- Email: [email protected]
A novel site of action of a high affinity A1 adenosine receptor antagonist
Vo1.153, No. 3,1988
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
June 30,1988
Pages 939-944
A NOVELSITE OF ACTION OF A HIGH AFFINITY AI ADENOSINERECEPTORANTAGONIST Vickram Ramkumar and Gary L. Stiles Departments of Medicine (Cardiology) and Biochemistry Duke University Medical Center Durham, NC 27710 Received March 28, 1988
XAC, a high a f f i n i t y antagonist of the AI adenosine receptor, enhances adenylate cyclase a c t i v i t y by 1.3-2 fold with~an ECBo of ~ 47 nM in adipocyte membranes pretreated with adenosine deaminase to eliminate adenosine and in the presence of total phosphodiesterase inhibition by 100 ~M papaverine. This effect of XAC is observed only at concentrations of GTP sufficient to activate Gi (~ 5 x 10-6 M GTP) and is not evident in the absence or presence of lower GTP concentrations. ADP ribosy!ation of G~ by pertussis toxin treatment also abolishes this stimulatory action of XAC. Furthermore, in the presence of GTP activation of inhibitory prostaglandin EI receptors diminishes the stimulatory effect of XAC on adenylate cyclase. In ~ddition, XAC interferes with GTP-mediated inhibition of forskolin-stimulated adenylate cyclase a c t i v i t y in a noncompetitive manner. Finally, XAC is only a weak inhibitor of the low K~ cyclic AMP phosphodiesterase, producing ~ 40% inhibition of phosphodiester~'se a c t i v i t y at a concentration of 100 ~M. These data suggest that XAC increases adenylate cyclase a c t i v i t y in absence of endogenous adenosine by inhibiting tonic Gi a c t i v i t y in a reversible manner. ©198BAcademicP~es~,In~.
XAC, 8-[4-[[[[(2-amino-ethyl)-amino]carbony!]methyl]oxy]phenyl]-l,3dipropylxanthine, is a relatively selective and high a f f i n i t y antagonist of the AI adenosine receptor (1).
This property apparently underlies i t s a b i l i t y to
block A1 adenosine receptor agonists mediated bradycardia (2) and their inhibition of adenylate cyclase. A1 adenosine receptors inhibit adenylate cyclase a c t i v i t y via the inhibitory guanine nucleotide regulatory protein (Gi) at GTP concentrations in the micromolar range (3). Agonists acting at AI receptors activate Gi leading to the exchange of bound GDP for GTP. This step represents the rate limiting step in the activation process. Activation leads to the dissociation of the oligomer into ai-GTP and By subunits.
I t appears that inhibition of adenylate cyclase
results from the binding of free as subunits by By, preventing their interaction with the catalytic unit. Termination of the activation process depends on GTPase a c t i v i t y intrinsic to ai (4). In preliminary studies in rat adipocyte membranes, we observed that XAC not only attenuates R-PIA mediated inhibition of adenylate cyclase as would be
939
0006-291X/88 $1.50 Copyright © 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 153, No. 3, 1 9 8 8
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
expected of an AI receptor antagonist but by i t s e l f actually stimulates the a c t i v i t y of this enzyme. This is a property not normally associated with adenosine receptor antagonists.
In this report we have characterized the
biochemical mechanisms responsible for this a c t i v i t y and demonstrate XAC's interaction at a site distinct from the A1 receptor, namely the inhibitory guanine nucleotide regulatory protein (Gi).
