- Email: [email protected]
Adenosine analogs: Potentiation of bradykinin-induced plasma exudation in rat skin and prevention by caffeine and theophylline
Pergamon Press
Life Sciences, Vol. 35, pp. 1575-1583 Printed in the U.S.A.
ADENOSINE ANALOGS: POTENTIATION OF BRADYKININ-INDUCED PLASMA EXUDATION IN RAT SKIN AND PREVENTION BY CAFFEINE AND THEOPHYLLINE Kazuo Sugio and John W. Daly Laboratory of Bioorganic Chemistry, Bldg. 4, Room 212, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20205, U.S.A. (Received in final form July 24, 1984) Summary Adenoslne and various analogslyytentiated plasma exudation elicited by I-labelled bovine serum albumin b a&ykinin In rat skin using (I I-BSA) as a tracer. L-N6-Phenylrsopropybadenoslne (L-PIA) was much more effective than D-PIA, adenoslne, N -cyclohexyladenoslne (CHA) and 2-chloroadenosine, all of which were comparable in activity. Adenoslne 5'-cyclopropylcarboxamide was the least effective analog. Caffeine and theophylllne had no effect on basal or bradykinln-ellclted plasma exudation, while inhibiting plasma exudatron eliclted by L-PIA, CHA or a combination of bradykinin and L-PIA. 8-Phenyltheophylline was more potent than caffeine or theophylllne versus the bradykinin and L-PIA combrnation. 2',5'-Dideoxyadenoslne, a P-site inhibitor of adenylate cyclase, had no effect on plasma exudation elicited by bradyklnin, L-PIA or a combination of bradykinln and L-PIA, but did inhibit plasma exudation eliclted by prostaglandin E lh(PGEa) Or a bradykinin-PGEl-combination. The antihistamine cypro epta ine slightly reduced plasma exudation elicited by a bradykinin-PGEl combination. The results suggest that adenosine potentiates bradykinin-induced plasma exudation via an adenosine receptor and that histamine may be involved to some extent in the phenomenon. Adenosine, a potent vasodilator, increases vascular permeability and potentiates the plasma exudation induced with bradykinin in rat skin (1) and enhances histamine release from mast cells (2). Furthermore adenosine receptor antagonlsts such as caffeine inhibit the inflammatory response induced with oxazolone in rat ear (3) and antagonize histamine release by mast cells (4). Such results suggest that adenoslne may play a significant role in the increase in the vascular permeability which is associated with Two classes of adenosine receptors have been proposed: An inflammation A2-receptor stimulatory to adenylate cyclase and and Al-receptor lnhlbitory to adenylate cyclase (5). Both classes of receptors are antagonized by methylxanthines such as caffeine and theophylline. The nature of the adenosine receptor which 1s involved in potentiatlon of bradykinininduced plasma exudation has been Investigated using a variety of adenosine analogs and adenosine receptor antagonlsts such as caffeine, theophylllne, Possible involvement of histamine was investiand 8-phenyltheophylline gated with an antihistamine, cyproheptadine.
0024-3205184 $3.00 + .OO
1576
Adenosine Analogs and Bradykinin
Vol. 35, No. 15, 1984
Materials and Methods Animals Male rats of Sprague-Dawley strain (body weight, 160-180 g) were used throughout the experiments. All rats were maintained on laboratory food and tap water ad libitum. Materials 125 The sourcea23f various agents were as follows: [ I]-labelled bovine serum albumin ( I-BSA, specific activity 60 uCi/mmol, New England Nuclear co., Boston, MA), caffeine, theophylline, 8-phenyltheophylline, adenosine and 2-chloroadenosine were from Calbiochem-Behring Co. (La Jolla, CA), LPIA and D-PIA were from Boehringer Co. (Sohn, IngelheimjRhein). Bradykinin and cyproheptadine were from Sigma Chemicals Inc. (St. Louis, MO). 2',5'Dideoxyadenosine was from ICN Chemicals and Radioisotope Division (Irvine, Adenosine 5'-cyclopropylcarboxamide was from Abbott Laboratories W. (Chicago, IL). Induction and measurement of vascular permeability response in rat skin Vascular permeability response was induced on shaved dorsal skin of the rats by the intradermal injection wit P23hlogistic agents (6,7). I-BSA as a tracer. PurifiVascular permeability waslysasured using cation of the commercial I-BSA was af53 rding to the method of Sugio et I-BSA In 0.2 ml of 0.9% NaCl al (8,9). About 1 uCi of the purified solution containing Evans Blue in 1.0% was injected into the tail vein of the rats. Immediately after the injection of tracer, phlogistic agents (0.1 ml) were inJected intradermally on the back under light ether anesthesia. Rats were killed by cutting the carotid arteries and the dorsal skin was peeled off and separated from adherent subcutaneous tissues. The blue area (20 mm diameter) of the rat skin at each site of inlection wf$Q phlogistic agents I-BSA of each rat was carefully cut as a sample. The radioactivity of n was measured in a gamma well-type scintillation counter. Exudation of 533 I-BSA was calculated by subtracting radioactivity of the control skin site inJected with Tyrode's solution from that of the rat skin sites injected with phlogistic agents and was expressed in terms of percent of total radioactivity injected into each rat. Drug treatment of animals Adenosine, its analogs, bradykinin, prostaglandin El, caffeine, theophylline, and 8-phenyltheophylline were dissolved or suspended in Tyrode's solution. These agents were inJected intradermally. Cyproheptadine which was dissolved in distilled water was administered orally 1 hr before the injection of the phlogistic agents. Agents were coinjected, and the volume injected within a set of experiments was constant. StatIstical significance of difference between control and the treated group was analyzed with the Student t-test. The n values represent injections into separate animals. Results __Effects of adenoslne_and Its analogs on plasma in rat skin -___ exudation ~________ The effects of adenoslne and Its analogs on bradyklnin-induced plasma exudation are shown In Fig. 1. Adenosine and all of the analogs except the
r
1
10
Adenoslne
01
1
10
1
D-PIA
01
FIG. 1
2-Chloro Ado
01 L-PIA
Ado Cyclopropylcarboxamlde
CHA
V
0
6
Effects of adenosine (Ado)lfvd analogs on plasma exudation induced vlth bradyklnin (BK) in I-BSA was measured 15 min after the lntradermal InJection of rat skin. The leakage of bradykinin (1 ug) and/or adenosine analogs (0.1, 1 and 10 up). The solid columns represent values without bradykinin. The open bars represent values with bradyklnin (1 ug). The vertical bar on each column represents standard error of the mean. N.S. indicates not significant for effects in the presence of bradykinln (~'0.05).
