Scand J Rheumatology 1989; Suppl. 17: 23-28

Effect of Nonsteroidal Antiinflammatory Drugs on Cartilage Destruction in Antigen Induced Arthritis in Mice

B.J. DE VRIES, W.B. VAN DEN BERG, E. VI’ITERS and L.B.A. VAN DE PU’ITE

Depclrtnient of Rheuniatology, University Hospital Sint Radboud, Nijntegen, The Netherlands

The nonsteroidal antiinflammatory drugs, salicylate, piroxicam and tiaprofenic acid, and the steroid pred- nisolone were investigated in a long-term study for their potential detrimental or kneficial effectson joint cartilage in mice with antigen induced monoarthritis. Daily drug treatment over a period of 4-7.5 weeks did not affect the histological characteristics of normal joints at all. Articular chondrocyte synthetic activity was even stimulated after salicylate and tiaprofenicacid treatment, but the significance of this finding is not yet clear. Cartilage damage, caused by inflammation in the knee joint, was neither markedly deteriorated nor attenuated by these drugs. Minor antiinflammatory properties as measured by decrease in edema using wmE-uptake and in the change of inflammatory cells were only evident with prednisolone, piroxicam and salicylate.

Key words: Articular cartilage, experimental arthritis, anti-inflammatory drugs.

B.J. dc Vries, Department of Rheumatolw, University Hmpual Shr Raa’bou~ l?O. Bar 9101, 6500 HB Nijmegen, The Netherlands

Introduction

Antigen-induced arthritis is an allergic type of joint inflammation, elicited by intra-articular injection of antigen into the knee joint of pre-immunized animals. This type of arthritis can be induced in mice with positively charged antigens like methylated bovine serum albumin (mBSA) (1-3). It bears histopathological resemblance to rheumatoid arthritis with regard to lymphocytic infiltrate in the synovial tissue and progressive cartilage destruction (44. One pharmacological study in this murine model has so far shown that the arthritis responds to steroids and some “second line drugs” (6,7). The nonsteroidal antiinflammatory drugs (NSAIDs), ibuprofen, flurbiprofen and indomethacin were without effect (6).

NSAIDs are commonly used in the treatment of various forms of arthritis, including rheu- matoid arthritis. However, recent experimental studies have shown that certain NSAIDs are detrimental to articular cartilage, especially osteoarthritic cartilage (8,9). On the one hand, drugs may theoretically exert a beneficial effects on arthritis by suppressing joint in- flammation, and therefore on inflammation-mediated cartilage damage (4,5). On the other hand, chondrocytes in already damaged cartilage of an arthritic joint may perhaps be more susceptible to adverse drug effects.

The present study was set up to investigate the overall effect of long-term treatment with various NSAIDs using the murine arthritis model. Animals with unilateral arthritis were used, which enabled a screen of the effect of drugs on both normal and arthritic cartilage.

This work was supported Ly Roussel BV (NL) and Pfizer International (USA).

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.

24 B.J de Vks, WB. van den Berg, E. Kttem and L B A . VM de Putte

Materials and methods

AniinaLr and mhritis induction Healthy, male C57l3110 mice, aged 6 7 weeks, were used for a 7.5 week drug treatment experiment (salicylate and tiaprofenic acid). Unilateral arthritis was induced in mBSA-immunized mice (9-12 weeks of age) as previously described (5) by intra-articular (i.a.) injection of 40 pg mBSA (in 6 pI physiological saline) into the right knee joint. These inflammatory monoarthritic mice were used to study the effects of orally administered drugs, over a 4week period, on both normal and arthritic car- tilage. 'Ib assess the severity of knee joint inflammation, %=-uptake measurements were performed as described (10). Mice were fed a standard pellet diet and tap water or drugcontaining water (see below) ad libitum. Prior to the start of each experiment, mice were matched by weight and age.

Drugs Mice received by oral gavage 250 pl of tap water (control) with dissolved drugs (prednisolone sodium phosphate, sodium salicylate) or with homogenically suspended drugs (piroxicam, tiaprofenic acid). In another experiment drugs were added to the drinking water. The drugs were dissolved in tap water or first in a negligible volume (less than 0.5% of the ultimate volume of drug water) of 0.15 M NaOH. All drug-waters had a pH of 8, identical to the pH of tap water at that time. Drink containers were protected from light; contents (70-150 ml) were routinely renewed every 24 h.

