Cellular and Molecular Neurobiotogy, Vol. 15, No. 4, 1995

The Amphiphilic Peptide Adenoregulin Enhances Agonist Binding to A1-Adenosine Receptors and [35S]GTP ,S to Brain Membranes Roger W. Moni, I~ Francisco S. Romero, 1 and John W. Daly 1-~ Received February 2, 1995, accepted March 1, 1995 KEY WORDS: peptides, adenosine receptors, adrenergic receptors, serotonin receptors, guanyl nucleotide exchange.

SUMMARY

1. Adenoregulin is an amphilic peptide isolated from skin mucus of the tree flog, Phyltomedusa bicotor. Synthetic adenoregulin enhanced the binding of agonists to several G-protein-coupled receptors in rat brain membranes. 2. The maximal enhancement of agonist binding, and in parentheses, the concentration of adenoregulin affording maximal enhancement were as follows: 60% (20/zM) for At-adenosine receptors, 30% (100/zM) for A2:adenosine receptors, 20% (2/zM) for a2-adrenergic receptors, and 30% (10/zM) for 5HT1A receptors. High affinity agonist binding for A1., a2_, and 5HT1A-receptors was virtually abolished by GTP3,S in the presence of adenoregulin, but was only partially abolished in its absence. Magnesium ions increased the binding of agonists to receptors and reduced the enhancement elicited by adenoregulin. 3. The effect of adenoregulin on binding of N6-cyclohexyladenosine ([3H]CHA) to Al-receptors was relatively slow and was irreversible. Adenoregulin increased the Bmax value for [3H]CHA binding sites, and the proportion of high affinity states, and slowed the rate of [3H]CHA dissociation. Binding of the I Laboratory of Bioorganic Chemistry, National Institute of Diabetes, Digestive and Kidney Diseases, National Intitues of Health, Bethesda, Maryland 20892. ~Present address: Queensland Pharmaceutical Research Institute, Mt. Gravatt Research Park, Nathan, Brisbane, Queensland, Australia. 3To whom correspondence should be addressed at Bldg. 8, Rm. 1A17, National Institutes of Health, Bethesda, Maryland 20892. 465 0272-4340t95t0800.0465507.5010~ 1995PlenumPublishingCorporation

466

Moni, Romero,and Daly

A~-selective antagonist, [3H]DPCPX, was maximally enhanced by only 13% at 2/xM adenoregulin. Basal and A,-adenosine receptor-stimulated binding of [35S]GTPyS were maximally enhanced 45% and 23%, respectively, by 50tzM adenoregulin. In CHAPS-solubilized membranes from rat cortex, the binding of both [3H]CHA and [3H]DPCPX were enhanced by adenoregulin. Binding of [3H]CHA to membranes from DDT~ MF-2 cells was maximally enhanced 17% at 20/.tM adenoregulin. In intact DDT1 MF-2 cells, 20/xM adenoregulin did not potentiate the inhibition of cyclic AMP accumulation mediated via the adenosine A~ receptor. 4. It is proposed that adenoregulin enhances agonist binding through a mechanism involving enhancement of guanyl nucleotide exchange at G-proteins, resulting in a conversion of receptors into a high affinity state complexed with guanyl nucleotide-free G-protein.

INTRODUCTION Adenosine regulates diverse biological events via at least three receptors types, widely distributed in mammalian tissues and distinguishable by binding and functional criteria (Abbrachio et aL, 1993). The Al-adenosine receptors couple to Gil-3 and Go proteins (Munshi et al., 1991) and thereby, regulate several effector systems, including adenylate cyclase (Van Calker et aL, 1978; Cooper et al., 1980) potassium channels (Trussel and Jackson, 1987; Tawfik et aL, 1989), and phospholipase C (Delahaunty et aL, 1988). Interaction of Al-adenosine receptors with G-proteins results in enhancement of a low Km GTPase of the G-protein (Hausleithner et al., 1991). The A2a-adenosine receptors couple to Gs-proteins and thereby regulate adenylate cyclase (Cooper et al., 1980; Daly et al., 1983). The Aa-adenosine receptors couple to G-proteins and thereby regulate adenylate cyclase and phospholipase C (Zhou et aL, 1992; Salvatore et al., 1993). The ubiquity of adenosine receptors makes them difficult therapeutic targets, and the clinical use of agonists/antagonists has been very limited (Jacobson et aL, 1992). A relatively new strategy is to indirectly stimulate the binding of endogenous adenosine, using ligands which bind to allosteric sites on the Al-adenosine receptor. Several 2-amino-3-benzoylthiophenes selectively enhance binding of agonists to As-adenosine receptors and potentiate the Al-mediated inhibition of forskolin-stimulated cAMP accumulation in FRTL-5 cells (Bruns and Fergus, 1990; Bruns et al., 1990). Such 2-amino-3-benzoylthiophenes appeared to promote a high affinity state of the receptor independent of G-proteins, since enhancement was greatest in the presence of GTP or N-ethylmaleimide, which would uncouple the receptor from the Grprotein. Allosteric enhancers of Al-adenosine receptors could act as "use-dependent" drugs to treat ischemia or seizures. Other examples of positive allosteric regulation of G-protein coupled receptors are rare. But recently a peptide, adenoregulin, that enhanced binding of an agonist to A~-adenosine receptors was discovered (Daly et al., 1993). Adenoregulin, a 33 amino acid basic amphiphilic peptide, was isolated from mucus of the skin of the neotropical tree frog, Phyllomedusa bicolor (Daly et al.,

Adenoregulinand AdenosineReceptors

467

1993). The peptide was detected, and named based on its ability to markedly enhance binding of the agonist [3H]N6-R-phenylisopropyladenosine to rat brain adenosine receptors. In preliminary studies, there was no significant effect of adenoregulin on the binding of the antagonist [3H]xanthine amine congener (XAC) to brain A~-adenosine receptors, nor any significant effect on binding of the antagonists [3H]naloxone and [3H]methylscopolamine to brain opioid or muscarinic receptors, respectively. Limited supplies of natural adenoregulin precluded further studies, but synthetic material also enhanced agonist binding to Al-adenosine receptors, although with somewhat lower potency (Daly et al., 1993). The synthetic adenoregulin also was not identical to the natural adenoregulin in chromatographic properties. It is possible that like other skin peptides from the frog Phyllomedusa bicolor (Erspamer et al., 1985), natural adenoregulin contains a D-amino acid or an amidated carboxy terminus. A cDNA encoding an 81 amino acid precursor of adenoregulin has been detected from skin of Phyllomedusa bicolor (Amiche et al., 1993). Synthetic adenoregulin has recently been shown to have antimicrobial properties (Mor et al., 1994). Adenoregulin like many basic amphiphilic peptides presumably assumes an a-helical structure with lysines on one face (see Mor et al., 1994). Adenoregulin, thus, shares structural analogies with the 14 amino acid amphiphilic peptide mastoparan. Mastoparan activates G-proteins leading to a stimulation of nucleotide exchange at ce-subunits (Higashijima et al., 1988, 1990). Mastoparan, other cationic amphiphilic peptides and polyamines all seem capable of directly activating G-proteins (Mousli et al., 1990), but effects of these agents on the receptor-G protein complex, as assessed by receptor ligand binding to not appear to have been investigated. Synthetic adenoregulin and mastoparan have now been shown to share many properties, including stimulation of agonist binding to Al-adenosine receptors, stimulation of nucleotide exchange and stimulation of phosphoinositide breakdown, calcium influx and norepinephrine release in cultured cells (Shin et aL, 1994). We now report a detailed study of the effects of synthetic adenoregulin on binding of ligands to A~-adenosine and other receptors. A mechanism by which adenoregulin might enhance agonist binding to A~adenosine receptors is proposed. EXPERIMENTAL PROCEDURES Materials. Radioligands were purchased from DuPont NEN (Boston, MA): [3H]CHA (30.2 Ci/mmol), [3H]DPCPX (108.3 Ci/mmol), [3H]CGS21680 (39.6 Ci/ mmol), [3H]clonidine (60 Ci/mmol), (:t:)-[3H]8-OH-DPAT (162.9 Ci/mmol) and [35S]GTPTS (1000 Ci/mmol). Compounds and reagents were from the following sources; 2-chloroadenosine (Fluka BioChemica, Switzerland), (-)-norepinephrine (Research Biochemicals Inc., Wayland, MA), CHAPS (Calbiochem, La Jolla, CA), serotonin hydrogen oxalate (Regis Chemical Company, Chicago, IL), GTPTS (Boehringer Mannheim GmbH. Germany), apoprotinin, leupeptin, PMSF, soy bean trypsin inhibitor, polyethylenimine and adenosine deaminase Type VI (Sigma Chemical Company, St. Louis, MO). The cyclic AMP assay kit

