PTN Webbed Reprint Collection
William H. Calvin
University of Washington
Box 351800
Seattle WA 98195-1800 USA
Email || Home Page
Journal of Neurophysiology
39:420-434 (1976).

copyright ©1976 by William H. Calvin, George W.Sypert, and the American Physiological Society
Scanned November 1996 WHC

scanned, OCR, webbed -- but NOT proofread -- 11 Jan 97
RealAudio file allows you
to listen to a sample of the audio monitor,
as the transition to double-spike firing is made.

Fast and Slow Pyramidal Tract Neurons:
An Intracellular Analysis of Their Contrasting
Repetitive Firing Properties in the Cat

Department of Neurological Surgery, University of Washington School of Medicine,
Seattle, Washington 98195


1. Intracellular recordings were made from an estimated 500 neurons in the sensorimotor cortex of barbiturate-anesthetized cats. Of those which were antidromically identified from the medullary pyramids, 70 were selected which also exhibited steady repetitive firing to steps of current injected through the recording electrode; 81% were "fast" (conduction velocity greater than 20 m/s) and 19% were "slow."

2. As shown by earlier workers, the spike duration is a function of conduction velocity; a spike duration of 1.0 ms is the dividing line between fast and slow.

3. Of the 57 fast pyramidal tract neurons (PTNs), 14 exhibited double spikes during otherwise rhythmic firing patterns to a step of injected current. These very short interspike intervals (usually 1.5-2.5 ms) were first seen interspersed in a rhythmic discharge (e.g., 50-ms intervals) but, with further increases in current strength, would come to dominate the firing pattern; e.g., double spikes every 40 ms. Further increases in current would typically shorten only the long intervals; e.g., 40-30 ms, but some fast PTNs developed triple spikes, etc.

4. The extra spike appears to arise from a large hump which follows most spikes in fast PTNs; while this humplike ''depolarizing afterpotential" can also be seen in slow PTNs, it is small. Extra spikes were seen only in fast PTNs with large postspike humps; in perhaps half of the fast PTNs, extra spikes probably contributed to "adaptation."

5. Slow PTNs often had frequency-current curves which were not repeatable; a "hysteresis" phenomenon could often be seen, where the proportionality constant relating current to firing rate decreased following high firing rates.

6. The B spike was distinguishable from the A spike in differentiated antidromic spikes in 77% of the slow PTNs, in only 14% of the fast PTNs which later exhibited double spikes during current-induced repetitive firing, and in 53% of the other fast PTNs.

7. The antidromic spike heights of doublet PTNs were not significantly different from those of other repetitively firing PTNs.


From both extracellular and intracellular recordings, it has been apparent during the past decade that pyramidal tract neurons (PTNs) with fast conduction velocities (>20 m/s) are quite different from the slow PTNs. We have summarized some of those differences relating to the cell's geometry and electrophysiological properties in Table 1. There are, in addition, other interesting features; e.g., the slow PTNs tend to provide recurrent excitation to the fast PTNs (42), and there are marked differences in spontaneous activity (20).

In comparing these differences to the well studied size spectrum of spinal motoneurons, it is apparent that PTNs (e.g., Koike et al., ref 26) have an even more exaggerated differentiation of cell properties than do motoneurons (e.g., Kuno et al., ref 29). Especially when one views motoneurons and PTNs from the framework of repetitive-firing properties, many features can be noted which go well beyond the usual categorization of fast PTNs as being "harder to fire" but also more "vigorous and variable" when they do fire.

Repetitive-firing properties of CNS neurons have been recently reviewed by Calvin (10). Essentially, a cell generates a spike train in response to various input waveforms using three different modes of repetitive firing: 1) occasional excursions of the membrane potential through threshold give rise to an occasional spike mode, 2 ) sustained depolarizing waveforms which attempt to hold the membrane potential above threshold give rise to a rhythmic firing mode with firing rate proportional to depolarizing current, and there is 3 ) an extra spike mode where depolarizing afterpotentials (large postspike humps) appear to rise through the falling threshold milliseconds after a spike to give rise to an "extra" spike. This extra-spike discharge is most easily detected when it arises during otherwise rhythmic discharge to a constant depolarizing current, i.e., double spikes in the midst of a rhythmic train of single spikes. In motoneurons, the double spikes were first seen with string galvanometer methods (23); their correlation with the postspike humps has only been more recently established (9, 13). In a preliminary report on our present data (15), we showed that the same phenomenon often occurs in fast PTNs and discussed the striking firing patterns produced, which are similar to those seen in "epileptic" neurons.


