[1] 16w CGP 55845A blocks baclofen, 7'-aminobutyric acid and inhibitory postsynaptic potassium currents in guinea pig CA3 neurons

[1] 118w Single electrode voltage-clamp recording from CA3 neurons in guinea pig hippocampal slices was applied to study effects of a new GABAB antagonist, CGP 55845A, on (-)baclofen (lR,c)-or ~'-aminobutyric acid (loABA)-induced potassium (K)-currents and on inhibitory postsynaptic K-currents (K-IPSCs) recorded in the presence of blockers for fast synaptic transmission. K-IPSCs were induced by bath application of 4-aminopyridine (4-AP). CGP 55845A, in 10 -s to 10 -7 M concentrations, blocked all these K-currents and was more potent than all GABAa antagonists known to date. However, onset of the CGP 55845A effect and recovery were slow. We conclude that a potent and selective GABAB antagonist is now available to study the physiological role of GABA a receptors in the mammalian brain.

[1] 115w In voltage-clamp recordings at a holding potential of -65 mV, (-)baclofen (0.5 or 5 ktM) induced an outward current (Iaac) accompanied by an increase in input conductance. Both effects were reduced by nanomolar concentrations of the GABAa antagonist CGP 55845A (Fig. 1A,B). When IBac induced by 0.5/IM (-)baclofen was measured 10 min after superfusion with CGP 55845A, the effect of the antagonist was concentration dependent (Fig. 1C). However, the effectiveness of CGP 55845A was also time dependent (Fig. 1A,B,D). 20 nM CGP 55845A was effective in reducing the (-)baclofen (0.5 /~M)-indueed currents, but it took >40 min to completely block them. In contrast, 200 nM CGP blocked Iaac within 10 rain (n=3: Fig. I D)

[2] 88w . When (-)baclofen and CGP 55845A were simultaneously applied, 1 /aM CGP 55845A blocked I~,~ within 30 s, and 10/aM blocked Iaa~ immediately (n=3). Not only was the washin of CGP 55845A slow, but also the washout. Only if IB, ~ was reduced by a low antagonist concentration could a full recovery be achieved within 60 rain of washing (Fig. 1E). If, however, IR,~ was completely blocked by the application of 200 nM CGP 55845A for 10 min or 10/aM CGP 55845A for 3 min, recovery was incomplete.

[3] 96w The slow time course of CGP 55845A effects might be due to a strong lipophilicity of the drug and, therefore, uptake into lipophilic compartments. Indeed, the lipophil(city of CGP 55845A as judged from the chemical structure [3] should be higher than that of CGP 35348, which also acts as a GABAR antagonist [16]. For comparison, we tested the time course of the effects of CGP 35348. CGP 35348 (20/aM) applied simultaneously with (-)baclofen became immediately effective, and strongly reduced the (-)baclofen current (Fig. 2A). The effect was reversible within 5 min of washing (Fig. 2A, n=9).

[4] 223w We tested the specifity of the new antagonist. CGP 55845A up to concentrations of l0/aM was a rather specific receptor antagonist. 10/aM CGP 55845A neither induced a current itself nor changed membrane resistance (cf. ref. 15). Serotonin (5/aM)-induced outward currents were unaffected by CGP 55845A (10/aM for 30 min, n=3). An intracellular site of action is also unlikely as 50/aM CGP 55845A in the recording microelectrode did not change lu~ within 30 rain. In summary, CGP 55845A is a potent GABAa receptor antagonist which, however, has the disadvantage of a low exchange rate when applied to brain slices. Because of its strong potency, we also tested CGP 55845A against GABA-induced outward currents (IGA~A)" GABA (500 /aM) was applied to the bath in the presence of GABAA antagonists and induced an outward current (Fig. 2B). CGP 55845A reduced IGASA in a concentration-dependent manner, l0/aM CGP 55845A immediately blocked the effects of simultaneously applied GABA (500/IM, n=3). When IGABA was measured before and after 20 min of superfusion with CGP 55845A, IGABA was blocked by a 500 nM concentration and reduced by 52+ 10% of control by a 200 nM concentration (Fig. 2B, n=5). In the same cells, IBac (0.5/IM) was blocked by a 200 nM concentration (Fig. 2B) after a shorter application time. Recovery upon washout was complete for IGABA but not for Bac"

