Neurochem. Int. Vol. 14, No. 4, pp. 413-417, 1989 0197-0186/89 $3.00 + 0.00 Printed in Great Britain. All rights reserved Copyright © 1989 Pergamon Press plc

CONUS VENOM INTERACTION WITH ~2-ADRENERGIC RECEPTORS IN CALF RETINA MEMBRANES*

EVA CZERWIEC l, WERNER DE POTTER 2, ANDRE CONVENTS 1 and GEORGES VAUQUELIN I tDepartment of Protein Chemistry, Institute for Molecular Biology, Free University Brussels and

2Department of Neuropharmacology, Universitaire Instelling Antwerpen, Belgium

(Accepted 18 November 1988)

A~traet----a2-Adrenergic receptors were identified in calf retina membranes by the specific binding of the radiolabelled antagonist [3H]RX 781094. Crude venoms from various Conus species did not interact with the radioligand but were able to inhibit radioligand binding to the *t2-receptors with the following order of potency: C. planorbis (IC50 = 2.1/~g protein/ml) ~ C. tessulatus (IC50 = 2.7) > C. eburneus (ICs0 = 19) > C. textile (IC50 = 54) > C. geographus (IC50 = 130). Venom from 17 other species was less or not active at all. Venom competition binding curves were steep and not affected by GTP. In contrast, the (-)-epinephrine competition binding curve was shallow and underwent a rightward shift and steepening in the presence of GTP. The venom-~2-receptor interaction was completely inhibited by the calcium chelating reagent EGTA. These data suggest that the venom of certain Conus species contains peptide toxins which are capable of shielding the binding site of ~2-receptors in an antagonistic manner.

Plant and animal toxins have proved to be extremely useful in defining key components of vital physio- logical systems, especially those related to nervous transmission. Recently, much attention has been focused on the action mechanism of peptide toxins isolated from Conus venom. Conidae form a large family of about 300 members of marine snails (Olivera et al., 1988). They possess potent venoms which are used primarily in the capture of prey organisms such as worms, other molluscs and fish. Interest in these venoms originally arose from the fact that Conidae can also use their venom apparatus defensively. In this respect, piscivorous species such as C. geographus and also molluscivorus species such as C. textile pose a serious threat to vertebrates, including man (Endean and Rudkin, 1965).

Two main strategies have been adopted for the initial identification of active toxins in Conus venom. Kobayashi et al. (1982a) compared the effect of the crude venoms of 29 different species upon the contractile status of muscle preparations such as diaphragm, atria, ileum and aorta. Alternatively, Olivera et al. (1985) investigated the physiological

*Parts of the results of this refereed paper were presented at the European Society of Neurochemistry in a workshop on Extracellular and lntracellular Messengers in the Retina held in G6teborg, Sweden on 12-17 June 1988. The workshop was organised by Dr N. N. Osborne, Oxford University, England.

properties of various peptide toxins purified from the venoms of C. geographus and C. magus. These experi- ments revealed that the venom of each Conus species contains a range of toxins with different specificities of action, and that the nature and occurrence of these toxins is species-dependent. So far, most of the attention has been devoted to the interaction between conotoxins and cation conductance channels. This has resulted in the discovery of a highly specific label for calcium channels: og-conotoxin (Abe and Saisu, 1987).

The presence of vasopressin-like peptides and of nicotinic acetylcholine receptor antagonists in Conus venom has also been documented (McManus et al., 1981; Olivera et al., 1985), indicating that part of the conotoxins might act directly at the level of hormone and neurotransmitter receptors. Since most of the previous research encompassed isolated organ experi- ments, the effects of such receptor-binding toxins in crude venom might have been masked by those directed against cation conductance channels. This potential problem can be overcome by the use of radioligand binding experiments, whereby each receptor type can be identified directly in cell membrane preparations by binding of a specific radioligand. Here, we report the ability of conotoxins to interfere with the interaction of the tritiated antagonist [3H]RX 781094 ([3H]idazoxan) with the ~2-adrenergic receptors in calf retina membranes.

413

414 E\ ,', (;/}:Rxvll:c ~'/ at.

EXPERIMENTAL PROCEDt/RES

( "hemica/s [3H]RX 781094 ([3H]idazoxan, 57 Ci/mmol) was obtained

from Amersham. Phentolamine hydrochloride was a kind gift from Ciba Geigy. (-)-Epinephrine bitartrate and ethyleneglycol - bis- (fl- aminoethylether)- N,N,N',N' - tetra - acetic acid (EGTA) were purchased from Sigma and gua- nosine triphosphate (GTP) from Boehringer, Mannheim.

