- Email: [email protected]
Intrinsic and synaptic properties of turtle red nucleus neurons in vitro
349
Brain Research, 608 (1993) 349-352 Elsevier Science Publishers B.V.
BRES 25614
Intrinsic and synaptic properties of turtle red nucleus neurons in vitro Joyce Keifer and James C. Houk Department of Physiology, Northwestern University Medical School, Chicago, IL 60611 (USA) (Accepted 19 January 1993)
Key words: Red nucleus; Turtle; In vitro; Slow EPSP; Excitatory amino acid
Burst discharges in the red nucleus are correlated with discrete limb movements. Intracellular recordings from red nucleus neurons in the in vitro turtle brainstem-cerebellum was performed to elucidate mechanisms underlying these bursts. Depolarizing intracellular current injection failed to demonstrate endogenous membrane currents that might produce burst discharges, and neurons did not exhibit significant spike frequency adaptation, which is a characteristic of synaptically driven bursts. Responses of red nucleus neurons to synaptic input demonstrated a late, slow depolarizing synaptic potential (slow EPSP) having a latency of 9-12 ms, and a maximal duration of 600 ms. it is concluded that neither intrinsic membrane responses, nor the duration of the slow EPSP, can fully account for the behavior of red nucleus neurons during burst discharge. We hypothesize that activity in the red nucleus is driven by a gradual recruitment of NMDA receptors, andlpr by polysynaptic excitatory pathways.
In behaving cats and monkeys, red nucleus neurons produce bursts of action potentials that are associated with discrete limb movements ~-4. The bursts correlate closely with parameters of movement, such as velocity and duration, and may represent descending motor commands. Because these bursts are not modulated when the movement is interrupted, the signals in the red nucleus are considered to be generated centrally and not by feedback from the periphery. The cellular mechanisms underlying red nucleus bursts were studied by Tsukahara and colleagues ~7'~8. Based on intracellular recordings in vivo, they proposed that prolonged depolarzations of red nucleus neurons were generated by reverberatory activity in circuits involving the cerebellum and reticular formation. In their studies, they could not rule out the possibility that the discharges were generated solely by intrinsic bursting properties of rubral neurons and not by synaptic inputs. Whether red nucleus activity is generated by synaptic interactions or by endogenous bursting properties is unknown. In a recent study, we demonstrated that red nucleus burst discharges in the in vitro turtle brainstem-cerebellum are mediated by excitatory amino acid neurotransmitters 7. The turtle was chosen for study as chelo-
nian tissue is extremely resistant to anoxia, thus allowing large portions of intact brain to be maintained in vitro 5. Red nucleus burst discharges were found to be blocked by the glutamate receptor antagonists APV or CNQX. Our pharmacological studies led to the hypothesis that bursts are triggered by afferent inputs that activate AMPA receptors. This initial depolarization then activates NMDA-mediated mechanisms that sustain the burst in the present study, intraceilular recordings were made from red nucleus neurons in the brainstem-cerebellum preparation, and we report a late, slow depolarizing synaptic potential (slow EPSP) that may contribute to red nucleus burst discharges. However, neither the duration of this slow EPSP, nor the responses to intracellular current injection, can account fully for the properties of prolonged rubral burst discharges. From these data, it is hypothesized that activity in the red nucleus is synaptically driven by a gradual recruitment of NMDA receptors, a n d / o r by polysynaptic excitatory pathways. A preliminary report has been published 6. Intracellular recordings from neurons in the red nucleus were performed using the in vitro brainstemcerebellum preparation from the turtle, Chrysemyspicta (n--7). The tissue was submerged in physiological
Correspondence: J. Keifer, Department of Physiology, Northwestern University Medical School, 303 E. Chicago Ave., Chicago, IL 60611 USA. Fax: (1) (312) 503-5101.
