Calcium- and potassium-channel blockers interact with α-adrenoceptors

Molecular and Cellular Endocrinology, 19 (1980) 243-251 0 Elsevier/NorthHolland Scientific Publishers, Ltd.

243

CALCIUM- AND POTASSIUM-CHANNEL BLOCKERS INTERACT WITH

a-ADRENOCEPTORS

Hartmut GLOSSMANN and Rainer HORNUNG * Rudolf-Buchheim-Institut fiir Pharmakologie, Frankfurter Strasse 107, D-6300 Giessen (F.R.G.)

Received 7 February 1980; accepted 16 May 1980

The so-called calcium antagonist DdOO and agents that enhance the influx of calcium ions into nerve terminals during action potentials (4-aminopyridine, tetraethylammonium, guanidine) interact with cu-adrenoceptors. The interaction was revealed by binding experiments in vitro with the (~a- and or-specific ligands [3H]clonidine and [3H]prazosin, resp. DdOO binds in a competitive manner to (~a- and at-adrenoceptors in rat-brain membranes. [3H]Prazosin was used to identify or-adrenoceptors in rat-heart-membrane fragments. DdOO inhibited the binding of the tritiated antagonist, in a manner similar to that seen in rat-brain membranes. 4-Ami- nopyridine, guanidme and tetraethylammonium blocked noncompetitively the binding of [3H]clonidine in brain membranes. There was little or no effect of these agents on the binding of [3H]prazosin to aradrenoceptors in rat heart or brain.

The results indicate that D-600 binds to a region common to or- and ora-adrenoceptors, whereas the potassium-channel blockers reveal a structural feature of the aa-adrenoceptors in brain membranes which is also present in voltage-dependent potassium channels but is not shared by or adrenoceptors.

Keywords: or-adrenoceptors; c~~adrenoceptors; clonidine; prazosin; ion-channel blockers; 4-aminopyridine; tetraethylammonium; guanidine; DdOO; or-receptor; brain; heart; calcium antagonist.

It is likely that a-adrenoceptors (e.g. in smooth muscle) operate through ion channels. (For a review, see Bolton, 1979.) Because relaxation as well as contraction are observed, two (o-adrenoreceptor-operated) channels are postulated. When the receptor causes relaxation it links to a channel that allows potassium to pass, resulting in hyperpolarization. The other channel allows sodium, chloride, potassium and other ions to pass, leading to depolarization and contraction.

If a receptor is connected to an ion channel, the ligand-binding characteristics of

the receptor may be altered by probing the channel with agents specific for the

* This is part of the thesis of R.H. to be presented to the Fachbereich humanmedizin, Justus- Liebig-Universitat Giessen, in partial fulfiient of the requirements for a Doctor of Medical Science degree.

244 Hartmut Glossmann, Rainer Hornung

channel. A recent example is the benzodiazepine receptor (believed to be tied to an anion channel) where several anions increased the affinity of diazepam binding to

brain membranes, and a specific anion-channel blocker (4,4’-diiosothiocyano-2,2’- stilbene disulphonic acid) reversed the anion enhancement of diazepam binding

(Costa et al., 1979; Rodbard et al., 1979). The concept of probing receptors with agents known to interact with ion chan-

nels may well turn out to be as useful as were the guanine nucleotides to show the linkage of membrane-bound hormone receptors to GTP-regulatory proteins (Rod- bell, 1980). We therefore investigated whether agents known to bind to ion channels can also interact with a-adrenoceptors in vitro.

D-600 and its parent compound verapamil are potent blockers of trans-mem-

brane calcium fluxes. These so-called calcium antagonists block calcium channels, e.g. in vascular smooth muscle and myocardium, at low concentrations (low7 to

10” M). (For a review, see Fleckenstein, 1977.) Voltage-dependent potassium channels are blocked by tetraethylammonium (TEA) (Hille, 1967; Katz and Miledi, 1969; Armstrong, 197.Q 4-aminopyridine (4-AP) (Lundh and Thesleff, 1977; Lundh, 1978) and guanidine (Otsuka and Endo, 1960; Lundh and Thesleff, 1977; Galvan et al., 1980). It is surprising, as will be reported below, that D600, TEA, 4-AP and guanidine also block the binding of the agonist clonidine to cus-adrenoceptors in brain membranes. DdOO (but not 4-AP, TEA or guanidine) inhibits the binding of the antagonist prazosin to ai-adrenoceptors in brain and heart membranes.