Histology and autoradwpphy The mice were killed by ether anaesthesia. The jointswere dissected and processed for histology. Total knee sections (7 pm) were prepared, mounted on gelatin-coated slides and stained with haematoxylin and eosin (H & E) or with Safranin 0. When chondrocyte function was studied, the mice were injected intraperitoneally with 2 pCi NaiSS04/g, 3 h before sacrifice. Paraffin sections of the total knee were dipped in K5 emulsion (Ilford, Basildon, &sex, England) and exposed for 3 weeks. After this period the slides were developed and stained with H & E. Cellular infiltration, exudation and proteoglycan loss (Safranin 0 loss) were scored semiquantitatively with light microscopy on five to seven serial sections.

In some experiments the patellae of 'SS-sulfate treated mice were excised and used to quantitate chondrocyte synthetic activity (11).

Results and Discussion

Effect of NSAZDs on developing arthritis Unilateral arthritis was induced in mBSA immunized mice by i.a. the injection of 40 pg mBSA into the right knee joint. Drug dosing commenced four hours prior to arthritis in- duction. From then on, drugs were given twice a day (at 9 a.m. and 6 p.m.), seven days a week for four weeks, by oral application. On days 7,14 and 21 99mR-uptake measurements were done to assess the severity of knee joint inflammation. Inflammation was expressed as the ratio of 99m'It-uptake in the right arthritic versus that in the left non-inflamed joint. O n day 28 the animals were sacrificed and inflammation and cartilage destruction were scored on whole joint sections. 'I?ible I shows the effects on 99mR-uptake. None of the drug treated groups showed a significant suppression compared to the group treated with water. How- ever, group l, i.e. mice not treated by daily gavage with water, showed significantly higher inflammation on day seven.

Histologic assessment of joint inflammation and cartilage damage is shown in nb le 11. No significant differences were seen between the various groups with respect to infiltrate, exudate and proteoglycan loss from the articular cartilage. The left non-arthritic joints of

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.

Effect of NonsteroidalAntiinjhnmatoy Drugs on Cartilage DestructiOn 25

Group

1 2 3 4 5

Table I: *%-uptake measurements.

?katment Infiltrate in* Exudate in* Cartilage PG** synovium joint space depletion

- l.1° 0.8 1.8 water 0.8 0.4 1.1 salicylate 1.4 0.5 1.3 piroxicam 0.6 0.3 1.0 tiaprofenic acid 1.2 0.3 1.5

Table ZZ: Histological grading of arthritis.

* Scored on a scale from &2. ** Scored on a scale from 0-4: 4 meaning complete depletion. a Values represent the mean calculated from groups of 12 mice.

all groups were completely normal in every aspect, indicating that 4 weeks drug treatment does not in any way cause gross cartilage damage.

The previous experiment suggested that the act of oral administration, twice a day, might of itself suppress joint inflammation. Group 1 showed higher 2-uptake on day 7 and carti- lage proteoglycan depletion also tended to be higher pointing to an antecedent inflamma- tion. The following experiment was set up to test this observation and to see whether the potential stress of daily oral gavage can be avoided by putting the drugs in the drinking wa- ter. Prednisolone was included as a positive control. a b l e I11 shows the =-uptake values in animals treated with drug waters. In one experiment *-uptake was only measured on day 4, in a second experiment measurements were done on days 4,7 and 14. The animals were sacrificed on day 28 and tissues processed for histology.

The results clearly indicated that stress induced by daily oral gavage did not necessarily Cause suppression of joint inflammation (comparison of groups 1 and 2). The R-uptake measurements in group 1 were even higher, although not reaching statistical significance. Furthermore, piroxicam (twice) and prednisolone (once) showed significant suppression on day 4 ( a b l e 111). On day 7 slightly decreased values were still noted for these two drugs, but no longer of reaching statistical significance anymore.

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.

26 B.J. de Vries, WB. van den Berg, E. Knem and L BA. van de Putte

B e a tmen t

, oralgavagec water salicylate piroxicam tiaprofenic acid prednisolone

Table HI: Effect of drugs in drinking water on arthritis.

Day 4b

1.97 f 0.32 1.86 f 0.16 207 f 0.42 1.41 f 0.12* 1.69 f 0.37 1.45 f 0.20.

