Effect of Gold Salts and Other Drugs on the Release and Activity of Lysosomal Hydrolases.код для вставкиСкачать
Effect of Gold Salts and Other Drugs on the Release and Activity of Lysosomal Hydrolases By ROBERT S. ENNIS,JOSE LUISGRANDA, AND AARON S. POSNER Acetylsalicylic acid, indomethacin, phenylbutazone, gold thiomalate and gold thioglucose had no protective effect on the release of hydrolases from lysosomes. Only the gold compounds produced a marked inhibition of acid phosphatase, 8-glucuronidase and cathepsin, obtained from both rabbit liver lysosomes and human synovial fluid. This inhibition is reversible by the addition of a sulfhydryl compound. It is suggested that a possible mechanism of gold action in rheumatoid arthritis is the inhibition of intra-articularhydrolases. I T has been postulated that the final stage in the development of rheumatoid joint disease is the disruption of intra-articular lysosomes, with subsequent release of free lysosomal hydrolases, followed by degradation of the joint cartilage.1.2.3Evidence supporting this concept has been obtained from both morphological and biochemical investigations. Histochemical methods h a w shown lysosomes in joints to be present in a ) the lining cells of the synovial membrane: b ) in wandering and fixed tissue macrophages,5 and c) in polymorphonuclear leukocyte^.^.^ Biochemical studies have shown increased titers of lysosomal acid hydrolases in rheumatoid synovial fluid. In particular, acid phosphatase,8 p-glucuronid a ~ eand , ~ P-acetylglucosaminaselO have all been identified in these pathological joint fluids. Z 8 et al.ll have shown the presence of a neutral protease in extracts of human polymorphonuclear leukocytes and rheumatoid synovial membrane capable of hydrolysing proteinpolysaccharides. In addition, Ali et al.12,13 and Fell and Dingle14 have shown that an acid protease, tentatively identified as cathepsin, is capable of degrading the extracellular matrix of joint cartilage. In addition, substances which cause the release of lysosomal enzymes, such as streptolysin S,'; and endotoxin,lGwhen injected directly into joints of experimental animals, produce a chronic arthritis characterized by synovial hyperplasia, focal concentrations of lymphocytes, and pannus formation. Thus, it has been proposed that various anti-inflammatory drugs exert their therapeutic effects by stabilizing lysosomes, thereby retarding the liberation of hydrolytic enzymes into the synovial fluid.3 In the case of gold salts, Persellin and Z s 5 suggested from studies on peritoneal macrophages, that direct inhibition of hydrolytic enzymes is a possible mechanism of their anti-inflammatory action. Publication Number 52 from The Laboratory of Ultrastructural Biochemistry, Hospital for Special Surgery, Cornell Unicersity Medical College, New York, N . Y. 10021. This work supported in part by PHS Grant DE01945 from the National Institute of Dental Research. ROBERTS . ESSIS, M.D.: Resident in Surgery, Neu; York Hospital. Cornell University Medical College. JOSE LUIS GRANDA,M.D., PH.D.: Assistant Professor, Department of Surgey, Cornell University Medical College. AARON S . POSNER, PH.D.: Associate Director of Research, Hospital for Special Surgery, Associate Professor of Biochemistry, Cornell University Medical College. Address reprint requests to Dr. Posner. 