Severity of murine collagen-induced arthritis correlates with increased CYP7B activityEnhancement of dehydroepiandrosterone metabolism by interleukin-1.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 10, October 2004, pp 3346–3353 DOI 10.1002/art.20509 © 2004, American College of Rheumatology Severity of Murine Collagen-Induced Arthritis Correlates With Increased CYP7B Activity Enhancement of Dehydroepiandrosterone Metabolism by Interleukin-1␤ John Dulos, Evert Verbraak, Wilma M. Bagchus, Annemieke M. H. Boots, and Allard Kaptein sulted in a dose-dependent increase in 7␣-OH-DHEA formation. In addition, IL-1␤ enhanced CYP7B mRNA and CYP7B protein levels in FLS. Conclusion. Disease progression in CIA is correlated with enhanced CYP7B activity, which leads to locally enhanced 7␣-OH-DHEA levels. Elevated IL-1␤ levels within the arthritic joint may regulate this increase in CYP7B activity. Objective. The endogenous steroid dehydroepiandrosterone (DHEA) has been reported to play a role in rheumatoid arthritis (RA). DHEA is metabolized by the P450 enzyme CYP7B into 7␣-OH-DHEA, which has immunostimulating properties. This study was undertaken to investigate the putative role of CYP7B in arthritis using murine collagen-induced arthritis (CIA), an interleukin-1␤ (IL-1␤)–dependent model. Methods. DBA/1J mice were immunized and administered a booster with type II collagen. The presence of 7␣-OH-DHEA was determined in both arthritic and nonarthritic joints and the serum of CIA mice by radioimmunoassay. CYP7B messenger RNA (mRNA) expression was analyzed in synovial biopsy samples, and in fibroblast-like synoviocytes (FLS) isolated from these synovial biopsy samples, by reverse transcriptase– polymerase chain reaction (RT-PCR). In addition, the regulatory role of IL-1␤ on CYP7B activity in FLS was determined using RT-PCR, Western blotting, and highperformance liquid chromatography. Results. In knee joint synovial biopsy samples from arthritic mice, 7␣-OH-DHEA levels were 5-fold higher than in nonarthritic mice. Elevated levels of 7␣-OH-DHEA were accompanied by an increase in CYP7B mRNA expression and were positively correlated with disease severity. In serum, no differences in 7␣OH-DHEA levels were observed between arthritic and nonarthritic mice. Incubation of FLS with IL-1␤ re- Rheumatoid arthritis (RA) is characterized by chronic inflammation of the joints, ultimately resulting in destruction of cartilage and bone. The pathogenesis of RA is still unknown. Mediators of the endocrine system, such as dehydroepiandrosterone (DHEA), androgen, estrogen, and glucocorticoids, are all known to influence the disease process in RA (1–3). Among these endocrine mediators, DHEA and its sulfate form, DHEAS, are the most abundant prohormones formed in vivo (4) and can be found in blood and synovial fluid (2). Both immunosuppressive (5–7) and immunostimulating (8–10) properties have been described for DHEA. The immunosuppressive effects may be explained by the conversion of DHEA into androgens and/or estrogens (1,2,11). The described immunostimulating effects of DHEA may be due to the conversion of DHEA into 7␣-OH-DHEA (9,12). Recent data suggest that 7␣-OH-DHEA acts as an antiglucocorticoid that can block the glucocorticoid-induced immunosuppression (13,14). It has been suggested that in inflammatory diseases the local balance between DHEA metabolites and endogenous glucocorticoids is disturbed (14). To date, 3 P450 7␣-hydroxylase isoenzymes are known, CYP7A, CYP39A, and CYP7B. CYP7A (15) is expressed in the liver and CYP39A1 (16) is expressed in the brain; both are involved in the elimination of cholesterol. CYP7B is not restricted to the liver or the brain John Dulos, MSc, Evert Verbraak, BSc, Wilma M. Bagchus, PhD, Annemieke M. H. Boots, PhD, Allard Kaptein, PhD: Organon NV, Oss, The Netherlands. Address correspondence and reprint requests to John Dulos, MSc, Department of Pharmacology, Section AutoImmunity, Organon NV, p.b. 20, Oss 5340 BH, The Netherlands. E-mail: [email protected] organon.com. Submitted for publication March 9, 2004; accepted in revised form June 4, 2004. 3346 CORRELATION OF CIA WITH CYP7B ACTIVITY and is expressed in various tissues and cells, including the thymus and lymphocytes (17). The DHEA metabolite 7␣-OH-DHEA and, to a lesser extent, 7␤-OHDHEA are formed by the activity of the cytochrome P450 enzyme CYP7B (18–20). We propose that enhanced CYP7B activity, resulting in enhanced levels of 7␣-OH-DHEA, might play a proinflammatory role in the disease process. Increased CYP7B activity is thought to result in increased 7␣-OH-DHEA levels but may also result in a decline in DHEA levels as observed in inflammatory diseases such as RA (11,21,22). To investigate the possible involvement of CYP7B in the arthritis process, the murine collageninduced arthritis (CIA) model was studied. First, we determined the level of 7␣-OH-DHEA in both serum and arthritic joints