[1]
42w
A NUMBER 15 0 O0 O0 lO o 9 O0 000 0000 00000 5 eo000 eeoooo ee0000 leo0 9149 0 @@OOOOO 1.0 2.0 B NUMBER 15 10 5 0 9 EPSP 0 FIRING O o 0 3,0 4'.0 rn s 9 EPSP
[2]
62w
5 2.0 2.5 3.0 m s LATENCY Fig. 3. A Latency histogram of CCN neurones in response to stimulation of dorsal root. The histogram was constructed for responses to segmental input. Latencies were measured from stimulus. Open and filled symbols indicate latencies of spike discharges and EPSPs, respectively. B Latency histogram of CCN neurones in response to stimulation of neck muscle afferents.
[3]
340w
When single shock did not evoke spikes, double shocks were used and the latency was measured from the effective stimulus. For latency measurement, stimulus intensity of 2.5T was used so that the minimal latency could be obtained 1. Segmental Input. Stimulation of the ipsilateral dorsal root elicited discharges in almost all CCN neurones in the same segment. This was evident from the fact that, when spikes of many CCN neurones were simultaneously recorded with a fairly coarse electrode, their antidromic spikes following supramaximal stimulation of the cerebellar peduncle could completely collide with preceding orthodromically induced spikes in response to dorsal root stimulation. However, individual neurones showed a great diversity of responses to dorsal root stimulation with respect to their threshold, latency, and firing probability. Figure 1A shows a representative example of responses of a single CCN neurone to stimulation of the cerebellar peduncle (a) and the C2 dorsal root ganglion (b-c). The threshold for evoking the response to dorsal root stimulation was 1.2 times the dorsal root threshold; it ranged from 1.1 to 1.5 times in other neurones. When stimulus intensity was relatively weak, the latency and the firing probability varied from trial to trial (Fig. lAb). As stimulus intensity increased, spikes were evoked in all trials and the latencies became shorter (Fig. 1Ac). The shortest latency corresponded roughly to the latency of the peak of negative field potential induced in the CCN (cf. Fig. 1). The shortest latency measured from stimulation of the dorsal root in Fig. 1A was 1.6 ms, and it ranged from 0.8 to 2.0 ms for 43 neurones examined, except for 3 neurones with latencies of more than 2.8 ms (Fig. 3A). The earliest volley arrived at the spinal cord as early as 0.35-0.4 ms after stimulation. Assuming that 0.2 ms is allowed for intraspinal conduction time, 0.3 ms for synaptic delay, and 1.0 ms or less for the time required for the EPSP to reach the firing level (see below), the latencies less than 2.0 ms after stimulation would be in a monosynaptic range.
[4]
82w
In addition to monosynaptic activation, late excitatory effects were also observed in about one third of CCN neurones when intensity of stimulation of the dorsal root ganglion was sufficiently high. The late as well as early excitatory input induced double or even triple discharges in these neurones (Fig. 1B). The existence of late responses would indicate polysynaptic connexions in the pathway from primary afferents to CCN neurones. In the remaining two thirds of cells, monosynaptic excitation was predominant without late excitatory effects.
[5]
134w
In 9 CCN neurones intracellular recording could be made successfully. In Fig. 2A spikes of a CCN neurone were first recorded extracellularly to observe its responses to dorsal root stimulation at various intensities (upper trace in each pair of records), and then the same cell was penetrated to record postsynaptic potentials (PSPs) in response to stimulation at the same intensities (lower traces). With weak stimulation inducing small EPSPs, spikes were generated at a time nearly corresponding to the top of the EPSP, so that the spike latencies were relatively long (2-3 ms in this case). As the stimulus intensity gradually increased, the latency of the EPSP was not changed (1.0 ms after stimulation), but its amplitude and the slope of the rising phase increased, rgsulting in shortening of the latencies of the evoked spikes.
