Carbonic anhydrase inhibitors are antiarthritic in the rat

Pharmacological Research, Vol . 24, No . 4 . 1991

377

CARBONIC ANHYDRASE INHIBITORS ARE ANTIARTHRITIC IN THE RAT JOSEPH C . NOLAN*, CAROL E . GATHRIGHT, CYNTHIA H . RADVANY, RICHARD J . BARRETT* and LAWRENCE F . SANCILIO Department of Pharmacology, A . H . Robins Co ., 1211 Sherwood Avenue, Richmond, VA 23220, USA Received in final form 2 May 1991

SUMMARY Adjuvant-induced arthritis in rats was attenuated by the therapeutic administration of carbonic anhydrase inhibitors . Female Lewis rats with established disease were treated daily (day 18 through day 50) with various carbonic anhydrase inhibitors ; oedema and joint integrity (X-ray) were determined post-treatment . Acetazolamide, ethoxzolamide, methazolamide, and dichlorphenamide reduced paw oedema and attenuated the deterioration of the joints of rats with adjuvant arthritis . However, no carbonic anhydrase inhibitor tested possessed significant, acute, anti-inflammatory activity in the carrageenan-paw oedema test . The activity of carbonic anhydrase inhibitors in the chronic model of inflammation may be due to their reported inhibition of bone resorption . KEY WORDS :

arthritis, carbonic anhydrase inhibitors .

INTRODUCTION Inflammatory arthritis is characterized by the loss of the integrity of connective tissue structures of the joint . This degeneration is mediated by a variety of inflammatory cells and soluble factors released from those cells [1] . Interleukin-1 and other cytokines have been shown to stimulate neutral protease and collagenase production in cultures of chondrocytes and human synovium [1], thus suggesting a rational hypothesis for the destruction of collagenous structures of the joint . Bone resorption is also an important clinical feature of the arthritic diseases [2, 3] . Osteoclasts in the vicinity of mononuclear cells are activated [4] . Potent bone resorbing agents, such as PGE 2 , are available to induce resorption [5] . Together, these results suggest that the cells and mediators potentially responsible for cartilage and bone destruction are present in the inflamed joint . Acetazolamide, a potent carbonic anhydrase inhibitor, inhibits bone resorption induced by disuse in the rat [6] and, in vitro, inhibits bone resorption induced by *Present address : Whitby Research Incorporated, 2801 Reserve Street, Richmond, VA 23220, USA . 1043-6618/91/080377-07/$03 .00/0

© 1991 The Italian Pharmacological Society

37 8

Pharmacological Research, Vol. 24, No . 4, 1991

parathyroid hormone, PGE 2 , or by 1,25-dihydroxyvitamin D 1 [7, 8] . Recently, these effects on bone resorption have been extended to other carbonic anhydrase inhibitors, such as ethoxzolamide and HTS (5-3-(hydroxybenzoyl)-2thiophenesulphonamide [9] . The availability of various carbonic anhydrase inhibitors and their inhibition of bone resorption in vivo and in vitro permitted us to determine the effects of these drugs in a chronic model of arthritis, the adjuvant arthritic rat . In the studies presented here, we report that carbonic anhydrase inhibitors show antiarthritic activity in rats .

MATERIALS AND METHODS Adjuvant arthritis

Female Lewis rats that weighed approximately 180 g (Charles River Laboratories, Wilmington, MA) were used in these studies . The animals were kept on a 12-h light/dark cycle with food and water available ad libitum . On day 0, 0 .05 ml of a suspension of 1 .5% Mycobacterium butyricum in mineral oil was injected into the subplantar surface of the right hind paw . On day 18, the hind paw volume of the uninjected hind limb was determined by mercury displacement ; rats with significant paw swelling (>2 .4 ml) of the uninjected hind limb were randomized by block design into groups of 7 . All drugs were dissolved or suspended in the vehicle (0 .5% Tween 80) and were given daily by gavage (10 ml/kg) to the rats from day 18 through day 50 after adjuvant injection (excluding weekends) . Oedema was determined by the difference (from day 18 paw volume) on days 29 and 50 . At the termination of the experiment, the rats were sacrificed by CO2 inhalation and the uninjected hind limb was removed, Xrayed, and scored on an arbitrary scale (1, no damage ; 10, maximum damage) in a blinded manner .