METHODS Pertussis vaccine, which contains the toxin (5) was administered intraperitoneally (~ 300 OPU/Kg; 0.3-0.5 ml) to rats (male Sprague Dawley, 250-300 g, Charles River Breeding Laboratories) three days prior to sacrifice. Control and pertussis-intoxicated rats were sacrificed by decapitation and adipocyte membranes were prepared from epididymal fat pads as described previously (6). Membranes were pretreated with adenosine deaminase before performing all assays. Adenylate cyclase a c t i v i t y in adipocyte membranes was measured as described previously (5). Briefly, 20 ul of membranes (~ 50 ~g protein) suspended in 75 mM Tris HCl buffer (pH 7.4 at 30°C), containing 12.5 mMMgCl~, 200 mM NaCl, 2.5 mM DTT, and 8 units/ml of adenosine deaminase were incubated with 20 ~l of Lomix (0.14 mM dATP, 5 mMphosphocreat!ne, 1 ~M cAMP, 30 units/ml creatine phosphokinase, ~ 1.5 ~Ci of [32P]ATP and 10 ~l of H20 or drugs. Papaverine (10-4 M) was used in all experiments to provide adequate inhibition of the low K~ cyclic AMP phosphodiesterase. Experiments were performed in the absence o~ GTP, or in the presence of 5 x 10-6 M GTP or at varying GTP concentrations (see Results). Cyclic AMP assays were incubated for 15 min at 30°C and terminated by the addition of 1 ml ice-cold stop solution containing ~ !5,000 cpm [3HI cAMP, 0.3 mM cAMPand 0.4 mMATP. Cyclic AMP was isolated by the method of Salomon et
a!. (7). Assay for the low K~ cyclic AMP phosphodiesterase was performed according to the method described byI~e Mazancourt and Guidicelli for rat adipocytes (8). [3HI cAMPlevels were measured by pouring the supernatant over Dowex 50 W-X4 columns (200-400 mesh). The f i r s t eluate and subsequent 1.5 ml wash were discarded. The columns were then eluted twice with 3 ml of H20 each and the eluates were collected and counted. Values obtained from tubes boiled without rior incubation of the reaction mixture were used to establish the maximal 100%) substrate level.
~
RESULTS XAC enhances adenylate cyclase a c t i v i t y in a dose-dependent manner with ECso for stimulation being 47 ± 11 nM (mean ± S.E.M.) and maximal a c t i v i t y being 148 ± 7% (range from 135-200% of control) (N = 9). The a b i l i t y of XAC to increase adenylate cyclase a c t i v i t y appears to be dependent on concentrations of GTP necessary to f a c i l i t a t e the a c t i v i t y of Gi (5 x 10-6 M) and is not observed in the absence of GTP (Figure 1) or at concentrations of GTP sufficient to activate Gs but not Gi (Figure 3). This indicates the essential role of an active Gi to promote stimulatory a c t i v i t y of XAC. This notion is supported by the finding that inactivation of Gi by pertussis toxin attenuates XAC enhancement of adenylate cyclase a c t i v i t y (Figure 1). Interestingly, adenylate cyclase a c t i v i t y 940
Vol. 153, No. 3, 1988
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
160 co
/•Control
o
140 0
< ¢0
o
r"o
<
"~" I
0
Figure ~.
//
"No
GTP
I
I
I
I
8
7
6
5
-Log [XAC] (M) Effect of XAC on basal adenylate cyclase activity.
Adenylate cyclase assayswere performed in control membranesin the absence (no GTP) and presence (control) of 5 x 10-6 M GTP or in pertussis-intoxicated membranes in presenceof GTP (pertussis toxin) as described in Methods. Basal adenylate cyclase activity in control, no GTP and pertussis toxin groups averaged 41.6, 77.4 and 75.0 pmol/min/mg protein. This is a representative of 3-9 experiments performed with similar results.
obtained in the absence and presence of 5 x 10-6 M GTP in control membranes and in pertussis toxin-intoxicated membranes (GTP at 5 x 10-6 M) were remarkably different, being 77.4, 41.6, and 75.0 pmol/mg/min, respectively (mean of 3-9 experiments).
This suggests that under basal conditions Gi is tonically active
in rat adipocyte membranes. We have previously demonstrated that XAC is a potent A1 adenosine receptor antagonist with Kd values in rat adipocyte ranging between I-2 nM (Ramkumar and Stiles, submitted).
In order to rule out A1 adenosine receptor blockade of
endogenous adenosine as the reason for the increase in adenylate cyclase a c t i v i t y observed with XAC, all experiments were performed by preincubating the membranes with about 8 units/ml of adenosine deaminase to degrade endogenous adenosine. This pretreatment was adequate since no change in a c t i v i t i e s of adenylate cyclase in the absence and presence of XAC were obtained by increasing the concentration of adenosine deaminase by up to 3-fold (data not shown). Furthermore, the increase in adenylate cyclase a c t i v i t y was not a consequence of inhibition of the low Km cyclic AMP phosphodiesterase by XAC. Significant inhibition of the l a t t e r enzyme was obtained only at concentrations of XAC close to the millimolar range. Inhibition by 100 uM XAC was ~ 40% of that observed with 100 uM papaverine (not shown). This concentration of papaverine was used in a l l experiments.