[6]
18w
The latencies of EPSPs measured from dorsal root stimulation ranged from 0.8 to 1.0 ms in 9 CCN
[7]
219w
A Cb ~~ E I OBL' CARCAUD" 1 " 5 T lo q~ ,,I.II ...... ,, .,, . =lh i, i ~-: --,-_ 9 0 ~f'" I B c2' SOL 1.2T F lo I 2.0 ..i,,l,, ' .................. o "~ tl o D 2.0 o h t" 0 5 lmV 2 ms 2.5 ,L .,hl , ,I , 1 N, I-10 15 20 ms H RCD 1.1 T I 1.4 iiiiitlll 2.0 ~v K ..... 2.5 i ~. -[lmV 2ms Fig. 4A-K. Responses of CCN neurones to muscle nerve stimulation. A Identification of a C2 CCN neurone by antidromic stimulation (10 g,A) of the cerebellum, liD Its responses to stimulation of C2 SPL nerve. Note a clear shortening of latency as nerve stimulation increased in strength. E-G PST histograms of response of extracellular spikes in a C1 CCN neurone to stimulation of OCC nerve. Arrow in abscissa indicates the time of stimulation, and ordinate the summed number of spikes during 50 trials. Both early and late responses increased with stimulus strength. H-K IntraceUular recording from a C1 CCN neurone. Monosynaptic EPSP was evoked by stimulation of RCD nerve. The amplitude of initial component of EPSP became maximum at stimulus strength around 1.5T, at which strength the second component was induced. Middle traces in J and K are extracellular potentials
[8]
198w
neurones (Fig. 3A), indicating that these EPSPs were generated monosynaptically. The threshold of dorsal root stimulation for evoking the EPSP was as low as 1.1 times the threshold for the incoming volley. The amplitude of EPSPs increased with an increase in stimulus intensity without saturation until the intensity reached 2.5-5 times the threshold for the dorsal root (Fig. 2Be and f). It was noted that a second component superimposed on the early monosynaptic EPSP was induced above 2.0 times the threshold and that this late EPSP contributed to the increase in amplitude of the whole EPSP. In contrast, in the neurone shown in Fig. 2A the EPSP was followed by a hyperpolarization when stimulus intensity exceeded 2.0-2.5 times the threshold. The amplitude and the slope of the hyperpolarization was increased in amplitude by passing depolarizing currents through the recording microelectrode, indicating that it was an IPSP. Absence of late spikes after strong stimulation of the dorsal root (Fig. 1Ac) may be attributed to the IPSP cutting off late spike generation. On the other hand, double or triple discharges found in the majority of CCN neurones may be caused by a long lasting depolarization as shown in Fig. 2B.
[9]
33w
the dorsal root to CCN neurones was not always limited to their own segment, but was extended over several segments for some CCN neurones. In this regard experiments were performed in three animals.
[10]
145w
Since it was technically difficult to dissect all dorsal root ganglia of the cervical segments for stimulation, C2, C3, C4 and C8 were prepared in one experiment, C2, C5 and C7 in another, and C2, C3, C5 and C8 in the rest. The results shown in Table 1 were obtained with 20 CCN neurones whose spike activity was recorded extracellularly. Extrasegmental inputs were most powerful from immediately adjacent segments, especially between C2 and C3. In some cases, however, CCN neurones received inputs from distant segments (Fig. 1C). The latency of spikes evoked from the dorsal roots in distant segments such as C7 or C8 ranged from 1.0 to 3.0 ms after the incoming volley recorded in the same segment as the location of investigated CCN neurones. This suggests that there is a monosynaptic as well as polysynaptic excitatory connexion with CCN neurones from distant segments.
[11]
67w
In order to reveal weak subliminal effects from distant segments, intracellular recording was made in 4 CCN neurones. In one neurone located in C2, monosynaptic EPSPs were induced from the C7 ganglion (Fig. 2Db), while in the remaining 3 cells no PSPs were observed in response to stimulation of the ganglia in C5-C8 segments (Fig. 2C). Polysynaptic EPSPs were not found in this limited number of cells.
[12]
606w
In this connexion, it is worth mentioning inputs other than from primary afferents. As shown in Fig. 2Ce, a monosynaptic EPSP with fairly large amplitude was evoked when stimulus was applied to the ipsilateral lateral funiculus at the C6 segment. It was evident that the EPSP was induced neither due to spread of stimulus currents to the dorsal funiculus nor to the C6 ganglion, since stimulation of the dorsal funiculus or the ganglion at this level did not evoke any PSPs in this neurone. The EPSPs was therefore likely to be elicited by activation of ascending or descending axons in the lateral funiculus that made direct connexion with CCN neurones. | I o 'J. , "'--.-,., ,,,LI ...... ,.. t 10[ C 2 VR 2.5T 101 C 3 SPLI 2.5T o ..I ............... J,l, J o ~ .i ....... i ..,I.. if,&... I J ~ l t I i i J i 0 5 10 15 20 ms 0 5 10 15 20 ms Fig. 5A-C. Muscle specificity of input to CCN neurones. A Cb: Identification of a C2 CCN neurone by antidromic stimulation of the cerebellar peduncle. Stimulation of C2 SPL nerve elicited monosynaptic and late excitation, while C2 BCC nerve and VR (ventral rami) stimulation did not evoke any clear response. B PST histogram of extracellular spikes of another C2 CCN neurone in response to stimulation of C2 neck afferents. Excitation was evoked from C2 SPL nerve, but not from BCC nerve and VR. C Same as B, but this C3 CCN neurone was activated only from C3 BCC nerve Input from Muscle Nerves 1. Muscle Afferents Responsible for Excitation of CCN Neurones In the previous section it was verified that low threshold afferents in the dorsal root exerted monosynaptic excitatory effects on CCN neurones. The possibility of involvement of neck muscle afferents in this excitation was investigated by stimulation of ipsilateral peripheral nerves innervating various neck muscles. Figures 4A-D and 4E-G exemplify spike discharges and PST histograms of spikes of two CCN neurones in response to single shock stimulation of nerves to C2 SPL and OCC, respectively. The neurone in Fig. 4A-D exhibited a single spike with short latencies. The threshold of the response was 1.2 times the threshold for nerve volley. The firing probability increased with an increase in stimulus strength and saturated below 2 times the threshold for nerve, suggesting involvement of low threshold muscle afferents. The threshold for the early responses was as low as 1.1-1.5 times for other neurones. In the CCN neurone in Fig. 4E-G, the early and late responses were clearly seen as in the case of dorsal root stimulation (Fig. 1B). The firing probability increased with an increase in stimulus intensity, the threshold for the early response being approximately 1.5 times the threshold for nerve volley in this case. The firing probability of both the early and late responses increased with an increase in stimulus intensity up to 2.5 times the threshold for the nerve volley (Fig. 4G), and in other examples further increase of stimuli up to 5 times the threshold enhanced the response, indicating the contribution of high as well as low threshold muscle afferents to excitation of CCN neurones. Latencies of the early responses were measured on the basis of PST histograms after stimulation of four muscle nerves; SPL, BCC, RCD, and OCC (Fig. 3B). They ranged from 0.9 to 2.8 msec after stimulation. Latencies measured from the volley arriving at the spinal cord ranged from 0.5 to 2.3 ms. The majority of responses, i.e. those with latencies below 2 ms measured from the stimulus, were very likely to have been evoked monosynaptically.
