Skip to main content
. 2017 Apr 3;6:e21105. doi: 10.7554/eLife.21105

Figure 6. DG-SOMIs provide weak inhibition on GABAergic and cholinergic but strong inhibition onto putative glutamatergic cells in the medial septum.

(A–C) Passive and active membrane properties and the neurochemical marker contents of the three main neuron types in the medial septum and vertical limb of the diagonal band of Broca (MSvDB). Left, characteristic discharge patterns of the three cells types (1 s, −100 and 400 pA current injection) classify them as fast-spiking (A), slow-firing (B) and cluster-firing (C) cells. Plots summarize the maximal discharge frequency of the three cell types. The discharge frequency between PVIs and slow-firing as well as cluster-firing cells was significantly different (***p<0.001 for both comparisons, t-test). Right, intracellular labeling of cells with biocytin (red) with subsequent antibody labeling (green). Note, all fast-firing cells expressed parvalbumin (PV). Slow-firing cells express choline acetyltransferase (ChAT) and cluster-firing cells have been previously identified as glutamatergic (Glut) cells (Manseau et al., 2005; Mattis et al., 2014). Pie chart summarizes the relative proportion of the recorded cell types (32 fast-spiking PVIs, 49 slow-firing ChATs, 23 cluster-firing putative Glut cells). (D) Left, schematic illustration of the experimental procedure. DG-septal projecting SOMIs expressing channelrhodopsin-2 (ChR2) after injection of rAAV-FLEX-ChR2-tdT in the dorsal DG were activated by light-pulses (5 ms, full-field illumination, 473 nm) applied to the MSvDB (red circles represent HIL cells). Middle, IPSCs recorded in the three neuron types. Individual IPSCs (grey traces) and average IPSCs (color-coded traces) are shown superimposed. Bath application of 10 µM SR59931 blocked IPSCs in three slow-firing ChAT cells. Right, summary plots show peak amplitudes of evoked IPSCs in PVIs (10 out of 32 cells), ChATs (7 out of 49 cells) and putative glutamatergic cells (5 out of 23 cells). Open circles are individual data points, filled circles are mean values with lines representing ± SEM; **p≤0.01, Mann-Whitney U test for pair-wise comparisons between PVIs vs putative glutamatergic cells and ChATs vs glutamatergic cells. ***p<0.001 for pair-wise comparisons between PVIs vs putative glutamatergic cells and PVIs vs ChATs. To compare three data sets for significant differences a Kruskal-Wallis one-way analysis of variance on Ranks was performed.

DOI: http://dx.doi.org/10.7554/eLife.21105.015

Figure 6.

Figure 6—figure supplement 1. Sequential recordings of fast-spiking PVIs and cluster-firing putative glutamatergic cells in the same slice preparation.

Figure 6—figure supplement 1.

(A) Confocal image stack of two biocytin-filled cells subsequently recorded in slice preparations of the medial septum and vertical limb of the diagonal band of Broca (MSvDB). Orange star depicts fast-spiking cell shown in (B). (B) Passive and active membrane properties of the two neuron types recorded in the MSvDB. Top, fast-spiking PVI; bottom, cluster-firing putative glutamatergic cell (1 s, −100 and 400 pA current injection). (C) Left, DG-septal projecting SOMIs expressing channelrhodopsin-2 (ChR2) were activated by light-pulses as indicated by the blue horizontal line (5 ms, 473 nm, full-field illumination) applied to the MSvDB, 2 weeks after injection of rAAV-FLEX-ChR2-tdT in the dorsal DG of SOM-Cre mice. Top, evoked IPSCs recorded in the fast-firing interneuron. Individual IPSCs (grey traces) and average IPSCs (orange trace) are shown superimposed. Bottom, light-induced IPSCs were recorded in a cluster-firing neuron. Average IPSC is shown in black. Right, graph summarizes mean peak amplitude of IPSCs subsequently recorded from a fast-spiking PVI and a cluster-firing putative glutamatergic cell in the same slice. Circles connected by lines represent individual data points. Color code represents individual mice (two mice).
HHS Vulnerability Disclosure