350 saline containing (in mM); NaC1 100, KCI 6, N a H C O 3 40, CaCI 2 2.6, MgC12, 1.6 and glucose 20, which was oxygenated with 95% 02/5% C O 2 at r o o m temperature (22-240°C) at p H 7.6. F u r t h e r details of the preparation have been described elsewhere 7. Intracellular microelectrodes were pulled from filamented glass capillary tubes and filled with 4 M KAc (DC resistance 8 0 - 1 0 0 M ~ ) . The recording electrode was m o u n t e d on a piezo-microdrive with remote stepping function (Nagel PM500) and was advanced using 2 ~ m steps. After impalement, t r a n s m e m b r a n e voltage was measured with a D a g a n 8800 in the bridge-balance mode. Neurons with a resting potential more negative than - 5 5 m V and with overshooting action potentials were accepted for data analysis. Electrical stimulation of the contralateral dorsolateral funiculus of the spinal cord using a concentric tungsten electrode (Rhodes) served to activate red nucleus neurons antidromically to unambiguously identify rubrospinal neurons. Some n e u r o n s could not be antidromically activated, and these were identified on the basis of short latency synaptic input presumably via the cerebellar nucleus, intracellular recordings were obtained from 12 neurons in the red nucleus which were identified as rubrospinal neurons. Eight cells responded antidromically ( 2 - 4 ms; ref. 8) to contralateral spinal cord stimulation, and 5 of these cells also responded synaptically. Four other cells responded at short synaptic latencies ( 5 - 7 ms; ref. 8) and were likely to have been rubrospinal neurons since they did not differ in any respects from antidromically identified neurons. Chelonian rubrospinal neurons are approximately 2 0 - 3 4 /~m in s o m a d i a m e t e r , w h e r e a s G A B A e r g i c interneurons in the red nucleus are 6 - 1 6 , m in diameter ]°. Because of their small size, we believe that the likelihood of obtaining good impalements of G A B A e r g i c interneurons is relatively small. The steady-state c u r r e n t - v o l t a g e properties of red nucleus neurons was fairly linear in the depolarizing and hyperpolarizing directions, although inward and outward rectification was evident. These results are similar to those reported for m a m m a l i a n red nucleus neurons by Kubota et al. 12 using the guinea-pig slice. Seven neurons d e m o n s t r a t e d a n o d e - b r e a k excitation following the offset of a hyperpolarizing current pulse. Fig. 1 illustrates two significant features of the current-voltage properties of red nucleus neurons. The first is that there was no afterdischarge that lasts beyond the duration of a suprathreshold depolazing current pulse, or e n d o g e n o u s non-linear response properties such as plateau potentials. The second feature is that in response to depolarizing current injection the neurons p r o d u c e d a steady discharge rate and
A
.........!........!,!,,!,,,!,!,( I I"''
'
' .
.
.
.
.
.
1
B E
5O
40 30
u.
20 lO
C
// i
0.2
i
0.4
20 1.0
I
mV nA
S
nA
~.. 10
0.8
v
8
0.6,
4
0.2
2
i
i
i
0.6
0.8
1.0
Current (hA)
200
400
Time
600 800 (ms)
1000
Fig. l. Effects of a long-duration intracellular depolarizing current step on the firing rate of a red nucleus neuron. This neuron responded to single-shock contralateral spinal cord stimulation with a synaptic latency of 7 ms. A: a current step of 0.6 nA and 3 s in duration produced a steady-state firing rate with no evidence of significant spike frequency adaptation. B: the frequency-current (f-l) plot for the same neuron shows a linear relationship. Six curves for the same neuron are plotted. C: minimal spike frequency adaptation is illustrated in response to depolazing current steps of increasing amplitude. Discharge frequency was measured at five intervals with respect to a 1 s current pulse. In C, data points for 0.2-0.6 nA are the average of 4 responses; 0.8 nA is 1 response. Membrane potential: -80 mV. did not d e m o n s t r a t e significant spike frequency adaptation. Quantitative data show that the f r e q u e n c y - c u r rent (f-l) relationship of the rubral n e u r o n in Fig. 1 was linear (Fig. 1B) and maintained a steady rate of discharge during a depolarizing current step of various amplitudes (Fig. 1C). T h e responses of red nucleus neurons to synaptic input d e m o n s t r a t e d the presence of a long-latency, slow depolarizing synaptic potential (slow EPSP) in 6 rubrospinal neurons. The cell shown in Fig. 2 was antidromically identified as a rubrospinal n e u r o n as it responded to stimulation of the contralateral spinal cord with a latency of 2 ms (Fig. 2A). It was further identified as antidromic because the action potential followed stimulus frequencies of up to 80 Hz, and the initial segment break was a p p a r e n t in the waveform. This n e u r o n was also excited at an antidromic latency by stimulation of the contralateral cerebellum (Fig. 2B), presumably by activation of rubrospinal axon collaterals that project as mossy fibers [3. A subthreshold stimulus applied to the contralateral spinal cord (Fig. 2C) evoked a slow E P S P having a latency of 9 ms. Higher intensity stimulation evoked an action potential followed by the E P S P (Fig. 2D). The duration of the E P S P for this neuron was 150-200 ms. Synaptic properties of a n o t h e r red nucleus n e u r o n are shown in Fig. 3. This n e u r o n fired an action poten-
351
A
B
m
2 ms
C
2
ms
D 20mY
20ms
20ms
Fig. 2. A slow EPSP is evoked in an antidromically identified rubrospinal neuron by single shock electrical stimulation of the contralateral spinal cord. A: this neuron responded to contralateral cord stimulation (2 mA) with an antidromic latency of 2 ms. B: stimulation of the contralateral cerebellum (4 mA) also evoked an antidromic action potential at a latency of less than 1 ms. C: a subthreshold spinal stimulus (1.8 mA) evoked a slow EPSP having a latency of 9 ms. D: higher intensity stimulation (2.2 mA) evoked an antidromic action potential followed by the EPSP which had a duration of 150-200 ms. Two traces are overlapped in A and B. Membrane potential: - 78 mV.