Conus venom preparations Specimens of C. aulicus, C. capitaneus, C. consors, C.

flavidus, C. eburneus (var. polyglotta), C. ,/hrvus, C. gener- alis, C. geographus, C. gloriamaris, C. imperialis, C. leop- ardus, C. litteratus, C. lividus, C. marmoreus, C. miles, C. mustellinus, C. planorbis, C. striatus, C. textile, C. vexillum and C. virgo were obtained live from Cebu, the Philippines, and specimens of C. tessulatus were obtained from the Seychelles, then frozen and shipped in dry ice via air to Brussels where they were stored at -20~'C until use. The following steps were carried out at O-4C. The venom ducts were dissected from the animals, homogenised in 100 vol of 30 mM ammonium acetate with a Polytron mixer and then sonicated 3 times for 10 s in a Soniprep 150 sonicator. The suspensions were centrifuged at 9000g for 10 rain and the resulting supernatants stored at -20 'C . The protein con- centrations were determined according to the method of Lowry et at. (1951) using bovine serum albumin as the standard.

Retina membrane preparations Calf eyes were obtained and dissected in a local slaughter-

house. All subsequent steps were performed at 0-4'C. Retinas were homogenised in 10vol of 10ram Tris-HCl (pH 7.5)/10 mM MgClff0.25 M sucrose with a motor-driven Potter-Elvehjem homogeniser (10 strokes at maximum speed). The homogenate was centrifuged at 2000g for 15 rain. The pellet was homogenised in sucrose buffer and recentrifuged at 2000g. All supernatants were pooled and centrifuged at 29,000 g for 20 min. The resulting pellets were washed 3 times by centrifugation in 10mM Tris-HCI (pH 7.5)/10 mM MgC12, suspended in the same buffer con- taining 10% (v/v) glycerol and stored in liquid nitrogen.

Binding of [3H]RX 781094 Retina membranes (0.25~.5 mg protein/ml) were incu-

bated for 15min at 3TC with 6nM [3H]RX 781094 in 50 mM Tris-HCl (pH 7.5)/10 mM MgCI,, in a final volume of 250 #1. At the end of the incubation, the samples were diluted in 4 ml of ice-cold buffer and filtered under reduced pressure through a glass fibre filter (Whatman GF/B, 2.5 cm dia). Filters were washed rapidly 4 times with 4 ml of ice-cold buffer, placed in polyethylene scintillation vials with I ml of 0. I N NaOH and 8 ml of scintillation fluid (picofluor 15 from Packard) and counted in a Packard liquid scintil- lation spectrometer. Determinations were performed in duplicate for each experiment. Specific binding was obtained by subtracting the nonspecific binding (i.e. binding in the presence of 10/~M phentolamine) from the total binding. Nonspecific binding was 5-10% of total binding. Results are expressed as means and SEM of 3 experiments.

RESI,:I,TS

~2-Adrenergic receptors were identilied in calf ret ina membranes by the specific b inding of the radiolabelled an tagonis t [~H]RX 781094 (Convents et al., 1987a,b).

Screening for the presence of c~2-receptor binding componen t s in (;.onus venom consisted of evaluat ing the ability of increasing concent ra t ions of' crude venom to inhibi t [3H]RX 781094 binding to the receptors. Figure l, which compares the venoms of C. tessulatus, C. textile and C. aulicus, clearly illus- trates the marked species difference in venom action. When active, the venoms were able to provoke complete inhibi t ion of the specific b inding of the radiol igand, but did not affect the nonspecific bind- ing. Inhibi t ion occurred with the following order of potency: C. planorbis (ICs0 = 2.1 ~g protein/ml) ~ C. tessulatus (2.7) > C. eburneus (19) > C. textile ( 5 4 ) > C. geographus (130). No more than 2-tbld differences in potency were observed for different venom prepara t ions of a single species. Venoms from the other Conus species tested ( i . e .C . aulicus, ('. capitaneus, C, consors, C. flavidus, C../urvus, C. generalis, C. gloriamaris, C. irnperialis, C. leopardus, C. litteratus, C. lividus, C. marmoreus, C. miles, ('. mustellinus, C. striatus, C. t:exillum and C. rirgo) provoked less than a 20% inhibi t ion of radiol igand binding at 100/~g/ml.

