NEURAL MECHANISMS GENERATING RESPIRATORY PATTERN IN MAMMALIAN BRAIN STEM-SPINAL CORD INVITRO .1. SPATIOTEMPORAL PATTERNS OF MOTOR AND MEDULLARY NEURON ACTIVITY

1. An analysis of the spatial and temporal patterns of activity of neurons of the respiratory motor-pattern generation system in an in vitro neonatal rat brain stem-spinal cord preparation is presented. Impulse discharge patterns of spinal and cranial motoneurons as well as respiratory neurons in the medulla were analyzed. Patterns of motoneuronal discharge were characterized at the population level from recordings of motor-nerve discharge and at the single-cell level from intracellular recordings. These patterns were compared to patterns generated in the neonatal rat and adult mammal in vivo to establish the correspondence between in vitro and in vivo states. 2. The in vitro system generated a complex spatiotemporal pattern of spinal and cranial motoneuron activity during inspiratory (I) and expiratory (E) phases of the respiratory cycle. The respiratory cycle consisted of three distinct phases of neuronal activity (I, early E, and late E phase) similar to the temporal organization of the cycle in the intact mammal. The spike discharge pattern of motoneurons during the I phase consisted of a rapidly peaking-slowly decrementing discharge envelope with a high degree of synchronization on a time scale of 25-50 ms (~ 20-40 Hz). A similar pattern was generated in the neonate in vivo under conditions comparable with the in vitro state (i.e., nervous system isolated from mechanosensory afferent imputs). However, the I-phase-motoneuron discharge pattern and cycle phase durations differed from those characteristic of the intact neonatal or adult systems in vivo. This difference could be accounted for primarily by removal of vagal mechanosensory afferent inputs. 3. The synaptic drive potentials of spinal motoneurons during the I phase in vitro consisted of a rapidly peaking-slowly decrementing potential envelope similar in shape to the spike-frequency histogram of single motoneurons and the envelope of the motoneuron-population discharge. The drive potentials had prominent high-frequency amplitude fluctuations superimposed on the slower drive potential envelope that were temporally correlated with the generation of motoneuron action potentials. The dominant frequency components of these fast-membrane-potential oscillations (20-35 Hz) were similar to the frequency components of the amplitude fluctuations in the motoneuron-population discharge. One class of medullary neurons with I-phase discharge also exhibited a rapidly peaking-slowly decrementing pattern of impulse discharge and synaptic drive potential with similar high-frequency components. These data indicate that the I-phase pattern of motoneuron depolarization and impulse discharge reflects in large part the temporal characteristics of the synaptic drive generated at the premotoneuron level. 4. The pattern of I-motoneuron synaptic drive in the neonatal system in vitro differs from the ramp-potential evelope and associated augmenting discharge pattern of spinal motoneurons and bulbospinal neurons in the adult rat in vivo. The analysis of the synaptic potentials in vitro suggests that the difference in motoneuron discharge pattern in vitro and in vivo is not due to differences in the intrinsic electrophysiological behavior of the motoneurons but to a perturbation of mechanisms involved in I-drive formation that occurs in medullary networks with removal of vagal mechanosensory afferent input. 5. Populations of respiratory neurons were localized in the ventrolateral reticular formation of the medulla in regions corresponding to subregions of the ventral respiratory group of the adult rat. Five distinct classes of neurons with different impulse discharge patterns during I or E phases were identified. The discharge of the various neuronal types showed a three-phase organization like the motor-output pattern. Correlates of several of these neuronal types exist in vivo, and the gross spatial distributions of the cells are similar to distributions in the adult rat. 6. We conclude that the respiratory pattern-generation mechanisms operating in the neonatal rat brain stem-spinal cord in vitro are a kernel of the mechanisms operating in the nervous system in vivo. The body of data presented further establishes the in vitro brain stem-spinal cord as a powerful model system for analysis of pattern-generation mechanisms in mammals.