The presence of a kinin in inflammatory synovial effusion from arthritides of varying etiologies.код для вставкиСкачать
The Presence of a Kinin in Inflammatory Synovial Effusion from Arthritides of Varying Etiologies By KENNETH L. MELMON,MARIONE. WEBSTER, STEPHENE. GOLDFINGER AND J. EDWINSEEGMILLER I certain vasoactive polypeptides (bradykinin and kallidin) may be invoIved in inflammatory reactions by virtue of their ability to produce vasodilatation, increased capillary permeability, leukotaxis, and pain in mamm a k 3 AIthough the kinins have the necessary properties of a biological mediator of inflammation, there is comparatively little information concerning their presence in various inflammatory e x ~ d a t e s .This ~ has been due, in part, to the lack of methods sufficiently sensitive and specific to detect physiological, as well as pathological, amounts of the kinins in biological fluids. Using a recently described method for the estimation of the kin in^,^ we have found these polypeptides in synovial fluid obtained from inflamed joints of patients with arthritides of various etiologies and from dogs with induced “arthritis.” This report presents evidence that kinins are present in synovial fluid of acutely inflamed joints. T HAS BEEN S U G G E S T E D ~ ~that ~ METHODS Twenty-five samples of synovial fluid were taken from 13 patients (7 men and 6 women. aqed 10 to 58 years) hospitalized at the Clinical Center. National Institutes of Health. All had synovial effusions of varying degrees of severity in the knees; the etiologies of the arthritis (four with gout, six with rheumatoid arthritis, one with psoriatic arthritis, and two with arthritis of unknown etiology) were determined by standard From the Experimental Therapeutics Branch and Laboratory of Cardiovascular Physiology, National Heart Institute; the Arthritis and Rheumatism Branch, National Institute of Arthritis and Metabolic Diseases; National Institutes of Health, Bethesda, Md.; and the Department of Medicine and Pharmacology and the Cardiovascular Re- clinical and laboratory procedures. Prior to aspirations of synovial fluid, the degree of warmth, swelling, and tenderness of each knee was evaluated by two different investigators. Each of these indices of inflammation was graded arbitrarily from 0 to 4 f . Synovial fluid for determination of kinin was aspirated from the knee shortly after the patients were admitted to the hospital; samples were taken again if the patient’s clinical course changed considerably. Aspirations were performed under sterile conditions without evidence of bleeding into the synovial fluid. The fluid (usually 5.0 ml.) was drawn within 15 seconds into a sterile siliconized syringe containing 1.2 ml. of 1 M phosphoric acid and 0.2 ml. soybean trypsin inhibitor (10 mg./ml., Worthington Biochemical Corp.). The polypeptide present in the synovial fluid was partially isolated and identified as previously described.5 This involved adsorption and elution of the kinin from IRC-50 (H-t), bioassay of the oxytocic activity on the isolated rat uterus. and destruction of the polypeptide by guinea pig plasma. Of the 25 samples assayed, 4 samples (from 3 patients with different arthritides) showed the presence of other unidentified oxytocic substancefs) which, unlike bradykinin, were not destroyed by the carboxypeptidase N in guinea pig plasma. Added polypeptide was recovered (500 ng.) and examined periodically and fell within the expected range (20 to 50 per cent). Samples of synovial fluid were also analyzed for their content of total protein and leukocytes and were always cultured for pathogens. Ailicrocrystalline Sodium Urate Injections Three male patients with gout, who were experiencing a prolonged remission, volunteered to omit medications and to permit injection of up to search Institute, University of California Sun Francisco Medical Center, Sun Francisco, Calif. Supported in part by USPHS Grants HE-03964 and HE-06285. +Present address: Massachusetts General Hospital, Boston, Massachusetts. 