LATE EXPIRATORY INHIBITION OF STAGE-2 EXPIRATORY NEURONS IN THE CAT - A CORRELATE OF EXPIRATORY TERMINATION

1. Intracellular recordings were made from stage 2 expiratory bulbospinal neurons (E2Ns) in the caudal part of the ventral respiratory group in pentobarbitone-anesthetized cats, to characterize changes in neuronal input resistance (R(n)) and synaptic inhibition occurring at the time of the expiratory-inspiratory phase transition of the respiratory cycle. 2. R(n) was maximal between 30-90% of stage 2 expiration, but decreased significantly during the last 10% of stage 2 expiration. Mean normalized R(n) for 60-90% of stage 2 expiration was 0.9 +/- 0.02, while mean R(n) during the last 10% of stage 2 expiration was 0.68 +/- 0.09 (n = 8). This decrease in R(n) began 200-300 ms before rapid hyperpolarization of E2N membrane potential and onset of phrenic nerve activity. 3. Under conditions of strong central respiratory drive, constant injection of positive current into E2Ns sometimes revealed a transient membrane hyperpolarization that straddled the expiratory-inspiratory phase transition. During this transient event, R(n) was markedly reduced. 4. Intracellular injection of Cl- or NO3- ions into E2Ns produced reversal of chloride-dependent inhibitory synaptic potentials (IPSPs). Comparison of averages of membrane potential pattern over the whole respiratory cycle during control conditions and IPSP reversal revealed several periods of synaptic inhibition: 1)weak but progressively increasing synaptic inhibition during the second half of stage 2 expiration, 2) strong transient synaptic inhibition beginning 200-300 ms before the onset of phrenic nerve activity and ending shortly after the onset of phrenic nerve activity, and 3) strong but progressively decreasing synaptic inhibition throughout inspiration. Measurement of R(n) during IPSP reversal showed that R(n) was most decreased during the inhibition associated with expiratory-inspiratory phase transition. The time course of IPSP reversal and differences in the amount of negative current injection required to reverse the different periods of synaptic inhibition indicated that inhibitory inputs activated during stage 2 expiration may be located more proximal than those activated during inspiration. 5. Firing rate records taken from intra- or extracellular recordings from early inspiratory (Early I) or pre-inspiratory (Pre I-also called expiratory-inspiratory phase-spanning) neurons were used to generate cumulative cycle histograms of firing activity. Early I neurons (n = 16) usually fired their first action potential immediately before phrenic nerve onset. Mean time between the first Early I discharge and phrenic nerve activity was 25 +/- 80 ms; only 2 Early I neurons fired their first action potential > 100 ms before phrenic nerve. Pre I neurons (n = 6) began firing at low rates during the first half of stage 2 expiration, steadily increasing their firing rate through this phase. A transient burst of firing activity in Pre I neurons occurred around the expiratory-inspiratory phase transition, with peak firing reached at times from 200 ms before to 100 ms after the onset of phrenic nerve activity. Pre I neurons then fired at declining rates through inspiration. 6. The time course and relative strength of synaptic inhibition in E2Ns was estimated by the relative difference between whole cycles averages of membrane potential during control conditions and IPSP reversal. Synaptic inhibition increased slowly throughout stage 2 expiration, showed a further rapid increase 200-300 ms before onset of phrenic nerve activity, and then declined steadily throughout inspiration. The possible effects of firing activity of Early I and Pre I neurons was assessed by calculation of average firing activity for each neuron type throughout the respiratory cycle, using equal numbers of each neuron type. These two averages were then summed and compared with the estimated synaptic inhibition observed in E2Ns. Summed Pre I and Early I firing activity corresponded well with the time course of synaptic inhibition during stage 2 expiration, but not as well with synaptic inhibition during inspiration. 7. In conclusion, stage 2 expiratory bulbospinal neurons receive previously undescribed synaptic inhibition during the later part of stage 2 expiration, which rapidly increases in strength 200-300 ms before the onset of phrenic nerve activity. The time course and strength of this inhibition corresponds well with firing activity in pre-inspiratory neurons. This inhibition may occur at relatively proximal neuronal sites and could potentially play a significant role in controlling E2N firing rate and expiratory termination. However, E2Ns also receive a declining pattern of synaptic inhibition during inspiration that presumptively arises from inputs from early inspiratory neurons. An overlap of stage 2 expiratory and inspiratory inhibition occurs at the expiratory-inspiratory phase transition and may be necessary to fully effect the termination of expiration.