756 ARTHRITIS AND RHEUMATISM, VOL. 11, NO. 6 (DECEMBER 1968) 75; DRUGS AND LYSOSOMAL HYDROLASES 80 - lL i' E ao- lu c 4.0 - 20- 2 3 4 5 PH Fig. 1-The pH dependance of the proteolytic activity of human rheumatoid synovial fluid. Enzyme activity in arbitrary units. It was the object of this work to determine whether proteolytic activity is present in the synovial fluid of patients with rheumatoid arthritis, and to study the effect of gold salts and other drugs used in the treatment of rheumatoid arthritis on the release and activity of lysosomal hydrolases. A preliminary report of these findings has appeared.17 MATERIALSAND METHODS Synovial fluid was aspirated from the knee joints of patients with rheumatoid arthritis and immediately cooled to 0 C. Samples grossly contaminated with blood were discarded. The fluid was centrifiiged withoiit delay at 2000 rpm. for 5 min- iites at 0 C. to remove cellular components. Enzyme assays were then performed in the manner described below. Intact lysosomes were separated from the fresh rabbit liver tissue by differential ultracentrifugation at 0" C. in 0.25M sucrose, according to the method of De Duve et al.18 The lysosomd stability of this fraction in the presence of a number of test substances, was measured by the technique of De Duve et al.19 One part by volume of the test substance dissolved in 0.25M sucrose was added to a solution containing 10 volumes of the liver particle suspension, and 10 volumes of 0.1M acetate buffer (pH = 5.0). When a non-aqueous solvent was used, a separate aliquot of particle suspension was incubated with solvent alone to serve as a control. The test mixtures were preincubated for 30 minutes at 37 C.after which en- 758 ENNIS, GRADA, POSNER 0Synovial Fluid t Cystrinr Synovial Fluid t W d Thiomalak Spwial Fluidt Qold Thiamolotrt Cystrinr ACID PHOSPMATASE CATMEPSIN 1 Fig. 2-Effect of gold thiomalate (1.0 mg./ml.) and of cysteine (10 mM/L) on the activities of synovial fluid acid phosphatase a n d cathepsin. zyme substrate prepared in 0.25M sucrose and 0.1M acetate buffer was added. Acid phosphatase activity was used as a measure of the degree of lysosomal disruption a t p H 5.0. The following test substances were used in the above procedure: 1.) Hydrocortisone sodium succinate (SolucortefB, The Upjohn Co.) in concentrations ranging from 5.0 mg./ml. to 0.1 mg./ml. solution in 95 per cent ethanol. 2.) Acetysalicylic acid in concentrations ranging from lO-3M to lO-GM, either as aqueous suspension or in solution of 50 per cent ethanol. 3.) Indoniethacin (Indocin@, Merck, Sharp, and Dohme Inc.) in aqueous solution was tested in concentrations ranging from 2 mg./ml. to 0.02 mg./ml. 4.) Phenylbutazone (ButazolidinB, Ceigy) was tested in aqueous suspension of 5.0 mg./ml. 5.) Gold Sodium Thiomalate (MyochrysineB, Merck, Sharp, and Dohme) and Gold Sodium Thioglucose (SolganalB) were tested in concentrations ranging from 5.0 mg./ml. to 0.01 mg./nil. The same analyses for enzyme content and inhibition were employed for both synovial and lysosomal solutions. Acid phosphatase was assayed by the method of Andersch and Szezypinski,20 f3-glucuronidase and cathepsin were measured according to the method of Gianetto and D e Duve.21 The cathepsin unit is defined as that quantity of enzyme which in 60 minutes produces a colored substance not precipitable by trichloracetic acid, equivalent to 1 milli-equivalent of trymine. The acid phosphatase and f3-glucuronidase units are defined as the mM of the product liberated per 60 minutes’ incubation. The pH dependance of the synovail fluid protease was tested with the following buffers: HCLKCL, pH 2.0; Glycine-HCL, p H 2.4 and p H 2.8; acetic acid-sodium acetate, pH’s: 3.6, 4.0 and 5.0. The total enzyme activity of lysosomal preparations was determined using the above assay procedures after addition of 0.1 per cent Triton X-100 detergent to disrupt intact particles. To assay enzyme inhibition, a series of gold thio- 759 DRUGS AND LYSOSOMAL HYDROLASES Table 1.