| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Rheumatology, Department of Internal Medicine I, University Hospital, Regensburg, Germany;
* Division of Rheumatology, Department of Internal Medicine, University of Genova, Italy; and
Department of Orthopedic Surgery, Medical Faculty of the University of Regensburg, in the Bavarian Red Cross Hospital, Bad Abbach, Germany
1Correspondence: Laboratory of Neuroendocrinoimmunology, Department of Internal Medicine I, University Hospital Regensburg, 93042 Regensburg, Germany. E-mail: rainer.straub{at}klinik.uni-r.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: RA • osteoarthritis • cortisol • NE • cytokine
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6
7
8
9)
. In untreated RA patients, we recently identified low levels of serum cortisol in relation to serum interleukin 6 (IL-6) and tumor necrosis factor (TNF) in association with a higher degree of joint inflammation (10)
. A higher degree of joint inflammation was associated with reduction of the synovial sympathetic innervation (SSI) in patients with RA (11)
. Both factors probably contribute to an overall proinflammatory situation in the synovial tissue due to a lack of anti-inflammatory cooperativity of cortisol and norepinephrine (NE).
In asthma therapy, parallel topical (or systemic) treatment with a ß-adrenergic agonist and corticosteroids (CS) has an additive anti-obstructive effect (12
, 13)
. Cooperative effects of cortisol and NE lead to an increase of glucocorticoid receptors, ß-adrenoreceptors, intracellular cAMP, protein kinase A, and CREB, a sequence of events that has been demonstrated in various cell types (14
15
16
17
18
19
20
21)
. Increase of these intracellular mediators is accompanied by a dramatic anti-inflammatory response in various immune cells (22
23
24
25
26
27
28)
. Furthermore, cortisol supports production of NE and epinephrine from sympathetic nerve terminals and adrenal medulla by inducing the synthesizing enzymes (29
, 30)
. Thus, one would expect that a parallel increase of cortisol and NE with local physiological concentrations of 10-8 to 10-6 M would be more anti-inflammatory than either substance alone. Presumably, a dissociation of these two factors may be deleterious in RA. However, respective studies to investigate anti-inflammatory cooperativity of cortisol and NE have never been conducted in synovial tissue of patients with RA or, for comparison, in patients with osteoarthritis (OA).
In this study, we aimed to directly demonstrate the anti-inflammatory cooperativity of CS and NE using two independent patient groups and study protocols. In an in vivo study in a large group of RA patients with or without CS treatment prior to knee joint replacement surgery, density of sympathetic nerve fibers and functional and histological markers of synovial inflammation was determined in explanted synovial tissue. Subgroups of patients with/without prior CS and with/without detectable SSI were compared. In an in vitro study, the cooperative effect of cortisol and NE was tested in primary early culture mixed synoviocytes and 3rd passage fibroblasts obtained from RA and OA patients with respect to secretion of TNF, IL-6, IL-8, and matrix metalloproteinase 3 (MMP-3).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. For RA patients in the in vivo study, we were not able to include a control group of OA patients because OA patients did not receive prior CS treatment. However, for the 10 RA patients in the in vitro study, 10 Caucasian patients with OA were included as control group. All patients in both studies underwent knee joint replacement surgery and were consecutively included without further selection. Basic characteristics of patients of the two independent studies are given in Table 1
. RA patients of the in vivo study were separated into four subgroups according to SSI (histological technique described below): A) patients with prior CS therapy/without SSI (n=7; without means absolutely no sympathetic nerve fiber was histologically visible); B) patients without prior CS therapy/with SSI (n=6); and C) patients with prior CS therapy/with SSI (n=19). Unfortunately, no patients were available for the fourth subgroup without prior CS therapy/without SSI. The patients were not different in age or gender. No patient received prior intra-articular injection of CS. Patients were informed about the purpose of the study and gave written consent. The study was approved by the Ethical Committee of the University of Regensburg. Parameters such as erythrocyte sedimentation rate, C-reactive protein, and blood leukocyte count were measured by standard techniques.
|
Synovial tissue preparation and isolation of synovial cells
Synovial tissue samples were obtained during surgery immediately after opening the knee joint capsule. Preparation of the tissue for histology has been described (11)
. One piece of 
9 cm2 of synovial tissue was dissected. Fat tissue and tissue with a large number of vessels were removed. Six 16 mm2 pieces were loaded into six superfusion chambers (see below) and eight 1 cm2 pieces of the same synovial area were used for histology. The samples intended for hematoxylin-eosin and alkaline phosphatase anti-alkaline phosphatase (APAAP) staining were immediately placed in freezing medium (Tissue-Tek, Sakura Finetek Europe, Zoeterwoude, The Netherlands) and quick-frozen floating on liquid nitrogen. The tissue samples used for detection of nerve fibers were fixed for 12–24 h in phosphate-buffered saline (PBS) containing 4% formaldehyde and incubated in PBS with 20% sucrose for 12 to 24 h. The tissue was then embedded in Tissue-Tek and quick-frozen. Each patient’s samples of synovial tissue were stored at -80°C.
