Convergence pattern of uncrossed excitatory and inhibitory semicircular canal-specific inputs onto second-order vestibular neurons of frogs - Organization of vestibular side loops

Second-order vestibular neurons of frogs receive converging monosynaptic ex citatory and disynaptic excitatory and inhibitory inputs following electric al pulse stimulation of an individual semicircular canal nerve on the ipsil ateral side. Here we revealed, in the in vitro frog brain, disynaptic inhib itory postsynaptic potentials (IPSPs) by bath application of antagonists sp ecific for glycine or gamma -aminobutyric acid-A (GABA(A)) receptors. Diffe rences in the response parameters between disynaptic IPSPs and excitatory p ostsynaptic potentials (EPSPs) suggested that disynaptic IPSPs originated f rom a more homogeneous subpopulation of thicker vestibular nerve efferent f ibers than mono- or disynaptic EPSPs. To investigate a possible size-relate d organization of these canal-specific, parallel pathways, we combined long lasting anodal currents of variable intensities with strong cathodal test p ulses, to block pulse-evoked responses reversibly in a graded manner accord ing to the size-related sensitivity of vestibular nerve afferent fibers. Th e anodal current intensity required to block a particular response componen t was about 15 times lower than the strength of the cathodal test pulse tha t activated this response component. These large threshold differences were exploited for a selective anodal suppression of the responses front thick vestibular nerve afferent fibers. Tn fact, response components known to ori ginate exclusively from thick-caliber afferent fibers such as the electrica lly transmitted monosynaptic EPSP component exhibited the lowest thresholds for cathodal test pulses and were the first to disappear in the presence o f small anodal polarization steps. Thresholds for the activation/inactivati on of responses and current intensities required for response saturation/bl ockade were used to assess the fiber spectrum that evoked the different res ponse components. Mono- and disynaptic EPSPs appeared to originate from a b road spectrum of thick and thin vestibular nerve afferent fibers. The spect rum of afferent fibers that activated disynaptic IPSPs on the other hand wa s more homogeneous and consisted of thick and intermediate fibers. Such a c anal-specific and fiber type-related organization of converging inputs of s econd-order vestibular neurons via feedforward projections was shown for th e first time by this study in frogs, but might also prevail in mammals. Sim ilar differences in these feedforward pathways have been proposed earlier i n a vestibular side-loop model. Our results are consistent with the basic a ssumptions of this model and relate to the processing and tuning of dynamic vestibular signals.