amelioration of rat experimental arthritides by treatment with the alkaloid sinomenine

Int. J. lmmunopharmac., Vol. 18, No. 10, pp. 529 543, 1996 Pergamon Copyright ~, 1997 International Society for lmmunopharmacology

Published by Elsevier Science Ltd. Printed in Great Britain 0192~)561 ;96 $15.00 + .00

PII: S0192--0561 (96)00025"--2

AMELIORATION OF RAT EXPERIMENTAL ARTHRITIDES BY TREATMENT WITH THE ALKALOID SINOMENINE

LIANG LIU,*t EBERHARD BUCHNER,*§ DENNIS BEITZE,+ + CARSTEN B. SCHMIDT-WEBER,* VOLKHARD KAEVER, + FRANK EMMR1CH§ and

RAIMUND W. KINNE*§ f

*Immunology Unit, Department of Medicine llI, University of Erlangen-Nuremberg, D-91054 Erlangen, Germany; tPharmacology Section at the Institute of Chinese Materia Medica Development, Guangzhou College of Traditional Chinese Medicine, 510407 Guangzhou, P. R. China; ~Institute of

Molecular Pharmacology, Medical School Hannover, D-30623 Hannover, Germany; §Institute of Clinical Immunology and Transfusion Medicine, University of Leipzig, D-04103 Leipzig, Germany

(Received 30 November 1995 and inJinaljorm 1 April 1996)

Abstract -The effects of treatment with sinomenine, a pure alkaloid extracted from the chinese medical plant Sinomenium acutum, were investigated in rat adjuvant arthritis (AA) and antigen-induced arthritis (AIA). In AA, long-term, intraperitoneal (i.p.) treatment induced significant improvement of arthritic score, hind paw swelling, body weight and erythrocyte sedimentation rate (ESR) beginning past the clinical peak of the disease. In acute AIA, short and middle-term treatment with sinomenine around and following induction of arthritis induced a dose-dependent decrease of both joint swelling and ESR, starting after the peak of arthritis, and a significant reduction of joint destruction on day 3. There was no rebound of the arthritic signs following suspension of treatment. Lony-term treatment of chronic AIA partially ameliorated clinical parameters and significantly counteracted joint destruction. Maximal plasma concentrations of 22.5 pg/ml, fast wash out (half- life 4.24___0.99 h; mean_+ S.E.M.) and no evidence of accumulation of sinomenine were observed following single or repeated i.p. injection of 150 mg/kg. In vitro, sinomenine markedly inhibited proliferation of synovial fibroblasts from AIA or normal rats, both at rest and following activation with either transforming growth factor/~2 (TGF-/~2) or interleukin-1/3 (IL-I/~). The effect was dose-dependent and half-maximal inhibition of proliferation occurred at 20.6/~g/ml, that is, within the in vivo therapeutic range of the drug. Late therapeutic effects of sinomenine in rat arthritic models despite early start of treatment may be related to its anti- proliferative effects on synovial fibroblasts in addition to its previously reported anti-inflammatory properties. ~'~ 1997 International Society for Immunopharmacology.

Keywords: adjuvant arthritis, antigen-induced arthritis, synovial fibroblasts.

The alkaloid sinomenine (7,8-didehydro-4-hydroxy- 3,7-dimethoxy-17-methylmorphinane-6-one) is extracted from the Chinese medical plant Sinomenium

acutum (Liu et al., 1994b), which has been utilized by Chinese doctors for over 2000 years to treat various rheumatic diseases (Shen & Shen, 1596). In open clinical trials the therapeutic efficacy of purified sinomenine in patients with rheumatoid arthritis (RA) has recently been confirmed (Shi et al., 1985; Ke & Xiu, 1986).

Previous pharmacological studies have demon- strated that sinomenine inhibits inflammatory reactions caused by various phlogistic agents (Irino, 1958; Chang et al., 1964; Hua &Chen , 1989). Recent in vitro investigations have shown that sinomenine significantly inhibits the production of prostaglandin Ez,

leukotriene C4, or nitric oxide by macrophages (Liu et al., 1994b), as well as the proliferation of human peripheral blood mononuclear cells and mouse spleen cells (Liu et al., 1994a). In vivo studies have revealed that sinomenine markedly ameliorates proteinuria and survival time of MRL- lp r lupus-prone mice (Emmend6rffer et al., submitted elsewhere). These effects suggest that sinomenine may act as an anti- rheumatic agent through its anti-inflammatory properties. However, a direct anti-rheumatic effect in experimental models of arthritis has yet to be proven. In the present study, therefore, it was investigated whether treatment with sinomenine influences the course of either adjuvant- or antigen-induced arthritis (AA and AIA, respectively), two experimental models of arthritis in the rat (Brackertz et al., 1977a,b:

¶ Raimund W. Kinne, Institute of Clinical Immunology and Transfusion Medicine, Delitzscher Str. 141, D-04129 Leipzig, Germany.

529

530 L. LIU et al.

Billingham, 1990; Griffiths, 1992; Buchner et al., 1995). Different protocols were performed in terms of dose, initiation and duration of treatments.

Because oral and intraperitoneal (i.p.) admin- istrations of sinomenine yield similar effects in vivo (Hojo et al., 1985), the i.p. route was chosen in this study in order to minimize variability in gas- trointestinal absorption of the drug. This was par- ticularly important in view of the fact that the two experimental arthritis models have rather different features, i.e. AA is a systemic disease associated with anorexia and severe wasting syndrome, whereas A1A is an exquisitely local synovitis with little systemic involvement. In addition, absorption and elimination of the drug proceed very rapidly following both oral and intravenous (i.v.) administration (Junbao et al., 1992), resulting in a bioavailability of approx. 73% of the orally injected dose.

As recent reports indicate that synovial fibroblasts may contribute to destruction of cartilage and bone in arthritic joints (Ritchlin & Winchester, 1989; Tra- bandt et al., 1990; Zvaifler & Firestein, 1994; Kinne et al., 1995), thereby representing additional targets of therapy, the effects of sinomenine on synovial fibroblasts in culture, both at rest and following stimu- lation with cytokines, were further explored.