[13]
288w
The above conclusion was confirmed by intracellular recording from CCN neurones. As exemplified in Fig. 4H-K, stimulation of the RCD nerve induced monosynaptic EPSPs with a latency of 0.9 ms after the stimulus and 0.6 ms after the nerve volley. The latencies of EPSPs after the stimulus ranged from 0.8 to 1.6 ms in 11 neurones (Fig. 3B). The threshold for the EPSP was below 1.1 times the threshold for the nerve volley (Fig. 4H). Amplitude of the early EPSP increased with stimulus intensity and was saturated at the intensity of 1.5 times. With a further increase in stimulus intensity, late EPSPs were superimposed on the peak of the early EPSP with a latency of approximately 1.4 ms after the stimulus (Fig. 4J-K). When a CCN neurone received excitatory input from the BCC nerve, it was generally not affected by SPL nerve stimulation (Fig. 5C). Such a specific input coming from one muscle and not from others in the same segment was found in 83 out of 92 neurones (Table 2). Thus, specificity of muscle input should be the principal pattern of connexion of neck muscle afferents with CCN neurones. This was further confirmed by intracellular recording from CCN neurones. In the C2 CCN neurone exemplified in Fig. 6A-C, stimulation of the C2 SPL nerve elicited a monosynaptic EPSP, but neither the C2 BCC nor C2 VR gave rise to any detectable PSPs at a stimulus intensity of 2.5 times the threshold for the nerve volley. In Fig. 6D-E, a C3 CCN neurone exhibited an EPSP after C3 BCC stimulation and no responses to C3 SPL stimulation. Similar results indicating muscle specificity of input were obtained with 5 other CCN neurones (2 SPL and 3 BCC nerve-excited neurones).
[14]
57w
Although the muscle specificity was the principal pattern of neck input, convergence from SPL and BCC nerves onto single CCN neurones was also observed in some cases (Table 2). However, the main input source was confined to one muscle and the excitatory convergence from the other muscle was in most cases very weak (Fig. 7A and B).
[15]
421w
In a few CCN neurones, suppression of their spontaneous activity was found after muscle nerve stimulation. In Fig. 7C, the neurone received activa- A spikes 40 30 20 10 0 iI C 3 SPL 2.5T ! C 3 BCC 2.5T .; Jh ......... , ...... / 0 5 10 15 20 ms B spc~s '~ I 0 30 2O 10 0 C2 SPL 2.5T 1 , ,= Bi ,, i | ! t C 2 BCC ||l L~a i i C spikes 30 C 2 SPL ,roD Ji=llM.i i . 9 I I 2.5T 10 i 0 im | i 2.5T / 10 I C 2 Bee 2.5T o/ ..... u J-,, L,L ...... I t t lo r SPONT '= ' 0 | lib .... Ll,,,hll,l,lkl..h.,,i L, ,11, I1,1 I' 1 20 m s 0 5 10 15 20 ms Fig. 7A-C. Convergence of segmental inputs from dorsal neck muscle afferents. A and B Convergence of monosynaptic excitatory inputs from C3 SPL and C3 BCC nerves on a C3 CCN neurone A and convergence between C2 SPL and C2 BCC nerves on a C2 CCN neurone B. C Convergence of excitatory input from C2 SPL and inhibitory input from C2 BCC nerve to a CCN neurone in C2 segment. Bottom diagram indicates spontaneous activity without stimulation as a control tion from the SPL nerve (top diagram) and suppression of spikes from the BCC nerve (middle) as compared with the control record of its spontaneous activity (bottom). There was no change of activity after stimulation of the VR. Such suppression of spontaneous activity was observed in two CCN neurones in response to stimulation of the BCC and in one neurone after VR stimulation. No suppression of spikes was found in any CCN neurone with stimulation of the SPL. That the suppression of spontaneous activity was due to production of IPSPs was confirmed by intracellular recording from two CCN neurones. In the neurone in Fig. 6F-G, located in the C2 segment, a monosynaptic EPSP was evoked by stimulation of the C2 SPL nerve and a hyperpolarizing response was elicited from the C2 BCC nerve with the latency of 2.2 ms. The hyperpolarization was increased in amplitude when the cell was depolarized by passing current through the recording electrode, indicating that it was an IPSP. The C3 segment also contains nerves innervating the muscles primarily associated with movements of the shoulder, such as the nerve to the occipitoscapularis and the spinal accessory nerve. Stimulation of these nerves exerted no effects on CCN neurones.