0
100
600
Adenoslne
1578
Analogs
and Bradyklnln
06
Vol.
35, No
15, 1984
Experiment II Theophylllne
$3
05
I= H 2 g
04
2s wq
03
2:
02
4 pb!
01 0
001
-
1
--+
10
-+
-+
0
01
_
--+
FIG.
1
10
-+
-+
L -PIA Xanthlne
2
Antagonism by caffeine and theophylllne of plasma exudation Induced with a ?yTblnatlon of bradyklnln (1 ug) and L-PIA in rat skin The leakage of I-BSA was measured 15 mln after the lntradermal inJectlons of bradyklnln L-PIA and/or adenoslne receptor antagonists. The solid columns represent values with caffeine (100 ug, experunent I) or theophylline (100 pg, expertment II) The vertical bar on each column represents the standard error of the mean (n=6) N.S lndlcates not significant (p>O 05)
06
04NS 03-
T
p
p
-I-
T
10
100
02Oln 0
1
I4
8-Phenyltheophyllme
FIG.
3
Antagonum by 8-phenyltheophylllne of plasma exudation Inducy$5wlth bradyI-BSA was The leakage of klnln (1 ug) and L-PIA (10 up) In rat skin. measured 15 mln after the lntradermal InJectIon of bradyklnln-L-PIA and 8The vertical bar on each column represents phenyltheophylllne (O-100 ug). u-tdxates not slgnlflcant the standard error of the mean (n=7) N S (p>O.O5).
Vol.
35, No.
15. 1984
Adenosine
Analogs
and Bradyklnln
1579
carboxamlde Increased the vascular permeability slightly when tested alone. L-PIA markedly potentlated the plasma exudation Induced with bradyklnln while CHA, D-PIA, adenosine and Z-chloroadenoslne had smaller effects The cyclopropylcarboxamlde potentiated the bradyklnin-Induced plasma exudation only slightly. Effects ___ of caffeine __--
and
theoph&llne ----___on plasma
exudation
in rat
skin
Caffeine and theophylllne at doses of 1, 10 and 100 ~g had no effect on vascular permeability, nor did these methylxanthines slgniflcantly effect plasma exudation induced with bradyklnln, or with a comblnatlon of bradyklnln and prostaglandln E (data not shown). Caffeine and theophylllne (100 pg) did signifi&antly inhlbrt the plasma exudation induced with L-PIA or CHA alone (Table 1) and with a comblnatlon of bradyklnln and L-PIA (Fig. 2). Lower doses (1 and 10 ug) of caffeine had no significant effect (data not shown). 8-Phenyltheophylline, a more potent adenoslne antagonist (5) inhibited plasma exudations elicited by the L-PIA and bradykinin comblnatlon at 1, 10 and 100 ug (Fig. 3). The lnhlbltlon was not statlstlcally significant at 1 pg 8-phenyltheophylllne TABLE Effects of caffeine and theophylllne PIA or with CHA in rat skin __.____~~--Phloglstlc
agents
(100 ug)
(10 1.18) (10 ug)+caffelne
(100 ug)
Experiment 2 (n=6) L-PIA (1 ug) L-PIA (1 ug)+theophylllne(lOO~g) L-PIA L-PIA
(10 ug) (10 ug)+theophylllne(lOOug)
Experiment 3 (n=6) CHA (1 pg) CHA (1 pg)+caffeine
(100 up)
CHA CHA __.
(10 ug (10 ug)+caffeine (100 pg) .~ 125 I-BSA was measured leakage of
The oE the agents.
N.S.
on plasma
exudations
lndlcates
not
Induced
with
L-
% of control
(X o,"*?~B~~~~:~Eed) mean _ + S.E.M
Experiment 1 (n=6) L-PIA (1 ng) L-PIA (1 ug)+caffeine L-PIA L-PIA
1
0.057+0.017 0.015+0.004
100 26 (pcO.05)
0.150+0.030 0.082+0.029
100 55 (N.S.)
0.079+0.012 0.030+0.011
100 38 (p
0.16O+o.017 0 llO+O 025
100 68 (N.S.)
0.054+0.023 0 006+0.003
100 11 (p
0.131+0.028 0.0922 0.028
100 69 (N.S.)
