Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Nov;222(8):3677-3690.
doi: 10.1007/s00429-017-1427-x. Epub 2017 May 2.

Differential Surface Density and Modulatory Effects of Presynaptic GABA B Receptors in Hippocampal Cholecystokinin and Parvalbumin Basket Cells

Affiliations
Free PMC article

Differential Surface Density and Modulatory Effects of Presynaptic GABA B Receptors in Hippocampal Cholecystokinin and Parvalbumin Basket Cells

Sam A Booker et al. Brain Struct Funct. .
Free PMC article

Abstract

The perisomatic domain of cortical neurons is under the control of two major GABAergic inhibitory interneuron types: regular-spiking cholecystokinin (CCK) basket cells (BCs) and fast-spiking parvalbumin (PV) BCs. CCK and PV BCs are different not only in their intrinsic physiological, anatomical and molecular characteristics, but also in their presynaptic modulation of their synaptic output. Most GABAergic terminals are known to contain GABAB receptors (GABABR), but their role in presynaptic inhibition and surface expression have not been comparatively characterized in the two BC types. To address this, we performed whole-cell recordings from CCK and PV BCs and postsynaptic pyramidal cells (PCs), as well as freeze-fracture replica-based quantitative immunogold electron microscopy of their synapses in the rat hippocampal CA1 area. Our results demonstrate that while both CCK and PV BCs contain functional presynaptic GABABRs, their modulatory effects and relative abundance are markedly different at these two synapses: GABA release is dramatically inhibited by the agonist baclofen at CCK BC synapses, whereas a moderate reduction in inhibitory transmission is observed at PV BC synapses. Furthermore, GABABR activation has divergent effects on synaptic dynamics: paired-pulse depression (PPD) is enhanced at CCK BC synapses, but abolished at PV BC synapses. Consistent with the quantitative differences in presynaptic inhibition, virtually all CCK BC terminals were found to contain GABABRs at high densities, but only 40% of PV BC axon terminals contain GABABRs at detectable levels. These findings add to an increasing list of differences between these two interneuron types, with implications for their network functions.

Keywords: Electron microscope; GABAergic interneurons; Presynaptic inhibition; Short-term plasticity; Synaptic transmission.