The ability of the venoms themselves to bind the radiol igand and, hence, to decrease its free concen- t rat ion, was ruled out by molecular sieving experi- ments. As shown in Fig. 2, the ~,-receptor inhibi tory

l z o c_

o°~ ~ ~ ~- ~= lOO6osO ~ t ao

t~

x ao

o -5 - 4 -3 - 2 -1 0

Venom concentrotion (Log mg/mL)

Fig. l. Species differences in Conus venom-[3H]RX 781094 competition binding. Calf retina membranes were incubated with 6 nM [3H]RX 781094 and the indicated concentrations (in mg protein/ml) of venom from C. tessulatus (Fq), C. textile (m) and C. aulicus (,~), after which specific binding was measured as described in the Experimental Procedures. Binding is expressed as percent of control

binding, i.e. binding in the presence of buffer only.

Conus venom interaction in calf retina membranes 415

- l O 1 2 0 - - .! o 100 -

B x

E o,,

l / / _i !

t i - , oL . . . . . . . . I~ . . . . . . . . . l , _ i - , - l + - , i o

0 5 40 15 20

Frac t i on number

Fig. 2. Lack of venom-[3H]RX 781094 interaction. C. eburneus venom (0.3 mg/ml) was incubated for 10 min at 37°C in Tris-magnesium buffer either alone (l-q) or with 6 nM [3H]RX 781094 (11) and chilled on ice. Samples (300 #1) were loaded on a small (18 x 0.7 cm) column of Sephadex G-50 (Pharmacia), and eluted at 4°C with buffer. Fractions of 300/~1 were collected. The specific binding of [3H]RX 781094 to calf retina membranes was measured in the presence of aliquots (100 #1) of the eluted fractions (I-q). Aliquots (50 #1) of the eluted fractions were also assayed for radioactivity ( I ) . Results are expressed as means and SEM of 3 experiments. Dextran Blue (Sigma), which is not retained by the Sephadex G-50

column, eluted in fraction 7. Fractions 1~ represent the void volume of the column.

activity present in the venom of C. eburneus was not retained by a Sephadex G-50 column. In contrast, elution of the radioactivity occurred well beyond the void volume when a preincubated mixture of venom and [3H]RX 781094 was applied onto the column

(Fig. 2). Similar results were obtained for the venom of the other Conus species of interest. Accordingly, the inhibitory activity of the venom can best be explained by the ability of one or more of their components to mask the binding site of the

r"

C

o~ o CO

X

I

100 - ~ I

8 0 -

100 f 60 - 80

6O

4 0 -- 40 •

,° t 20 ° e 7 -6 --5 -4 -3 ~ J .

EGTA concentration ( Log M)

O I I I I - 4 - 3 - 2 - I 0

C. t e x t i l e venom c o n c e n t r a t i o n ( l o g m g / m l )

Fig. 3. Suppression of venom-receptor interaction by EGTA. Retina membranes were incubated with 6 nM [3H]RX 781094 and the indicated concentrations (in mg protein/ml) of venom from C. textile either alone ( n ) or in the presence of 0.1 m'M EGTA (A). Insert: membranes and radioligand were incubated

in the presence of C. textile venom (100 pg protein/ml) and increasing concentrations of EGTA.

416 l{x~ ;A. (.~IZ[CR~.~rIE( " ¢'f ~I[.

r

.E 8 o -

=~ ,oo , . ~ "o c • ~ 6 0 eo -

6 0 -

0 4 0 40 --

X 20 -- er

o

, - , 0 - 9 -2, ILM-; -' T [epi phr e] (

0 5 - 4 - 3 - 2 - I 0

C . e b u r n e u s v e n o m c o n c e n t r a t i o n ( L o g m g / m t )

Fig. 4. Lack of GTP modulation of venom-[]H]RX 781094 competition binding, Retina membranes were incubated with 6 nM [3H]RX 781094 and the indicated concentrations (in _mg protein/ml) of C. eburneus venom either alone ( l ) or in the presence of 0.1 mM GTP (D). The curve was drawn according to a one-site model. Insert: membranes and radioligand were incubated with increasing concentrations of

(-)-epinephrine either alone ( I ) or in the presence of 0.1 mM GTP (l-l).

72-receptor. Migrating in the void volume of the Sephadex G-50 column, these components have an apparent molecular weight exceeding 10 kDa.

