Electrophysiological and morphological characteristics of GABAergic respiratory neurons in the mouse pre-Boetzinger complex

The characteristics of GABAergic neurons involved in respiratory control have not been fully understood because identification of GABAergic neurons has so far been difficult in living tissues. In the present in vitro study, we succeeded in analysing the electrophysiological as well as morphological characteristics of GABAergic neurons in the pre-Boetzinger complex. We used 67-kDa isoform of glutamic acid decarboxylase-green fluorescence protein (GAD67-GFP) ( Delta neo) knock-in (GAD67 super(GFP-+)) mice, which enabled us to identify GABAergic neurons in living tissues. We prepared medullary transverse slices that contained the pre-Boetzinger complex from these neonatal mice. The fluorescence intensity of the pre-Boetzinger complex region was relatively high among areas of the ventral medulla. Activities of GFP-positive neurons in the pre-Boetzinger complex were recorded in a perforated whole-cell patch-clamp mode. Six of 32 GFP-positive neurons were respiratory and the remaining 26 neurons were non-respiratory; the respiratory neurons were exclusively inspiratory, receiving excitatory post-synaptic potentials during the inspiratory phase. In addition, six inspiratory and one expiratory neuron of 30 GFP-negative neurons were recorded in the pre-Boetzinger complex. GFP-positive inspiratory neurons showed high membrane resistance and mild adaptation of spike frequency in response to depolarizing current pulses. GFP-positive inspiratory neurons had bipolar, triangular or crescent-shaped somata and GFP-negative inspiratory neurons had multipolar-shaped somata. The somata of GFP-positive inspiratory neurons were smaller than those of GFP-negative inspiratory neurons. We suggest that GABAergic inhibition not by expiratory neurons but by inspiratory neurons that have particular electrophysiological and morphological properties is involved in the respiratory neuronal network of the pre-Boetzinger complex.