-Relative Concentrations of Acid Phosphatase and Cathepsin Denionstrated in Human Knee Pathological Synooial Fluid Volume in ml. Patient 5.2 B.I. 2.5 R.G. 3.4 1l.R. 6.2 L.K. 4.1 M.W. 4.2 J1.A. 4.0 E.G. Diagnosis Osteoarthritis (with effusion) Posetraumatic arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Cathepsin units */ml. $ Acid phosphatasc unitst/ml. 1.07 1.19 2.38 0.96 2.42 - 2.34 1.49 3.72 2.20 2.86 2.40 3.4 2.33 *Cathepin unit =milliequivalents of tryosine std. released in 60 minutes a t p H 3.6 and 37" C. tAcid phosphatase unit =mM of p-nitrophenol released in 60 minutes p H 5.0 and 37" C. $ml. of synovial fluid. malate solutions (ranging from 0.01 to 5.0 mg./ml.) were incubated with enzyme solutions (30minutes pH 5.0, 37 C) prior to the substrate addition. Finally, to test the reversibility of the gold inhibition, cysteine a t a final concentration of 10 mM, was added to the enzyme system which had been preincubated with gold thiomalate. In an effort to establish the mechanism of this inhibition, detailed kinetic studies (using Lineweaver and Burk plots) were performed on acid phosphatase and P-glucuronidase liberated from rabbit liver lysosonies after treatment with double distilled, deionized water. The following inhibitors were added to the enzyme system in each of the experiments: gold thiomalate, 2.5 mM, gold thioglucose, 2.5 mM, iodoacetate, 5.0 mM, mercuric acetate, 5.0 mM, pchloro-mercuric benzoate, 5.0 mM and sodium fluoride, 5.0 mM. After adding the inhibitor the enzyme solutions were preincuhated for 30 minutes a t 37 C, and pII 5.0. The solutions containing iodoacetate and p-chloro-mercriric benzoate, on the other hand, were preincubated in glycine buffer at pH 9.0 for 33 minutes at 37 C: and prior to the addition of substrate the solution p H was returned to 5.0 by the addition of 0.2M acetic acid. RESULTS Stynooiul Fluid Hydrolases A cathepsin-like acid protease can be demonstrated in human rheumatoid synovia1 fluid. The pH dependence of this protease is shown in Fig. 1. It can be seen that it has a pH optimum of 2.5-2.8. The enzyme is slightly activated by the addition of lOmM cysteine (Fig. 2). The relative concentrations of cathepsin and acid phosphatase, obtained from pathological synovial fluids, are shown in Table 1. E f e c t of Anti-injlammatory Drugs on the Release and Activity of Lysosomal Hydrolases The effects of hydrocortisone, acetylsalicylic acid, indomethacin, phenylbutazone and gold salts on the release of acid phosphatase from rabbit liver lysosomes are summarized in Table 2. Actually, when gold thiomalate, in aqueous solution, was incubated with intact lysosomal particles in concentrations ranging from 5.0 mg./ml. to 0.01 mg./ml., the free enzyme activity was found to be markedly diminished. This apparent protective effect of gold on the lysosoma1 membrane, however, could be ac- 760 ENNIS, GRADA, POSNER Table t.--Eaect of Anti-inflammatory Drugs on the Release Acid Phosphatase from Rabbit Liver Lysosomes Drug Control Hydrocortisone Acetylsalicylic acid Indomethacin Phenylbutazone Gold thiomalate of No. exp. Free activity (% of total) 13 6 6 8 4 14 35.5 f4.7 27.5 k 3.5 40.0 f4.2 45.8 f5.3 85.1 f5.0 43.0 *4.6* *Corrected for direct enzyme inhibition as shown in Table 3. All other drugs showed no inhibition. ? =Standard deviation. counted for entirely by a direct inhibitory effect upon the enzymes. All other drugs in Table 2 showed no free enzyme inhibition and thus did