelial cells. 11β-HSD, 11β-hydroxysteroid dehydrogenase type 1; vWF, von Willebrand factor.

Figure 4

Site-specific variations in human fibroblastic 11β-HSD1 expression versus other components of glucocorticoid metabolism and signalling. Expression of mRNA for (a) 11β-HSD1, (b) H6PDH, (c) GRα and (d) GRβ in bone marrow (B), and dermal (D) and synovial (S) fibroblasts in the presence or absence of TNF-α (10 ng/ml). For each gene product data were normalized for levels of the housekeeping gene 18S rRNA and are presented as fold change in expression relative to vehicle-treated bone marrow fibroblasts. *P < 0.01 versus vehicle control; ^#P < 0.01 versus equivalent bone marrow fibroblasts; ^&P < 0.01 versus equivalent dermal fibroblasts. 11β-HSD, 11β-hydroxysteroid dehydrogenase type 1; GR, glucocorticoid receptor; H6PDH, hexose-6-phosphate dehydrogenase; TNF, tumour necrosis factor.

Figure 5

Induction of 11β-HSD1 expression is specific for inflammatory cytokines. (a) Bone marrow, and (b) dermal and (c) synovial fibroblasts (n = 3 for each) were cultured with or without TNF-α (10 ng/ml), IL-1 (10 ng/ml), IFN-γ (100 iU), or IL-4 (10 ng/ml) for 24 hours. Total RNA from these cells was then used to assess expression of mRNA for 11β-HSD1. In each case mRNA levels for 11β-HSD1 were normalized for the housekeeping gene 18S rRNA and presented as fold change (mean ± standard deviation) in expression relative to vehicle treated bone marrow, dermal, or synovial fibroblasts. *P < 0.05, **P < 0.01 versus vehicle-treated cells. 11β-HSD, 11β-hydroxysteroid dehydrogenase type 1; C, control; IFN, interferon; IL, interleukin; TNF, tumour necrosis factor.

Figure 6

Expression of 11β-HSD1 in fibroblasts correlates with C/EBPα and C/EBPβ. (a) Correlation between ΔCt values for C/EBPα and C/EBPβ with ΔCt values for 11β-HSD1 in all fibroblast populations. (b) Fold change in C/EBP expression in bone marrow and synovial fibroblasts relative to dermal fibroblasts (mean ± standard deviation). *P < 0.01 versus dermal and bone marrow fibroblasts (statistical analysis carried out on ΔCt values). C/EBP = CCAAT/enhancer binding protein; Ct = the cycle number at which logarithmic PCR plots cross a calculated threshold line; 11β-HSD = 11β-hydroxysteroid dehydrogenase.