Inflammation, Vol. 3, No. 3, 1979
REGULATION OF BRADYKININ-INDUCED CYCLIC AMP RESPONSE BY QUINACRINE AND PROSTAGLANDIN E2 AND Fza IN HUMAN SYNOVIAL FIBROBLASTS JOHN V. FAHEY and DAVID S. NEWCOMBE Department of Medicine, University of Vermont, College of Medicine, Burlington, Vermont, and Departments of Medicine and Environmental Health Sciences, Johns Hopkins University, Baltimore, Maryland, 21205
Abstract--Bradykinin induces an increment in intracellular cyclic AMP concentrations of human synovial fibroblasts and evokes the release of [3H]arachidonic acid and [3H]-E prostaglandins from human synovial fibroblasts prelabeled in their phospholipids. Both these bradykinin-induced reactions are inhibited by quinacrine, an inhibitor of phospholipase A activity. The cyclic AMP response of human synovial fibroblasts to bradykinin is potentiated by prostaglandin E2 and inhibited by prostaglandin F2~. These data emphasize the critical role of the prostaglandin system in reactions induced by bradykinin and suggest mechanisms by which inflammatory reactions due to bradykinin may be modulated.
INTRODUCTION Bradykinin, a potent algesic nonapeptide with properties capable of inducing chemotaxis and cell proliferation (1, 2), evokes an increment in the intracellular cyclic AMP concentrations in human synovial fibroblasts but does not stimulate adenylate cyclase activity in membrane preparations of these cells (3). Bradykinin concentrations are higher in synovial fluids from patients with chronic proliferative and acute inflammatory arthritides than in those with noninflammatory joint disease (4-8). This same nonapeptide evokes an increase in intracellular cyclic adenosine 3',5"monophosphate (cyclic AMP) concentrations in human synovial fibroblasts; it also stimulates the release of E prostagtandins from these cells (3). E prostaglandins are found in increased quantities in rheumatoid synovial effusions and are produced in large quantities by rheumatoid synovial tissues (9, 10). The antiinflammatory agents, indomethacin and hydrocortisone, inhibit both the 235 0360-3997/79/0700-9235503.00/09 1979PlenumPublishingCorporation
236
Fahey and Newcomhe
kinin-induced cyclic AMP response and E prostagland~n release (3). These findings support the hypothesis that bradykinin is a key mediator of the tissue responses observed in certain inflammatory arthritides. Further, pharmacologic agents regulating bradykinin activated processes could be used to examine the role of kinins in inflammatory arthritis. We report here the inhibition of the bradykinin-induced cyclic AMP response in intact human synovial fibroblasts by quinacrine and the modulation of this cyclic AMP response by E2 and F2~ prostaglandins.
MATERIALS AND METHODS Cell Culture, Human synovial membrane was obtained during surgical procedures and synovial fibroblast cultures were initiated and maintained as previously described (3, 11). Cells were cultured in minimal essential medium containing Earle's salts and L-glutamine, supplemented with 10% fetal bovine serum (Reheis Chemical Company) and 100 units/ml penicillin and 100 /zg/ml streptomycin (GIBCO). For the experiments described, synovial fibroblasts obtained from patients without systemic disease were used. Cell viability was tested by dye exclusion using trypan blue. Cell Treatment with Pharmacoactive Agents. Cells were grown to monolayers in 60- • 15-mm tissue culture dishes (Corning). Media was changed 1 h prior to bradykinin treatment unless otherwise specified. All reagents were solubilized and prepared as 10• solutions in serumfree media. Prostaglandins were solubilized in absolute ethanol; the final concentration of ethanol was always less than 0.5%. Bradykinin and quinacrine were solubilized directly in ice-cold serum-free media, Cell Processing and Cyclic AMP Assay. After incubations were terminated with trichloroacetic acid, the cells were scraped and assayed for protein by the Lowry method (12). Intracellular cyclic AMP concentrations were determined by the Gilman binding-protein assay without addition of protein kinase inhibitor (13), as described previously (3, 11). Purified beef heart phosphodiesterase treatment of processed cells resulted in a greater than 95% decrease in the intracellular cyclic AMP concentration. Data are expressed as picomoles of cyclic AMP per milligram of cell protein + one standard deviation. Preparation and Use of Radioactively Labeled Synovial Fibroblasts. The labeling of human synovial fibroblasts in their phospholipids and the extraction and separation of labeled lipids from the media on human synovial fibroblasts has been previously described in detail (3). For the experiments shown in Table 1, human synovial fibroblast monolayers were preincubated for 24 h with 5.2 • l0 s counts per minute of [~H]arachidonic acid containing 3 ml of media with I0% fetal calf serum, Approximately 42% of the label was taken up into the phospholipids of the human synovial fibroblasts (3). Since isotope uptake varies from cell culture to cell culture, data using this technique are reported as representative experiments rather than mean values. All experiments were repeated three times to confirm results. The remaining radioactive label in the media not taken up by the ceils was removed, and the ceils were washed twice with serum-free media and then replenished with fresh media with or without fetal calf serum, and with or without 1 • 10 -4 M quinacrine l h before the addition of 8.5 • 10 -6 M bradykinin or control solution (serum-free media only) for 2 min. Incubations were terminated by transferring the medium to a cold test tube containing 0.5 ml of 0.2 M citric acid; washes were added to the cold citric acid. The lipids from the acidic media were extracted with three volumes of a cold mixture of ethyl ether-methanol (9:1, v/v) following the addition of prostaglandin
Bradykinin and Quinacrine
237
Table 1. The Effect of Quinacrine on Cyclic AMP Response Induced by BK in H S F Quinacrine (M) pretreatment
5.0 2.5 1.0 7.5 5.0 2.5 1.0 7.5 5.0
•
10 -4
• • • • • • •
10 -4
X
Cyclic AMP (pmol/mg of cell protein)
Inhibition (%)
911.7 _+ 21.8 24.4 _+ 4.5 21.0 _+ 3.5 19.2 ___ 1.2 20.2 _+ 1.7 17.6 _+ 1.3 30.3 + 0.4 411.0 _+ 60.5 857.3 _+ 48.5 877.4 _+ 91.5
99.9 100 100 100 100 99.3 56.4 6.1 3.9
10-4 10-5 10 s 10-5 10-s 10-6 10-6
~The media on HSF monolayer cultures was changed to fresh EMEM supplemented with 10% FCS and the appropriate concentration of quinacrine 1 h before the addition of 8.5 • 10-6 M BK for 3 rain. The cyclic AMP content of untreated cells was 23.8 + 1.7 pmol/mg of cell protein. Quinacrine alone had no effect on intracellular cyclic AMP concentrations.
