- Email: [email protected]
The entire trajectory of single climbing and mossy fibers in the cerebellar nuclei and cortex
N.M. Gerrits, T.J.H. Ruigrok and C.I. De Zeeuw (Eds.) Progressin BrainResearch,Vol 124 © 2000 Elsevier Science BV. All rights reserved.
CHAPTER
12
The entire trajectory of single climbing and mossy fibers in the cerebellar nuclei and cortex Y. Shinoda*, I. Sugihara, H.-S. Wu and Y. Sugiuchi Department of Systems Neurophysiology, School of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo, 113-8519, Japan
Introduction
Afferents of the cerebellum from many different origins, including brain stem precerebellar nuclei except the inferior olive (IO), take the form of mossy fibers in the cerebellar cortex (Cx), while olivocerebellar (OC) axons take the form of climbing fibers (CFs). Since the discovery of CFs in the cerebellar cortex (Ram6n y Cajal, 1911), the morphology of these fibers has been extensively studied by Golgi staining, electron-microscopy, and recently anterograde labeling. Despite abundant anatomical and physiological studies on the OC system, there is little information available on the organization of OC projection at the level of single axons. For example, the exact number of CFs that a single OC axon gives rise to is not known, although the average number of CFs per OC axon (about seven in the rat) was inferred by counting the total numbers of PCs and IO neurons (Schild, 1970). Electrophysiological studies (Armstrong et al., 1973; Ekerot and Larson, 1982) and retrograde labeling studies (Brodal et al., 1980; Rosina and Provini, 1983) have shown that single OC axons project to separate sites in the Cx, but the general extent of the spread of single OC axons in the cerebellar cortex remains unknown. Due to uniformity of the local circuit of the Cx, the *Corresponding author. Tel.: 81-3-5803-5153; Fax: 81-3-5803-5155; e-mail: [email protected]
projection pattern of afferents to the Cx including the mossy fiber system is significant in determining functional compartmentation and organized action of the cerebellum (Eccles et al., 1967; Palay and Chan-Palay, 1974; Ito, 1984). Recent studies of mapping receptive fields through the olivocerebellar climbing fiber projection have suggested that the Cx is composed of many longitudinal functional compartments (Andersson and Oscarsson, 1978; Ito et al, 1982; Ito, 1984). This electrophysiologically defined organization is consistent with the anatomically defined longitudinal compartments inthe Cx (Groenewegen and Voogd, 1977; Voogd et al., 1996). Some mossy fiber projections also seem to show longitudinal organization in the vestibulofloccular system (Sato et al., 1983) and the spinocerebellar system (Oscarsson, 1976). Whether or not there is a common principle in the cortical organization of the projection of mossy fibers conveying somatosensory information is still elusive. Besides the projection to the Cx, climbing and mossy fiber projections to the deep cerebellar nucleus (DCN) have not been fully investigated. The DCN gives rise to cerebellar output to target structures outside the cerebellum, and efferent neurons in the DCN receive inhibitory input from cortical Purkinje cells (Eccles et al., 1967; Ito, 1984). Therefore, the existence or absence of excitatory inputs to the DCN from the brainstem must be important for our understanding of the
174
cerebellar function (Shinoda et al., 1993, 1997). It has been generally assumed that excitatory inputs to the DCN are supplied through collaterals of mossy and climbing fiber afferents to the Cx. In spite of a wealth of anatomical reports on the afferent pathways to the Cx, there have been far fewer studies on afferent projections to the DCN until recently. The existence of the projections of mossy fibers to the DCN had been controversial (Chan-Palay, 1977; Dietriches et al., 1983; Ito, 1984), and reliable anatomical data on the projections from the precerebellar nuclei to the DCN had not been available until recently. (see Dietrichs et al. (1983) for