EXPERIMENTAL PROCEDURES

were freshly prepared in Ringer's solution for i.p. injection into rats, or in DMEM with 0.1% dime- thylsulfoxide (DMSO) for /n vitro experiments with synovial fibroblasts.

Animals

Female Lewis rats (140-170 g; 7-10 weeks old) were purchased from Charles River Laboratories, Sulzfeld, Germany. They were housed two per cage, with food and water ad libitum, and maintained at 2 1 C tem- perature, 60% relative humidity, and a 12 h light/dark cycle. Two weeks were allowed after arrival before starting the experiments.

ADJUVANT ARTHRITIS

Induction

Rats were intradermally injected into the tail base with 0.1 ml of a suspension containing 500/~g heat- killed MT in paraffin oil (Pearson & Wood, 1963).

Treatment

AA rats (n = 8-9 for each group) were treated by daily i.p. injection of different doses of sinomenine (15, 50 or 150mg/kg) from the day of induction of arthritis (day 0) until day 36. Matched volumes of Ringer's solution were used as negative control.

Materials

Heat-killed Mycobacterium tuberculosis (MT) was obtained from Difco, Detroit, Michigan, U.S.A.; paraffin oil from Riedel de H/ien, Seelze, Germany; methylated bovine serum albumin (mBSA) from Serva & Co., Heidelberg, Germany. Dulbecco's modified Eagle medium (DMEM) with Glutamine X-1 was purchased from Life Technologies, Paisley, U.K.; collagenase P from Boehringer Mannheim, Mannheim, Germany; trypsin, fetal calf serum (FCS), and N- 2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) from Gibco, Eggenstein, Germany. Trans- forming growth factor f12 (TGF-fl2) was obtained from Biermann, Bad Nauheim, Germany. Human recombinant interleukin-lfl (IL-lfl) was purchased from Genzyme, Cambridge, U.K. [3H]thymidine (spec. act. 25 Ci/mmol) was purchased from Amer- sham, Braunschweig, Germany. Sinomenine (7,8- d i d e h y d r o - 4 - h y d r o x y - 3 , 7 - d i m e t h o x y - 1 7- methylmorphinane-6-one; MW 329.4) was kindly provided by L. D. Zhang, Shanxi Institute of Pharma- ceutical Industry, Xi-an, China. Purity of the drug was confirmed by reverse phase-HPLC as described elsewhere (Liu et al., 1994b). Solutions of sinomenine

Clinical and hematoloyical parameters

The effects of treatment were monitored by daily measurements of arthritic score, hind paw volume and body weight. Lesions of the four paws were graded from 0 to 4 according to the extent of both erythema and edema of the periarticular tissue (Wood et al., 1969); 16 was the maximum arthritic score per animal. The hind paw volume was measured using a ple- thysmometer chamber (U. Basile, Comerio, Italy), and expressed as the mean volume of both hindpaws. The body weight of the rats was monitored with a 1 g precision balance (Sartorius, G6ttingen, Germany). Erythrocyte sedimentation rate (ESR), as well as total leucocyte count were measured in blood samples obtained from the tail vein on days 13, 24, 30 and 37. The ESR was determined as described previously (Buchner et al., 1995). The total leucocyte count was determined in peripheral blood using a Sysmex E-5000 M/CS cell counter (Sysmex TOA Medical Electronic, Hamburg, Germany).

Histopatholoyical evaluation

After killing the rats, both ankle joints were excised and the skin removed. The samples were frozen in

Sinomenine in Arthritis Models 531

methylbutane cooled in liquid nitrogen and embedded in 8% gelatine. Cryosectioning of undecalcified joint samples was performed as described previously (Van Noorden & Vogels, 1986), using a motor-driven Bright cryostat equipped with a retraction microtome. Serial 8-#m-thick sections were attached to adhesive tape and stained with Giemsa solution. Grading of joint destruction, inflammatory infiltration, fibrosis and bone formation was performed by two independent observers using a semiquantitative scale from 0 to 4; in the case of joint destruction 0 = no joint destruction; 1 = destruction of < 10%, 2 = 10-25%, 3 = 2 5 50% or 4 = >50% of cartilage and bone cross sections; in the other cases 0 = none, 1 = weak, 2 = moderate, 3 = strong and 4 = very strong degree of inflammatory infiltration, fibrosis or bone formation. Following onset of arthritis on days 9 13 and peak of joint inflammation on days 21-25, the clinical signs declined while progressive cartilage and bone destruction, together with fibrosis, led to complete ankylosis of the joints (data not shown).

each group). After suspension of this treatment it was evaluated whether a clinical rebound occurred. (3) Long-term treatment past the peak of arthritis (150mg/kg from day 8 to day 41; n -- 6). A flare-up was induced on day 42 to analyze whether treatment with sinomenine influenced its severity; the rats were subsequently killed and dissected on day 51.

Clinical and hematologieal parameters

The degree of arthritis was monitored by daily measurements of the knee width with a caliper; swelling was expressed as the difference between arthritic and control knees. Blood samples were taken on day 2 and 8 after induction of arthritis from vehicle- and middle-term treated AIA rats. ESR and blood total leucocyte counts were determined as described above for AA.

Histopathological evaluation

The histopathological evaluation of the arthritic knee joint was performed as described above for the AA model.

A N T I G E N - I N D U C E D ARTHRITIS

Immunization

On days -21 and - 1 4 rats were subcutaneously injected with a total volume of 2.0 ml of a suspension containing equal volumes of 0.5 mg/ml mBSA in phos- phate-buffered saline (PBS) and Freund's complete adjuvant (2 mg MT/ml; Griffiths, 1992).

Induction t?[primary arthritis

On day 0, 0.1 mg mBSA in 50 #l of PBS were intra- articularly (i.a.) injected into the right knee (Griffiths, 1992). The left knee joint was injected with PBS only and used as control.