MATERIALS AND METHODS

[3H]Clonidine (spec. act., 26.7 Ci/mmole) and [3H]prazosin (spec. act., 33 Ci/ mmole) were gifts from Boehringer and Sohne, Ingelheim (FRG), and Pfizer GmbH,

Karlsruhe (FRG), resp. D-600 was a gift from Knoll AG, Ludwigshafen (FRG). 4Aminopyridine was from Merck, Darmstadt (FRG); guanidine from Fluka, Base1 (Switzerland), and tetraethylammonium (purified) a gift from Prof. Vogel (Physio- logisches Institut, Giessen). Brain membranes from rat cerebral cortices were pre- pared as described (Glossmann and Presek, 1979), and rat-heart particulate fractions according to Williams et al. (1976).

Specific binding of [3H]clonidine and [3H]prazosin to particulate preparations was determined as described (Hornung et al., 1979) with the filtration technique. In brief: rat-brain or rat-heart particles (l-3 mg of protein/ml) were incubated in buffer (0.5 mM EDTA, 50 mM Tris-HCl, pH 7.4, 4 /.IM phenylmethanesulfonyl fluoride) and the tritiated ligands in the presence or absence of competitors for 20-30 min at 30°C in a final volume of 0.25 ml. Subsequent separation of bound and free ligand was performed as previously described. Unspecific binding (which was subtracted from total binding to give specific binding) was determined with 5 PM unlabeled clonidine for [ 3H] clonidine and 10 ,uM unlabeled phentolamine for [3H]prazosin experiments. Unspecific binding was less than 10% of the total bind-

Calcium- and potassium-channel blockers 245

ing for competition experiments with brain membranes (0.4-l nM [3H]prazosin, 2-6 nM [3H]clonidine). For rat-heart particulate fractions (0.3-0.6 nM [3H]prazosin) 15% of the total ligand binding represented unspecific binding.

RESULTS

Experiments with D-600

Rat-brain membranes have saturable high-affinity sites for the 2 adrenoceptor

ligands clonidine (Greenberg et al., 1976; Glossmann and Presek, 1979) and prazo-

sin (Hornung et al., 1979). Clonidine labels olz-adrenoceptors whereas the prazosin- binding sites (similar to WB4101 sites, U’Prichard and Snyder, 1979), have similari-

ties with al-adrenoceptors. Fig. 1 shows that increasing concentrations of D-600 inhibited the specific bind-

ing of [3H]clonidine (as-adrenoceptors) and of [3H]prazosin (or-adrenoceptors) in rat-brain membranes. DdOO was a competitive inhibitor for a*-adrenoceptors

and cwl-adrenoceptors as revealed by saturation analysis. An example of a saturation analysis of a*-adrenoceptors in the presence or absence of D&O0 is shown in Fig. 2.

To see whether the interaction of DdOO with a-adrenoceptors was tissue- specific, we investigated its effects on aI-adrenoceptors in rat-heart membranes.

80 t

60 x

60-

a,-ADRENOCEPTORS

ZO-

0 0.1 I.0 IO 100 1000 0 01 1.0 IO 100 1000

D-600 [AA] Fig. 1. Inhibition of specific [3H]clonidine binding (cq-adrenoceptors) and [3H]prazosin binding (aI-adrenoceptors) to rat-brain membranes by DdOO. Points are means from triplicate incubations, which differed less than 7%. B, is the fraction of tracer specifically bound in the absence, and B in the presence, of competing ligand. [3H]Clonidine was 5.8 nM (Bo = 0.11) and [3H]prazosin 0.6 nM (B. = 0.22).