markers. After mixing and centrifugation, the top (solvent) layer was removed and evaporated to dryness. The residue was dissolved in 0.1 ml of a cold ethyl ether-methanol mixture (9:1, v / v) and applied to silica gel thin-layer chromatography sheets. Arachidonic acid and prostaglandins were separated in a solvent system of ethyl acetate-acetone-acetic acid (90:10:1, v/v) and counted for the 3H label. Initial experiments used four different solvent systems for the separation of prostaglandins and the alkaline-treated PGE2 fractions had an identical Ry value to PGBz (14, 15). These experiments confirmed the identity of the compounds and subsequent experiments used the aforementioned solvent system. Duplicate plates were used, and the values obtained agreed within 5% of the mean value shown. The arachidonic acid, prostaglandin E, and prostaglandin F values are in counts per minute; background counts from the thinlayer chromatogram were subtracted. Reagents. Prostaglandins were kindly supplied by Dr. John E. Pike or Upjohn. Quinacrine (atabrine) was obtained from Sterling Winthrop. Bradykinin triacetate and [3H]arachidonic acid were purchased from Sigma and New England Nuclear, respectively.
RESULTS
AND
DISCUSSION
Q u i n a c r i n e ( a t a b r i n e ) s u p p r e s s e s p r o s t a g l a n d i n E synthesis a n d b l o c k s m e c h a n i c a l l y s t i m u l a t e d release o f a r a c h i d o n i c acid a n d p r o s t a g l a n d i n s f r o m g u i n e a pig s p l e e n s (16), i n h i b i t s the b r a d y k i n i n - i n d u c e d release o f r a b bit a o r t a c o n t r a c t i n g s u b s t a n c e f r o m g u i n e a pig l u n g s (17), a n d is a n inh i b i t o r o f p h o s p h o l i p a s e A a c t i v i t y (18). Since r e c e n t r e p o r t s h a v e p r o p o s e d t h a t b r a d y k i n i n d e a c y l a t e s p h o s p h o l i p i d s (19, 20), q u i n a c r i n e was tested for its effect as a n i n h i b i t o r of the h u m a n s y n o v i a l f i b r o b l a s t r e s p o n s e s to bradykinin.
238
Fahey and Newcombe
Human synovial fibrobtasts were pretreated with quinacrine (5 X 10 -4 to 5 X 10 -6 M ) prior to the addition of bradykinin and then intracellular cyclic AMP was assayed. No alterations in human synovial fibroblast viability were noted. Quinacrine concentrations of greater than 2.5 • 10-5 M completely blocked bradykinin-induced cyclic AMP responses. The concentration of quinacrine resulting in a 50% inhibition of the bradykinininduced stimulation of human synovial fibroblast cyclic AMP was 9.7 X 10 -6 M (Table 1). As shown previously, fetal calf serum evoked the release of [3H]arachidonic acid and [3H]prostaglandins from prelabeled human synovial fibroblasts but did not alter the cyclic AMP content of these cells (3). The release of [3H]arachidonic acid and [3H]-E prostaglandins induced by media which contained fetal calf serum was completely abrogated by 1 X 10-4 M quinacrine (Table 2). When human synovial fibroblasts were incubated in serum-supplemented media, 1 X 10 -4 M quinacrine inhibited the bradykinin-evoked release of labeled arachidonic acid by 44%. Similarly, 1 X 10-4 M quinacrine suppressed the bradykinin-induced prostaglandin E response by more than half in human synovial fibroblasts incubated in both serum-free and serumsupplemented media. Thus, quinacrine inhibited the fetal calf serum- and bradykinin-induced release of arachidonic acid and prostaglandins from prelabeled human synovial fibroblasts, and 2.5 X 10 -4 M quinacrine completely blocked the cyclic AMP response to bradykinin. These data substantiate other evidence (3) that the factor responsible for the increase in intracellular cyclic AMP concentrations observed after the treatment of human synovial fibroblasts with bradykinin is dependent upon the synthesis and release of arachidonic acid and an intact prostaglandin synthetase pathway. In addition to inhibitors as regulators of the cyclic AMP and prostaglandin responses evoked by bradykinin on human synovial fibroblasts,
Table 2. Representative Effect of Quinacrine on Bradykinin-Induced Release of Arachidonic Acid on Prostaglandins from Prelabeled Human Synovial Fibroblasts
Serum-free media (cpm)
Control 10-4 M Quinacrine
Bradykinin 10-4 M Quinacrine + Bradykinin
Serum-supplemented media (cpm)
ARA
PGE
PGF
ARA
PGE
PGF
1243 1293 1590 1214
52 94 244 103
38 59 96 50
2432 1222 3396 2437
258 82 454 202
200 81 252 113
Bradykinin and Quinacrine
239