[16]
173w
CCN Neurones in the C1 Segment. In the C1 segment, nerve branches innervating the SPL and BCC were thin or sometimes difficult to find. The C1 nerves mainly supply branches to many small muscles; i.e. RCD major, medius and minor, OCC, and obliquus capitis cranialis. Since it was difficult to dissect all small branches innervating the muscles listed above in one experiment, various combinations of these nerves were dissected for stimulation in different experiments. The nerves thus examined were those innervating the BCC, RCD medius plus minor, RCD major, and OCC. These nerves exerted a monosynaptic excitatory action on most CCN neurones in the C1 segment. In contrast to CCN neurones in the C2 and C3 segments, afferents from more than one muscle innervated by the C1 nerve usually converged upon single C1 CCN neurones (Table 2). Such a convergence was always observed at least between functionally synergistic muscles; i.e. between the RCD major and RCD medius plus minor or between the RCD and BCC. All these muscles are elevators of the head.
[17]
98w
Late Excitation. As with dorsal root stimulation, late excitatory effects often followed monosynaptic excitation, as shown in Figs. 4E-G and 7A. Out of 101 CCN neurones which received excitation from neck muscle nerves, late excitation without preceding monosynaptic response was found in 7 neurones in the C2 segment and one neurone in the C3 segment. All of them responded to stimulation of the SPL nerve with latencies ranging from 3.1 to 4.6 ms, suggesting a di-or polysynaptic nature. Convergence of excitatory action from SPL and BCC nerves with such long latencies was observed in only two CCN neurones.
[18]
161w
The PSPs representing late excitatory effects were recorded in 4 CCN neurones. In the neurone shown in Fig. 6E, a monosynaptic EPSP evoked from the C3 BCC nerve was followed by a long-lasting EPSP (arrow). The latency of the latter was approximately 1.3 ms longer than the former, suggesting that at least the initial part of the late long-lasting EPSP could be evoked disynaptically, but its later part might be evoked polysynaptically from the primary afferent. Late EPSPs without a preceding monosynaptic EPSP were also observed. In the case of Fig. 8, both intra-and extracellular responses could be examined on the same single neurone. A burst of extracellular spikes was elicited by double shocks to the C3 SPL nerve with a latency of 3.5 ms after the first shock (the effective shock for the earliest responses in this example, Fig. 8C). When the cell was penetrated, EPSPs had a slow rising phase and generated action potentials with a similar latency to
[19]
44w
A Cb E C 3 SPL .... --% .... " :!~:LL B Cb C c 3 SPL 4.0T F --v --[ 9 -: f~,~W=, D c 3 SPL 3.2 H Z j lmV (A,C) 2mV (D-I) 40mY (B) 2ms ( A,B,E-I ) 4ms (C,D)
[20]
187w
Fig. 8A-I. Late excitation of a CCN neurone from dorsal neck muscle afferents. Extracellular A, C and intracellular B, D--I recording from the same cell. A-B Identification of a C3 CCN neurone by antidromic activation from the cerebellum. C Late evocation of extracellular spikes without preceding early excitation with double shocks to C3 SPL nerve. D Same as in C, but the recording was intracellular. E-I Temporal facilitation of late EPSP with double shocks. The EPSP was evoked by the second shock when C3 SPL nerve was stimulated at 2T H that of extracellular spikes (Fig. 8D). The latency of the EPSPs was 2.0 ms (Fig. 8D and E). When stimulated at near threshold intensity, temporal facilitation was observed in that an EPSP was evoked by the second and not the first stimulus (Fig. 8H). This strongly suggests that it was evoked disynaptically. The threshold for the EPSP was about twice the nerve threshold and the amplitude of the EPSP increased with an increase in stimulus strength (Fig. 8E-I), indicating that the afferents responsible for the EPSP were higher threshold muscle afferents, probably from spindle secondary endings.