of mice with CIA. Second, expression levels of CYP7B (responsible for the conversion of DHEA into 7␣-OH-DHEA) were studied in synovial biopsy samples from mice with CIA and in fibroblastlike synoviocytes (FLS) derived from these biopsy samples. Third, regulatory properties of the proinflammatory cytokine interleukin-1␤ (IL-1␤) were studied with respect to the regulation of expression of CYP7B messenger RNA (mRNA) and CYP7B protein in FLS. MATERIALS AND METHODS Murine CIA. Male DBA1/J mice were obtained from Bomholtgård (Ry, Denmark). Animals were routinely maintained at 23°C with water and food provided ad libitum. Mice were immunized at the base of the tail on day 0 (at the age of 8 weeks) with 100 g bovine type II collagen (CII) in Freund’s complete adjuvant enriched with 2 mg/ml Mycobacterium tuberculosis H37Ra. On day 21, the animals were administered a booster with an intraperitoneal injection of 100 g CII dissolved in saline. Animals were euthanized and at the end of the experiment (days 35–47), knee synovial biopsy samples, hind paws, and serum samples were obtained as previously described (23). For histopathologic analysis, knee joints and knee synovial biopsy samples were fixed in 4% formaldehyde and processed further, as previously described (23). Knee joints were also used for radiographic analysis, as previously described (23). To analyze 7␣- and 7␤-OH-DHEA levels by radioimmunoassay (RIA), individual hind paws (with the skin intact) and pooled knee synovial biopsy samples were isolated (23), frozen in liquid nitrogen, and stored at ⫺70°C. Detection of 7␣- and 7␤-OH-DHEA levels by RIA. To determine 7␣- and 7␤-OH-DHEA levels in hind paws and knee synovial biopsy samples, an RIA was performed using antiserum against 7␣-OH-DHEA and 7␤-OH-DHEA. These metabolites are formed by the activity of the CYP7B enzyme. The RIA was performed at the Institute of Endocrinology at Prague (Czech Republic) in cooperation with Dr. R. Hampl (24). After pulverization in a microdismembrator, the hind paws and knee synovial biopsy tissue powder were weighed and ultrapure water (Milli-Q; Millipore, Bedford, MA) was added. 3347 Levels of 7␣-OH-DHEA and 7␤-OH-DHEA were measured by RIA in extracts of hind paws from 3 animals, pooled knee joint synovial biopsy samples (30 joints) with an arthritis score of 0 (103 mg total weight), pooled knee joint biopsy samples (30 joints) with an arthritis score of 1–2 (116 mg total weight), and serum extracts from 3 animals, as previously described (24). Extraction was carried out using diethyl ether. Diethyl ether extracts containing 7␣-OH-DHEA and 7␤-OH-DHEA were evaporated with nitrogen, and the dry residue was dissolved in assay buffer and measured by RIA, as previously described (24). FLS isolation from CIA joints, culturing, and IL-1␤ incubation. Knee joint biopsy samples were isolated from mice with severe arthritis, as previously described (23). The synovial tissue biopsy samples were digested with collagenase, and the mixture of cells was filtered through a fine sieve (200 m; NPBI, Amsterdam, The Netherlands). After overnight culturing, nonadherent lymphoid cells were removed and adherent FLS were cultured in medium (Dulbecco’s modified Eagle’s medium–Ham’s F-12 plus 10% fetal calf serum [FCS]; Life Technologies, Breda, The Netherlands). Adherent FLS between passages 3 and 8 were used. Adherent FLS were harvested with 0.05% trypsin/EDTA (Flow, Irvine, UK). FLS (1 ⫻ 105) were cultured in the presence or absence of IL-1␤ and cycloheximide in X-Vivo 15 medium (BioWhittaker Europe, Verviers, Belgium). FLS were also used for total RNA or protein isolation or to determine CYP7B activity by measuring 7␣-OH-DHEA by high-performance liquid chromatography (HPLC). CYP7B activity in FLS. FLS derived and cultured by CIA were incubated for 24 hours with IL-1␤ and 1,2,6,7-3HDHEA (1 ⫻ 10⫺8M; NEN Life Science Products, Boston, MA) in X-Vivo 15 medium. Steroid-containing medium (1 ml) was collected and passed over a C-18 SPE cartridge (Sopachem, Woerden, The Netherlands) to determine the conversion of 1,2,6,7-3H-DHEA into 3H-labeled 7␣-OH-DHEA, representing CYP7B activity. Steroids were eluted from the column with 300 l methanol. Next, 3H-labeled 7␣-OH-DHEA and 3HDHEA were measured by HPLC. The amount of 7␣-OHDHEA is expressed as the percentage of 3H-labeled 7␣-OHDHEA of the total amount of 3H-label measured. Recoveries after extraction were in the range of 85–95%. Identification of 7␣-OH-DHEA was confirmed by gas chromatography mass spectrometry (data not shown). Western blotting for CYP7B in FLS microsomes. FLS were collected in ice-cold phosphate buffered saline containing 1 mM EDTA and protease inhibitors (Roche, Almere, The Netherlands). Homogenization was carried out at 4°C with 10 strokes in a Potter-Elvehjem homogenizer (New Brunswig, Nijmegen, The Netherlands) set at 600 revolutions per minute. The homogenate was centrifuged at 800g for 