prostaglandins with opposing effects were tested to determine their effect on these reactions. We have shown previously that prostaglandin E~ greatly enhanced the cyclic AMP response of human synovial fibroblasts to bradykinin (21), and bradykinin (21) and prostaglandin E2 (22) induce maximal increases in human synovial fibroblast cyclic AMP concentrations 3 and 15 min after addition of the respective hormone. When human synovial fibroblast cyclic AMP content is compared after treatment of cells with saturating doses of either prostaglandin E2 alone for 15 min, bradykjnin alone for 3 min, or both drugs together for 15 and 3 min, respectively, the combination of prostaglandin E2 and bradykinin gave a more than additive response (Table 3A). The cyclic AMP response to both hormones was 65% greater than the response to each hormone alone. Bradykinin evokes the release of E prostaglandins from human synovial fibroblasts (3, Table 2) and cultured transformed fibroblasts (19, 20, 23), and some investigators have suggested that bradykinin exerts its activity solely through the release of E prostaglandins. Since saturating concentrations of bradykinin and prostaglandin E2 gave a more than additive cyclic AMP response in human synovial fibroblasts, it is unlikely that this response results from bradykinin stimulating the synthesis and release of superfluous quantities of prostaglandin E2. Prostaglandin E~ also potentiates the bradykinin-induced cyclic AMP response in an identical fashion (21). These data indicate that the receptor sites on human synovial fibroblasts for the E prostaglandins and the effector agent of the bradykinin response are most likely different and the bradykinininduced cyclic AMP response in human synovial fibroblasts is not completely dependent on E prostaglandins. Table 3. The Effect of Prostaglandins E2 and Fz~ on the Bradykinininduced Cyclic AMP Response of Human Synovial Fibroblasts Cyclic AMP (pmol/mg cell protein) -Bradykinin
+Bradykinina
23.6 + 2.5 376.4 + 28.2
'189.2 _+ 8.7 915.3 + 81.6
A None Prostaglandin E2 B None Prostaglandin F2c~
17.0 + 20.2 +
1.9 2.8
246.3 + 7.7 136.5 + 25.3
~Equimolar concentrations (8.5 • 10-6 M) of bradykinin (3 min) or the indicated prostaglandins (15 min) were added to human synovial fibroblast monolayers. With mixed hormone treatments, the prostaglandins were each added for 12 min before the addition of bradykinin for 3 min at the same concentrations.
240
Fahey and Newcombe
Using a similar protocol, the effect of prostaglandin F2~ on the bradykinin-induced cyclic AMP response was studied (Table 3B). Prostaglandin F2~ inhibited the cyclic AMP response to bradykinin by 48%. F prostaglandins have antagonistic effects on E prostaglandins in some tissues (24, 25), and bradykinin administration can cause the release of prostaglandin F2~ as well as the E prostaglandins. Therefore, it is possible that the E and F prostaglandins modulate the human synovial fibroblast response to bradykinin in vivo, since prostaglandin E2 potentiated and prostaglandin F2~ inhibited bradykinin responses in the human synovial fibroblasts in a tissue culture system. Bradykinin (4-8) and prostaglandin E2 (10, 26, 27) have been found in high concentrations in synovial fluids of patients with inflammatory arthritides, and the potent effects of prostaglandins E1 and E2 on the inflammatory response (28), bone resorption (29), pain production (30), and leukocyte chemotaxis (31) demonstrates the need for agents that can modify the cellular reactivity to bradykinin and E prostaglandins. The capacity of quinacrine to inhibit both the release of cellular phospholipids by bradykinin and the subsequent synthesis and release of E prostaglandins and the modulation of bradykinin responsiveness of F prostaglandins provide models for the study of therapeutic approaches to inflammatory arthritides by the use of these compounds or their analogs. Although the actual roles of bradykinin and prostaglandins in the pathogenesis of inflammatory arthritides have not been completely defined, prostaglandin E2 is produced in increased quantities by rheumatoid synovia (9). This same class of compounds suppresses both humoral and cell-mediated immune responses (32) which are known to participate in the synovial reaction seen in certain forms of arthritis. Bradykinin, as a chemotactic agent, vasodilator, and algesic agent, has the potential to augment an inflammatory reaction. Thus, pharmacologic agents such as quinacrine and prostaglandin F2~ that can modulate the activity of bradykinin and prostaglandin E2 will provide a means for investigating their role in inflammatory arthritides. Data have also been presented to support the case against E prostaglandins as the sole effectors of the bradykinin-induced cyclic AMP response and to suggest that the receptor for E prostaglandins and the effector agent for the bradykinin-induced cyclic AMP response are separate. The identification of this kinin effector agent would permit investigations concerned with the regulation of kinin receptors and bradykinin-induced, cyclic AMPregulated responses of synovial cells. Acknowledgments--The authors gratefully acknowledge support from the Arthritis Foundation (Vermont Chapter), American Heart Association (grant 77-784).
