Brain Research, 568 (1991) 178-184 (~) 1991 Elsevier Science Pubhshers B.V. All rights reserved. 0006-8993/91/$03.50
178
BRES 17294
The cerebellopontine system: an electrophysiological study in the rat Sabina Berretta, Gianfranco Bosco, Giuseppe Smecca, and Vincenzo Perciavalle Instuute of Human Physiology, Umverstty of Catama, Catama (Italy) (Accepted 13 August 1991)
Key words: Basilar pontme nucleus; Cerebellar lateral nucleus; 7-Aminobutyric acid; Reticulotegmental nucleus; Rat
We examined the effects of electric stmaulation of the cerebellar lateral nucleus (LN) in the rat on the activity of single pontocerebellar neurons in the basilar pontme nuclei (BPN) and the reticulotegmental nucleus (RtTg). We found that about half of the cells of these nuclei that were influenced by LN stimulation were inhibited. A significant fraction of both excitatory and mhibRory responses had latencies of less than 4 ms and were able to follow high frequency stimulation, compatible with a monosynaptic linkage. ExtraceUular field potential recordings within the BPN and RtTg were interpreted as arising from impulses propagating along inhibitory axons projecting in a bundle from the cerebellum to these pontine structures. Microionophoretlc administration of GABA antagonists bicuculline or pmrotoxin abolished or attenuated most inhibitory effects. Therefore, we conclude that LN-mduced inhibition is most likely mediated by cerebellopontine GABAerglc fibers. The functional significance of this cerebellopontine inhibitory circuit is discussed. INTRODUCTION It is well k n o w n 24 that the s u p e r i o r c e r e b e l l a r p e d u n cle contains d e s c e n d i n g fibers that r e a c h t w o i m p o r t a n t p r e c e r e b e l l a r structures, t h e basilar p o n t i n e nuclei ( B P N ) and the r e t i c u l o t e g m e n t a l nucleus ( R t T g ) .
In several
species t h e s e d e s c e n d i n g p r o j e c t i o n s o r i g i n a t e m a i n l y f r o m the c e r e b e l l a r lateral nucleus ( L N ) 1'3'5'6'9'20'26'30, with a s m a l l e r c o n t r i b u t i o n f r o m interpositus nucleus and v e r y few fibers f r o m the m e d i a l nucleus. F e w studies h a v e focused o n the functional p r o p e r t i e s of t h e s e p r o jections. E a r l y e l e c t r o p h y s i o l o g i c a l e x p e r i m e n t s indic a t e d that the c e r e b e l l o p o n t i n e system is e x c i t a t o r y 14"28, and s u b s e q u e n t ultrastructural studies r e p o r t e d that cerebellopontine terminal boutons contain mostly round synaptic vescicles 29. H o w e v e r , it was r e c e n t l y o b s e r v e d with d o u b l e - l a b e l i n g t e c h n i q u e s that s o m e c e r e b e l l o p o n tine n e u r o n s are i m m u n o r e a c t i v e for g l u t a m a t e - d e c a r b o x y l a t e 4 as well as y - a m i n o b u t y r i c acid ( G A B A ) 1. This suggests that at least a p a r t of the c e r e b e l l o p o n t i n e syst e m m a y be inhibitory. T h e p r e s e n t study was u n d e r t a k e n to r e - e x a m i n e the effects o f electric s t i m u l a t i o n of the L N o n the spontan e o u s activity o f single n e u r o n s of the B P N as well as the R t T g in the rat. MATERIALS AND METHODS Extracelhilar responses of single pontme cells were recorded on
24 male Sprague-Dawley rats (230-280 g) anesthetized with urethane (1.2 g/kg l.p.). Recordings were made using glass microplpettes (7-18 Mf~ resistance) filled with a 4% solution of Pontamine sky blue in 1 M KC1. The microelectrodes were positioned within the BPN and RtTg according to the stereotaxic coordinates of the atlas of Paxinos and Watson 24. A screw posmoned in the skull served as indifferent electrode. Action potentials were regarded as ansmg from the cell somata when they appeared initially negativepositive and then, as the electrode advanced, positive-negative with a clear inflection on the ascending phase 1°. The end of each electrode penetration was marked with a small (50-100/zm) iontophoretic injection of pontamine (10/~A for 20 mm) for subsequent histological identification. A monopolar stimulating electrode (0.6-1 2 Mf~ resistance nickel-chrome wire) was stereotaxically placed in the contralateral LN. Stimuli were 1-5 cathodal shocks (0.05-0.3 ms, 100-700 Hz) delivered at random intervals from 1.5 to 5 s For two experiments in which field potentials were analyzed, a concentric electrode having a shaft diameter of 150/~m and a central stainless steel w~re was used to stimulate the LN. The electrode was insulated with varmsh (Epoxylite) except for an exposed tip of about 200/zm. The cathode was formed by e~ther external or internal pole. In some experiments, 5-barreled glass mlcropipettes were used for recording and ionophoretic application of drugs. The central barrel (impedance 4-8 MQ) contained a 4% solution of Pontamine sky blue m 1 M KC1. An outer barrel, filled with NaC1 3 M, served as balance channel, while the remammg barrels were filled with the following drugs: GABA (Sigma: 0.4 M, pH 4.0); bicuculhne (Sigma; 20 mM in NaC1 150 mM, pH 3.0); picrotoxm (Sigma; 10 mM in NaC1 150 mM, pH 3.8) Three constant current sources and an automatic balance unit (Neurophore) were used for drug ejection Anodal currents of up to 150 nA were used to lonophorese the substances, while a cathodal current of 10-15 nA was used for retaining them during the penetration After the recording sessions, the s~te of stimulation was marked with an electrolytic reference les~on (20/~A cathodal current for 30 s) and the animals were perfused with a 4% paraformaldehyde solution while under deep urethane (2 g/kg i.p.) anesthesia. Recon-
Correspondence: V. Percaavalle, Istituto di Flslologla Umana, Universlta" di Catanla, Vlale Andrea Dona 6, 95125 Catama, Italy. Fax: (39) (95) 330645.
179 BPN
RESULTS
RtTg
2O"
20-6122 SD .4001 n -41
16-
~2"
1612"
8-
X -5143 SO -4 036 n .14
8" ~
4-
EXCITATIONS .
0
o
E
I
[]
4"
INHIBITIONS
812'
12-
~ . 7 722 SD - 5 ~ n .54
16. 20-
~-5556 SD.3203 n.18
24
i
I
I
I
I
I
I
I
I
I
I
I d ~
I I I I I I I I I I I I 0 2 4 6 8 10121416162022
0 2 4 6 8 101214161820223032
L A T E N C Y (ms)
Fig. 1. Histograms showing latency of excitation (upper) and inhibition (lower) displayed by neurons belonging to the BPN and to the RtTg following stimulation of the contralateral cerebellar LN.
structlons of the stimulation site and electrode penetrations were made from frozen sections of 30 gm thickness, stained with Neutral red. To exclude the possibility that the effects of LN stimulation were due to activation of afferent terminals and/or of fibers of passage, the LN neurons in 6 rats were chemically ablated prior to testing. An injection of i gg of kainic acid (Sigma) dissolved in I #1 of 0.05 M buffered phosphate solution (pH 7.4), was made unilaterally into the LN area 15 days prior to the electrophysiological experiment. The evaluation of unitary discharges was performed by converting 60-90 responses into post-stimulus time histograms (PSTHs) and cumulative frequency distributions (CFDs). Excitatory and inhibitory responses were sequences of at least 3 bins (bin width 0.5-2 ms), which showed frequency values of more than twice the standard deviation (S.D.) above or below the mean value during spontaneous activity. Solitary bins between these sequences were not considered. Latency was the time interval between the last shock of the stimulating train and the first bin of the sequence, and the duration was the distance between first and last bin TM.