Induction ~f exacerbations (flare-ups)

On day 42, rats were re-injected into the right knee joint with 0. I mg mBSA in 50/~1 of PBS. The left joint was injected with PBS and used as control Buchner et al., 1995).

]r'reatnlen l

AIA rats were administered sinomenine 1.p. daily according to three different protocols: (1) short-term treatment before and during the peak of primary arthritis (150mg/kg from day - 1 to day 6; n=6); on day 7 the animals were killed and dissected for histological examination. (2) Middle-term treatment before and during the peak of primary arthritis (1, 15, 50 or 150mg/kg from day - 1 to day 14; n = 5 for

P LASMA CLEARANCE OF SINOMENINE

Surgery

The day before the experiment a 7-cm-long poly- ethylene catheter (Bardikath, Bard Limited, Sunder- land, U.K.) was inserted into the right common jugular vein of normal rats to allow sampling of central venous blood (Bodziony & Schwille, 1985).

Dru 9 administration

Two protocols were performed: (1) Single i.p. administration: 150mg/kg sinomenine into normal female rats (n = 11 ); central venous blood was sampled 0.25, 0.5, I, 1.5, 3, 6, 12, 24, 48 and 96h following administration. (2) Repeated i.p. administration: in six of the above rats a second dose of 150mg/kg was administered 24 h later; the blood was drawn according to the previous sampling schedule.

Measurement ofplasma levels of sinomenine

Sera (50pl) were mixed with 200/11 methanol (HPLC grade), vigorously shaken and finally cen- trifuged (16,000 g, 5 min, RT). Reverse phase-HPLC (LC41D, Bruker, Bremen, Germany) was performed using a LiChroCART 250/4 column (Supersphere 100 RP-18 end-capped; Merck, Darmstadt, Germany). The mobile phase consisted of methanol/ acetonitrile/sodium acetate (0.2 mol/l)/water (9/9/12/70, v/v) with a pH of 5.5. The flow rate was

532 L. LIU et al.

held constantly at 1.0 ml/min and the UV absorbance was set to 234nm. Authentic sinomenine (Aldrich, Steinheim, Germany) was used to generate a standard curve. The recovery rate of sinomenine from serum was 97.2_+3.9% (mean_+ S.E.M.; n = 4).

Plasma half-lives were determined using the TOPFIT computer program, version 2.0 (Schering, Berlin, Germany).

ISOLATION AND CULTURE OF SYNOVIAL FIBROBLASTS

Fibroblast isolation

Synovial membranes were dissected from the knee joints of either normal rats, or AIA rats on day 2 following induction of arthritis. The tissues were immediately transferred into sterile PBS containing 0.1% trypsin and finely minced with scissors. The resulting suspension was mixed and digested for 30 min at 37°C with 0.1% trypsin in PBS, under con- stant stirring. After removal of the trypsin/PBS, freshly prepared 0.1% collagenase P in DMEM/10% FCS was added at 3T'C for 2h under shaking to further digest the tissues. The cell suspension was then filtered through a sterile steel sieve and the cells were collected by centrifugation (338 g, 6min, RT). After washing twice with serum-flee medium, the cells were resuspended and cultured in DMEM supplemented with penicillin (100 U/ml), streptomycin (100#g/ml), HEPES (0.05 M), and 10% FCS (hereafter complete medium) in 50ml culture flasks (Greiner, Erlangen, Germany) at 37°C in 5% CO2. The adherent cells became uniformly confluent in 7-10 days; they were then trypsinized (0.25% trypsin/0.02% EDTA in PBS) and further passaged every 5 to 6 days approxi- mately at a 1:4 dilution (Lafyatis et al., 1989). Cells from passage 1 to 5 were used for proliferation assays. Cells at passage 2 contained virtually no macrophages or lymphocytes, as determined by immuno- histochemical staining of cytospins (data not shown).

Fibroblast proliferation

Cells derived by trypsinization (5 x 103/0.2 ml/well) were seeded in 96-well microtiter flat bottom tissue culture plates (Nunc, Wiesbaden, Germany) in complete medium and allowed to adhere overnight. The medium was replaced by serum-free medium, and the cells were cultured for an additional 3 days to deprive them of growth factors. Following removal of the old medium, new medium with 1% FCS was added, containing increasing doses of sinomenine (5.1, 10.3, 20.6, 41.2, 82.4, 164.7 or 329.4/~g/ml, that is 15.5/tM to l mM in 0.1% DMSO), and either TGF-fl2

(20 ng/ml) or IL-1/~ (100 U/ml; 17 ng/ml). In the case of resting fibroblasts or in control cultures, an equal volume of medium with 1% FCS and 0.1% DMSO without TGF-/~2 or IL-I/3 was added. The incubation was continued for 3 days in each case. Then 1/~Ci/well [3H]thymidine was added. Eighteen hours later the medium was gently aspirated and the cells trypsinized with 50/~1 of 0.25% trypsin/0.02% EDTA for 15 min. After detachment, the cells were harvested onto fiber filters with a cell harvester. [3H]thymidine incorporation was measured by/~-scintillation counting.

Time course of the anti-proliJerative effects

Synovial fibroblasts were cultured with TGF-fl2 (20 ng/ml) as described above; 1 mM (329.4 #g/ml) sinomenine was added at time points of 0, 0.5, 1, 2, 4, 8, 24 and 48 h following the beginning of incubation with TGF-/32.

Cytotoxicity and reversibility of anti-proliJerative effects

To determine cytotoxic effects of sinomenine, synovial fibroblasts were cultured with TGF-/32 (20ng/ml) at a cell density of l x 104/96-well plate in the presence of increasing doses of sinomenine as above. After incubation for 3 days, the medium was aspirated and 25/fl of 0.25% trypsin/0.02% EDTA together with 25/fl of trypan-blue were added to each well for cell counting. The ratio of viable:dead cells was calculated in each well.