The
addition of 100 vM XAC to 100 uM papaverine did not improve inhibition of the 941
Voi. 153, No. 3, 1 9 8 8
160
:_~
Prostaglandin
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
//
E 1 (M)
.o
"~
•
10 "9
•
1 0 "8
100 <0 80
_
,
/
~
° 8 g 12o e(D
"a <
®
60 >, e'10
100
,< I#
I
I
I
I
I
0
10
9
8
7
6
-Log Concentration XAC (M)
[] 10"6
40 ~jl
0
8
Q
I
I
I
7
6
5
-Log [GTP] (M)
Figure_2. Prostaglandin E1 mediated inhibition of the stimulatory effect of XAC on adenylate cyclase. Membranes were incubated with an assay mixture as described in Methods containing 5 x 10-6 M GTP and different concentrations of XAC and prostaglandin E1. Each point is the average of duplicate determinations from a single experiment. Experiments were repeated twice with similar results. Figure 3.
GTP inhibition of forskolin-stimulated adenylate cyclase activity in absence and presence of XAC.
GTP inhibition curves were performed in absence and presence of 10 and 1000 nM of XAC as described in the Methods. Asterisks (*) denote statistically significant difference from control (no XAC). Eachpoint is the average of three experiments performed in duplicate. Forskolin-stimulated (5 uM forskolin) adenylate cyclase activity in the absence of GTP and XAC averaged 174 pmol/min/mg protein and did not change significantly with addition of XAC.
phosphodiesterase a c t i v i t y compared with that observed with 100 uM papaverine alone (not shown). Further experiments were performed to test whether the effect of XAC was reversible by simultanenous activation of Gi via the inhibitory PGE1 receptors known to be present in adipocyte membranes (6,9).
Figure 2 demonstrates that the
stimulatory effect of XAC is clearly diminished in a dose-dependent fashion by prostaglandin EI.
In a series of three experiments, inhibition of the effect of
XAC (I uM) by I , 10 and 100 nM of PGE1 averaged 90.1, 32.1 and 24.7~ of control, respectively.
These l a t t e r experiments attest to the r e v e r s i b i l i t y of the
stimulatory effect of XAC on adenylate cyclase.
In two separate experiments
performed i t was shown that PGE1 does not i n h i b i t [3H]XAC binding, ruling out a direct effect of PGE1 at A1 adenosine receptors to explain the above finding. To probe the action of XAC on Gi functioning more closely, i t s effect on GTP-dependent inhibition of adenylate cyclase was determined. Maximal inhibition of adenylate cyclase was evident at ~ 5-10 pM GTP (Figure 3).
Increasing the
concentration of XAC to 10 and 1000 nM led to significant attenuation of GTP inhibition.
However, the concentration of GTP producing maximal inhibition
942
Vol. 153, No. 3, 1988
BIOCHEMICAL AND BIOPHYSICAL RESEARCHCOMMUNICATIONS
remained unchanged. This finding suggests a noncompetitive mode of interaction between XAC and the GTP binding site on Gi . DISCUSSION The present data demonstrate that XAC can interact with multiple components of the AI adenosine receptor-adenylate cyclase complex. These include the A1 adenosine receptor, the Gi protein and the low Km cyclic AMP phosphodiesterase. XAC binds to A1 receptors with high a f f i n i t y in the low nanomolar range and to Gi in the mid-nanomolar range. However, inhibition of the low Km cyclic AMP phosphodiesterase is not evident until near millimolar concentrations of XAC are attained. Several pieces of evidence support a direct interaction of XAC with the Gi protein.
First, XAC elevates adenylate a c t i v i t y in adipocyte membrane which is
under tonic inhibition by Gi (10).
This suggests that this drug in some manner
interrupts the normal functioning of Gi .
Second, this stimulatory action of XAC
is evident only at concentrations of GTP sufficient to activate Gi and is not observed in the absence of GTP or at the lower concentrations sufficient for optimal Gs a c t i v i t y (3,11). Third, pertussis toxin treatment which ADP ribosylates and inactivates Gi (and Go) (12) abolishes the effect of XAC on adenylate cyclase.
