Respiratory drive to thoracic motoneurones

Abstract

The paper by Saywell et al. (2007) in this issue of The Journal of Physiology provides a detailed study of the projection of the expiratory bulbospinal neurones to identified thoracic motoneurones in the anaesthetized cat. The importance of the paper lies first in its clarification of the debate regarding the strength of the monosynaptic excitation from this pathway and secondly, the deduction of the significance of the pathway in driving natural rhythmic motoneurone firing. The methods developed and the conclusions derived are relevant not only to the special case of respiration, but also to the more general consideration of the part played by particular sources of excitation in generating motoneurone firing patterns in other systems. It has long been known that both the inspiratory and the expiratory phases of breathing are dependent on activity in groups of cells in the hindbrain (Pitts et al. 1939). The most extensively studied group are in the nucleus retroambigualis and their output is carried to various motoneurones involved in respiration via crossed bulbospinal fibres. Previous studies have indicated that both phrenic and inspiratory intercostal motoneurones receive monosynaptic excitation from this source, but probably depend to a large degree on multisynaptic connections for their natural firing pattern (Davies et al. 1985). The relative contributions of mono- and multisynaptic connections to expiratory motoneurones, on the other hand, have not been clear. Using intracellular spike-triggered averaging (STA) from expiratory bulbospinal neurones (EBSNs) to internal intercostal motoneurones Merrill & Lipski (1987) found only very few and weak monosynaptic excitatory postsynaptic potentials (EPSPs), which could account for neither the rhythmic motoneurone firing nor the observed central respiratory drive potentials (CRDPs). Cross-correlation of EBSN spikes with expiratory nerve filaments also rarely showed the short duration peak expected for a monosynaptic connection. It was surprising therefore that when a spike correlation study was repeated by Kirkwood (1995) most EBSN neurones showed clear evidence of monosynaptic projection, which was estimated to be producing most of the depolarization required to drive the rhythmic firing of the expiratory motoneurones. A valuable review of this background was provided by Kirkwood et al. (1999) together with preliminary data now reported in detail. This new study has confirmed, by intracellular STA, the presence of a potent monosynaptic connection, especially to expiratory neurones with their axons in the internal intercostal nerves. Computations, taking account of directly measured EPSP duration and amplitude and reasonable estimates of numbers of EBSNs and of their firing frequencies, indicate that the sum of the unitary EPSPs could account for the observed CRDPs. Indeed the data imply a projection at least 10 times as strong as that found by Merrill & Lipski (1987). Of course the question arises as to how there can be such a disparity in data from basically similar experiments. Factors such as bias in selection of motoneurone types, rostral or caudal recording sites in a segment or variation according to the part of the thorax supplied were all considered but rejected. A final possibility arises from the observation of a correlation between EPSP size and CRDP amplitude, which suggests that some state-dependent modulation in EPSP amplitude may vary between different preparations, for example as a result of differing anaesthesia. Although the same agent (pentobarbitone) was employed in both cases, Saywell et al. (2007) appear to have controlled the dosage to give a lighter level. Nevertheless, this seems unlikely to be a very important factor and other causes should be considered. It is worth recalling the observations of Nelson et al. (1979) showing acute increases in Ia EPSP amplitudes following spinal transection. The mechanism of such state-dependent modulation is not considered, but two factors might be examined in future work. There could well be a presynaptic inhibitory mechanism involved (see Rudomin & Schmidt, 1999) which could vary due to small differences in the preparation. Also, the incidence of failures of transmission (Grimwood et al. 1992) could differ according to the condition of the animal. It is interesting to speculate as to the functional significance of a potentially strong monosynaptic EBSN projection to expiratory motoneurones. We have come to associate monosynaptic descending control with the need for precision, as in hand movements in primates. Perhaps similar precision may be required in some expiratory movements such as vocalization (Sears, 1977), though cats are not noted for their musical skills. The more serious message is the importance of detailed quantitative studies, such as reported by Saywell et al. (2007) in establishing the relative significance for normal function of various neural projections, which are generally only established qualitatively.