Different subtypes of GABAB receptors are present at pre- and postsynaptic sites within the rat dorsolateral septal nucleus.

GABAB receptor activation modulates neuronal activity mediated by multiple CNS transmitters and can occur at pre- and postsynaptic sites. In low concentrations, baclofen acts presynaptically to diminish transmitter release via both hetero- and autoreceptors, whereas at increasing concentrations, the same compound alters postsynaptic membrane excitability by inducing a membrane hyperpolarization. We have utilized electrophysiological techniques in vitro to focus on the possibility that pharmacologically different subtypes of GABAB receptors are present on presynaptic sites of glutamatergic terminals when compared with GABAB receptors on postsynaptic sites within the dorsolateral septal nucleus (DLSN). The glutamatergic terminal within the DLSN originates from a pyramidal cell body located within the hippocampus and most likely terminates on a GABAergic neuron from which recordings were made. Whole cell patch voltage-clamp methods were employed to record pharmacologically isolated excitatory postsynaptic currents (EPSCs) from DLSN neurons as an index of glutamatergic transmission. Using a modified internal pipette solution containing QX-314 and in which CsGluconate and GDPbetaS replaced Kgluconate and GTP, respectively, we recorded isolated monosynaptic EPSCs. The GABAA receptor antagonists bicuculline and picrotoxin were included in the external standard superfusion solution. Application of the GABAB receptor agonists, (+/-)-baclofen, CGP44533, and CGP35024 (10 nM to 10 microM) depressed glutamate-mediated EPSCs in a concentration-dependent manner. With the use of this combination of solutions, CGP44533 did not produce postsynaptic membrane property changes. Under these conditions, both (+/-)-baclofen and CGP35024 still induced increases of postsynaptic membrane conductance associated with an outward current. The GABAB receptor antagonist CGP55845A (1 microM) blocked the presynaptic CGP44533-mediated depressant effects of EPSCs, whereas CGP35348 (100 microM) or barium (2 mM) was ineffective. Furthermore, both CGP35348 (100 microM) and CGP55845A (1 microM) were effective in blocking the postsynaptic conductance changes associated with baclofen and CGP35024, whereas barium was ineffective. Our results demonstrate a distinct pharmacology for GABAB agonists acting at putative subtypes of GABAB receptors located on presynaptic sites of a glutamatergic terminal versus GABAB receptors on postsynaptic sites of a DLSN neuron. Furthermore, our results also suggest a different pharmacology and/or coupling of a GABAB receptor to different effectors at postsynaptic sites within the DLSN. Thus there may be three or more pharmacologically distinct GABAB receptors or receptor complexes associated with DLSN neurons: at least one pre- and two postsynaptic. If this distinct pharmacology and GABAB receptor distribution also extends to other CNS structures, such differences could provide development of selective drugs to act at these multiple sites.