Unit activity was recorded from more than 600 pontine cells, 261 of which were located within the B P N (193) and RtTg (68). A b o u t half of the cells in these two nuclei were influenced by stimulation of the contralateral LN, 95/193 (49.2%) in the B P N and 32/68 (47.1%) in the RtTg. In the BPN, the prevailing LN-induced effect was inhibitory. As illustrated by the latency distributions in the left part of Fig. 1, inhibitory responses having latencies ranging from 2 to 32 ms (mean value 7.72 ms -+ 5.50 S.D.) were recorded in 54 of the 95 B P N ceils (56.8%) while excitatory responses were recorded in the remainhag 41 cells (43.2%). The excitatory responses had latencies ranging from 2 to 18 ms (mean value 6.12 ms -+ 4.01 S.D.). A significant fraction of the excitation (18/41; 43.9%) and inhibition (9/54; 16.7%) had latencies less than 4 ms. All of these responses were elicited with a single shock and followed stimulation frequencies up to 70-80 Hz. Increasing the pulse number in the stimulation train did not significantly change the latency of these responses. The effects elicited by L N stimulation on two neurons located within the medial part of the B P N are illustrated in Fig. 2. It can be seen that the first B P N neuron was excited and the second inhibited by stimulation of contralateral LN with a single 0.2 ms cathodal pulse (80
~A). A similar preponderance of inhibitory responses to L N stimulation was observed in the RtTg (Fig. 1, right). Although the sample size was smaller, similar fractions of cells displayed inhibitory (18/32 cells; 56.3%) and excitatory (14/32; 43.7%) responses. Response latencies were slightly shorter on average and slightly larger frac-
I II II
-I
i
II
III
LJIIItllllllllJIIII
-I
] III ! IIIIII ~l I11t11~
,Om.JO: Fig. 2. Evoked responses displayed by two neurons located in the medial part of the BPN following stimulation (pulse duration 0.2 ms; pulse intensity 80 #A) of the contralateral LN. The records are formed by 10 superimposed traces. Calibration bar in the camera lucida drawing: 1 ram.
180 tions had latencies shorter than 4 ms (8/14; 57.1% for excitatory and 6/18; 33.3% for inhibitory responses). The responses were also capable of following stimulation frequencies up to 70-80 Hz. Examples of excitatory and inhibitory effects on two RtTg cells evoked by stimulation of the LN with a single 0.2 ms cathodal pulse (90 #A) is illustrated by the PSTHs and CFDs of Fig. 3. It is worth noting that in these cells as well as in the remaining cells examined in this study the frequency of the prestimulus neuronal activity was similar to that in the late part of PSTHs and CFDs. The location of cells recorded within the BPN and RtTg are illustrated in Fig. 4. Cells that were inhibited by LN stimulation are marked by filled triangles and those that were excited by open triangles. Cells that did not respond to LN stimulation are marked by dots. It can be seen that, although inhibited and excited cells are present throughout the BPN and RtTg area, inhibited cells appear more concentrated in the anterior part and the excited cells in the more posterior part of the BPN. The results reported above show that LN stimulation elicits short latency responses in BPN and RtTg neurons that could result from monosynaptic excitation or inhibition. In order to rule out possible contributions from the activation of LN afferent terminals or fibers of passage, the cells of the LN were chemically ablated using kainic acid in 6 rats 15 days before the electrophysiological recording. In 3 of the rats, histological examination showed more than 70% fewer cells in the contralateral
LN compared to the ipsilateral control. In these rats, stimulation of contralateral LN area was tested on 13 BPN and 9 RtTg neurons. Only two BPN neurons displayed discharge changes that were excitatory in nature and appeared with a latency of 1.8 and 2.2 ms, respectively. Short latency inhibitory effects were not observed. Further evidence for a monosynaptic activation from LN was obtained from field potential recordings. Fig. 5 shows the results of a systematic exploration of the field potentials induced in the BPN and RtTg areas by LN stimulation. In the region labeled 1 in Fig. 5, just dorsal to the two nuclei, the field potentials consist of a small negative deflection followed by a relatively small positivity and a second small negative deflection. As the recording electrode was shifted ventrally across the RtTg and BPN, two peaks, called a and fl spikes, became more evident. The a spike started as early as 1.05 ms and attained its peak at 1.60 ms after stimulation of the LN, while the fl spike began at 2.04 ms with summit at 2.24 ms. The deflection equivalent to the a spike was greater in the rostral tract than in the caudal tract, while in both tracts there were less prominent delayed negative deflections, which apparently belonged to the fl spike. These spikes could be discriminated by varying the stimulus intensity. The a spike appeared at a relatively low stimulation intensity and attained its maximum before the fl spike was produced by a further increase in stimulus strength. We tested directly whether G A B A is involved in the 45
. . . .
• .....
~....+,,I
I
....
i. . . . .
.
.+
i
.
o9
"~. 180
~ E
5 ms/d=v
18
e-
, .i+,m ,,,,Li,+,I.,+,,J+.JIJ+, ..... +,i, i
18o
I
10 ms/dtv
Fig. 3. PSTHs a n d CFDs illustrating the response patterns displayed by two RtTg neurons following stimulauon (pulse duration 0.2 ms; pulse intensity 90 #A) of the contralateral LN (same animal as Fig. 2). Histograms were compiled from 70 sweeps. Calibration bar m the camera lumda drawing: 1 ram.
181
L01
Rt'r~_ _
B
A
D
D •
•
z
•
•
•
, •
~,•
BPN
•
• &
~
•
~AA °
z
.
.
a
.
•
A
.
z
.
z
.
•
~
•
At=/
, V
.
RtTg
-'.',~':'::-.L0,4 "j~, a , ,A.z= a"
1 mm t
A
I
B +
1
LO.9
2
F-
3
L14
1 m6
!
!