Fourth, activation of Gi via PGE1 receptors reverses this
stimulatory effect of XAC, implying that this drug is acting at a site on or closely associated with Gi to disrupt i t s normal functioning.
I t is unlikely
that XAC increases adenylate cyclase a c t i v i t y by stimulating B-adrenergic receptors or by activating the catalytic unit, owing to the GTP dependence of XAC's effect and i t s abolition by pertussis toxin (Figure 1).
Furthermore, the
increase in adenylate cyclase a c t i v i t y is apparently not due to antagonism of adenosine at A1 receptors since an adequate amount of adenosine deaminase was used to degrade endogenous adenosine. Attentuation of Gi a c t i v i t y is a novel action of this AI adenosine antagonist not previously described. Its effects are similar to those described for several of the new inotropic agents like sulmazole (10). Unlike sulmazole, however, XAC is a potent inhibitor of Gi (4-5 orders of magnitude higher a f f i n i t y ) . Other agents known to attenuate inhibition of adenylate cyclase a c t i v i t y include phorbol esters (13) and divalent cations such as Mg~+ and Mn2+ (14). In contrast to XAC, the effects of these agents are not fully reversible by inhibitory agonists (13,14). The i n a b i l i t y of GTP to produce maximal inhibition of forskolin-stimulated adenylate cyclase a c t i v i t y in presence of XAC suggests that inhibition of Gi function derives from a reduction in the binding of GTP to this G protein. Furthermore, the noncompetitive nature of this interaction suggests that XAC influences GTP binding to Gi via an allosteric site. The fact that XAC-mediated 943
VoI. 153, No. 3, 1 9 8 8
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
enhancement in adenylate cyclase activity is offset by increasing GTP turnover at the Gi protein (by PGE1) implies a close association between the GTP binding site and the site occupied by XAC such that a conformational change induced by the binding of one ligand will greatly affect the binding of the second ligand to its site. The relative contributions of AI adenosine receptor antagonism and Gi blockade to the pharmacological effects of XAC is unclear at present. Nevertheless, the high potency and reversibility of binding at the Gi protein may make i t a useful biochemical tool to explore the functions of this G protein. ACKNOWLEDGEMENTS The authors would like to thank Linda Scherich for her excellent secretarial assistance. G.L.S. is an Established Investigator of the American Heart Association and is supported in part by NHLBI grant ROIHL35134 and Grant-in-aid 850612 from the American Heart Association, with funds contributed in part by the North Carolina Affliate. REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
Jacobson, K.A., Ukena, D., Kirk, K.L. and Daly, J.W. (1986) Proc. Natl. Acad. Sci. USA 83, 4089-4093. Evoniuk, G., Jacobson, K.A., Shamim, M.T., Daly, J.W. and Wurtman, R.J. (1987) J. Pharmacol. Exp. Ther. 242, 882-887. Cooper, D.M.F., Schlegel, W., Lin, M.C. and Rodbell, M. (1979) J. Biol. Chem. 254, 8927-8931. \ Gilman, A.G. (1987) Ann. Rev. Biochem. 56, 615-649. Wolff, J., Cook, G.H., Goldhamner, A.R., Londos, C. and Hewlett, E.L. (1984) In Advances in Cyclic Nucleotide Protein Phosphorylation Research (P. Greengard, Ed.), Vol. 17, pp. 161-172. Raven Press, New York. Parsons, W.J. and Stiles, J.L. (1987) J. Biol. Chem. 262, 841-847. Salomon, Y., Londos, C. and Rodbell, M. (1974) Anal. Biochem. 58, 541-548. De Mazancourt, P. and Guidicelli, Y. (1984) FEBSLett. 173, 385-388. Murayama, T. and Ui, M. (1984) J. Biol. Chem. 259, 761-769. Parsons, W.3., Ramkumar, V. and Stiles, G.L. (1988) Mol. Pharmacol. (in press). Murayama, T. and Ui, M. (1983) J. Biol. Chem. 258, 3319-3326. Ui, M. (1984) Trends Pharmacol. Sci. 5, 277-279. Bell, J.D. and Buxton, L.L. (1986) J. Biol. Chem. 261, 12036-12041. Katada, T., Gilman, A.G., Watanabe, Y., Baver, S. and Jakobs, K.H. (1985) Eur. 3. Biochem. 151, 431-437.
