Search

The pre- and postsynaptic effects of baclofen, a broad-spectrum gamma-aminobutyric acid (GABA)B receptor agonist, and gabapentin, a selective agonist at GABA(B) receptors composed of GABA(B)(1a,2) heterodimers, were examined in CA1 pyramidal cells using whole-cell patch-clamp recordings in hippocampal slices from different strains of mice. In slices from C57BL/6 mice, by means of GABA(B) receptors, gabapentin and baclofen activated outward K+ currents at resting membrane potential. In weaver mice with a Kir3.2 channel mutation, baclofen and gabapentin failed to activate postsynaptic K+ currents. However, in littermate controls of weaver mice, gabapentin failed to evoke K+ currents, whereas baclofen activated currents in the same cells. Thus, postsynaptic actions of gabapentin and baclofen on K+ currents are different in this mouse strain. Via presynaptic GABA(B) receptors, baclofen significantly reduced GABA(A) inhibitory postsynaptic currents (IPSCs) in slices from C57BL/6 mice, as well as weaver and control mice. In contrast, gabapentin did not affect IPSCs significantly in any group of mice. These results indicate that although baclofen and gabapentin are agonists at postsynaptic GABA(B) receptors positively coupled to K+ channels, their mechanism of action differs in certain strains of mice, including the weaver wild-type mice, suggesting a dissociation in their signaling mechanism and coupling to K+ channels.

The mechanisms of Ca2+ responses evoked in hippocampal glial cells in situ, by local application of glutamate and by synaptic activation, were studied in slices from juvenile rats using the membrane permeant fluorescent Ca2+ indicator fluo-3AM and confocal microscopy. Ca2+ responses induced by local application of glutamate were unaffected by the sodium channel blocker tetrodotoxin and were therefore due to direct actions on glial cells. Glutamate-evoked responses were significantly reduced by the L-type Ca2+ channel blocker nimodipine, the group I/II metabotropic glutamate receptor antagonist (S)-alpha-methyl-4-carboxyphenylglycine (MCPG), and the N-methyl-D-aspartate (NMDA) receptor antagonist (+/-)2-amino-5-phosphonopentanoic acid (APV). However, glutamate-induced Ca2+ responses were not significantly reduced by the non-NMDA receptor antagonist 6-cyano-7-nitro-quinoxaline-2,3-dione (CNQX). These results indicate that local application of glutamate increases intracellular Ca2+ levels in glial cells via the activation of L-type Ca2+ channels, NMDA receptors, and metabotropic glutamate receptors. Brief (1 s) tetanization of Schaffer collaterals produced increases in intracellular Ca2+ levels in glial cells that were dependent on the frequency of stimulation (> or =50 Hz) and on synaptic transmission (abolished by tetrodotoxin). These Ca2+ responses were also antagonized by the L-type Ca2+ channel blocker nimodipine and the metabotropic glutamate receptor antagonist MCPG. However, the non-NMDA receptor antagonist CNQX significantly reduced the Schaffer collateral-evoked Ca2+ responses, while the NMDA antagonist APV did not. Thus, these synaptically mediated Ca2+ responses in glial cells involve the activation of L-type Ca2+ channels, group I/II metabotropic glutamate receptors, and non-NMDA receptors. These findings indicate that increases in intracellular Ca2+ levels induced in glial cells by local glutamate application and by synaptic activity share similar mechanisms (activation of L-type Ca2+ channels and group I/II metabotropic glutamate receptors) but also have distinct components (NMDA vs. non-NMDA receptor activation, respectively). Therefore, neuron-glia interactions in rat hippocampus in situ involve multiple, complex Ca2+-mediated processes that may not be mimicked by local glutamate application.

Long-term potentiation (LTP) of excitatory synaptic responses of principal neurons in the hippocampus is accompanied by changes in GABAergic inhibition mediated by interneurons. The impact of inhibition on LTP of excitatory postsynaptic responses in CA1 pyramidal cells was assessed by monitoring changes in field potentials evoked by Schaffer collateral stimulation in hippocampal slices in vitro. First, to determine the effect of inhibition on population EPSPs, slices were exposed to the GABA(A) receptor antagonist bicuculline (10 microM). Both the slope and amplitude of field EPSPs (fEPSPs) were significantly enhanced by bicuculline indicating that inhibition modulates excitatory postsynaptic responses of pyramidal cells. To assess if stimulation-dependent changes in inhibition influence LTP of excitatory responses of pyramidal cells, LTP was examined in the presence and absence of bicuculline (20 microM) following either 100 Hz tetanization, or theta-patterned stimulation (short bursts delivered at 5 Hz). In normal medium, 100 Hz stimulation produced marked short-term potentiation that decayed 5-10 min post-tetanus and both stimulation paradigms produced similar LTP at 30 min post-tetanus. In comparison, LTP of the fEPSP slope and amplitude was significantly enhanced after theta-patterned stimulation, but not after 100 Hz stimulation, in bicuculline. The greater potentiation of field responses following theta-patterned stimulation in the presence of bicuculline indicates that a larger potentiation of excitatory responses was unmasked during suppression of inhibitory inputs. These results suggest that a long-lasting enhancement of inhibition in pyramidal cells was also induced following theta-patterned stimulation in normal ACSF. Since suppression of inhibition did not uncover a significantly larger potentiation following 100 Hz tetanization, the influence of inhibition on LTP of excitatory responses appears to be stimulation-dependent. In conclusion, theta-patterned stimulation appears to be more effective at inducing plasticity within inhibitory circuits, and this plasticity may partially offset concurrent increases in the excitability of the CA1 network.