10 minutes, after which the supernatant was used. With supernatant, a second centrifugation for 20 minutes at 11,000g (4°C) was performed to remove cell debris, nuclei, and mitochondria. The last supernatant was centrifuged at 150,000g for 60 minutes. The microsomal pellet was suspended in buffer containing 0.1M sodium pyrophosphate and 1 mM EDTA (pH 7.4). The amount of protein was determined by bicinchoninic acid (BCA) assay, and the microsomal fraction was stored at ⫺70°C until further use. Amido black staining was performed as an extra control, in addition to the BCA assay, to determine the 3348 amount of protein loaded. Microsomal lysate (5 g protein) was fractionated on Tris-buffered 10% polyacrylamide ready gel (Bio-Rad, Veenendaal, The Netherlands) and transferred to a polyvinylidene difluoride membrane (Bio-Rad). Membranes were blocked with blocking buffer (1 gm/liter bovine serum albumin, 1% FCS, 5% Tween 20, 1 mole/liter NaCl in 100 mmoles/liter Tris HCl, pH 7.4) for 30 minutes and incubated overnight (4°C) with a polyclonal rabbit anti-mouse CYP7B antibody. Detection was performed by chemiluminescence following anti-rabbit peroxidase incubation. The antibody used was raised by us against a peptide sequence from mouse CYP7B (amino acids 266–281). The peptide used was identical to the peptide used by Schwarz et al for antibody generation (20). CYP7B and IL-1␤ mRNA levels. To investigate the presence of CYP7B and IL-1␤ mRNA in synovial biopsy samples, total RNA was isolated in RNAzol (Campro, Veenendaal, The Netherlands). FLS were lysed in RNAzol (1 ⫻ 106 cells per 1 ml RNAzol) and the suspension was then collected by scraping the bottom of the plate. RNA was isolated by chloroform and water separation. Synthesis of complementary DNA (cDNA) was carried out according to the manufacturer’s protocol using random hexa primers (Pharmacia, Piscataway, NJ) and reverse transcriptase (Superscript RT; Pharmacia). Semiquantitative RT–polymerase chain reaction (PCR) was performed by removing aliquots (20 l) of the PCR product after 25, 30, and 35 cycles. The PCR product was analyzed on 1.5% agarose gel for 1 hour at 180V. As primers, the following were used: for GAPDH, CCCTTCATTGACCTCAACTACATGG (forward) and GGTCCACCACCCTGTTGCTGTAGCC (reverse); for mouse CYP7B, GCCCTCTTTCCTCCACTCA (forward) and TGAGCTTCTCCAAGATTTTGC (reverse); for mouse IL-1 ␤ , GCAACTGTTCCTGAACTCA (forward) and CTCGGAGCCTGTAGTGCAG (reverse). RESULTS Arthritis score. To investigate the role of CYP7B in arthritis, the CIA model in DBA/1J mice was used. Eighty-five percent of the DBA/1J mice developed CIA after immunization and booster with bovine CII. To quantify the severity of knee joint inflammation, an arthritis scoring system for knee joints of mice with CIA was used as previously described (25). For this purpose, the skin was removed and the underlying tissue of the knee was scored macroscopically. In mice with clinical signs of arthritis (scores 1–2), both increased swelling and newly formed blood vessels were observed (Figures 1B and C). In addition, histologic analysis of knee synovial biopsy samples showed no signs of inflammation in animals without clinical signs of arthritis (score 0) but did show large infiltrates of inflammatory cells in animals with a clinical score of 1–2 (Figures 1D–F). Increased 7␣-OH-DHEA levels in arthritic joints. We studied 7␣- and 7␤-OH-DHEA levels in knee synovial biopsy samples from arthritic versus nonar- DULOS ET AL Figure 1. Macroscopic and histopathologic scoring of arthritis. Macroscopic appearance of knee joints with no arthritis (A) and with arthritis scores of 1 (B) and 2 (C). Synovial biopsy samples were isolated and processed for histopathologic study. Samples exhibit no arthritis (D), or arthritic changes scored as 1 (E) and 2 (F). Arrows in A–C indicate the sites from which the samples shown in D–E, respectively, were obtained. thritic CII-immunized mice. A 5-fold increase in 7␣-OHDHEA and a 3-fold increase in 7␤-OH-DHEA levels was observed in knee joint synovial biopsy samples from arthritic mice versus nonarthritic mice (Figure 2). In addition, the hind paws of arthritic mice showed a significantly (3-fold) increased 7␣-OH-DHEA level, whereas the level of 7␤-OH-DHEA was nearly identical in arthritic and nonarthritic hind paws (Figure 2). Interestingly, both the 7␣-OH-DHEA and the 7␤-OHDHEA levels in the knee synovial biopsy samples were higher than the levels in the hind paws, suggesting enhanced CYP7B activity at the site of inflammation. No major changes were observed when serum levels in 7␣-OH-DHEA and 7␤-OH-DHEA in arthritic mice were compared with those of nonarthritic mice. Enhanced CYP7B mRNA levels in knee synovial biopsy samples. Figure 3 shows the clinical evaluation of the hind paws from arthritic and nonarthritic mice. In addition, radiographic and histopathologic findings in the knee joints of arthritic versus nonarthritic mice