Limits...
Hippocampal Theta Input to the Amygdala Shapes Feedforward Inhibition to Gate Heterosynaptic Plasticity.

Bazelot M, Bocchio M, Kasugai Y, Fischer D, Dodson PD, Ferraguti F, Capogna M - Neuron (2015)

Bottom Line: These effects are mediated by GABAB receptors and change in the Cl(-) driving force.Hence, feedforward inhibition, known to enforce temporal fidelity of excitatory inputs, dominates hippocampus-amygdala interactions to gate heterosynaptic plasticity.VIDEO ABSTRACT.

View Article: PubMed Central - PubMed

Affiliation: MRC Brain Network Dynamics Unit, Department of Pharmacology, University of Oxford, Mansfield Road, Oxford OX1 3TH, UK.

Show MeSH

Related in: MedlinePlus

Presynaptic GABAB Receptors at BA IN-PN Synapses Mediate Depression of FFI(A) Left, schematic of the experimental configuration. Local electrical stimulation is applied while recording from PNs. Right, representative current clamp recording (−65 mV) showing monosynaptic IPSP induced by local electrical stimulation in presence of NBQX/APV. The IPSP was abolished by the GABAB receptor agonist baclofen (10 μM). During baclofen application, neurons hyperpolarized but were held at −65 mV to avoid changes in driving force for Cl−.(B) Pooled data, (n = 6).(C) PPR of IPSCs recorded in PNs and evoked by local electrical stimulation. PPR reduced with shorter ISIs.(D) The depression of IPSPs during TBS occurred in a PN recorded at 0 mV with cesium-containing intracellular solution.(E) Pooled data (n = 7).(F) Trains of ten single pulses were delivered with variable ISIs (0.5, 1, 2, 5, 10, 15, and 20 s) while recording from PNs. Top, representative trace during which light pulses were delivered at 0.5-s intervals. Depression of the IPSP was largely blocked by CGP54626. Bottom, average of the last three IPSPs evoked for each ISI normalized to the first IPSP. The IPSP amplitude became smaller with shorter ISIs. This depression was reduced when stimulation evoked only GABAA receptor IPSPs or in presence of CGP54626 (5 μM).(G) Averaged amplitudes of last three IPSPs elicited with a train of ten single pulses delivered every 0.5 s: IPSP amplitude in control conditions (n = 9) was significantly lower than in presence of CGP54626 (p < 0.001 versus control, n = 7) or in PNs where only a GABAA IPSP was evoked (p < 0.001 versus control, n = 6).(H1) Electron micrograph of a GABAergic terminal in the BA as obtained with the SDS-digested FRIL method. Scale bar: 500 nm.(H2) Enlarged view of the area within the dashed lines in (H1). Larger gold particles (10 nm) reveal VGAT, whereas smaller gold particles (5 nm; arrowheads) identify GABAB receptors. Scale bar: 500 nm.(I) Frequency distribution of VGAT+ axon terminals with different density of GABAB receptor labeling (n = 179 synapses obtained from three mice). Data are presented as means ± SEM.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4590554&req=5

fig7: Presynaptic GABAB Receptors at BA IN-PN Synapses Mediate Depression of FFI(A) Left, schematic of the experimental configuration. Local electrical stimulation is applied while recording from PNs. Right, representative current clamp recording (−65 mV) showing monosynaptic IPSP induced by local electrical stimulation in presence of NBQX/APV. The IPSP was abolished by the GABAB receptor agonist baclofen (10 μM). During baclofen application, neurons hyperpolarized but were held at −65 mV to avoid changes in driving force for Cl−.(B) Pooled data, (n = 6).(C) PPR of IPSCs recorded in PNs and evoked by local electrical stimulation. PPR reduced with shorter ISIs.(D) The depression of IPSPs during TBS occurred in a PN recorded at 0 mV with cesium-containing intracellular solution.(E) Pooled data (n = 7).(F) Trains of ten single pulses were delivered with variable ISIs (0.5, 1, 2, 5, 10, 15, and 20 s) while recording from PNs. Top, representative trace during which light pulses were delivered at 0.5-s intervals. Depression of the IPSP was largely blocked by CGP54626. Bottom, average of the last three IPSPs evoked for each ISI normalized to the first IPSP. The IPSP amplitude became smaller with shorter ISIs. This depression was reduced when stimulation evoked only GABAA receptor IPSPs or in presence of CGP54626 (5 μM).(G) Averaged amplitudes of last three IPSPs elicited with a train of ten single pulses delivered every 0.5 s: IPSP amplitude in control conditions (n = 9) was significantly lower than in presence of CGP54626 (p < 0.001 versus control, n = 7) or in PNs where only a GABAA IPSP was evoked (p < 0.001 versus control, n = 6).(H1) Electron micrograph of a GABAergic terminal in the BA as obtained with the SDS-digested FRIL method. Scale bar: 500 nm.(H2) Enlarged view of the area within the dashed lines in (H1). Larger gold particles (10 nm) reveal VGAT, whereas smaller gold particles (5 nm; arrowheads) identify GABAB receptors. Scale bar: 500 nm.(I) Frequency distribution of VGAT+ axon terminals with different density of GABAB receptor labeling (n = 179 synapses obtained from three mice). Data are presented as means ± SEM.