468

Moni, Romero, and Daly

was from Amersham Life Science (UK). PD81,723 was a kind gift from Dr. Robert F. Bruns (formerly Warner-Lambert, Parke-Davis, MI). Adenoregulin composed entirely of L-amino acids and with a free acid carboxy terminus was synthesized at the W.M. Keck Foundation Biotechnology Resource Laboratory (New Haven, CT). Preparation of Membranes. Membranes were prepared from the cortex, striatum, cerebellum and hippocampus of adult rat brains (Pel-Freeze Biologicals Co., Rogers, AR) and from confluent subcultures of DDT~ MF-2 cells. Cortical tissue was homogenized in approximately 10 volumes of ice-cold 0.34 M sucrose (Polytron setting 5, ten seconds) and centrifuged (1,000 x g, 20 minutes, 4°C). The supernatant was centrifuged (35,000× g, 25 minutes, 4°C) and the resultant membrane fraction washed twice in 20 volumes of ice-cold 50 mM Tris HC1 buffer (pH 7.4). All other brain tissues and DDT~ MF-2 cells were homogenized and washed twice in 50 mM Tris HC1 (pH 7.4) by centrifugation. All membranes were finally resuspended in 50mMTrisHC1 (pH7.4) and the protein concentration determined using the bicinchoninic acid kit at 37°C (Pierce Chemical Co., Rockford, IL). Membranes were stored at minus 70°C for up to three months without change of binding characteristics. Receptor Binding Assays. Experiments were performed in 12 × 75 mm glass vials containing 100/xg of membrane protein from rat brain or DDT~ MF-2 cells, radioligand, adenoregulin and other specified reagents in an assay volume of 250/xL. Steady state incubations were as follows: l nM [3H]CHA, 90 min; 0.2 nM [3H]DPCPX, 120 min; 5 nM [3H]CGS21680, 120 min; 1 n M ( + )-[3H]8OH-DPAT, 60 min and 4nM [3H]clonidine, 90 min. Nonspecific binding was determined in the presence of 10/zM 2-chloroadenosine ([3H]CHA and [3H]DPCPX), 20/xM 2-chloroadenosine ([3H]CGS21680), 10/xM serotonin ([3H]8-OH-DPAT) and 10/xM (-)-norepinephrine ([3H]clonidine). Reactions were initiated by the addition of membrane protein and conducted at 25°C unless stated otherwise. For adenosine receptor binding assays, membrane suspensions were preincubated with adenosine deaminase at 0.3 U/mg for 10 min at 37°C. Unless stated otherwise, experiments with A2a-adenosine receptors included 10 mM MgC12 in the assay. Bound and unbound radioligand were separated by filtration (Whatman FP-100, GF/B) using a Brandel M24R cell Harvester (Gaithersburg, MD). Membrane residues were washed three times with 4 mL of ice-cold 50 mM Tris HC1 (pH 7.4). Wet filters were then immersed in 4 ml of scintillant (Hydrofluor, National Diagnostics, NJ), equilibrated and counted by liquid scintillation spectrometry for two min. Saturation experiments were performed with Al-adenosine receptors using eight concentrations of [3H]CHA (0.1-20 nM) for 90 rain or [3H]DPCX (0.02-10 nM) for 120 min in the absence and presence of adenoregulin at specified concentrations. Kinetic experiments were performed with and without adenoreuglin. The time course to steady state was determined for 1 nM [3H]CHA binding to A~-adenosine receptors in rat cortical membranes. Dissociation of 1 nM [3H]CHA from A~-adenosine receptors in rat cortical membranes was initiated by the addition of 2-chloroadenosine to a final concentration of 10/xM and followed for 90 min. In all experiments,

Adenoregulin and Adenosine Receptors

469

receptor binding was defined by subtracting nonspecific binding from total binding. The Al-adenosine receptor from rat cerebral cortex was solubilized as follows: Membrane suspensions, treated with 1 U/mL adenosine deaminase were gently mixed with a 1% solution of CHAPS (protein suspension : detergent, 1:2) for 30 min on ice. The extract was diluted 1:4 with ice-cold 50mMTrisHC1 buffer (pH 7.4) and centrifuged (100,000 × g, 60 min, 4°C). Protein recovery in the supernatant was determined as previously described. Approximately 60/zg of solubilized membrane protein was used for binding studies with [3H]CHA and [3H]DPCPX as above, except that filters were pretreated with 0.3% polyethylenimine for 120 rain. Recovery of membrane protein after solubilization was 38%. Protein recovery in the supernatant was determined with the bicinchoninic acid kit. Binding of [35S]GTPTS to Brain Membranes. Membranes from rat cortex were prepared and treated with 0.3 U / m L adenosine deaminase as described (Lorenzen et al., 1993). Triplicate measurements of [3sS]GTPTS binding were performed in a 100/zL assay volume containing 50mMTrisHC1 (pH 7.4), 5 mM MgC12, i mM EDTA, 10/zM GDP, 1 mM dithiothreitol, 100 mM NaCl and 0.6nM [35S]GTP-/S (about 105dpm per vial). Reactions were initiated by the addition of 10/zg of membranes in the presence and absence of 1 p~M CHA and adenoregulin at specified concentrations. Incubations (150 rain, 25°C) were terminated by the addition of 3mL of ice-cold buffer (50mMTrisHC1, 5mMMgC12, pH7.4) and rapid filtration over Whatman GF/B filters, followed by two washes, using a Brandel cell harvester. Nonspecific binding of the radioligand was determined in the presence of 10/zM GTPyS. Measurements of Cyclic AMP Accumulation in DDTI MF-2 Cells. DDT1 MF-2 cells were grown at 37°C with 20% 02 and 7% CO2 to confluence in Dulbecco's Modified Eagle Medium-high glucose, (DMEM) supplemented with 10% fetal bovine serum, 100U/mL penicillin and 100p.g/mL streptomycin. Subcultured cells were washed three times (1,000 x g, 5 rain) in 20 mM HEPESDMEM buffer (pH 7.2) and viability determined by exclusion of 0.04% Trypan Blue. Cells were incubated at 25°C with 30/xM rolipram and 1 U/mL adenosine deaminase for 10 rain, followed by preincubation in the presence or absence of 20/~M adenoregulin for 90 min. Incubation were then initiated by the addition of cells (3 × 105 cells) to 250/zL with HEPES-DMEM buffer containing 10/zM forskolin and increasing concentrations of CHA. After 10 rain incubation at 25°C, reactions were stopped by rapid heating of cells at 100°C for 3 min. Cyclic AMP accumulation was determined in a 50/xL aliquot of the supernatant using a commercial cyclic AMP kit. Data Analysis. Arithmetic means-4-SEM were calculated from three experiments, each performed with triplicate measurements. Saturation and kinetic data were analyzed using the INPLOT IV program. The model best describing the results was selected by conducting an F-test. Estimation of EC50 values for increases in agonist binding were estimated directly from the curve.

470

Moni, Romero, and Daly

RESULTS Agonist Binding to G-Coupled Receptors Binding Profiles to G~-Coupled Receptors. Adenoregulin enhanced the specific binding of the agonist [3H]CHA to Al-adenosine receptors in cerebral cortical membranes with a threshold of 0.5/zM and maximal stimulation of 60% at 20/zM (Fig. 1A). Nonspecific binding was not altered by adenoregulin. Specific binding of [3H]8-OH-DPAT to hippocampal 5HTlA-receptors (Fig. 1B) and [3H]clonidine to cortical a2-receptors (Fig. 1C) had similar bell-shaped profiles. The stimulatory phase was followed by a steep inhibitory phase at higher concentrations of adenoregulin. Values for threshold, ECs0, ECmax, and maximal enhancement of agonist binding to the G-coupled receptors are reported in Table I. Effects of GTP3,S and MgCl2. Low and high agonist affinity states for G-coupled receptors can be interconverted using guanine nucleotides or magnesium salts. Inclusion of micromolar concentrations of GTP or GTP3,S (an essentially nonhydrolysable analog of GTP) converts most receptors to a low affinity binding state. This was confirmed by determining agonist binding to the AI-, aa- and 5HTla-receptors with increasing concentrations of GTP3,S. Agonist binding was maximally reduced at 10/zM GTP3,S, but receptors differed greatly in the maximum decrease of binding. The receptor, radioligand, and maximal inhibition were as follows: A~: [3H]CHA, 90; a2: [3H]clonidine, 43%; 5HT~A: [3H]8-OH-DPAT, 60%; data not shown]. In the presence of 10/zM GTP3,S, adenoregulin caused a further reduction of [3H]CHA binding to Al-adenosine receptors (Fig. 2A), [3H]8-OHDPAT to 5HT~A receptors (Fig. 2B) and [3H]clonidine binding to a2-receptors (Fig. 2C). Inclusion of millimolar concentrations of magnesium ions converts most receptors to a high affinity binding state. With the inclusion of 10 mM MgC12, the enhancement mediated by adenoregulin was greatly reduced for [3H]CHA binding to A~-receptors (Fig. 2A), abolished for [3H]8-OH-DPAT binding to 5HT~A-receptors (Fig. 2B) and little affected for [3H]clonidine binding to a2-receptors (Fig. 2C). Binding Profile to a Gs-Coupled Receptor. The binding of [3H]CGS21680 to A2a-adenosine receptors is usually performed in the presence of 10 mM MgC12 in order to both increase specific binding and to reduce nonspecific binding. Under these conditions, adenoregulin at 100/zM caused a 29% increase in [3H]CGS 21680 binding (Fig. 3A). The inclusion of 10/zM GTP3,S reduced [3H]CGS21680 binding to A2a-adenosine receptors by 53%. In the presence of 10/zM GTPTS, the adenoregulin had no effect on [3H]CGS21680 binding to Aza-adenosine receptors (Fig. 3a). When no magnesium ions were added to the assay, [3H]CGS21680 specific binding was reduced by about 80% and this was further reduced by about 25% by 10 tzM GTP3,S (Fig. 3B). With no added magnesium salts, adenoregulin caused a robust enhancement of [3H]CGS21680 binding to A2~-adenosine receptors (108% above control values at 200/zM), which was