Experimental procedures

We anesthetized 23 adult cats with pentobarbital (initial dose, 35-50 mg/kg intraperitoneally, with supplements intravenously to maintain deep anesthesia) and immobilized them with gallamine triethiodide. Arterial pressure was monitored and rectal temperature was automatically maintained by hot pad and heat lamp. The artificial respiration was adjusted to moderately hyperventilate the cats (end-expired CO2 about 3.5%). A bilateral craniectomy exposed both pericruciate areas; we usually made a small slit in the aura of the right side to aid in CSF drainage and reflected the aura on the left side for recording.

We opened the dura at the cisterna magna, both for CSF drainage and to place a concentric bipolar stimulating electrode in the medullary pyramids via a dorsal approach (an approach at about 45, about I mm left of midline, will intersect the pyramids before their decussation even with the entry point several millimeters caudal to the obex). We temporarily recorded antidromic surface potentials from the pericruciate gyri with silver ball electrodes while adjusting the position of this stimulating electrode. While this dorsal placement is not as good at eliminating orthodromic driving of PTNs as the more exacting ventral approach to the pyramids, one has little difficulty in distinguishing antidromic from orthodromic activation with intracellular recording if the shock is adjusted to straddle the threshold.

To further reduce cerebral pulsations, we performed a bilateral thoracotomy and held the rib cage in an extended position. We used a thin Plexiglas "pressor foot" of 10 mm diameter with a central 1-mm hole. This was positioned over the pericruciate area, angled so as to press harder posteriorly and to barely touch at the hole. We observed the surface blood vessels through a dissecting microscope while adjusting the pressure, we avoided signs of pial blanching within several millimeters of the recording site. This pressor foot also contained three Ag-AgCI electrodes embedded flush with the ventral surface of the Plexiglas; we used one to monitor the surface electrocorticogram (ECoG) near the recording site. The other two were occasionally used for attempts (largely unsuccessful) to modify cell responses by bipolar surface stimulation.

Microelectrode methods

We used single-barrel micropipettes which we beveled (2) and then back-filled with 2.7 M KCI using drawn-out PE tubing. These micropipettes were capable of transmitting large currents without the usual troublesome resistance fluctuations (1). Our tip diameters were probably about I micron and our electrode resistances were usually on the order of 4-8 megohms when measured with a 1-kHz square wave. We injected various combinations of pulse-, step-, and ramp-current waveforms through the recording microelectrode; we approximately balanced the bridge of the neutralized capacitance amplifier (Bioelectric Instruments, Pl) using spike-height and firing-level methods rather than by compensating for the make-and-break discontinuities. We were careful not to overcompensate for capacity. We recorded the injected-current waveform, the microelectrode and ECoG recordings, and various synchronization pulses on an FM tape recorder having a band pass of direct current to 5 kHz. This 5-kHz band pass was capable of resolving the usual components of the differentiated spikes (293 in the playback, even for very narrow spikes from fast PTNs. Other aspects of the experimental methods were analogous to those we have previously described (11, 15, 39).

Cell selection

We estimate that we encountered over 500 neurons in the 15 good experiments, i.e., cells where the intracellular penetration would have been at least good enough to allow for the brief study of synaptic potentials (if not spikes). The reduction of this sample to the present 70 PTNs (14%) is discussed later. We first tested for repetitive firing by injecting a step of current through the recording electrode; if successful, we turned on our tape recorder and tried to obtain antidromic identification. We then used various combinations of injected current to explore the repetitive firing behavior. A typical sequence was 400-ms pulses repeated every 1,200 ms, with the pulse size incremented gradually until very high firing rates were obtained and then gradually decremented to zero.