To determine the reversibility of the effects of sinomenine, the fibroblasts were incubated for 24 h with increasing doses of sinomenine as above. The cells were then trypsinized and washed twice with drug- free DMEM/1% FCS; the cells were then seeded again in 96-well-plates and incubated in complete medium for 48 h in the absence of sinomenine or following re- addition of increasing doses of sinomenine as above; subsequently, [3H]thymidine incorporation was determined.

STATISTICS

The non-parametric Mann Whitney (U) test was applied to analyze differences in clinical, serological and histological parameters of AA and AIA as well as in the in vitro results with synovial fibroblasts. Stat- istical significance was accepted for P ~< 0.05.

RESULTS

Adjuvant arthritis

In agreement with previous studies (Pearson & Wood, 1963), arthritis began to develop from day 9


to 13 following injection of MT (Fig. IA, B, C); the arthritic score peaked from day 21 to 25 and declined thereafter (Fig. 1A), while the paw volume reached a plateau as late as day 29 (Fig. I B). The ESR and the total leucocyte count were also augmented (Fig. 2A, B), and the diseased animals displayed marked loss of body weight (Fig. 1C).

While i.p. treatment with low doses of sinomenine (either 15 or 50mg/kg) was ineffective, daily administration of 150 mg/kg from day 0 to 36 significantly ameliorated clinical parameters of AA, inducing significantly decreased arthritic scores from day 23 (Fig. 1A), reduced hind paw swelling from day 27 (Fig. IB), and improvement of the body weight from day 23 (Fig. IC) in comparison to vehicle-treated arthritic animals. Sinomenine also induced a significant decrease of the ESR on days 24, 30 and 37 (Fig. 2A), but did not influence the total leucocyte count (Fig. 2B). There was no significant influence of the treatment on joint destruction, inflammatory infiltration or bone formation in the arthritic joint on day 38 of AA, while fibrosis of the synovial membrane was significantly increased by treatment with 150mg/kg sinomenine, as analyzed by examination of Giemsa- stained histological sections (data not shown).

Anti,qen-induced arthritis

Following induction of AIA, swelling of the knee injected with mBSA increased rapidly, reaching a peak 2 days following injection; it decreased within 2 weeks to a low grade, chronic swelling (Fig. 3A, Fig. 4A, and Fig. 5A). There was an elevation of the ESR in the acute phase of the disease (day 2; Fig. 4B), and an augmentation of total leucocyte counts in comparison to normal rats (Fig. 4C).

A flare-up was successfully induced on day 42 by a second i.a. injection of mBSA; time course and severity of joint swelling were comparable to those of the primary arthritis (Fig. 5A).

Short-term, daily i.p. treatment of acute AIA with 150 mg/kg from day - I to day 6 led to marked amelioration of the clinical signs of A1A, with significant decrease of articular swelling from day 6 (Fig. 3A). This treatment schedule also significantly reduced joint destruction in the arthritic knee joint on day 7, while inflammatory infiltration, fibrosis or bone formation were not significantly altered (Fig. 3B).

While rniddle-term, daily i,p. treatment of acute AIA with low doses of sinomenine (1 or 15mg/kg from day - 1 to day 14) was ineffective (Fig. 4A), treatment with 50 or 150 mg/kg from day - 1 to day 14 led to significant amelioration of the clinical signs of AIA from day 8 or day 3, respectively (Fig. 4A).

Suspension of treatment after 2weeks was not fol- lowed by rebound of arthritis. Both clinically effective doses (50 and 150mg/kg) significantly counteracted the transient elevation of ESR already on day 2 (Fig. 4B), that is, one day before the onset of clinical improvement. On day 8 the ESR of both sinomenine- and vehicle-treated arthritic animals had returned to almost normal levels (Fig. 4B). Middle-term-treatment of acute AIA with sinomenine did not have any influence on the total leucocyte count at any time point (Fig. 4C).

Long-term treatment of chronic AIA with 150mg/kg from day 8 to 41 reduced the swelling of the knee joint during chronic phase and flare-up, although in both cases the differences were not statis- tically significant due to the large variation in the vehicle-treated A1A group (Fig. 5A). There was a significant reduction of the degree of joint destruction on day 51 of the disease (Fig. 5B). Inflammatory infiltration, fibrosis and bone formation were not altered by this treatment schedule (Fig. 5B).

Plasma clearance ~?f sinomenine

Because of the rapid absorption and elimination of sinomenine (Junbao et al., 1992), the plasma clearance of the drug was investigated in normal rats, on the basis of the assumption that arthritis per se would not greatly influence this parameter.

Maximal plasma concentrations of sinomenine (22.5 pg/ml; 68.3 pM) were reached within 30 rain following both first or second i.p. injection of 150mg/kg into normal female Lewis rats (Fig. 6). Half-maximal plasma concentrations were reached within 2h following injection. The final plasma half-life of the drug was calculated as 4.24_+0.99h (mean+S.E.M., n= 17). Following both first and second injection, plasma levels of sinomenine reached less than 2 #g/ml within 12 h, and were below the detection limit at 24 h; there was, therefore, no indication for accumulation of the drug.