Adenoregutin and Adenosine Receptors

4'71

4000.

3500,

3000,

'U

2500,

z t'~

2O00

z r~

.I'-

R -I-

1500,

1~ ~ 500

0 •001

.01

.1

1

10

100

1000

[ADENOREGULINI(Idd) r

g00o

B I1000

E

Q. "U

7000

=000

o z

¢3

SOOO

rrl p. ,<

4OO0

z

-r

3O0O

o, ao -r-

20O0

O3 1000

0

.01

.1

1

10

100

1000

[ ADENOREGULIN ] ( ldVI ) Fig. 1. Effect of adenoregulin on binding of [3H]agonists to (A) Al-adenosine (B) 5HT]A-serotonin and (C) c~2-adrenergic receptors in rat cerebral cortical membranes. Membranes were incubated at 25°C with 1 nM [3H]CHA (A]), 1 nM [3H]8-OH-DPAT (5HT1A) or 4riM [3 H]clonidine (a2), as described in Methods. Specific binding (O). Nonspecific binding (0). Data are expressed as means + SEM of 3 experiments, each with triplicate determinations.

Moni, Romero, and Daly

472

22S0 A

2000 17S0

z

o

/

tO

z

S -r

0

.001

.01

.1

1

[ ADENOREGUUN Fig. 1.

10

100

1000

] ( pJVl )

(Continued.)

unaffected by the presence of 10/zM GTPyS (Fig. 3B). ArAdenosine Receptor Binding

Binding Profiles. Adenoregulin at 20/zM caused a 60% enhancement of [3H]CHA binding to cortical Al-adenosine receptors, nearly twice that observed for the other receptors (Table I). Enhancement of [3H]CHA binding to membrane preparations from rat cerebellum, hippocampus and striatum was similar (Table I). By contrast, binding of the Al-selective antagonist [3H]DPCPX to rat cortical membranes was increased only 13%, being maximal at 2 ~M adenoregulin. Nonspecific binding of [3H]DPCPX was reduced by 70% at 200/zM of adenoregulin (Fig. 4). The effects of adenoregulin on [3H]CHA and Table I. The Effects of Adenoregulin in Binding of [3H]-Labelled Agonists to Receptors in Rat Brain Membranesa Thres-

Receptor A1

5HT]A c~2

A2a

Ligand CHA

8-OH-DPAT clonidine CGS 21680

hold

ECma x

Maximum stimulation

Brain region

(p.M)

ECs0 (~M)

(aM)

(% Control)

HIPPO CBM CTX STR CTX CTX STR

0.5 0.5 0.5 0.2 0.2 0.5 10

6 3 5 5 0.7 0.7 20

50 50 20 20 10 2 100

76 62 60 60 30 20 29

a Binding experiments were performed as described in Materials and Methods using hippocampus (HIPPO), cerebellum (CBM), cortex (CTX) and striatum (STR).

Adenoregulin and Adenosine Receptors

z

473

2000.

m

IS00 • "r e~

1000.

SO0. q 0 .01

.1

1

t0

100

1000

[ ADENOREGULIN ] ( I~M )

10000. B ~

0

v

I v

.

0

. . . . .

~

.......

.01

~

.1

.......

,

I

[ ~ENOREGUUN

.......

•

. . . . . .

40

~

100

J.....

....

1000

] (~ )

Fig. 2. Effect of adenoregulin, GTPyS and magnesium ions on specific binding of [3H]-agonists in rat cerebral cortical membranes. Effect of adenoregulin on specific binding was determined in the absence (©) or presence of 10~M GTPyS (0) or with 10mMMgC12 (z~). Membranes were incubated at 25°C with (A) 1 nM [3H]CHA (Aj), (B) 1 nM [3H]8OH-DPAT (5HT1A) or (C) 4 nM [~H]clonidine (ct2) as described in Methods. Data are expressed as means ± SEM of 3 experiments, each with triplicate determinations.

474

Monk Romero, and Daly

-t c ~50 • 2000-

1750

Z~"

1500, 1250-

z 0

~5o !

"r

s~.

.,J

250 "

0 ,001

.01

.1

1

10

100

1000

[ ADENOREGULIN ] ( ~ ) Fig. 2.

(Continued.)

[3H]DPCPX binding were studied at 12°C (180 min incubations), 25°C and 37°C (90 rain incubations). While incubation temperature clearly affected control [3H]CHA binding (Fig. 5A) and control [3H]DPCPX binding (Fig. 5B), the effects of adenoregulin on the agonist and the antagonist binding were unchanged. The enhancement of [3H]CHA binding mediated by adenoregulin was 100% at 12°C and 50 to 55% at 25°C and 37°C. For [3H]DPCPX, the enhancement of binding mediated by adenoregulin was about 5% at 12°C and about 10% at 25°C and 37°C. The enhancement of [3H]CHA binding caused by adenoregulin was unaffected by the inclusion of protease inhibitors--10txg/mL leupeptin, 10/xg/mL aprotinin, 10 gM soy bean trypsin inhibitor and 200 txM PMSF (data not shown). When cortical A]-adenosine receptors were solubilised (CHAPS), adenoregulin increased both the binding of the agonist [3H]CHA and the antagonist [3H]DPCPX by 120% (Fig. 6). The response curves for effects of adenoregulin on binding of the agonist and antagonist were almost identical. Saturation Experiments. In saturation experiments, adenoregulin at 5, 50, and 200 txM was coincubated with [3H]CHA and cortical membranes. Isotherms for [3H]CHA specific binding were monophasic both in the presence and absence of adenoregulin. These were transformed to Scatchard plots (Fig. 7A). Adenoregulin slightly increased the affinity of Al-adenosine receptors for CHA, the largest decrease in Ka being at 50txM adenoregulin. More pronounced was the increase in binding sites for CHA, the Bmax of [3H]CHA binding being 38% above control values at 200 txM adenoregulin. In other experiments, adenoregulin at 5, 50 and 200 IxM was preincubated with adenosine deaminase-treated cortical membranes for 90 rain at 25°C. Membranes were then washed twice in buffer by centrifugation (35,000 x g, 25 rain, 4°C), followed by saturation experiments with

Adenoregulin and Adenosine Receptors

475

8OOO

r'~ "0 v

2500

o Z

2000

_z o a:) (D

150o,

O3 10oo, I

r~ 500.

o

" .

. . . . . .

.

. . . . . . .

•01

.

. . . . . . .

.

,1

. . . . . . .

1

~

. . . . . . .

10

n

. . . . . .

100

1000

[ADENOREGUUN](IdVI) B 8,.,0 1

"U

°

_z ¢'~

S0O

i

ra

o -r

0

.01

.1

1

10

100

1000

[ ~)ENOREGUUN ] ( ~ ) Fig. 3. Effect of adenoregulin on specific binding of (3H]CGS21680 to A2~-adenosine receptors in rat striatal membranes. Membranes were incubated with 5nM [3H]CGS21680 for 120 min at 25°C in (A) the presence or (B) the absence of 10 mM MgC12 as described in Methods. Specific binding in the absence of 10p.M GTPTS (©). Specific binding in the presence of 10/.tM GTPTS (e). Data are expressed as means + SEM of 3 experiments, each with triplicate determinations.