tial in response to contralateral spinal cord stimulation with a synaptic latency of 5 ms (Fig. 3A). Subthreshold stimulation of the contralateral cord evoked an E P S P having a latency of 12 ms (Fig. 3B). A higher stimulus intensity evoked an action potential followed by a slow E P S P (Fig. 3C). A t the slower sweep speed in Fig. 3D, the duration of this E P S P can be seen to be about 500
B
A
20ms
5ms
C
D
_
--
I
T
--
20~s
100ms
Fig. 3. Slow EPSP from a different red nucleus neuron evoked by contralateral spinal cord stimulation. A: neuron responded to cord stimulation with a synaptic latency of 5 ms. B: subthreshold stimulation (1 mA) evoked an EPSP having a latency of 12 ms. C: a higher stimulus intensity (5 mA) evoked a synaptically activated action potential followed by the slow EPSP. The baseline membrane potential is superimposed to illustrate the amplitude of the EPSP. D: shown at a slower sweep speed, the maximal duration of the EPSP was about 500 ms. Membrane potential: - 82 mV.
ms. T h e choppy waveform of the slow EPSP, as shown in Fig. 3D, suggests the possibility of fast EPSPs superimposed on the voltage envelope of the slow response. In 6 neurons, the slow E P S P had a latency ranging from 9 to 12 ms and a duration of 150-600 ms. In 3 cells in which the E P S P was observed, burst discharge evoked by contralateral cord stimulation was also recorded, however, high-quality recordings from these cells was not maintained long e n o u g h to permit examination of the bursts in detail. T h e results in Fig. 1 show that rubral responses to intracellular depolarizing current injection do not mimic the activity of rubrospinal neurons during synaptically evoked bursting. R e d nucleus neurons fail to show repetitive afterdischarges that outlast the duration of the current pulse. These data suggest that red nucleus neurons do not have regenerative m e m b r a n e currents, such as plateau potentials, that have been shown to be present in other intrinsically bursting n e u r o n s 11,14,16. A n alternative possibility is that specialized currents are located on distal dendrites and are not affected by the intracellular current pulse or are not observed from the p r e s u m e d somatic recordings. Rubrospinal n e u r o n s also do not d e m o n s t r a t e significant spike frequency adaptation during depolarizing current injection. A d a p t a t i o n is a prominent feature of the synaptically evoked burst response of red nucleus n e u r o n s r e c o r d e d extracellularly (see Fig. 3 in ref. 7). Intracellular recordings have d e m o n s t r a t e d the presence of a long-latency, slow E P S P in red nucleus n e u r o n s that is p r o d u c e d in response to synaptic input. Since N M D A and n o n - N M D A ( A M P A / K A ) receptors are known to be involved in the m e c h a n i s m of rubral burst discharge 7, it is important to further investigate w h e t h e r the slow E P S P is generated by N M D A receptors, or by a barrage of fast A M P A - m e d i a t e d events. It is also possible that both mechanisms are superimposed onto one another. T h e duration of the E P S P also does not account fully for the prolonged burst discharges. T h e maximum duration of the E P S P r e c o r d e d was 600 ms. T h e duration of the rubral burst discharges recorded extracellularly and evoked by spinal cord stimulation was on the o r d e r of several seconds, frequently 5 - 1 0 s (ref. 7). In light of these data, several hypotheses regarding the mechanisms of red nucleus burst discharge can be advanced. First, activity might be generated by non-recurrent synaptic drive. While the present data suggest that the maximal duration of the slow E P S P is several h u n d r e d milliseconds, one possibility is that increasing the stimulus amplitude would result in a gradual recruitment of the contribution of N M D A receptors n e e d e d to p r o d u c e a burst T h e second hypothesis is