[21]
26w
A s~kes 10 I C 2 SPL 2.5T 0 ~ L 9 , ,,, i ...... I . * .h ,.I...I 2010f L C 3 SPL
[22]
233w
d, ~I .I,= 4=,Ibl 9 ,,, I o I0 I " c4 SPL o iJ i i|~l ml ..Jmm, i 9 n i L n | 0 5 10 1520 ms B spikes 101 C 1 BCC 2.5T 0 / L-A ,. . I, I, =,* . ,,, ,, m 10 t C 2 BCC o/,-.,I ....u ,,. .li.. I...,.~ .... I ...... 1~ I ~1 5[ c 3 BCC o .... ,--L,...L .... L I ........... lo I C 4 BCC 0 [ * ,. 11 tel ........ ' ...... n n L r n 0 5 10 15 20 ms C spikes i i 20 t L OBL. CAP. CAUD. 2.5 T r L 10 o l.u, Lh.u..~.l.I..a.[ LI ..n.,. o ,,.. .,.mJl~, ~ .... i.aL ...... ,Lk..,.L t ...i -M.li,,.~ ..... 4 ~.u, li.~.. 0 t lo I C 4 SPL o ,.L.-,k,.,=, ,.,li.. '-'"L,ktkL.i,.i",L,l ' ,:' ' ' 'o 0 10 15 2 ms Fig. 9A-C. Segmental and extrasegmental inputs from dorsal neck muscle afferents. A An example of a C3 CCN neurone which received input only from C3 SPL nerve. B Another C3 CCN neurone which received extrasegmental input from C4 BCC nerve as well as segmental input from C3 BCC nerve. C A rostral C2 CCN neurone excited by stimulation of OCC, C2 SPL, and C3 SPL nerves. Note stronger excitation from OCC than from C2 SPL
[23]
167w
The afferent nerves from SPL or BCC enter the spinal cord through the dorsal roots of C1-C4 segments. It was investigated whether afferents from the same muscle entering different segments converged onto the same CCN neurones. In Fig. 9A, a CCN neurone in the C3 segment was excited only from the C3 SPL nerve and neither from the C2 nor from the C4 SPL nerve. Since there might be weak convergence which could not be revealed with extracellular spikes, intracellular recording was also made to assess this segment specificity of input. In the C3 CCN neurone exemplified in Fig. 10A-D, EPSPs were induced only after stimulation of the C3 SPL nerve and not from the SPL nerve in different segments, confirming the existence of segment specificity of CCN input. Such a segment specificity of neck muscle input was observed in 30 of 69 CCN neurones recorded extracellularly and tested for inputs from at least 3 different segments (9/14 C3 neurones, 11/39 C2 neurones and 10/16 C1 neurones).
[24]
111w
In the remaining 39 CCN neurones (57%), excitatory effects were induced from afferents of different segments as well (Fig. 9B). Such extrasegmental excitation was evoked from afferents of rostral as well as caudal segments. As shown in Table 3, however, excitatory inputs from rostral segments were confined to afferents of the immediately rostral segment, while caudal inputs extended over one to three segments. Moreover, any CCN neurone which received input from a rostral segment was found only in the very rostral part of its own segment. Thus, it appeared to be a general tendency that extrasegmental inputs to CCN neurones from caudal segments predominated in frequency over those from rostral segments.
[25]
105w
The origin of extrasegmental input was confined to the same muscle so far as the C2, C3 and C4 segments were concerned, except for two CCN neurones which received convergence from BCC and SPL nerves (Table 2). Such a muscle specificity of extrasegmental inputs as well as segmental was verified by intracellular recording. The neurone in Fig. 10E-J, located in the rostral C3, received convergent inputs from C2-C4 BCC nerves. Monoand polysynaptic EPSPs were elicited, the largest being the EPSP from the C3 BCC nerve. Muscle specificity of extrasegmental inputs was ascertained by the absence of any clear PSPs following stimulation of C2-C4 SPL nerves.
[26]
51w
Input from C1 nerves to CCN neurones located in the rostral C2 segment was quite unique in various aspects. Four such neurones received muscle afferent input from C1 nerves and not from C2 nerves. Rostral C2 CCN neurones, which received input from the C2 SPL nerve, were excited from the C1
[27]
122w
~mm,,tml/.~-m~ 2ms OCC nerve or from both OCC and RCD nerves. The C1 OCC input to rostral C2 CCN neurones was especially strong, even stronger than C2 SPL input (Fig. 9C). SPL is a lateral flexor of the head, RCD is a head elevator, and OCC subserves head rotation, thus each of them has an independent function (Reighard and Jennings 1935). Extrasegmental convergence onto single CCN neurones from functionally different muscles was extremely rare among C2, C3, and C4 segments. Latencies of excitation of CCN neurones by extrasegmental inputs were 0-0.6 ms longer than those of excitation by segmental inputs, except for C1 input, especially OCC input, which often had shorter latencies than those from C2 nerves for activating C2 CCN neurones.