Synovial fibroblast prol!/eration

Sinomenine markedly and dose-dependently decreased [~H]thymidine incorporation of both resting and TGF-fl2- (20ng/ml) or IL-l/~-stimulated (100U/ml) synovial fibroblasts. Half-maximal inhibition of proliferation occurred at concentrations of 62.5#M (20.6pg/ml; Fig. 7A). There were no differences in the responses of synovial fibroblasts obtained from normal or AIA synovial membranes (data not shown). The concentration able to inhibit proliferation of fibroblasts in vitro by 50% was similar to the plasma concentration achieved in vivo 1.5h

.= 0 0

0

P

m

14

12

10

8"

6 -

4 -

2

0 0

i u m m i

10 20 30 40

u

E

E 0 >

O=

"0 e- .1¢

1.3

1.2"

1.1"

1.0"

0.9

0.8

0.7

0.6

0.5 0

~ ~ . k 'k "k '/r

IB

10 20 30 40

1 9 0

~ t ~ 'k 'k'k /

~_. 180 . " ~ ~71" ~'

Z- ._=

1 7 0 - _ T

"Q 1 6 0 '

150 , , , 0 10 20 30 40

treatment r I

time (days) Fig. 1. Effects of different doses of sinomenine on clinical parameters of AA. Rats were treated i.p. daily with 15mg (Ul, n = 9), 50rag ( ~ , n = 9), or 150 mg/kg (&, n = 8), respectively, from the day of the induction of arthritis (day 0) until day 36. The effects of each dose on arthritic score (A), hind paw volume (B), and body weight (C), were compared to those of treatment of AA with vehicle ( i , n = 9). Significant amelioration of arthritic score, hind paw volume and body weight were observed with the highest dose starting from day 23 (A), 27 (B) and 23 (C), respectively. Data are means+S.E.M.;

*P ~< 0.05; **P ~< 0.01 in comparison to vehicle-treated AA rats.

A J ¢

E v

r r

W

20

15

10

0 0

Sinomenine in Arthritis Models

i m

10 20 30 40

535

E o o

x

U )

o

_¢

25

20

15

10 0

IB

10 20 30 40 treatment ~' v I

t i m e ( d a y s )

Fig. 2. Influence of different doses of sinomenine on ESR (A) and total leucocyte count (B) in AA. Treatments, symbols and numbers of rats are the same as in Fig. 1. Peripheral blood was drawn on days 13, 24, 30 and 37. Treatment with 150 mg/kg (A/n = 6), but not with lower doses, induced a late, significant reduction of ESR in comparison to vehicle-treated rats (A). Treatment with either dose of sinomenine had no influence on the total leucocyte count (B). Data are means+_ S.E.M.;

*P ~< 0.05; **P ~< 0.01 in comparison to vehicle-treated AA rats.

following single i.p. injection of 150 mg/kg sinomenine (Fig. 7A), that is, within the therapeutic range of the drug.

In time course experiments, sinomenine (1 raM; 329.4#g/ml) only reduced [3H]thymidine incorporat ion of TGF-/~2-stimulated fibroblasts to a significant degree when administered within the first 24 h after initiation of cell culture (Fig. 7B).

The percentage of viable, TGF-/~2-activated fibroblasts, as determined by trypan blue exclusion, remained at >_-90% upon exposure to concentrations of up to 1 mM sinomenine (329.4 #g/ml; Fig. 8A).

Synovial fibroblasts exposed to different concentrations of sinomenine (15/~M to l mM; 5.1 329.4/~g/ml) for 24 h, and subsequently cultured in the absence of sinomenine for additional 24 h, showed a [3H]thymidine incorporat ion indistinguishable from that of control cultures never exposed to sinomenine (Fig. 8B).

Re-addit ion of increasing doses of sinomenine as above following a washing period of 24 h resulted in significant inhibition of cell proliferation, comparable to that seen upon initial addition of different doses of sinomenine (Fig. 8B).

536 L. LIU et al.

2

1

0 -1

P¢

u

0 1 2 3 4 5 6 7

treatment

time (days)

A

E E

v

o~ c ,m

.~ 3 0

t~ .0 2 o~ o

IB

I

Joint Destruction Inflammatory Infiltration

Fibrosis Bone Formation

Fig. 3. Effects of short- term treatment with sinomenine on clinical (A) or histopathological (B) parameters of acute AIA. Rats were treated i.p. daily with 150 mg/kg from day -- 1 to day 6 following induction of arthritis. Swelling of the knee joints from sinomenine-treated rats (A, n = 6) was compared to that of vehicle-treated AIA rats (11, n = 6). Following treatment, a significant amelioration of clinical parameters was observed after the peak of arthritis, i.e. from day 6 (A). In addition, treatment of AIA (empty bars) significantly counteracted joint destruction in the arthritic knee joint on day 7 (B). Data are

means _+ S.E.M.; *P ~< 0.05; ** P ~< 0.01 in comparison to vehicle-treated AIA rats.

Fig. 4. Effects of middle-term treatment with sinomenine on clinical (A) or hematological (B, C) parameters of acute AIA, Rats were treated i.p. daily with I mg/kg (O, n = 5), 15 mg/kg (11, n = 5), 50 mg/kg (A, n = 5) or 150 mg/kg (A, n = 5) from day - 1 to day 14. Clinical scores of sinomenine-treated rats were compared to those of vehicle-treated AIA rats (11, n = 5). Following treatment, a significant, dose-dependent amelioration of clinical parameters was observed after the peak of arthritis, i.e. from day 3 (A). Two and 8 days after induction of arthritis peripheral blood was drawn from the animals. ESR (B) and the total leucocyte count (C) were compared to those of vehicle-treated AIA rats (11, n = 5). A significant decrease of the ESR on day 2 was observed following treatment with 50 or 150mg/kg sinomenine (B). On day 8 there remained no differences among groups (B). There were no changes in the total leucocyte count following treatment (C), Data are

means + S.E.M.; *P ~< 0.05; **P ~< 0.01 in comparison to vehicle-treated AIA rats.

0 -1

10

I

,k

6

q) IlJ


- r

4 9 14 19 treatment

time (days)

24

18

537

day 2 day 8

2

0 0

x

U3

0 0 (J

_e

25

20"

10

T

day 2 day 8

Ic

538 L. LIU et al.

DISCUSSION

Treatment with sinomenine induced significant amelioration of clinical and hematological signs in both A A and AIA, two widely accepted experimental arthritis models in the rat (Billingham, 1990; Griffiths, 1992; Buchner et al., 1995). The effects of sinomenine developed late in the course of the disease, i.e. from day 23 in AA and from day 3 in AIA, in spite of continuous treatment from the day of induction of

arthritis. In AIA, the treatment was efficacious whether begun in the acute or in the chronic phase of the disease; however, due to large variation in the control group, amelioration following treatment in the chronic phase did not attain statistical significance. These effects indicate that the drug predominantly influences the late phase of experimental arthritides, i.e. after the clinical peak of arthritis, which is day 21 for AA and day 2 for AIA.