Synovial cells were isolated by enzymatic digestion of fresh synovial tissue for 1–2 h at 37°C (Dispase Grade II, Boehringer, Mannheim, Germany). The cells were seeded in RPMI 1640 (Sigma, Deisenhofen, Germany), 10% FCS (Sigma), 1% penicillin/streptomycin (Life Technologies, Inc., Paisley, U.K.), 0.1% amphotericin B (Bristol-Myers Squibb, Munich, Germany) and cultured for 12 h (primary early culture mixed synoviocytes) or for 3–6 wk (3rd passage fibroblasts). The percentage of different types of synoviocytes was tested by specific antibodies against prolyl 4-hydroxylase (for the synoviocyte type B=fibroblasts; Calbiochem, Bad Soden) and CD163 (synoviocyte type A=activated macrophages; Dako, Hamburg). Twenty thousand cultured synovial cells were used for stimulatory tests with NE and cortisol alone or in combination as described below.
Histological evaluation of inflammation and determination of synovial sympathetic innervation (SSI)
Histological evaluation was used for the in vivo study and the techniques have been described earlier (11)
. Tissue samples were cut into 6–8 µm thick sections and cell density and lining layer thickness were evaluated using a standard hematoxylin-eosin staining of 45 sections. At a magnification of 400x, the extent of the lining layer thickness was determined by averaging the number of cells in a lining layer cross section at nine different locations. Cell density in the synovial tissue was determined by counting all stained cells in 17 randomly selected high-power fields (400x) and expressed per mm2. To determine the number of T cells (CD3, Dako), macrophages (CD163, Dako), and capillary vessels (collagen IV, Dako) in the synovial tissue of each patient, eight cryosections were investigated using APAAP staining; the number of identified structures was averaged from 17 randomly selected high-power fields (400x) and expressed per mm2.
For determination of SSI, six to eight 7–9 µm thick cryosections were used for immunohistochemistry with primary antibodies against tyrosine hydroxylase (TH, the key enzyme for NE production in sympathetic nerve endings; Chemicon, Temecula, CA). A Cy3-labeled secondary antibody (Dianova, Hamburg, Germany) was used to achieve an immunofluorescent staining. The numbers of stained nerve fibers per mm2 were determined by averaging the number of stained nerve fibers with a minimum length of 50 µm (indicated by a small micrometer eyepiece) in 17 randomly selected high-power fields (400x).
Inflammation index
For the in vivo study in RA patients, morphological and functional inflammatory parameters were combined in order to establish a previously published mathematical index of an individual RA patient (11)
:
 |
100=less inflammation than the mean value of all RA patients, which is 100).
Cytokine analysis in the superfusate and in culture supernatants
In the superfusion experiments (see superfusion technique below), cytokine concentrations were determined in a superfusate fraction of 1 ml (collected over 15 min). Human IL-6 and IL-8 (detection limit in the two assays: <2 pg/ml; Endogen, Boston) were determined by enzyme immunometric assay. We did not include TNF and MMP-3 as read-out parameters, because in superfusion experiments TNF and MMP-3 were not measurable in 50% of patients.
In culture experiments for the in vitro study with primary early culture mixed synoviocytes and 3rd passage synovial fibroblasts, supernatant was collected 16 h after the start of incubation with NE and/or cortisol and stored in adequate aliquots at -80°C for determination of cytokines. Enzyme immunometric assays for human TNF (detection limit: 0.15 pg/ml, HS Quantikine, R&D, Wiesbaden, Germany), IL-6 (Endogen), IL-8 (Endogen), and MMP-3 (detection limit: 3 ng/ml, Amersham, Braunschweig, Germany) were used. Intra-and inter-assay coefficients of variation for all mentioned ELISAs were < 10%.