[28]
241w
The present findings obtained by intraaxonal staining with HRP are in agreement with the above described previous studies. An important addition was that axons identified as low threshold muscle afferents (SPL and BCC) and projecting to the CCN in the same segment as the nerve entry were invariably collaterals of axons projecting to the motor nucleus. Thresholds of these afferents were near the threshold for the incoming volley, indicating that they belonged to the lowest threshold group of muscle afferents. Further, these afferents showed a clear dynamic response (cf. Richmond and Abrahams 1979). Thus, all of the HRP-stained single axons projecting to both the CCN and the motor nuclei (Fig. 11) were most likely from spindle primary endings. Previous studies showed that monosynaptic EPSPs evoked in SPL or BCC motoneurones by stimulation of the SPL or BCC muscle nerve had thresholds near the threshold of the nerve, suggesting that these EPSPs were evoked mainly by stimulation of afferents from muscle spindle primary endings, i.e. group Ia fibres (Wilson and Maeda 1974;Anderson 1977;Ezure et al. 1978;Rapoport 1979;Brink et al. 1981). Our results also showed that monosynaptic EPSPs evoked in CCN neurones had thresholds near the threshold of the nerve. The present morphological findings further suggest that monosynaptic EPSPs in CCN neurones following stimulation of the muscle nerve are induced mainly from axon collaterals of afferents from spindle primaries which are responsible for generation of monosynaptic EPSPs in motoneurones in the same segment.
[29]
38w
Afferents from tendon organs are known to exist in dorsal neck muscles (Richmond and Abrahams 1979). The present study cannot provide evidence for the contribution of these afferents to excitation of CCN neurones following stimulation of muscle nerves.
[30]
24w
The firing probability of early excitation increased as the intensity of stimulation of the muscle nerve increased up to 2.5-5.0 times the nerve threshold.
[31]
306w
The late EPSP, which was superimposed on the early EPSP with a delay of 0.5-0.7 ms from the onset of the early EPSP, had threshold at 1.5-2.0 times the nerve threshold. This gives the possibility that higher threshold muscle afferents such as those from spindle secondaries might contribute to the late EPSP. Although threshold separation between groups I and II in the neck muscle afferents has not been investigated, Richmond and Abrahams (1979) reported that the conduction velocity of afferents from SPL spindle primaries and secondaries had a mean value of 65 m/s and 29 m/s, respectively. Given the conduction distance of approximately 3 cm in the peripheral nerve in the present experiments, the difference in conduction time between the two kinds of afferent fibres would be 0.6 ms, the value being similar to the latency difference between early and late EPSPs mentioned above. These results are in agreement with the hypothesis that spindle afferents from not only primaries but also secondaries project to CCN neurones monosynaptically, as has been demonstrated for hindlimb group II fibres from spindle secondary endings to motoneurones (Kirkwood and Sears 1975;Stauffer et al. 1976;Fyffe 1979;Mannen et al. 1981;Brown 1981) as well as to dorsal spinocerebellar tract cells (Eccles et al. 1961). In the histogram for EPSP latencies in Fig. 3B, the range of latencies could be divided into two groups; i.e. 0.8-1.2 ms and 1.4-1.6 ms. The latter group (see also Fig. 10B, C3-SPL) corresponds well to the latency after stimulation of the above described late EPSP. This suggests that afferents from spindle secondaries could project to a population of CCN neurones without a conjoint projection from spindle primaries. However, the hypothesis of projection from spindle secondaries to CCN neurones requires some reservation, because the possibility of disynaptic connexions from spindle primaries for generation of late EPSPs cannot entirely be excluded.
[1]
189w
The intramedullary course of the dorsal root fibres of the first three cervical nerves has been studied in Marchi preparations of the cat (Ranson et al. 1932;Corbin and Hinsey 1935). They showed that the most conspicuous group of collaterals went to the CCN region and that scattered fibres could be traced forward into the ventral horn. Degeneration experiments further showed that preterminal fibres were found in the CCN and the motor nucleus in the ventral horn (Escolar 1948;Petras 1965;Shriver et al. 1968;Imai and Kusama 1969;Cummings and Petras 1977). Electron microscopic evidence has been presented for the presence of degenerated dorsal root terminals upon the dendrites and somata of CCN neurones (Cummings and Petras 1977). In a Golgi study, most afferent fibres to the CCN were shown to come from the dorsal funiculus, though some were traced from the lateral funiculus (Wiksten 1979a). The present study provides evidence for muscle afferent projection to the CCN, whereas previous anatomical observations have given no information on the source of the afferents. Considerations on receptor origin of the muscle afferents and the pattern of their convergence onto CCN neurones will be given below.
[2]
146w
In C2 and C3, stimulation of the dorsal root ganglia excited virtually all CCN neurones in the same segment, and the majority of neurones were also activated from the ganglia of other segments (Table 1). The powerful action from the dorsal root of the immediately adjacent segment was consistent with anatomical findings (Imai and Kusama 1969;Shriver et al. 1968). In contrast, when muscle nerves were stimulated, extrasegmental inputs between C2 and C3 (Table 2) as well as segmental inputs were less effective. It would be reasonable to suggest that the difference in effects between the dorsal root and the muscle afferent stimulation originated from joint afferents at least in part (Hirai et al. 1978), though effects of afferents from small, intervertebral muscle have yet to be studied. The existence of afferents originating from around the neck joint region has been shown anatomically (Richmond and Bakker 1982).
