Background synaptic activity in rat entorhinal cortical neurones: differential control of transmitter release by presynaptic receptors

Abstract

The entorhinal cortex (EC) is a key brain area controlling both hippocampal input and output via neurones in layer II and layer V, respectively. It is also a pivotal area in the generation and propagation of epilepsies involving the temporal lobe. We have previously shown that within the network of the EC, neurones in layer V are subject to powerful synaptic excitation but weak inhibition, whereas the reverse is true in layer II. The deep layers are also highly susceptible to acutely provoked epileptogenesis. Considerable evidence now points to a role of spontaneous background synaptic activity in control of neuronal, and hence network, excitability. In the present article we describe results of studies where we have compared background release of the excitatory transmitter, glutamate, and the inhibitory transmitter, GABA, in the two layers, the role of this background release in the balance of excitability, and its control by presynaptic auto- and heteroreceptors on presynaptic terminals.

Figure 1. Frequency-dependent changes in synaptic potentials in the EC

Intracellular recordings were made from neurones in either layer II or layer V in interface slices of rat EC. Synaptic responses were evoked by stimulating afferent pathways, and using appropriate receptor antagonists to isolate different components of the synaptic EPSPS and IPSPs (e.g. 2-AP5, bicuculline and CGP55845A for AMPAr-mediated EPSPs). Amplitudes of each component were determined as a percentage of the control response evoked at low frequency (0.1 Hz). At least 7 neurones were tested in each case. A, the facilitation of AMPA receptor-mediated EPSPs was slightly but not significantly greater in layer V at all three higher frequencies. NMDA-dependent EPSPs, however, showed a greater facilitation in layer V, which was significant (P > 0.05, paired t test) at 2 and 3 Hz. B, frequency-dependent depression of both GABA_A- and GABA_B-mediated IPSPs was significantly (P > 0.05, paired t test) greater in layer V at both 2 and 3 Hz. (Data adapted from Dhillon, 1996 and A. Dhillon & R.S.G. Jones, unpublished observations).

Figure 2. Spontaneous synaptic excitation in the EC

All recordings in this and subsequent figures (3–6) were made in submerged slices of rat EC using whole cell patch clamp recordings at a holding potential of −60 mV, to monitor transmitter release. A, records at the top show sEPSCs in individual layer V and layer II neurones. sEPSCs were on average larger and more frequent in layer V neurones compared to layer II. Note also the burst of sEPSCs at the beginning of the upper trace in the layer V neurone, which were prevalent in deep but not superficial neurones. The graphs show cumulative probability curves for pooled data (200 events per neurone, n = 17 neurones in each layer) for interevent interval and peak amplitudes. B, the traces are averaged sEPSCs (n = 20 in each case). In the layer V neurone, addition of the NMDAr antagonist, 2-AP5, reduced both the peak and decay phase of the sEPSCs, whereas neither was greatly altered in the layer II neurone. These effects are confirmed by the cumulative probability analyses of peak amplitude and time to 50% decay of sEPSCs in pooled data (at least 150 events per neurone both before and after drug application, n = 7 neurones in each case) shown in the graphs below the traces. C, pure NMDA-mediated sEPSCs could be recorded. The records are from a layer V neurone. Addition of the AMPA receptor antagonist, NBQX, removed most of the sEPSCs although small occasional events with slow kinetics remained. These were increased in number and amplitude by removal of the voltage-dependent block of the NMDAr in Mg²⁺-free medium. Subsequent addition of the NMDAr antagonist abolished the slow sEPSCs. The frequency histogram shows the distribution of event amplitudes in the presence of the AMPAr antagonist alone and following removal of Mg²⁺. (Data adapted from Berretta & Jones, 1996a.)

Figure 3. Spontaneous synaptic inhibition in the EC

The voltage clamp traces were recorded at a holding potential of −60 mV with symmetrical intra- and extracellular Cl⁻ concentrations and show sIPSCs mediated by activation of GABA_A receptors (see Bailey et al. 2004). The frequency of sIPSCs was dramatically higher in layer II compared to layer V. The pooled analyses of data from 15 neurones shown in the graphs confirm the much greater frequency in layer II and show that there was a slightly greater preponderance of larger amplitude events in the superficial neurones, and this was reflected by slightly larger mean and median amplitude values in layer V. (Data from Woodhall et al. 2004.)

Figure 4. Presynaptic NMDA receptors in the EC

All recordings in this figure were conducted with the open channel NMDAr blocker, MK801 in the patch pipette. This approach allows blockade of postsynaptic NMDAr on the recorded neurone (see Berretta & Jones, 1996b). A, the records show AMPAr-mediated miniature EPSCs in a layer V neurone recorded in the presence of 1 μm TTX. Addition of an NMDA receptor agonist, homoquinolinic acid (HQA), dramatically increased the frequency of sEPSCs and resulted in a shift towards larger amplitude events. The graphs show pooled data from 5 layer V neurones. B, the facilitatory effect of the presynaptic NMDAr on glutamate release was tonically operative, since the NMDAr antagonist, 2-AP5, reduced the frequency of mEPSCs with little effect on amplitude. C, presynaptic NMDAr are present in both layer V and layer II. However, the summary data (n = 5, layer V and n = 4 layer II) show a greater effect of the agonist in increasing EPSC frequency in layer V and a greater reduction in frequency with the antagonist, compared to layer II neurones. D, presynaptic NMDAr also act to facilitate GABA release in the EC. The records are GABA_Ar-mediated mIPSCs recorded in TTX in a layer II neurone. Addition of HQA increased the frequency with little effect on amplitude. This effect was weak compared to that seen with mEPSCs, and the pooled summary data in E show that this effect is restricted to layer II, and that the NMDAr antagonist did not alter mIPSC frequency in either layer. (Data adapted from Woodhall et al. 2001a.)

Figure 5. Presynaptic mGlurs in the EC

A, addition of ACPT-1 (20 μm), a specific agonist at the group III receptors, greatly increased the frequency of mEPSCs recorded in the presence of TTX (1 μm) in a layer V neurone. In contrast, the same agonist significantly decreased the frequency of mEPSCs in layer II. The summary data (14 layer V neurones and 5 layer II) show a doubling of mEPSC frequency in the deep layer and around a 30% decrease in layer II. Neither effect is tonically operative since the group III antagonist, CPPG had no effect on frequency in either layer. B, mGlur activation reduces the frequency of sIPSCs in layer V neurones, but not in layer II. The summary data (n = 10 in each layer) show this differential effect and also that again the group III antagonist does not affect frequency in either layer. (Data adapted from Evans et al. 2000 and Woodhall et al. 2001b.)

Figure 6. Presynaptic GABA_B receptors in the EC

A, summary data show that activation of GABA_Br with the agonist CGP 44533 (10 μm) reduced the frequency of mIPSCs recorded in both layer II and layer V to a similar extent (n = 17 in each case). A similar reduction in mEPSC frequency was seen in layer V, but in layer II, this heteroreceptor effect was significantly more pronounced. B, the GABA_Br was tonically operative at autoreceptors in layer V neurones, as the antagonist CGP 55845 increased the frequency of sIPSCs. C, the summary data show that the tonic autoreceptor effect is restricted to layer V, and is not present in layer II. In addition, the same antagonist, CGP 55845 (10 μm), failed to affect the frequency of mEPSCs in either layer. (Data adapted from Bailey et al. 2004 and Thompson, 2004 including unpublished observations from S.E. Thompson & R.S.G. Jones).