**2.2.3 Radioimmunoassay for ADM**

The RIA for ADM was performed as described previously (Kitamura et al., 1994). The incubation buffer for RIA was 0.05 M sodium phosphate buffer (pH 7.4), containing 0.5% BSA, 0.5% Triton X-100, 0.08 M NaCl, 0.025 M EDTA 2Na, 0.05% NaN3, and 500 KIU/ml trasylol. A disposable plastic tube (10 × 75 mm) was used for assaying. All assay procedures were performed at 4°C. Both standard ADM and unknown samples (100 μl) were incubated with anti-ADM antiserum diluent (200 μl) for 12 h before the tracer solution (125I-AM, 18,000–20,000 counts/min in 100 μl) was added. After incubation for 16 h, anti-rabbit IgG goat serum diluent (100 μl) was added. After standing for 24 h, the tubes were centrifuged at 3000 rpm for 30 min at 4°C and the radioactivity of the precipitate was measured using an Aloka ARC-600 gamma counter.

(68 ±8), and 10 healthy volunteer controls aged 50–66 years (57 ± 5). All subjects were

To clarify further the reason for the increase in plasma ADM, joint fluid, synovial tissue, and cartilage were measured by radioimmunoassay (RIA). These were acquired from surgical subjects during total knee arthroplasty in patients with RA (n = 6) and OA (n = 6). All patients with RA were classified as stage 4, functional class 2, according to the criteria of the American Rheumatism Association (ARA) (Clegg & Ward, 1987); they were medicated with only DMARDs and NSAIDs without steroid. All patients with OA were classified as stage 4 to 5, according to the Kellgren-Lawrence radiographic staging system (Kellgren & Lawrence, 1958). Since plasma ADM concentration has been reported to be elevated in patients with hypertension, renal failure, systemic infections, myocardial infarction, and heart failure (Ishimitsu et al., 1994; Nishikimi et al., 1995; Kobayashi et al., 1996), patients

**2.2 Methods of measuring ADM levels in plasma, joint fluid and joint tissues** 

Blood and joint fluid samples were transferred into tubes containing 1 mg/ml EDTA-2Na and 500 kallikrein inhibitory units/ml of aprotinin for measurement of ADM. The plasma was kept at –30°C until assayed. Levels of plasma ADM and joint fluid ADM were measured by IRMA using specific kits (AM RIA Shionogi) developed by Shionogi Pharmaceutical Co. Ltd., Osaka, Japan. The limit of detection of human AM is 0.5 pmol/l for

For measuring ADM levels of acquired joint fluids, samples were acidified with acetic acid to a final concentration of 1.0 M and centrifuged at 3000 rpm for 5 min, while synovium and cartilage specimens were acidified with acetic acid to a final concentration of 1.0 M and boiled for 10 min to inactivate proteases. The samples were then homogenized and centrifuged for 90 min at 12,000 rpm. The supernatant of samples was applied to a Sep-Pak C18 cartridge (Millipore-Waters, Milford, MA, USA). After the cartridge was washed with 10% CH3CN in 0.1% trifluoroacetic acid, the absorbed materials were eluted with 50% CH3CN in 0.1% trifluoroacetic acid. The eluted samples were dried by speed vacuum,

The RIA for ADM was performed as described previously (Kitamura et al., 1994). The incubation buffer for RIA was 0.05 M sodium phosphate buffer (pH 7.4), containing 0.5% BSA, 0.5% Triton X-100, 0.08 M NaCl, 0.025 M EDTA 2Na, 0.05% NaN3, and 500 KIU/ml trasylol. A disposable plastic tube (10 × 75 mm) was used for assaying. All assay procedures were performed at 4°C. Both standard ADM and unknown samples (100 μl) were incubated with anti-ADM antiserum diluent (200 μl) for 12 h before the tracer solution (125I-AM, 18,000–20,000 counts/min in 100 μl) was added. After incubation for 16 h, anti-rabbit IgG goat serum diluent (100 μl) was added. After standing for 24 h, the tubes were centrifuged at 3000 rpm for 30 min at 4°C and the radioactivity of the precipitate was measured using an

**2.2.1 Methods measuring the plasma and joint fluid ADM level** 

and controls with these conditions were excluded.

