A Study of the Different Phases of Refinement of an Inhibitory Circuit in the Mouse Auditory Brainstem

Crudely assembled neuronal circuits with exuberant innervations are refined into precise adult-like circuits by functional silencing and structural pruning of the surplus connections during early development. Such reorganizations are evident in several excitatory circuits as well as in the inhibitory circuit between the medial nucleus of trapezoid body (MNTB) and the lateral superior olive (LSO). LSO neurons integrate excitation from the glutamatergic inputs from ipsilateral anteroventral cochlear nucleus (AVCN) and inhibition from the contralateral AVCN via GABA/glycinergic inputs from the ipsilateral MNTB (Cant and Casseday, 1986; Spangler et al., 1985). The tonotopic arrangement of the circuit ensures that the excitatory and inhibitory inputs corresponding to the same sound frequency converge on to the same population of LSO neurons. Due to this tonotopic arrangement, the medial part of the circuit responds to high frequency sound stimulus and the lateral part of the circuit responds to low frequency sound stimulus (Kandler and Friauf, 1993; Friauf, 1992; Tsuchitani, 1977). In the first two postnatal weeks of pre-hearing development, the medial part of the circuit undergoes extensive functional refinement during which single LSO neurons lose 75 % of their initial MNTB connections and the remaining inputs are strengthened by 8 fold (Kim and Kandler, 2003; Noh et al., 2010). The neurotransmitter phenotype of this circuit also transforms from a GABA/Glycine/Glutamate co-release to primarily glycinergic release during this period (Kotak et al., 1998; Gillespie et al., 2005). However, it is still unclear whether functional elimination and strengthening of inputs occurs simultaneously or in succession and which factors influence these processes. In this study I investigated the course of refinement of the medial and lateral MNTB inputs to the LSO in acute brainstem slices of E 18 to P 13 mice using whole-cell patch clamp technique. I determined the strength of single MNTB inputs using minimal stimulation technique. I estimated the number of inputs on to each LSO cell by calculating the convergence ratio derived from the postsynaptic response amplitudes to minimal and maximal stimulation of MNTB fibers. I also investigated the role of transient glutamatergic and GABAergic co-release from the MNTB inputs in the process of refinement and strengthening of the inputs. I observed three distinct phases in the pre-hearing refinement of the MNTB-LSO circuit. The first phase or the proliferation phase (between embryonic day 18 and postnatal day (P) 2-3) is followed by a functional elimination phase (between P 3 and P5). This is then followed by a strengthening phase (betweenP6 and P9) of the retained MNTB inputs. I observed extensive proliferation and functional elimination of inputs in the medial part of the circuit while no significant elimination occurred in the lateral part. Both medial and lateral inputs strengthened about 3-4 fold in the first two weeks of development. I also observed that the transient glutamate co-release, which is essential for the refinement and strengthening of the medial MNTB-2 LSO projections, does not seem to play an important role in the strengthening of the lateral MNTB inputs. I further investigated the role of GABA co-release in the process of refinement of the MNTB-LSO circuit using a conditional Gad1 knockout mouse. However, the study of the role of GABA co-release in the MNTB-LSO circuit was inconclusive due to the lack of phenotypic alteration of the GABAergic input in the MNTB-LSO circuit of the Gad1 knockout mice. Thus from the present study it is clear that, during the pre-hearing development, functional elimination of specific inputs occurs before strengthening of the inputs in the medial MNTB-LSO circuit. So the strength of single inputs does not seem to be a key factor in determining which inputs are functionally eliminated. Significant strengthening occurs in the lateral MNTB-LSO circuit even in the absence of any functional elimination. Further, while glutamate co-release in the circuit seems to be crucial for the elimination and strengthening of medial MNTB-LSO projections, it does not seem to be important for the lateral part of the circuit. These observations collectively suggest that the functional elimination phase and the strengthening phase in the pre-hearing refinement of the MNTB-LSO circuit seem to be independent of each other with different underlying mechanisms.