Bradykinin and Quinacrine
241
REFERENCES 1. BAZIN S., M. PELLETIER, and A. DELAUNAY. 1973. The influence of chemical mediators of acute inflammation on the cells of a subacute inflammation. Agents Actions 3:317-322. 2. BOUCEK, R.J., N. L. NOBEL. 1973. Histamine, norepinephrine, and bradykinin stimulation of flbroblast growth and modification of serotonin response. Proc. Soc. Exp. Biol. Med. 144:929-933. 3. NEWCOMBE,D. S., J. V. FAHEY, and Y. 1SHIKAWA. 1977. Hydrocortisone inhibition of the bradykinin activation of human synovial fibroblasts. Prostaglandins 13:235-244. 4. MELMON,K. L., M. E. WEBSTER, S. E. GOLDFINGER, and J. E. SEEGMILLER. 1967. The presence of a kinin in inflammatory synovial effusion from arthritides of varying etiologies. Arthritis Rheum 10:13-20. 5. WEBSTER,M. E., and H.M. MALING. 1970. Evidence for and against the kinins as endogenous mediators of arthritis. In Bradykinin and Related Kinins. F. Sicuteri, M. Rocha e Silva, and N. Back, editors. Plenum Press, New York, pp. 493-501. 6. KEELE, C.A., and V. EISEN. 1970. Plasma Kinin formation in rheumatoid arthritis. In Bradykinin and Related Kinins. F. Sicuteri, M. Rocha e Silva, and N. Back, editors. Plenum Press, New York, pp. 471-475. 7. EISEN, V. 1970. Plasma kinins in synovial exudates. Br. J. exp. Pathol. 51:322-327. 8. GOLDFINGER, S., K.L. MELMON, M. E. WEBSTER, and J. E. SEEGMILLER. 1964. The presence of a kinin-peptide in inflammatory synovial effusions. Arthritis Rheum 7:311-312. 9. DAYER, J. M., S. M. KRANE, R. G. G. RUSSELL, and D. R. ROBINSON. 1976. Production of collagenase and prostaglandins by isolated adherent rheumatoid synovial cells. Proc. NatL Acad. Sci. U.S.A. 73:945-949. 10. ROBINSON, D. R., and L. LEVINE. 1974. Prostagtandin concentrations in synovial fluid in rheumatic diseases: Action of indomethacin and aspirin. In Prostaglandin Synthetase Inhibitors. J. H. Robinson and J. R. Vane, editors. Raven Press, New York, pp. 223-228. 11. CIOSEK, C. P. JR., J. V. FAHEY, and D. S. NEWCOM~E. 1975. PGEl-mediated cyclic AMP refractoriness: Effects of cycloheximide and indomethacin. J. Cyclic Nucleotide Res. 1:229-235. 12. LowRy, O. H., N. J. ROSEBROUGH,A. L. FARR, and R. J. RANDALL. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. 13. GILMAN, A.G. 1970. A Protein binding assay for adenosine 3':5'-cyclic monophosphate. Proc. Natl. Acad. Sci. U.S.A. 67:305-312. 14. ISHIKAWA, Y., C.P. CIOSEK, JR., J. V. EAHEY, and D. S. NEWCOMBE. 1976. Prostaglandin synthetase activity and hormone responsiveness in normal and SV40 transformed WI - 38 fibroblasts. J. Cyclic Nucleotide Res. 2:115-128. 15. WALLACH,D. P., and E. G. DANIELS. 1971. Properties of a novel preparation of prostaglandin synthetase from sheep seminal vesicles. Biochim. Biophys. Acta 231:445-457. 16. FLOWER, R.J., and G.J. BLACRWELL. 1976. The importance of phospholipase-A2 in prostaglandin biosynthesis. Biochem. Pharmacol. 25:285-291. 17. VARGAFTIG,B. B., and N. D. HAI. 1972. Selective inhibition by mepacrine of the release of "rabbit aorta contracting substance" evoked by the administration of bradykinin. J. PharmacoL 24:159-161. 18. MARKUS, H. B., and E. G. BALL. 1969. Inhibition of lipolytic processes in rat adipose tissue by antimalaria drugs. Biochim. Biophys. Acta 187:486-491. 19. HONG, S.-C. L., R. POLSKY-CYNKIN, and L. LEVlNE. 1976. Stimulation of prostaglandin synthesis by bradykinin and thrombin and their mechanisms of action of MC5-5 fibroblasts. J. Biol. Chem. 251:5814-5816.
242
Fahey and Newcombe
20. PONG, S.-S., S.-C. L. HONG,and L. LEVINE. 1977. Prostaglandin production by methyl-
21.
22.
23.
24. 25.
26.
27. 28. 29.
30.
31.
32.
cholanthrene-transformed mouse BALB/3T3. Requirement for protein synthesis. J. BioL Chem. 252:1408-1413. FAHEY,J. V., C. P. CIOSEK,JR., and D. S. NEWCOMBE.1977. Human synovial fibroblasts: The relationship between cyclic AMP, bradykinin, and prostaglandins. Agents Actions 7:255-265. NEWCOMBE,D. S., C. P. C1OSEK,JR., Y. ISHIKAWA,and J. V. FAHEY.1975. Human synoviocytes: Activation and desensitization by prostaglandins and /-epinephrine. Proc. Natl. Aead. Sci. U.S.A. 72:3124-3128. HONG, S.-C. L., R. POLSKY-CYNKIN,and L. LEVINE. 1976. Stimulation of prostaglandin biosynthesis by vasoactive substances in methylcholanthrene-transformed mouse BALB/ 3T3. J. Biol. Chem. 251:776-780. CRUN~:HORN,P., and A. L. WILLlS. 1971. Cutaneous reactions to intradermal prostaglandins. Br. J. Pharmacol. 41:507-512. JUAN, H., and R. LEMBEC~. 1977. Prostaglandin F2~ reduces the algesic effect of bradykinin by antagonizing the pain enhancing action of endogenously released prostaglandin E. Br. J. Pharmacol. 59:385-391. HmGS, G.A., J.R. VANE, F.D. HART, and J.A. WOJTULEWSKI. 1974. Effects of antiinflammatory drugs on prostaglandins in rheumatoid arthritis. In Prostaglandin Synthetase Inhibitors. J. H. Robinson and J.R. Vane, editors. Raven Press, New York, pp. 165-173. ROBINSON, H.J., and J. L. GRANDA. 1974. Prostaglandins in synovial inflammatory disease. Surg. Forum 25:476-477. ZURIER, R.B. 1974. Prostaglandins. In Mediators of Inflammation. G. Weissman, editor. Plenum Press, New York, pp. 163-180. ROBINSON, D. R., A. H. TASHJ1AN,JR., and L. LEVINE. 1975. Prostaglandin-stimulated bone resorption by rheumatoid synovia. A possible mechanism for bone destruction in rheumatoid arthritis. J. Clin. Invest. 56:1181-1188. GILLESPrE, A. 1972. Prostaglandin-oxytocin enhancement and protentiation and their clinical applications. Br. Med. J. 1:150-152. McCLATCttEY,W., and R. SYNDERNAN.1976. Prostaglandins and inflammation: Enhancement of monocyte chemotactic responsiveness by prostaglandin E2. Prostaglandins 12:415425. PELUS, L. M., and H. R. STRAUSSER.1972. Prostaglandins and the immune response. Life Sci. 20:903-914.