Superfusion technique of synovial tissue and measurement of spontaneous cytokine secretion
As described previously in detail for spleen slices (32
, 33)
, we used a superfusion chamber (80 µl) apparatus to superfuse pieces of synovial tissue with culture medium (RPMI 1640, 25 mM HEPES, 5% FCS, 1% penicillin/streptomycin, 30 µM mercaptoethanol, 0.57 mM ascorbic acid, 1.3 mM calcium, all additions from Sigma, Deisenhofen, Germany). Superfusion was performed for 2 h at 37°C and a flow rate of 66 µl/min (one piece per chamber, six chambers in parallel). At 120 min, superfusate was collected in order to measure spontaneous cytokine release (1 ml collected over 15 min). The superfusate was immediately frozen and stored for short periods at -20°C.
Modulation of cytokine secretion by NE and/or cortisol
In the in vitro study, NE (Arterenol®, Höchst, Frankfurt, Germany) and cortisol (hydrocortison® 100, Pharmacia & Upjohn, Erlangen, Germany) were used. Since benzyl alcohol was an additive of hydrocortison® 100, we added the respective amount of this vehicle to control cultures and NE cultures. Twenty thousand primary early culture mixed synoviocytes and 3rd passage synovial fibroblasts were stimulated for 16 h in 1 ml Dulbecco’s MEM plus 100 µl medium with the adequate concentration of NE, cortisol or a combination of NE plus cortisol to obtain the mentioned final concentrations. The cells were not further stimulated by additional substances because they spontaneously produced large amounts of the tested cytokines. Due to the different tissue digestion time of RA tissue (needs more time) in comparison to OA tissue (needs less time), we believe that absolute values of cytokine supernatant concentrations in the in vitro study should not be used for inter-disease comparison of the cytokine secreting capability of RA or OA synovial cells. Relative changes of cytokine secretion in relation to control conditions were used in order to compare the two disease groups.
Presentation of the data and statistical analysis
All data are given as mean ± SE. Two group means were compared by the nonparametric Mann-Whitney test (SPSS/PC, Advanced Statistics, V10.0.1, SPSS Inc., Chicago). Percentages were compared by 
2 test using Yates continuity correction (SPSS). Differences between three group means were tested by the nonparametric Kruskal-Wallis test (SPSS). Means of two groups (OA, RA) over more than one condition (different concentrations of cortisol or NE in Fig. 4
) were compared using the General Linear Model analysis (GLM, SPSS). This analysis investigated influence of group allocation and influence of concentration in parallel (earlier called MANOVA). P < 0.05 was the significance level.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. This did not reach a significant level for number of cells in the lining layer, overall cell density, and vessel density although group C tended to have lower values than the other subgroups (Table 2)
. It is obvious that the subgroup of patients with prior CS therapy/with SSI had significantly lower synovial inflammatory activity as assessed by functional and histological markers.
|
Anti-inflammatory cooperativity of CS and NE in vitro in primary early culture mixed synoviocytes of RA and OA patients
In preliminary experiments with primary early culture mixed synoviocytes of three patients with RA and three patients with OA, we detected that 26 ± 3% of cells were positive for CD163 (i.e., activated macrophage) and 37 ± 3% were positive for prolyl 4-hydroxylase (i.e., fibroblast). There was a significant difference between RA and OA patients with respect to percentage of CD163 positive cells (36±3 vs. 15±3%, P<0.001), which was not observed for prolyl 4-hydroxylase positive cells (37±4 vs. 38±4%, n.s.). This suggests that the following results with primary early culture mixed synoviocytes of RA patients were influenced more by activated macrophages (CD 163) than cultures of OA patients.
Cooperativity together with NE has been examined using two different concentrations of cortisol at 10-6 M (Fig. 1
A–D) and 10-7 M (Fig. 2
A–D). In most instances, cortisol had a strong inhibitory effect on secretion of TNF and IL-8 from cells of OA and RA patients, leading to a reduction of cytokine secretion by 15–30% (Figs. 1
, 2)
. This cortisol effect was similar with respect to secretion of IL-6 and MMP-3 (data not shown). Similar to cortisol, NE strongly inhibited secretion of TNF and IL-8 at 10-6 to 10-8 M (Figs. 1
, 2)
. In experiments with simultaneous administration of cortisol and NE to synovial cells of RA patients, stronger inhibition of TNF and IL-8 secretion was observed than with either substance alone (Fig. 1C, D
, Fig. 2C, D
). Although it was similar in direction, this did not reach a significant level in OA patients (Figs. 1A, B
, 2A, B
). A combination of cortisol (10-6 M and 10-7 M) and NE (10-6 M) induced a stronger inhibition of TNF secretion than the other combinations (Figs. 1C
, 2C
). Similarly, only in cells of RA but not of OA patients, a combination of cortisol (10-6 M) and NE (10-6 to 10-8 M) induced a stronger inhibitory effect on IL-6 secretion than either substance alone (Fig. 3
), which was not significant for the combination of cortisol at 10-7 M and different NE concentrations. Concerning secretion of MMP-3 from RA and OA synovial cells, no such inhibitory cooperative effects were observed.