944
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
June 30,1988
Pages 939-944
A NOVELSITE OF ACTION OF A HIGH AFFINITY AI ADENOSINERECEPTORANTAGONIST Vickram Ramkumar and Gary L. Stiles Departments of Medicine (Cardiology) and Biochemistry Duke University Medical Center Durham, NC 27710 Received March 28, 1988
XAC, a high a f f i n i t y antagonist of the AI adenosine receptor, enhances adenylate cyclase a c t i v i t y by 1.3-2 fold with~an ECBo of ~ 47 nM in adipocyte membranes pretreated with adenosine deaminase to eliminate adenosine and in the presence of total phosphodiesterase inhibition by 100 ~M papaverine. This effect of XAC is observed only at concentrations of GTP sufficient to activate Gi (~ 5 x 10-6 M GTP) and is not evident in the absence or presence of lower GTP concentrations. ADP ribosy!ation of G~ by pertussis toxin treatment also abolishes this stimulatory action of XAC. Furthermore, in the presence of GTP activation of inhibitory prostaglandin EI receptors diminishes the stimulatory effect of XAC on adenylate cyclase. In ~ddition, XAC interferes with GTP-mediated inhibition of forskolin-stimulated adenylate cyclase a c t i v i t y in a noncompetitive manner. Finally, XAC is only a weak inhibitor of the low K~ cyclic AMP phosphodiesterase, producing ~ 40% inhibition of phosphodiester~'se a c t i v i t y at a concentration of 100 ~M. These data suggest that XAC increases adenylate cyclase a c t i v i t y in absence of endogenous adenosine by inhibiting tonic Gi a c t i v i t y in a reversible manner. ©198BAcademicP~es~,In~.
XAC, 8-[4-[[[[(2-amino-ethyl)-amino]carbony!]methyl]oxy]phenyl]-l,3dipropylxanthine, is a relatively selective and high a f f i n i t y antagonist of the AI adenosine receptor (1).
This property apparently underlies i t s a b i l i t y to
block A1 adenosine receptor agonists mediated bradycardia (2) and their inhibition of adenylate cyclase. A1 adenosine receptors inhibit adenylate cyclase a c t i v i t y via the inhibitory guanine nucleotide regulatory protein (Gi) at GTP concentrations in the micromolar range (3). Agonists acting at AI receptors activate Gi leading to the exchange of bound GDP for GTP. This step represents the rate limiting step in the activation process. Activation leads to the dissociation of the oligomer into ai-GTP and By subunits.
I t appears that inhibition of adenylate cyclase
results from the binding of free as subunits by By, preventing their interaction with the catalytic unit. Termination of the activation process depends on GTPase a c t i v i t y intrinsic to ai (4). In preliminary studies in rat adipocyte membranes, we observed that XAC not only attenuates R-PIA mediated inhibition of adenylate cyclase as would be
939
0006-291X/88 $1.50 Copyright © 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 153, No. 3, 1 9 8 8
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
expected of an AI receptor antagonist but by i t s e l f actually stimulates the a c t i v i t y of this enzyme. This is a property not normally associated with adenosine receptor antagonists.
In this report we have characterized the
biochemical mechanisms responsible for this a c t i v i t y and demonstrate XAC's interaction at a site distinct from the A1 receptor, namely the inhibitory guanine nucleotide regulatory protein (Gi).