1 mm Fig. 4. Four serial saglttal sections (redrawn from Paxinos and Watson~) summarizing the location within the BPN and RtTg of cells inhibited (filled triangles), excited (open triangles) or non-influenced by LN stimulation. A, anterior; D, dorsal; P, posterior; V, ventral.
inhibitory responses of these neurons by local iontophoretic injection of G A B A and the G A B A antagonists bicuculline and picrotoxin at the recording sites in 6 rats. Recordings were obtained from 46 neurons (32 belonging to the BPN and 14 to the RtTg). Spontaneous firing of all neurons was abolished or greatly attenuated within 1-3 s following the application of 30-40 nA injection of G A B A . The depressant effects of G A B A were reversed in 96% of cases (44/46) by an iontophoretic application (70-80 nA) of bicucuUine, which alone increased the spontaneous firing of 2/3 of the tested neurons by 1540%, suggesting a disinhibition of a tonic G A B A input to these neurons. The depressant effect of G A B A was also abolished in 87% of cases (40/46) by application of picrotoxin which, however, did not significantly increase the spontaneous discharge of tested neurons. The sensitivity to these drugs was also tested on 11 neurons which displayed an LN-induced inhibition. Of these cells, bicuculline administration abolished or greatly attenuated
Fig. 5. Field potentials induced by LN stimulation (pulse duration 0.1 ms; pulse intensity 250/zA) recorded above and in the RtTg and BPN during two electrode penetrations. Electrode tracks and location of the recording sites are illustrated on the saglttal section of brainstem at the top of the figure. The records are formed by 10 superimposed traces. In the lower record of track A, a and fl denote the two peaks of the field potential.
the post-stimulus inhibition in 7 (63.6%). In addition, 5 cells were studied for a sufficiently long interval to observe a return of the post-stimulus inhibitory effect after cessation of bicuculline administration. The time interval required for the return of the LN-induced inhibition varied from 2 up to 11 min. Moreover, the same 11 neurons were tested for the effects of picrotoxin; it was observed that in 6/11 cells (54.5%) picrotoxin also abolished the inhibitory effect elicited by LN stimulation. An example of the effects elicited by bicuculline and picrotoxin on the LN-induced inhibitions displayed by a BPN and a RtTg neuron is illustrated in Fig. 6. DISCUSSION The present study demonstrates that about half of the neurons in the BPN and RtTg are significantly affected by LN stimulation, and more than 50% of those cells were inhibited. Many of the responses, both excitatory and inhibitory, were compatible with a monosynaptic
182
181
BPN
18
before I
,L,
j,==,
Rt Tg before
,I,L= ,,~ ,hjL-~II ~ ,~l,,,,, I~,IIL,,,, ,,J,I, I
I
-,,._..
---__
300
180
181
~
blcuculline (60 nA)
18 I.
I ,1~,.,~.......... h~, J..... ~.~,, ,L.~ l .......
-'~.
300 18
I
plcrotoxm (80 hA)
.......
1
11 mm after
!
,
18
I
•
300
5 mm after
,~ ~J~,,, i,, k,,,~iH,, ~a, ~,,,,,I, ,,,,J,~,l
,
180
i
,
10ms/d=v
Fig. 6. PSTHs and CFDs illustrating the effect of the iontophoretic application of the GABA-antagonists bicuculline (5 mM in NaCI 150 raM, pH 3.0) and picrotoxin (10 raM in NaCI 150 raM, pH 3.8) on the inhibition displayed by a BPN and a RtTg neuron, respectively. Inhibitory responses were elicited by stimulating the LN with a single cathodal 0.2-ms pulse (intensity 110 and 90/~A, respectively). Note in both cases that, during delivering of the drug, the LN stimulation did not evoke a significant discharge change, while the LN-induced inhibition reappeared 11 and 5 rain after the cessation of administration, respectively. Histograms were compiled from 70 sweeps and have bin-width of 1 ms.
linkage; they had short latencies (< 4 ms) and were capable of following high frequency stimulation. Longer latency responses were probably due to polysynaptic linkages, e.g. via the cerebcllo-thalamo-cortico-pontine pathway. These results suggest that inhibitory LN neurons synapse directly on pontine neurons. It was still possible, though, that the inhibitory effects were not directly mediated by LN neurons but rather by fibers of passage in the LN or by cells having afferent terminals within the LN. However, after at least 70% of the cells in the LN were destroyed by kainic acid, leaving fibers of passage
or afferents intact7'19, stimulation of the LN no longer evoked short latency responses in the pontine nuclei. We conclude, therefore, that the activation of the LN cells is responsible for the short latency inhibitory responses observed in the BPN and RtTg. The majority of the BPN and RtTg cells displayed a significant decrease of their spontaneous discharge upon microiontophoretic application of GABA and LN-induced inhibition was abolished or greatly reduced following application of the classic GABA antagonists bicuculline and picrotoxin (cf. ref. 13). Therefore, a significant percentage of cerebellopontine projections are
183 inhibitory in nature and the majority of them appear to be GABA-mediated. Monosynaptic inhibition of cells in the BPN or RtTg from the cerebellar nuclei has not been described before. However, some cerebellopontine neurons have been shown to be glutamate-decarboxylase positive4 or GABA-Iike immunoreactive ~, and several small cell bodies containing G A B A have been found scattered within the deep cerebellar nuclei, with the highest incidence in the LN m22. In the eat, about 10% of cerebellar nuclear cells projecting to the BPN and RtTg are GABA-Iike immunoreactive 1. In the rat, we now show that about 10% of the cells in the BPN (9/95) that were influenced by LN stimulation were inhibited at short latency. In the RtTg the fraction was closer to 20% (6/32). Our data also imply that GABAergic neurons in the LN exert an inhibitory influence on neurons in the pons. Cells in the BPN and RtTg were found to react to local iontophoretic injection of G A B A which significantly reduced their spontaneous discharge. Moreover, LN-induced inhibition was abolished or greatly reduced following application of the classic G A B A antagonists, bieueulline and picrotoxin (el. ref. 13). Nevertheless, a significant fraction (more than a third) of the short latency inhibitory responses were not affected by these antagonists. Since no other inhibitory transmitter has been thus far implicated immunohistochemically, and there are only few intrinsic neurons within the BPN 21, we suggest that some of the non-GABAergic inhibitory responses may be due to presynaptic inhibition. That is, they may have resulted from an excitatory action of cerebellopontine projections upon fiber terminals (corticopontine) exciting PBN or RtTg neurons. Thus we would estimate that about two-thirds of the short latency inhibitory responses in these pontine nuclei are the result of postsynaptic inhibition mediated monosynaptically by GABAergic neurons in the LN and the remaining onethird could be dependent on inhibitory synapses far from the electrode tip and unreached by the applied antagonists, or they are the result of presynaptic inhibition mediated by excitatory neurons. We would suggest, therefore, that there is a dual control exerted by the cerebellar nuclear cells on BPN and RtTg, an inhibitory control providing negative feedback and an excitatory control providing both positive and negative feedback. The results of the field potential recordings also support the suggestion of a dual innervation by LN neurons.
REFERENCES 1 Aas, ,I.E. and Brodal, P., GABA and glycine as putative neurotransmitters in subcortical pathways to the pontine nuclei. A combined immunocytochemicaland retrograde tracing study in
The a and fl spikes of the field potential evoked by LN stimulation appear to arise from the presynaptic impulses generated from two distinctive groups of axons (of. ref. 11). The a spikes appear to be generated by rapidly conducting axons and the fl spikes by slower axons. Since the putative GABAergic cells of LN are small m22, they are likely to have slower conduction velocities. Thus, one could propose that the a spike is generated by the axons of excitatory cells and the fl spike by inhibitory cells. If so, the total latency expected for an inhibitory postsynaptic response from the slower axons (2 ms for the presynaptic conduction and about 1 ms synaptic delay 11) is consistent with our suggestion that latencies less than 4 ms are due to monosynaptic inhibition. A recent retrograde double-labeling study 16 showed that cerebellopontine fibers from relatively small LN cells are collateral branches of cerebello-olivary axons, while those from intermediate to large LN cells are collaterals of axons projecting to superior colliculus and/or thalamus. As stated above, many of the small cells in LN are GABAergic 15 and the cerebello-olivary projection is, at least in part, GABAergic 23. On the other hand, cerebellothalamic and cerebellocollicular projections appear to be aspartatergic and/or glutamatergic s. Thus, it seems likely that the inhibitory projections to the pons from the LN are collateral branches of cerebello-olivary projections and the excitatory projections are collaterals of the rostral projections to thalamus and collieulus. It has been proposed that the circuitry connecting the pons and cerebellar nuclei serves to generate a tonic discharge in the cerebellar nuclear cells which may play a role in the signal carried by cerebellofugal messages 2s and may also be involved in motor learning ~7. The inhibitory cerebellopontine projection could also be part of a feedback pathway that assists in augmenting the excitability of BPN cells. Pontocerebellar neurons excite, via the granule cells, the Purkinje cells which, in turn, inhibit the cerebellar nuclear cells 12. The resulting disinhibition of pontocerebellar neurons could help to maintain a capability for transmission of high-frequency information from cerebral cortex to cerebellum2.