15 min
after
significant
the
intradermal
(p>O.O5).
lnlectlon
1580
Adenosine Analogs and Bradykinin
Vol. 35, No. 15. 1984
Effect of P-sJte inhibitor 2'.5 '-dideoxyadenosine on plasma exudation in rat skin -The P-site inhibitor of adenylate cyclase, 2',5'-dideoxyadenosine, had no significant effect on plasma exudations elicited by bradykinin, L-PIA or a bradykinin-L-PIA combination (Table 2). 2',5'-Dideoxyadenosine markedly inhibited plasma exudations induced by prostaglandin E and slightly inhibited plasma exudations elicited by a combination Af bradykinin and prostaglandin El. 2',5'-Dideoxyadenosine had no effect on plasma exudation elicited by a combination of bradykinin and forskolin, the latter a general activator of adenylate cyclase (10). Table 2 Effect of 2',5'-dideoxyadenosine on plasma exudations induced with bradykinin (BK), prostaglandin E (PGE ), and combinations of bradykinin with L-PIA, prostaglandin El or forskol& (FK) in rat skin Phlogistic agents
(% ofP"??B~~?~:~~:ed) Mean + _ S.E.M.
coztzil
Experiment 1 (n=6) Bradykinin(1 ug) BK(1 ug)+2',5'-dideoxyadenosine(1 Ug) BK(~ ug)+2',5'-dideoxyadenosine(l0 ug) BK(~ ug)+2',5'-dideoxyadenosine(lO0 ug)
0.172+0.029 0.1255.047 0.175+_0.037 0.2795.032
100 71(N.S.) 103(N.S.) 172(N.S.)
Experiment 2 (n=6) BK(l ug)+L-PIA(10 ug) BK(1 ug)+L-PIA(10 ug)+2',5'-dideoxyadenosine(lUg) BK(lug)+L-PIA(lOug)+2',5'-dideoxyadenosine(lOug) BK(lug)+L-PIA(lO~g)+2',5'-dideoxyadenosine(lOO~g)
0.567kO.023 0.353TO.045 0.440+_0.103 0.462TO.124
100 62(N.S.) 77(N.S.) 81(N.S.)
Experiment 3 (n=6) Prostaglandin El (lug) PGE1(lug)+2',5'-dideoxyadenosine(lug) PGE1(lug)+2',5'-dideoxyadenosine(lOUg) PGEl(lpg)+2',5'-dideoxyadenosine(lOOug)
0.128+0.038 0.011+0.005 0.067+0.022 0.058+0.014
100 8(p
Experiment 4 (n=6) PGEl(lug)+BK(lUg) PGE (lug)+BK(lug)+2',5'-dldeoxyadenoslne(lUg) PGE'(lug)+BK(lug)+2',5'-dideoxyadenosine (lO!Jg) PGEi(lpg)+BK(lpg)+2',5'-dideoxyadenosine(lOO~g)
0.24+_0.027 0.15+o.O18 O.lS+a.O23 0.17fp.011
100 6O(p<0.025) 6O(p
Experiment 5 (n=6) BK(lug)+Forskolin(lug) BK(lug)+FK(lpg)+2',5'-dideoxyadenosine(lUg) BK(lpg)+FK(lug)+2',5'-dldeoxyadenosine(lOug) BK(lug)+FK(lug)+2',5'-dideoxyadenoslne(lOOug)
0.46+0.083 0.3850.045 0.425.054 0.46+0.046
100 82(N.S.) 92(N.S.) lOO(N.S.)
125 I-BSA was measured 15 min after the intradenal The leakage of the agents. N.S. indicates not significant (p>O.O5).
inJection of
Vol. 35, No. 15, 1Y64
1581
Adenosine Analogs and Bradykinin
Effect of the antihistamine cyproheptadine on plasma exudation in rat skin The antihistamine cyproheptadine has been reported to completely inhibit the vascular permeability response induced with histamine in rat skin (11). An oral dose of cyproheptadine of 10 mg/kg was used in the present experiments. Cyproheptadine had no effect on plasma exudation elicited by bradykinin or by a combination of bradykinin and L-PIA while markedly inhibiting plasma exudation elicited with histamine (Fig. 4).
T
I-N.S.
--I-
P <0.05
Braclykinin
Bradykinin
tlistamine
L-AA
FIG. 4 ne cyproheptadine on plasma exudation in rat I-BSA was measured 15 min after the intradermal injection of bradykinin (1 up), histamine (1 up), and a combination of bradyklnin and L-PIA (1 up). Cyproheptadine (10 mgfkg) was administered orally 1 hr before the injection of the phlogistic agents. The solid columns represent values with cyproheptadine. N.S. indicates not significant (p>O.O5). Discussion Adenosine and its analogs potentiated the plasma exudation induced with bradykinin in rat skin (Fig. 1). The potent Al-adenosine agonist L-PIA caused the greatest potentiation. Adenosine and the other analogs including D-PIA caused a smaller potentiation with adenosine 5'-cyclopropylcarboxamide, a potent A2-adenosine agonist, causing the smallest potentiation. The marked activity of L-PIA and CHA and the lesser activity of D-PIA are consonant with the involvement of an Al-adenoslne receptor in the potentiation of bradykinin-induced Plasma exudation by adenosine. However, L-PIA is only about 10 fold more potent than D-PIA and a 40-200 fold selectivity is more usually seen at Al-receptors (5). Antagonism of responses to L-PIA by the methylxanthines caffeine and theophylline (Fig. 2, Table 1) indicate the involvement of classical adenosine receptors but provides no evidence as to Its nature since these methylxanthines are not selective as adenosine receptor antagonists (12). The very potent adenosine antagonist 8-phenyltheophylline (5) was at least 10 fold more potent
1582
than
Adenoslne
theophylline
Analogs
or caffeine
Vol
and Bradyklnln
in antagonlzlng
L-PIA
responses
35, No
(Fig.