[3]
168w
It is important to compare the pattern of convergence of afferents from different muscles onto single CCN neurones with that onto single motoneurones. Wilson and Maeda (1974) and Anderson (1977) found that there were few Ia connexions between SPL and BCC motoneurones: no PSPs were observed in motoneurones innervating one muscle group while stimulating the nerve of the other. Similar relations between SPL and BCC were observed in CCN neurones in the C2 or C3 segment: when a CCN neurone received excitation from the nerve of one muscle, it was rarely affected by stimulation of the nerve of the other. The similarity between inputs to motoneurones and CCN neurones is in accord with the present finding that motoneurones and CCN neurones in the same segment receive common input from the same fibres through their axon collaterals. Such muscle specificity of input has been found in the dorsal spinocerebellar tract cells in Clarke's column as well (Holmqvist et al. 1956;Eccles et al. 1961;Hongo and Okada 1967;Kuno et al. 1973).
[4]
134w
With respect to the extrasegmental input, its connectivity with CCN neurones and that with motoneurones is also very similar. Brink et al. (1981) found that monosynaptic EPSPs in motoneurones of the SPL or biventer cervicis were produced by stimulation of the muscle nerve containing the motoneurone's axon and often by stimulation of more caudal nerves to the same muscle. Frequency of EPSP occurence and average EPSP amplitude decreased as nerves in more distant segments were stimulated. Stimulation of nerves rostral to the motoneurones was less effective. The rostrocaudal asymmetry of inputs observed for motoneurones resembled that for CCN neurones (Table 3). This similarity suggests that spindle afferents coming from different segments may also project to CCN neurones and motoneurones in common through their axon collaterals, though this is not yet verified by intraaxonal staining.
[5]
128w
In addition to monosynaptic excitatory pathways, there may be polysynaptic, including disynaptic, excitatory and inhibitory pathways to CCN neurones, in view of the relatively long latencies of their responses after stimulation of the dorsal root ganglion or the muscle nerve. Abrahams et al. (1979) reported excitation of CCN neurones with rather long latencies (2-30 ms); they seem to have observed effects mainly via polysynaptic pathways. For spinocerebellar tract neurones in the hindlimb segments, polysynaptic pathways from muscle afferents are known to exist, and interneurones mediating these polysynaptic effects involve common interneurones which mediate both the segmental reflex and the spinocerebellar transmission (Lundberg and Weight 1971;Lundberg 1971;Gustafsson and Lind-str6m 1973;Lindstr6m and Schomburg 1973;Hongo et al. 1983a, b). In the neck segments, location of such interneurones has not been defined.
[6]
103w
A few comments may be given with respect to polysynaptic pathways to CCN neurones from the forelimb segments. Wiksten (1979a) has described axons that enter the CCN from the lateral funiculus. We have found CCN neurones that are monosynaptically excited by stimulation of the lateral funiculus at C6. Although we could not identify whether the axons activated were ascending or descending in the lateral funiculus, it appears possible that polysynaptic effects on CCN neurones from afferents of the forelimb segments are mediated by interneurones located in these segments and their axon ascends in the lateral funiculus to give collateral connexions to CCN neurones.