Acknowledgements. The authors which to thank Dr. Richard E. Poppele for his comments on the manuscript and Mr. S. Bentivegna for the figures.
the eat with some observations in the rat, Neuroscience, 34 (1990) 149-162. 2 Allen, G.I. and Tsukahara, N., CerebrocerebeUarcommunication systems, Physiol. Rev., 54 (1974) 957-1006. 3 Angaut, P., Cicirata, E and Pant6, M.R., An autoradiographic
184 study of the cerebello pontine projections from the interposed and lateral cerebellar nuclei in the rat, J. Hirnforsch., 26 (1985) 463-470. 4 Border, B.G., Kosinski, R.J., Azizi, S.A. and Mihailoff, G.A., Certain basilar pontine afferent systems are GABA-ergic: combined HRP and immunocytochemical studies in the rat, Brain Res. Bull., 17 (1986) 169-179. 5 Brodal, A., Destombes, J., Lacerda, A.M. and Angaut, P., A cerebellar projection onto the pontine nuclei. An experimental anatomical study in the cat, Exp. Brain Res., 16 (1972) 543558. 6 Cicirata, F., Pant6, M.R. and Angant, P., An autoradiographic study of the cerebello pontine projections in the rat. I. Projections from the medial cerebellar nucleus, Brain Research, 253 (1982) 303-308. 7 Coyle, J.T. and Schwarcz, R., Lesion of striatal neurones with kalnic acid provides a model for Huntington's chorea, Nature, 263 (1976) 244-246. 8 Fag,g, G.E. and Foster, A.C., Amino acid neurotransmitters and their pathways in the mammalian central nervous system, Neuroscience, 9 (1983) 701-719. 9 Ho, C.P. and Leong, S.K., A cerebellar projection onto the pontine nuclei in the albino rat, Exp. Brain Res., 30 (1977) 149-154. 10 Hubei, D.H., Single umt activity in lateral geniculate body and optic tract of unrestrained cats, J. Physiol., 150 (1960) 91-104. 11 Ito, M. and Yoshida, M., The ongm of cerebellar-induced inhibition of Deiters neurones. I. Monosynaptic initiation of the inhibitory postsynaptic potentials, Exp. Brain Res., 2 (1966) 330-349. 12 Ito, M., Yoshida, M. and Obata, K., Monosynaptic inhibition of intracerebellar nuclei induced from the cerebellar cortex, Experientia, 20 (1964) 575-576. 13 Iversen, L.L., Biochemical pharmacology of GABA. In M.A. Lipton, A. DiMascio and K.F. Killam K.E (Eds.,), Psychopharmacology - A Generation in Progress, Raven, New York, 1978, pp. 25-38. 14 Kital, S.T., Kocsis, J.D. and Kiyohara, T., Electrophysiological properties of nucleus reticularis tegmenti pontis cells: antidromic and synaptic activation, Exp. Brain Res., 24 (1976) 295309. 15 Kumoi, K., Saito, N., Kuno, T. and Tanaka, C., Immunohlstochemical localization of gamma-aminobutyric-acid- and aspartate-containing neurons in the rat deep cerebellar nuclei, Brain Research, 439 (1988) 302-310. 16 Lee, H.S., Kosinski, R.J. and Mlhalloff, G.A., Collateral branches of cerebeUopontme axons reach the thalamus, superior colliculus or inferior olive: a double-fluorescence and combined fluorescence-horseradish peroxidase study m the rat,
Neurosczence, 28 (1989) 725-734. 17 Llcata, E, Perciavalle, V., Urbano, A. and Viscuso, A., Learnmg and recalling of motor sequences: a theoretical model. In G. Hauske and E. Butenandt (Eds.), Kybernetik '77, Oldembourg, Munchen, 1977, pp. 373-375. 18 Li Volsi, G., Pacitti, C., Perciavalle, V., Sapienza, S. and Urbano, A., Interpositus nucleus influences on pyramidal tract neurons in the cat, Neuroscience, 7 (1982) 1929-1936. 19 McGeer, E.G. and McGeer, EL., Duplication of biochemical changes of Huntington's chorea by mtrastriatal injections of glutarmc and kainic acids, Nature, 263 (1976) 517-519 20 Martin, G.E, King, J.S. and Dora, R., The projection of the deep cerebellar nuclei of the opossum, Didelphis marsupialis virginiana, J. Hirnforch., 15 (1974) 545-573. 21 Mihailoff, G.A., McAxdle, G.B. and Adams, C.E., The cytoarchitecture, cytology, and synaptic organizaUon of the basilar pontine nuclei in the rat. I. Nissl and Golgi studies, J. Comp. Neurol., 195 (1981) 181-201. 22 Mugnaini, E. and Oertel, W.H., An atlas of the distributaon of GABAergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry. In A. Bjtrklund and T. Htkfeld (Eds.), Handbook of Chermcal Anatomy, Elsevier, Amsterdam, 1985, pp. 536-608. 23 Nelson, B., Barmack, N.H. and Mugnaini, E , A GABA-erglc cerebello-olivary projection m the rat, Soc. Neurosci. Abstr., 10 (1984) 539. 24 Paxmos, G. and Watson, C., The Rat Brain m Stereotaxie Coordinates, 2nd edn., Academic, Sydney, 1986. 25 Ram6n y Cajal, S., La double via descendente nacida del pedunculo cerebeloso superior, Trab. Lab. Invest. Biol. Unw Madrid, 2 (1903) 23-29. 26 Torigoe, Y., Blanks, R.H. and Precht, W., Anatomical studies on the nucleus reticularis tegmenti pontis in the pigmented rat. II. Subcortical afferents demonstrated by the retrograde transport of horseradish peroxidase, J Comp. Neurol., 243 (1986) 88-105. 27 Tsnkahara, N., Bando, T., Kitai, S.T. and Kiyohara, T., CerebeUo-pontine reverberating circuit, Brain Research, 33 (1971) 233-237. 28 Tsukahara, N. and Bando, T., Red nuclear and interpositonuclear excitation of pontine nuclear cells, Brain Research, 19 (1970) 295-298. 29 Watt, C.B. and Mihatloff, G.A., The cerebellopontme system in the rat. II. Electron microscopic studies, J. Comp. Neurol, 216 (1983) 429-437. 30 Yuen, H., Dom, R.M. and Martin, G.E, Cerebellopontine projections m the American opossum. A study of their origin, distribution and overlap with fibers from the cerebral cortex, J. Comp. Neurol, 154 (1974) 257-285.
