BACKGROUNDPrevious studies have suggested that primary degeneration of hair cells causes secondary degeneration of spiral ganglion neurons (SGNs), but the effect of SGN degeneration on hair cells has not been studied. In the adult mouse inner ear ouabain can selectively and permanently induce the degeneration of type 1 SGNs while leaving type 2 SGNs, efferent fibers, and sensory hair cells relatively intact. This study aimed to investigate the dynamic changes in hair cell ribbon synapse induced by loss of SGNs using ouabain application to the round window niche of adult mice.

METHODSIn the analysis, 24 CBA/CAJ mice aged 8-10 weeks, were used, of which 6 normal mice were used as the control group. After ouabain application in the round window niche 6 times in an hour, ABR threshold shifts at least 30 dB in the three experimental groups which had six mice for 1-week group, six for 1-month group, and six for 3-month group. All 24 animals underwent function test at 1 week and then immunostaining at 1 week, 1 month, and 3 months.

RESULTSThe loss of neurons was followed by degeneration of postsynaptic specializations at the afferent synapse with hair cells. One week after ouabain treatment, the nerve endings of type 1 SGNs and postsynaptic densities, as measured by Na/K ATPase and PSD-95, were affected but not entirely missing, but their partial loss had consequences for synaptic ribbons that form the presynaptic specialization at the synapse between hair cells and primary afferent neurons. Ribbon numbers in inner hair cells decreased (some of them broken and the ribbon number much decreased), and the arrangement of the synaptic ribbons had undergone a dynamic reorganization: ribbons with or without associated postsynaptic densities moved from their normal location in the basal membrane of the cell to a more apical location and the neural endings alone were also found at more apical locations without associated ribbons. After 1 month, when the neural postsynaptic densities had completed their degeneration, most ribbons were lost and the remaining ribbons had no contact with postsynaptic densities; after 3 months, the ribbon synapses were gone except for an occasional remnant of a CtBP2-positive vesicle. Hair cells were intact other than the loss of ribbons (based on immunohistochemistry and DPOAE).

CONCLUSIONThese findings define the effect of SGN loss on the precise spatiotemporal size and location of ribbons and the time course of synaptic degeneration and provide a model for studying plasticity and regeneration.