|
|
|
Anti-inflammatory cooperativity of CS and NE in vitro in 3rd passage synovial fibroblasts of patients with RA and OA
In preliminary experiments with 3rd passage fibroblasts of three patients with RA and three patients with OA, we detected that 0% of cells were positive for CD163 (i.e., activated macrophage) and 100% were positive for prolyl 4-hydroxylase (i.e., fibroblast). There was no significant difference between RA and OA synovial cells for both antigens. Furthermore, in supernatants of 3rd passage fibroblasts TNF and MMP-3 were not detectable, which also indicates that activated macrophages were absent.
Using an identical experimental design as mentioned for primary early culture mixed synoviocytes, experiments with 3rd passage fibroblasts did not reveal any cooperative effect of cortisol and NE on secretion of IL-6 or IL-8 (data not shown; TNF and MMP-3 were not measurable). However, cortisol induced a concentration-dependent inhibition of IL-6 (P<0.001) and IL-8 (P<0.001) in cells of RA and OA patients with a maximum effect at 10-6 M (IL-6: 43.9% of control in RA, 33.5% of control in OA; IL-8: 46.5%, 47.1%, respectively). NE did not inhibit IL-6 or IL-8 secretion of OA cells, but inhibited cytokine secretion of RA cells without a clear concentration-dependent effect (to 80% of control for both cytokines, P<0.001).
Responsiveness of primary early culture mixed synoviocytes and 3rd passage fibroblasts to cortisol and NE in patients with RA and OA
Cortisol concentration-dependently inhibited secretion of IL-6, IL-8, TNF, and MMP-3 (Fig. 4
). Inhibition of secretion of IL-8, TNF, and MMP-3 was stronger in synoviocytes of RA than in OA patients (Fig. 4)
. Similarly, NE-induced inhibition of IL-8 and MMP-3 was stronger in RA synoviocytes than OA cells (Fig. 4B, D
). A concentration-dependent inhibition was observed only for TNF (Fig. 4C
).
In 3rd passage fibroblasts, cortisol exerted a concentration-dependent inhibition of IL-6 and IL-8 (for both, P<0.001) but responsiveness of cells from OA patients was not different from cells of RA patients (data not shown). In contrast, NE-induced inhibition of both cytokines was stronger when using cells of RA patients than cells of OA patients (for both cytokines, P<0.001), although a concentration-dependent effect was not present.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
15
16
17
18
19
20
21)
. In patients with asthma, this has led to a more effective combination therapy with local CS and local ß-adrenergic agents than either substance alone (12
, 13)
. This permissive effect of cortisol is due to ß-adrenoreceptor up-regulation and stabilization of the cAMP signaling pathway (34)
. However, such a cooperative effect has never been demonstrated in primary synovial cells of patients with RA and OA. This study demonstrated anti-inflammatory cooperativity of cortisol and NE in vivo and in vitro in primary early culture mixed synoviocytes in patients with RA. In the in vivo study, RA patients with prior prednisolone treatment and presence of SSI had significantly lower levels of synovial inflammation than patients without prior prednisolone treatment and presence of SSI or than patients with prior prednisolone treatment but without SSI. This indicates that a combination of CS and NE leads to a more anti-inflammatory situation in the synovial microenvironment. To better characterize these associative in vivo findings, we wanted to study a cooperative effect of cortisol and NE in vitro using primary early culture mixed synoviocytes and 3rd passage fibroblasts. These experiments supported the in vivo data by showing an inhibitory cooperative effect of cortisol and NE on secretion of TNF, IL-6, and IL-8 from mixed RA synoviocytes. Inhibition did not reach the level of an additive or synergistic effect but it was demonstrated that it is at least a subadditive cooperative effect. Although the findings were similar in OA synoviocytes, the anti-inflammatory cooperative effect did not reach the level of significance. Furthermore, responsiveness to cortisol and NE was more pronounced in primary early culture mixed synoviocytes of patients with RA vs. OA (depending on the read-out parameter between 10 and 20% difference). One may speculate that due to a higher percentage of activated macrophages (and other cells) in RA than OA cultures, the anti-inflammatory cooperativity and responsiveness to cortisol and NE were more marked. This is supported by the fact that an anti-inflammatory cooperative effect was not observed in 3rd passage fibroblasts, suggesting that fibroblasts are not the primary target of cooperative effects of cortisol and NE. One may speculate that stronger cortisol- and NE-induced inhibition of cytokine secretion demonstrates up-regulation of respective receptors in RA more than in OA, possibly due to an earlier decrease in respective local concentrations of these two mediators.