Carrageenan oedema

Carrageenan-induced oedema was determined in female Lewis rats (150-175 g) . Groups of six fed rats were dosed orally with test compounds on four consecutive days prior to injection of 0 .1 ml of a 1% carrageenan solution (in saline) into the subplantar surface of the right hind paw . The last dose of drug was given 1 h prior to the injection of the carrageenan solution . Oedema was determined by the difference in paw volumes measured (by mercury displacement) prior to and 3 h post-carrageenan challenge .

Carbonic anhydrase assay

The inhibitory activity of the compounds against carbonic anhydrase (from bovine erythrocytes) was determined in the following manner . Approximately 10 units of bovine carbonic anhydrase in 6 ml of barbitone (barbital) buffer were incubated for 10-15 s with various amounts of the test drugs . The reaction was initiated by the addition of substrate (4 ml C0 2 -saturated H 2 O) . The rate of

Pharmacological Research, Vol. 24, No . 4, 1991

379

enzyme-catalysed change in pH was followed, and the IC 50 was calculated (graphically) from the concentration-% inhibition curves .

Materials and statistical analyses

Acetazolamide, dichlorphenamide, indomethacin, and carbonic anhydrase were purchased from the Sigma Chemical Company, St Louis, Missouri . Ethoxzolamide and methazolamide were synthesized in the Chemical Research Department of the A . H . Robins Company, Richmond, Virginia . Data were analysed by using the Dunnett's t-test to compare the effects of drug treatments to a control .

RESULTS Various carbonic anhydrase inhibitors (acetazolamide, ethoxzolamide, methazolamide, and dichlorphenamide), as well as the reference non-steroidal anti-inflammatory drug (NSAID), indomethacin, were given to groups of seven rats with established adjuvant arthritis . The results in Table I represent the pooling of several experiments . All of the carbonic anhydrase inhibitors tested produced anti-inflammatory activity as determined on day 29 or on day 50 (Table I) . They also attenuated the changes seen in the X-ray score of the arthritic joint on day 50 (Table I) . Significant activity was observed with the lowest dose of each drug tested . From the data presented, it is apparent that the potent carbonic anhydrase

Table I Antiarthritic activity of carbonic anhydrase inhibitors and indomethacin in adjuvant-arthritic rats Treatment

Vehicle Indomethacin Acetazolamide

n

55 55 28 35 7 7 7 7 14 7

Dose (mglkg) p .o .

3 .16 3 .16 10 .0 31 .6 1 .0 3 .16 3 .16 31 .6 100

Oedema' day 29 (ml±so)

Oedema" day 50 (ml±so)

X-Ra v score (mean±so)

C .A ." iC :,, (pM)

0 .19±0 .28 -0 .05±0 .43 8 .57±1.03 3 .80±0 .72` -0 .73±0 .22` -0 .63±0.30` >34 -0 .21±0 .23` -0 .42±0.20` 6 .30±1 .49` 0 .072 -0 .32±0 .21` -0 .53±0 .28` 6 .37±0 .79` -0 .43±0 .20` -0 .69±0 .18` 6 .79±0 .74` Ethoxzolamide -0 .03±0 .20` -0 .36±0 .17` 6 .89±1 .07` 0 .016 -0 .29±0 .20` -0 .43±0 .35` 6 .11±1 .31` Methazolamide -0 .24±0 .35` -0 .43±0 .24` 6 .71±I .01` 0 .040 Dichlorphenamide -0 .10±0 .20` -0 .34±0 .27` 7 .93±1 .30` 0 .233 -0 .29±0 .25` -0 .54±0 .25` 5 .89±0 .45` Compounds were given to the rats daily (excluding weekends), by gavage, beginning on day 18 after adjuvant injection and continuing through day 50 after adjuvant injection . Vehicle-treated animals received 0 .5% Tween in H 2 O (10 ml/kg) . X-rays of the uninjected hind limb were made on day 50 . Oedema=difference between paw volume on day 29 or day 50 and paw volume on day 18 (first day of drug treatment) . hC .A .=carbonic anhydrase in vitro . `P<0 .05, compared to vehicle control, Dunnett's t-test .