[5] 183w Having established that CGP 55845A antagonized IBac and IGABA, we next examined its potency against K-IPSCs. K-IPSCs were elicited by the application of 4-AP (50/.tM) for 3-5 min [6, l 1]. 1-10 pM CGP 55845A blocked K-IPSCs almost immediately (<30 s). The time course of the effect of 100 nM CGP 55845A is shown in Fig. 3A. Subsequent applications of 4-AP in the presence of CGP 55845A failed to induce K-IPSCs. There was no recovery upon washout. 4-AP-induced K-IPSCs were also completely blocked after 20 min of superfusion with 50 nM CGP 55845A (Fig. 3B; n=3) and reduced in amplitude by 46_+7% of control after 35 min in 20 nM CGP 55845A (Fig. 2C, n=3). In all cells, in which GABA B receptor-mediated inhibition was blocked (n=8), 4-AP induced repetitive inward currents followed by an outward current (Fig. 3B), despite the presence of antagonists for fast synaptic transmission. The inward currents most likely are burst currents followed by a Ca-activated K-current [l 1]. With low concentrations of CGP 55845A, recovery of K-IPSCs was easier to achieve than for IBac (Figs. 2C and 3B).

[1] 99w In the mammalian brain, T-aminobutyric acid (GABA) acts on bicuculline-sensitive GABA A receptors and on bicuculline-resistant GABAB receptors [1,14,17]. The GABAB agonist baclofen induces a potassium (K) conductance increase in hippocampal cells [4,13], and the effect can be blocked by GABAB antagonists [2,7,8,16]. Up until now only low affinity GABAB antagonists have been available [3] which are rather ineffective in blocking GABA-induced K conductance increases [5,6]. In this paper, we describe the effects of a new, high affinity GABAB receptor antagonist [3,15] upon (-)baclofen (IBac)-and GABA (IGABA)-induced K-currents and upon inhibitory postsynaptic K-currents (K-IPSCs) induced by 4-aminopyridine (4-AP) [11].

[2] 191w Guinea pig hippocampal slices (thickness 300-400 Hm) were prepared and single electrode voltage-clamp recordings (switching frequency 10-12 kHz) of CA3 pyramidal neurons were made as previously described [6,9,10]. All drugs were applied by the bath to the slices held in a submersion chamber. Exchange of solution was complete within 40 s [6]. The standard solution consisted of(in mM): NaCl 127, KCI 2, MgSO4 1. lar microelectrodes were filled with 3 M KC1 and had resistances between 35 and 80 MfL All recordings were made in the presence of picrotoxin (50/,tM) and bicuculline (50 HM) to block GABAA responses and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 HM) to prevent bursting of CA3 neurons in the presence of GABAA antagonists [11]. Cell input resistance was >50 MfL resting membrane potential was more negative than -60 mV and spike amplitudes >80 mV. All drugs were from Sigma (FRG), except 4-AP which was from Merck (FRG), CNQX was from Tocris Neuramin (UK). (-)Baclofen, CGP 35348 and CGP 55845A were kindly provided by Ciba-Geigy Ltd. (Basel, Switzerland). CGP 55845A is the pure S,S-diastereoisomer of the 1:1 mixture of S,S and S,R-diastereoisomers, CGP 54062, described by Olpe et al. [15].

[3] 75w Olpe et al. [15] have demonstrated that CGP 54062 which contains CGP 55845A used in the present study blocks the locally evoked slow K-IPSP in rat CA1 neurons, but does not affect the GABA g receptor-mediated CI-IPSE We could confirm this effect for CGP 55845A in guinea pig CA3 neurons. CGP 55845A in a concentration as low as 50 nM blocked the slow K-IPSP evoked by local electrical stimulation after 20 min of superfusion (n=2).

[4] 175w Low affinity GABA-antagonists currently in use are less potent blockers OflGABA than oflBa c [18]. Also, 4-APinduced K-IPSCs are blocked only by higher concentrations of these antagonists than they are required for the blockade of Iaa ~ [6]. This was taken to indicate that a heterogeneity of postsynaptic GABAB receptors might exist [18]. In line with this assumed heterogeneity is the observation that the K-conductance underlying IBac is blocked by caesium in a voltage-dependent manner while 4-AP-induced K-IPSCs are not [6]. Further, Iaa c is modulated by muscarinic receptor stimulation, but electrically evoked K-IPSCs are not [12]. CGP 55845A is not a useful drug for the discrimination of postsynaptic GABA B receptor subtypes, although, there were small differences in the concentrations in which it blocked /Bac, /GABA and K-IPSCs. Instead, CGP 55845A is a very potent, specific GABA B antagonist which strongly blocks Iaac,/GABA and K-IPSCs. In experiments in which slow reversibility is of no concern this drug is, therefore, very well suited for the study of the physiological role of GABAB receptor-mediated postsynaptic inhibition.