Mentions: The GABAB-receptor-mediated effects could be due to the depression of either glutamatergic synapses from vCA1 fibers onto BA INs and/or GABAergic synapses from BA INs onto PNs. We distinguished between these two possibilities by recording directly from GABAergic INs with the soma in the BA and ITCs (Figure S5). The EPSPs recorded in these cells displayed a gradual increase in amplitude through the TBS (Figure 5F, tenth versus first train p < 0.05, n = 23). Hence, vCA1-BA INs excitatory synapses could not be responsible for the transient depression of the IPSPs recorded from PNs. Furthermore, the amplitude of the EPSPs recorded from PNs was stable in presence of antagonists for GABA receptors (Figure 5A, F, tenth versus first train p > 0.05, n = 6), indicating that vCA1-BA PNs excitatory synapses do not undergo depression during TBS. Next, we tested whether presynaptic mechanisms contribute to the depression of the IPSP amplitude. Monosynaptic IPSPs were triggered in PNs through local electrical stimulation in the presence of NBQX (10 μM) and APV (100 μM). The application of the GABAB receptor agonist baclofen (10 μM) abolished the monosynaptic IPSPs (Figures 7A and 7B, n = 5, p < 0.01). The magnitude of this effect suggests that activation of presynaptic GABAB receptors and an increase in postsynaptic membrane conductance induced by baclofen accounted for this effect. To further corroborate the involvement of presynaptic GABAB receptors in the depression of the IPSP amplitude, additional experiments were carried out. First, electrical stimulation was delivered while recording postsynaptic neurons in voltage-clamp mode. The paired pulse ratio (PPR) of electrically evoked inhibitory postsynaptic currents (IPSCs), a parameter critically dependent on presynaptic GABAB receptors (Davies et al., 1990), became smaller with shorter inter-stimuli intervals (ISIs) (Figure 7C, n = 5). Second, we observed depression of IPSPs during optical TBS using a cesium-based intracellular solution to block postsynaptic cesium-sensitive K+ channels linked to GABAB receptors (Figures 7D and 7E, n = 7) (Gähwiler and Brown, 1985). Third, we delivered trains of ten single light pulses with varying ISI; the IPSP amplitude was inversely related to the ISIs (Figure 7F). In the presence of 5 μM CGP54626, such use-dependent depression of the IPSPs was significantly reduced (p < 0.001 versus control, Figures 7F and 7G). Likewise, TBS-induced depression of the IPSPs was attenuated in PNs in which GABAA-only IPSPs were evoked (p < 0.001 versus control, Figures 7F and 7G). This suggests that when spillover of GABA on extrasynaptic GABAB receptors at postsynaptic site is weak or absent, resulting in GABAA only IPSP, spillover of GABA on presynaptic GABAB receptor (López-Bendito et al., 2004) is also less likely to occur, ensuing more stable IPSPs. Fourth, we detected GABAB receptors at the axon terminals of INs using the SDS-digested freeze-fracture replica immunolabeling (FRIL) method (Figures 7H and 7I). Taken together, these data demonstrate that vCA1-driven FFI on BA PNs is dynamically modulated by GABAB receptors located on the axon terminals of GABAergic INs. What could be the consequences of FFI dynamics for synaptic integration and plasticity in PNs?


