INPUTS FROM REGULARLY AND IRREGULARLY DISCHARGING VESTIBULAR NERVE AFFERENTS TO SECONDARY NEURONS IN SQUIRREL-MONKEY VESTIBULAR NUCLEI .3. CORRELATION WITH VESTIBULOSPINAL AND VESTIBULOOCULAR OUTPUT PATHWAYS

1. A previous study measured the relative contributions made by regularly and irregularly discharging afferents to the monosynaptic vestibular nerve (V(i)) input of individual secondary neurons located in and around the superior vestibular nucleus of barbiturate-anesthetized squirrel monkeys. Here, the analysis is extended to more caudal regions of the vestibular nuclei, which are a major source of both vestibuloocular and vestibulospinal pathways. As in the previous study, antidromic stimulation techniques are used to classify secondary neurons as oculomotor or spinal projecting. In addition, spinal-projecting neurons are distinguished by their descending pathways, their termination levels in the spinal cord, and their collateral projections to the IIIrd nucleus. 2. Monosynaptic excitatory postsynaptic potentials (EPSPs) were recorded intracellularly from secondary neurons as shocks of increasing strength were applied to V(i). Shocks were normalized in terms of the threshold (T) required to evoke field potentials in the vestibular nuclei. As shown previously, the relative contribution of irregular afferents to the total monosynaptic V(i) input of each secondary neuron can be expressed as a %I index, the ratio (x 100) of the relative sizes of the EPSPs evoked by shocks of 4 x T and 16 x T. 3. Antidromic stimulation was used to type secondary neurons as 1) medial vestibulospinal tract (MVST) cells projecting to spinal segments C1 or C6; 2) lateral vestibulospinal tract (LVST) cells projecting to C1, C6, or L1; 3) vestibulooculo-collic (VOC) cells projecting both to the IIIrd nucleus and by way of the MVST to C1 or C6; and 4) vestibuloocular (VOR) neurons projecting to the IIIrd nucleus but not to the spinal cord. Most of the neurons were located in the lateral vestibular nucleus (LV), including its dorsal (dLV) and ventral (vLV) divisions, and adjacent parts of the medial MV) and descending nuclei (DV). Cells receiving quite different proportions of their direct inputs from regular and irregular afferents were intermingled in all regions explored. 4. LVST neurons are restricted to LV and DV and show a somatotopic organization. Those destined for the cervical and thoracic cord come from vLV, from a transition zone between vLV and DV, and to a lesser extent from dLV. Lumbar-projecting neurons are located more dorsally in dLV and more caudally in DV. MVST neurons reside in MV and in the vLV-DV transition zone. VOR and VOC neurons are concentrated in ventrolateral MV and the adjacent part of vLV, including its border with DV. 5. V(i) inputs varied among the different classes of relay neurons. MVST and LVST cells projecting to C1 and LVST cells projecting to C6 received mostly irregular inputs. Regular inputs predominated in VOC cells and in C6-projecting MVST cells. Lumbar-projecting LVST cells and VOR cells received a heterogeneous input from both afferent classes. 6. Axonal conduction velocities were measured for vestibulospinal neurons projecting to C6 or L1. VOC cells were more slowly conducting than were MVST or LVST cells. Relatively few spinal-projecting neurons conducted >75 m/s. Within any projection class, there was no correlation between the %I index of a neuron and its conduction velocity. 7. Eight secondary neurons were intrasomatically labeled with horseradish peroxidase or biocytin. The axonal trajectory of each cell was consistent with the antidromic classification of its output tract. 8. The results are discussed in terms of the operation of vestibular reflexes. In the case of the vestibuloocular reflex, the present study shows that there is considerable irregular afferent input to secondary VOR neurons, yet other results demonstrate that the net afferent input to the overall reflex is entirely regular. A discrepancy apparently also occurs in the vestibulocollic reflex. Here, there is evidence that the net afferent input to the reflex is irregular. At the same time, our results imply that secondary neurons projecting to neck motoneurons are likely to carry inputs from both regular and irregular afferents. One way to reconcile the results is to assume that polysynaptic pathways modify the vestibular signals carried by secondary neurons.