the details of the problems). Using modem reliable anatomical staining methods, the projections from the pontine nucleus, nucleus reticularis tegmenti pontis, IO and spinal cord to the DCN have been definitely confirmed (Gerritis and Voogd, 1987; Van der Want et al., 1989; Shinoda et al., 1992; Mihailoff, 1993; Matsushita and Yaginuma, 1995). Intra-axonal staining with horseradish peroxidase of electrophysiologically identified neurons has been successfully used by our group to analyze the trajectories of single axons of various neurons (Futami et al., 1979; Shinoda et al., 1981, 1986, 1992). With this method, we have revealed the detailed morphology of single mossy fibers in the Cx (Krieger et al., 1985) and the existence of collateral projection of mossy fibers of the pontine nucleus and the nucleus reticularis tegmenti pontis to the DCN (Shinoda et al., t992, 1997). In the pontine nucleus and the nucleus reticularis tegmenti pontis, only a portion of mossy fibers projecting to the Cx have axon collaterals to the DCN, and many mossy fibers project only to the Cx. However, a question still remains as to whether there exist fibers that only project to the DCN without projecting to the Cx. In addition, it is essential for understanding the neural mechanism of the cerebellar function to determine the exact percentage of fibers that have axon collaterals to the DCN in different cerebellar afferent fiber systems. The purpose of the present study was to analyze the entire trajectories of single OC axons and single axons of the lateral reticular nucleus (LRN) that had been labeled by the injection of biotinylated dextran amine into the IO, and the LRN in the rat, respectively. An anterograde tracer, biotinylated
dextran amine dissolved in physiological saline to give a concentration of 10-15% was injected either with pressure (0.001-0.05 Ixl) or electrophoresis (2 IxA positive current pulses of 1 sec duration at 0.5 Hz for 5-20 min to the IO or the LRN). By reconstructing entire axons on serial sections, we examined the detailed morphological characteristics of single OC axons and mossy fibers of the LRN (Wu et al., 1999; Sugihara et al., 1999).
Morphology of single olivocerebellar axons The trajectory of 16 completely reconstructed OC axons and 46 partially reconstructed axons terminating in the vermis, intermediate area, hemisphere and flocculus were used for the present analysis. Axons left an injection site in the IO toward the contralateral side, crossed the midline, ran transversely above or through the contralateral IO, and entered the white matter of the inferior cerebellar peduncle (ICP) (Fig. 1). Then they ran longitudinally through the dorsolateral ICP to enter the cerebellum. After entering the cerebellum, stem axons frequently ramified into many branches in the deep cerebellar white matter; these ramifications often occurred when the stem axons reached the semiparasagittal plane in which branches terminated as CFs. Stem axons and branches usually ran in the white matter rostral and dorsal to the cerebellar nuclei, but sometimes they ran through or beneath the cerebellar nuclei. Branches entered the folial white matter, ramified and ran along each other in the semi-parasagittal plane to reach the areas for their termination in the cerebellar cortex, in which they further ramified. The branches of OC axons were grouped into thick branches (0.7-1.4 Ixm in diameter) and thin collaterals (0.2-0.5 ixm in diameter). A single OC axon generated 2-17 final thick branches (6.1 _ 3.7, mean _+S.D., n = 16), each of which innervated a PC as a CE We observed thin retrograde collaterals (Scheibel and Scheibel, 1954) originating from a CF terminal arborization and reentering the PC and granular layers, and some swellings of the retrograde collaterals in the PC layer apparently touched a PC which was not the target of the CF terminal arborization.
175
n Z
o.~= \
Z
\
\ \\
E E
c~
(..)
/
/
/
-J
.....
\
~,.;>
/
¢)
© .-
J '
"
:
~i'i ¸
•. . . .
0
.
•
, ~.U .-~ ~ ,.
.~.~ v/
/
.
....