476

Moni, Romero, and Daly 10000

II000, P, "ID

z z_

7000. 8000. 5000.

rr~ X

4000-

o 3000. 2000

•

1000.

0

.01

.1

.2

.5

~

2

$

10

20

60 1 0 0 2 0 0

[ ADENOREGULIN ] ( gM )

Fig,_a4, Effect of adenoregulin on the binding of [ H]DPCPX to Al-adenosine receptors in rat cerebral cortical membranes. Membranes were incubated with 0.2 nM [3H]DPCPX for 90 min at 25°C as described in Methods. Specific binding (O). Nonspecific binding (e). Data are expressed as means + SEM of 3 experiments, each with triplicate determinations.

[3H]CHA. Compared with control values, pretreatment with adenoregulin caused very similar changes to the Kd and Bmax as those observed in coincubation experiments (data not shown). The Kd was lowest in membranes pretreated with 50/zM adenoregulin, while the Bm~x was highest in membranes pretreated with 200 IzM adenoregulin, being increased by 54% (Table II). Saturation experiments were also performed with [3H]DPCPX coincubated with adenoregulin 2, 50, or 200/zM (Fig. 7B). In the presence of 2/zM adenoregulin, no change in receptor affinity for the radioligand was observed, but the Bmax was increased by 14%. At higher concentrations of adenoregulin, the Kd of [3H]-DPCX was increased by 30 to 40% and the Bmaxwas increased by 20% above the control value. Kinetics of[sH]CHA Binding. Adenoregulin up to 50/zM had no effect on the association rate of [3H]CHA binding to cortical Al-adenosine receptors (data not shown). The presence of adenoregulin at 10/zM caused an increase of [3H]CHA binding only after 15 to 20 min. Binding reached a plateau level both in the presence or absence of adenoregulin after about 60 min). Preincubation of membranes with adenoregulin for 90 min at 25°C, followed by two centrifugation washes, afforded a similar time course for binding of [3H]CHA (data not shown). Dissociation of [3H]CHA from cortical Al-adenosine receptors in the absence of adenoregulin appeared biphasic, having a rapid (t0.5, 1.8 min) and a slow component (t0.5, 33 rain) (see Table III). Coaddition of 10/~M adenoregulin with 10tzM 2-chloroadenosine made no difference to the dissociation of [3H]CHA from Al-adenosine receptors (data not shown). However, the rate 'of [3H]CHA dissociation from Al-adenosine receptors initiated with 2-chloroadenosine, was slowed when membranes had been preincubated with adenoregulin for 90 min at 25°C (data not shown). The [3H]CHA dissociation in the presence of 10/zM adenoregulin had a single slow component with a to.s of about 51 min (Table III). Effect of GTPyS and MgCI2 on [3H]CHA Binding. Membranes were preincubated with 5, 50 or 200/zM adenoregulin for 90 min at 25°C, prior to the addition of 1 nM [3H]CHA and GTPyS for a further 90 min incubation. GTPyS

Adenoregulin and Adenosine Receptors

477

4000-

Z~

~000-

0

....... 0

,01

i .1

.......

, 1

......

;i 10

......

~ ....... 100 1000

[ ADENOREGUUN l ( IJM ) 10000-

,B

B000

.~.

,ooo SOOO-

--.

0

.01

.1

1

t0

100

1000

[ ADENOREGUUN ] (iaM ) Fig. 5. Effect of temperature on the specific binding of [3H]CHA and [3H]DPCPX to Al-adenosine receptors from rat cerebral cortical membranes. Membranes were incubated at 12°C (A), 25°C (O) and 37"C (0) with either (A) [3H]CHA (lnM) or (B) [3H]DPCPX (0.2nM) in the presence of increasing concentrations of adenoregulin. Data are expressed as means :1:SEM of 3 experiments, each with triplicate determinations.

decreased [3H]CHA binding both in the absence and presence of adenoregulin. The maximal inhibition of [3H]CHA binding by GTPyS greater in the presence of adenoregulin (Fig. 8A). The presence of adenoregulin did not significantly alter the IC50 value for GTPyS. Washing of the membranes prior to addition of

478

Moni, Romero, and Daly 4000-

3SO0-

2500z

2ooo-.

.J

tr 500

¸

±

.

• 01

.1

.

.

.

.

.

1

.

.

_

~10

-

100

_

1000

[ ADENOREGULIN ] ( 14M ) Fig. 6. Effect of adenoregulin on binding of [3H]CHA and [3H]DPCPX to solubilised A Iadenosine receptors from rat cerebral cortical membranes. Solubilized receptors were incubated with 1 nM [3H]CHA (O, Q) or 0.2 nM [3H]DPCPX (A, A) at 25°C as described in Methods. Specific binding (open symbols). Nonspecific binding (closed symbols). Data are expressed as means .4. SEM of 2 experiments, each with triplicate determinations.

[3H]CHA and GTPTS, had no effect (data not shown). Similar experiments were performed in the presence of 10 mM MgC12. The presence of magnesium ions caused a 10-fold shift to the right in the GTPyS inhibition curve (control ICs0, 0.7/~M; IC50 with 10 mM MgC12, 6/~M). The coaddition of 50/~M adenoregulin with 10 mM MgC12 had no effect on this result (ICs0 with MgC12 plus 50 ~M adenoregulin, 6 ~M; Fig. 8B). Displacement of [3H]DPCPX Binding by an Agonist CHA. Adenoregulin enhanced the potency of CHA towards displacement of binding of the antagonist [3H]DPCPX from cortical Al-adenosine receptors (Fig. 9A). The percentage of Al-adenosine receptors in the high affinity state increased from 50% to 74% in the presence of 10 lzM adenoregulin (Table IV). With the inclusion of 10 p+M GTPTS, two affinity states of the receptor to CHA were defined. Seventeen percent of receptors were in the high affinity state (Ki 14/~M) and 83% in the low affinity state (K, 280#M). However, with the inclusion of 10 p.M GTPyS and 10 tzM adenoregulin, only one low affinity receptor state (Ki 180/~M) could be measured (Fig. 9B). Effect of NaCl. The enhancement of [3H]CHA binding caused by the presence of adenoregulin at 5, 50 or 100tzM was reduced by increasing the concentration of NaC1 and was abolished at about 300 mM NaC1 (Fig. 10A). With the inclusion of 150mMNaCI in the assay, adenoregulin increased binding of [3H]CHA to Al-adenosine receptors by only 25% with a maximal enhancement

Adenoregulin and Adenosine Receptors

479

A

1750i

~

~

15000

1250"

111

1000

A

er z O

rn

O

200

400

600

800

1000

t200

BOUND ( fmol/mg )

1600

B

1400 A 1200

~s E UJ 111

1000 t~00600-

z O m

4002000

0

100

200

100

400

100

800

?00

100

BOUND ( frnoVmg )

Fig. 7. Scatchard plots for specific binding of [3H]CHA and [3H]DPCPX to rat cerebral cortical membranes. Membranes were incubated with either (A) [3H]CHA (0.1-20 nM) without adenoregulin (O), with 5/zM adenoregulin (O), with 50/.tM adenoregulin (A) and with 200~tM adenoregulin (&) for 90 minutes at 25°C as described in Methods; or (B) [3H]DPCPX (0.02-10 nM) without adenoregulin (©), with 2 g M adenoregulin (O), with 50/.tM adenoregulin (A) and with 200/.tM adenoregulin (&) for 120 min at 25°C as described in Methods. Data are averaged results from 3 different experiments, each with triplicate determinations.

480

Moni, Romero, and Daly

Table 11. Effects of Adenoregutin on Binding of [3H]CHA to A1-Adenosine Receptors in Rat Cortical Membranes ° Coincubation

Control Adenoregulin 5/zM 50 p.M 200 ~.M

Pretreatment

K d (nM)

Bm~ (fmol/mg)

K d (nM)

Bmax (fmol/mg)

1.04 + 0.07

731 + 10

0.79 ± 0.09

452 ± 10

0.81 ± 0.05 0.62 ± 0.06 0.80 ± 0.08

874 ± 11 992 ± 19 1010 • 22

0.75 ± 0.07 0.36 ± 0.03 0.51 + 0.04

516 ± 9 617 ± 9 692 + 10

a Adenoregulin was incubated with [3H]CHA and membranes for 90 min at 25°C (coincubation) or membranes were incubated with adenoregulin for 90 min at 25°C before addition of [3H]CHA for a further 90 min (pretreatment). Data are expressed as means + SEM of three experiments, each with triplicate determinations.

at 50/zM (Fig. 10B). The ability of adenoregulin to decrease [3H]CHA binding in the presence of 10/zM GTP3,S was reduced when 154 mM NaC1 was added to the assay (ICso with 10/zM GTPTS, 3/zM; ICs0 GTPTS plus 150 mM NaC1, 40/zM; Fig. 10C). Comparison to the Allosteric Enhancer PD81,723. The binding characteristics of adenoregulin were compared with those of the aminobenzoylthiophene, PD81,723. This compound is known to be both an allosteric enhancer and an antagonist at Al-adenosine receptors (Bruns and Fergus, 1990; Bruns et al., 1990). PD81,723 at 20/zM increased [3H]CHA binding to Al-receptor~ in cortical membranes by 20% (Fig. 11). In the presence of 10mMMgC12, PD81,723 at 10/zM and higher inhibited [3H]CHA binding. In the presence of 100 p.M GTP or 10/zM GTP3,S binding of [3H]CHA was greatly reduced and PD 81,723 now increased binding by 200 to 300%. In keeping with its known antagonist properties (Bruns and Fergus, 1990), PD81,723 at about 5 ~M and above, inhibited [3H]DPCPX binding to Al-adenosine receptors (data not shown). When coincubated with 20/zM PD81,723, adenoregulin increased [3H]CHA binding by 26% with maximal enhancement occuring at 10/xM adenoregulin (data not shown). Binding o f [35S]GTp3,S to Brain Membranes. Adenoregulin at 50p.M increased the specific binding of [35S]GTP'yS to crude brain membranes by Table II1.