3 80

Pharmacological Research, Vol . 24, No . 4, 1991

Table II Acute anti-inflammatory activity of carbonic anhydrase inhibitors and indomethacin in the carrageenan-induced paw oedema test in rats Treatment

n

Vehicle Indomethacin

6 6 6 6 6 6 6 6

Acetazolamide Ethoxzolamide Methazolamide Dichlorphenamide

Dose (mg/kg) p .o .

Oedema (ml±sD)

% Change from control

3 .16 10 .0 31 .6 10 .0 10 .0 31 .6 100 .0

1 .13±0 .18 0 .65±0 .23° 0.97±0 .23 1 .18±0 .16 1 .02±0 .26 1 .28±0 .35 1.20±0 .19 1 .02±0 .29

-42 .6 -14 .7 4 .4 -10 .3 13 .2 5 .9 -10 .3

Drugs were administered once daily for 4 consecutive days to groups of fed rats . The last dose was given I h prior to carrageenan challenge . Oedema was determined by the paw volume difference prior to and 3 h post challenge . 'P<0.05, compared to vehicle control, Dunnett's t-test .

inhibitors (acetazolamide, ethoxzolamide, and methazolamide) produced significant activity at doses of 1 and 3 .16 mg/kg . Dichlorphenamide, a weaker carbonic anhydrase inhibitor, produced comparable results at 31 .6 and 100 mg/kg . The reference NSAID, indomethacin, at 3 .16 mg/kg produced the expected oedema and X-ray score changes (Table I) . These drugs were tested for their effect on acute inflammation induced by carrageenan . When given for four consecutive days prior to challenge, the carbonic anhydrase inhibitors at doses equal to or greater than those tested in the adjuvant rat did not produce any significant anti-inflammatory activity (Table II) . Indomethacin did produce significant, acute, anti-inflammatory activity at 3 .16 mg/kg, which was the dose tested in the adjuvant rat (Table 11) . The IC50 values for the drugs tested against carbonic anhydrase are given in Table I .

DISCUSSION Bone resorption is an important clinical feature in patients with arthritic disease [2, 4] . The loss of collagenous structures can lead to permanent functional impairment of the affected joint(s) . Cell culture and tissue culture work have revealed that enzymes and mediators (IL-1, PGE 2 , etc .) of connective tissue destruction and bone resorption are found in synovial fluid or synovial explants [3, 5] . These results suggest that inhibitors or antagonists of the actions of these mediators might be of benefit in arthritic disease . NSAIDs, for example, indomethacin, are routinely used to treat patients with inflammatory joint disease . PGE,, a major lipid mediator, is a potent proinflammatory agent as well as a potent stimulator of bone resorption in vitro [10] . The activity of NSAIDs in models of chronic inflammatory joint disease