METHODS Pertussis vaccine, which contains the toxin (5) was administered intraperitoneally (~ 300 OPU/Kg; 0.3-0.5 ml) to rats (male Sprague Dawley, 250-300 g, Charles River Breeding Laboratories) three days prior to sacrifice. Control and pertussis-intoxicated rats were sacrificed by decapitation and adipocyte membranes were prepared from epididymal fat pads as described previously (6). Membranes were pretreated with adenosine deaminase before performing all assays. Adenylate cyclase a c t i v i t y in adipocyte membranes was measured as described previously (5). Briefly, 20 ul of membranes (~ 50 ~g protein) suspended in 75 mM Tris HCl buffer (pH 7.4 at 30°C), containing 12.5 mMMgCl~, 200 mM NaCl, 2.5 mM DTT, and 8 units/ml of adenosine deaminase were incubated with 20 ~l of Lomix (0.14 mM dATP, 5 mMphosphocreat!ne, 1 ~M cAMP, 30 units/ml creatine phosphokinase, ~ 1.5 ~Ci of [32P]ATP and 10 ~l of H20 or drugs. Papaverine (10-4 M) was used in all experiments to provide adequate inhibition of the low K~ cyclic AMP phosphodiesterase. Experiments were performed in the absence o~ GTP, or in the presence of 5 x 10-6 M GTP or at varying GTP concentrations (see Results). Cyclic AMP assays were incubated for 15 min at 30°C and terminated by the addition of 1 ml ice-cold stop solution containing ~ !5,000 cpm [3HI cAMP, 0.3 mM cAMPand 0.4 mMATP. Cyclic AMP was isolated by the method of Salomon et
a!. (7). Assay for the low K~ cyclic AMP phosphodiesterase was performed according to the method described byI~e Mazancourt and Guidicelli for rat adipocytes (8). [3HI cAMPlevels were measured by pouring the supernatant over Dowex 50 W-X4 columns (200-400 mesh). The f i r s t eluate and subsequent 1.5 ml wash were discarded. The columns were then eluted twice with 3 ml of H20 each and the eluates were collected and counted. Values obtained from tubes boiled without rior incubation of the reaction mixture were used to establish the maximal 100%) substrate level.
~
RESULTS XAC enhances adenylate cyclase a c t i v i t y in a dose-dependent manner with ECso for stimulation being 47 ± 11 nM (mean ± S.E.M.) and maximal a c t i v i t y being 148 ± 7% (range from 135-200% of control) (N = 9). The a b i l i t y of XAC to increase adenylate cyclase a c t i v i t y appears to be dependent on concentrations of GTP necessary to f a c i l i t a t e the a c t i v i t y of Gi (5 x 10-6 M) and is not observed in the absence of GTP (Figure 1) or at concentrations of GTP sufficient to activate Gs but not Gi (Figure 3). This indicates the essential role of an active Gi to promote stimulatory a c t i v i t y of XAC. This notion is supported by the finding that inactivation of Gi by pertussis toxin attenuates XAC enhancement of adenylate cyclase a c t i v i t y (Figure 1). Interestingly, adenylate cyclase a c t i v i t y 940
Vol. 153, No. 3, 1988
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
160 co
/•Control
o
140 0
< ¢0
o
r"o
<
"~" I
0
Figure ~.
//
"No
GTP
I
I
I
I
8
7
6
5
-Log [XAC] (M) Effect of XAC on basal adenylate cyclase activity.
Adenylate cyclase assayswere performed in control membranesin the absence (no GTP) and presence (control) of 5 x 10-6 M GTP or in pertussis-intoxicated membranes in presenceof GTP (pertussis toxin) as described in Methods. Basal adenylate cyclase activity in control, no GTP and pertussis toxin groups averaged 41.6, 77.4 and 75.0 pmol/min/mg protein. This is a representative of 3-9 experiments performed with similar results.
obtained in the absence and presence of 5 x 10-6 M GTP in control membranes and in pertussis toxin-intoxicated membranes (GTP at 5 x 10-6 M) were remarkably different, being 77.4, 41.6, and 75.0 pmol/mg/min, respectively (mean of 3-9 experiments).
This suggests that under basal conditions Gi is tonically active
in rat adipocyte membranes. We have previously demonstrated that XAC is a potent A1 adenosine receptor antagonist with Kd values in rat adipocyte ranging between I-2 nM (Ramkumar and Stiles, submitted).
In order to rule out A1 adenosine receptor blockade of
endogenous adenosine as the reason for the increase in adenylate cyclase a c t i v i t y observed with XAC, all experiments were performed by preincubating the membranes with about 8 units/ml of adenosine deaminase to degrade endogenous adenosine. This pretreatment was adequate since no change in a c t i v i t i e s of adenylate cyclase in the absence and presence of XAC were obtained by increasing the concentration of adenosine deaminase by up to 3-fold (data not shown). Furthermore, the increase in adenylate cyclase a c t i v i t y was not a consequence of inhibition of the low Km cyclic AMP phosphodiesterase by XAC. Significant inhibition of the l a t t e r enzyme was obtained only at concentrations of XAC close to the millimolar range. Inhibition by 100 uM XAC was ~ 40% of that observed with 100 uM papaverine (not shown). This concentration of papaverine was used in a l l experiments.