not require t h i s correction. The experiments reported in Table 2 represent the average of the three different drug concentrations outlined above in Materials and Methods. All of the results, for a given drug, were averaged, since no significant difference in free activity of acid phosphatase was found at the different drug concentrations. Thus, it can be seen that only hydrocortisone produces any significant protective effect, while phenylbutazone alone has a marked disruptive effect, on the lysosomal membrane. Inhibition of Hydrolases by Gold Salts The relative effectiveness of gold thiomalate on the inhibition of synovial fluid and rabbit liver lysosomal acid hydrolases is shown in Table 3. Gold thioglucose and gold chloride were found to be identical to gold thiomalate in the inhibition of lysosoma1 hydrolases in the range of concentrations tested. The effect of the concentration of gold in the inhibition of acid phosphatase and p-glucuronidase is shown in Fig. 3. It can be seen that at concentrations as low as 0.01 mg./ml. there is still s i w c a n t inhibition of P-glucuronidase. Mechanism of Gold Inhibition Enzyme inhibition by the gold compounds was found to be independent of pH or precipitation effects. In addition, the chelating agent, EDTA (10 mM/L) had no effect on the activity of the lysosomal hydrolases, suggesting that the gold is not competing with other metals required for enzyme activity. The sulfhydryl-binding agents, mercuric acetate, p-chloro-mercury benzoate and iodoacetate produced a similar inhibition when incubated with the free enzymes. Acid phosphatase was reduced in activity Table 3.-Enzyme Inhibition b y Gold Thiomalate, 1.0 mg./ml., at pH 5.0,37C, Pre-incubated for 30 Minutes Enzyme Acid phosphatase p-glucuronidase Cathepsin Percent inhibition Rabbit liver lysosomes 64.5%f3.9% (7)* 32.5%+5.2% (5) 30.0%k4.3% (4) *Number of experiments shown in parentheses. +Standard deviation. H~~~~ synovial fluid 48.4%f6.1% (5) 50.2%f3.1% (2) 62.5%f4.5% (5) 761 DRUGS AND LYSOSOMAL HYDROLASES Pre-Incubation, pH 4.8, 37OC, 4 0 min. Acid Phosphotose p-Glucuronidose Control 0 5.0 mg Gold g Liver 2.5 mg Gold g Liver 1.0mg Gold g Liver 0.5 mg Gold 0.1mg Gold g Liver 1 Li rl g Liver 1 inhibition o I.-Effect of gold concentration in thc acid phospharase and fl-glucuronidase of rabbit liver lysosomes. The enzymes from a gram of liver tissue are dissolved in 10 ml. of buffer for each gold concentration. by 82.5 per cent, 57.0 per cent and 45.0 per cent respectively, by these agents. In like manner, it was shown that the addition of a sulf-hydryl compound (cysteine, 10 mM/L) to the system, following preincubation with gold thiomalate, produced reversal of the enzyme inhibition (Fig. 2 ) . Acid phosphatase assayed at pH 5.0, 37 C and 10 minute incubation with p-nitrophenylphosphate yielded a K,,, of 8.48 x 10-4M. The K, for P-glucuronidase, incubated for 30 minutes at pH 5.0, 37 C, was found to be 1.76 X 10-3M. A plot of S-l versus V-l (Fig. 4) demonstrates that iodoacetate, mercuric acetate, gold thioglucose and sodium fluoride all act as noncompetitive inhibitors of acid phosphatase. The action of sodium fluoride, a known noncompetitive inhibitor of acid phosphatase, is not reversed by the addition of sulfhydryl compounds. Finally, Table 4 shows that gold compounds are weak inhibitors of acid phosphatase when compared to mercuric acetate, iodoacetate and sodium fluoride. DISCWSSION These investigations show that a proteolytic enzyme, which has been identified in human rheumatoid synovial tissue,22 is also present in significant amounts in