13 ARTHRITIS AND RHEUMATI~M, VVL. 10, No. 1 (FEBRUARY, 1967) 14 100 mg. of sterile microcrystalline sodium uratefi suspended in 5 ml. saline into both knees. After receiving the injection, one patient experienced severe gouty effusions in both knees. Joint fluid prior to the injection of sodium urate could not be obtained since the knee was entirely quiescent, but was obtained 6, 24, and 54 hours after injection. The 54 hour sample was taken after colchicine appeared to have exerted a beneficial affect. Injection of Synthetic Bradykinin into the Knees of Dogs Synthetic hradykinin (500 kg in 1.0 ml. saline) was injected intra-articularly into the synovial space of one knee of pentobarbital-anesthetized dogs (60 mg./Kg.). Saline was injected into the other knee as a control. The skin temperature medial to the patella was measured at brief intervals by a thermistor (Yellow Springs Instrument Co.), and the circumference of the knee was measured frequently. After 40 minutes, the animals were killed and the joints opened and inspected for the presence of fluid, as well as for the degree of hyperemia of synovial tissue. In two dogs, both knees were injected with 500 kg bradykinin, followed within 4 minutes by injection of 10 mg. carboxypeptidase B in one knee and equivalent volumes of saline in the other. RESULTS The Presence of Kinin in the Synovial Fluid of Acute Arthritis Detectable concentrations of kinin (Tables 1 and 2 ) were found in knee effusions of all patients regardless of the etiology of their arthritis. Synovial peptide ranged from 1.6 ng./ml. (the lower level of sensitivity for the method used) to 58 ng./ml. fluid. There was, however, little correlation between the concentration of kinin and the degree of warmth, pain and tenderness of the joints. The kinin levels taken from a single patient (W. R., Table 1) may show such a correlation. A similar correlation, however, was not found in synovial effusions from patient J. F., possibly due to the minimal degree of inflammation of this joint. In any event the data are too limited to resolve these questions. The volume of synovial fluid from quiescent joints was insufficient for assay. Since normal kinin levels in human plasma have been less than 3.0 ng./ MELMON ET AL. ml.5 and synovial fluid contains a lower protein concentration, it is to be expected that the kinin levels would be correspondingly less and possibly undetectable. In two gouty patients, W. J. and R. J., an acute attack of arthritis resulted in an increase in polypeptide 24 hours after the onset of symptoms (Table 2). Patient W. J. suffered a spontaneous attack of gout in the left knee whereas the attack of patient R. J. was induced by the injection of microcrystalline sodium urate into both knees. Patient R. J. (Fig. 1) had such severe reaction in both knees that he was unable to walk 24 hours after the injections. Colchicine therapy was then initiated and proved moderately effective in reducing the clinical symptoms. As the symptoms receeded, the kinin levels were lowered. Although the acute process was rapidly ameliorated by colchicine, both knees of this patient remained somewhat stiff for about two weeks. Induction of Arthritis by Injection of Synthetic Bradykinin into the Knees of Dogs When synthetic bradykinin was injected into the knees of four dogs, definite warmth appeared within six minutes, followed by swelling and, on two occasions, the knee became so painful that the animal could be roused from stage three pentobarbital anesthesia by rather gentle manipulation of the joint. The temperature was high for about 20 minutes (Fig. 2) and then began returning to control levels. Swelling, however, (Table 3 ) persisted until the time of sacrifice (40 minutes). All of the changes produced by the polypeptide could be prevented or aborted if carboxypeptidase B, which rapidly inactivates the kinins,7 was injected shortly after the administration of the polypeptide (Fig. 2, bottom, and Table 3 ) . The knees injected with bradykinin showed increased synovial fluid and hyperemia of the synovial membrane when compared to saline-injected knees. No synovialfliuid leiikocytosis occurred in either group, 15 KININ I N INFLAMMATORY SYNOVIAL EFFUSION Table 1.-The Patient W. R. Presence of a Kinin in Injlummatory Synovial Effusions of Acute Arthritis Kinin' ng./ml. Arthritis A. C. Swelling + + ++ + + + ++ ++ + ++ ++ -++ ++ ++ 4.9 12 Unknown N. S. v. s. A . M. A. E. ++ -+ ++ 0 + 0 0 ++ 4.0 3.1 4.3 3.8 -+- 0 -t- Total Protein Leukocytes Gm./100 ml. X l @ / m ~ n . ~ 3.7 4.7 -t ++ * + +++ + + 2.9 7.6 18 26 27 44 Rheumatoid 0 + 0 + ++ 0 0 10 14 Psoriatic *ZI'I 1). 13. D. D. Warmth 1.6 2.1 4.3 1.7 2.3 3.1 3.4 Gout J. F. C. F. Tenderness +++ +++ ++ ++ ++ +++ ++++ 5.3 4.3 3.7 9.4 3.2 4.0 2.6 5.6 + ++ + 3.8 42 t++ 3.5 2.2 19 24 11 26 16 .4 25 27 21 19 18 1.7 8.0 "Calculated as bradykinin (Sandoz Pharmaceuticals). Table 2.