Pre- and postsynaptic inhibition mediated by GABA(B) receptors in cerebellar inhibitory interneurons.

Abstract

The inhibitory interneurons in the molecular layer of the cerebellar cortex form a complex network, interconnected by both chemical and electrotonic synapses. Previous work, using voltage optical imaging in an isolated cerebellum, has indicated that these interneurons also form presynaptic inhibitory interconnections. Here we examine the participation of GABA(B) receptors in the proposed presynaptic inhibition by recording from the molecular layer interneurons (MLI) in cerebellar slices. The GABA(B) agonist, baclofen, profoundly depressed synaptic transmission; a concentration of 10 microM decreased the frequency of spontaneous inhibitory synaptic potentials by 82 +/- 15% and of miniature synaptic potentials by 75 +/- 13%. In simultaneous recording from two synaptically interconnected MLIs, baclofen (10 microM) increased the failure rate of synaptic transmission by a factor of 3, confirming a presynaptic mechanism, most likely mediated by a decrease in calcium conductance. A postsynaptic effect of baclofen was also found; 10 microM decreased the spontaneous firing rate by 55 +/- 19% even in the presence of synaptic blockers. One hundred micromolar baclofen induced an averaged hyperpolarization of 6 +/- 2 mV or an averaged 7.8 +/- 3 pA net outward current that can account for the decrease in firing rate. The outward current reflects a reduction in a tonic Ca(2+) current, since it was abolished by blocking Ca(2+) currents and remained unchanged in the presence of Ba(2+). Application of the specific GABA(B) blocker, CGP 55845A (1 microM), not only reversed the effects of baclofen but also increased the spontaneous firing rate and synaptic activity when applied alone. Thus in slice preparations, GABA(B) receptors are tonically activated by endogenous GABA. The temporal role of GABA(B) receptors was tested using the paired-pulse paradigm. Recording from two synaptically interconnected MLIs showed a 3.5 times lower probability of release for the second stimulus. In the isolated cerebellar preparation, a robust depression of the second inhibitory response was observed. This depression was partially blocked by CGP 55845A (2 microM). We conclude that both the pre- and postsynaptic effects of baclofen are mediated by GABA(B) receptors that decrease Ca(2+) currents. These can serve a modulatory role as well as participating in shaping the temporal interactions between consecutive inputs.