The
addition of 100 vM XAC to 100 uM papaverine did not improve inhibition of the 941
Voi. 153, No. 3, 1 9 8 8
160
:_~
Prostaglandin
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
//
E 1 (M)
.o
"~
•
10 "9
•
1 0 "8
100 <0 80
_
,
/
~
° 8 g 12o e(D
"a <
®
60 >, e'10
100
,< I#
I
I
I
I
I
0
10
9
8
7
6
-Log Concentration XAC (M)
[] 10"6
40 ~jl
0
8
Q
I
I
I
7
6
5
-Log [GTP] (M)
Figure_2. Prostaglandin E1 mediated inhibition of the stimulatory effect of XAC on adenylate cyclase. Membranes were incubated with an assay mixture as described in Methods containing 5 x 10-6 M GTP and different concentrations of XAC and prostaglandin E1. Each point is the average of duplicate determinations from a single experiment. Experiments were repeated twice with similar results. Figure 3.
GTP inhibition of forskolin-stimulated adenylate cyclase activity in absence and presence of XAC.
GTP inhibition curves were performed in absence and presence of 10 and 1000 nM of XAC as described in the Methods. Asterisks (*) denote statistically significant difference from control (no XAC). Eachpoint is the average of three experiments performed in duplicate. Forskolin-stimulated (5 uM forskolin) adenylate cyclase activity in the absence of GTP and XAC averaged 174 pmol/min/mg protein and did not change significantly with addition of XAC.
phosphodiesterase a c t i v i t y compared with that observed with 100 uM papaverine alone (not shown). Further experiments were performed to test whether the effect of XAC was reversible by simultanenous activation of Gi via the inhibitory PGE1 receptors known to be present in adipocyte membranes (6,9).
Figure 2 demonstrates that the
stimulatory effect of XAC is clearly diminished in a dose-dependent fashion by prostaglandin EI.
In a series of three experiments, inhibition of the effect of
XAC (I uM) by I , 10 and 100 nM of PGE1 averaged 90.1, 32.1 and 24.7~ of control, respectively.
These l a t t e r experiments attest to the r e v e r s i b i l i t y of the
stimulatory effect of XAC on adenylate cyclase.
In two separate experiments
performed i t was shown that PGE1 does not i n h i b i t [3H]XAC binding, ruling out a direct effect of PGE1 at A1 adenosine receptors to explain the above finding. To probe the action of XAC on Gi functioning more closely, i t s effect on GTP-dependent inhibition of adenylate cyclase was determined. Maximal inhibition of adenylate cyclase was evident at ~ 5-10 pM GTP (Figure 3).
Increasing the
concentration of XAC to 10 and 1000 nM led to significant attenuation of GTP inhibition.
However, the concentration of GTP producing maximal inhibition
942
Vol. 153, No. 3, 1988
BIOCHEMICAL AND BIOPHYSICAL RESEARCHCOMMUNICATIONS
remained unchanged. This finding suggests a noncompetitive mode of interaction between XAC and the GTP binding site on Gi . DISCUSSION The present data demonstrate that XAC can interact with multiple components of the AI adenosine receptor-adenylate cyclase complex. These include the A1 adenosine receptor, the Gi protein and the low Km cyclic AMP phosphodiesterase. XAC binds to A1 receptors with high a f f i n i t y in the low nanomolar range and to Gi in the mid-nanomolar range. However, inhibition of the low Km cyclic AMP phosphodiesterase is not evident until near millimolar concentrations of XAC are attained. Several pieces of evidence support a direct interaction of XAC with the Gi protein.
First, XAC elevates adenylate a c t i v i t y in adipocyte membrane which is
under tonic inhibition by Gi (10).
This suggests that this drug in some manner
interrupts the normal functioning of Gi .
Second, this stimulatory action of XAC
is evident only at concentrations of GTP sufficient to activate Gi and is not observed in the absence of GTP or at the lower concentrations sufficient for optimal Gs a c t i v i t y (3,11). Third, pertussis toxin treatment which ADP ribosylates and inactivates Gi (and Go) (12) abolishes the effect of XAC on adenylate cyclase.