178
BRES 17294
The cerebellopontine system: an electrophysiological study in the rat Sabina Berretta, Gianfranco Bosco, Giuseppe Smecca, and Vincenzo Perciavalle Instuute of Human Physiology, Umverstty of Catama, Catama (Italy) (Accepted 13 August 1991)
Key words: Basilar pontme nucleus; Cerebellar lateral nucleus; 7-Aminobutyric acid; Reticulotegmental nucleus; Rat
We examined the effects of electric stmaulation of the cerebellar lateral nucleus (LN) in the rat on the activity of single pontocerebellar neurons in the basilar pontme nuclei (BPN) and the reticulotegmental nucleus (RtTg). We found that about half of the cells of these nuclei that were influenced by LN stimulation were inhibited. A significant fraction of both excitatory and mhibRory responses had latencies of less than 4 ms and were able to follow high frequency stimulation, compatible with a monosynaptic linkage. ExtraceUular field potential recordings within the BPN and RtTg were interpreted as arising from impulses propagating along inhibitory axons projecting in a bundle from the cerebellum to these pontine structures. Microionophoretlc administration of GABA antagonists bicuculline or pmrotoxin abolished or attenuated most inhibitory effects. Therefore, we conclude that LN-mduced inhibition is most likely mediated by cerebellopontine GABAerglc fibers. The functional significance of this cerebellopontine inhibitory circuit is discussed. INTRODUCTION It is well k n o w n 24 that the s u p e r i o r c e r e b e l l a r p e d u n cle contains d e s c e n d i n g fibers that r e a c h t w o i m p o r t a n t p r e c e r e b e l l a r structures, t h e basilar p o n t i n e nuclei ( B P N ) and the r e t i c u l o t e g m e n t a l nucleus ( R t T g ) .
In several
species t h e s e d e s c e n d i n g p r o j e c t i o n s o r i g i n a t e m a i n l y f r o m the c e r e b e l l a r lateral nucleus ( L N ) 1'3'5'6'9'20'26'30, with a s m a l l e r c o n t r i b u t i o n f r o m interpositus nucleus and v e r y few fibers f r o m the m e d i a l nucleus. F e w studies h a v e focused o n the functional p r o p e r t i e s of t h e s e p r o jections. E a r l y e l e c t r o p h y s i o l o g i c a l e x p e r i m e n t s indic a t e d that the c e r e b e l l o p o n t i n e system is e x c i t a t o r y 14"28, and s u b s e q u e n t ultrastructural studies r e p o r t e d that cerebellopontine terminal boutons contain mostly round synaptic vescicles 29. H o w e v e r , it was r e c e n t l y o b s e r v e d with d o u b l e - l a b e l i n g t e c h n i q u e s that s o m e c e r e b e l l o p o n tine n e u r o n s are i m m u n o r e a c t i v e for g l u t a m a t e - d e c a r b o x y l a t e 4 as well as y - a m i n o b u t y r i c acid ( G A B A ) 1. This suggests that at least a p a r t of the c e r e b e l l o p o n t i n e syst e m m a y be inhibitory. T h e p r e s e n t study was u n d e r t a k e n to r e - e x a m i n e the effects o f electric s t i m u l a t i o n of the L N o n the spontan e o u s activity o f single n e u r o n s of the B P N as well as the R t T g in the rat. MATERIALS AND METHODS Extracelhilar responses of single pontme cells were recorded on
24 male Sprague-Dawley rats (230-280 g) anesthetized with urethane (1.2 g/kg l.p.). Recordings were made using glass microplpettes (7-18 Mf~ resistance) filled with a 4% solution of Pontamine sky blue in 1 M KC1. The microelectrodes were positioned within the BPN and RtTg according to the stereotaxic coordinates of the atlas of Paxinos and Watson 24. A screw posmoned in the skull served as indifferent electrode. Action potentials were regarded as ansmg from the cell somata when they appeared initially negativepositive and then, as the electrode advanced, positive-negative with a clear inflection on the ascending phase 1°. The end of each electrode penetration was marked with a small (50-100/zm) iontophoretic injection of pontamine (10/~A for 20 mm) for subsequent histological identification. A monopolar stimulating electrode (0.6-1 2 Mf~ resistance nickel-chrome wire) was stereotaxically placed in the contralateral LN. Stimuli were 1-5 cathodal shocks (0.05-0.3 ms, 100-700 Hz) delivered at random intervals from 1.5 to 5 s For two experiments in which field potentials were analyzed, a concentric electrode having a shaft diameter of 150/~m and a central stainless steel w~re was used to stimulate the LN. The electrode was insulated with varmsh (Epoxylite) except for an exposed tip of about 200/zm. The cathode was formed by e~ther external or internal pole. In some experiments, 5-barreled glass mlcropipettes were used for recording and ionophoretic application of drugs. The central barrel (impedance 4-8 MQ) contained a 4% solution of Pontamine sky blue m 1 M KC1. An outer barrel, filled with NaC1 3 M, served as balance channel, while the remammg barrels were filled with the following drugs: GABA (Sigma: 0.4 M, pH 4.0); bicuculhne (Sigma; 20 mM in NaC1 150 mM, pH 3.0); picrotoxm (Sigma; 10 mM in NaC1 150 mM, pH 3.8) Three constant current sources and an automatic balance unit (Neurophore) were used for drug ejection Anodal currents of up to 150 nA were used to lonophorese the substances, while a cathodal current of 10-15 nA was used for retaining them during the penetration After the recording sessions, the s~te of stimulation was marked with an electrolytic reference les~on (20/~A cathodal current for 30 s) and the animals were perfused with a 4% paraformaldehyde solution while under deep urethane (2 g/kg i.p.) anesthesia. Recon-
Correspondence: V. Percaavalle, Istituto di Flslologla Umana, Universlta" di Catanla, Vlale Andrea Dona 6, 95125 Catama, Italy. Fax: (39) (95) 330645.
179 BPN
RESULTS
RtTg
2O"
20-6122 SD .4001 n -41
16-
~2"
1612"
8-
X -5143 SO -4 036 n .14
8" ~
4-
EXCITATIONS .
0
o
E
I
[]
4"
INHIBITIONS
812'
12-
~ . 7 722 SD - 5 ~ n .54
16. 20-
~-5556 SD.3203 n.18
24
i
I
I
I
I
I
I
I
I
I
I
I d ~
I I I I I I I I I I I I 0 2 4 6 8 10121416162022
0 2 4 6 8 101214161820223032
L A T E N C Y (ms)
Fig. 1. Histograms showing latency of excitation (upper) and inhibition (lower) displayed by neurons belonging to the BPN and to the RtTg following stimulation of the contralateral cerebellar LN.
structlons of the stimulation site and electrode penetrations were made from frozen sections of 30 gm thickness, stained with Neutral red. To exclude the possibility that the effects of LN stimulation were due to activation of afferent terminals and/or of fibers of passage, the LN neurons in 6 rats were chemically ablated prior to testing. An injection of i gg of kainic acid (Sigma) dissolved in I #1 of 0.05 M buffered phosphate solution (pH 7.4), was made unilaterally into the LN area 15 days prior to the electrophysiological experiment. The evaluation of unitary discharges was performed by converting 60-90 responses into post-stimulus time histograms (PSTHs) and cumulative frequency distributions (CFDs). Excitatory and inhibitory responses were sequences of at least 3 bins (bin width 0.5-2 ms), which showed frequency values of more than twice the standard deviation (S.D.) above or below the mean value during spontaneous activity. Solitary bins between these sequences were not considered. Latency was the time interval between the last shock of the stimulating train and the first bin of the sequence, and the duration was the distance between first and last bin TM.