:,
E
~
.~
-~
_~~ •
"~
~
176
(1) Climbing fibers Most of the reconstructed OC axons terminated on the side contralateral to the IO from which they originated (Fig. 1B). The most distal portion of a thick branch of an OC axon, which was equivalent to a CF, reached a target PC usually at its soma. The number of swellings on the terminal arborization of a single CF was 154-321 (249.8_+42.2, n = 3 2 terminal arborizations). The diameter of swellings ranged from 0.5 to 2.0 p.m, mostly 1.2-1.6 p.m. Assuming 6.1 CFs per OC axon, 249.8 swellings per CF corresponded to 1524 swellings on CF terminal arborizations per OC axon, on average. To examine whether a terminal arborization on a single PC is always formed by a single thick branch of an OC axon, we observed 523 CFs in three experiments. A terminal arborization always originated from a single thick branch of an OC axon. Spatial distribution of climbing fibers originating from a single olivocerebellar axon had a characteristic feature as expected from the anatomical analysis of the olivocerebellar mass projection. All CFs of a single olivocerebellar axon terminated in a single strip-shaped area in the cerebellar cortex. This area was narrow in the medio-lateral direction (usually about 200 p.m) and wide in the rostrocaudal direction, covering single or multiple lobules (Fig. 2). Strip shaped termination areas of
Via Vlb
~~J//ll ' " ~ ~~~ VII 1ram½
~
Fig. 2. Lateral view of the entire trajectory of a single OC axon in the cerebellum. The mediolateal extent of this axon in the Cx was less than 300 ~ m (unpublished data).
single axons originating from olivary neurons located in adjacency in the inferior olive tended to have significant overlap in the cerebellar cortex. These projection patterns of olivocerebellar axons stood in sharp contrast to those of LRN neurons described in the next section. (2) Thin collaterals of OC axons Thin collaterals, which were given off from stem axons and thick branches of OC axons before they formed CF terminal arborizations, were more abundant in terms of the number per OC axon than thick branches. For convenience of description, they were classified into three types according to their main termination sites: (1) white matter in the ICP, (2) the cerebellar nuclei, and (3) the cerebellar granular layer. When labeled thin axons bearing swellings in the ICP (n=23), cerebellar nuclei (n=42) and the granular layer (n = 81) were traced proximally, they were always found to be collaterals of OC axons which gave rise to some CFs. This indicated that all of the axon terminals in the ICE the cerebellar nuclei and the granular layer were derived from OC axons terminating as CFs in the cerebellar cortex, and none of the OC axons specifically terminated in the ICE the cerebellar nuclei or the granular layer without sending CFs to the cerebellar cortex. About half of the OC axons reconstructed throughout the ICP had a single or occasionally a few thin collaterals in the ICP. Thin collaterals to the cerebellar nuclei were observed in 20 out of 22 OC axons that were reconstructed throughout the deep cerebellar white matter. Each of these 20 axons had one (n= 15), two (n=3), three (n= 1), or six collaterals (n= 1) (1.4 + 1.2 collaterals per OC axon). Each OC axon innervated only a single cerebellar nucleus. Since thin collaterals terminating in the granular layer were also sometimes given off at this portion of OC axons, collaterals could not be identified as terminating in the cerebellar nuclei without tracing them to their terminals. The diameters of nuclear collaterals were 0.2-0.3 p,m. Terminal branches of nuclear collaterals bore several en-passant swellings and terminal swellings. The average number of swellings per OC axon was 54.0 +_66.0 (n = 22).
177
In counterstained sections, some swellings touched somata and proximal dendrites of both large and small neurons. We observed that some OC axons gave rise to thin collaterals terminating in the i vestibular nucleus. Thin collaterals terminating mainly in the granular layer were given off from stem or thick branches of OC axons in the deep white matter and in the folial white matter (Fig. 1A). Some of these thin collaterals terminated in the white matter with a solitary terminal swelling or a few en-passant and. a terminal swelling, while the rest of these thin collaterals entered the granular layer. Swellings were most dense in the upper portion of the granular layer, but were seen at all depths in that layer. In counterstained preparations, some swellings in the granular layer seemed to make a contact with a soma of a presumed Golgi cell. Other swellings were located among a dense aggregation of blue-colored granule cells. Thin collaterals usually terminated in the same lobule and in the same thin parasagittal zone as CFs originating from the same OC axons. The number of granular layer thin collaterals ranged from 3 to 16 (average, 8.5) axons per OC axon.