Pharmacological Research, Vol . 24, No . 4, 1991

38 1

could result from both its acute effects (reduction of oedema due to PGE 2mediated vasodilation) as well as its inhibition of PGE 2 -mediated bone resorption . Since these NSAIDs, through their inhibition of PGE 2 formation, would attenuate both the early (vasodilation, oedema) and late (bone resorption, joint destruction) phases, it is difficult to assess the role of these drugs on only one of these phases . Carbonic anhydrase inhibitors in rats have been shown to inhibit bone resorption induced by disuse [6], and by parathyroid hormone (PTH), PGE 2 , or 1,25-dihydroxyvitamin D 3 in vitro [7, 9] . The local production of protons by the osteoclast plays an important role in the resorption of mineralized matrix . Carbonic anhydrase inhibitors, by inhibiting the local secretion of H + , are believed to control this resorption [11] . It appeared possible to us to separate the acute inflammatory parameters from the chronic, bone-resorbing parameters with these drugs . Our results indicate that carbonic anhydrase inhibitors, although devoid of acute anti-inflammatory properties (Table II), do attenuate the lesions seen in a chronic model of inflammation that involves connective tissue destruction in the adjuvant rat (Table I) . This result cannot be due to a simple diuretic (loss of fluid) action since, in the acute carrageenan experiment, the carbonic anhydrase inhibitors were given for four consecutive days prior to testing and were not active . In support of this are our unpublished results that showed, when given chronically, these agents do not produce a diuretic effect (because of a systemic acidosis) . Also . noncarbonic anhydrase inhibitor diuretics, such as hydrochlorthiazide and frusemide (furosemide), were not antiarthritic (results not shown) . Thus, the protective effects of carbonic anhydrase inhibitors in arthritic rats cannot be due to the inability of the rats to mount an inflammatory reaction . Other results we have obtained, but not reported here, support our belief that the effects of these drugs are specific . For example, acetazolamide at doses of 10-100 mg/kg does not inhibit oedematous responses induced by platelet activating factor (PAF), by serotonin, or by IgE (passive anaphylactic response) in rats, thus indicating that these agents are devoid of non-specific anti-inflammatory activity . When given daily to rats beginning 18 days prior to the induction of adjuvant disease and continuing for 18 days after adjuvant injection, acetazolamide did not prevent the oedema associated with arthritis (that is, the oedema developed by day 18 after adjuvant injection) as did NSAIDs, steroids, or immunomodulators, such as cyclosporin or cyclophosphamide (data not shown) . Once the arthritic state was obtained (day 18), continued administration of acetazolamide reduced the paw oedema as indicated in Table I . Also, acetazolamide had no effect on a DTH (delayed type hypersensitivity) response in mice when administered prior to sensitization or prior to challenge (data not shown) . Also, none of the carbonic anhydrase inhibitors had a significant effect on weight changes of the rats . These results also support our conclusions that the effects of these drugs are not nonspecific . In an effort to learn more about the actions of the carbonic anhydrase inhibitors in the arthritic rats, the X-ray lesions were scored for proliferative new bone (PNB), and the limbs (day 50) were fixed, sectioned, and analysed by an independent pathologist . Proliferative new bone (PNB) was measured at three sites (distal tibia, posterior calcaneus, and the number five metatarsal) on the X-rays of