are shown. Swelling was evident in the hind paws of CIA mice (Figures 3A and B) and was associated with bone erosion in knee joints, as observed by radiographic analysis (Figure 3D). Besides bone erosions, we found histopathologic evidence of inflammatory infiltrates and cartilage destruction in the knee joints of arthritic mice (Figure 3F). As expected, inflammatory infiltrates and cartilage destruction were absent in nonarthritic but CII-immunized mice (Figure 3E). To investigate whether a local increase in 7␣-OHDHEA levels resulted from an increase in CYP7B levels, CYP7B mRNA expression levels in knee synovial biopsy CORRELATION OF CIA WITH CYP7B ACTIVITY 3349 synovial biopsy samples from knees with different arthritis scores was used. IL-1␤ mRNA levels were measured, in addition to CYP7B, as a reference cytokine for inflammation in murine CIA (26). GAPDH was used as an internal control to check for possible differences in the amount of RNA per sample. The mRNA level of GAPDH was compared with those of CYP7B and IL-1␤, Figure 2. Levels of 7␣-OH-dehydroepiandrosterone (7␣-OH-DHEA) and 7␤-OH-DHEA in serum, hind paws, and knee joint synovial biopsy samples from arthritic and nonarthritic type II collagen (CII)– immunized mice. DBA1/J mice were immunized on day 0 and administered a booster of CII on day 21. Levels of A, 7␣-OH-DHEA and B, 7␤-OH-DHEA were measured in extracts of hind paws from 3 animals, pooled knee joint synovial biopsy samples, and serum extracts from 3 animals, by radioimmunoassay. Thirty biopsy samples with a knee score of 0 (103 mg total weight) and 30 biopsy samples with a knee score of 1–2 (116 mg total weight) were pooled. Hind paws and knee joints of arthritic mice with a knee score of 1–2 were compared with those of nonarthritic mice (knee score 0). The level of 7␣- and 7␤-OH-DHEA in joint synovial biopsy samples and hind paws is shown as nM, normalized to 7␣- or 7␤-OH-DHEA concentration per 100 mg tissue. The concentration of 7␣- and 7␤-OH-DHEA in serum is shown as nM. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus nonarthritic mice, by Student’s t-test. samples from arthritic mice were compared with those from nonarthritic mice. A marked increase of CYP7B mRNA was found in the inflamed synovium of arthritic mice (Figure 3H versus Figure 3G). This elevated CYP7B mRNA level in knee synovial biopsy samples from arthritic mice, together with increased CYP7B activity, correlated with CIA disease progression. Correlation of arthritis scores with increases in synovial CYP7B and IL-1␤ mRNA expression. To investigate whether there is a relationship between the joint score and the level of CYP7B mRNA, cDNA from Figure 3. CYP7B mRNA expression in arthritic and nonarthritic mice. DBA1/J mice were immunized on day 0 and administered a booster of type II collagen on day 21. Mice were thereafter divided into an arthritic group and a nonarthritic group. A and B, Paw swelling, C and D, radiographic features, and E and F, histopathologic features were assessed as described in Materials and Methods. Findings in nonarthritic (control: score 0) mice and mice with collagen-induced arthritis (score 2), representative of 5 independent experiments, are shown. Arrows indicate areas of bone destruction. Inflammatory infiltrates in synovium (S) and joint space (JS) and cartilage destruction in arthritic and nonarthritic mice were compared. Individual synovial biopsy samples were obtained and semiquantitative polymerase chain reaction (PCR) was performed by removing aliquots (20 l) of the PCR product after increasing the number of cycles (25, 30, and 35) in which the CYP7B signal was compared with the GAPDH signal. Boxes in E and F indicate the areas from which the biopsy samples were obtained. Knee synovial biopsy cDNA from G, nonarthritic mice was compared with cDNA from H, mice with severe arthritis, using CYP7B- and GAPDH-specific primers. Reverse transcriptase–PCR results are representative of 3 independent experiments in which, per experiment, 1 synovial biopsy sample of score 0 and 3 synovial biopsy samples of score 2 were used. P ⫽ patella; F ⫽ femur; Ti ⫽ tibia; Fi ⫽ fibula; C ⫽ cartilage. 3350 using the same amount of cDNA. No signal or only a weak signal for CYP7B and IL-1␤ mRNA was observed in nonarthritic mice (score 0) (Figure 4). When arthritis developed, an increase in the arthritis score of the knee joints (score 1, 1.5, 2) was associated with a gradual increase in both CYP7B and IL-1␤ mRNA levels in the synovium (Figure 4). IL-1␤-induced increase of CYP7B activity. Apart from the analysis of 7␣-OH-DHEA and CYP7B mRNA in vivo, the presence and activity of CYP7B in cells isolated from knee synovial biopsy samples was examined. For this purpose, FLS, which are increased in the synovium in arthritis, were isolated from knee synovial biopsy samples and cultured (passages 3–8) in vitro. In addition, FLS were incubated with the cytokine IL-1␤ to determine the effect of IL-1␤ on the activity of CYP7B in the formation of 7␣-OH-DHEA. The conversion of DHEA into 7␣-OH-DHEA was observed in cultured mouse FLS. Following stimulation with IL-1␤ (1, 10, or 100 pg/ml), a dose-dependent increase in the formation of 7␣-OH-DHEA occurred, indicating enhanced CYP7B activity (Figure 5A). To investigate whether the IL-1␤–dependent increase in CYP7B activity resulted from enhanced CYP7B protein levels in FLS, Western blot analysis (using our CYP7B peptide–specific antibody) was carried out on the microsome fraction of FLS known to contain CYP7B. FLS cultured without IL-1␤ showed a weak CYP7B protein signal (Figure 5B). Consistent with the increased CYP7B activity, IL-1␤ (100 pg/ml) stimulation of FLS resulted in an increase in CYP7B protein (Figure 5B). When studied at the mRNA level, IL-1␤ increased the CYP7B transcript in a dose-dependent manner (results not shown). Thus, the cytokine IL-1␤ dose-dependently increased 7␣-OH-DHEA formation, which implied that Figure 4. Correlation between knee joint score and CYP7B and interleukin-1␤ (IL-1␤) mRNA levels in collagen-induced arthritis. Synovial biopsy samples from 5 mice were pooled for each knee joint score. RNA was isolated, cDNA was made, and reverse transcriptase– polymerase chain reaction was performed with GAPDH-, CYP7B-, and IL-1␤–specific primers, as described in Materials and Methods. Results shown are representative of 3 experiments with 3 mice (2 synovial biopsy samples per mouse). DULOS ET AL Figure 5. Dose-dependent increase of CYP7B activity and CYP7B protein expression in fibroblast-like synoviocytes (FLS) with interleukin-1␤ (IL-1␤) treatment. Mouse knee FLS were stimulated for 24 hours in the presence or absence of 1, 10, or 100 pg/ml IL-1␤ and with or without 1.5 ⫻ 10⫺8M 3H-dehydroepiandrosterone (DHEA). Supernatant and cells were processed by high-performance liquid chromatography (A) and Western blot analysis (B), respectively. Results in A are the mean and SEM from 2 independent experiments done in triplicate for each condition. The percentages of 3H-labeled 7␣-OH-DHEA formed in the incubations without IL-1␤ were 2.4 ⫾ 0.6 and 8.0 ⫾ 0.6 in the 2 experiments, respectively, those in incubations with 1 pg/ml IL-1␤ were 4.4 ⫾ 0.3 (P ⱕ 0.05) and 6.3 ⫾ 0.4 (P ⱕ 0.05), those in incubations with 10 pg/ml were 6.2 ⫾ 0.2 (P ⱕ 0.01) and 11.1 ⫾ 2.3 (P ⱕ 0.08), and those in incubations with 100 pg/ml were 10.7 ⫾ 0.4 (P ⱕ 0.0005) and 18.1 ⫾ 0.79 (P ⱕ 0.0003) (P values versus medium control, by Student’s t-test). Western blot analysis was performed with 5 g protein of stimulated (100 pg/ml IL-1␤) and unstimulated FLS microsomes, as described in Materials and Methods. The apparent molecular weight of CYP7B is 58 kd (B). IL-1␤ can enhance the expression of CYP7B at both the mRNA and the protein levels. Kinetics of CYP7B mRNA expression after IL-1␤ stimulation. RT-PCR experiments were performed to investigate the kinetics of IL-1␤–induced CYP7B mRNA levels. A weak CYP7B mRNA signal was found CORRELATION OF CIA WITH CYP7B ACTIVITY Figure 6. Time-dependent increase in CYP7B mRNA levels. Murine fibroblast-like synoviocytes were left untreated or were stimulated with interleukin-1␤ (IL-1␤; 100 pg/ml) in the presence or absence of cycloheximide (CHX; 15 g/ml). Reverse transcriptase–polymerase chain reaction was performed with murine CYP7B- and GAPDHspecific primers (35 cycles). in unstimulated FLS (Figure 6). Stimulation with 100 pg/ml IL-1␤ for 2 hours resulted in a strong increase in CYP7B mRNA levels, which was maintained after the incubation of FLS for 4, 8, and 24 hours. Using the specified PCR amplification cycles, IL-1␤ did not affect the GAPDH mRNA level (Figure 6). Cycloheximide was used to analyze whether the induction of CYP7B mRNA by IL-1␤ is dependent on de novo protein synthesis. The results presented in Figure 6 show that cycloheximide did not affect the IL-1␤–induced increase in CYP7B mRNA levels in murine FLS. Therefore, the IL-1␤–induced increase in CYP7B production was caused by either a direct effect of IL-1␤ on CYP7B mRNA expression or an indirect effect dependent on a protein in the resident pool. DISCUSSION Our study is the first to show that the severity of murine CIA is correlated with increased CYP7B mRNA levels in the joint. As a result of increased CYP7B activity in the joint, increased formation of immunostimulating 7␣-OH-DHEA was found. Moreover, IL-1␤, an important proinflammatory cytokine in arthritis, was found to positively regulate CYP7B activity. These findings suggest a role for CYP7B in arthritis pathology. Our data show a disease-dependent increase in the amount of 7␣-OH-DHEA and, to a lesser extent, 7␤-OH-DHEA in the affected joints of arthritic mice. The increase in 7␣-OH-DHEA proved to be a local phenomenon since enhanced levels of 7␣-OH-DHEA were not found in the serum of these mice. The observed low serum and synovial biopsy