0 0.2 0.4 0.6 0.6 IC

60

50 t

40

30 i

BOUND LIGAND [rb] D-600 [MM]

Fig. 2. Scatchard transfo~ation of results of 13H]cIonid~ne equ~brium binding to brain membranes. The Expt. was performed in the absence (0) or presence (0) of 5 PM D-600. [3H]Cloni- dine ranged from 0.5 to 25 nM and rat-brain-membrane protein was 3 mg/ml. The association constant (Ka) in the absence of DdOO was 0.206 nM_’ ; Kg was 0.128 in the presence of 5 PM DdOO. The Ki value, assuming competitive inhibition, was 8.2 MM, in excellent agreement with the data from displacement studies (Table 1).

Fig. 3. Inhibition of specific [3H]prazosin binding to rat-heart membranes by D-600. The data, which are means of triplicate incubations, are shown in the logit-log transformation. Percent- age is the S/B, ratio,multip~ed by 100 (So = 0.15).

These receptors bind [3H]prazosin with high af~nity (the association ~onstant,~~, is 6 nM_’ at 3O*C) and have a density of about 50 fmoles per mg of crude rat-heart particulate protein (Glossmann et al., in press), The results of a representative competition experiment with D&O0 are shown in Fig, 3 and indicate that the calcium antagonist also binds to ~~~drenoceptors in rat heart. The Ki values of D-600 for 02- and aradrenoceptors in rat-brain and rat-heart membranes are given in Table 1.

Experiments with potassium-evangel blockers 4-~inopyridine (4-A%‘), tetraethylammonium (TEA) and guanidine inhibited

preferentially the binding of [3H]clonidine to rat-brain membranes (Fig. 4). When the same concentrations of the channel blockers were used to probe all-adrenoceptors, little or no effect was seen (not shown). This was observed with rat-brain membranes as well as with rat-heart particulate fractions and with various concen-


Table 1 The table summarizes the Ki values of D-600 and the potassium-channel blockers. The results for DdOO are mean values from 3 Expts. IC so values (5-7 concentrations) were determined by converting the binding-inhibition data into logit-log plots (as shown in Fig. 3) and computer fitting with a Wang 2200 B laboratory computer. Ki values were calculated from: Ki = ICso/ (1 +L. K,), where the ICso value is the concentration of the competitor that reduced the specific binding of the labeled ligand by 50%; L is the total labeled ligand concentration and K, is the association constant of the tritiated ligand. L was 2-6 nM for Expts. with [3H]clonidine (as-adrenoceptors) and 0.3-0.6 nM for experiments with [3H]prazosin (q-adrenoceptors). The Ki values for the potassium channel blockers derived from secondary plots of slopes from primary plots of l/B versus l/F (as shown in Fig. 5) are given. Values in parentheses are those obtained from plots of l/B versus inhibitor concentration.

Channel blocker Inhibition type era-Adrenoceptors or -Adrenoceptors (rat brain)

Rat brain Rat heart

D-600 competitive 9.3 /.LM 2.8 MM 0.34 /.rM 4-Aminopyridine non-competitive 78nM(llSrM) - Guanidine non-competitive 650/~M(1130~M) - -

Tetraethylammonium non-competitive 12 200 NM (13 900 rM)

trations of [3H]prazosin (0.1-t nM). It appears that the inhibitory action of the potassium-channel blockers is restricted to a2-adrenoceptors.

In contrast with D-600, inhibition by these agents was non-competitive. The noncompetitive inhibition was revealed by saturation analysis. The experiments are illustrated for 4-AP and guanidine in Fig. 5a, b. In addition to Scatchard analysis we used primary double-reciprocal plots, secondary plots of slope versus [I] (where the slope is calculated by linear-regression analysis of the primary plot and [I] is the concentration of the inhibitor) and plots of [I/b] versus [I] where [b] is the con-

0.01 0.1 1.0 IO wo

DRUG [mM]

Fig. 4. Inhibition of [3H]clonidine binding to brain membranes by 4-aminopyridine (4-AP), guanidine and tetraethylammonium (TEA). The drugs were used as their chloride salts. Points are means from triplicate incubations; [3H]clonidine was 4.8 nM (B, = 0.13).