Most anti-rheumatic agents, both steroids and non-

3

0 - - 0

induction of primary arthritis

10 20 30 40 50 treatment -'~

t i m e ( d a y s ) induction of flare-up

3 O O

m t~

.o 2

O

¢0 om 1- 1

i

Joint Destruction Inflammatory Infiltration

Fibrosm

IB

Bone Formation

Fig. 5. Effects of lony-term treatment with sinomenine on clinical (A) and histopathological (B) parameters of chronic A1A. Rats were treated i.p. daily with 150mg/kg from day 8 to day 41 following induction of arthritis. On day 42 an exacerbation was induced by i.a. injection of the inducing antigen mBSA. The knee swelling of sinomenine-treated AIA rats (A, n = 6) was compared to that of vehicle-treated AIA rats ( I , n = 6). Improvements in chronic phase and flare-up following post- peak treatment with sinomenine (A) did not attain statistical significance due to large variability in the vehicle-treated group. There was, however, a significant counteraction of joint destruction by treatment (open bars) on day 51 of AIA (B). Data

are means_+ S.E.M.: *P ~< 0.05 in comparison to vehicle-treated AIA rats.

30

2 o - c

~ o E e" 10

~.0 S


i i i r i

539

0 2 4 6 8 10 12

t ime (h) fo l lowing i.p. in ject ion Fig. 6. Plasma concentrations of sinomenine 0.25, 0.5, 1.0, 1.5, 3.0, 6.0, 12.0 and 24.0 h following single (A) or repeated (A) i.p. injection of 150 mg/kg into normal female Lewis rats. Peak plasma levels of the drug were reached within 30 min following i.p. injection. Half-maximal plasma concentrations were reached within 2 h following injection; at 24 h the concentration of

the drug was already below detection limits.

steroids, as well as monoclonal antibodies directed against T-cell surface molecules (e.g. anti-CD4, OX35; anti-pan T-cell, W3/13; anti-TCR ~/~, R73), are effective in the early phase of experimental arthritides, when treatment is started around the induction of arthritis (Newbould, 1963; Billingham et al., 1990; Larsson et al., 1985; Buchner et al., 1994). The effects induced by sinomenine do not resemble those of any of the above described classes of agents, in particular those aimed at modulating the functions of inflammatory cells. Therefore, the inhibiting effects of sinomenine on lymphocyte proliferation and macrophage function observed in vitro (Liu et al., 1994a; Liu et al., 1994b), as well as the early reduction of the ESR in the AIA model in vivo (Fig. 4A), may not be as relevant for its therapeutic efficacy. In contrast, the late ben- eficial effects of sinomenine may be related to its action on other cell types more relevant during the chronic phase of arthritis, The in vitro effects on fibroblasts observed in the present study suggest that synovial and pannus tissue fibroblasts may be good candidates for such effects.

The present in vitro investigations demonstrate that sinomenine significantly inhibits [3H]thymidine incorporation of synovial fibroblasts from AIA rats, either in a resting state or after activation with TGF-/~2 or IL-1/~. There was a good correlation between the plasma levels of sinomenine following single or repeated i.p. injection of 150 mg/kg sinomenine into normal rats and the concentrations used for in vitro

experiments. This suggests that the effects of sinomenine on fibroblasts observed in vitro may be responsible for the late therapeutic effects of the drug in vivo. In fact, histological examination of joints from AIA rats, in particular following long- term t rea tment during the chronic phase, revealed a significant counteraction of joint destruction, a predominant feature of this phase of A1A (Buchner et al., 1995; Lafyatis et al., 1989). These findings are also consistent with previous studies showing that sinornenine can inhibit the formation of granulation tissue (Saeki et al., 1975a; Saeki et al., 1975b), and underline the pathogenetic importance of fibroblast-like cells for the progress of joint destruction in chronic arthritis (Ritchlin & Winchester, 1989; Trabandt et al., 1990: Zvaifler & Firestein, 1994; Sack et al., 1994; Kinne et al., 1995). Interestingly, counteraction of joint destruction by treatment with sinomenine was only observed in AIA, the experimental model with a more chronic, smouldering time course, indicating that sinomenine may be more effective when acting on a long- term basis against chronically activated, aggressive fibroblasts.

There was no evidence of toxicity in vivo despite the use of relatively high doses and the extended length of treatment with sinomenine. In contrast, the loss of body weight typical of AA was significantly counteracted, indicating not only good tolerance to the drug, but also good general conditions of the animals. Pat- ients with RA also show good tolerance to long-term

100

¢ = C

g 80

I,= o

60

• ; oo I¢ ~,..

o 40

E o

~, 2o -1-

o lO

t 1 o0 1 ooo

I*

540 L. LIU et al.

s i n o m e n i n e ( ~ g / m l )

3 t- O

o ~

o

8o° 2 ._~

x e..

E ~ 1 > , p .