levels of 7␤-OH-DHEA, when compared with 7␣-OH-DHEA, suggest that 7␣- 3351 OH-DHEA is the main CYP7B-catalyzed metabolite formed (18–20). The importance of 7␣-OH-DHEA as an immunostimulating metabolite has been reported before (9,12). Moreover, data suggest that 7␣-OH-DHEA shows antiglucocorticoid activity (13,14,27). Chmielewski and colleagues demonstrated that dexamethasoneinduced thymocyte apoptosis was counteracted by 7␣OH-DHEA but not by 7␤-OH-DHEA (13). Recently, it has been proposed that glucocorticoids and 7␣-OHDHEA, by virtue of the hydroxy group at positions 11␤ and 7␣, respectively, display rotational symmetry and thus may bind to the glucocorticoid receptor (17). In addition, it was suggested that autoimmunity could result from elevated antiglucocorticoid 7␣-OH-DHEA levels (14). Since antiglucocorticoid 7␣-OH-DHEA might have stimulatory effects on the inflammatory process, studies of the administration of 7␣-OH-DHEA in murine arthritis are clearly needed to elucidate the mechanism by which 7␣-OH-DHEA interferes with the inflammatory process. In this respect, it would be of interest to investigate whether inflammation and arthritis development are reduced in CYP7B-knockout mice (28). Increased levels of 7␣-OH-DHEA locally in the arthritic joint are thought to result from a conversion of DHEA into 7␣-OH-DHEA by increased CYP7B activity. This activity may enhance inflammation not only via an increase in 7␣-OH-DHEA, but also via a decline in the level of DHEA (29,30). A number of studies have shown that, in patients with inflammatory diseases, DHEA concentrations in serum are low (11,21,22). The immunosuppressive effects of DHEA that have been described could be due to the conversion of DHEA into the steroids testosterone and/or 17-␤estradiol (1,2). The administration of DHEA seems to be beneficial in chronic inflammatory diseases in humans and mice (5–7,11). In addition, decreased Th1 cytokine production in vitro was observed in the presence of high doses of DHEA (31). However, other authors reported no effect on Th1 cytokines (32,33) or observed an increase in Th1 cytokine production (8,34). Immunosuppression was also observed following DHEA treatment of mice with CIA (6). In these mice, both IgG1 and IgG2a anti-CII antibodies were found to be reduced, suggesting down-modulation of both Th1 and Th2 responses (6). Besides immunosuppression, several authors have reported an immunostimulating role for DHEA in vitro and in vivo (8–10). One explanation for the discrepancy in the immunomodulatory effects of DHEA is 3352 that the 7␣-OH-DHEA metabolite, rather than DHEA, may be responsible for the effects observed (9,12,17). In this respect, it is of interest that 7␣-OH-DHEA appears to be more active than DHEA in stimulating the production of antilysozyme antibodies (9) and in counteracting the antigen-specific induction of nasal tolerance (10). It became clear from various studies that fibroblast-like cells display CYP7B activity by their ability to convert DHEA into 7␣-OH-DHEA (35–37). In our studies, CYP7B activity and CYP7B mRNA expression were observed with FLS isolated from the arthritic synovium. Recent studies also indicate that other cell types, such as macrophages (38) and lymphocytes (17), display CYP7B activity. Therefore, apart from FLS, macrophages and lymphocytes could also be involved in the formation of 7␣-OH-DHEA at the site of (chronic) inflammation. To investigate whether CYP7B is regulated by inflammatory cytokines, the proinflammatory cytokine IL-1␤ was used. IL-1␤ is recognized as a key cytokine in arthritis because it is found at high levels in arthritic synovium, and treatment of established arthritis with anti–IL-1␤ can block inflammation and thereby cartilage and bone destruction (26,39). Indeed, we show here that IL-1␤ stimulates CYP7B activity, CYP7B mRNA, and CYP7B protein levels in FLS. Our results regarding IL-1␤ stimulation of CYP7B activity are consistent with data from Payne et al (40) showing increased CYP7B activity induced by IL-1␤ in ovarian cells. It is likely that IL-1␤ is not the only mediator in inducing CYP7B activity, because tumor necrosis factor ␣ (TNF␣) treatment also induces such activity (40). Thus, TNF␣, which is, in addition to IL-1␤, a key cytokine in RA, may also be involved in enhanced formation of immunostimulating 7␣-OH-DHEA. Notably, activation of the P450 enzyme CYP7B by IL-1␤/ TNF␣ is a unique phenomenon since other P450 enzymes, such as CYP7A, CYP2C11, CYP3A2, and CYP1A1, are down-regulated by TNF␣ (41–44). Further research is needed to investigate whether CYP7B is present in synoviocytes from RA patients and whether CYP7B activity is regulated by inflammatory cytokines such as IL-1␤ and TNF␣ in human arthritis. The CYP7B promoter contains NF-B–responsive elements (45). It is known that NF-B is activated by inflammatory cytokines such as TNF␣ and IL-1␤. Studies are needed to clarify whether