To examine the inhibitory postsynaptic potentials (IPSPs) elicited in pyramidal cells by interneurons situated at the stratum oriens/alveus border (O/A), glutamate was applied by micropressure to this area during intracellular recordings from CA1 pyramidal cells. Glutamate stimulation evoked IPSPs (glut-IPSPs) of small amplitude (4 mV), delayed peak latency (100-110 ms), and long duration (300-400 ms). Recurrent activation of interneurons via glutamate stimulation of pyramidal cells by local application in stratum pyramidale (PYR) evoked recurrent IPSPs (PYR glut-IPSPs) with similar amplitude and time course as O/A glut-IPSPs. The mean equilibrium potential of O/A glut-IPSPs (-77 mV) was significantly different from that of the PYR glut-IPSPs (-71 mV), however, neither equilibrium potential was significantly different from that of the electrically evoked early IPSP in the same cells. Glutamate-evoked IPSPs elicited from O/A displayed some response reversal (27% reversal) like those evoked from PYR (41% reversal). The early IPSP evoked by electrical stimulation displayed significantly more response reversal (67% reversal) than glut-IPSPs. Both types of glut-IPSPs (O/A and PYR) were associated with moderate increases in membrane conductance (5.9 and 6.6 nS, respectively), which were significantly less than the conductance change associated with the early IPSP (45.8 nS). In interneurons within PYR, glutamate stimulation in PYR readily elicited a flurry of excitatory postsynaptic potentials, whereas glutamate stimulation in O/A elicited IPSPs. The electrophysiological properties of IPSPs elicited in pyramidal cells by glutamate stimulation of interneurons in O/A were similar to those of recurrent IPSPs evoked from PYR. Given that both of these types of glutamate-evoked IPSPs were mostly mediated via GABAA receptor channels (Samulack DD, Lacaille J-C, 1993, Hippocampus 3:345-358), the small differences observed between equilibrium potentials, response reversals, and conductance changes could be due to a more electronically distant location from the soma of the synapses involved in O/A glut-IPSPs as compared to those of recurrent IPSPs elicited from PYR.

The receptor type mediating the inhibitory postsynaptic potentials (glut-IPSPs), recorded in CA1 pyramidal cells, as a result of glutamate stimulation of interneurons in stratum oriens near the alveus (O/A) was assessed and compared to the type mediating recurrent IPSPs evoked by recurrent activation of interneurons through glutamate stimulation of pyramidal cells in stratum pyramidale (PYR). In response to repetitive electrical stimulation, the peak amplitude of both the O/A glut-IPSP and the PYR glut-IPSP was attenuated (n = 5) in parallel to the reduction in amplitude of the early and late components of the electrically evoked response (stimulus-evoked disinhibition). This suggested the involvement of GABAergic receptors and attested that the interneurons activated during glut-IPSPs were also involved in the circuitry of the electrically evoked IPSPs. The local application of the selective GABAA antagonist bicuculline (100-200 microM) to the slice resulted in a significant reduction in the amplitude of both the O/A (by 76.5%; n = 9) and PYR (by 86.2%; n = 5) glut-IPSPs, in parallel to a decrease of the electrically evoked early IPSP, but not of the late IPSP. The presence of the GABAB antagonist 2-hydroxy-saclofen (1 mM) was able to significantly reduce the amplitude of the O/A glut-IPSPs (by 27.5%; n = 7) and of the electrically evoked late IPSP, but not the PYR glut-IPSP (n = 3). Although the application of phaclofen (20 mM) to the slice reduced the amplitude of the O/A glut-IPSPs (n = 3), the reduction was not statistically significant. These results suggest that recurrent IPSPs elicited from activation of interneurons by stimulation of pyramidal cells are mediated solely via GABAA receptors. Inhibitory postsynaptic potentials elicited from stimulation of interneurons in O/A were also mediated mostly by GABAA receptors, but in addition, displayed a minor component mediated by GABAB receptors. Therefore, since a large proportion of interneurons in O/A are recurrently excited by pyramidal cells (Lacaille J-C et al., 1987, J Neurosci 7: 1979-1993), and since recurrent IPSPs appeared mediated by GABAA receptors, a subpopulation of interneurons activated from O/A might exist that do not receive recurrent excitation but can inhibit pyramidal cells via GABAB receptors.