-lncrease of Kinin Concentrations in Synovial Effusions During Spontaneous and Induced Acilte Gouty Arthritis Patient Attack W. J . Spontaneous R. J. Induced Time' (hr.) Knee? Kinin ng./ml. 6 LK 24 LK 58 6 LK RK LK RK 8.4 12 19 43 4.0 7.3 24 54 LK RK Tenderness 3.5 Warmth +++ +++ + +++ +++ ++++ +++ ++++ ++++ ++++ ++++ ++++ +++ ++ +++ ++++ Swellinz LeukoTotal cytes Protein X loJ/ G m . i l 0 0 ml. mm:' ++ ++ ++ ++ ++++ +++ ++ ++I+ 42 44 3.1 3.0 3.6 3.4 3.8 4.2 19 21 3.8 6.7 3.6 3.7 'Time in hours after initiation of symptoms (spontaneous attack) or after injection of 100 mg. microcrystalline sodium urate (induced attack). Colchicine therapy given to patient H. J. after 24 hours. iLK-left knee; RK-right knee. and routine sections showed no histologic alteration of synovium. DISCUSSION The present studies show that kinins are present in synovial effusions from patients with arthritides of various etiology and that in gout the levels of polypeptide increase in response either to a spontaneous acute attack of gouty arthritis or to the intraarticular injection of urate crystals. Also, certain aspects of the acute inflammatory process-such as warmth, tenderness, and swelling-can be reproduced in dogs after intra-articular injection of bradykinin. Activation of the plasma kallikrein-kininogenkinin system provides an attractive hypothesis to explain certain aspects of the acute 16 MELMON ET AL. 0 I 2 TIME (days) Fig. 1.-Levels of kinin in knee joint fluid following injection of urate crystals. The experiment was initiated by injection of 100 mg. microcrystalline sodium urate in both knees of it gouty volunteer. Right knee 0-0;left knee 0 - 0 . After the development of symptoms in day 1, the patient was given intravenously 2 mg. colchicine; treatment was repeated after a 10 hour interval. inflammatory response. Plasma kallikrein is present in blood and lymph as an inactive precursor (prekallikrein). This kallikrein acts specifically on an a2 globulin in human plasma (kininogen) to release the nonapeptide bradykinin. The initial activator4 of the prekallikrein-kallikrein system is thought to be Hageman factor, a clotpromoting substance activated by glass or other negatively charged particles. These observations, together with the recent revivals of Garrods original hypothesis that sodium urate crystals initiate the typical acute attack of gouty arthritis, provide a mechanism for the activation of the prekallikrein-kallikrein system. Thus deposition of monosodium urate microcrystals in the joint could activate Hageman factor; activated Hageman factor, either directly or through a series of intermediate enzymatic steps, would activate prekallikrein; the active kallikrein would act on kininogen to release the polypeptide. This hypothesis is essentially identical to that recently proposed by Kellermeyer and Breckenridge.$ These authors, referring to a preliminary report of these studies,l(’ have recently established that the microcrystalline sodium urate crystals are negatively charged; moreover, they have shown that these and other negatively charged crystals are capable of activating Hageman factor. The presence of possible inhibitors to plasma kallikrein, and of enzymes destructive to the kinins, may lengthen the time required to produce sufficient quantities of bradykinin to overcome the body’s defensive mechanisms. Also, as the inflammatory reaction progresses, plasma proteins migrate into the synovial space, thus increasing the substrate arid enzymes available for production of detectable concentrations of bradykinin. This could explain the latent period of four or more hours before the inflammatory response is evident in humans following injection of urate crystals, as compared to the prompt appearance of the inflammatory reaction following (f3-25 minutes) the intra-articular injection of bradykinin in dogs. I n related types of arthritis the production of kinins could be due to the deposition of other negatively charged particles, such as calcium pyrophosphate crystals, which have been foundll in the synovial fluid of patients with chondrocalcinosis.12 The polypeptide found in the synovial fluid has not yet been identified by chromatographic technics. It is presumed to be bradykinin. If plasma leaks into the synovial space during arthritis, the kinin released by kallikrein would be bradykinin. This would be in agreement with the findings 17 KININ IN INFLAMMATORY SYNOVIAL EFFUSION 36.5 - 36.0 - Fig. 2.-Effect of intra-articular injection of bradykinin in the dog. Top illustrates the temperature response to the injection (1)of 0.5 mg. bradykinin in 1.0 ml. ( 0 - 0 ) and 1.0 ml. saline ( 0-0). Bottom illustrates the temperature response to the injection of 0.5 mg. bradykinin (J1) followed by injection of 10 mg. carboxypeptidase B (12) ( 0 - 0 ) or saline (0-0). 