Unit activity was recorded from more than 600 pontine cells, 261 of which were located within the B P N (193) and RtTg (68). A b o u t half of the cells in these two nuclei were influenced by stimulation of the contralateral LN, 95/193 (49.2%) in the B P N and 32/68 (47.1%) in the RtTg. In the BPN, the prevailing LN-induced effect was inhibitory. As illustrated by the latency distributions in the left part of Fig. 1, inhibitory responses having latencies ranging from 2 to 32 ms (mean value 7.72 ms -+ 5.50 S.D.) were recorded in 54 of the 95 B P N ceils (56.8%) while excitatory responses were recorded in the remainhag 41 cells (43.2%). The excitatory responses had latencies ranging from 2 to 18 ms (mean value 6.12 ms -+ 4.01 S.D.). A significant fraction of the excitation (18/41; 43.9%) and inhibition (9/54; 16.7%) had latencies less than 4 ms. All of these responses were elicited with a single shock and followed stimulation frequencies up to 70-80 Hz. Increasing the pulse number in the stimulation train did not significantly change the latency of these responses. The effects elicited by L N stimulation on two neurons located within the medial part of the B P N are illustrated in Fig. 2. It can be seen that the first B P N neuron was excited and the second inhibited by stimulation of contralateral LN with a single 0.2 ms cathodal pulse (80
~A). A similar preponderance of inhibitory responses to L N stimulation was observed in the RtTg (Fig. 1, right). Although the sample size was smaller, similar fractions of cells displayed inhibitory (18/32 cells; 56.3%) and excitatory (14/32; 43.7%) responses. Response latencies were slightly shorter on average and slightly larger frac-
I II II
-I
i
II
III
LJIIItllllllllJIIII
-I
] III ! IIIIII ~l I11t11~
,Om.JO: Fig. 2. Evoked responses displayed by two neurons located in the medial part of the BPN following stimulation (pulse duration 0.2 ms; pulse intensity 80 #A) of the contralateral LN. The records are formed by 10 superimposed traces. Calibration bar in the camera lucida drawing: 1 ram.
180 tions had latencies shorter than 4 ms (8/14; 57.1% for excitatory and 6/18; 33.3% for inhibitory responses). The responses were also capable of following stimulation frequencies up to 70-80 Hz. Examples of excitatory and inhibitory effects on two RtTg cells evoked by stimulation of the LN with a single 0.2 ms cathodal pulse (90 #A) is illustrated by the PSTHs and CFDs of Fig. 3. It is worth noting that in these cells as well as in the remaining cells examined in this study the frequency of the prestimulus neuronal activity was similar to that in the late part of PSTHs and CFDs. The location of cells recorded within the BPN and RtTg are illustrated in Fig. 4. Cells that were inhibited by LN stimulation are marked by filled triangles and those that were excited by open triangles. Cells that did not respond to LN stimulation are marked by dots. It can be seen that, although inhibited and excited cells are present throughout the BPN and RtTg area, inhibited cells appear more concentrated in the anterior part and the excited cells in the more posterior part of the BPN. The results reported above show that LN stimulation elicits short latency responses in BPN and RtTg neurons that could result from monosynaptic excitation or inhibition. In order to rule out possible contributions from the activation of LN afferent terminals or fibers of passage, the cells of the LN were chemically ablated using kainic acid in 6 rats 15 days before the electrophysiological recording. In 3 of the rats, histological examination showed more than 70% fewer cells in the contralateral
LN compared to the ipsilateral control. In these rats, stimulation of contralateral LN area was tested on 13 BPN and 9 RtTg neurons. Only two BPN neurons displayed discharge changes that were excitatory in nature and appeared with a latency of 1.8 and 2.2 ms, respectively. Short latency inhibitory effects were not observed. Further evidence for a monosynaptic activation from LN was obtained from field potential recordings. Fig. 5 shows the results of a systematic exploration of the field potentials induced in the BPN and RtTg areas by LN stimulation. In the region labeled 1 in Fig. 5, just dorsal to the two nuclei, the field potentials consist of a small negative deflection followed by a relatively small positivity and a second small negative deflection. As the recording electrode was shifted ventrally across the RtTg and BPN, two peaks, called a and fl spikes, became more evident. The a spike started as early as 1.05 ms and attained its peak at 1.60 ms after stimulation of the LN, while the fl spike began at 2.04 ms with summit at 2.24 ms. The deflection equivalent to the a spike was greater in the rostral tract than in the caudal tract, while in both tracts there were less prominent delayed negative deflections, which apparently belonged to the fl spike. These spikes could be discriminated by varying the stimulus intensity. The a spike appeared at a relatively low stimulation intensity and attained its maximum before the fl spike was produced by a further increase in stimulus strength. We tested directly whether G A B A is involved in the 45
. . . .
• .....
~....+,,I
I
....
i. . . . .
.
.+
i
.
o9
"~. 180
~ E
5 ms/d=v
18
e-
, .i+,m ,,,,Li,+,I.,+,,J+.JIJ+, ..... +,i, i
18o
I
10 ms/dtv
Fig. 3. PSTHs a n d CFDs illustrating the response patterns displayed by two RtTg neurons following stimulauon (pulse duration 0.2 ms; pulse intensity 90 #A) of the contralateral LN (same animal as Fig. 2). Histograms were compiled from 70 sweeps. Calibration bar m the camera lumda drawing: 1 ram.
181
L01
Rt'r~_ _
B
A
D
D •
•
z
•
•
•
, •
~,•
BPN
•
• &
~
•
~AA °
z
.
.
a
.
•
A
.
z
.
z
.
•
~
•
At=/
, V
.
RtTg
-'.',~':'::-.L0,4 "j~, a , ,A.z= a"
1 mm t
A
I
B +
1
LO.9
2
F-
3
L14
1 m6
!
!
1 mm Fig. 4. Four serial saglttal sections (redrawn from Paxinos and Watson~) summarizing the location within the BPN and RtTg of cells inhibited (filled triangles), excited (open triangles) or non-influenced by LN stimulation. A, anterior; D, dorsal; P, posterior; V, ventral.
inhibitory responses of these neurons by local iontophoretic injection of G A B A and the G A B A antagonists bicuculline and picrotoxin at the recording sites in 6 rats. Recordings were obtained from 46 neurons (32 belonging to the BPN and 14 to the RtTg). Spontaneous firing of all neurons was abolished or greatly attenuated within 1-3 s following the application of 30-40 nA injection of G A B A . The depressant effects of G A B A were reversed in 96% of cases (44/46) by an iontophoretic application (70-80 nA) of bicucuUine, which alone increased the spontaneous firing of 2/3 of the tested neurons by 1540%, suggesting a disinhibition of a tonic G A B A input to these neurons. The depressant effect of G A B A was also abolished in 87% of cases (40/46) by application of picrotoxin which, however, did not significantly increase the spontaneous discharge of tested neurons. The sensitivity to these drugs was also tested on 11 neurons which displayed an LN-induced inhibition. Of these cells, bicuculline administration abolished or greatly attenuated
Fig. 5. Field potentials induced by LN stimulation (pulse duration 0.1 ms; pulse intensity 250/zA) recorded above and in the RtTg and BPN during two electrode penetrations. Electrode tracks and location of the recording sites are illustrated on the saglttal section of brainstem at the top of the figure. The records are formed by 10 superimposed traces. In the lower record of track A, a and fl denote the two peaks of the field potential.