REGULATION OF BRADYKININ-INDUCED CYCLIC AMP RESPONSE BY QUINACRINE AND PROSTAGLANDIN E2 AND Fza IN HUMAN SYNOVIAL FIBROBLASTS JOHN V. FAHEY and DAVID S. NEWCOMBE Department of Medicine, University of Vermont, College of Medicine, Burlington, Vermont, and Departments of Medicine and Environmental Health Sciences, Johns Hopkins University, Baltimore, Maryland, 21205
Abstract--Bradykinin induces an increment in intracellular cyclic AMP concentrations of human synovial fibroblasts and evokes the release of [3H]arachidonic acid and [3H]-E prostaglandins from human synovial fibroblasts prelabeled in their phospholipids. Both these bradykinin-induced reactions are inhibited by quinacrine, an inhibitor of phospholipase A activity. The cyclic AMP response of human synovial fibroblasts to bradykinin is potentiated by prostaglandin E2 and inhibited by prostaglandin F2~. These data emphasize the critical role of the prostaglandin system in reactions induced by bradykinin and suggest mechanisms by which inflammatory reactions due to bradykinin may be modulated.
INTRODUCTION Bradykinin, a potent algesic nonapeptide with properties capable of inducing chemotaxis and cell proliferation (1, 2), evokes an increment in the intracellular cyclic AMP concentrations in human synovial fibroblasts but does not stimulate adenylate cyclase activity in membrane preparations of these cells (3). Bradykinin concentrations are higher in synovial fluids from patients with chronic proliferative and acute inflammatory arthritides than in those with noninflammatory joint disease (4-8). This same nonapeptide evokes an increase in intracellular cyclic adenosine 3',5"monophosphate (cyclic AMP) concentrations in human synovial fibroblasts; it also stimulates the release of E prostagtandins from these cells (3). E prostaglandins are found in increased quantities in rheumatoid synovial effusions and are produced in large quantities by rheumatoid synovial tissues (9, 10). The antiinflammatory agents, indomethacin and hydrocortisone, inhibit both the 235 0360-3997/79/0700-9235503.00/09 1979PlenumPublishingCorporation
236
Fahey and Newcomhe
kinin-induced cyclic AMP response and E prostagland~n release (3). These findings support the hypothesis that bradykinin is a key mediator of the tissue responses observed in certain inflammatory arthritides. Further, pharmacologic agents regulating bradykinin activated processes could be used to examine the role of kinins in inflammatory arthritis. We report here the inhibition of the bradykinin-induced cyclic AMP response in intact human synovial fibroblasts by quinacrine and the modulation of this cyclic AMP response by E2 and F2~ prostaglandins.
MATERIALS AND METHODS Cell Culture, Human synovial membrane was obtained during surgical procedures and synovial fibroblast cultures were initiated and maintained as previously described (3, 11). Cells were cultured in minimal essential medium containing Earle's salts and L-glutamine, supplemented with 10% fetal bovine serum (Reheis Chemical Company) and 100 units/ml penicillin and 100 /zg/ml streptomycin (GIBCO). For the experiments described, synovial fibroblasts obtained from patients without systemic disease were used. Cell viability was tested by dye exclusion using trypan blue. Cell Treatment with Pharmacoactive Agents. Cells were grown to monolayers in 60- • 15-mm tissue culture dishes (Corning). Media was changed 1 h prior to bradykinin treatment unless otherwise specified. All reagents were solubilized and prepared as 10• solutions in serumfree media. Prostaglandins were solubilized in absolute ethanol; the final concentration of ethanol was always less than 0.5%. Bradykinin and quinacrine were solubilized directly in ice-cold serum-free media, Cell Processing and Cyclic AMP Assay. After incubations were terminated with trichloroacetic acid, the cells were scraped and assayed for protein by the Lowry method (12). Intracellular cyclic AMP concentrations were determined by the Gilman binding-protein assay without addition of protein kinase inhibitor (13), as described previously (3, 11). Purified beef heart phosphodiesterase treatment of processed cells resulted in a greater than 95% decrease in the intracellular cyclic AMP concentration. Data are expressed as picomoles of cyclic AMP per milligram of cell protein + one standard deviation. Preparation and Use of Radioactively Labeled Synovial Fibroblasts. The labeling of human synovial fibroblasts in their phospholipids and the extraction and separation of labeled lipids from the media on human synovial fibroblasts has been previously described in detail (3). For the experiments shown in Table 1, human synovial fibroblast monolayers were preincubated for 24 h with 5.2 • l0 s counts per minute of [~H]arachidonic acid containing 3 ml of media with I0% fetal calf serum, Approximately 42% of the label was taken up into the phospholipids of the human synovial fibroblasts (3). Since isotope uptake varies from cell culture to cell culture, data using this technique are reported as representative experiments rather than mean values. All experiments were repeated three times to confirm results. The remaining radioactive label in the media not taken up by the ceils was removed, and the ceils were washed twice with serum-free media and then replenished with fresh media with or without fetal calf serum, and with or without 1 • 10 -4 M quinacrine l h before the addition of 8.5 • 10 -6 M bradykinin or control solution (serum-free media only) for 2 min. Incubations were terminated by transferring the medium to a cold test tube containing 0.5 ml of 0.2 M citric acid; washes were added to the cold citric acid. The lipids from the acidic media were extracted with three volumes of a cold mixture of ethyl ether-methanol (9:1, v/v) following the addition of prostaglandin
Bradykinin and Quinacrine
237
Table 1. The Effect of Quinacrine on Cyclic AMP Response Induced by BK in H S F Quinacrine (M) pretreatment
5.0 2.5 1.0 7.5 5.0 2.5 1.0 7.5 5.0
•
10 -4
• • • • • • •
10 -4
X
Cyclic AMP (pmol/mg of cell protein)
Inhibition (%)
911.7 _+ 21.8 24.4 _+ 4.5 21.0 _+ 3.5 19.2 ___ 1.2 20.2 _+ 1.7 17.6 _+ 1.3 30.3 + 0.4 411.0 _+ 60.5 857.3 _+ 48.5 877.4 _+ 91.5
99.9 100 100 100 100 99.3 56.4 6.1 3.9
10-4 10-5 10 s 10-5 10-s 10-6 10-6
~The media on HSF monolayer cultures was changed to fresh EMEM supplemented with 10% FCS and the appropriate concentration of quinacrine 1 h before the addition of 8.5 • 10-6 M BK for 3 rain. The cyclic AMP content of untreated cells was 23.8 + 1.7 pmol/mg of cell protein. Quinacrine alone had no effect on intracellular cyclic AMP concentrations.