352 that polysynaptic excitatory pathways contribute to the generation of red nucleus burst discharges. These pathways might arise from neuroanatomically identified recurrent connections involving the red nucleus, cerebellum and reticular formation, as was originally proposed by Tsukahara and colleagues 17'18 (also ref. 15). A role for polysynaptic circuits is also supported by our studies using the activity-dependent dye sulforhodamine to label pathways active during red nucleus bursting 9. Finally, and most likely, a combination of these cellular and circuit mechanisms might interact to produce long-duration red nucleus burst discharges. Considerable further study will be needed to unravel the different contributions of these mechanisms to rubral bursting. We thank Traverse Slater for helpful discussions and for comments on the manuscript. Supported by National Institutes of Health Fellowship NS-08661 to J.K., and Grant NS-21015 to J.C.H. 1 Cheney, P.D., Mewes, K. and Fetz, E.E., Encoding of motor parameters by corticomotoneuronal (CM) and rubromotoneuronal (RM) cells producing postspike facilitation of forelimb muscles in the behaving monkey, Behau. Brain Res., 28 (1988) 181-191. 2 Ghez, C., and Kubota, K., Activity of red nucleus neurons associated with a skilled forelimb movement in the cat. Brain Research., 131 (1977) 383-388. 3 Gibson, A.R., Houk, J.C. and Kohlerman, N.J., Magnocellular red nucleus activity during different types of limb movement in the macaque monkey, J. Physiol., 358 (1985) 527-549. 4 Gibson, A.R., Houk, J.C. and Kohlerman, N.J., Relation between red nucleus discharge and movement parameters in trained macaque monkeys, J. Physiol., 358 (1985) 551-570. 5 Hounsgaard, J. and Nicholson, C., The isolated turtle brain and the physiology of neuronal circuits. In H. Jahnsen (Ed.), Prepara-
tions of Vertebrate Central Neruous System In Vitro, Wiley, New York, 1990, pp. 155-181. 6 Keller, J. and Houk, J.C., A slow EPSP contributes to burst discharge of turtle red nucleus neurons revealed by intracellular recording in vitro. Soc. Neurosci. Abstr., 17 (1991) 1024. 7 Keifer, J. and Houk, J.C., Role of excitatory amino acids in mediating burst discharge of red nucleus neurons in the in vitro turtle brainstem-cerebellum, J. Neurophysiol., 65 (1991) 454-467. 8 Keifer, J. and Houk, J.C., An in vitro preparation for studying motor pattern generation in the cerebellorubrospinal circuit of the turtle, Neurosei. Lett., 97 (1989) 123-128. 9 Keifer, J., Vyas, D. and Houk J.C., Sulforhodamine labeling of neural circuits engaged in motor pattern generation in the in vitro turtle brainstem-cerebellum, J. Neurosci., 12 (1992) 31873199. 10 Keller, J., Vyas, D., Houk, J.C., Berrebi, A.S. and Mugnaini, E., Evidence for GABAergic interneurons in the red nucleus of the painted turtle, Synapse, 11 (1992) 197-213. 11 Kiehn, O., Plateau potentials and active integration in the 'final common pathway' for motor behaviour, Trends Neurosci., 14 (1991) 68-73. 12 Kubota, M., Nakamura, M. and Tsukahara, N., Ionic conductance associated with electrical activity of guinea-pig red nucleus neurones in vitro, J. Physiol., 362 (1985) 161-171. 13 Kunzle, H., Supraspinal cell populations projecting to the cerebellar cortex in the turtle ( Pseudemys scripta elegans ), Exp. Brain Res., 49 (1983) 1-12. 14 Llinas, R. and Sugimori, M., Electrophysiological properties of in vitro Purkinje cell somata in mammalian cerebellar slices, 3.. Physiol., 305 (1980) 171-195. 15 Sarrafizadeh, R., Keifer, J. and Houk, J.C., Anatomy of the turtle cerebellorubral circuit studied in vitro using neurobiotin and biocytin, Neurosci. Lett., 149 (1993) 59-62. 16 Silva, L.R., Amitai, Y. and Connors, B.W., Intrinsic oscillations of neocortex generated by layer 5 pyramidal neurons, Science, 251 (1991) 432-435. 17 Tsukahara, N., Bando, T. and Kiyohara, T., The properties of the reverberating circuit in the brain. In K. Yagi and S. Yoshida (Eds.), Neuroendocrine Control, Tokyo University Press, 1973, pp. 3-26. 18 Tsukahara, N., Bando, T., Murakami, F. and Oda, Y., Properties of cerebello-precerebellar reverberating circuits, Brain Res., 274 (1983) 249-259.