Group

6

Day 4 Day 7

Group

I 2 3 4 5 6

1.73 f 0.21 1.55 f 0.25 1.39 f 0.18 1.27 f 0.07. 1.51 f 0.29 1.48 f 0.24

'Iteatment Conc. Water intake D W Mouse weight(mg): m g h l mVmouseltlh mgntg/Bh day-1 day28

oral gavage - 2.6 f 0.5 - 26f2 26f2 water - 3.3 f 0.4 - 27f2 29f2 salicylate 3 2.2 f 0.4' 255 26f2 28f2 piroxicam 0.2 2.7 f 0.4, 2o 28f2 28f2 tiaprofenic acid 0.2 3.3 f 0.3 25 27f2 28f2 prednisolone 0.02 3.2 f 0.3 25 28f2 25f2**

1.33 f 0.17 1.23 f 0.12 1.30 f 0.11 1.16 f 0.11 1.29 f 0.12 1.16 f 0.14

Day 14

1.16 f 0.07 1.14 f 0.10 1.16 f 0.0s

1.13 f 0.06 1.18 f 0.09

1.11 f 0.10

a Mean f S D calculated from groups of at least 9 mice. * Day 4 measured in a separate experiment. 'Mce a day 250 p1 tap water/mouse per a. Statistically significant difference compared to group 2.

In order to find out what the average drug intake, the water consumption was measured daily. Average values over the whole 4 week period are shown in n b l e IV. The daily doscs (ml) of group 1 (inclusive 500 pl by oral gavage), the tiaprofenic acid and prednisolone group approximated the amounts of water intake by control mice (group 2). However, the salicylate and piroxicam treated mice drunk significantly less than those of group 2. Undoubtedly, the taste of this drug water was the reason for this. The average weight of the mice at the beginning (day -1) and at the end (day 28) of the experiment are also given in n b l c IV Only prednisolone treatment appeared to result in significant weight loss.

Table Il4 Average drug dase and water intake during, and weight change after, a 4 week period."

Histological grading of the inflammation and of cartilage damage is shown in nblc V. Interestingly none of the drugs caused a significant aggravation of the cartilage damage. Antiinflammatory effects in terms of decreased values for infiltration and exudation seemcd apparent for prednisolone and to a lesser extent for salicylate. Piroxicam seemed only 10 suppress exsudation.

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.

Effect of Nons temida lAnt i in~a tory Drugs on Cartilage Destruction' 27

Group

1 2 3 4 5 6

Table I? Histological grading of arthritis.*

lteatment** Infiltrate Exudate in in synovium joint space

oral gavage 1.9 0.6 water 1.9 0.7 salicylate 1.2 0.2 piroxicam 1.7 0.2 tiaprofenic acid 1.6 1 .o prednisolone 0.7 0.0

1 2 3 4 5 6

Cartilage PG depletion

tiaprofenic acid 32 16 8 4

salicyla ie m water -

1 .o 0.7 0.5 1.3 1 .o 0.4

See for histologic paranletem Tible 11. * * Drugs given in drinking water.

Table H: Weight change of drug treated mice and "S-sulfate incorporation in articular knee carti- lage.

weight

day 0 day 52

2 0 f 2 25fl 2 0 f 2 2 5 f l m f 2 2 4 f 2 1 9 f 2 25fl 1 9 f 2 25fl m f 2 2 3 f 2

3 s ~ incorporationC

patellae (%I ~

302f 25 131. 275 f 29 119 252 f 27 109 349f48 151. 338 f 50 146.

231f25 100

O Oncc a day by oral gavage. mg f SD of at least 12 mice. cpm f S D in the patellar cartilage of groups of 6 mice.

* p < 0.01.

Apart from effects on cartilage proteoglycan depletion, we also looked for potential drug effecls on chondrocyte integrity. Scores for chondrocyte death were similar for groups 1-5 and tended to be slightly lower for group 6 (data not shown).

Analysis of chondrocytefinction As stated previously no abnormalities were found in the overall observation of the articular cartilage of the left normal joints of mice treated for 4 weeks with the various drugs. TI ex- amine potential effects on chondrocyte synthetic function in more detail, additional groups of normal mice were treated with tiaprofenic acid and salicylate for 7.5 weeks. "henty- four hours after the last drug treatment, the mice received an intraperitoneal injection of Na2 35S04 and 3 hours later patellae were isolated.