markers. After mixing and centrifugation, the top (solvent) layer was removed and evaporated to dryness. The residue was dissolved in 0.1 ml of a cold ethyl ether-methanol mixture (9:1, v / v) and applied to silica gel thin-layer chromatography sheets. Arachidonic acid and prostaglandins were separated in a solvent system of ethyl acetate-acetone-acetic acid (90:10:1, v/v) and counted for the 3H label. Initial experiments used four different solvent systems for the separation of prostaglandins and the alkaline-treated PGE2 fractions had an identical Ry value to PGBz (14, 15). These experiments confirmed the identity of the compounds and subsequent experiments used the aforementioned solvent system. Duplicate plates were used, and the values obtained agreed within 5% of the mean value shown. The arachidonic acid, prostaglandin E, and prostaglandin F values are in counts per minute; background counts from the thinlayer chromatogram were subtracted. Reagents. Prostaglandins were kindly supplied by Dr. John E. Pike or Upjohn. Quinacrine (atabrine) was obtained from Sterling Winthrop. Bradykinin triacetate and [3H]arachidonic acid were purchased from Sigma and New England Nuclear, respectively.
RESULTS
AND
DISCUSSION
Q u i n a c r i n e ( a t a b r i n e ) s u p p r e s s e s p r o s t a g l a n d i n E synthesis a n d b l o c k s m e c h a n i c a l l y s t i m u l a t e d release o f a r a c h i d o n i c acid a n d p r o s t a g l a n d i n s f r o m g u i n e a pig s p l e e n s (16), i n h i b i t s the b r a d y k i n i n - i n d u c e d release o f r a b bit a o r t a c o n t r a c t i n g s u b s t a n c e f r o m g u i n e a pig l u n g s (17), a n d is a n inh i b i t o r o f p h o s p h o l i p a s e A a c t i v i t y (18). Since r e c e n t r e p o r t s h a v e p r o p o s e d t h a t b r a d y k i n i n d e a c y l a t e s p h o s p h o l i p i d s (19, 20), q u i n a c r i n e was tested for its effect as a n i n h i b i t o r of the h u m a n s y n o v i a l f i b r o b l a s t r e s p o n s e s to bradykinin.
238
Fahey and Newcombe
Human synovial fibrobtasts were pretreated with quinacrine (5 X 10 -4 to 5 X 10 -6 M ) prior to the addition of bradykinin and then intracellular cyclic AMP was assayed. No alterations in human synovial fibroblast viability were noted. Quinacrine concentrations of greater than 2.5 • 10-5 M completely blocked bradykinin-induced cyclic AMP responses. The concentration of quinacrine resulting in a 50% inhibition of the bradykinininduced stimulation of human synovial fibroblast cyclic AMP was 9.7 X 10 -6 M (Table 1). As shown previously, fetal calf serum evoked the release of [3H]arachidonic acid and [3H]prostaglandins from prelabeled human synovial fibroblasts but did not alter the cyclic AMP content of these cells (3). The release of [3H]arachidonic acid and [3H]-E prostaglandins induced by media which contained fetal calf serum was completely abrogated by 1 X 10-4 M quinacrine (Table 2). When human synovial fibroblasts were incubated in serum-supplemented media, 1 X 10 -4 M quinacrine inhibited the bradykinin-evoked release of labeled arachidonic acid by 44%. Similarly, 1 X 10-4 M quinacrine suppressed the bradykinin-induced prostaglandin E response by more than half in human synovial fibroblasts incubated in both serum-free and serumsupplemented media. Thus, quinacrine inhibited the fetal calf serum- and bradykinin-induced release of arachidonic acid and prostaglandins from prelabeled human synovial fibroblasts, and 2.5 X 10 -4 M quinacrine completely blocked the cyclic AMP response to bradykinin. These data substantiate other evidence (3) that the factor responsible for the increase in intracellular cyclic AMP concentrations observed after the treatment of human synovial fibroblasts with bradykinin is dependent upon the synthesis and release of arachidonic acid and an intact prostaglandin synthetase pathway. In addition to inhibitors as regulators of the cyclic AMP and prostaglandin responses evoked by bradykinin on human synovial fibroblasts,
Table 2. Representative Effect of Quinacrine on Bradykinin-Induced Release of Arachidonic Acid on Prostaglandins from Prelabeled Human Synovial Fibroblasts
Serum-free media (cpm)
Control 10-4 M Quinacrine
Bradykinin 10-4 M Quinacrine + Bradykinin
Serum-supplemented media (cpm)
ARA
PGE
PGF
ARA
PGE
PGF
1243 1293 1590 1214
52 94 244 103
38 59 96 50
2432 1222 3396 2437
258 82 454 202
200 81 252 113
Bradykinin and Quinacrine
239