**2.2.2 Extraction of ADM in joint tissues** 

freeze-dried and stored at –30°C until assayed.

**2.2.3 Radioimmunoassay for ADM** 

Aloka ARC-600 gamma counter.

female.

these kits.

#### **2.3 Result of ADM in plasma, joint fluid and joint tissues in patients with RA Discussion of joint ADM level in patients with RA**

It has been reported that some collagenous disorders show increased levels of plasma ADM (Yudoh et al., 1999). We measured and compared plasma ADM concentrations in patients with RA and healthy controls, finding that patients with RA exhibited a 1.7-fold increase in plasma ADM levels (Table 1).


Table 1. Blood adrenomedullin (ADM) levels in RA and OA patients and healthy controls

All values are expressed as means ± SD.※p < 0.01. Patients with RA demonstrated high plasma concentration of ADM (18.35 ± 6.9 fmol/ml) compared to healthy controls (11.64 ± 2.8 fmol/ml) and OA patients (12.88 ± 1.9 fmol/ml)

Moreover, plasma ADM levels in patients with RA were also found to be significantly correlated to CRP levels (Fig. 1). In RA, CRP correlates with disease activity and response to therapy. Our patient data did not include RA disease activity, excluding RA functional class classification. Hamada et al. (2010) stated that no autologous antibody such as rheumatoid factor or anti-CCP antibody showed significant correlation with plasma ADM level. In addition, there was no correlation with disease activity scores such as DAS-28. DAS-28 is not only dependent on the inflammatory level, so the plasma ADM level may escalate before escalation of the activity in synovitis or the amount of actual synovial inflammation (Hamada et al., 2010). Therefore, our results suggest that ADM levels are increased in patients with RA and investigated that they might be correlated with disease activity. Moreover, ADM concentration in the joint fluid of RA patients (10.8 ± 4.3 fmol/ml) was significantly higher than that of OA patients (7.2 ± 1.8 fmol/ml) (Fig. 2).

Fig. 1. Correlation between plasma AM and C-reactive protein

A significant positive correlation was observed between AM and CRP (correlation coefficient = 0.685, p < 0.01). Plasma ADM and plasma CRP levels were found to be well correlated. The correlation coefficient between CRP and AM was 0.685, p < 0.01.

Role of Adrenomedullin in Patients with Rheumatoid Arthritis 177

to host defense systems (Ueda et al., 1999; Elsasser & Kahl, 2002). From these observations, we speculate that ADM may play a role in the pathophysiology of inflammation as well as

Synovitis inevitably plays a role in the destruction of joint surface, so we measured the concentration of ADM in articular cartilage. We found the ADM concentration in cartilage was lower than that of other tissues (Fig. 2). While the concentration of ADM in normal human articular cartilage has not been determined, an immunohistochemical study has reported that normal human articular chondrocytes produce ADM (Fig. 4) (Asada et al., 1999). No statistically significant difference was found in the articular cartilage concentration of ADM in RA and OA patients (Fig. 3). We therefore speculate that this may be because our samples were from cases that consisted of end-stage arthritis, with advanced degeneration and differentiation of the cartilage. A thorough investigation of the role of

Fig. 3. Concentration of ADM for synovium and articular cartilage in patients with OA and RA RA patients showed higher concentrations of ADM in synovium compared to OA patients. \*p < 0.01; \*\*not significant. In cartilage, there is no significantly difference between the

Fig. 4. Novel distribution of adrenomedullin-immunoreactive cells in human cartilage

Hyaline cartilage: Chondrocytes are positive for ADM. (donated from Prof. Asada, Miyazaki

in the regulation of joint disorders.

concentration of OA and RA

University)

ADM in cartilage pathophysiology is necessary.