Effect of 10/zM Adenoregulin on the Rate of Dissociation of [3H]CHA from A~-Adenosine Receptors in Rat Brain Membranes °

Assay Control Adenoregulin

Model (number of sites) 2 1

k - i (rain -1) High 0.02 0.01

to.s (min)

Low

High

0.38

32.93

Low 1.81 50.85

° Rat cortical membranes and 1 nM [3H]CHA were preincubated with and without 10/.tM adenoregulin for 90 min at 25°C. Dissociation of [~H]CHA was initiated by the addition of 2-chloroadenosine to a final concentration of 10p.M. Results are the means of triplicate determinations and representative of three experiments.

Adenoregulin and Adenosine Receptors

481

4OOO

A

asoo

A

3000

E

z e'~

_z

2o00.

err

.~-

15o0.

•.r

Iooo500.

...... ~ .001

. ', "~, - 7

. . . . . .

.01

~

. . . . .

.1

~

. . . . .

1

~

10

. . . . . .

•

. . . .

100

1000

100

1000

[ G T P .f S ] ( f t M )

I=

L~-o0'

ea

t.9 _z a z

2000, 1500 '

<

"r" "r

looo

500

0

.001

.01

tGTP7

.1

1

S]

10

(pM)

Fig. 8. Effect of GTPyS on specific binding of [3H]CHA to Al-adenosine receptors in rat cerebral membranes. (A) Membranes were incubated in the absence of adenoregulin (O), with 5 ktM adenoregulin (0) or with 50/zM adenoregulin (&) or with 200/.tM adenoregulin (•) for 90 rain at 25°C as described in Methods. Membranes were then incubated with 1 nM [3H]CHA and increasing concentrations of GTPyS for a further 90 min at 25°C. Data are the results of one representative experiment with triplicate determinations. (B) Membranes were incubated in the absence of adenoregulin (O), with 10 mM MgC12 (A) and with 50/zM adenoregulin plus 10mMMgCI2 (0) for 90 min at 25°C. Membranes were then incubated with l nM [3H]CHA and increasing concentrations of GTPyS for a further 90 rain at 25°C. Data are expressed as means ± S E M of 3 experiments, each with triplicate determinations.

Moni, Romero, and Daly

482

approximately 45% (Fig. 12). CHA at 1 p~M increased the specific binding of [3SS]GTP3,S by 80% in the absence of adenoregulin. In the presence of CHA, adenoregulin at 50/zM, caused a further 23% increase in binding of [35S]GTPTS. Binding of [3H]CHA to DDTz MF-2 Cell Membranes. In membranes from DDT1 MF-2 smooth muscle cells, 20/zM adenoregulin increased the binding of [3H]CHA (Fig. 13A) and [3H]DPCPX (Fig. 13B) to Al-adenosine receptors by 17%. GTP3,S at 10/xM maximally reduced [3H]CHA binding to Ai-adenosine receptors by 54%. This was further decreased in the presence of adenoregulin. A~-Adenosine Receptor-mediated Inhibition of Cyclic AMP Formation in DDT~ MF-2 Cells. CHA inhibited forskolin-mediated accumulation of cyclic AMP in DDT~ MF-2 cells with an ICs0 of 0.5/xM. Adenoregulin at 20/zM had no significant effect on cyclic AMP accumulation in DDT1 MF-2 cells either when added either at the same time as the CHA or when cells were preincubated for 90 min with the peptide (data not shown). By contrast, PD81,723 at 20/zM potentiated the inhibition of cyclic AMP accumulation caused by CHA (data not shown). The IC50 for CHA was about 0.6/xM in the absence and about 0.25/xM in the presence of PD81,723.

DISCUSSION

The Al-adenosine receptor is one member of large super family of G-protein-coupled receptors consisting of homologous proteins sharing common structural motifs. Binding of agonists to transmembrane sites in such receptors cause conformational changes resulting in altered interaction of intracellular loops of the receptors with GDP-liganded G-proteins. Such altered interactions result in exchange of GDP for GTP and dissociation of the GTP-liganded G-protein from the receptor into a- and/3-/-subunits (Gilman, 1987). The GTPase activity of the a-subunit is increased and a resultant GDP-liganded or guanyl nucleotide-free subunit is able to reassociate with receptor to complete the cycle. Experimentally, it is possible to measure in membranes high and low affinity states for agonists, high affinity states for antagonists, density of binding sites for agonists and antagonists, and effects of agonists on binding of radiolabelled GTP3'S to G-proteins. The high affinity state of receptors for agonists is thought to be the receptor-G-protein complex in which either no guanyl nucleotide or GDP is bound to the G-protein. Addition of GTP or the non-hydrolyzable analog GTPyS results in uncoupling of the receptor from the G-protein yielding a low affinity state of receptors for agonists. Antagonist binding to receptors is little affected by receptor-G-protein coupling. Magnesium ions enhance coupling of receptor and G-protein and thereby promote the high affinity agonist binding state. Sodium ions reduce agonist but not antagonist binding, presumably by decreasing receptor-G-protein coupling. The effects of synthetic adenoregulin on binding of agonists to four different G-protein coupled receptors both in the absence or presence and GTPyS and magnesium ions were investigated with rat cortical membranes. For the AIadenosine receptor at which the degree of enhancement of agonist binding was

Adenoregulin and Adenosine Receptors

483

A

6.~00

A

E

8OOO

A

:t%.

5SO0

• O

O• • ~A O O

50004500

_z

c) z X e~

0;6o

-

4000-

•&O O

~O0-

•A

3OOO

O

e A

O

eA O • A 0

2000.'r

e6A ° • &O +

1500, IC00, 500+ 0

•

6 e e

a .01

.00!

,1

1

10

100

1000

10000

[(:::HA} (nM) 7g~

B

8OO0 i

O.

S000-

z_ z

4000-

r~

X

3000.

U 2000.

1000,

0

. . . . .

~

.001

. . . .

~

.01

. . . . .

~

.1

. . . .

~

. . . .

1

~

.

10

• - - 9

100

. . . .

~

1000

-

--

10000

[CHA] (nM) Fig. 9. Effect of adenoregulin and GTP3,S on the displacement of [3H]DPCPX by CHA from A1adenosine receptors in rat cerebral cortical membranes. (A) Membranes were incubated for 120 min at 25°C with 0.2nM [3H]DPCPX and increasing concentrations of CHA in the absence of adenoregulin (O), with 10 u.M adenoregulin (A) and with 20nM adenoregulin (@) as described in Methods. Data are expressed as the means ± SEM of 3 experiments, each with triplicate determinations. (B) Membranes were incubated for 120 rain at 25°C with 0.2nM [3H]DPCPX and increasing concentrations of CHA and with 10/zM GTPyS (@) and 10p.M GTPyS plus 10/.tM adenoregulin (A) as described in Methods. Data are expressed as means + SEM of 3 experiments, each with triplicate determinations.