Figures

Fig. 1
Fig. 1
Presynaptic GABABRs inhibit transmission at CB1R-expressing perisomatic GABAergic terminals more strongly than M2R-positive terminals. a Representative WIN-sensitive IPSCs in a CA1 PC evoked by minimal stimulation delivered to the CA1 str. pyramidale (top traces) under control conditions (Ctrl) and during sequential bath application of WIN-55212 (WIN, 1 µM, 2 min), baclofen (Bac, 10 µM, 5 min) and CGP-55845 (CGP, 10 µM, 5 min); control trace is underlain for all conditions for comparison. Time-course plots of the mean I WC (middle) and normalized amplitude of the WIN-sensitive IPSCs recorded from 10 CA1 PCs (bottom). b Effects of baclofen application on WIN-insensitive IPSCs presented as in a. These IPSCs were consistently inhibited by subsequent application of the selective M2R agonist, ABET (10 µM, 5 cells). Summary bar charts of the mean IPSC amplitudes elicited by WIN-sensitive (c) and insensitive afferents (d) measured at the end of each pharmacological epoch. Numbers of recorded PCs indicated in brackets. e Bar chart of the normalized IPSC amplitudes elicited by WIN-sensitive (WIN +ve) and insensitive afferents (WIN−ve) during Bac and CGP application. Statistics shown: ns P > 0.05, * P < 0.05, *** P < 0.001, Mann–Whitney or Wilcoxon signed rank tests, corrected for multiple comparisons
Fig. 2
Fig. 2
Unitary IPSCs produced in CA1 PCs by identified CCK and PV BCs show differential presynaptic GABABR-mediated inhibition. a Reconstruction of a synaptically coupled CCK BC (soma and dendrites in black, axon in red) and CA1 PC (soma and dendrites in blue) pair. Inset, top, voltage responses elicited in the CCK BCs and the CA1 PCs to hyper- to depolarizing current steps (−500 to 500 pA, 50 pA steps, 500 ms duration). Note the regular-spiking phenotype of the CCK BCs. Inset, right, immunolabeling for CCK (in green pseudocolor) in the biocytin-filled soma of the BC (in gray). Pyr., stratum pyramidale; Ori., stratum oriens; Rad., stratum radiatum; L-M, stratum lacunosum-moleculare. b upper, representative pairs of action potentials elicited in the same CCK BC as in a (upper traces, suprathreshold current pulses of 1–2 nA, 1 ms duration, 50 ms interval) resulted in unitary IPSCs in the post-synaptic CA1 PC (lower traces) under control conditions (left panel), followed by sequential bath application of baclofen (10 µM, middle panel), and CGP (10 µM; right panel); baclofen and CGP traces are underlain by control traces (gray). b lower, Time-course plot of the postsynaptic whole-cell current (I WC; top) and unitary IPSC amplitude (lower) from the same CCK BC–CA1 PC pair over the course of the experiment, including the initial control conditions and following application of baclofen and then CGP to the bath. Summary bar charts of synaptic parameters for CCK BC–CA1 PC pairs (8 cells) under control conditions and during baclofen and CGP bath application: IPSC amplitude (c), apparent failures of transmission (d), paired-pulse ratio (e) and latency distribution (CV, f). g–i Bar charts of synaptic parameters for PV BC–CA1 PC pairs (11 cells) under control conditions, baclofen and CGP application: IPSC amplitude (g), failure rate of IPSCs (h) and the paired-pulse ratio (i). j Summary bar chart of the normalized IPSC amplitudes for a comparison of baclofen-induced inhibition at CCK and PV BC synapses. CGP recovery is shown for both cell types. Statistics shown: ns P > 0.05, * P < 0.05, ** P < 0.01, *** P < 0.001, results from repeated measures ANOVA with Sidak’s multiple comparison test or Mann–Whitney tests (j), all data is derived from 8 CCK BC/CA1 PC and 11 PV BC/CA1 PC pairs
Fig. 3
Fig. 3
Identified CCK SCA cells show strong presynaptic GABABR-mediated inhibition. a Reconstruction of a synaptically coupled CCK SCA (soma and dendrites in black, axon in red) and a CA1 PC (soma and dendrites in blue, axon in green). Inset (top), voltage responses elicited in the CCK SCA cells and the CA1 PCs to hyper- to depolarizing current steps (−500 to 500 pA, 50 pA steps, 500 ms duration). Inset (right), immunolabeling for CCK (in green pseudocolor) in the soma of the SCA cell (in gray). Pyr., stratum pyramidale; Ori., stratum oriens; Rad., stratum radiatum. b, upper, Representative pairs of action potentials elicited in the SCA cell shown in a (upper traces, suprathreshold current pulses of 1–2 nA, 1 ms duration, 50 ms interval) were followed by short-latency IPSCs in the recorded CA1 PC (lower traces) under control conditions (left panel), during bath application of baclofen (10 µM, middle panel), and CGP (10 µM; right panel); baclofen and CGP are underlain by control traces (gray). b, lower, time-course plot of the postsynaptic whole-cell current (I WC; top) and unitary IPSC amplitude (lower) from the same CCK SCA cell–CA1 PC pair during the experiment, including control period, and subsequent baclofen and CGP applications. Summary bar charts of synaptic properties of CCK SCA–CA1 PC pairs (six pairs) during each pharmacological epoch: IPSC amplitude (c), apparent failures of transmission (d) and paired-pulse ratio (PPR, e). f Summary bar chart of the normalized IPSC amplitudes during baclofen and CGP applications in CCK SCA cells. Statistics shown: ns P > 0.05, * P < 0.05, ** P < 0.01, results from Wilcoxon signed rank tests, all data are derived from 6 CCK SCA–CA1 PC pairs
Fig. 4
Fig. 4
Surface expression of GABAB1 differs on CB1R- and M2R-positive axon terminals. a, b Electron micrographs of CB1R-positive (10 nm gold particles) putative CCK BC axon terminals (b) forming synapses either with somata (S in A) or proximal dendritic shafts (den in B) of putative CA1 PCs. Note, that CB1R-positive terminals show consistently high levels of immunogold labeling for GABAB1 (5 nm, arrows) (synaptic vesicles are indicated by arrowheads in a). c, d Electron micrographs of M2R-positive (10 nm immunogold) putative PV BC axon terminals (b) contacting somata (S) of putative CA1 PCs. M2R-positive terminals showed a lower GABAB1 density (5 nm particles, arrows in c) and also included immunonegative terminals (b in d). Insets in ac show differences in particle size, with small 5 nm gold particles for GABAB1 (arrows) and the larger 10 nm particles for the interneuron-specific marker CB1R and M2R, respectively. e Summary bar chart of the proportion of GABAB1 on double-labeled CB1R- or M2R-containing axon terminals. The total number of examined CB1R- and M2R-immunopositive terminals is indicated in parenthesis. f Quantification of the surface density of immunogold particles for GABAB1 subunit on CB1R- and M2R-positive axon terminals (with GABAB1 immuno-negative profiles exclude) in comparison to PC dendrites. Note the divergence of labeling density between CB1R- and M2R-positive terminals. Statistics shown: ns P > 0.05, *** P < 0.0001, Fisher’s exact test and 1-way ANOVA with multiple comparisons. Scale bars 200 nm

Similar articles

See all similar articles

Cited by 2 articles

References

    1. Althof D, Baehrens D, Watanabe M, Suzuki N, Fakler B, Kulik Á. Inhibitory and excitatory axon terminals share a common nano-architecture of their Cav2.1 (P/Q-type) Ca2+ channels. Front Cell Neurosci. 2015;9:315. doi: 10.3389/fncel.2015.00315. - DOI - PMC - PubMed
    1. Armstrong C, Soltesz I. Basket cell dichotomy in microcircuit function. J Physiol. 2012;590:683–694. doi: 10.1113/jphysiol.2011.223669. - DOI - PMC - PubMed
    1. Ault B, Nadler JV. Effects of baclofen on synaptically-induced cell firing in the rat hippocampal slice. Br J Pharmacol. 1983;80:211–219. doi: 10.1111/j.1476-5381.1983.tb11068.x. - DOI - PMC - PubMed
    1. Bartos M, Vida I, Frotscher M, Geiger JR, Jonas P. Rapid signaling at inhibitory synapses in a dentate gyrus interneuron network. J Neurosci. 2001;21:2687–2698. - PMC - PubMed
    1. Bartos M, Vida I, Jonas P. Synaptic mechanisms of synchronized gamma oscillations in inhibitory interneuron networks. Nat Rev Neurosci. 2007;8:45–56. doi: 10.1038/nrn2044. - DOI - PubMed

MeSH terms

LinkOut - more resources

Feedback