rheumatoid synovial fluid. This enzyme has a different pH optimum from the one described by ZifE et a1.I1 in polymorphonuclears and rheumatoid synovial membrane and it is probably a different enzyme. The difference between the pH optimum (2.5-2.8) of this rheumatoid synovial fluid protease 'and the pH optimum (4.0) of a similar protease found in polymorphon~clears,~~ is probably due to the different substrates employed. The pH dependance of the enzyme was not measured above a pH of 5.0 762 ES-NIS, GRADA, POSNER 9 30 - 20 - 10 I 0- -n -10 -6 - 2 0 2 6 n ~ I I I I6 al 26 s" I d and Burk Plot of inhibitors of acid phosphatase from rabbit liver lysosomes. Pre-incubation 30 minutes, followed by 10 minute incubation at pH 5.0,37" C, using p-nitrophenylphosphatesubstrate. Fig. 4-Lineweaver because of technical difficulties. When heatdenatured albumin is used as a substrate above a pH of 5.0, the filtration time of the TCA precipitate is so long (24 hours) that acid hydrolysis of the protein occurs. Finally, the proteolytic activity found in the synovial fluid is tentatively identified as cathepsin D on the basis of pH optimum and slight activation by sulphydryl groups. There were no differences in the levels of cathepsin in the synovial fluid of patients with rheumatoid and post-traumatic arthriConstants, Ki,of Certain Compounds on Acid Phosphatase from Rabbit Liver Lysosomes Table 4.-Inhibidon Ki ComDound Gold thiomalate 2.5 mM Gold thioglucose 2.5 mM Mercuric acetate 5.0 mM Iodoacetate 5.0 mM Sodium Fluoride 5.0 mM 1.39 X 3.15 X 1.82 x 1.69 x 6.92 x 1O-zM 10-2M 10-5~ 10-5~ 10-5~ tis, and even in the case of a patient with osteoarthritis, who had an inflammatory component, there was considerable cathep tic activity in the synovial fluid. These results stress the point that the liberation of hydrolytic enzymes is a consequence of the inflammatory process regardless of the origin. The levels of cathepsin found in the pathological synovial fluids suggest strongly that this enzyme may play a role in cartilage destruction. Concomitant acid phosphatase measurements are in agreement with levels previously reported in the literatureSs Of the drugs tested, only hydrocortisone was capable of producing a significant protective effect upon intact lysosomes subjected to artificial stress conditions. These findings with hydrocortisone are in agreement with the in uitro studies of De Duve,lB and the in vivo work of Weissmann and 763 DRUGS A N D LYSOSOMAL HYDROLASES Th0mas.2~On the other hand, acetylsalicylic acid was found to have no protective effect upon isolated lysosomes at pH 5.0, in agreement with the report of Weissmann.25 However, a previous report indicates that enzyme release may be retarded by this drug, presumably, at a higher pH value.26The activation of lysosomes by phenylbutazone is difficult to interpret since the drug was incubated with the lysosomes as a suspension and not in solution. Another example of the disruption of lysosomal membranes by suspended particles has been given in an earlier study of the cytotoxic effects of silica on macrophages.27 Finally, it should be stressed that the experimental results suggest that anti-inflammatory drugs effective in treating rheumatoid arthritis do not necessarily act through direct stabilization of lysosomal membrane. The use of gold oompounds in the treatment of rheumatoid arthritis has been thoroughly reviewed by Freyberg.28 Investigations by Persellin et al.29have shown that gold salts do not alter or retard the immune response in rabbits challenged by bacterial products. Early studies indicated that gold salts