15, 1984
3)
5'-cyclopropylcarboxamlde The remarkably low efficiency of the adenosine suggests either that the adenoslne receptor has unusual characteristics, namely insensltlvlty to the carboxamlde or more likely that the results are not explicable in terms of only one adenoslne receptor. Adenoslne 5'-cyclopropylcarboxamlde is very potent at stlmulatory A2-receptors coupled to adenylate cyclase, It is thus possible but also has high activity at inhibitory Al-receptors (5) that at high doses marked activation of the stimulatory A -receptors by the receptor-mediated poten $ iation of bradykinincarboxamide begins to blunt an of adenylate cyclase In ,$,vlous studies , activation elicited plasma exudation by forskolln was shown to block adenoslne and prostaglandln-ellcited plasma Adenoslne and other analogs are at least lo-fold less potent at exudation (7). A2-receptors than the carboxamlde and hence might not, even at their highest the stimulatory A -receptor to an extent sufficient to blunt an dose, activate 2 Al-response. The present results emphasize the need to develop selective antagonists Caffeine, theophylllne and most other for Al- and A -adenosine receptors. xanthlnes inc 1 udlng 8-phenyltheophylllne are relatively nonselective as even the effects of caffeine and theophylline However, antagonists (12). on plasma exudation are not readily explained merely by blockade of one The methylxanthlnes had no effect on prostaclass of adenosine receptors. glandln-elicited or bradykinin-elicited plasma exudations, but did inhibit plasma exudation elicited by L-PIA and CHA What is not readily explained is why caffeine reduces the L-PIA-potentiated bradykinin response to a level much below that seen with bradyklnin alone (Fig. 2) It would appear that a very low dose of L-PIA is actually inhibitory to bradykinin-induced plasma exudation in the presence of a methylxanthlne. Another approach to the mechanism of adenosine effects was to investigate the effects of a general inhibitor of adenylate cyclase, namely 2',5'dideoxyadenoslne. This analog has virtually no activity at adenoslne receptors, but through interaction with a so-called P-site inhibits virtually all adenylate cyclases (5). It was felt that 2',5'-dldeoxyadenosine should blunt the effect of agents which potentlate plasma exudation through activation of adenylate cyclases Indeed, while having no effect alone, 2',5'dldeoxyadenoslne did inhibit plasma exudation ellclted by prostaglandin El and partially reversed, the potentlatlon of bradyklnin responses by prostaProstaglandin El does activate adenylate cyclase in glandin El (Table 2). readily reverses prostaglandln many cell types, and 2' ,5'-dldeoxyadenosine El-elicited activation of cyclic AMP-generating systems in platelets (13) However, 2',5'-dldeoxyadenoslne did not significantly affect plasma exudation elicited by bradyklnln, L-PIA or comblnatlon of bradyklnin and L-PIA 2',5'-Dldeoxyadenosine also did not reverse the potentiatlon of the bradykinin-response by a general activator of adenylate cyclase, forskolin. forskolln-activated cyclic Although at first this would seem anomalous, AMP-generating systems in platelets were found to be relatively refractory to inhibition by 2',5'-dldeoxyadenosine (13,14). The present results suggest that the potentiatory effects of adenoslne and its analogs on the bradykinin-induced plasma exudation is induced through an inhibitory A -adenoslne receptor in rat skin. The possibility that an A21
Vol. 35. No. 15, 1984
Adenosine Analogs and Bradyklnin
1583
receptor is involved cannot, however, be excluded. Indeed, that an inhibitory Al-adenosine type is involved in potentiation of bradykinin responses might 5 prlori be thought unlikely since forskolln, its analogs, phosphodiesterase inhibitors and 8-bromo-cyclic AMP, all of which activate cyclic AMP-systems, potentiated bradykinin-induced plasma exudation in rat skin (16). Such results Indicate that the Increase in cyclic AMP level in some compartment in rat skin can potentiate the bradykinin-induced plasma exudation. Prostaglandin E also potentiates bradykinin-elicited plasma exudation and is well known to ac &ivate many adenylate cyclases (1). In the present study an adenylate cyclase inhibitory, 2',5'-dideoxyadenosine, was shown to blunt the prostaglandin-response in rat skin. The precise mechanism of the potentiatory effects of adenosine and its analogs on bradykinin-induced plasma exudation clearly remains to be elucidated. Bradykinin has been proposed to increase the vascular permeability through the formation of intracellular gaps betwen endothelial cells in rat skin (15). Therefore, the present results suggest the existence of adenosine receptors in vascular endothellal cells in rat skin, which in some manner potentlate bradykinin effects on these cells. References ~--1.
2. 3. 4. 5 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
T.J. WILLIAMS and M.J. PECK, Nature 270 530-532 (1977) B.B. FREDHOLM, F. LANEFELT, P. KJEMDAHL, B. SIDHAGEN and A. SYDBOM, Progress in Pharmacol. 5 109 (1980). M.K. DUNLAP, R J., KAPLAR and E.W. ROSENBERG, Clin. Res. -30 5828 (1982). A. SYDBOM and B.B. FREDHOLM, Acta Physiol. Stand. 114 243-251 (1982) J.W. DALY, R.F. BRUNS and S.H. SNYDER, Life Scl. mO83-2097 (1981). K. SUGIO, K. OHUCHI, M. SUGATA and S. TSURUFUJI, Prostaglandins 21 649-653 (1981). K. SUGIO and J.W. DALY, Life Scl 22 65-73 (1983). K. SUGIO and S. TSURUFUJI, Br. J Pharmacol. 73 605-608 (1981). S. TSURUFUJI, K. SUGIO and Y. ENDO, Biochem. Pharmacol. 6 1131-1136 (1977). K.B. SEAMON AND J.W. DALY, J. Cyclic Nucleotlde Res. 7 201-224 (1981). I L. BONTA and C.J. DEVOS, Acta Endocrinol. 49 403-411 (1965). J.W. DALY, P. BUTTS-IAMB and W. PADGETT, Cell. Mol. Neurobiol. 3 69-80 (1983). K.C AGARWAL and R.E. PARKS, Jr., Biochem. Pharmacol. -31 3713-3716 (1982). A.M. SIEGL and J.W. DALY, Pharmacologist 25 140 (1983). G. MAJNO, G.E. PALADE and G.I. SCHOEFLE, J. Biophys. Biochem. Cyto. -11 607-626 (1961). K. SUGIO and J.W. DALY, Life Sci. 24 123-132 (1984).