34.5 0 I 5 I 10 I I 15 20 TIME (min) of Melchiorri,l" who isolated a polypeptide from pooled and dialyzed synovial fluid and found it to be identical with bradykinin. Melchiorri also found the concentration of the polypeptide to be higher in pooled and dialyzed synovial fluid from normal subjects. However, he could not have been measuring the concentration of bradykinin in intact synovial fluid because dialysis removes the polypeptide. Rather the data would indicate increased levels of substrate and enzymes in the synovial effusion. Although in the present studies a kinin 1 25 I 30 was found in synovial fluid, it might be noted that intra-articular injection of synthetic bradykinin does not mimic all of the events usually associated with arthritis. No synovial fluid leukocytosis occurred, and there was no histologic alteration of synovium. The lack of such events does not exclude a role for the kinins in joint inflammation. Kinins may not be the only stimulus to leukocyte accumulation, and although the high concentrations of kinin injected into the joints in dogs produced symptoms rapidly, time may have been insufficient to allow granulocyte accumula- 18 MELMON ET AL. Table 3.-SweZZing articukw Injection of as the Result of ZntraBradykinin in the Dog Substance Time* (min.) 0 Bradykinin 0 18 23 14.8 15.7 15.8 14.8 14.8 14.8 0 10 15 29 0 10 15 29 14.8 15.6 16.4 15.8 14.8 14.5 14.9 14.8 18 25 Saline Bradykinin + Saline Bradykinin + COB# Cireumference (cm.) ‘Time in minutes after iniection. fCOB-carboxypeptidase B ( Worthington Biochemical Gorp.) tion. Because granulocytes are key to the pathogenesis of crystal-induced acute arthritis,14 and because granulocytes may themselves be an important source of kallikrein,15 other stimuli may be primarily responsible for leukotaxis and histologic changes in synovial tissue. The accumulated leukocytes may then produce kinins which contribute to further swelling, warmth, and pain. The exact contribution of the kinins to the inflammatory reaction in humans remains to be determined. Changes in synovial fluid kinin concentration seemed to correlate well with the severity of symptoms (tenderness, warmth, and swelling) in selected individual patients. On the other hand, the clinical severity of arthritis from person to person could not have been predicted by knowing the kinin concentration in synovial fluid. This seems to diminish the importance of the role of kinin peptides in the pathogenesis of the process. However, several factors must be considered before such a conclusion can be supported: 1. Single determinations of kinin in synoviai fluid do not reflect the activity of the kallikrein- kininogen-kinin system in the synovial tissue or accumulated granulocytes. 2. The determinations in this study reflect a single point in time and neglect the rate and extent of synovial fluid accumulation, activity of the kinin-destroying system, levels of substrate (kininogen) in the locale of kinin production, and accumulation or destruction of kallikrein or kininase inhibitors. 3. Because the total volume of fluid could not be easily assessed, no attempt was made to determine the absolute amounts of kinin in a synovial space. 4. No specific antikinin agents are available for use in human synovial spaces. PheIps e t al.16 recently reported that carboxypeptidase B failed to block synovitis in dogs induced with microcrystalline monosodium urate. While it is true that these crystals have been shown to activate Hageman factor on the synovial fluid of dogs,l7 there is considerable evidence that the prekallikrein-kallikrein system in dogs is not activated by glass beads and is thus quite different from humans.18 Whether the data obtained in the canine system can be extrapolated to man remains to be determined. Another unanswered question is how various therapeutic agents work. In the course of these studies, agents commonly used in the treatment of arthritis, such as phenylbutazone, hydrocortisone, colchicine, and acetylsalicylic acid, were added to the smooth muscle chamber in concentrations of 10-7 to 10d3M. None of these agents inhibited the response of the uterus used for assay to a standard amount ( 4 to 10 ng.) of synthetic bradykinin. There can be no conclusions, however, concerning the ability of these agents to alter kinin concentrations in synovial effusions, for inhibitors to the polypeptide may be effective in one biological system and not in another.lQ Also, these agents might have no direct effect on the polypeptide but may alter the rate of its formation, as has been suggested by recent data,20 or increase its rate of destruction or alter its peripheral activity. ACKNOWLEDGMENTS The technical assistance