Hippocampal Theta Input to the Amygdala Shapes Feedforward Inhibition to Gate Heterosynaptic Plasticity.

Bazelot M, Bocchio M, Kasugai Y, Fischer D, Dodson PD, Ferraguti F, Capogna M - Neuron (2015)

Presynaptic GABAB Receptors at BA IN-PN Synapses Mediate Depression of FFI(A) Left, schematic of the experimental configuration. Local electrical stimulation is applied while recording from PNs. Right, representative current clamp recording (−65 mV) showing monosynaptic IPSP induced by local electrical stimulation in presence of NBQX/APV. The IPSP was abolished by the GABAB receptor agonist baclofen (10 μM). During baclofen application, neurons hyperpolarized but were held at −65 mV to avoid changes in driving force for Cl−.(B) Pooled data, (n = 6).(C) PPR of IPSCs recorded in PNs and evoked by local electrical stimulation. PPR reduced with shorter ISIs.(D) The depression of IPSPs during TBS occurred in a PN recorded at 0 mV with cesium-containing intracellular solution.(E) Pooled data (n = 7).(F) Trains of ten single pulses were delivered with variable ISIs (0.5, 1, 2, 5, 10, 15, and 20 s) while recording from PNs. Top, representative trace during which light pulses were delivered at 0.5-s intervals. Depression of the IPSP was largely blocked by CGP54626. Bottom, average of the last three IPSPs evoked for each ISI normalized to the first IPSP. The IPSP amplitude became smaller with shorter ISIs. This depression was reduced when stimulation evoked only GABAA receptor IPSPs or in presence of CGP54626 (5 μM).(G) Averaged amplitudes of last three IPSPs elicited with a train of ten single pulses delivered every 0.5 s: IPSP amplitude in control conditions (n = 9) was significantly lower than in presence of CGP54626 (p < 0.001 versus control, n = 7) or in PNs where only a GABAA IPSP was evoked (p < 0.001 versus control, n = 6).(H1) Electron micrograph of a GABAergic terminal in the BA as obtained with the SDS-digested FRIL method. Scale bar: 500 nm.(H2) Enlarged view of the area within the dashed lines in (H1). Larger gold particles (10 nm) reveal VGAT, whereas smaller gold particles (5 nm; arrowheads) identify GABAB receptors. Scale bar: 500 nm.(I) Frequency distribution of VGAT+ axon terminals with different density of GABAB receptor labeling (n = 179 synapses obtained from three mice). Data are presented as means ± SEM.
© Copyright Policy - CC BY
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4590554&req=5