0.06

LL , 0.04 m

0 02 04 06 OJ3

BOUND [d.i]

-IW 0 200 403 60;,

4-AP [vM]

-100 0 200 400 6M3

4-AP w]

Fig.5a. Saturation analysis (Scatchard plot) of oa-adrenoceptors with [3H]clonidine in rat- brain membranes without or with 4dP present at the indicated concentrations. Brain-membrane protein was 2.8 mg/ml. Each point is a mean of 3 determinations. Points were fitted by linear regression with a Wang 2200 B laboratory computer. B/F is the ratio of bound/free ligand. A replot of the data obtained with the indicated concentrations of [3H]clonidine versus the wncen~ation of 4-AP is shown on the left. (Ki = 115 + 35 &f.) The secondary plot of the slopes of primary plots using l/lp versus l/F is shown on the right (Kt = 78 MM.).

centration of specifically bound labeled ligand. In all cases strait-one fits were obtained with correlation coefficients between 098 and 0.99. The Ki values are sum- marized in Table 1,

DISCUSSION

Our results show that certain agents that are known to block ion channels also interact with Ir-adrenoceptors. Although one is tempted to dismiss the results as un-


0 02 04 0.6 0.8

BOUND [nM]

-20 4 e

200

160

w 120 a

s I” cn 60

40

-20 4 8

GUANIDINE [mM] GUANIDINE [mM]

Fig. 5b. The same as for Fig. 5a, but with guanidine instead of 4-AP. The Ki value derived from the replot l/B versus the indicated concentrations of guanidine was 1130 * 290 PM. The secondary plot of the slopes of primary plots using l/B versus l/Fgave a Ki value of 650 PM.

specific phenomena the following points are worth considering. The interaction of

D400 with al- and cr+drenoceptors is of the competitive type. The Ki values of D&O0 are in the pmolar range and are lower for a1 -adrenoceptors in brain and heart compared with cr2-adrenoceptors. Blackmore et al. (1979) have shown that verapamil, the parent compound of D&00, is also an ar-adrenoceptor blocker on rat-liver cells. The findings strongly suggest that these calcium antagonists block a-adrenoceptors regardless of subtype and localization. Because DdOO and verapamil are

highly lipophilic drugs, a structural hydrophobic component common to both CY~- and al-adrenoceptors @artly overlapping with the neurotransmitter binding site?) is most likely the target.

The calcium channels in smooth and cardiac muscIe (Fleckenstein, 1977) are most sensitive to DdOO and verapamil. However, the calcium channels on terminal sympathetic nerve membranes (e.g. GGthert et al., 1979) and on synaptosomes from brain (Nachshen and Blaustein, 1979) have sensitivities even lower than those


observed by us for cw-adrenoceptors. The question therefore arises what have calcium channels and aadrenoceptors in common to display competitive antagonism for adrenergic drugs in the latter case and for calcium in the former?

4-AP, guanidine and TEA blocked the binding of clonidine to as-adrenoceptors in rat-brain membranes. The inhibition was of the non-competitive type and specific in the sense that concentrations that were inhibitory to arz-adrenoceptors had

no effect on ar-adrenoceptors in rat brain and heart labeled with [3H]prazosin. 4-AP, guanidine and TEA are blockers of the voltage-dependent potassium channels. They prolong the pre-synaptic action potential, thereby increasing the amount of free calcium within the nerve terminal, and augment transmitter release (Galvan

et al., 1980; Armstrong, 1975;Molgo et al., 1977). It is interesting that the concentrations of these agents found to be effective in neurophysiological experiments are

in the range of the Ki values found by us for a2-adrenoceptors. Although a pre- synaptic specificity of 4-AP, the most widely used agent, is often claimed, addi- tional post-synaptic actions are also seen. Low concentrations of 4-AP (10 PM) reduce the noradrenalineinduced contractile response of rabbit vas deferens -but only at low concentrations of the neurotransmitter. High concentrations of 4-AP