have a definite chemotherapeutic effect against hemolytic streptococci, tubercle bacilli and other bacteria.28 The ability of gold salts to prevent PPLO-induced chronic arthritis in mice, has also been known for some time, but the same gold compounds have no effect on the in vitro growth of these organisms.2BThese observations have led some investigators28 to postulate that the gold salts may act non-specificallyby altering the enzyme systems of host cells. Persellin and Zig5 have shown that gold is taken up by tissue macrophages, and that it is accumulated in lysosomes. They found in addition that the gold salts inhibited the activities of acid phosphatase and P-glucuronidase. The present experiments show quantitatively that gold compounds are potent inhibitors of acid phosphatase, /3-glucuronidase and cathepsin. I n vitro inhibition has been demonstrated in the case of /3-glucuronidase (Fig. 3) at gold concentrations comparable to those found in the plasma during gold therapy.2Y From these experiments, the mechanism of gold inhibition is presumed to be via the binding of sulfhydryl groups. This cunclusion is based on a ) the similarity in a c tion of gold salts to enzyme inhibitors known to act by sulfhydryl binding, and b ) the reversibility of this inhibition by the addition of cysteine. In this light, Libensonv2reported that the toxicity of a series of gold compounds could be correlated with their relative sulfhydryl binding capacities. Additionally, he showed that reduced glutathione produced a decreased toxicity of gold salts in the rat. The lack of activation of acid phosphatase or p-glucuronidase, and the minimal activation of cathepsin D by sulfhydryl compounds suggest that this inhibition is due to a non-specific alteration of enzyme protein structure. SUMMARIOIN INTERLINGUA Esseva constatate que acido acetylsalicylic, indomethacina, phenylbutazona, thiomalato de auro, e thioglucosa de auro exerce nulle efFecto protective super le liberation de hydrolases ab lysosomas. Inter le agentes mentionate, solo le compositos a auro effectuava un inhibition marcate de phosphatase acide, p-glucuronidase, e cathepsina obtenite ab lysosomas hepatic de conilio e ab human fluido synovial. Iste inhibition es reversibile per le addition de un composito sulphhydrylic. Es suggestionate que le inhibition de hydrolase intra-articular es un mechanism0 possibile subjacente le effect0 de auro in arthritis rheumatoide. REFERENCES 1. Weissmann, G.: Lysosomes. New Eng. J. hled. 273:1143,1965. 2. Hollander, J. L., McCarty, D. J., Astorga, G., and Castro-Murillo, E.: Studies on the pathogene- 764 sis of rheumatoid joint inflammation and the “R. A. Cell” and a working hypothesis. Ann. Intern. Med. 62:271, 196.5. 3. Weissnann, G.: Lysosomes and joint disease. Arthritis Rheum. 9834,1966. 4. Barland, P., Novikoff, A. B., and Hamerman, D.: Electron microscopy of human synovial membrane. J. Cell Biol. 14:207, 1962, 5. Persellin, R. H., and ZifF, M.: The effect of gold salts on lysosomal enzymes of the peritoneal macrophage. Arthritis Rheum. 957, 1966. 6. Astorga, G., and Bollet, A. J.: Diagnostic specificity and possible significance of inclusions in synovial leukocytes. Arthritis Rheum. 8611, 1965. 7. Cohn, Z. A., and Hirsh, J. G.: Isolation and properties of specific cytoplasmic granules of rabbit polymorphonuclear leukocytes. J. Exp. Med. 112:983, 1960. 8. Smith, C., and Hamerman, D.: Acid phosphatase in human synovial fluid. Arthritis Rheum. 5: 411, 1962. 9. Jacox, R. F., and Feldman, A.: Variations of 0-glucuronidase in abnormal human synovial fluid. J. Clin. Invest. 