Brain Research, 608 (1993) 349-352 Elsevier Science Publishers B.V.
BRES 25614
Intrinsic and synaptic properties of turtle red nucleus neurons in vitro Joyce Keifer and James C. Houk Department of Physiology, Northwestern University Medical School, Chicago, IL 60611 (USA) (Accepted 19 January 1993)
Key words: Red nucleus; Turtle; In vitro; Slow EPSP; Excitatory amino acid
Burst discharges in the red nucleus are correlated with discrete limb movements. Intracellular recordings from red nucleus neurons in the in vitro turtle brainstem-cerebellum was performed to elucidate mechanisms underlying these bursts. Depolarizing intracellular current injection failed to demonstrate endogenous membrane currents that might produce burst discharges, and neurons did not exhibit significant spike frequency adaptation, which is a characteristic of synaptically driven bursts. Responses of red nucleus neurons to synaptic input demonstrated a late, slow depolarizing synaptic potential (slow EPSP) having a latency of 9-12 ms, and a maximal duration of 600 ms. it is concluded that neither intrinsic membrane responses, nor the duration of the slow EPSP, can fully account for the behavior of red nucleus neurons during burst discharge. We hypothesize that activity in the red nucleus is driven by a gradual recruitment of NMDA receptors, andlpr by polysynaptic excitatory pathways.
In behaving cats and monkeys, red nucleus neurons produce bursts of action potentials that are associated with discrete limb movements ~-4. The bursts correlate closely with parameters of movement, such as velocity and duration, and may represent descending motor commands. Because these bursts are not modulated when the movement is interrupted, the signals in the red nucleus are considered to be generated centrally and not by feedback from the periphery. The cellular mechanisms underlying red nucleus bursts were studied by Tsukahara and colleagues ~7'~8. Based on intracellular recordings in vivo, they proposed that prolonged depolarzations of red nucleus neurons were generated by reverberatory activity in circuits involving the cerebellum and reticular formation. In their studies, they could not rule out the possibility that the discharges were generated solely by intrinsic bursting properties of rubral neurons and not by synaptic inputs. Whether red nucleus activity is generated by synaptic interactions or by endogenous bursting properties is unknown. In a recent study, we demonstrated that red nucleus burst discharges in the in vitro turtle brainstem-cerebellum are mediated by excitatory amino acid neurotransmitters 7. The turtle was chosen for study as chelo-
nian tissue is extremely resistant to anoxia, thus allowing large portions of intact brain to be maintained in vitro 5. Red nucleus burst discharges were found to be blocked by the glutamate receptor antagonists APV or CNQX. Our pharmacological studies led to the hypothesis that bursts are triggered by afferent inputs that activate AMPA receptors. This initial depolarization then activates NMDA-mediated mechanisms that sustain the burst in the present study, intraceilular recordings were made from red nucleus neurons in the brainstem-cerebellum preparation, and we report a late, slow depolarizing synaptic potential (slow EPSP) that may contribute to red nucleus burst discharges. However, neither the duration of this slow EPSP, nor the responses to intracellular current injection, can account fully for the properties of prolonged rubral burst discharges. From these data, it is hypothesized that activity in the red nucleus is synaptically driven by a gradual recruitment of NMDA receptors, a n d / o r by polysynaptic excitatory pathways. A preliminary report has been published 6. Intracellular recordings from neurons in the red nucleus were performed using the in vitro brainstemcerebellum preparation from the turtle, Chrysemyspicta (n--7). The tissue was submerged in physiological
Correspondence: J. Keifer, Department of Physiology, Northwestern University Medical School, 303 E. Chicago Ave., Chicago, IL 60611 USA. Fax: (1) (312) 503-5101.
