Projections from upper cervical inspiratory neurons to thoracic and lumbar expiratory motor nuclei in the cat.

Previous studies have demonstrated the existence of axonal projections from propriospinal respiratory neurons in the rostral cervical cord of the cat (upper cervical inspiratory neurons) to the vicinity of the phrenic and rostral thoracic inspiratory (external) intercostal motoneurons. However, no synaptic targets of the upper cervical inspiratory neurons have been identified. This study investigated the axonal projections to the caudal thoracic and upper lumbar cord and the possible existence of inhibitory connections to the expiratory intercostal and abdominal motoneurons. The connections from upper cervical inspiratory neurons to expiratory motoneurons in the lower thoracic cord were examined using the methods of antidromic mapping and the spike-triggered averaging of intracellular potentials. Of the 70 upper cervical inspiratory neurons examined, only four (5.7%) could not be antidromically activated from the T9 segment of the spinal cord. The axons of 66 upper cervical inspiratory antidromic activation at less than 5 microV, and the presence of collaterals was confirmed by antidromic mapping in 30 cases (49.2%). Of 21 axons tested for lumbar projections, 13 (61.9%) projected as far as T12, seven (33.3%) as far as L1, three (14.3%) as far as L2, and one (4.8%) was antidromically activated from L3. Spike-triggered averaging of the synaptic potentials recorded intracellularly from expiratory intercostal motoneurons in T9 and T10 spinal segments was done for 27 upper cervical inspiratory neurons, 17 of these with 4 or more motoneurons, for a total of 111 expiratory motoneurons. In 16 cases the motoneurons were injected with Cl- to reverse IPSPs and the spike-triggered averaging was repeated. No monosynaptic or disynaptic post-synaptic potentials were seen in any of the spike-triggered averages. We concluded that the upper cervical inspiratory neurons may provide inspiratory inhibition to expiratory motoneurons via a di- or oligosynaptic pathway involving segmental inhibitory interneurons and that either the spike-triggered averaging technique was not sensitive enough to detect the ipsilateral connections or the interneuron pathway was to the contralateral side of the spinal cord.