increased formation of 7␣-OH-DHEA via IL-1␤ and/or TNF␣ is regulated via the NF-B pathway. In conclusion, disease progression in CIA is correlated with an IL-1␤–dependent increase in CYP7B DULOS ET AL mRNA expression and CYP7B enzyme activity, which produce enhanced levels of 7␣-OH-DHEA. The latter conclusion is underlined by the fact that IL-1␤ regulates CYP7B mRNA expression in FLS, isolated from arthritic joints. More studies are needed to investigate whether CYP7B is a relevant target for the treatment of RA. ACKNOWLEDGMENTS We thank Mrs. M. van der Vleuten for generating and characterizing the CYP7B antibody, Mrs. T. Coenen-de Roo for radiographic analysis, and Mr. W. van der Heijden and Mrs. N. Bisseling for photographic reproductions. Dr. L. Joosten (University Medical Center St. Radboud, Nijmegen, The Netherlands) is acknowledged for helping with the scoring system for CIA knee joints. We thank Dr. A. Kavelaars and Professor Dr. C. J. Heijnen for stimulating discussions and for helpful suggestions regarding the manuscript. REFERENCES 1. Straub RH, Cutolo M. Involvement of the hypothalamic– pituitary–adrenal/gonadal axis and the peripheral nervous system in rheumatoid arthritis: viewpoint based on a systemic pathogenetic role [review]. Arthritis Rheum 2001;44:493–507. 2. Cutolo M, Seriolo B, Villaggio B, Pizzorni C, Craviotto C, Sulli A. Androgens and estrogens modulate the immune and inflammatory responses in rheumatoid arthritis [review]. Ann N Y Acad Sci 2002;966:131–42. 3. Dulos GJ, Bagchus WM. Androgens indirectly accelerate thymocyte apoptosis. Int Immunopharmacol 2001;1:321–8. 4. Bradlow HL, Murphy J, Byrne JJ. Immunological properties of dehydroepiandrosterone, its conjugates, and metabolites. Ann N Y Acad Sci 1999;876:91–101. 5. Giltay EJ, van Schaardenburg D, Gooren LJ, von Blomberg BM, Fonk JC, Touw DJ, et al. Effects of dehydroepiandrosterone administration on disease activity in patients with rheumatoid arthritis. Br J Rheumatol 1998;37:705–6. 6. Williams PJ, Jones RH, Rademacher TW. Reduction in the incidence and severity of collagen-induced arthritis in DBA/1 mice, using exogenous dehydroepiandrosterone. Arthritis Rheum 1997;40:907–11. 7. Du C, Khalil MW, Sriram S. Administration of dehydroepiandrosterone suppresses experimental allergic encephalomyelitis in SJL/J mice. J Immunol 2001;167:7094–101. 8. Daynes RA, Dudley DJ, Araneo BA. Regulation of murine lymphokine production in vivo. II. Dehydroepiandrosterone is a natural enhancer of interleukin 2 synthesis by helper T cells. Eur J Immunol 1990;20:793–802. 9. Morfin R, Courchay G. Pregnenolone and dehydroepiandrosterone as precursors of native 7-hydroxylated metabolites which increase the immune response in mice. J Steroid Biochem Mol Biol 1994;50:91–100. 10. Wolvers DA, Bakker JM, Bagchus WM, Kraal G. The steroid hormone dehydroepiandrosterone (DHEA) breaks intranasally induced tolerance, when administered at time of systemic immunization. J Neuroimmunol 1998;89:19–25. 11. Wilder RL. Adrenal and gonadal steroid hormone deficiency in the pathogenesis of rheumatoid arthritis. J Rheumatol 1996;44 Suppl:10–2. 12. Morfin R, Lafaye P, Cotillon AC, Nato F, Chmielewski V, CORRELATION OF CIA WITH CYP7B ACTIVITY 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. Pompon D. 7␣-hydroxy-dehydroepiandrosterone and immune response [review]. Ann N Y Acad Sci 2000;917:971–82. Chmielewski V, Drupt F, Morfin R. Dexamethasone-induced apoptosis of mouse thymocytes: prevention by native 7␣-hydroxysteroids. Immunol Cell Biol 2000;78:238–46. Morfin R. Involvement of steroids and cytochromes P(450) species in the triggering of immune defenses [review]. J Steroid Biochem Mol Biol 2002;80:273–90. Chiang JY. Regulation of bile acid synthesis. Front Biosci 1998;3: D176–93. Bjorkhem I, Eggertsen G. Genes involved in initial steps of bile acid synthesis [review]. Curr Opin Lipidol 2001;12:97–103. Lathe R. Steroid and sterol 7-hydroxylation: ancient pathways. Steroids 2002;67:967–77. Stapleton G, Steel M, Richardson M, Mason JO, Rose KA, Morris RG, et al. A novel cytochrome P450 expressed primarily in brain. J Biol Chem 1995;270:29739–45. Rose KA, Stapleton G, Dott K, Kieny MP, Best R, Schwarz M, et al. Cyp7b, a novel brain cytochrome P450, catalyzes the synthesis of neurosteroids 7␣-hydroxy dehydroepiandrosterone and 7␣hydroxy pregnenolone. Proc Natl Acad Sci U S A 1997;94: 4925–30. Schwarz M, Lund EG, Lathe R, Bjorkhem I, Russell DW. Identification and characterization of a mouse oxysterol 7␣hydroxylase cDNA. J Biol Chem 1997;272:23995–4001. Cutolo M, Sulli A, Villaggio B, Seriolo B, Accardo S. Relations between steroid hormones and cytokines in rheumatoid arthritis and systemic lupus erythematosus [review]. Ann Rheum Dis 1998;57:573–7. Straub RH, Lehle K, Herfarth H, Weber M, Falk W, Preuner J, et al. Dehydroepiandrosterone in relation to other adrenal hormones during an acute inflammatory stressful disease state compared with chronic inflammatory disease: role of interleukin-6 and tumour necrosis factor. Eur J Endocrinol 2002;146:365–74. Lubberts E, Joosten LA, Oppers B, van den Bersselaar L, Coenen-de Roo CJ, Kolls JK, et al. IL-1-independent role of IL-17 in synovial inflammation and joint destruction during collageninduced arthritis. J Immunol 2001;167:1004–13. Lapcik O, Hampl R, Hill M, Starka L. Immunoassay of 7-hydroxysteroids. 