However, no suppression of chondrocyte synthetic function was found in the drug treated groups ("hble VI). In addition to quantitative analysis of chondrocyte function shown in 711- ble VI, qualitative analysis was done by autoradiography. In neither test was any evidence

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.

28 B.J. de Vnes, WB. van den Berg E. Yuiem and LBA. van de f i t r e

found for any local change in synthetic activity in the patellar cartilage, nor for the occur- rence of dead chondrocytes.

Perhaps the increased synthetic activity found in some tiaprofenic acid groups and in the salicylate group p b l e VI) points to an attempt a t repair after a period of suppressed syn- thetic activity. For example, if drugs cause short-term suppression followed by a period of enhanced synthesis, the overall effect is probably negligible. Such reversibility had already been demonstrated for the salicylate mediated suppression of 3SS-GAG synthesis after a single dose. Another aspect, apart from the amounts of GAG present in the cartilage, is the quality. If, in the phase' of suppressed synthesis, molecules of a changed composition are made, e.g. undersulfated GAGS, this could disturb cartilage integrity. Methods have now been developed to enable this issue to be addressed shortly.

Acknowledgements The staff of the Central Animal Laboratory is gratefully thanked for their excellent animal care.

References 1. Urackertz D, Mitchell GF, MacKay IR. Antigen-induced arthritis in mice. I. Induction of arthritis in variorrs

strains of mice. Arthritis Rheum 1977; u):841-50. 2. Van den Berg WB , van Beusekom HJ, van de Putte LBA, Zwarts WA, van der Sluis M.Antigen handlirig iti

antigen-induced arthritis in mice An autoradiographic and immunopuorescetice smc& using whole joint sectiotrs. Am J Path 1982; 1089-16.

3. Van den Berg WB, van de Putte LBA, Zwarts WA, Joosten LAB. Ekctrical charge of the antigen deterrnitrcs inw-curicular antigcn handling and chroniciy of arthritis in mike. J Clin Invest 1984,741850-9.

4. Van den Berg WB , Kruijsen MWM, van de Putte LBA, van der Sluis-van der Pol M, Zwarts WA. Antigeti- induced andzyman-indufcd arthritis in mice: studies on in vivo conilogeproteog&an synthesis and chondro- cyre death. Br J Exp Path 1981; 62308-16.

5. Kruijsen MWM, van den Berg WB, van de Putte L B k Influence of theseveriry and duration of murine anrigti- itiduced arthritis on cartihgcproteoglycan synthesis and chondrocyte death. Arthritis Rheum 1985; 28:813-9.

6. Hunneyball IM, Crossley MB, Spowage M. Pharmacological studies of antigen-induced arihritis in BALBIc mice. I, Characterization of the arthritis and the fleets of steroidal andnon-steroidal antiinflammatory agents. Agents Actions 1986; 18384-93.

7. I lunneyball IM, Crossley MJ, Spowage M. Phamcological studes of antigen-induced arthritis in BALllk mice. II. The effects of second-he antirheumatic dtup and cytam*c agents on the histopathological chattges. Agents Actions 1986; 18394-400.

8. Palmoski MJ, Brandt KD. Roteoglycan depletiofi rather than jibrillation, determines rhe effecrs ofsalicylarc and indomethacin on ostewnhnric cartihge. Arthritis Rheum 1985; 28548-53.

9. Palmoski MJ, Brandt KD. I n vivo efect of aspirin on canine OsleODnhrilic carfihgc. Arthritis Rheum 1983; 2694-1001.

10. Lens Jw, van den Berg WB, van de Putte LBA. QLtantitath of arthritis by *Tc-uptake measurements ;ti rlrc mouse knee joint: correlation with hitologicaljoint infammarion scores. Agents Actions 1984,14723-8.

11. De Vries BJ, van den Berg WB, Vitters E, van de Putte LBk Quanriration of&cosaninog&an rnetabolisin iti anatomicalty intact articular cartilage of the mouse patella: in v i m and in viw studies with 3sS-suva~e, 3H- ghcosamine, and 3H-acetate. Rheumatol Int 1% 6273-81.

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/17/

14Fo

r pe

rson

al u

se o

nly.