[7]
864w
Ezure K, Sasaki S, Uchino Y, Wilson VJ (1978) Frequencyresponse analysis of vestibular-induced neck reflex in cat. II. Functional significance of cervical afferents and polysynaptic descending pathways. J Neurophysiol 41:459-471 Fyffe REW (1979) The morphology of group II muscle afferent fibre collaterals. J Physiol (Lond) 296: 39P-40P Gustafsson B, Lindstrfm S (1973) Recurrent control from motor axon collaterals of Ia inhibitory pathways to ventral spinocerebellar tract neurones. Acta Physiol Scand 89: 457-481 Hirai N, Hongo T, Sasaki S (1978) Cerebellar projection and input organization of the spinocerebellar tract arising from the central cervical nucleus in the cat. Brain Res 157:341-345 Hirai N, Hongo T, Sasaki S, Yoshida K (1979) The neck and labyrinthine influences on cervical spinocerebellar tract neurones of the central cervical nucleus in the cat. Prog Brain Res 50:529-536 Hirai N, Hongo T, Sasaki S (1984) A physiological study of identification, axonal course and cerebellar projection of spinocerebellar tract cells in the central cervical nucleus in the cat. Exp Brain Res 55:272-285 Holmqvist B, Lundberg A, Oscarsson O (1956) Functional organization of the dorsal spinocerebellar tract in the cat. V. Further experiments on convergence of excitatory and inhibitory actions. Acta Physiol Scand 38:76-90 Hongo T, Jankowska E, Ohno T, Sasaki S, Yamashita M, Yoshida K (1983a) Inhibition of dorsal spinocerebellar tract cells by interneurones in upper and lower lumbar segments in the cat. J Physiol (Lond) 342:145-159 Hongo T, Jankowska E, Ohno T, Sasaki S, Yamashita M, Yoshida K (1983b) The same interneurones mediate inhibition of dorsal spinocerebellar tract cells and lumbar motoneurones in the cat. J Physiol (Lond) 342:161-180 Hongo T, Okada Y (1967) Cortically evoked pre-and postsynaptic inhibition of impulse transmission to the dorsal spinocerebellar tract. Exp Brain Res 3:163-177 Imai Y, Kusama T (1969) Distribution of the dorsal root fibers in the cat. An experimental study with the Nauta method. Brain Res 13:338--359 Ishizuka N, Mannen H, Hongo T, Sasaki S (1979) Trajectory of group Ia afferent fibers stained with horseradish peroxidase in the lumbosacral spinal cord of the cat: Three dimensional reconstructions from serial sections. J Comp Neurol 186: 189--211 Kirkwood PA, Sears TA (1975) Monosynaptic excitation of motoneurones from muscle spindle secondary endings of intercostal and triceps surae muscles in the cat. J Physiol (Lond) 245: 64P-66P Kuno M, Munoz-Martinez E J, Randic M (1973) Sensory inputs to neurones in Clarke's column from muscle, cutaneous and joint receptors. J Physiol (Lond) 228:327-342 LindstrOm S, Schomburg ED (1973) Group I inhibition in Ib excited ventral spinocerebellar tract neurones. Acta Physiol Scand 90:166-185 Lundberg A, Weight F (1981) Functional organization of connexions to the ventral spinocerebellar tract. Exp Brain Res 12: 295-316 Lundberg A (1971) Function of ventral spinocerebellar tract. A new Hypothesis. Exp Brain Res 12:317-330 Mannen H, lshizuka N, Hongo T, Kudo N, Sasaki S, Yamashita M (1981) Quantitative analysis of the morphology of group II fibers in the spinal cord of the cat. Neurosci Lett Suppl 6:$96 Matsushita M, Ikeda M (1975) The central cervical nucleus as cell origin of a spinocerebellar tract arising from the cervical cord: a study in the cat using horseradish peroxidase. Brain Res 100:412-417 Matsushita M, Hosoya Y, Ikeda M (1979) Anatomical organization of the spinocerebellar system in the cat, as studied by retrograde transport of horseradish peroxidase. J Comp Neurol 184:81-106 Petras JM (1965) Afferent peripheral nerve fibers to the spinal cord and dorsal column nuclei in the cat. An analysis and comparison with the distribution of terminal efferent brain fibers to the spinal cord. Anat Record 151:399-400 Petras JM (1977) Spinocerebellar neurons in the rhesus monkey. Brain Res 130:146-151 Ranson SW, Davenport HK, Doles EA (1932) Intramedullary course of the dorsal root fibers of the first three cervical nerves. J Comp Neurol 54:1-12 Rapoport S (1979) Reflex connexions of motoneurones of muscles involved in head movement in the cat. J Physiol (Lond) 289: 311-327 Reighard J, Jennings HS (1935) In: Elliott R (ed) Anatomy of the Cat. Holt, Rinehart and Winston, New York, pp 131-136 Rexed B (1954) A cytoarchitectonic atlas of the spinal cord in the cat. J Comp Neurol 100:297-379 Richmond FJR, Abrahams VC (1979) Physiological properties of muscle spindles in dorsal neck muscles of the cat. J Neurophysiol 42:604-617 Richmond FJR, Bakker DA (1982) Anatomical organization and sensory receptor content of soft tissues surrounding upper cervical vertebrae in the cat. J Neurophysiol 48:49-61 Shriver JE, Stein BM, Carpenter MB (1968) Central projections of spinal dorsal roots in the monkey. I. Cervical and upper thoracic dorsal roots. Am J Anat 123:27-74 Stauffer EK, Watt DGD, Taylor A, Reinking RM, Stuart DG (1976) Analysis of muscle receptor connections by spike triggered averaging. 2. Spindle group II afferents. J Neurophysiol 39:1393--1402 Wiksten B (1975) The central cervical nucleus -a source of spinocerebellar fibres, demonstrated by retrograde transport of horseradish peroxidase. Neurosci Lett 1:81-84 Wiksten B (1979a) The central cervical nucleus in the cat. I. A Golgi study. Exp Brain Res 36:143--154 Wiksten B (1979b) The central cervical nucleus in the cat. II. The cerebellar connections studied with retrograde transport of horseradish peroxidase. Exp Brain Res 36:155-173 Wilson VJ, Maeda M (1974) Connections between semicircular canals and neck motoneurons in the cat. J Neurophysiol 37: 346-357 Received August 17, 1983