Fig. 2. Concentration of ADM between RA and OA in plasma and joint fluid

Patients with RA demonstrated high plasma concentration of ADM (18.35 ± 6.9 fmol/ml) compared to OA patients (12.88 ± 1.9 fmol/ml). ADM concentration in the joint fluid of RA patients (10.8 ± 4.3 fmol/ml) was significantly higher than that of OA patients (7.2 ± 1.8 fmol/ml)

We applied conventional radioimmunoassay (RIA) to determine ADM concentration in tissues. RA is characterized by the presence of inflammatory synovitis accompanied by destruction of joint cartilage and bone. The concentration of ADM in the synovium of RA patients was 3.2-fold higher than that of OA patients (Fig. 3). In addition, the concentration of ADM in plasma of RA patients was 1.4-fold higher than that of OA patients (Table 2). To determine the relationship between plasma ADM and arthritis in patients with RA, we compared ADM levels in joint fluid of RA and OA patients. Joint fluid is similar in composition to plasma, which explains the similarly significant increases in ADM concentration in both plasma and joint fluid in patients with RA compared with those with OA (p < 0.01) (Fig. 2). These observations indicate that the reason for this high concentration of ADM might be secretion from synovial stromal cells or secretion from synovial vascular wall cells. However, the mechanism by which plasma and joint fluid ADM levels increase in patients with RA remains unknown.

Several tissues including vessels secrete ADM, and elevated ADM could conceivably be caused by secretion from vascular cells in general. When we consider the results, in which ADM levels in the synovium of RA patients were higher than those of OA patients, we may assume that synovitis is one reason for the increased ADM concentration in synovium and joint fluid, and may partially contribute to the increase in ADM levels in plasma. Our results indicate the possibility that ADM participates in the pathophysiology of joint lesions in patients with RA. Together with the fact that ADM is known to inhibit the secretion of cytokines from several cell lines (Isumi et al., 1999; Kamoi et al., 1995), the findings seem to validate the assumption that production and secretion of ADM in synovium are strongly correlated with anti-inflammation in arthritis. Thus, the elevation of plasma ADM levels may be related to the anti-inflammatory response. Clementi et al. (1999) investigated the anti-inflammatory effect of ADM in rats, finding that ADM production in several cell lines was strongly induced by stimulation of a group of inflammatory cytokines including interleukin 1 and tumor necrosis factor-α (TNF-α) (Hofbauer et al., 2002). Some studies report that ADM acts as a circulating vasoactive hormone in blood, and plays an antiinflammatory role in the prevention of local infection and inflammation, thus contributing

Fig. 2. Concentration of ADM between RA and OA in plasma and joint fluid

fmol/ml)

patients with RA remains unknown.

Patients with RA demonstrated high plasma concentration of ADM (18.35 ± 6.9 fmol/ml) compared to OA patients (12.88 ± 1.9 fmol/ml). ADM concentration in the joint fluid of RA patients (10.8 ± 4.3 fmol/ml) was significantly higher than that of OA patients (7.2 ± 1.8

We applied conventional radioimmunoassay (RIA) to determine ADM concentration in tissues. RA is characterized by the presence of inflammatory synovitis accompanied by destruction of joint cartilage and bone. The concentration of ADM in the synovium of RA patients was 3.2-fold higher than that of OA patients (Fig. 3). In addition, the concentration of ADM in plasma of RA patients was 1.4-fold higher than that of OA patients (Table 2). To determine the relationship between plasma ADM and arthritis in patients with RA, we compared ADM levels in joint fluid of RA and OA patients. Joint fluid is similar in composition to plasma, which explains the similarly significant increases in ADM concentration in both plasma and joint fluid in patients with RA compared with those with OA (p < 0.01) (Fig. 2). These observations indicate that the reason for this high concentration of ADM might be secretion from synovial stromal cells or secretion from synovial vascular wall cells. However, the mechanism by which plasma and joint fluid ADM levels increase in