484

Moni, Romero, and Daly

Table IV. Effectof 10/zM and 20 p.M Adenoregulin on Displacement of [3H]DPCPXby CHA from A1-Adenosine Receptors in Rat Brain Membranesa Bmax(fmol/mg) (% receptors)

Assay

Model (number of sites)

High

Low

High

Low

Control

2

0.91

50.50

GTPrS

2

13.78

281.05

Adenoregulin

2

0.85

71.66

390.81 (50.5) 119.61 (16.8) 621.31 (73.7)

383.07 (49.5) 592.33 (83.2) 221.10 (26.3)

Adenoregulin/GTPyS

1

Kt (nM)

149.38

875.53

a Rat cortical membranes were incubated for 120 min at 25°C with 0.2 nM [3H]DPCPX and CHA without and with 10#.M GTPyS, with 10tzM adenoregulin and with 10/zM adenoregulin plus 10/.tMGTPyS. Data are expressed as means of three experiments, each with triplicate determinations.

the greatest, the effects of adenoregulin were investigated in detail. Effects of adenoregulin on agonist binding to A~-adenosine receptors in membranes from DDT1 MF-2 cells and on A~-receptor-mediated inhibition of cyclic AMP accumulation in the same cells were also probed. The Al-adenosine receptor couples to both Gn.3 and Go proteins. Two further receptors, namely the aE-adrenergic and the 5HT1g-receptor that couple to G~ receptors were investigated (Fig. 1). Adenoregulin stimulated agonist binding to all three of these receptors with the greatest enhancement (60%) at the Aradenosine receptor, and lesser enhancement (20-30%) at the other two receptors. Maximal enhancement occurred at 1 0 - 2 0 ~ M adenoregulin for the A~-adenosine and 5HT1g-serotonin receptors, while maximal enhancement occurred at 2/xM for the a2-receptor. An inhibitory leg to the dose response curve for adenoregulin was clearly manifest at 200/xM for the A~-adenosine receptor, 10/xM for the a2-receptor and 50/xM for the 5HT~g-receptor. Such inhibition, presumably by a second site/mechanism, could clearly affect the maximal stimulation observed at the receptors. The results suggest that adenoregulin will enhance binding of agonists to many Gi/G0-coupled receptors and that the dose-response relationships for such enhancement will vary from receptor to receptor. Magnesium ions only slightly enhance binding of agonist to Al-adenosine receptors perhaps reflective of already tight coupling of these receptors with G-proteins in brain membranes (Munshi and Linden, 1989). However, in the presence of magnesium, adenoregulin had greatly reduced effects on agonist binding with no significant effects at the 10/xM concentration that had caused maximal enhancement in the absence of magnesium. G T P y S caused 80% inhibition of agonist binding to Al-adenosine receptors in the absence of adenoregulin, but caused virtual complete inhibition in the presence of 20 p,M adenoregulin. A GTP~/S-resistant component of high affinity

Adenoregulin and Adenosine Receptors

485

4OO0

360O

3O0O

2500-

z_

+

2000-

i1 10oo

5

,;o

0

=~o [ NaCI ] mM

s;*

400

25OO St250,

~2000.

E

IZ "O

17~. 15~.

¢3

z 2:

1250' 1000'

"r O3

500, 2S0,

0

. . . . . . . . ......~ -.-..-..~ -7 .--..-5 ~..-.--+~ " .01

.1

t

10

100

. 1000

[ ADENOREGULIN ] ( Idd ) Fig. 10. Effect of sodium ions on specific binding of {3H]CHA to Al-adenosine receptors to rat cerebral cortical membranes. (A) Membranes were incubated with 1 nM [3H]CHA and NaCI (50-400 mM) without adenoregulin (O), with 5/~M adenoregulin (e), with 50/~M adenoregulin (&) and with 100/zM adenoregulin (A) for 90 min at 25°C. Data are expressed as means + SEM of 3 experiments, each with triplicate determinations. (B) Membranes were incubated with l nM [3H]CHA and 154mMNaCI and increasing concentrations of adenoregulin for 90 minutes at 25°C. Specific binding (O). Nonspecific binding (0). (C) Membranes were incubated for 90 minutes at 25°C with 1 nM [3H]CHA and increasing concentrations of adenoregulin with 10/zM GTPyS (O), and with 10 p.M GTPTS plus 154mM NaCI (O).

486

Moni, Romero, and Daly

A

E ~3 z_ C~ _Z

250. I

'X

200.

m

<: -r-

lS0,

"r 100.

.SO

0

. . . . . . .

•

. . . . . . .

•1

•

. . . . . . . .

I

J

. . . . . . . .

10 '

,

.

100

.

,w,,u,L . .

1000

[ ADENOREGULIN ] ( p.M )

(Continued.)

Fig. 10.

275O

1750

25O

0

. . . . . .

~

.OI

. . . . . . . . . . . . . . .

.t

,

1

.......

;

10

......

~

100

. . . . . .

1000

[ PD81,723 ] ( pM ) Fig. 11. Effect of PD81,723 on binding of [3H]CHA to A]-adenosine receptors in rat cerebral cortical membranes. Membranes were incubated for 90 min at 25°C with 1 nM [3H]CHA and increasing concentrations of PD81,723 (©) as described in Methods. Further additions included 10mMMgCI 2 (&), 100/.tM GTP (0) and 10 ~ M GTPTS (A). Nonspecific binding, determined with the inclusion of 10/zM 2-chloroadenosine, was not altered. Data are expressed as means ± SEM of 3 experiments, each with triplicate determinations.

Adenoregulin and Adenosine Receptors

487

a0000,

200OO

O z z_

15000

O~ pQ.

6000

0

.01

.1

1

10

100

1000

[ ADENOREGUUN l ( I,tM ) Fig. 12. Effect of adenoregulin on the binding of [35-S]GTP-yS to rat cerebral cortical membranes. Membranes were incubated with 0.6nM [35S]GTP'),S and increasing concentrations of adenoregulin without CHA (©, @) or in the presence of 1/zM CHA (A, A) as described in Methods. Specific binding (open symbols). Nonspecific binding (closed symbols). Data are expressed as means + SEM of 3 experiments, each with tripdcate determinations.

agonist binding has been observed for many receptors and has not been adequately explained (Prater et al., 1992). Agonist binding at 5HTIA receptors is markedly enhanced by magnesium and adenoregulin has no enhancing effect in the presence of magnesium (Fig. 2B). The inhibitory leg is also reduced with only a slight inhibition being seen at 200/zM adenoregulin. GTP3,S caused only a 60% inhibition of agonist binding to 5HT1A receptors in the absence of adenoregulin. In the presence of increasing concentrations of adenoregulin, the GTP3~S inhibition of agonist binding increased to near 90%. Agonist binding at a2-receptors was only marginally affected by magnesium and adenoregulin still caused a marked stimulation of agonist binding (Fig. 2C). Indeed, the major effect of magnesium was to reduce the inhibitory leg of the adenoregulin dose-response curve. GTPTS caused about a 40% inhibition of agonist binding to t~2-receptors, which was increased by adenoregulin. However, the inhibitory leg of the adenoregulin dose-response curve in the absence of GTP~/S complicates any interpretation of this result. The effects of adenoregulin on agonist binding to the three receptors that couple to Gi/G0 proteins are similar, but not identical. In all cases, two effects are seen, stimulation of agonist binding with thresholds at 1 ~M or less and an inhibition first manifest at from 20 to 200/zM, dependent on the receptor. Magnesium virtually eliminated adenoregulin-elicited stimulation of agonist binding at Al-adenosine and 5HTtm-serotonin receptors, while having no effect

488

Moni, Romero, and Daty 2000-

A

~750 -

A

lS00

E

O,. 'ID v

12S0

_z c~

_z <~ ,,.r "r

0

.01

,1

I

10

100

1000

t00

IO00

[ ADENOREGULIN ] ( ~ ) 5000-

4000

ED . '10

........

~00

c0 z

3000-

z i:o x o. to KL D 2:: ¢,:,

2500 -

20001500o

1000-

500-

O

.01

.1

1

10

[ ADENOREGULIN ] ( pM ) Fig. 13. Effect of adenoregulin on binding of [3H]CHA and [3H]DPCPX to Aradenosine receptors in membranes from DDT] MF-2 cells. Membranes were incubated at 25°(:: with either (A) [3H)CHA (1 nM) in the absence (O) or presence of 10/xM GTP3,S (0) or with 10 mM MgCl2 (A) with increasing concentrations of adenoregulin, or (B) [3H]DPCPX (0.2 nM) with increasing concentrations of adenoregulin, as described in Methods. Specific binding (O). Nonspecific binding (0). Data are expressed as means ± SEM of 3 experiments, each with triplicate determinations.