350 saline containing (in mM); NaC1 100, KCI 6, N a H C O 3 40, CaCI 2 2.6, MgC12, 1.6 and glucose 20, which was oxygenated with 95% 02/5% C O 2 at r o o m temperature (22-240°C) at p H 7.6. F u r t h e r details of the preparation have been described elsewhere 7. Intracellular microelectrodes were pulled from filamented glass capillary tubes and filled with 4 M KAc (DC resistance 8 0 - 1 0 0 M ~ ) . The recording electrode was m o u n t e d on a piezo-microdrive with remote stepping function (Nagel PM500) and was advanced using 2 ~ m steps. After impalement, t r a n s m e m b r a n e voltage was measured with a D a g a n 8800 in the bridge-balance mode. Neurons with a resting potential more negative than - 5 5 m V and with overshooting action potentials were accepted for data analysis. Electrical stimulation of the contralateral dorsolateral funiculus of the spinal cord using a concentric tungsten electrode (Rhodes) served to activate red nucleus neurons antidromically to unambiguously identify rubrospinal neurons. Some n e u r o n s could not be antidromically activated, and these were identified on the basis of short latency synaptic input presumably via the cerebellar nucleus, intracellular recordings were obtained from 12 neurons in the red nucleus which were identified as rubrospinal neurons. Eight cells responded antidromically ( 2 - 4 ms; ref. 8) to contralateral spinal cord stimulation, and 5 of these cells also responded synaptically. Four other cells responded at short synaptic latencies ( 5 - 7 ms; ref. 8) and were likely to have been rubrospinal neurons since they did not differ in any respects from antidromically identified neurons. Chelonian rubrospinal neurons are approximately 2 0 - 3 4 /~m in s o m a d i a m e t e r , w h e r e a s G A B A e r g i c interneurons in the red nucleus are 6 - 1 6 , m in diameter ]°. Because of their small size, we believe that the likelihood of obtaining good impalements of G A B A e r g i c interneurons is relatively small. The steady-state c u r r e n t - v o l t a g e properties of red nucleus neurons was fairly linear in the depolarizing and hyperpolarizing directions, although inward and outward rectification was evident. These results are similar to those reported for m a m m a l i a n red nucleus neurons by Kubota et al. 12 using the guinea-pig slice. Seven neurons d e m o n s t r a t e d a n o d e - b r e a k excitation following the offset of a hyperpolarizing current pulse. Fig. 1 illustrates two significant features of the current-voltage properties of red nucleus neurons. The first is that there was no afterdischarge that lasts beyond the duration of a suprathreshold depolazing current pulse, or e n d o g e n o u s non-linear response properties such as plateau potentials. The second feature is that in response to depolarizing current injection the neurons p r o d u c e d a steady discharge rate and
A
.........!........!,!,,!,,,!,!,( I I"''
'
' .
.
.
.
.
.
1
B E
5O
40 30
u.
20 lO
C
// i
0.2
i
0.4
20 1.0
I
mV nA
S
nA
~.. 10
0.8
v
8
0.6,
4
0.2
2
i
i
i
0.6
0.8
1.0
Current (hA)
200
400
Time
600 800 (ms)
1000
Fig. l. Effects of a long-duration intracellular depolarizing current step on the firing rate of a red nucleus neuron. This neuron responded to single-shock contralateral spinal cord stimulation with a synaptic latency of 7 ms. A: a current step of 0.6 nA and 3 s in duration produced a steady-state firing rate with no evidence of significant spike frequency adaptation. B: the frequency-current (f-l) plot for the same neuron shows a linear relationship. Six curves for the same neuron are plotted. C: minimal spike frequency adaptation is illustrated in response to depolazing current steps of increasing amplitude. Discharge frequency was measured at five intervals with respect to a 1 s current pulse. In C, data points for 0.2-0.6 nA are the average of 4 responses; 0.8 nA is 1 response. Membrane potential: -80 mV. did not d e m o n s t r a t e significant spike frequency adaptation. Quantitative data show that the f r e q u e n c y - c u r rent (f-l) relationship of the rubral n e u r o n in Fig. 1 was linear (Fig. 1B) and maintained a steady rate of discharge during a depolarizing current step of various amplitudes (Fig. 1C). T h e responses of red nucleus neurons to synaptic input d e m o n s t r a t e d the presence of a long-latency, slow depolarizing synaptic potential (slow EPSP) in 6 rubrospinal neurons. The cell shown in Fig. 2 was antidromically identified as a rubrospinal n e u r o n as it responded to stimulation of the contralateral spinal cord with a latency of 2 ms (Fig. 2A). It was further identified as antidromic because the action potential followed stimulus frequencies of up to 80 Hz, and the initial segment break was a p p a r e n t in the waveform. This n e u r o n was also excited at an antidromic latency by stimulation of the contralateral cerebellum (Fig. 2B), presumably by activation of rubrospinal axon collaterals that project as mossy fibers [3. A subthreshold stimulus applied to the contralateral spinal cord (Fig. 2C) evoked a slow E P S P having a latency of 9 ms. Higher intensity stimulation evoked an action potential followed by the E P S P (Fig. 2D). The duration of the E P S P for this neuron was 150-200 ms. Synaptic properties of a n o t h e r red nucleus n e u r o n are shown in Fig. 3. This n e u r o n fired an action poten-
351
A
B
m
2 ms
C
2
ms
D 20mY
20ms
20ms
Fig. 2. A slow EPSP is evoked in an antidromically identified rubrospinal neuron by single shock electrical stimulation of the contralateral spinal cord. A: this neuron responded to contralateral cord stimulation (2 mA) with an antidromic latency of 2 ms. B: stimulation of the contralateral cerebellum (4 mA) also evoked an antidromic action potential at a latency of less than 1 ms. C: a subthreshold spinal stimulus (1.8 mA) evoked a slow EPSP having a latency of 9 ms. D: higher intensity stimulation (2.2 mA) evoked an antidromic action potential followed by the EPSP which had a duration of 150-200 ms. Two traces are overlapped in A and B. Membrane potential: - 78 mV.