2. Radioimmunoassay of 7␣-hydroxy-dehydroepiandrosterone. J Steroid Biochem Mol Biol 1999;71:231–7. Van Meurs JB, van Lent PL, Joosten LA, van der Kraan PM, van den Berg WB. Quantification of mRNA levels in joint capsule and articular cartilage of the murine knee joint by RT-PCR: kinetics of stromelysin and IL-1 mRNA levels during arthritis. Rheumatol Int 1997;16:197–205. Joosten LA, Helsen MM, Saxne T, van de Loo FA, Heinegard D, van den Berg WB. IL-1 ␣␤ blockade prevents cartilage and bone destruction in murine type II collagen-induced arthritis, whereas TNF-␣ blockade only ameliorates joint inflammation. J Immunol 1999;163:5049–55. Sterzl I, Hampl R, Sterzl J, Votruba J, Starka L. 7␤-OH-DHEA counteracts dexamethasone induced suppression of primary immune response in murine splenocytes. J Steroid Biochem Mol Biol 1999;71:133–7. Li-Hawkins J, Lund EG, Turley SD, Russell DW. Disruption of the oxysterol 7␣-hydroxylase gene in mice. J Biol Chem 2000;275: 16536–42. Belanger A, Candas B, Dupont A, Cusan L, Diamond P, Gomez JL, et al. Changes in serum concentrations of conjugated and unconjugated steroids in 40- to 80-year-old men. J Clin Endocrinol Metab 1994;79:1086–90. 3353 30. Corsini E, Lucchi L, Meroni M, Racchi M, Solerte B, Fioravanti M, et al. In vivo dehydroepiandrosterone restores age-associated defects in the protein kinase C signal transduction pathway and related functional responses. J Immunol 2002;168:1753–8. 31. Du C, Guan Q, Khalil MW, Sriram S. Stimulation of Th2 response by high doses of dehydroepiandrosterone in KLH-primed splenocytes. Exp Biol Med (Maywood) 2001;226:1051–60. 32. MacPhee IA, Turner DR, Oliveira DB. The role of endogenous steroid hormones in the generation of T helper 2-mediated autoimmunity in mercuric chloride-treated Brown-Norway rats. Immunology 2000;99:141–6. 33. Moynihan JA, Callahan TA, Kelley SP, Campbell LM. Adrenal hormone modulation of type 1 and type 2 cytokine production by spleen cells: dexamethasone and dehydroepiandrosterone suppress interleukin-2, interleukin-4, and interferon-␥ production in vitro. Cell Immunol 1998;184:58–64. 34. Suzuki T, Suzuki N, Daynes RA, Engleman EG. Dehydroepiandrosterone enhances IL2 production and cytotoxic effector function of human T cells. Clin Immunol Immunopathol 1991;61: 202–11. 35. Rose K, Allan A, Gauldie S, Stapleton G, Dobbie L, Dott K, et al. Neurosteroid hydroxylase CYP7B: vivid reporter activity in dentate gyrus of gene-targeted mice and abolition of a widespread pathway of steroid and oxysterol hydroxylation. J Biol Chem 2001;276:23937–44. 36. Khalil MW, Strutt B, Vachon D, Killinger DW. Metabolism of dehydroepiandrosterone by cultured human adipose stromal cells: identification of 7␣-hydroxydehydroepiandrosterone as a major metabolite using high performance liquid chromatography and mass spectrometry. J Steroid Biochem Mol Biol 1993;46:585–95. 37. Zhang J, Larsson O, Sjovall J. 7␣-hydroxylation of 25-hydroxycholesterol and 27-hydroxycholesterol in human fibroblasts. Biochim Biophys Acta 1995;1256:353–9. 38. Schmidt M, Kreutz M, Loffler G, Scholmerich J, Straub RH. Conversion of dehydroepiandrosterone to downstream steroid hormones in macrophages. J Endocrinol 2000;164:161–9. 39. Dayer JM, Feige U, Edwards CK III, Burger D. Anti-interleukin-1 therapy in rheumatic diseases [review]. Curr Opin Rheumatol 2001;13:170–6. 40. Payne DW, Shackleton C, Toms H, Ben-Shlomo I, Kol S, deMoura M, et al. A novel nonhepatic hydroxycholesterol 7␣hydroxylase that is markedly stimulated by interleukin-1␤: characterization in the immature rat ovary. J Biol Chem 1995;270: 18888–96. 41. Feingold KR, Spady DK, Pollock AS, Moser AH, Grunfeld C. Endotoxin, TNF, and IL-1 decrease cholesterol 7␣-hydroxylase mRNA levels and activity. J Lipid Res 1996;37:223–8. 42. Nadin L, Butler AM, Farrell GC, Murray M. Pretranslational down-regulation of cytochromes P450 2C11 and 3A2 in male rat liver by tumor necrosis factor ␣. Gastroenterology 1995;109: 198–205. 43. Paton TE, Renton KW. Cytokine-mediated down-regulation of CYP1A1 in Hepa1 cells. Biochem Pharmacol 1998;55:1791–6. 44. Zhao Y, Nichols JE, Valdez R, Mendelson CR, Simpson ER. Tumor necrosis factor-␣ stimulates aromatase gene expression in human adipose stromal cells through use of an activating protein-1 binding site upstream of promoter 1.4. Mol Endocrinol 1996;10: 1350–7. 45. Wu Z, Martin KO, Javitt NB, Chiang JY. Structure and functions of human oxysterol 7␣-hydroxylase cDNAs and gene CYP7B1. J Lipid Res 1999;40:2195–203.