Life Sciences, Vol. 35, pp. 1575-1583 Printed in the U.S.A.
ADENOSINE ANALOGS: POTENTIATION OF BRADYKININ-INDUCED PLASMA EXUDATION IN RAT SKIN AND PREVENTION BY CAFFEINE AND THEOPHYLLINE Kazuo Sugio and John W. Daly Laboratory of Bioorganic Chemistry, Bldg. 4, Room 212, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20205, U.S.A. (Received in final form July 24, 1984) Summary Adenoslne and various analogslyytentiated plasma exudation elicited by I-labelled bovine serum albumin b a&ykinin In rat skin using (I I-BSA) as a tracer. L-N6-Phenylrsopropybadenoslne (L-PIA) was much more effective than D-PIA, adenoslne, N -cyclohexyladenoslne (CHA) and 2-chloroadenosine, all of which were comparable in activity. Adenoslne 5'-cyclopropylcarboxamide was the least effective analog. Caffeine and theophylllne had no effect on basal or bradykinln-ellclted plasma exudation, while inhibiting plasma exudatron eliclted by L-PIA, CHA or a combination of bradykinin and L-PIA. 8-Phenyltheophylline was more potent than caffeine or theophylllne versus the bradykinin and L-PIA combrnation. 2',5'-Dideoxyadenoslne, a P-site inhibitor of adenylate cyclase, had no effect on plasma exudation elicited by bradyklnin, L-PIA or a combination of bradykinln and L-PIA, but did inhibit plasma exudation eliclted by prostaglandin E lh(PGEa) Or a bradykinin-PGEl-combination. The antihistamine cypro epta ine slightly reduced plasma exudation elicited by a bradykinin-PGEl combination. The results suggest that adenosine potentiates bradykinin-induced plasma exudation via an adenosine receptor and that histamine may be involved to some extent in the phenomenon. Adenosine, a potent vasodilator, increases vascular permeability and potentiates the plasma exudation induced with bradykinin in rat skin (1) and enhances histamine release from mast cells (2). Furthermore adenosine receptor antagonlsts such as caffeine inhibit the inflammatory response induced with oxazolone in rat ear (3) and antagonize histamine release by mast cells (4). Such results suggest that adenoslne may play a significant role in the increase in the vascular permeability which is associated with Two classes of adenosine receptors have been proposed: An inflammation A2-receptor stimulatory to adenylate cyclase and and Al-receptor lnhlbitory to adenylate cyclase (5). Both classes of receptors are antagonized by methylxanthines such as caffeine and theophylline. The nature of the adenosine receptor which 1s involved in potentiatlon of bradykinininduced plasma exudation has been Investigated using a variety of adenosine analogs and adenosine receptor antagonlsts such as caffeine, theophylllne, Possible involvement of histamine was investiand 8-phenyltheophylline gated with an antihistamine, cyproheptadine.
0024-3205184 $3.00 + .OO
1576
Adenosine Analogs and Bradykinin
Vol. 35, No. 15, 1984
Materials and Methods Animals Male rats of Sprague-Dawley strain (body weight, 160-180 g) were used throughout the experiments. All rats were maintained on laboratory food and tap water ad libitum. Materials 125 The sourcea23f various agents were as follows: [ I]-labelled bovine serum albumin ( I-BSA, specific activity 60 uCi/mmol, New England Nuclear co., Boston, MA), caffeine, theophylline, 8-phenyltheophylline, adenosine and 2-chloroadenosine were from Calbiochem-Behring Co. (La Jolla, CA), LPIA and D-PIA were from Boehringer Co. (Sohn, IngelheimjRhein). Bradykinin and cyproheptadine were from Sigma Chemicals Inc. (St. Louis, MO). 2',5'Dideoxyadenosine was from ICN Chemicals and Radioisotope Division (Irvine, Adenosine 5'-cyclopropylcarboxamide was from Abbott Laboratories W. (Chicago, IL). Induction and measurement of vascular permeability response in rat skin Vascular permeability response was induced on shaved dorsal skin of the rats by the intradermal injection wit P23hlogistic agents (6,7). I-BSA as a tracer. PurifiVascular permeability waslysasured using cation of the commercial I-BSA was af53 rding to the method of Sugio et I-BSA In 0.2 ml of 0.9% NaCl al (8,9). About 1 uCi of the purified solution containing Evans Blue in 1.0% was injected into the tail vein of the rats. Immediately after the injection of tracer, phlogistic agents (0.1 ml) were inJected intradermally on the back under light ether anesthesia. Rats were killed by cutting the carotid arteries and the dorsal skin was peeled off and separated from adherent subcutaneous tissues. The blue area (20 mm diameter) of the rat skin at each site of inlection wf$Q phlogistic agents I-BSA of each rat was carefully cut as a sample. The radioactivity of n was measured in a gamma well-type scintillation counter. Exudation of 533 I-BSA was calculated by subtracting radioactivity of the control skin site inJected with Tyrode's solution from that of the rat skin sites injected with phlogistic agents and was expressed in terms of percent of total radioactivity injected into each rat. Drug treatment of animals Adenosine, its analogs, bradykinin, prostaglandin El, caffeine, theophylline, and 8-phenyltheophylline were dissolved or suspended in Tyrode's solution. These agents were inJected intradermally. Cyproheptadine which was dissolved in distilled water was administered orally 1 hr before the injection of the phlogistic agents. Agents were coinjected, and the volume injected within a set of experiments was constant. StatIstical significance of difference between control and the treated group was analyzed with the Student t-test. The n values represent injections into separate animals. Results __Effects of adenoslne_and Its analogs on plasma in rat skin -___ exudation ~________ The effects of adenoslne and Its analogs on bradyklnin-induced plasma exudation are shown In Fig. 1. Adenosine and all of the analogs except the
r
1
10
Adenoslne
01
1
10
1
D-PIA
01
FIG. 1
2-Chloro Ado
01 L-PIA
Ado Cyclopropylcarboxamlde
CHA
V
0
6
Effects of adenosine (Ado)lfvd analogs on plasma exudation induced vlth bradyklnin (BK) in I-BSA was measured 15 min after the lntradermal InJection of rat skin. The leakage of bradykinin (1 ug) and/or adenosine analogs (0.1, 1 and 10 up). The solid columns represent values without bradykinin. The open bars represent values with bradyklnin (1 ug). The vertical bar on each column represents standard error of the mean. N.S. indicates not significant for effects in the presence of bradykinln (~'0.05).