-I-

T

C1 C2 0 0.5 1 2 4 8 24

t i m e (h) f o l l o w i n g addi t ion of TGF-[32

nsS

48

Fig. 7. (A) Inhibitory effects of sinomenine on synovial fibroblast proliferation & t, itro. The incorporation of [3H]thymidine was assessed in resting (11) synovial fibroblasts, or following activation with TGF-//2 (A; 20 ng/ml) or IL-1/~ ((3; 100 U/ml), in the presence of sinomenine at the indicated concentrations. The percent inhibition of incorporation in comparison to controls was calculated following 3days of incubation. Control values were 629+385cpm for resting fibroblasts, 2435+585cpm for TGF-//2-stimulated fibroblasts and 3075+339cpm for IL-1/~-stimulated fibroblasts. Data are means_+ S.E.M. of triplicate incubations. The arrow in A indicates the plasma level of sinomenine 1.5 h following single i.p. injection of 150 mg/kg. Sinomenine (EDs0 62.5 #M; 20.6/~g/ml) induced a dose-dependent inhibition of fibroblast proliferation. (B) Time course of the anti-proliferative effects of 1 mM sinomenine on synovial fibroblast cultured with TGF-//2 (20 ng/ml). Data are means+S.E.M, from sextuplicate incubations. CI, control incubation in the absence of both TGF-/~2 and sinomenine; C2, control in the presence of TGF-//2 alone. Significant differences between the CPM of C2 and cell cultures exposed to increasing concentrations of sinomenine in the presence of TGF-//2 are indicated. ***P ~< 0.005; n.s. not

significant.

t rea tment with s inomenine, with skin rash and i tching appear ing only at the highest doses (Ke & Xiu, 1986; Shi et al., 1985).

The absence of toxic effects of s inomenine on fibroblasts in vitro, and the complete reversibility of

the inhibi t ion of f ibroblast proliferat ion, demons t ra te tha t the drug is no t cytotoxic for these cells, con- firming the lack of toxicity on Jurkat , SV80 and HeLa cell lines (Liu et al., 1994a). There was evidence tha t s inomenine had to be adminis tered within the first

Sinomenine in Arthritis Models 54t

120

lOO t 80

• ; o 60- :Eo "~= ~ 40 > . ",=.,

JC

X, 20

100

80

60

40-

20.

0

. \ ' N

0

m

.~-:_..__.-.

. - - - - - -

5.2

m

~ N\\\\N

10.3 20.6 41.2 82.4

sinomenine(gg/ml) 164.7

I,

N N ~ N N 3

~.xxx}? N N N N N N N

~ N N N N % N

N ~ X N N X %

329.4

_¢ J~

>

0 . . . . . . . . , . . . . . . . . , . . . . . . . .

10 100 1000

sinomenine (~g/ml) Fig. 8. Lack of cytotoxicity of sinomenine (A) and reversibility of its anti-proliferative effects (B) on synovial fibroblasts of AIA rats. In (A) synovial fibroblasts stimulated with TGF-,82 (20 ng/ml) were incubated in the absence or in the presence of increasing concentrations of sinomenine for 3 days; viable and dead cells were then counted following staining with trypan- blue. There were no cytotoxic effects at any dose of sinomenine. Data are means-I-S.E.M, of triplicate experiments. In (B), increasing concentrations of sinomenine were initially added for 24 h and then washed out for 24 h; the culture was contin ued for 2 days in the absence (©) or following re-addition (11) of 1 mM sinomenine. The mean cpm control value was 3121 + 908 (mean_+ S.E.M. of triplicate assays). Data are percentages of the cpm of fibroblasts incubated with medium only for the

entire duration of the experiment.

24 h fo l lowing initiation o f cell culture with TGF-/~2 in order to achieve a significant reduction o f [3H]thymidine incorporation. These results suggest that, in vitro, the drug may be effective in early signal trans- duct ion pathways necessary for fibroblast proliferation in synchronized cultures. Therefore, the drug could also inhibit the proliferation o f newly activated fibroblasts in vivo and thereby exert its therapeutic effects.

In view of its clearcut effects on synovial fibroblasts

in vitro and the potential importance o f activated fibroblast-like cells in human RA, s inomenine may represent an interesting therapeutic agent not only in experimental arthritides but also in human RA, either alone or in combinat ion with other anti- inflammatory agents (Panayi, 1993).

Acknowledgements We are grateful to J. Harzendorf, F. KIibel, B. M~iller and B. Niescher for skilful technical assist- ance, and to Dr. E. Palombo-Kinne, J. B. Wang, F. Beier,

542 L. LIU et al.

C. Pelegri, K. Wenk, R. Voll and Prof. H. Rodemann for valuable advice. We also thank L. D. Zhang, Shanxi Institute of Pharmaceutical Industry, China, who kindly provided pure sinomenine. Prof. Emmrich and Dr Kinne were sup-

ported by a grant from the German Ministry for Research and Technology (BMFT; FKZ 01245607/1 and 01VM9311/3); E. Buchner and C.B. Schmidt-Weber by the Graduierten-Kolleg Erlangen.

REFERENCES

Billingham, M. E. J. (1990). Models of arthritis and the search for anti-arthritic drugs. Anti-rheumatic Drugs, pp. 1~,7. Pergamon Press Inc., New York.

Billingham, M. E. J., Hicks, C. & Carney, S. (1990) Monoclonal antibodies and arthritis. Agents Actions, 29, 77-87. Brackertz, D., Mitchell, G. F. & Mackay, I. R. (1977a) Antigen-induced arthritis in mice. I. Induction of arthritis in various

strains of mice. Arthritis Rheum., 20, 841-850. Brackertz, D., Mitchell, G. F., Vadas, M. A. & Mackay, 1. R. (1977b) Studies on antigen-induced arthritis in mice. Ill. Cell

and serum transfer experiments. J. Immunol., 118, 1645-1648. Bodziony, J. & Schwille, P. O. (1985) Subcutaneous cannula in the jugular and femoral vein a tool for frequent blood

sampling and infusions in the rat. Z. Versuchstierk., 27, 29-32. Buchner, E., Schmidt, C., Lehmann, M., Emmrich, F. & Kinne, R.W. (1994). Systemic treatment of established rat

antigen-induced arthritis with a combination of anti-T-cell monoclonal antibodies and dexamethasone-containing lipid microspheres. 12th European Immunology Meeting, 14~17 June, Barcelona, Spain (abstract).

Buchner, E., Braeuer, S., Schmidt, C., Emmrich, F. & Kinne, R. W. (1995) Induction of flare-up reactions in rat antigen- induced arthritis. J. Autoimmunity, 8, 61 74.