prostaglandins with opposing effects were tested to determine their effect on these reactions. We have shown previously that prostaglandin E~ greatly enhanced the cyclic AMP response of human synovial fibroblasts to bradykinin (21), and bradykinin (21) and prostaglandin E2 (22) induce maximal increases in human synovial fibroblast cyclic AMP concentrations 3 and 15 min after addition of the respective hormone. When human synovial fibroblast cyclic AMP content is compared after treatment of cells with saturating doses of either prostaglandin E2 alone for 15 min, bradykjnin alone for 3 min, or both drugs together for 15 and 3 min, respectively, the combination of prostaglandin E2 and bradykinin gave a more than additive response (Table 3A). The cyclic AMP response to both hormones was 65% greater than the response to each hormone alone. Bradykinin evokes the release of E prostaglandins from human synovial fibroblasts (3, Table 2) and cultured transformed fibroblasts (19, 20, 23), and some investigators have suggested that bradykinin exerts its activity solely through the release of E prostaglandins. Since saturating concentrations of bradykinin and prostaglandin E2 gave a more than additive cyclic AMP response in human synovial fibroblasts, it is unlikely that this response results from bradykinin stimulating the synthesis and release of superfluous quantities of prostaglandin E2. Prostaglandin E~ also potentiates the bradykinin-induced cyclic AMP response in an identical fashion (21). These data indicate that the receptor sites on human synovial fibroblasts for the E prostaglandins and the effector agent of the bradykinin response are most likely different and the bradykinininduced cyclic AMP response in human synovial fibroblasts is not completely dependent on E prostaglandins. Table 3. The Effect of Prostaglandins E2 and Fz~ on the Bradykinininduced Cyclic AMP Response of Human Synovial Fibroblasts Cyclic AMP (pmol/mg cell protein) -Bradykinin
+Bradykinina
23.6 + 2.5 376.4 + 28.2
'189.2 _+ 8.7 915.3 + 81.6
A None Prostaglandin E2 B None Prostaglandin F2c~
17.0 + 20.2 +
1.9 2.8
246.3 + 7.7 136.5 + 25.3
~Equimolar concentrations (8.5 • 10-6 M) of bradykinin (3 min) or the indicated prostaglandins (15 min) were added to human synovial fibroblast monolayers. With mixed hormone treatments, the prostaglandins were each added for 12 min before the addition of bradykinin for 3 min at the same concentrations.
240
Fahey and Newcombe
Using a similar protocol, the effect of prostaglandin F2~ on the bradykinin-induced cyclic AMP response was studied (Table 3B). Prostaglandin F2~ inhibited the cyclic AMP response to bradykinin by 48%. F prostaglandins have antagonistic effects on E prostaglandins in some tissues (24, 25), and bradykinin administration can cause the release of prostaglandin F2~ as well as the E prostaglandins. Therefore, it is possible that the E and F prostaglandins modulate the human synovial fibroblast response to bradykinin in vivo, since prostaglandin E2 potentiated and prostaglandin F2~ inhibited bradykinin responses in the human synovial fibroblasts in a tissue culture system. Bradykinin (4-8) and prostaglandin E2 (10, 26, 27) have been found in high concentrations in synovial fluids of patients with inflammatory arthritides, and the potent effects of prostaglandins E1 and E2 on the inflammatory response (28), bone resorption (29), pain production (30), and leukocyte chemotaxis (31) demonstrates the need for agents that can modify the cellular reactivity to bradykinin and E prostaglandins. The capacity of quinacrine to inhibit both the release of cellular phospholipids by bradykinin and the subsequent synthesis and release of E prostaglandins and the modulation of bradykinin responsiveness of F prostaglandins provide models for the study of therapeutic approaches to inflammatory arthritides by the use of these compounds or their analogs. Although the actual roles of bradykinin and prostaglandins in the pathogenesis of inflammatory arthritides have not been completely defined, prostaglandin E2 is produced in increased quantities by rheumatoid synovia (9). This same class of compounds suppresses both humoral and cell-mediated immune responses (32) which are known to participate in the synovial reaction seen in certain forms of arthritis. Bradykinin, as a chemotactic agent, vasodilator, and algesic agent, has the potential to augment an inflammatory reaction. Thus, pharmacologic agents such as quinacrine and prostaglandin F2~ that can modulate the activity of bradykinin and prostaglandin E2 will provide a means for investigating their role in inflammatory arthritides. Data have also been presented to support the case against E prostaglandins as the sole effectors of the bradykinin-induced cyclic AMP response and to suggest that the receptor for E prostaglandins and the effector agent for the bradykinin-induced cyclic AMP response are separate. The identification of this kinin effector agent would permit investigations concerned with the regulation of kinin receptors and bradykinin-induced, cyclic AMPregulated responses of synovial cells. Acknowledgments--The authors gratefully acknowledge support from the Arthritis Foundation (Vermont Chapter), American Heart Association (grant 77-784).