Adenoregulin and Adenosine Receptors

489

on stimulation of agonist binding at a2-receptors. The inhibition by adenoregulin appeared reduced in the presence of magnesium at all. three receptors. At all three receptors adenoregulin increased the magnitude of GTPTS-elicited conversion of receptors to a low affinity state. The similar response profiles for the three receptors suggest that adenoregulin modulates agonist binding by a common mechanism, perhaps involving coupling of receptor and Gi/G0-protein. The stimulatory phase might represent an enhanced coupling, while the inhibitory phase might represent a reduction in coupling caused by high concentrations of adenoregulin. Only one receptor-coupled to Gs proteins was investigated. This was the A2aadenosine receptor present in high density in rat striatal membranes (Bruns et al., 1986). Preliminary studies had indicated no significant effect of natural adenoregulin on binding of the agonist [3H]N-ethylcarboxamidoadenosine to this receptor (Daly et al., 1993). With the more selective agonist [3H]CGS21680, a significant enhancement was observed with synthetic adenoregulin. However, higher concentrations of adenoregulin were required for maximal enhancement of binding (Fig. 3A). An inhibitory leg was not observed, but perhaps that also requires higher concentrations of adenoregulin than those required for receptors coupled to Gi/Go-proteins. These results were obtained in the presence of magnesium, an almost obligatory requirement for high affinity agonist binding to receptors coupled to Gs-proteins. Thus, unlike Al-adenosine and 5HTtAserotonin receptors, adenoregulin can cause enhancement in agonist binding to A2~-adenosine receptors even in the presence of marked enhancement by magnesium. In the absence of magnesium, agonist binding to A2a-adenosine receptors is reduced by nearly 10-fold (Fig. 3B). Remarkably, non-specific binding also appears reduced in the absence of magnesium. Under these conditions adenoregulin enhanced agonist binding by 200% with maximal effect apparently not reached even at 200 ~M adenoregulin. The enhancement is relatively small compared to the enhancement caused by magnesium. It would appear that adenoregulin is less potent and perhaps less efficacious in affecting agonist binding when receptors coupled to Gs-proteins are involved, compared to effects on receptors coupled to G~ proteins. The estimated threshold, ECso, Efmax and maximal stimulation for each receptor is presented in Table I. The maximal stimulation of agonist binding to Al-adenosine receptors varied only slightly from 60-70% in four brain regions. The effects of adenoregulin on receptor/G protein interactions now was examined in detail for the Al-adenosine receptor, where the effects were greatest in magnitude. Antagonist binding, effect of solubilization, Rosenthal (Scatchard) analysis, effects on kinetics of on- and off-rates of binding, effects on GTPTS and sodium ion inhibition of agonist binding, and combination with the A~-receptor allosteric enhancer PD81,723 were all examined. A slight but significant enhancement of binding by adenoregulin of the selective Al-receptor antagonist, [3H]DPCPX was found. The maximal enhancement occurred at 2/a.M adenoregulin and was only 13% (Fig. 4). It should be noted that adenoregulin significantly reduced nonspecific binding of [3H]DPCPX. Antagonist binding at A~-receptors is relatively unaffected by, guanine

490

Moni, Romero, and Daly

nucleotides (Goodman et al., 1982). Thus, the slight effect of adenoregulin on antagonist binding would suggest a direct action on the receptor. The effect of temperature on adenoregulin-eIicited changes in agonist and antagonist binding was determined (Fig. 5). CHA binding was greatest at 25°C, but the effects of adenoregulin were similar at 12, 25 and 37°C. Stimulatory effects of adenoregulin on antagonist binding were minimal except at 37°C. Rosenthal analysis of both agonist and antagonist binding was conducted. Adenoregulin causes a marked increase in the Bm~xfor agonist binding with marginal effects on the Kd (Fig. 7A, Table II). In contrast at concentrations of 50/zM, adenoregulin has complex effects on antagonist binding--increasing Bronx, but also increasing the I~ (Fig. 7B, Table II). The time course for binding of agonist to Al-adenosine receptors appeared little affected by adenoregulin, but the dissociation of agonist was significantly retarded (Table III). Inhibition of antagonist binding by the agonist CHA was significantly altered by adenoregulin as summarized in Table IV. Adenoregulin increased the percent of receptors with high affinity for the agonist CHA from 50% to 74%, consonant with an enhanced formation of receptor-G-protein complexes. Only low affinity receptors were present when GTPyS was present in addition to adenoregulin. Adenoregulin appeared to bind strongly to brain membrane preparations in that it was not possible, even after thorough washing of the tissue, to demonstrate a reduction in the actions of the peptide. The time course for enhancement of agonist binding to Aradenosine receptors was slow. These findings suggest slow, but near irreversible interactions of adenoregulin with membranes. It appears most likely that the peptide is slow to partition into the membrane. However, it remains possible that interactions of adenoregulin with some critical component that regulates agonist binding to receptors, for example, some form of the receptor-G protein complex, is rate-limiting. Solubilization of A~-adenosine receptors from brain membranes had little effect on adenoregulin-elicited enhancement of agonist binding (Fig. 6). It is known that Al-adenosine rec@tors remain complexed with G-proteins after solubilization (Munshi and Linden, 1989). Unexpectedly, antagonist binding was now markedly enhanced by adenoregulin in solubilized preparations. An explanation is not apparent. The enhancement in agonist binding to As-adenosine receptors by adenoregulin was fully reversed by GTP3,S (Fig. 8A). Thus, if adenoregulin promotes a high affinity state by enhancing receptor-G-protein coupling this can be readily reversed by GTP3,S. The ICso for GTP3,S with respect to agonist binding was not affected by adenoregulin. In contrast, magnesium, which enhances agonist binding by increasing receptor-G-protein coupling does result, as previously reported by others (Lohse et al., 1984), in an increase in the ICs0 of GTP~/S for inhibition of agonist binding (Fig. 8B). Adenoregulin has no effect on the magnesium-elicited increase in the ICs0 of GTPTS. The contrast between lack of effect of adenoregulin and the marked effect of magnesium on potency of GTPTS as an inhibitor of agonist binding is surprising if both agents enhance binding by promoting receptor-G-protein coupling. It is possible that the ability of adenoregulin to ehhance binding of GTPyS (Figure 12), offsets the enhanced stability of a receptor-G-protein complex.

Adenoregulin and Adenosine Receptors

491

Sodium ions are known to reduce agonist binding to Al-adenosine receptors (Lohse et al., 1984). Like GTP3,S, sodium ions could eliminate adenoregulinelicited increases in agonist binding to Aradenosine receptor (Fig. 10). However, sodium ions at 154mM did not completely eliminate adenoregulin-elicited enhancement of agonist binding (Fig. 11B), nor did 154 mM sodium ions prevent the ability of adenoregulin to convert all Al-adenosine receptors into a GTP3,S-sensitive state. The effects of adenoregulin on agonist binding to Al-adenosine receptors are markedly different than the effects of the allosteric enhancer PD81,723. The effects of PD81,723 appear greater in the presence of GTP or GTP3,S (Fig. 11), suggesting the enhancement represents a direct effect on receptors. In the presence of PD81,723, adenoregulin had nearly the same effects on agonist binding as it did in the absence of PD81,723. The effect of adenoregulin on agonist binding to Al-adenosine receptors was also examined in membranes from DDT~ MF-2 cells. In contrast to the robust effects in brain membranes, adenoregulin caused only a small enhancement in DDT~ MF-2 membranes with maximal of enhancement 13% and 17% observed at 1 IzM and 20/xM adenoregulin, respectively, followed by an inhibitory leg (Fig. 13A). Antagonist binding was enhanced to a similar small extent (Fig. 13B). Effects of adenoregulin on A~-receptor-mediated inhibition of cyclic AMP accumulation in DDT~ MF-2 cells was investigated. There was no effect on CHA-elicited inhibition by adenoregulin, in marked contrast to the increased potency of CHA seen in the presence of the allosteric agent PD81,723. Thus, adenoregulin does not function as an allosteric enhancer in DDT1 MF-2 cells, nor does it block A~-receptor-mediated inhibition of adenylate cyclase. In toto, the results suggest that the effects of the amphiphilic peptide adenoregulin on agonist binding to G-protein-coupled receptors is dependent both on the receptor and on the nature of the G-Protein i.e. Gi/Go versus Gs. The effects are also dependent on the source of the receptor with effects being very robust for Al-receptors in rat brain membranes and being minimal in DDT~ MF-2 membranes. It is noteworthy that the degree to which adenoregulin enhances agonist binding to a receptor shows a strong correlation ( r = 0.82) with the maximal inhibition of agonist binding elicited by GTP3'S i.e., receptors like the brain Al-receptor that appear tightly coupled to G-proteins show the greatest enhancement of agonist binding by adenoregulin. The effects of adenoregulin are biphasic with thresholds for enhancement of binding of agonists to receptors coupled to G~/G0 proteins occurring at 1/zM or less, while inhibitory effects were first detected at 50-200/zM. The enhancement is due to an increase in total number of high affinity binding sites for agonists. There appear to be minimal effects on antagonist binding to A~-receptors (Daly et al., 1993 and this paper) or of antagonist binding to muscarinic or opioid receptors (Daly et al., 1993). The remarkable enhancement of specific binding of both agonists and antagonists to A~-receptors after solubilization is not explicable and deserves further study. In membranes, adenoregulin not only enhances agonist binding, but appears to convert all high-affinity agonist binding to a state sensitive to inhibition by GTP3,S.

