Skip to main content
. 2015 May 12;4:e06262. doi: 10.7554/eLife.06262

Figure 1. Targeted viral labeling of the GABAergic and glycinergic neurons in the CN reveals dense and wide-spread network of nucleo-cortical axons in the cerebellar cortex.

(A1–B1) Confocal images of coronal cerebellar sections in mice, where floxed virus was injected into the cerebellar nuclei of GAD-cre and GlyT2-cre mice. (A1) Flocculus, coronal view, 40× confocal scan tiles. (B1) Posterior vermis, horizontal view. (A2B2) Higher magnification confocal images of the cerebellar nuclei show transfected GABAergic and glycinergic neurons, respectively. Note the lower density of labeled neurons in GlyT2-cre brain. In the GAD-cre mice, both small, globular and larger, multipolar neurons (arrows and arrowheads in A2, respectively) were seen. In the GlyT2-cre mice, only large, multipolar neurons were observed, characteristic to the glycinergic CN neurons (arrowheads in B2). In the GAD-cre mice, the transfected neurons included the GABAergic NO neurons, as evidenced by the fluorescent axons in the contralateral IO (A3). (C) Injection of hSyn-GFP-virus into the IO retrogradely labeled small, round NO neurons in the contralateral CN. (D) Comparison of soma sizes among the three labeled populations. The histograms are fitted with single (GlyT2+ and NO) or double (GAD+) Gaussians (thick lines), showing that the GlyT2+ neurons distribution matches the second peak in GAD + fit and that the GlyT2+ and NO neurons form distinct populations that contribute to the GAD + population. (E) Confocal composite images showing virally transfected CN neurons in GAD-cre (E1) and GlyT2-cre (E2) mice, and their axons (arrows) extending across the WM surrounding the CN into the GrCL of the cerebellar cortex. (F1) Confocal composite image of a caudal coronal section of GlyT2-cre cerebellum where the lateral CN was virally transfected (location of the CN is drawn schematically on top of the image). The wide distribution of the NC axons in medio-lateral direction, including parts of the contralateral cerebellum, is shown in yellow color. (F2) Confocal composite image of a horizontal section at the level of the CN in GlyT2-cre cerebellum with transfection of the glycinergic neurons in the medial CN. The axons of the labeled neurons can be seen extending through wide areas of the vermal cortex. The inset shows a single inhibitory nucleo-cortical (iNC) axon forming axonal swellings across several hundreds of μm in the GrCL. Abbreviations: GrCL, granule cell layer; CN, cerebellar nuclei; mCN, medial cerebellar nuclei; IO, inferior olive; NO, nucleo-olivary; WM, white matter. Scale bars: A1, B1: 50 μm; A2, B2, C, 10 μm; E, 100 μm, F1, 2: 400 μm; F2, inset: 100 μm. See also Figure 1—figure supplement 1.

DOI: http://dx.doi.org/10.7554/eLife.06262.003

Figure 1.

Figure 1—figure supplement 1. iNC neurons receive functional GABAergic Purkinje Neuron inputs.

Figure 1—figure supplement 1.

(A) In the GlyT2-eGFP mouse, GFP-positive neurons (in green) are contacted by VIAAT-positive varicosities (in red). Points of contacts are indicated by arrows. (Z-projection thickness: 4.6 µm, scale bar: 10 µm). (B) Optogenetical stimulation of Purkinje neurons (PNs) axonal varicosities was performed by whole-field LED (470 nm) illumination of the cerebellar nuclei of L7-ChR2-YFP mice bred with GlyT2-eGFP mice. In these double-positive mice, PNs expressed specifically the channelrhodopsin (ChR2) (Chaumont et al., 2013) while iNC neurons can be easily targeted for patch-clamp recording using epifluorescence for GlyT2-eGFP. (C) One millisecond illumination (indicated by blue box) elicited large inhibitory responses in the iNC neurons (mean amplitude of 416.5 ± 332.1 pA, n = 10 cells), in presence of glutamate receptors blockers (APV 50 µM, NBQX 10 µM), which were blocked by 1 µM gabazine (98.1 ± 1.4 % block, n = 9 cells). The kinetics of the PN-originating inhibitory post-synaptic currents (IPSCs) in glycinergic CN cells (decay time constant 3.28 ± 0.72 ms, n = 10 cells) were similar to those of IPSCs at Purkinje cell synapses on glutamatergic projection neurons (Telgkamp and Raman, 2002; Person and Raman, 2012; Husson et al, 2014; Kawaguchi and Sakaba, 2015) but much faster than at Purkinje cell synapses on NO cells (Najac and Raman, 2015), suggesting differential control of the GABAergic NO cells and the larger, GABA-glycinergic CN cells by PNs.