Fourth, activation of Gi via PGE1 receptors reverses this
stimulatory effect of XAC, implying that this drug is acting at a site on or closely associated with Gi to disrupt i t s normal functioning.
I t is unlikely
that XAC increases adenylate cyclase a c t i v i t y by stimulating B-adrenergic receptors or by activating the catalytic unit, owing to the GTP dependence of XAC's effect and i t s abolition by pertussis toxin (Figure 1).
Furthermore, the
increase in adenylate cyclase a c t i v i t y is apparently not due to antagonism of adenosine at A1 receptors since an adequate amount of adenosine deaminase was used to degrade endogenous adenosine. Attentuation of Gi a c t i v i t y is a novel action of this AI adenosine antagonist not previously described. Its effects are similar to those described for several of the new inotropic agents like sulmazole (10). Unlike sulmazole, however, XAC is a potent inhibitor of Gi (4-5 orders of magnitude higher a f f i n i t y ) . Other agents known to attenuate inhibition of adenylate cyclase a c t i v i t y include phorbol esters (13) and divalent cations such as Mg~+ and Mn2+ (14). In contrast to XAC, the effects of these agents are not fully reversible by inhibitory agonists (13,14). The i n a b i l i t y of GTP to produce maximal inhibition of forskolin-stimulated adenylate cyclase a c t i v i t y in presence of XAC suggests that inhibition of Gi function derives from a reduction in the binding of GTP to this G protein. Furthermore, the noncompetitive nature of this interaction suggests that XAC influences GTP binding to Gi via an allosteric site. The fact that XAC-mediated 943
VoI. 153, No. 3, 1 9 8 8
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
enhancement in adenylate cyclase activity is offset by increasing GTP turnover at the Gi protein (by PGE1) implies a close association between the GTP binding site and the site occupied by XAC such that a conformational change induced by the binding of one ligand will greatly affect the binding of the second ligand to its site. The relative contributions of AI adenosine receptor antagonism and Gi blockade to the pharmacological effects of XAC is unclear at present. Nevertheless, the high potency and reversibility of binding at the Gi protein may make i t a useful biochemical tool to explore the functions of this G protein. ACKNOWLEDGEMENTS The authors would like to thank Linda Scherich for her excellent secretarial assistance. G.L.S. is an Established Investigator of the American Heart Association and is supported in part by NHLBI grant ROIHL35134 and Grant-in-aid 850612 from the American Heart Association, with funds contributed in part by the North Carolina Affliate. REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
Jacobson, K.A., Ukena, D., Kirk, K.L. and Daly, J.W. (1986) Proc. Natl. Acad. Sci. USA 83, 4089-4093. Evoniuk, G., Jacobson, K.A., Shamim, M.T., Daly, J.W. and Wurtman, R.J. (1987) J. Pharmacol. Exp. Ther. 242, 882-887. Cooper, D.M.F., Schlegel, W., Lin, M.C. and Rodbell, M. (1979) J. Biol. Chem. 254, 8927-8931. \ Gilman, A.G. (1987) Ann. Rev. Biochem. 56, 615-649. Wolff, J., Cook, G.H., Goldhamner, A.R., Londos, C. and Hewlett, E.L. (1984) In Advances in Cyclic Nucleotide Protein Phosphorylation Research (P. Greengard, Ed.), Vol. 17, pp. 161-172. Raven Press, New York. Parsons, W.J. and Stiles, J.L. (1987) J. Biol. Chem. 262, 841-847. Salomon, Y., Londos, C. and Rodbell, M. (1974) Anal. Biochem. 58, 541-548. De Mazancourt, P. and Guidicelli, Y. (1984) FEBSLett. 173, 385-388. Murayama, T. and Ui, M. (1984) J. Biol. Chem. 259, 761-769. Parsons, W.3., Ramkumar, V. and Stiles, G.L. (1988) Mol. Pharmacol. (in press). Murayama, T. and Ui, M. (1983) J. Biol. Chem. 258, 3319-3326. Ui, M. (1984) Trends Pharmacol. Sci. 5, 277-279. Bell, J.D. and Buxton, L.L. (1986) J. Biol. Chem. 261, 12036-12041. Katada, T., Gilman, A.G., Watanabe, Y., Baver, S. and Jakobs, K.H. (1985) Eur. 3. Biochem. 151, 431-437.
944