382

Pharmacological Research, Vol . 24, No . 4, 1991

the uninjected hind limb of all rats at the termination of the experiment . A blinded investigator made all readings and a PNB score was determined as follows : the readings at the three sites were summed (to the nearest 0 .1 mm) and averaged (over seven animals in that particular treatment group) to yield the score for that group . The pathological PNB score was determined by subtracting the PNB score for the negative control group (usually about 15) from the score for the treatment group . The arthritic control (positive) had a pathological PNB score of 15-21 . Indomethacin-treated (3 .16 mg/kg) rats had PNB scores of 3-4, while the carbonic anhydrase inhibitor-treated rats had scores as follows : acetazolamide, 3 .16 mg/kg, 7 .39, 10 mg/kg, 6 .34, 31 .6 mg/kg, 5 .14 ; ethoxzolamide, l mg/kg, 7 .71, 3 .16 mg/kg, 5 .43 ; methazolamide, 3 .16 mg/kg, 6 .86 ; and dichlorphenamide, 31 .6 mg/kg, 8 .36, 100 mg/kg, 5 .43 . Thus the PNB scores paralleled the oedema and X-ray scores (Table I) . Although the limbs of all treatment groups were not histologically graded, those for the control groups (positive and negative), indomethacin (3 .16 mg/kg) and acetazolamide (10 mg/kg) were scored on a subjective 0-3 (0, none ; 1, minimal ; 2, moderate ; and 3, marked) scale by a pathologist who was blinded to the treatment . The following parameters were scored : connective tissue proliferation, osteolysis, and chondrolysis ; the cellular inflammatory response and synovitis . The scores for these parameters (in above order) were as follows : arthritic positive control, 3 .0, 2 .71, 2 .71, 2 .86 ; negative control, 0, 0, 0, 0 ; indomethacin (3 .16 mg/kg) 1 .43, 0 .57, 1 .14, 1 .0 ; and acetazolamide (10 mg/kg) 1 .86, 1 .57, 1 .57, 1 .0 . The results thus substantiated our oedema and X-ray results (Table I) but did not shed further light on a mechanism distinct from that of the NSAID, indomethacin . These results, however, further support the observation that the carbonic anhydrase inhibitors possess antiarthritic activity . Our results suggest that the chronic phase (established disease) of the disease is dependent upon products of connective tissue degradation . That is, while the early acute events that result in oedema are regulated in part by the vasodilatory effects of PGE2, later (as cells and mediators come into the joint) the local formation of connective tissue resorbing agents (PGE 2 , IL-1, etc .) become important features of the clinical picture . The release of connective tissue fragments then may play a role in perpetuating the disease . Indomethacin, by inhibiting PGE 2 formation, inhibits both early and late PGE 2-dependent reactions . The carbonic anhydrase inhibitors, however, appear to affect only the bone resorption phase (late) through a non-PGE 2 -dependent mechanism . This may explain why the carbonic anhydrase inhibitors produce significant activity but activity that is somewhat less than that obtained with indomethacin . Our results suggest a role for bone resorption in the pathology of joint destruction, as well as a new therapeutic treatment of joint disease with carbonic anhydrase inhibitors .

REFERENCES 1 . Harris ED Jr . Pathogenesis of rheumatoid arthritis . Am J Med 1986 ; 80 (Suppl . 4B) : 4-10 . 2 . Sambrook PN, Reeve J . Bone disease in rheumatoid arthritis . Clin Sci 1988 ; 74 : 225-30 .

Pharmacological Research, Vol. 24, No . 4, 1991

383

3 . Alwan WH, Dieppe PA, Elson CJ, Brodfield JWB . Bone resorbing activity in synovial fluids in destructive osteoarthritis and rheumatoid arthritis. Ann Rheum Dis 1988 : 47 : 198-205 . 4 . Duncan H . Cellular mechanisms of bone damage and repair in the arthritic joint . .1 Rheumatol 1983 ; 11 (Suppl .) : 29-42 . 5 . Robinson DR, Tashjian AH Jr, Levine L . Prostaglandin-stimulated bone resorption by rheumatoid synovia . J Clin Invest 1975 ; 56 : 1181-8 . 6 . Kenny AD . Role of carbonic anhydrase in bone : partial inhibition of disuse atrophy of bone by parenteral acetazolamide . Calcif Tissue Int 1985 ; 37 : 126-33 . 7 . Hall GE, Kenny AD . Role of carbonic anhydrase in bone resorption induced by 1,25dihydroxyvitamin D 3 in vitro . Calcif Tissue Int 1985 ; 37 : 134-42 . 8 . Minkin C, Jennings JM . Carbonic anhydrase and bone remodeling : sulfonamide inhibition of bone resorption in organ culture . Science 1972 ; 176 : 1031-3 . 9 . Raisz LG, Simmons HA, Thompson WJ, Shepard KL, Anderson PS, Rodon GA . Effects of a potent carbonic anhydrase inhibitor on bone resorption in organ culture . Endocrinology 1988 ; 122(3) : 1083-6 . 10 . Raisz LG, Martin TJ . Prostaglandins in bone and mineral metabolism . In : Peck WA, ed . Bone and mineral research . Annual 2 . Amsterdam : Elsevier, 1984 : 286-304 . 11 . Waite LC, Volkert WA, Kenny AD . Inhibition of bone resorption by acetazolamide in the rat . Endocrinology 1970 ; 87 : 1129-39 .