We typically recorded from area praecentralis gigantopyramidalis (4 gamma), judging from the maps of Hassler and Muhs-Clement (22). Our microelectrode usually recorded first from surface posterior sigmoid gyrus and then from buried cortex along the cruciate sulcus. We did not typically test synaptic inputs.

Data-analysis methods

Some experiments were done with the aid of an on-line computer (LINC-8, Digital Equipment Corp.) which generated current steps and plotted the current against the resulting firing rate so that we could immediately judge the responses. A variation of the same program could be used during replay of FM tapes for data analysis off-line. The program also generated plots of instantaneous firing rate (reciprocal of interspike interval) versus time. It plotted frequency-current (f-l) curves using a different alphanumeric symbol (0-9, A-Z) for each successive point to allow reconstruction of the sequence in which points were obtained; this feature was important when attempting to judge shifts in f-l curves, such as in the case of the hysteresis seen in slow PTNs (Fig. 4).

The local variations of the raster method of Wall (44) and the joint interval density of Rodieck et al. (33) have been previously described (4, 7, 8). Spikes were detected using a rate-of-change method within the computer program rather than by using the oscilloscope sweep circuit previously described (8).


We have reduced our population of PTNs, for the purposes of this paper, to a total of 70 antidromically identified cells, of which 57 (81%) were fast (antidromic conduction time less than 2.3 ms, following Takahashi, ref 41) and 13 (19%) were slow. The latency spectrum shown in Fig. 1C demonstrates the distribution of latencies across the range 0.78-4.70 ms. These cells were antidromically identified and fired repetitively to steps of injected current; on most, we were able to obtain enough current steps to plot all or part of an f-l curve. Some cortical neurons on which we could not obtain antidromic identification have also been used in selected figures, as noted.

Spike heights

The scatter plot of Fig. 1A shows that the spike heights from the slow PTNs were, on the average, larger than those of the fast PTNs, in agreement with spinal motoneuron data (29). The 13 slow PTNs antidromic spikes ranged from 58 to 104 mV (mean and SD, 83 + 13 mV), while the 57 fast PTNs ranged from 48 to 100 mV (77 + 13 mV).

For the purposes of Fig. 1 we have includedcells down to 48-mV spike heights, provided that they exhibited repetitive firing to injected currents. This serves to demonstrate that fast PTNs exhibiting doublets (15), as shown in Fig. 6, were not preferentially distributed the doublet fast PTNs are shown as triangles in Fig. 1A and B, and as the shaded areas of the latency histogram in C. It is apparent that doublets occur throughout the entire range of fast latencies; also, they are not concentrated in the low spike heights, as might be expected if injury played a prominent role in their production.

The spike heights shown in Fig. 1A were typically obtained within the first 60 s followingpenetration. They often improved with time; for example, the 89-mV antidromic spike and 122-mV repetitive spikes in Fig. 2A are from a cell contributing a 7~mV point to Fig. 1A because the heights increased. Of course spike heights more often deteriorated, but this often meant that the cell was discarded from our sample for failure to obtain repetitive firing (see DISCUSSION). Thus, the spike heights in Fig. 1A probably represent an underestimate of our spike heights.

Spike width

Fast and slow PTNs can usually be instantly identified from their sound in the loudspeaker;the fast PTN spike can be 2-3 times narrower than those of slow PTNs, as shown in Figs. 1B and 2. The fast PTN spikes range from 0.38 to 1.00 ms in duration, while the slow PTN spikes are uniformly longer than 1 ms. The spikes were measured near their base; while it would seem that the end point of the width measurement might be somewhat arbitrary, the scatter about the least-squares line in Fig. 1B is proportionately no worse than that of a similar plot made with the spike width measured half way up the spike (the width at half-height is about 45% of the base width). Extrapolating our least-squares fit line (Fig. 1B; W = 0.272 + 0.295L) to the 12-ms latencies expected for the

TO BE COMPLETED.... || Home Page || Calvin publication list || My Science Surf column || The Calvin Bookshelf

scanned, OCR, webbed -- but NOT proofread -- 11 Jan 97