the post-stimulus inhibition in 7 (63.6%). In addition, 5 cells were studied for a sufficiently long interval to observe a return of the post-stimulus inhibitory effect after cessation of bicuculline administration. The time interval required for the return of the LN-induced inhibition varied from 2 up to 11 min. Moreover, the same 11 neurons were tested for the effects of picrotoxin; it was observed that in 6/11 cells (54.5%) picrotoxin also abolished the inhibitory effect elicited by LN stimulation. An example of the effects elicited by bicuculline and picrotoxin on the LN-induced inhibitions displayed by a BPN and a RtTg neuron is illustrated in Fig. 6. DISCUSSION The present study demonstrates that about half of the neurons in the BPN and RtTg are significantly affected by LN stimulation, and more than 50% of those cells were inhibited. Many of the responses, both excitatory and inhibitory, were compatible with a monosynaptic
182
181
BPN
18
before I
,L,
j,==,
Rt Tg before
,I,L= ,,~ ,hjL-~II ~ ,~l,,,,, I~,IIL,,,, ,,J,I, I
I
-,,._..
---__
300
180
181
~
blcuculline (60 nA)
18 I.
I ,1~,.,~.......... h~, J..... ~.~,, ,L.~ l .......
-'~.
300 18
I
plcrotoxm (80 hA)
.......
1
11 mm after
!
,
18
I
•
300
5 mm after
,~ ~J~,,, i,, k,,,~iH,, ~a, ~,,,,,I, ,,,,J,~,l
,
180
i
,
10ms/d=v
Fig. 6. PSTHs and CFDs illustrating the effect of the iontophoretic application of the GABA-antagonists bicuculline (5 mM in NaCI 150 raM, pH 3.0) and picrotoxin (10 raM in NaCI 150 raM, pH 3.8) on the inhibition displayed by a BPN and a RtTg neuron, respectively. Inhibitory responses were elicited by stimulating the LN with a single cathodal 0.2-ms pulse (intensity 110 and 90/~A, respectively). Note in both cases that, during delivering of the drug, the LN stimulation did not evoke a significant discharge change, while the LN-induced inhibition reappeared 11 and 5 rain after the cessation of administration, respectively. Histograms were compiled from 70 sweeps and have bin-width of 1 ms.
linkage; they had short latencies (< 4 ms) and were capable of following high frequency stimulation. Longer latency responses were probably due to polysynaptic linkages, e.g. via the cerebcllo-thalamo-cortico-pontine pathway. These results suggest that inhibitory LN neurons synapse directly on pontine neurons. It was still possible, though, that the inhibitory effects were not directly mediated by LN neurons but rather by fibers of passage in the LN or by cells having afferent terminals within the LN. However, after at least 70% of the cells in the LN were destroyed by kainic acid, leaving fibers of passage
or afferents intact7'19, stimulation of the LN no longer evoked short latency responses in the pontine nuclei. We conclude, therefore, that the activation of the LN cells is responsible for the short latency inhibitory responses observed in the BPN and RtTg. The majority of the BPN and RtTg cells displayed a significant decrease of their spontaneous discharge upon microiontophoretic application of GABA and LN-induced inhibition was abolished or greatly reduced following application of the classic GABA antagonists bicuculline and picrotoxin (cf. ref. 13). Therefore, a significant percentage of cerebellopontine projections are
183 inhibitory in nature and the majority of them appear to be GABA-mediated. Monosynaptic inhibition of cells in the BPN or RtTg from the cerebellar nuclei has not been described before. However, some cerebellopontine neurons have been shown to be glutamate-decarboxylase positive4 or GABA-Iike immunoreactive ~, and several small cell bodies containing G A B A have been found scattered within the deep cerebellar nuclei, with the highest incidence in the LN m22. In the eat, about 10% of cerebellar nuclear cells projecting to the BPN and RtTg are GABA-Iike immunoreactive 1. In the rat, we now show that about 10% of the cells in the BPN (9/95) that were influenced by LN stimulation were inhibited at short latency. In the RtTg the fraction was closer to 20% (6/32). Our data also imply that GABAergic neurons in the LN exert an inhibitory influence on neurons in the pons. Cells in the BPN and RtTg were found to react to local iontophoretic injection of G A B A which significantly reduced their spontaneous discharge. Moreover, LN-induced inhibition was abolished or greatly reduced following application of the classic G A B A antagonists, bieueulline and picrotoxin (el. ref. 13). Nevertheless, a significant fraction (more than a third) of the short latency inhibitory responses were not affected by these antagonists. Since no other inhibitory transmitter has been thus far implicated immunohistochemically, and there are only few intrinsic neurons within the BPN 21, we suggest that some of the non-GABAergic inhibitory responses may be due to presynaptic inhibition. That is, they may have resulted from an excitatory action of cerebellopontine projections upon fiber terminals (corticopontine) exciting PBN or RtTg neurons. Thus we would estimate that about two-thirds of the short latency inhibitory responses in these pontine nuclei are the result of postsynaptic inhibition mediated monosynaptically by GABAergic neurons in the LN and the remaining onethird could be dependent on inhibitory synapses far from the electrode tip and unreached by the applied antagonists, or they are the result of presynaptic inhibition mediated by excitatory neurons. We would suggest, therefore, that there is a dual control exerted by the cerebellar nuclear cells on BPN and RtTg, an inhibitory control providing negative feedback and an excitatory control providing both positive and negative feedback. The results of the field potential recordings also support the suggestion of a dual innervation by LN neurons.
REFERENCES 1 Aas, ,I.E. and Brodal, P., GABA and glycine as putative neurotransmitters in subcortical pathways to the pontine nuclei. A combined immunocytochemicaland retrograde tracing study in
The a and fl spikes of the field potential evoked by LN stimulation appear to arise from the presynaptic impulses generated from two distinctive groups of axons (of. ref. 11). The a spikes appear to be generated by rapidly conducting axons and the fl spikes by slower axons. Since the putative GABAergic cells of LN are small m22, they are likely to have slower conduction velocities. Thus, one could propose that the a spike is generated by the axons of excitatory cells and the fl spike by inhibitory cells. If so, the total latency expected for an inhibitory postsynaptic response from the slower axons (2 ms for the presynaptic conduction and about 1 ms synaptic delay 11) is consistent with our suggestion that latencies less than 4 ms are due to monosynaptic inhibition. A recent retrograde double-labeling study 16 showed that cerebellopontine fibers from relatively small LN cells are collateral branches of cerebello-olivary axons, while those from intermediate to large LN cells are collaterals of axons projecting to superior colliculus and/or thalamus. As stated above, many of the small cells in LN are GABAergic 15 and the cerebello-olivary projection is, at least in part, GABAergic 23. On the other hand, cerebellothalamic and cerebellocollicular projections appear to be aspartatergic and/or glutamatergic s. Thus, it seems likely that the inhibitory projections to the pons from the LN are collateral branches of cerebello-olivary projections and the excitatory projections are collaterals of the rostral projections to thalamus and collieulus. It has been proposed that the circuitry connecting the pons and cerebellar nuclei serves to generate a tonic discharge in the cerebellar nuclear cells which may play a role in the signal carried by cerebellofugal messages 2s and may also be involved in motor learning ~7. The inhibitory cerebellopontine projection could also be part of a feedback pathway that assists in augmenting the excitability of BPN cells. Pontocerebellar neurons excite, via the granule cells, the Purkinje cells which, in turn, inhibit the cerebellar nuclear cells 12. The resulting disinhibition of pontocerebellar neurons could help to maintain a capability for transmission of high-frequency information from cerebral cortex to cerebellum2.