34:2.63, 1955. 10. Caygill, J. C, and Pitkeathly, D. A.: A study of fLacetylglucosaminase and acid phosphatase in pathological joint fluids. Ann. Rheum. Dis. 25: 137, 1966. 11. ZifF, M., Gribetz, H. J., and Lospalluto, J.: Effect of leukocyte and synovial membrane extracts on cartilage mucoprotein. J. Clin. Invest. 39: 405,1980. 12. Ali, S. Y.: Degradation of the cartilage matrix by an intracellular protease. Biochem. J. 93: 611,1964. 13. Ali, S. Y., Evans, L., Stainthorpe, E. and Lack, C. H.: Characterization of cathepsins in cartilage. Biochem. J. 105:549, 1967. 14. Fell, H. B., and Dingle, J. T.: Studies on the mode of action of excess Vitamin A, 6, Lysosomal protease and the degradation of the cartilage matrix. Biochem. J. 87:403, 1963. 15. Weissmann, G., Becher, B., Wiedermann, G., and Bernheimer, A.: Studies on lysosomes. VII. acute and chronic arthritis produced by intraarticular injections of streptolysin S in rabbits. Amer. J. Path. 46:129, 1965. 16. Hollingsworth, J. W. and Atkins, E.: Synovial inflammatory response to bacteria, endotoxin. Yale J. Biol. Med. 38:11, 1965. 17. Ends, R. S., Granda, J. L., and Posner, A. S In oitro effect of gold on lysosomal hydrolases. Arthritis Rheum. (Abstr.) 10:276, 1967. 18. De Duve, C., Pressman, B., Gianetto, R., Wattiaux, R., and Appelmans, F.: Intracellular dis- ENNIS, GRADA, POSNER tribution patterns of enzymes in rat liver tissue. Biochem. J. 6Q:604,1955. 19. De Duve, C., Wattiaux, R., and Wibo, M.: Effect of fat soluble compounds on lysosomes in aitm. Biochem. Pharmacol. 9:97, 1961. 20. Andersch, M. A., and Szezypinski, A. G.: Use of p-nitrophenyl phosphate as the substrate in determination of serum acid phosphatase. h e r . J. Clin. Path. 17:571, 1947. 21. Gianetto, R., and De Duve, C.: Tissue fractionation studies. V.: Comparative study of the binding of acid phosphatase, fLglucuronidase, and cathepsin by rat liver particles. Biochem. J. 59: 433, 1955. 22. Luscombe, M.: Acid phosphase and cathepsin activity in rheumatoid synovial tissue. Nature 197:1010, 1963. 23. Weissmann, G., and Spilberg, I.: Breakdown of cartilage protein-polysaccharide by lysosomes. Arthritis Rheum. 11:162,1968. 24. Weissmann, G., and Thomas, L.: Studies on lysosomes. 11. The effect of cortisone on the release of acid hydrolases from a large granule fraction of rabbit liver induced by an excess of vitamin A. J. Clin. Invest. 42:661, 1983. 25. Weissmann, G.: Effects on lysosomes of drugs useful in connective tissue disease. In: Campbell, P. N. (Ed.): The Interaction of Drugs and Subcellular Components in Animal Cell. Boston, Little Brown, 1968, p. 203. 28. Miller, W. S., and Smith, J. G.: Effect of acetysalicylic acid on lysosomes. Proc. Soc. Exp. Biol. Med. 122:634, 1 W . 27. Allison, A. C., Harington, J. S., and Birbeck, M.: An examination of the cytotoxic effects of silica on macrophages. J. Exp. Med. 124:141, 1966. 28. Freyberg, R. H.: Gold therapy for rheumatoid arthritis. In: Hollander, J. L. (Ed.): Arthritis and Allied Conditions. Philadelphia, Lea and Febger, 1966. 29. Persellin, R. H., Hess, E. V., and ZifF, M.: Effect of gold on the immune response. Arthritis Rheum. 10:99,1967. 30. Preston, W. S., Block, W. D., and Freyberg, R. H.: Chemotherapy of chronic progressive arthritis of mice. Role of sulfur in gold containing compounds. Proc. SOC. Exp. Biol. Med. 1:253, 1942. 31. Sabin, A. B., and Warren, J.: The curative effect of certain gold compounds on experimental, proliferative, chronic arthritis in mice. J. Bact. 40: 823,1940. 32. Libenson, L.: Toxicity and mode of action of gold salts. Exp. Med. Surg. 3:146, 1945.