tial in response to contralateral spinal cord stimulation with a synaptic latency of 5 ms (Fig. 3A). Subthreshold stimulation of the contralateral cord evoked an E P S P having a latency of 12 ms (Fig. 3B). A higher stimulus intensity evoked an action potential followed by a slow E P S P (Fig. 3C). A t the slower sweep speed in Fig. 3D, the duration of this E P S P can be seen to be about 500
B
A
20ms
5ms
C
D
_
--
I
T
--
20~s
100ms
Fig. 3. Slow EPSP from a different red nucleus neuron evoked by contralateral spinal cord stimulation. A: neuron responded to cord stimulation with a synaptic latency of 5 ms. B: subthreshold stimulation (1 mA) evoked an EPSP having a latency of 12 ms. C: a higher stimulus intensity (5 mA) evoked a synaptically activated action potential followed by the slow EPSP. The baseline membrane potential is superimposed to illustrate the amplitude of the EPSP. D: shown at a slower sweep speed, the maximal duration of the EPSP was about 500 ms. Membrane potential: - 82 mV.
ms. T h e choppy waveform of the slow EPSP, as shown in Fig. 3D, suggests the possibility of fast EPSPs superimposed on the voltage envelope of the slow response. In 6 neurons, the slow E P S P had a latency ranging from 9 to 12 ms and a duration of 150-600 ms. In 3 cells in which the E P S P was observed, burst discharge evoked by contralateral cord stimulation was also recorded, however, high-quality recordings from these cells was not maintained long e n o u g h to permit examination of the bursts in detail. T h e results in Fig. 1 show that rubral responses to intracellular depolarizing current injection do not mimic the activity of rubrospinal neurons during synaptically evoked bursting. R e d nucleus neurons fail to show repetitive afterdischarges that outlast the duration of the current pulse. These data suggest that red nucleus neurons do not have regenerative m e m b r a n e currents, such as plateau potentials, that have been shown to be present in other intrinsically bursting n e u r o n s 11,14,16. A n alternative possibility is that specialized currents are located on distal dendrites and are not affected by the intracellular current pulse or are not observed from the p r e s u m e d somatic recordings. Rubrospinal n e u r o n s also do not d e m o n s t r a t e significant spike frequency adaptation during depolarizing current injection. A d a p t a t i o n is a prominent feature of the synaptically evoked burst response of red nucleus n e u r o n s r e c o r d e d extracellularly (see Fig. 3 in ref. 7). Intracellular recordings have d e m o n s t r a t e d the presence of a long-latency, slow E P S P in red nucleus n e u r o n s that is p r o d u c e d in response to synaptic input. Since N M D A and n o n - N M D A ( A M P A / K A ) receptors are known to be involved in the m e c h a n i s m of rubral burst discharge 7, it is important to further investigate w h e t h e r the slow E P S P is generated by N M D A receptors, or by a barrage of fast A M P A - m e d i a t e d events. It is also possible that both mechanisms are superimposed onto one another. T h e duration of the E P S P also does not account fully for the prolonged burst discharges. T h e maximum duration of the E P S P r e c o r d e d was 600 ms. T h e duration of the rubral burst discharges recorded extracellularly and evoked by spinal cord stimulation was on the o r d e r of several seconds, frequently 5 - 1 0 s (ref. 7). In light of these data, several hypotheses regarding the mechanisms of red nucleus burst discharge can be advanced. First, activity might be generated by non-recurrent synaptic drive. While the present data suggest that the maximal duration of the slow E P S P is several h u n d r e d milliseconds, one possibility is that increasing the stimulus amplitude would result in a gradual recruitment of the contribution of N M D A receptors n e e d e d to p r o d u c e a burst T h e second hypothesis is