fig7: Presynaptic GABAB Receptors at BA IN-PN Synapses Mediate Depression of FFI(A) Left, schematic of the experimental configuration. Local electrical stimulation is applied while recording from PNs. Right, representative current clamp recording (−65 mV) showing monosynaptic IPSP induced by local electrical stimulation in presence of NBQX/APV. The IPSP was abolished by the GABAB receptor agonist baclofen (10 μM). During baclofen application, neurons hyperpolarized but were held at −65 mV to avoid changes in driving force for Cl−.(B) Pooled data, (n = 6).(C) PPR of IPSCs recorded in PNs and evoked by local electrical stimulation. PPR reduced with shorter ISIs.(D) The depression of IPSPs during TBS occurred in a PN recorded at 0 mV with cesium-containing intracellular solution.(E) Pooled data (n = 7).(F) Trains of ten single pulses were delivered with variable ISIs (0.5, 1, 2, 5, 10, 15, and 20 s) while recording from PNs. Top, representative trace during which light pulses were delivered at 0.5-s intervals. Depression of the IPSP was largely blocked by CGP54626. Bottom, average of the last three IPSPs evoked for each ISI normalized to the first IPSP. The IPSP amplitude became smaller with shorter ISIs. This depression was reduced when stimulation evoked only GABAA receptor IPSPs or in presence of CGP54626 (5 μM).(G) Averaged amplitudes of last three IPSPs elicited with a train of ten single pulses delivered every 0.5 s: IPSP amplitude in control conditions (n = 9) was significantly lower than in presence of CGP54626 (p < 0.001 versus control, n = 7) or in PNs where only a GABAA IPSP was evoked (p < 0.001 versus control, n = 6).(H1) Electron micrograph of a GABAergic terminal in the BA as obtained with the SDS-digested FRIL method. Scale bar: 500 nm.(H2) Enlarged view of the area within the dashed lines in (H1). Larger gold particles (10 nm) reveal VGAT, whereas smaller gold particles (5 nm; arrowheads) identify GABAB receptors. Scale bar: 500 nm.(I) Frequency distribution of VGAT+ axon terminals with different density of GABAB receptor labeling (n = 179 synapses obtained from three mice). Data are presented as means ± SEM.
Mentions: The GABAB-receptor-mediated effects could be due to the depression of either glutamatergic synapses from vCA1 fibers onto BA INs and/or GABAergic synapses from BA INs onto PNs. We distinguished between these two possibilities by recording directly from GABAergic INs with the soma in the BA and ITCs (Figure S5). The EPSPs recorded in these cells displayed a gradual increase in amplitude through the TBS (Figure 5F, tenth versus first train p < 0.05, n = 23). Hence, vCA1-BA INs excitatory synapses could not be responsible for the transient depression of the IPSPs recorded from PNs. Furthermore, the amplitude of the EPSPs recorded from PNs was stable in presence of antagonists for GABA receptors (Figure 5A, F, tenth versus first train p > 0.05, n = 6), indicating that vCA1-BA PNs excitatory synapses do not undergo depression during TBS. Next, we tested whether presynaptic mechanisms contribute to the depression of the IPSP amplitude. Monosynaptic IPSPs were triggered in PNs through local electrical stimulation in the presence of NBQX (10 μM) and APV (100 μM). The application of the GABAB receptor agonist baclofen (10 μM) abolished the monosynaptic IPSPs (Figures 7A and 7B, n = 5, p < 0.01). The magnitude of this effect suggests that activation of presynaptic GABAB receptors and an increase in postsynaptic membrane conductance induced by baclofen accounted for this effect. To further corroborate the involvement of presynaptic GABAB receptors in the depression of the IPSP amplitude, additional experiments were carried out. First, electrical stimulation was delivered while recording postsynaptic neurons in voltage-clamp mode. The paired pulse ratio (PPR) of electrically evoked inhibitory postsynaptic currents (IPSCs), a parameter critically dependent on presynaptic GABAB receptors (Davies et al., 1990), became smaller with shorter inter-stimuli intervals (ISIs) (Figure 7C, n = 5). Second, we observed depression of IPSPs during optical TBS using a cesium-based intracellular solution to block postsynaptic cesium-sensitive K+ channels linked to GABAB receptors (Figures 7D and 7E, n = 7) (Gähwiler and Brown, 1985). Third, we delivered trains of ten single light pulses with varying ISI; the IPSP amplitude was inversely related to the ISIs (Figure 7F). In the presence of 5 μM CGP54626, such use-dependent depression of the IPSPs was significantly reduced (p < 0.001 versus control, Figures 7F and 7G). Likewise, TBS-induced depression of the IPSPs was attenuated in PNs in which GABAA-only IPSPs were evoked (p < 0.001 versus control, Figures 7F and 7G). This suggests that when spillover of GABA on extrasynaptic GABAB receptors at postsynaptic site is weak or absent, resulting in GABAA only IPSP, spillover of GABA on presynaptic GABAB receptor (López-Bendito et al., 2004) is also less likely to occur, ensuing more stable IPSPs. Fourth, we detected GABAB receptors at the axon terminals of INs using the SDS-digested freeze-fracture replica immunolabeling (FRIL) method (Figures 7H and 7I). Taken together, these data demonstrate that vCA1-driven FFI on BA PNs is dynamically modulated by GABAB receptors located on the axon terminals of GABAergic INs. What could be the consequences of FFI dynamics for synaptic integration and plasticity in PNs?

Bottom Line: These effects are mediated by GABAB receptors and change in the Cl(-) driving force.Hence, feedforward inhibition, known to enforce temporal fidelity of excitatory inputs, dominates hippocampus-amygdala interactions to gate heterosynaptic plasticity.VIDEO ABSTRACT.

View Article: PubMed Central - PubMed

Affiliation: MRC Brain Network Dynamics Unit, Department of Pharmacology, University of Oxford, Mansfield Road, Oxford OX1 3TH, UK.

Show MeSH
Related in: MedlinePlus