In conclusion, anti-inflammatory cooperativity of cortisol and NE seems to play an important role to suppress local synovial inflammation. Concerning RA therapy, a combination of CS and ß-adrenergic agonists may represent an interesting new option that has already been demonstrated in asthma therapy. Indeed, in a similar way, cooperativity of CS and methotrexate may be achieved through elevation of local adenosine that increases cAMP via A2 adenosine receptors (35)
. These known therapeutic effects of CS and methotrexate and our own data indicate that the simultaneous presence of cortisol and cAMP-enhancing substances exerts a stronger anti-inflammatory capacity that may be a new principle of RA therapy, at least at the synovial level.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 28, 2002. Accepted for publication February 28, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R H Straub, R Wiest, U G Strauch, P Harle, and J Scholmerich The role of the sympathetic nervous system in intestinal inflammation. Gut, November 1, 2006; 55(11): 1640 - 1649. [Full Text] [PDF]
|
|
|
|
 |
|
 O. Lian, J. Dahl, P. W. Ackermann, F. Frihagen, L. Engebretsen, and R. Bahr Pronociceptive and Antinociceptive Neuromediators in Patellar Tendinopathy Am. J. Sports Med., November 1, 2006; 34(11): 1801 - 1808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Harle, R H Straub, R Wiest, A Mayer, J Scholmerich, F Atzeni, M Carrabba, M Cutolo, and P Sarzi-Puttini Increase of sympathetic outflow measured by neuropeptide Y and decrease of the hypothalamic-pituitary-adrenal axis tone in patients with systemic lupus erythematosus and rheumatoid arthritis: another example of uncoupling of response systems Ann Rheum Dis, January 1, 2006; 65(1): 51 - 56. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Del Rosso, M. Cinelli, S. Guiducci, A. Pignone, G. Fibbi, F. Margheri, A. Gabrielli, R. Giacomelli, A. Coppini, M. Del Rosso, et al. Deflazacort modulates the fibrinolytic pattern and reduces uPA-dependent chemioinvasion and proliferation in rheumatoid arthritis synoviocytes Rheumatology, October 1, 2005; 44(10): 1255 - 1262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R H Straub, F Buttgereit, and M Cutolo Benefit of pregnancy in inflammatory arthritis Ann Rheum Dis, June 1, 2005; 64(6): 801 - 803. [Full Text] [PDF]
|
|
|
|
|
|
P. Harle, T. Bongartz, J. Scholmerich, U. Muller-Ladner, and R. H. Straub Predictive and potentially predictive factors in early arthritis: a multidisciplinary approach Rheumatology, April 1, 2005; 44(4): 426 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C Weidler, C Holzer, M Harbuz, R Hofbauer, P Angele, J Scholmerich, and R H Straub Low density of sympathetic nerve fibres and increased density of brain derived neurotrophic factor positive cells in RA synovium Ann Rheum Dis, January 1, 2005; 64(1): 13 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Juarranz, B. Santiago, M. Torroba, I. Gutierrez-Canas, G. Palao, M. Galindo, C. Abad, C. Martinez, J. Leceta, J. L. Pablos, et al. Vasoactive intestinal peptide modulates proinflammatory mediator synthesis in osteoarthritic and rheumatoid synovial cells Rheumatology, April 1, 2004; 43(4): 416 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. STRAUB and H. O. BESEDOVSKY Integrated evolutionary, immunological, and neuroendocrine framework for the pathogenesis of chronic disabling inflammatory diseases FASEB J, December 1, 2003; 17(15): 2176 - 2183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. De Bosscher, W. Vanden Berghe, and G. Haegeman The Interplay between the Glucocorticoid Receptor and Nuclear Factor-{kappa}B or Activator Protein-1: Molecular Mechanisms for Gene Repression Endocr. Rev., August 1, 2003; 24(4): 488 - 522. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