of W. Anderson, Jr., and D. England js gratefnll~ acknowledged. The 19 KININ I N INFLAMMATORY SYNOVIAL EFFUSION authors are indebted to Dr. E. D. Nicolaides, Parke, Davis and Co., Ann Arbor, Michigan, and Sandoz Pharmaceuticals, Hanover, New Jersey, for gifts of synthetic bradykinin and to Dr. L. Sokoloff for histologic examination of synovial tissue obtained from dogs. SUMMARY Detectable concentrations of a kinin (1.6-58 ng./ml. fluid) were found in effusions from the knees of patients with arthritides of varying etiology. In two patients suffering from gouty arthritis, a n acute attack (spontaneous or induced by microcrystalline sodium urate) resulted in a n increase in polypeptide 24 hours after the onset of symptoms. Intra-articular injection of synthetic bradykinin in the dog produced warmth, swelling, and pain. It is tentatively suggested that the kinins may contribute to the inflammatory synovial reaction seen in arthritides. SUMMARIO IN INTERLINGUA Detegibile concentrationes d e un kinina (1,6 a 58 ng/ml) esseva trovate in effusiones de articulationes del genus in patientes con arthritis d e varie etiologias. I n duo patientes con arthritis guttose, un attacco acute (spontanee o inducite per microcrystallin urato de natrium) resultava in un augmentate concentration d e polypeptida 24 horas post le declaration del symptomas. Le injection intra-articular de bradykinina synthetic in canes produceva calor, tumescentia, e dolor. Es suggestionate que le kininas contribue possibilemente a1 reaction synovio-inflammatori incontrate in certe arthritides. REFERENCES 1. Collier, H. 0. J.: Kinins. Sci. Amer. 207:111, 1962. 2. Schachter, M.: Kinins-a group of active peptides. Ann. Rev. Pharmacol. 4:281, 1964. 3. Lewis, G. P.: Bradykinin. Nature (London) 1922596, 1961. 4. Webster, M. E., and Innerfield, J.: Interrelationship of human plasma kallikrein and plasmin in inflammation. Enzymol. Biol. Clin. 5: 129, 1965. 5. Webster, M. E., and Gilmore, J. P.: The estimation of the kallidins in blood and urine. Biochem. Pharm. 14:1161, 1965. 6. Seegmiller, J. E., Howell, R. R., and Malawista, S. E.: The inflammatory reaction to sodium urate. J.A.M.A. 180:469, 1962. 7. Erdos, E. G., Wohler, J. R., and Levine, M. I.: Blocking of the in V ~ V Oeffects of bradykinin and kallidin with carboxypeptidase B. J. Pharmacol. Exp. Ther. 142:327, 1963. 8. Seegmiller, J. E., Laster, L., and Howell, R. R.: Biochemistry of uric acid and its relation to gout. New Eng. J. Med. 268:712, 764, 821, 1963. 9. Kellermeyer, R. W., and Breckenridge, R. T.: The inflammatory process in acute gouty arthritis. I. Activation of Hageman factor by sodium urate crystals. J. Lab. Clin. Med. 65: 307,1965. 10. Goldfinger, S., Melmon, K. L., Webster, M. E., Sjoerdsma, A., and Seegmiller, J. E.: The presence of a kinin-peptide in inflammatory synovial effusions. Arthritis Rheum. 7:311, 1964. 11. McCarty, D. J., Kohn. N. N., and Faires, J. S.: Significance of calcium phosphate crystals in synovial fluid of arthritic patients: “pseudogout syndrome.” I. Clinical Aspects. Ann. Intern. Med. 56:711, 1962. 12. Zitnan, D., and Sitaj, S.: Calcifications multiples du cartilage articnlaire. 9th International Congress sur les hlaladies Rheumatismales 2:291, 1957. 13. Melchiorri, P.: Occurrence of bradykinin in the synovia of subjects affected with rheumatoid arthritis. In: Bradykinin and its Precursors. F. Sicuteri, Ed., Rome, Centro Editoriale Publicitario Italiano. 1963, p. 65. 14. Phelps, P., and RlcCarty. D. J., Jr.: Crystal induced inflammation in canine joints. 11. Importance of polymorphonudear leukocytes. J. Exp. Med. 124:115, 1966. 15. Melmon, K. L., and Cline, M. J. The interaction of plasma kinins and granulocytes. Nature (London). In press. 16. Phelps, P., Prockop, D. J., and hlccarty, D. I., Jr.: Crystal induced inflammation in canine joints. 111. Evidence against bradykinin as a 20 mediator of inflammation. J. Lab. Clin. Med. 58:433, 1966. 17. KellermeYer, R. w., and Breckenridge, R. T. The inflammatory process in acute gouty arthritis. 11. The presence of Hageman factor and plasma thromboplastin antecedent in synovial fluid. J. Lab. clin.~ ~ 67:455, d . 1966. 18. Armstrong, D., Jepson, J. B., Keele, C. A., and Stewart, J. W.: Pain-producing substance in MELMON ET AL. human inflammatory exudates and plasma. J. Physiol. 135350, 1957. 19. Collier, H. 0. J.: The action and antagonism of kinins on bronchioles. Ann. N. Y. Acad. Sci. 104:290, 1963. 20. Cline, M. J., and Melmon, K. L. Plasma kinins and cortisol: A possible explanation of the anti-inflammatory action of cortisol. Science. In press.