(100-1000 nM) prevent desensitization and potentiate the response to noradrenal- ine (Johns et al., 1976). These effects of 4-AP are as yet unexplained but may indicate post-synaptic and intracellular actions of 4-AP. It is interesting that agents which block voltage-dependent potassium channels act with a similar rank order of po- tency on as-adrenoceptors. Perhaps the ionic pores and the receptors have similar

structural components or charged groups. A molecular interpretation of our findings with the ion-channel blockers is dif-

ficult. One firm conclusion, however, can be made. Caution should be exerted in the interpretation of experiments in biological systems where these drugs are used. When the role of calcium in a-adrenoceptor-mediated effects is investigated with the help of D-600 or verapamil, the competitive antagonism of the calcium antagonists at the receptor level must be taken into account. If the potassium-channel blockers are used on tissue preparations with sympathetic nerve endings (which

have pre-synaptic oz-adrenoceptors) the blockade of these receptors should lead to

increased transmitter output.

ACKNOWLEDGEMENT

We thank F. Dreyer , of this Institute, for helpful suggestions and discussion. The

Deutsche Forschungsgemeinschaft supported us with a grant (Gl48/10).


REFERENCES

Armstrong,C.M. (1975) Quart. Rev. Biophys. 7,179-210. Blackmore, P.F., El-Refai, M.F., and Exton, J.H. (1979) Mol. Pharmacol. 15,598-606. Bolton,T.B. (1979) Physiol. Rev. 59,606-718. Costa, T., Rodbard, D., and Pert, C.B. (1979) Nature (London) 277,315-317. Fleckenstein, A., (1977) Annu. Rev. Pharmacol. Toxicol. 17,149-166. Galvan, M., Grafe, P., and Ten Bruggencate, G, (1980) Exp. Neural. 67,234-246. Glossmann, H., and Presek, P. (1979) Naunyn8chmiedeberg’s Arch. Pharmacol. 306, 67-73. Glossmann, H., Hornung, R., and Ptesek, P,, J, Cardiovasc. Pharmacol., in press. GSthert, M., Nawroth, P., and Neumeyer, H. (1979) Naunyn~chmiedeberg’s Arch. Pharmacol.

310,11-19. Greenberg, D.D., II’Prichard, D.C., and Snyder, 8.H. (1976) Life Sci. 19,69-76. HiBe, B. (1967) J. Gen. Physiot. SO, 1287-1302. Hornung, R., Presek, P., and Glossmann, H. (1979) NaunynSchmiedeberg’s Arch. Pharmacol.

308,223-230. Johns, A., Golko, D.S., Lauzon, P.A.,and Paton, D.M. (1976) Eur. J. Pharmacol. 38,71-78. Katz, B., and Miledi, R. (1969) J. Physiol. 203,689-706. Lundh, H., and Thedeff, S. (lg?7) E&r, I, Pharmacol. 42,411-412. Lundh, M. (1978) Brain Res. 153,307-318. MoJgo, J., Lemeignan, M., and Lechat, P. (1977) J. Pharmacol. Exp. Ther. 203,653-663. Nachshen, D.A., and Blaustein, M.P. (1979) Mol. Pharmacol. 16,5?9-586. z;;$ M., and Endo, M. (1960) J. Pharmacol. Exp. Ther. 128,273-283. *,- Rodbard, D+,costa, T., and Pert,C.B. (1979) Nature (London) 280,173-174.

Rodbell, M. (1980) Nature (Lolidon) 284,17-22. ~‘&RtIM& B&J,, and Snyder, S.H. (1979) Life 8ci. 24,79-88. ~~a~~~ j,,T:, ~~~~~ D,, and Leflcowitz, R.J. (1976) J. Biol. Chem. 251,6YlS-6923.

Calcium- and potassium-channel blockers interact with α-adrenoceptors

Documents

Transcript of Calcium- and potassium-channel blockers interact with α-adrenoceptors