The inhibitory activity of all the venoms appears to be dependent on the presence of divalent cations such as Ca 2+. The decrease in activity of C. textile venom upon the addition of increasing concentrations of the Ca2+-chelating agent EGTA is shown in the insert of Fig. 3. A maximally effective concentration of EGTA was able to provoke the complete suppression of competition binding for the venom of C. textile (Fig. 3) as well as for the other species investigated (data not shown).

The competition binding curves for the venoms are steep and GTP-insensitive (Fig. 4). In contrast, the curve of the agonist ( - ) -epinephr ine is shallow and undergoes a marked steepening and rightward shift upon addition of GTP (Fig. 4, insert). These data suggest that the venom of certain Conus species contains peptide toxins which are capable of shielding the binding site of c(2-receptors in a Ca2+-dependent and antagonistic manner.

D I S C U S S I O N

Calf retina membranes contain a large amount of ~2-adrenergic receptors: 0.9-1.2pmol/mg protein. Moreover, an appreciable portion (75%) of these receptors is able to undergo agonist-mediated coup- ling to the adenylate cyclase regulatory component Ni (Convents et al., 1987a,b). As shown in the insert

of Fig. 4, these receptors display high agonist affinity when coupled and low agonist affinity when coupling is reversed by guanine nucleotides such as GTP. Since antagonists do not mediate receptor-Ni coupling, their competition curves are steep and not affected by GTP (Hoffman et al., 1980). This difference allows an easy discrimination between agonists and antagonists by means of competition binding experiments. The high amount of receptor sites and the high pro- portion of coupling make calf retina membranes an excellent model system for the investigation of the molecular properties of ct2-adrenergic receptors as well as their ability to interact with natural and synthetic substances.

Whereas the interaction between a2-receptors and natural alkaloids and their derivatives is well known (Weiner, 1980; Motulsky et al., 1984), the findings presented in this report suggest that these receptors are also able to bind toxins present in the venom of different Conus species. Conidae, a family of gastro- pods living in tropical waters, have developed their radular system into an ingenious venom apparatus to catch their prey (Olivera et al., 1988). According to their feeding habits, they have been divided into fish-, molusc- and worm-eating species (Endean and Rudkin, 1965).

Crude venoms of only 5 of the 22 species tested provoked an appreciable inhibition of binding of the radiolabelled antagonist [3H]RX 781094 to the calf retina ~t2-receptors. The potency decreased in the order: C. planorbis (feeding habits unknown)~

Conus venom interaction in calf retina membranes 417

C. tessulatus (f ish-eating)> C. eburneus (worm- eating) > C. textile (mollusc-eating) > C. geograhus (fish-eating). The competi t ion curves of these venoms were all steep and GTP-insensitive (Fig. 4), indicating that they contain components which are able to shield the binding site of ~t2-receptors in an antagonistic manner. Proteolytic degradation of the receptors has not been detected so far (Czerwiec et al., manuscript in preparation). As illustrated in Fig. 2, this in- hibitory activity was not retained by a Sephadex G-50 molecular sieving column, suggesting that this activ- ity is borne by proteins with a molecular weight exceeding 10 kDa. Based on this consideration and the fact that only crude venoms were tested, it appears that, in C. planorbis and C. tessulatus venom, the IC5o values of these proteins for the receptor must be well below 10 -7 M.

The described Conus venom toxins often encom- pass small peptides but the isolation of components with higher molecular weights has been reported as well; i.e. a 13 kDa peptide with convulsant action from C. geographus (Clark et al., 1981), a cardiotonic 25 kDa glycoprotein from C. striatus (Kobayashi et al., 1982b) and vasoactive 28 and 55 kDa proteins from C. eburneus (Kobayashi et al., 1982c) and C. tessulatus (Schweitz et al., 1986), respectively. These vasoactive proteins were reported to interact with calcium transport systems in the plasma membrane. Interestingly, direct modulat ion of some membrane calcium transport activity by ctz-receptors is in- creasingly suggested (Van Zwieten and Timmerman, 1987; Nielsson et al., 1988). In this context, it is noteworthy that the venoms of C. eburneus and C. tessulatus are amongst the most potent in our study and that their action on ctz-receptors is completely suppressed by the Ca2+-chelating agent EGTA. We are therefore tempted to speculate that the involved venom components are able to shield the radioligand binding site of ct2-receptors by binding to adjacent calcium transport systems. If so, these purified com- ponents might be useful tools for the investigation of ~,.-recentor-calcium transport system interactions. Acknowledgements--We are very grateful to Mr G. Poppe for obtaining and identifying the various Conus specimens, to Mr J. P. De Backer for technical assistance and to the slaughterhouse of Geel for the facilities offered for obtain- ing and dissecting retinas. G.V. is Onderzoeksleider of the Nationaal Fonds voor Wetenschappelijk Onderzoek, Belgium. This work was supported by a grant from the Fonds voor Geneeskundig Wetenschappelijk Onderzoek, Belgium.