Moni, Romero, and Daly

492

A tentative explanation for the effects of adenoregulin on agonist binding to receptors is that adenoregulin, perhaps in part through enhancement of guanyl nucleotide exchange, converts all receptors into a complex with guanyl nucleotide-free G-protein. Such a complex is known to have high affinity for agonists (Gilman, 1987). Such a complex would then be dissociable on binding of GTP3,S, thereby, resulting in free receptor with a low affinity for agonists. Mastoparan, another amphiphilic cationic peptide, has similar effects to adenoregulin on agonist binding to receptors (Shin et a t , 1994) and presumably enhances binding by a similar mechanisms. The inhibitory effects of adenoregulin on agonist binding observed at higher concentrations of adenoregulin appear likely to be closely related to the stimulatory effects. Perhaps at higher concentrations, adenoregulin through further binding to G-proteins prevents receptor-G-protein complexes from forming. The effects of adenoregulin do not appear readily reversible, nor does it appear to act as an allosteric enhancer or agonist at least with respect to adenylate cyclase inhibition in DDT1 MF-2 cells. Both adenoregulin and mastoparan do cause phosphoinositide breakdown in intact cells, but in membranes both inhibit phospholipase C (Shin et al., 1994). Adenoregulin does appear to be a useful tool for further investigation of receptor-G-protein interactions.

REFERENCES Abbrachio, M. P., Cattabeni, F., Fredholm, B. B., and Williams, M. (1993). Purinoceptor nomenclature: A status report. Drug Develop. Res. 28".207-213. Amiche, M., Ducancel, F., Lajeanesse, E., Boulain, J-C., Menez, A., and Nicolas, P. (1993). Molecular cloning of a cDNA encoding the precursor of adenoregulin from frog skin. Relationships with the vertebrate defensive peptides, dermaseptins. Biochem. Biophys. Res. Commun. 191:983-990. Bruns, R. F., and Fergus, J. H. (1990). Allosteric enhancement of adenosine A~ receptor binding and function by 2-amino-3-benzoylthiophenes. Mol. PharmacoL 38:939-949. Bruns, R. F., Lu, G. H., and Pugsley, T. A. (1986). Characterization of the A 2 adenosine receptor labeled by [3H]NECA in rat striatal membranes. MoL Pharmacol. 29:331-346. Bruns, R. F., Fergus, J. H., Coughenour, L. L., Courtland, G. G., Pugsley, T. A., Dodd, J. H., and Tinney, F. J. (1990). Structure-activity relationships for enhancement of adenosine A~ receptor binding by 2-amino-3-benzoylthiophenes. MoL Pharmacol. 38:950-958. Cooper, D. M. F., Londos, C., and Rodbell, M. (1980). Adenosine receptor-mediated inhibition of rat cerebral cortical adenylate cyclase by a GTP-dependent process. MoL Pharmacol. 18:598-601. Daly, J. W., Butts-Lamb, P., and Padgett, W. (1983). Subclasses of adenosine receptors in the central nervous system: Interaction with caffeine and related methylxanthines. Cell. Mol. Neurobiol. 3:69-80. Daly, J. W., Caceres, J., Mori, R. W., Gusovsky, F., Moos, M., Jr., Seamon, K. B., Milton, K., and Myers, C. W. (1993). Frog secretions and hunting magic in the upper Amazon: Identification of a peptide interacting with an adenosine receptor. Proc. Natl. Acad. Sci. USA 89:10960-10963. Delahaunty, T. M., Cronin, M. J., and Linden, J. (1988). Regulation of GH3-cell function via adenosine A~ receptors. Inhibition of prolactin release, cyclic AMP production and inositol phosphate generation. Biochem. J. 255:69-77. Erspamer, V., Melchiorri, P., Falconieri-Erspamer, G., Montecucchi, P. C., and De Castiglione, R. (1985). Phyllomedusa skin: A huge factory and store-house of a variety of active peptides. Peptides 6:7-12. Gilman, A. G. (1987). G proteins: Transducers of receptor-generated signals. Annu. Rev. Biochem. 58:615-649. Goodman, R. R., Cooper, M. J., Gavish, M., and Snyder, S. H. (1982). Guanine nucleotide and

Adenoregulin and Adenosine Receptors

493

cation regulation of the binding of [3H]cyclohexyladenosine and [3H]cyclohexyladenosine and [3H]diethylphenylxanthine to adenosine A t receptors in brain membranes. Mol. Pharmacol. 21:329-335. Hausleithner, V., Freissmuth, M., and SchUtz, W. (1985). Adenosine-receptor-mediated stimulation of Iow-Km GTPase in guinea-pig cerebral cortex. Biochem. J. 232:501-504. Higashijima, T., Uzu, S., Nakijima, T., and Ross, E. M. (1988). Mastoparan, a peptide toxin from wasp venom, mimics receptors by activating GTP-binding regulatory proteins (G proteins). J. Biol. Chem. 263:6491-6494. Higashijima, T., Burnier, J., and Ross, E. M. (1990). Regulation of Gi and Go by mastoparan, related amphiphilic peptides, and hydrophobic amines. J. Biol. Chem. 265:14176-14186. Jacobson, K. A., van Galen, P. J. M., and Williams, M. (1992). Adenosine receptors: Pharmacology, structure-activity relationships, and therapeutic potential. J. Med. Chem. 35:407-422. Lohse, M. J., Lenschow, V., and Schwabe, U. (1984). Two affinity states of R i adenosine receptors in brain membranes. Analysis of guanine nucleotide and temperature effects on radioligand binding. Mol. Pharmacol. 2.6:1-9. Lorenzen, A., Fuss, M., Vogt, H., and Schwabe, U. (1993). Measurement of guanine nucleotidebinding protein activation by A1 adenosine receptor agonists in bovine brain membranes: Stimulation of guanosine-5'-0-(3-[35S]thio)triphosphate binding. Mol. Pharmacol. 44:115-123. Mor, A., Amiche, M., and Nicolas, P. (1994). Structure, syntheses, and activity of dermaseptin b, a novel vertebrate defensive peptide from frog skin: Relationship with adenoregulin. Biochemistry 33:6642-6650. Mousli, M., Bueb, J-L., Bronner, C., Rouot, B., and Landry, Y. (1990). G protein activation: A receptor-independent mode of action for cationic amphiphilic neuropeptides and venom peptides. TiPS 11:358-362. Munshi, R., and Linden, J. (1989). Co-purification of A 1 adenosine receptors and guanine nucleotide-binding proteins from bovine brain. J. Biol. Chem. 264:14853-14859. Munshi, R., Pang, I-H., Sternweiss, P. C., and Linden, J. (1991). Al-adenosine receptors of bovine brain couple to guanine nucleotide-binding proteins G,1, Gi2 and G o. J. Biol. Chem. 266:2228522289. Prater, M. R., Taylor, H., Munshi, R., and Linden, J. (1992). Indirect effect of guanine nucleotides on antagonist binding to A~ adpenosine receptors: Occupation of cryptic binding sites by endogenous vesicular adenosine. Mol. Pharmacol. 42:765-772. Salvatore, C. A., Jacobson, M. A , Taylor, H. E., Linden, J., and Johnson, R. G. (1993). Molecular cloning and characterization of the human A 3 adenosine receptor. Proc. Natl. Acad. Sci. USA 90:.10365-10369. Shin, Y., Moni, R. W., Lueders, J. E., and Daly, J. W. (1994). Effects of the amphiphilic peptides mastoparan and adenoregulin on receptor binding, G proteins, phosphoinositide breakdown, cyclic AMP generation, and calcium influx. Cell. Mol. NeurobioL 14:133-158. Tawfik-Schlieper, H., Klotz, K-N., Kreye, V. A. W., and Schwabe, U. (1989). Characterization of the K*-channel-coupled adenosine receptor in guinea pig atria Naunyn-Schmiedeberg's. Arch. Pharmacol. 340:.684-699. Trussel, L. O., and Jackson, M. B. (1987). Dependence of an adenosine-activated potassium current on a GTP-binding protein in mammalian central neurons. J. Neurosci. 7:3306-3316. Van Calker, D., MUller, M., and Hamprecht, B. (1978). Adenosine inhibits the accumulation of cyclic AMP in cultured brain cells. Nature 276:839-841. Zhou, Q-Y., Li, C., Olah, M. E., Johnson, R. A., Stiles, G. L., and Civelli, O. (1992). Molecular cloning and characterization of an adenosine receptor: The A 3 adenosine receptor. Proc. Natl. Acad. Sci. USA 89:.7432-7436.