Acknowledgements. The authors which to thank Dr. Richard E. Poppele for his comments on the manuscript and Mr. S. Bentivegna for the figures.
the eat with some observations in the rat, Neuroscience, 34 (1990) 149-162. 2 Allen, G.I. and Tsukahara, N., CerebrocerebeUarcommunication systems, Physiol. Rev., 54 (1974) 957-1006. 3 Angaut, P., Cicirata, E and Pant6, M.R., An autoradiographic
184 study of the cerebello pontine projections from the interposed and lateral cerebellar nuclei in the rat, J. Hirnforsch., 26 (1985) 463-470. 4 Border, B.G., Kosinski, R.J., Azizi, S.A. and Mihailoff, G.A., Certain basilar pontine afferent systems are GABA-ergic: combined HRP and immunocytochemical studies in the rat, Brain Res. Bull., 17 (1986) 169-179. 5 Brodal, A., Destombes, J., Lacerda, A.M. and Angaut, P., A cerebellar projection onto the pontine nuclei. An experimental anatomical study in the cat, Exp. Brain Res., 16 (1972) 543558. 6 Cicirata, F., Pant6, M.R. and Angant, P., An autoradiographic study of the cerebello pontine projections in the rat. I. Projections from the medial cerebellar nucleus, Brain Research, 253 (1982) 303-308. 7 Coyle, J.T. and Schwarcz, R., Lesion of striatal neurones with kalnic acid provides a model for Huntington's chorea, Nature, 263 (1976) 244-246. 8 Fag,g, G.E. and Foster, A.C., Amino acid neurotransmitters and their pathways in the mammalian central nervous system, Neuroscience, 9 (1983) 701-719. 9 Ho, C.P. and Leong, S.K., A cerebellar projection onto the pontine nuclei in the albino rat, Exp. Brain Res., 30 (1977) 149-154. 10 Hubei, D.H., Single umt activity in lateral geniculate body and optic tract of unrestrained cats, J. Physiol., 150 (1960) 91-104. 11 Ito, M. and Yoshida, M., The ongm of cerebellar-induced inhibition of Deiters neurones. I. Monosynaptic initiation of the inhibitory postsynaptic potentials, Exp. Brain Res., 2 (1966) 330-349. 12 Ito, M., Yoshida, M. and Obata, K., Monosynaptic inhibition of intracerebellar nuclei induced from the cerebellar cortex, Experientia, 20 (1964) 575-576. 13 Iversen, L.L., Biochemical pharmacology of GABA. In M.A. Lipton, A. DiMascio and K.F. Killam K.E (Eds.,), Psychopharmacology - A Generation in Progress, Raven, New York, 1978, pp. 25-38. 14 Kital, S.T., Kocsis, J.D. and Kiyohara, T., Electrophysiological properties of nucleus reticularis tegmenti pontis cells: antidromic and synaptic activation, Exp. Brain Res., 24 (1976) 295309. 15 Kumoi, K., Saito, N., Kuno, T. and Tanaka, C., Immunohlstochemical localization of gamma-aminobutyric-acid- and aspartate-containing neurons in the rat deep cerebellar nuclei, Brain Research, 439 (1988) 302-310. 16 Lee, H.S., Kosinski, R.J. and Mlhalloff, G.A., Collateral branches of cerebeUopontme axons reach the thalamus, superior colliculus or inferior olive: a double-fluorescence and combined fluorescence-horseradish peroxidase study m the rat,
Neurosczence, 28 (1989) 725-734. 17 Llcata, E, Perciavalle, V., Urbano, A. and Viscuso, A., Learnmg and recalling of motor sequences: a theoretical model. In G. Hauske and E. Butenandt (Eds.), Kybernetik '77, Oldembourg, Munchen, 1977, pp. 373-375. 18 Li Volsi, G., Pacitti, C., Perciavalle, V., Sapienza, S. and Urbano, A., Interpositus nucleus influences on pyramidal tract neurons in the cat, Neuroscience, 7 (1982) 1929-1936. 19 McGeer, E.G. and McGeer, EL., Duplication of biochemical changes of Huntington's chorea by mtrastriatal injections of glutarmc and kainic acids, Nature, 263 (1976) 517-519 20 Martin, G.E, King, J.S. and Dora, R., The projection of the deep cerebellar nuclei of the opossum, Didelphis marsupialis virginiana, J. Hirnforch., 15 (1974) 545-573. 21 Mihailoff, G.A., McAxdle, G.B. and Adams, C.E., The cytoarchitecture, cytology, and synaptic organizaUon of the basilar pontine nuclei in the rat. I. Nissl and Golgi studies, J. Comp. Neurol., 195 (1981) 181-201. 22 Mugnaini, E. and Oertel, W.H., An atlas of the distributaon of GABAergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry. In A. Bjtrklund and T. Htkfeld (Eds.), Handbook of Chermcal Anatomy, Elsevier, Amsterdam, 1985, pp. 536-608. 23 Nelson, B., Barmack, N.H. and Mugnaini, E , A GABA-erglc cerebello-olivary projection m the rat, Soc. Neurosci. Abstr., 10 (1984) 539. 24 Paxmos, G. and Watson, C., The Rat Brain m Stereotaxie Coordinates, 2nd edn., Academic, Sydney, 1986. 25 Ram6n y Cajal, S., La double via descendente nacida del pedunculo cerebeloso superior, Trab. Lab. Invest. Biol. Unw Madrid, 2 (1903) 23-29. 26 Torigoe, Y., Blanks, R.H. and Precht, W., Anatomical studies on the nucleus reticularis tegmenti pontis in the pigmented rat. II. Subcortical afferents demonstrated by the retrograde transport of horseradish peroxidase, J Comp. Neurol., 243 (1986) 88-105. 27 Tsnkahara, N., Bando, T., Kitai, S.T. and Kiyohara, T., CerebeUo-pontine reverberating circuit, Brain Research, 33 (1971) 233-237. 28 Tsukahara, N. and Bando, T., Red nuclear and interpositonuclear excitation of pontine nuclear cells, Brain Research, 19 (1970) 295-298. 29 Watt, C.B. and Mihatloff, G.A., The cerebellopontme system in the rat. II. Electron microscopic studies, J. Comp. Neurol, 216 (1983) 429-437. 30 Yuen, H., Dom, R.M. and Martin, G.E, Cerebellopontine projections m the American opossum. A study of their origin, distribution and overlap with fibers from the cerebral cortex, J. Comp. Neurol, 154 (1974) 257-285.
