Webbed Reprint Collection|
William H. Calvin
University of Washington
Seattle WA 98195-1800 USA
Email || Home Page
|Calvin, W. H., and Schwindt, P. C. (1972). |
Steps in production of motoneuron spikes during rhythmic firing.
Journal of Neurophysiology 35:297-310.
copyright ©1972 by authors and publisher
Steps in Production of Motoneuron Spikes During Rhythmic Firing
WILLIAM H. CALVIN and PETER C. SCHWINDT
Departments of Neurological Surgery and Physiology and Biophysics,
University of Washington School of Medicine,
Seattle, Washington 98195
ALTHOUGH A TRAIN of action potentials is the means by which most neurons encode messages for transmission down a long axon, little is known about the mechanisms which control this repetitive firing. While the Hodgkin-Huxley formulation for squid axon (2, 43) does not predict a very wide range of repetitive-firing behavior, more successful voltage-clamp formulations have recently been made for invertebrate somata which do exhibit repetitive firing (11-13), correlated with a prolonged potassium-conductance change lacking in squid axon.
Such investigations of isopotential membrane regions do not, however, address themselves to the question: What is the normal mechanism which controls repetitive firing in the region of the cell which originates the spikes? Neural membranes with a latent capability for repetitive firing, or those forced to isopotentiality by a space clamp, may utilize mechanisms quite different from those actually used by the typical spike-origination regions. Unlike sites of spike replication (such as the middle of an axon), the spike-origination regions (initial segment, some, dendrites) of many CNS neurons exhibit spikes with multiple components. These components represent the activity of different spatial subregions of the cell membrane, so that the overall spike-origination properties may be a composite of the activity of different subregions of the cell. In cat spinal motoneurons (14, 21) the threshold property comes from the A spike subregion (perhaps the initial segment), the afterhyperpolarization (16, 29) comes from the B spike (usually associated with the "somadendritic" area), and a delayed depolarization (27, 31, 38) is thought to be associated with the dendrites.
The primary inference which has thus far been drawn from this spatial arrangement is that synapses do not normally set up spikes at the synaptic sites in the dendrites, but rather contribute to spike origination by the electrotonic spread of synaptic currents to a low-threshold trigger zone, e.g., the initial segment. The subsequent antidromic invasion of the somadendritic area by this initial segment spike, however, should be very important in determining when the next spike can occur because of the afterhyperpolarization and delayed depolarization which this invasion causes. It is the afterpotential currents which one naturally suspects of a role in rhythmic-firing control. The initial segment spike is reported to lack an afterpotential (16, 29), suggesting that both synaptic currents and afterpotential currents spread electrotonically to the spike trigger zone from the some and dendrites. This leads to the concept of the trigger zone as a relatively passive "summing point" for currents generated elsewhere (Fig. 1). The steady barrage of somatic and dendritic synaptic currents should sum together to produce a steady depolarization (9, 17) at the trigger zone, and the afterhyperpolarization current should subtract from it. The net IR drop across the trigger zone should thus build up after a spike as the afterhyperpolarization current wears off; when large enough, the net IR drop should cross threshold and generate an A spike. The subsequent antidromic invasion of the somadendritic area would then reset the process by causing another hyperpolarizing current,....
[intervening pages not scanned yet, jumps to summary]SUMMARY
1. Spike components (A spike, B spike, delayed depolarization, afterhypcrpolarizatiorl) of rhythmically firing cat spinal motoneurons were compared to similar components seen in single-spike experiments.
2. The B spike seems to produce most of the afterhyperpolarization during rhythmic firing. Because this mcmbranc-potential trajectory between spikes determines when the next spike can occur, it would seem as if the antidromic invasion of the somadendritic region were essential for rhythmic firing in the regions of the motoneuron where spikes are normally generated.
3. The delayed depolarization is often seen following rhythmic spikes, just as in single-spike experiments. It may change with successive spikes of a rhythmic train, either becoming larger or smaller. It may occasionally cross threshold to cause a second spike (doublet firing pattern), demonstrating that the delayed depolarization affects the site of spike initiation.
4. The afterhyperpolarization trajectories exhibit summation when an extra spike is caused between two rhythmic spikes. The afterhyperpolarization is deeper when the extra spike is closer to the preceding rhythmic spike; the time course of this enhancement is not unlike that of the afterhyperpolarization itself.
5. While the spike components seen in single-spike experiments seem to be analogous to rhytllmic-spike components, their importance to the function of the neuron is more apparent in the context of rhytllmic firing. For example, afterpotential summation may provide a mechanism for firing-rate adaptation.
6. Repetitive-firing theories are conceptually simplified if the threshold property resides at one site (for example, the initial segment), while the afterhyperpolarization currents are generated elsewhere (the somadendritic region) and spread passively to this trigger zone along with the synaptic currents from the dendrites. If the trigger zone is thus not generating pacemaker curtcnts itself during tile interspike interval, there is a physical, spatial separation of the threshold and pacemaker properties of tile oscillator mechanism.