Bradykinin and Quinacrine
241
REFERENCES 1. BAZIN S., M. PELLETIER, and A. DELAUNAY. 1973. The influence of chemical mediators of acute inflammation on the cells of a subacute inflammation. Agents Actions 3:317-322. 2. BOUCEK, R.J., N. L. NOBEL. 1973. Histamine, norepinephrine, and bradykinin stimulation of flbroblast growth and modification of serotonin response. Proc. Soc. Exp. Biol. Med. 144:929-933. 3. NEWCOMBE,D. S., J. V. FAHEY, and Y. 1SHIKAWA. 1977. Hydrocortisone inhibition of the bradykinin activation of human synovial fibroblasts. Prostaglandins 13:235-244. 4. MELMON,K. L., M. E. WEBSTER, S. E. GOLDFINGER, and J. E. SEEGMILLER. 1967. The presence of a kinin in inflammatory synovial effusion from arthritides of varying etiologies. Arthritis Rheum 10:13-20. 5. WEBSTER,M. E., and H.M. MALING. 1970. Evidence for and against the kinins as endogenous mediators of arthritis. In Bradykinin and Related Kinins. F. Sicuteri, M. Rocha e Silva, and N. Back, editors. Plenum Press, New York, pp. 493-501. 6. KEELE, C.A., and V. EISEN. 1970. Plasma Kinin formation in rheumatoid arthritis. In Bradykinin and Related Kinins. F. Sicuteri, M. Rocha e Silva, and N. Back, editors. Plenum Press, New York, pp. 471-475. 7. EISEN, V. 1970. Plasma kinins in synovial exudates. Br. J. exp. Pathol. 51:322-327. 8. GOLDFINGER, S., K.L. MELMON, M. E. WEBSTER, and J. E. SEEGMILLER. 1964. The presence of a kinin-peptide in inflammatory synovial effusions. Arthritis Rheum 7:311-312. 9. DAYER, J. M., S. M. KRANE, R. G. G. RUSSELL, and D. R. ROBINSON. 1976. Production of collagenase and prostaglandins by isolated adherent rheumatoid synovial cells. Proc. NatL Acad. Sci. U.S.A. 73:945-949. 10. ROBINSON, D. R., and L. LEVINE. 1974. Prostagtandin concentrations in synovial fluid in rheumatic diseases: Action of indomethacin and aspirin. In Prostaglandin Synthetase Inhibitors. J. H. Robinson and J. R. Vane, editors. Raven Press, New York, pp. 223-228. 11. CIOSEK, C. P. JR., J. V. FAHEY, and D. S. NEWCOM~E. 1975. PGEl-mediated cyclic AMP refractoriness: Effects of cycloheximide and indomethacin. J. Cyclic Nucleotide Res. 1:229-235. 12. LowRy, O. H., N. J. ROSEBROUGH,A. L. FARR, and R. J. RANDALL. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. 13. GILMAN, A.G. 1970. A Protein binding assay for adenosine 3':5'-cyclic monophosphate. Proc. Natl. Acad. Sci. U.S.A. 67:305-312. 14. ISHIKAWA, Y., C.P. CIOSEK, JR., J. V. EAHEY, and D. S. NEWCOMBE. 1976. Prostaglandin synthetase activity and hormone responsiveness in normal and SV40 transformed WI - 38 fibroblasts. J. Cyclic Nucleotide Res. 2:115-128. 15. WALLACH,D. P., and E. G. DANIELS. 1971. Properties of a novel preparation of prostaglandin synthetase from sheep seminal vesicles. Biochim. Biophys. Acta 231:445-457. 16. FLOWER, R.J., and G.J. BLACRWELL. 1976. The importance of phospholipase-A2 in prostaglandin biosynthesis. Biochem. Pharmacol. 25:285-291. 17. VARGAFTIG,B. B., and N. D. HAI. 1972. Selective inhibition by mepacrine of the release of "rabbit aorta contracting substance" evoked by the administration of bradykinin. J. PharmacoL 24:159-161. 18. MARKUS, H. B., and E. G. BALL. 1969. Inhibition of lipolytic processes in rat adipose tissue by antimalaria drugs. Biochim. Biophys. Acta 187:486-491. 19. HONG, S.-C. L., R. POLSKY-CYNKIN, and L. LEVlNE. 1976. Stimulation of prostaglandin synthesis by bradykinin and thrombin and their mechanisms of action of MC5-5 fibroblasts. J. Biol. Chem. 251:5814-5816.
242
Fahey and Newcombe
20. PONG, S.-S., S.-C. L. HONG,and L. LEVINE. 1977. Prostaglandin production by methyl-
21.
22.
23.
24. 25.
26.
27. 28. 29.
30.
31.
32.
cholanthrene-transformed mouse BALB/3T3. Requirement for protein synthesis. J. BioL Chem. 252:1408-1413. FAHEY,J. V., C. P. CIOSEK,JR., and D. S. NEWCOMBE.1977. Human synovial fibroblasts: The relationship between cyclic AMP, bradykinin, and prostaglandins. Agents Actions 7:255-265. NEWCOMBE,D. S., C. P. C1OSEK,JR., Y. ISHIKAWA,and J. V. FAHEY.1975. Human synoviocytes: Activation and desensitization by prostaglandins and /-epinephrine. Proc. Natl. Aead. Sci. U.S.A. 72:3124-3128. HONG, S.-C. L., R. POLSKY-CYNKIN,and L. LEVINE. 1976. Stimulation of prostaglandin biosynthesis by vasoactive substances in methylcholanthrene-transformed mouse BALB/ 3T3. J. Biol. Chem. 251:776-780. CRUN~:HORN,P., and A. L. WILLlS. 1971. Cutaneous reactions to intradermal prostaglandins. Br. J. Pharmacol. 41:507-512. JUAN, H., and R. LEMBEC~. 1977. Prostaglandin F2~ reduces the algesic effect of bradykinin by antagonizing the pain enhancing action of endogenously released prostaglandin E. Br. J. Pharmacol. 59:385-391. HmGS, G.A., J.R. VANE, F.D. HART, and J.A. WOJTULEWSKI. 1974. Effects of antiinflammatory drugs on prostaglandins in rheumatoid arthritis. In Prostaglandin Synthetase Inhibitors. J. H. Robinson and J.R. Vane, editors. Raven Press, New York, pp. 165-173. ROBINSON, H.J., and J. L. GRANDA. 1974. Prostaglandins in synovial inflammatory disease. Surg. Forum 25:476-477. ZURIER, R.B. 1974. Prostaglandins. In Mediators of Inflammation. G. Weissman, editor. Plenum Press, New York, pp. 163-180. ROBINSON, D. R., A. H. TASHJ1AN,JR., and L. LEVINE. 1975. Prostaglandin-stimulated bone resorption by rheumatoid synovia. A possible mechanism for bone destruction in rheumatoid arthritis. J. Clin. Invest. 56:1181-1188. GILLESPrE, A. 1972. Prostaglandin-oxytocin enhancement and protentiation and their clinical applications. Br. Med. J. 1:150-152. McCLATCttEY,W., and R. SYNDERNAN.1976. Prostaglandins and inflammation: Enhancement of monocyte chemotactic responsiveness by prostaglandin E2. Prostaglandins 12:415425. PELUS, L. M., and H. R. STRAUSSER.1972. Prostaglandins and the immune response. Life Sci. 20:903-914.