0
100
600
Adenoslne
1578
Analogs
and Bradyklnln
06
Vol.
35, No
15, 1984
Experiment II Theophylllne
$3
05
I= H 2 g
04
2s wq
03
2:
02
4 pb!
01 0
001
-
1
--+
10
-+
-+
0
01
_
--+
FIG.
1
10
-+
-+
L -PIA Xanthlne
2
Antagonism by caffeine and theophylllne of plasma exudation Induced with a ?yTblnatlon of bradyklnln (1 ug) and L-PIA in rat skin The leakage of I-BSA was measured 15 mln after the lntradermal inJectlons of bradyklnln L-PIA and/or adenoslne receptor antagonists. The solid columns represent values with caffeine (100 ug, experunent I) or theophylline (100 pg, expertment II) The vertical bar on each column represents the standard error of the mean (n=6) N.S lndlcates not significant (p>O 05)
06
04NS 03-
T
p
p
-I-
T
10
100
02Oln 0
1
I4
8-Phenyltheophyllme
FIG.
3
Antagonum by 8-phenyltheophylllne of plasma exudation Inducy$5wlth bradyI-BSA was The leakage of klnln (1 ug) and L-PIA (10 up) In rat skin. measured 15 mln after the lntradermal InJectIon of bradyklnln-L-PIA and 8The vertical bar on each column represents phenyltheophylllne (O-100 ug). u-tdxates not slgnlflcant the standard error of the mean (n=7) N S (p>O.O5).
Vol.
35, No.
15. 1984
Adenosine
Analogs
and Bradyklnln
1579
carboxamlde Increased the vascular permeability slightly when tested alone. L-PIA markedly potentlated the plasma exudation Induced with bradyklnln while CHA, D-PIA, adenosine and Z-chloroadenoslne had smaller effects The cyclopropylcarboxamlde potentiated the bradyklnin-Induced plasma exudation only slightly. Effects ___ of caffeine __--
and
theoph&llne ----___on plasma
exudation
in rat
skin
Caffeine and theophylllne at doses of 1, 10 and 100 ~g had no effect on vascular permeability, nor did these methylxanthines slgniflcantly effect plasma exudation induced with bradyklnln, or with a comblnatlon of bradyklnln and prostaglandln E (data not shown). Caffeine and theophylllne (100 pg) did signifi&antly inhlbrt the plasma exudation induced with L-PIA or CHA alone (Table 1) and with a comblnatlon of bradyklnln and L-PIA (Fig. 2). Lower doses (1 and 10 ug) of caffeine had no significant effect (data not shown). 8-Phenyltheophylline, a more potent adenoslne antagonist (5) inhibited plasma exudations elicited by the L-PIA and bradykinin comblnatlon at 1, 10 and 100 ug (Fig. 3). The lnhlbltlon was not statlstlcally significant at 1 pg 8-phenyltheophylllne TABLE Effects of caffeine and theophylllne PIA or with CHA in rat skin __.____~~--Phloglstlc
agents
(100 ug)
(10 1.18) (10 ug)+caffelne
(100 ug)
Experiment 2 (n=6) L-PIA (1 ug) L-PIA (1 ug)+theophylllne(lOO~g) L-PIA L-PIA
(10 ug) (10 ug)+theophylllne(lOOug)
Experiment 3 (n=6) CHA (1 pg) CHA (1 pg)+caffeine
(100 up)
CHA CHA __.
(10 ug (10 ug)+caffeine (100 pg) .~ 125 I-BSA was measured leakage of
The oE the agents.
N.S.
on plasma
exudations
lndlcates
not
Induced
with
L-
% of control
(X o,"*?~B~~~~:~Eed) mean _ + S.E.M
Experiment 1 (n=6) L-PIA (1 ng) L-PIA (1 ug)+caffeine L-PIA L-PIA
1
0.057+0.017 0.015+0.004
100 26 (pcO.05)
0.150+0.030 0.082+0.029
100 55 (N.S.)
0.079+0.012 0.030+0.011
100 38 (p
0.16O+o.017 0 llO+O 025
100 68 (N.S.)
0.054+0.023 0 006+0.003
100 11 (p
0.131+0.028 0.0922 0.028
100 69 (N.S.)
15 min
after
significant
the
intradermal
(p>O.O5).
lnlectlon
1580
Adenosine Analogs and Bradykinin
Vol. 35, No. 15. 1984
Effect of P-sJte inhibitor 2'.5 '-dideoxyadenosine on plasma exudation in rat skin -The P-site inhibitor of adenylate cyclase, 2',5'-dideoxyadenosine, had no significant effect on plasma exudations elicited by bradykinin, L-PIA or a bradykinin-L-PIA combination (Table 2). 2',5'-Dideoxyadenosine markedly inhibited plasma exudations induced by prostaglandin E and slightly inhibited plasma exudations elicited by a combination Af bradykinin and prostaglandin El. 2',5'-Dideoxyadenosine had no effect on plasma exudation elicited by a combination of bradykinin and forskolin, the latter a general activator of adenylate cyclase (10). Table 2 Effect of 2',5'-dideoxyadenosine on plasma exudations induced with bradykinin (BK), prostaglandin E (PGE ), and combinations of bradykinin with L-PIA, prostaglandin El or forskol& (FK) in rat skin Phlogistic agents
(% ofP"??B~~?~:~~:ed) Mean + _ S.E.M.
coztzil
Experiment 1 (n=6) Bradykinin(1 ug) BK(1 ug)+2',5'-dideoxyadenosine(1 Ug) BK(~ ug)+2',5'-dideoxyadenosine(l0 ug) BK(~ ug)+2',5'-dideoxyadenosine(lO0 ug)
0.172+0.029 0.1255.047 0.175+_0.037 0.2795.032
100 71(N.S.) 103(N.S.) 172(N.S.)