REFERENCES

Abe T. and Saisu H. (1987) Identification of the receptor for to-conotoxin in brain. J. biol. Chem. 262, 9877-9882.

Clark C., Olivera B. M. and Cruz L. J. (1981) A toxin from the venom of the marine snail Conus geographus which acts on the vertebrate central nervous system. Toxicon 19, 691-699.

Convents A., De Backer J.-P. and Vauquelin G. (1987a) Characterization of alpha2-adrenergic receptors of calf retina membranes by [3H]-Rauwolscine and [3H]-RX 781094 binding. Biochem. Pharmac. 346, 2497-2503.

Convents A., De Backer J.-P. and Vauquelin G. (1987b) Tight agonist binding may prevent the correct inter- pretation of agonist competition binding curves of ~%-adrenergic receptors. Molec. Pharmac. 32, 65-72.

Endean R. and Rudkin C. (1965) Further studies on venoms of Conidae. Toxicon 2, 225-249.

Hoffman B. B., Mullikin-Kilpatrick D. and Lefkowitz R. J. (1980) Heterogeneity of radioligand binding to ~t-adrenergic receptors. J. biol. Chem. 255, 4645-4652.

Kobayashi J., Nakamura H., Hirata Y. and Ohizumi Y. (1982a) Effect of venoms from Conidae on skeletal, cardiac and smooth muscles. Toxicon 20, 823-830.

Kobayashi J., Nakamura H., Hiram Y. and Ohizumi Y. (1982b) Isolation of a cardiotonic glycoprotein, stria- toxin, from the venom of the marine snail Conus striatus. Biochem. biophys. Res. Commun. 105, 1389-1395.

Kobayashi J., Nakamura H., Hirata Y. and Ohizumi Y. (1982c) Isolation of eburneotoxin, a vasoactive substance from Conus eburneus venom. Life Sci. 31, 1085-t091.

Lowry O. H., Rosebrough N. J., Farr A. C. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275.

McManus O. B., Musick J. R. and Gonzalez C. (1981) Peptides isolated from the venom of Conus geographus block neuromuscular transmission. Neurosci. Lett. 24, 57-62.

Motulsky H. J., Maisel A. S., Snavely M. D. and Insel P. A. (1984) Quinidine is a competitive antagonist at ~t~- and c%-adrenergic receptors. Circ. Res. 55, 376-381.

Nielsson T., Arkhammer P., Rorsman P. and Berggen P. O. (1988) Inhibition of glucose-stimulated insulin release by ct2-adrenoceptor activation is paralleled by both a re- polarisation and the reduction in cytoplasmic free Ca 2+ concentration. J. biol. Chem. 263, 1855-1860.

Olivera B. M., Gray W. R. and Cruz L. J. (1988) Marine toxins and venoms. In: Handbook of Natural Toxins (Tu A. T., ed.), Vol. 3, pp. 327-352. Dekker, New York.

Olivera B. M., Gray W. R., Zeikus R., Mclntosh J. M., Varga J., Rivier J., de Santos V. and Cruz L. J. (1985) Peptide neurotoxins from fish-hunting Cone snails. Science 230, 1338-1343.

Schweitz H., Renaud J. F., Randimbivololona N., Preau C., Schmid A., Romey G., Rakotovao L. and Lazdunsky M. (1986) Purification, subunit structure and pharmaco- logical effects on cardiac and smooth muscle cells of a polypeptide toxin isolated from the marine snail Conus tessulatus. Eur. J. Biochem. 161, 787-792.

Van Zwieten P. A. and Timmermans P. B. M. W. M. (1987) Interactions between calcium entry blockers and vascular ct-adrenoceptors. In: Pharmacology; Proceedings of the Xth International Congress of Pharmacology (Rand M. J. and Raper C., eds), pp. 527-536. Elsevier, New York.

Weiner N. (1980) Drugs that inhibit adrenergic nerves and block adrenergic receptors. In: The Pharmacological Basis of Therapeutics (Goodman Gilman A., Goodman L. S. and Gilman A., eds), pp. 176-210. Macmillan, New York.

NC[ 14/4~C