Chang, S. S., Fu, S. X., LI, Y. S. & Wang, N. C. (1964) The pharmacology of sinomenine. I. The analgesic and anti-phlogistic actions and acute toxicity. A cta Pharmaceutica Sinaca, 4, 177-180.

Griffiths, R. J. (1992) Characterisation and pharmacological sensitivity of antigen arthritis induced by methylated bovine albumin in the rat. Agents Actions, 35, 88 95.

Hojo, H., Kondo, Y., Umeda, H., Tahira, T. & Hashimoto, Y. (1985) Effect of sinomenine on antibody responses in mice. J. lmmunopharmac., 7, 33~42.

Hua, H. R. &Chen, X. P. (1989) Study on mechanisms of sinomenine on analgesic and anti-inflammatory actions. J. Xian Medical University, 4, 346-349.

Irino, S. (1958) Effect of histamine releasers and anti-inflammatory drugs on the egg-white edema of rat hind paws in relation to skin histamine. Acta Med. Okayama, 12, 93 111.

Junbao, L., Quiang, F., Gengsheng, Z., Zhiquan, L. & Zhexun, W. (1992) Pharmacokinetic parameters and bioavailability of sinomenine hydrochloride. J. Xian Medical University, 4, 16-19.

Ke, X. Y. & Xiu, M. X. (1986) Treatment of rheumatoid arthritis by sinomenine. Beijing Yi Xue., 3, 186-187. Kinne, R. W., Boehm, S., Iftner, T., Aigner, T., Vornehm, S., Weseloh, G., Bravo, R., Emmrich, F. & Kroczeck, R. A.

(1995) Synovial fibroblast-like cells strongly express j un-B and c-fos proto-oncogenes in rheumatoid- and osteoarthritis. Stand. J. Rheum., 24(Suppl. 101), 121-125.

Lafyatis, R., Remmers, E. F., Roberts, A. B., Yocum, D. E., Sporn, M. B. & Wilder, R. L. (1989) Anchorage-independent growth of synoviocytes from arthritic and normal joints. J. clin. Invest., 83, 1267 1276.

Larsson, P., Holmdahl, R., Dencker, L. & Klareskog, L. (1985) In vivo treatment with W3/13 (anti-pan T) but not with OX8 (anti-suppressor/cytotoxic T) monoclonal antibodies impedes the development of adjuvant arthritis in rats. Immunol., 56, 383-391.

Liu, L., Resch, K. & Kaever, V. (1994) Inhibition of lymphocyte proliferation by the anti-rheumatic drug sinomenine. Int. J. Immunopharmac., 16, 685-691.

Liu, L., Riese, J., Resch, K. & Kaever, V. (1994) Impairment of macrophage eicosanoid and nitric oxide production by sinomenine, an alkaloid from Sinomenium acutum. Arzneim.-forsch./Drug Res., 44((11) 11), 1223-1226.

Newbould, B. B. (1963) Chemotherapy of arthritis induced in rats by injection of mycobacterial adjuvant. Br. J. Pharmac., 21, 127-136.

Panayi, G. S. (1993) The immunopathogenesis of rheumatoid arthritis. Br. J. Rheumatol., 32(Suppl. 1), 4~14. Pearson, C. M. & Wood, F. D. (1963) Studies of arthritis and other lesions induced in rats by the injection of mycobacterial

adjuvant. Am. J. Pathol., 42, 7345. Ritchlin, C. T. & Winchester, R. J. (1989) Potential mechanisms for coordinate gene activation in the rheumatoid synoviocyte.

Springer Sem. Immunopathol., 11, 219-234. Sack, U., Kuhn, H., Ermann, J., Kinne, R. W., Vogt, S., Jungmichel, D. & Emmrich, F. (1994) Synovial tissue implants

from patients with rheumatoid arthritis cause cartilage destruction in knee joints of SCID.bg mice. J. Rheumatol., 21, 10-16.

Saeki, K., Yokayama, J. & Wake, K. (1975a) Inhibition of granulation tissue growth by histamine. J. Pharmac. exp. Med., 193, 910 917.

Saeki, K., Yokayama, J. & Yamasaki, K. (1975b) Mechanism of histamine inhibition of granulation tissue growth. Jpn. J. Pharmac., 25(Suppl), 106-107.

Shen, N. & Shen, N.P.T., Quoted by Li, S.C. (1596). In Pen-Tsdto Kang-Mu 18 (in chinese) and quoted by Wong, K.C. & Wu, L.T. (1936). In History of Chinese Medicine, 2nd edn, p. 119. National Quarantine Service, Shanghai.


Shi, P. M., Ma, Z. X. & Zhang, W. Z. (1985) Therapeutic efficacy of sinomenine on 60 rheumatoid arthritis cases. Xien Yi Xue., 3, 292 293.

Trabandt, A., Aicher, W. K., Gay, R. E., Sukhame, V. P., Nilson-Hamilton, M., Hamilton, R. T., McGhee, J. R., Fassbender, H. G., Gay, S. (1990) Expression of the collagenolytic and ras-induced cysteine proteinase cathepsin L and proliferation- associated oncogenes in synovial cells of MRL/1 mice and patients with rheumatoid arthritis. Matrix, 10, 349-361.

Van Noorden, C. J. F. & Vogels, I. M. C. (1986) Enzyme histochemical reactions in unfixed and undecalcified cryostat sections of mouse knee joints with special reference to arthritic lesions. Histochemisto,, 86, 127-133.

Wood, F. D., Pearson, C. M. & Tanaka, A. (1969) Capacity of mycobacterial wax D and its subfractions to induce adjuvant arthritis in rats. Int. Arch. Allergy, 35, 456-467.

Zvaifler, N. J. & Firestein, G. (1994) Pannus and pannocytes. Alternative model of joint destruction in rheumatoid arthritis. Arthritis Rheum., 37, 783 789.

amelioration of rat experimental arthritides by treatment with the alkaloid sinomenine

Documents