352 that polysynaptic excitatory pathways contribute to the generation of red nucleus burst discharges. These pathways might arise from neuroanatomically identified recurrent connections involving the red nucleus, cerebellum and reticular formation, as was originally proposed by Tsukahara and colleagues 17'18 (also ref. 15). A role for polysynaptic circuits is also supported by our studies using the activity-dependent dye sulforhodamine to label pathways active during red nucleus bursting 9. Finally, and most likely, a combination of these cellular and circuit mechanisms might interact to produce long-duration red nucleus burst discharges. Considerable further study will be needed to unravel the different contributions of these mechanisms to rubral bursting. We thank Traverse Slater for helpful discussions and for comments on the manuscript. Supported by National Institutes of Health Fellowship NS-08661 to J.K., and Grant NS-21015 to J.C.H. 1 Cheney, P.D., Mewes, K. and Fetz, E.E., Encoding of motor parameters by corticomotoneuronal (CM) and rubromotoneuronal (RM) cells producing postspike facilitation of forelimb muscles in the behaving monkey, Behau. Brain Res., 28 (1988) 181-191. 2 Ghez, C., and Kubota, K., Activity of red nucleus neurons associated with a skilled forelimb movement in the cat. Brain Research., 131 (1977) 383-388. 3 Gibson, A.R., Houk, J.C. and Kohlerman, N.J., Magnocellular red nucleus activity during different types of limb movement in the macaque monkey, J. Physiol., 358 (1985) 527-549. 4 Gibson, A.R., Houk, J.C. and Kohlerman, N.J., Relation between red nucleus discharge and movement parameters in trained macaque monkeys, J. Physiol., 358 (1985) 551-570. 5 Hounsgaard, J. and Nicholson, C., The isolated turtle brain and the physiology of neuronal circuits. In H. Jahnsen (Ed.), Prepara-
tions of Vertebrate Central Neruous System In Vitro, Wiley, New York, 1990, pp. 155-181. 6 Keller, J. and Houk, J.C., A slow EPSP contributes to burst discharge of turtle red nucleus neurons revealed by intracellular recording in vitro. Soc. Neurosci. Abstr., 17 (1991) 1024. 7 Keifer, J. and Houk, J.C., Role of excitatory amino acids in mediating burst discharge of red nucleus neurons in the in vitro turtle brainstem-cerebellum, J. Neurophysiol., 65 (1991) 454-467. 8 Keifer, J. and Houk, J.C., An in vitro preparation for studying motor pattern generation in the cerebellorubrospinal circuit of the turtle, Neurosei. Lett., 97 (1989) 123-128. 9 Keifer, J., Vyas, D. and Houk J.C., Sulforhodamine labeling of neural circuits engaged in motor pattern generation in the in vitro turtle brainstem-cerebellum, J. Neurosci., 12 (1992) 31873199. 10 Keller, J., Vyas, D., Houk, J.C., Berrebi, A.S. and Mugnaini, E., Evidence for GABAergic interneurons in the red nucleus of the painted turtle, Synapse, 11 (1992) 197-213. 11 Kiehn, O., Plateau potentials and active integration in the 'final common pathway' for motor behaviour, Trends Neurosci., 14 (1991) 68-73. 12 Kubota, M., Nakamura, M. and Tsukahara, N., Ionic conductance associated with electrical activity of guinea-pig red nucleus neurones in vitro, J. Physiol., 362 (1985) 161-171. 13 Kunzle, H., Supraspinal cell populations projecting to the cerebellar cortex in the turtle ( Pseudemys scripta elegans ), Exp. Brain Res., 49 (1983) 1-12. 14 Llinas, R. and Sugimori, M., Electrophysiological properties of in vitro Purkinje cell somata in mammalian cerebellar slices, 3.. Physiol., 305 (1980) 171-195. 15 Sarrafizadeh, R., Keifer, J. and Houk, J.C., Anatomy of the turtle cerebellorubral circuit studied in vitro using neurobiotin and biocytin, Neurosci. Lett., 149 (1993) 59-62. 16 Silva, L.R., Amitai, Y. and Connors, B.W., Intrinsic oscillations of neocortex generated by layer 5 pyramidal neurons, Science, 251 (1991) 432-435. 17 Tsukahara, N., Bando, T. and Kiyohara, T., The properties of the reverberating circuit in the brain. In K. Yagi and S. Yoshida (Eds.), Neuroendocrine Control, Tokyo University Press, 1973, pp. 3-26. 18 Tsukahara, N., Bando, T., Murakami, F. and Oda, Y., Properties of cerebello-precerebellar reverberating circuits, Brain Res., 274 (1983) 249-259.