Experiment 2 (n=6) BK(l ug)+L-PIA(10 ug) BK(1 ug)+L-PIA(10 ug)+2',5'-dideoxyadenosine(lUg) BK(lug)+L-PIA(lOug)+2',5'-dideoxyadenosine(lOug) BK(lug)+L-PIA(lO~g)+2',5'-dideoxyadenosine(lOO~g)
0.567kO.023 0.353TO.045 0.440+_0.103 0.462TO.124
100 62(N.S.) 77(N.S.) 81(N.S.)
Experiment 3 (n=6) Prostaglandin El (lug) PGE1(lug)+2',5'-dideoxyadenosine(lug) PGE1(lug)+2',5'-dideoxyadenosine(lOUg) PGEl(lpg)+2',5'-dideoxyadenosine(lOOug)
0.128+0.038 0.011+0.005 0.067+0.022 0.058+0.014
100 8(p
Experiment 4 (n=6) PGEl(lug)+BK(lUg) PGE (lug)+BK(lug)+2',5'-dldeoxyadenoslne(lUg) PGE'(lug)+BK(lug)+2',5'-dideoxyadenosine (lO!Jg) PGEi(lpg)+BK(lpg)+2',5'-dideoxyadenosine(lOO~g)
0.24+_0.027 0.15+o.O18 O.lS+a.O23 0.17fp.011
100 6O(p<0.025) 6O(p
Experiment 5 (n=6) BK(lug)+Forskolin(lug) BK(lug)+FK(lpg)+2',5'-dideoxyadenosine(lUg) BK(lpg)+FK(lug)+2',5'-dldeoxyadenosine(lOug) BK(lug)+FK(lug)+2',5'-dideoxyadenoslne(lOOug)
0.46+0.083 0.3850.045 0.425.054 0.46+0.046
100 82(N.S.) 92(N.S.) lOO(N.S.)
125 I-BSA was measured 15 min after the intradenal The leakage of the agents. N.S. indicates not significant (p>O.O5).
inJection of
Vol. 35, No. 15, 1Y64
1581
Adenosine Analogs and Bradykinin
Effect of the antihistamine cyproheptadine on plasma exudation in rat skin The antihistamine cyproheptadine has been reported to completely inhibit the vascular permeability response induced with histamine in rat skin (11). An oral dose of cyproheptadine of 10 mg/kg was used in the present experiments. Cyproheptadine had no effect on plasma exudation elicited by bradykinin or by a combination of bradykinin and L-PIA while markedly inhibiting plasma exudation elicited with histamine (Fig. 4).
T
I-N.S.
--I-
P <0.05
Braclykinin
Bradykinin
tlistamine
L-AA
FIG. 4 ne cyproheptadine on plasma exudation in rat I-BSA was measured 15 min after the intradermal injection of bradykinin (1 up), histamine (1 up), and a combination of bradyklnin and L-PIA (1 up). Cyproheptadine (10 mgfkg) was administered orally 1 hr before the injection of the phlogistic agents. The solid columns represent values with cyproheptadine. N.S. indicates not significant (p>O.O5). Discussion Adenosine and its analogs potentiated the plasma exudation induced with bradykinin in rat skin (Fig. 1). The potent Al-adenosine agonist L-PIA caused the greatest potentiation. Adenosine and the other analogs including D-PIA caused a smaller potentiation with adenosine 5'-cyclopropylcarboxamide, a potent A2-adenosine agonist, causing the smallest potentiation. The marked activity of L-PIA and CHA and the lesser activity of D-PIA are consonant with the involvement of an Al-adenoslne receptor in the potentiation of bradykinin-induced Plasma exudation by adenosine. However, L-PIA is only about 10 fold more potent than D-PIA and a 40-200 fold selectivity is more usually seen at Al-receptors (5). Antagonism of responses to L-PIA by the methylxanthines caffeine and theophylline (Fig. 2, Table 1) indicate the involvement of classical adenosine receptors but provides no evidence as to Its nature since these methylxanthines are not selective as adenosine receptor antagonists (12). The very potent adenosine antagonist 8-phenyltheophylline (5) was at least 10 fold more potent
1582
than
Adenoslne
theophylline
Analogs
or caffeine
Vol
and Bradyklnln
in antagonlzlng
L-PIA
responses
35, No
(Fig.