Several tissues including vessels secrete ADM, and elevated ADM could conceivably be caused by secretion from vascular cells in general. When we consider the results, in which ADM levels in the synovium of RA patients were higher than those of OA patients, we may assume that synovitis is one reason for the increased ADM concentration in synovium and joint fluid, and may partially contribute to the increase in ADM levels in plasma. Our results indicate the possibility that ADM participates in the pathophysiology of joint lesions in patients with RA. Together with the fact that ADM is known to inhibit the secretion of cytokines from several cell lines (Isumi et al., 1999; Kamoi et al., 1995), the findings seem to validate the assumption that production and secretion of ADM in synovium are strongly correlated with anti-inflammation in arthritis. Thus, the elevation of plasma ADM levels may be related to the anti-inflammatory response. Clementi et al. (1999) investigated the anti-inflammatory effect of ADM in rats, finding that ADM production in several cell lines was strongly induced by stimulation of a group of inflammatory cytokines including interleukin 1 and tumor necrosis factor-α (TNF-α) (Hofbauer et al., 2002). Some studies report that ADM acts as a circulating vasoactive hormone in blood, and plays an antiinflammatory role in the prevention of local infection and inflammation, thus contributing to host defense systems (Ueda et al., 1999; Elsasser & Kahl, 2002). From these observations, we speculate that ADM may play a role in the pathophysiology of inflammation as well as in the regulation of joint disorders.

Synovitis inevitably plays a role in the destruction of joint surface, so we measured the concentration of ADM in articular cartilage. We found the ADM concentration in cartilage was lower than that of other tissues (Fig. 2). While the concentration of ADM in normal human articular cartilage has not been determined, an immunohistochemical study has reported that normal human articular chondrocytes produce ADM (Fig. 4) (Asada et al., 1999). No statistically significant difference was found in the articular cartilage concentration of ADM in RA and OA patients (Fig. 3). We therefore speculate that this may be because our samples were from cases that consisted of end-stage arthritis, with advanced degeneration and differentiation of the cartilage. A thorough investigation of the role of ADM in cartilage pathophysiology is necessary.

RA patients showed higher concentrations of ADM in synovium compared to OA patients. \*p < 0.01; \*\*not significant. In cartilage, there is no significantly difference between the concentration of OA and RA

Fig. 4. Novel distribution of adrenomedullin-immunoreactive cells in human cartilage

Hyaline cartilage: Chondrocytes are positive for ADM. (donated from Prof. Asada, Miyazaki University)

Role of Adrenomedullin in Patients with Rheumatoid Arthritis 179

identical elution position to that of authentic human ADM (1-52), which is the full-length human ADM peptide (Fig. 6). The minor peak that was eluted earlier was thought to be oxidized ADM containing methione sulfoxide. The intracellular ADM concentration remained extremely low in RA synoviocytes (Fig. 5), suggesting that these cells constitutively secrete ADM. Thus, RA synoviocytes seem to secrete ADM rapidly after its

Fig. 6. Analysis by reverse-phase HPLC of immunoreactive adrenomedullin (ir-ADM)

Fig. 7. Inhibitory effect of ADM about the IL-6 secretion into the media in cultured FLS

Effects of ADM 22-52 10-6M (ADM blocker) on control level of IL-6 secretion in cultured fibroblast like synoviocytes (FLS) donated from RA patients. Serum-starved FLS were incubated 24 hours. Values are the means±SEM of six wells examined. Each was compared with the cells incubated with 1%FBS media (control). Each set of experiments was repeated

When the synoviocytes reached confluence, ADM (10-8 M) was added to the culture media. After 24 hours, the level of IL-6 in culture media significantly decreased. Moreover, ADM (22-52), which is an ADM antagonist, elevated the level of IL-6 in culture media (Fig. 7).

A linear gradient of acetonitrile of 10% to 60% was made in 0.1% trifluoroacetic acid for 60 minutes at a flow rate of 1.0mL/min. The arrow indicates the elution position of synthetic

synthesis, with little intracellular storage.

human AM. The specificity of the RIA was clarified.

three times and identical results were obtained.

secreted into the media

We observed that plasma ADM concentration in patients with RA was higher than that of healthy controls, and plasma ADM and plasma CRP levels were found to be well correlated. Our data suggest that plasma ADM levels increase with the activity of RA. We conclude that ADM probably plays a part in the regulation of the inflammatory process of RA, and its plasma and/or joint fluid levels could be used as an index of the degree of RA. Therefore, we investigated the role of ADM in inflammation in cultured RA synoviocytes (fibroblastlike synovium) in a subsequent investigation.