15, 1984
3)
5'-cyclopropylcarboxamlde The remarkably low efficiency of the adenosine suggests either that the adenoslne receptor has unusual characteristics, namely insensltlvlty to the carboxamlde or more likely that the results are not explicable in terms of only one adenoslne receptor. Adenoslne 5'-cyclopropylcarboxamlde is very potent at stlmulatory A2-receptors coupled to adenylate cyclase, It is thus possible but also has high activity at inhibitory Al-receptors (5) that at high doses marked activation of the stimulatory A -receptors by the receptor-mediated poten $ iation of bradykinincarboxamide begins to blunt an of adenylate cyclase In ,$,vlous studies , activation elicited plasma exudation by forskolln was shown to block adenoslne and prostaglandln-ellcited plasma Adenoslne and other analogs are at least lo-fold less potent at exudation (7). A2-receptors than the carboxamlde and hence might not, even at their highest the stimulatory A -receptor to an extent sufficient to blunt an dose, activate 2 Al-response. The present results emphasize the need to develop selective antagonists Caffeine, theophylllne and most other for Al- and A -adenosine receptors. xanthlnes inc 1 udlng 8-phenyltheophylllne are relatively nonselective as even the effects of caffeine and theophylline However, antagonists (12). on plasma exudation are not readily explained merely by blockade of one The methylxanthlnes had no effect on prostaclass of adenosine receptors. glandln-elicited or bradykinin-elicited plasma exudations, but did inhibit plasma exudation elicited by L-PIA and CHA What is not readily explained is why caffeine reduces the L-PIA-potentiated bradykinin response to a level much below that seen with bradyklnin alone (Fig. 2) It would appear that a very low dose of L-PIA is actually inhibitory to bradykinin-induced plasma exudation in the presence of a methylxanthlne. Another approach to the mechanism of adenosine effects was to investigate the effects of a general inhibitor of adenylate cyclase, namely 2',5'dideoxyadenoslne. This analog has virtually no activity at adenoslne receptors, but through interaction with a so-called P-site inhibits virtually all adenylate cyclases (5). It was felt that 2',5'-dldeoxyadenosine should blunt the effect of agents which potentlate plasma exudation through activation of adenylate cyclases Indeed, while having no effect alone, 2',5'dldeoxyadenoslne did inhibit plasma exudation ellclted by prostaglandin El and partially reversed, the potentlatlon of bradyklnin responses by prostaProstaglandin El does activate adenylate cyclase in glandin El (Table 2). readily reverses prostaglandln many cell types, and 2' ,5'-dldeoxyadenosine El-elicited activation of cyclic AMP-generating systems in platelets (13) However, 2',5'-dldeoxyadenoslne did not significantly affect plasma exudation elicited by bradyklnln, L-PIA or comblnatlon of bradyklnin and L-PIA 2',5'-Dldeoxyadenosine also did not reverse the potentiatlon of the bradykinin-response by a general activator of adenylate cyclase, forskolin. forskolln-activated cyclic Although at first this would seem anomalous, AMP-generating systems in platelets were found to be relatively refractory to inhibition by 2',5'-dldeoxyadenosine (13,14). The present results suggest that the potentiatory effects of adenoslne and its analogs on the bradykinin-induced plasma exudation is induced through an inhibitory A -adenoslne receptor in rat skin. The possibility that an A21
Vol. 35. No. 15, 1984
Adenosine Analogs and Bradyklnin
1583
receptor is involved cannot, however, be excluded. Indeed, that an inhibitory Al-adenosine type is involved in potentiation of bradykinin responses might 5 prlori be thought unlikely since forskolln, its analogs, phosphodiesterase inhibitors and 8-bromo-cyclic AMP, all of which activate cyclic AMP-systems, potentiated bradykinin-induced plasma exudation in rat skin (16). Such results Indicate that the Increase in cyclic AMP level in some compartment in rat skin can potentiate the bradykinin-induced plasma exudation. Prostaglandin E also potentiates bradykinin-elicited plasma exudation and is well known to ac &ivate many adenylate cyclases (1). In the present study an adenylate cyclase inhibitory, 2',5'-dideoxyadenosine, was shown to blunt the prostaglandin-response in rat skin. The precise mechanism of the potentiatory effects of adenosine and its analogs on bradykinin-induced plasma exudation clearly remains to be elucidated. Bradykinin has been proposed to increase the vascular permeability through the formation of intracellular gaps betwen endothelial cells in rat skin (15). Therefore, the present results suggest the existence of adenosine receptors in vascular endothellal cells in rat skin, which in some manner potentlate bradykinin effects on these cells. References ~--1.
2. 3. 4. 5 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
T.J. WILLIAMS and M.J. PECK, Nature 270 530-532 (1977) B.B. FREDHOLM, F. LANEFELT, P. KJEMDAHL, B. SIDHAGEN and A. SYDBOM, Progress in Pharmacol. 5 109 (1980). M.K. DUNLAP, R J., KAPLAR and E.W. ROSENBERG, Clin. Res. -30 5828 (1982). A. SYDBOM and B.B. FREDHOLM, Acta Physiol. Stand. 114 243-251 (1982) J.W. DALY, R.F. BRUNS and S.H. SNYDER, Life Scl. mO83-2097 (1981). K. SUGIO, K. OHUCHI, M. SUGATA and S. TSURUFUJI, Prostaglandins 21 649-653 (1981). K. SUGIO and J.W. DALY, Life Scl 22 65-73 (1983). K. SUGIO and S. TSURUFUJI, Br. J Pharmacol. 73 605-608 (1981). S. TSURUFUJI, K. SUGIO and Y. ENDO, Biochem. Pharmacol. 6 1131-1136 (1977). K.B. SEAMON AND J.W. DALY, J. Cyclic Nucleotlde Res. 7 201-224 (1981). I L. BONTA and C.J. DEVOS, Acta Endocrinol. 49 403-411 (1965). J.W. DALY, P. BUTTS-IAMB and W. PADGETT, Cell. Mol. Neurobiol. 3 69-80 (1983). K.C AGARWAL and R.E. PARKS, Jr., Biochem. Pharmacol. -31 3713-3716 (1982). A.M. SIEGL and J.W. DALY, Pharmacologist 25 140 (1983). G. MAJNO, G.E. PALADE and G.I. SCHOEFLE, J. Biophys. Biochem. Cyto. -11 607-626 (1961). K. SUGIO and J.W. DALY, Life Sci. 24 123-132 (1984).