Morphology of single mossy fibers of lateral reticular nucleus neurons Complete reconstruction of the whole trajectory of single LRN axons could be made from an injection site to terminations of every branch in 11 identified LRN axons, and nearly- complete reconstruction except for distal portions of some branches in the cerebellum was made in 18 identified LRN axons. A majority of reconstructed axons of LRN neurons passed through the ipsilateral ICP to the cerebellum (n=25 out of 29) (Fig. 3), but some axons (n =4) passed the midline, ran through or under the contralateral IO, and entered the cerebellum through the contralateral ICE Generally, an arbor of an LRN axon within the cerebellum could be regarded as consisting of a thick stem axon running transversely, cortical branches of various diameters, and very thin nuclear branches. The stem axons ran medially toward the midline in the deep white matter rostral and dorsal to the DCN. Some stem axons did not cross the midline in
the cerebellum and made projections only on the same side as the ICP which they passed through (n = 4 of 29), but the other stem axons crossed the midline within the cerebellum to make bilateral projections. All completely reconstructed axons had a larger number of terminals in the Cx on the same side as the ICP through which they ran. While running transversely in the cerebellar white matter towards the contralateral side, stem axons gave rise to several primary collaterals to the Cx and DCN. Individual primary collaterals arose at an almost right angle from the stem axon, and ran largely in the parasagittal plane in the Cx. On their way, individual primary collaterals widely ramified mainly in a dorsoventral direction, but did not spread so much in a mediolateral direction, so that each collateral terminated as mossy fiber rosettes in the granular layer in a relatively narrow longitudinal zonal area covering one to four lobules (Fig. 4). The number of cortical primary collaterals given off from a transverse stem axon in the deep cerebellar white matter was 5-9 (7.0_+ 1.0, n--29). In each of bilaterally projecting neurons, the number of ipsilateral primary collaterals was 3-5 (3.8 _+0.8, n--25) and the number of contralateral primary collaterals was 2-6 (3.7_+ 1.4, n =25). In each of unilateral projecting neurons, the number of primary collaterals was 5-8 (6.0_+ 1.2, n=4). These primary collaterals further ramified to innervate 1-4 lobules. Single longitudinal arrangement of terminals in the Cx was originated usually from a single primary collateral, but occasionally from multiple primary collaterals. (1) Distribution of terminals of single LRN axons in the cerebellar cortex Each terminal branch of a single LRN axon always terminated as rosette-type terminals. They were borne either at the end (terminal type) or in the middle of the fiber (en-passant type) in the granular layer of the Cx. The number of rosettes per a single LRN axon ranged from 84 to 219 (154.0+37.0, n = 11). In each of the bilaterally projecting axons, the ratio of axon terminals in the ipsilateral Cx to those in the contralateral Cx ranged from 1.3 to 2.3 (1.7_+0.4, n=7). In the Cx, the terminations of reconstructed axons were mostly seen in the vermis
178
in the anterior lobe and often in lobule VI, especially Via. Cortical primary collaterals sent terminal branches sometimes to a single lobule but often to multiple lobules. Furthermore, terminal branches given off from each cortical primary collateral generally spread rather widely in the semi-parasagittal plane (width, 400-2000 ~m, usually more than 1000 Ixm) within each lobule, and extended to adjacent one to four lobules (Fig. 4, lower draw-
//
~
ings). In contrast, the spread of the terminal branches belonging to a single primary collateral was relatively restricted in the transverse direction (mediolateral width, usually 300-650 p~m). As a result, the distribution of terminals of a single LRN axon plotted in the coronal plane roughly showed a pattern of multiple longitudinal bands. Some cortical collaterals appeared to give rise to terminal branches which were distributed rather widely up to about 1 mm in the transverse plane. However, the
"
1mm
Fig. 3. Reconstruction of the entire trajectory of an axon originating from a lateral reticular nucleus neuron in the cerebellum and brain stem of the rat. The reconstruction of the axon labeled with extracellular iontophoretic application of biotinylated dextran amine was made from 158 serial sections of 50 I~m thickness. DN, IE FN and LRN, dentate, posterior interpositus, fastigial and lateral reticular nucleus; icp, inferior cerebellar peduncle; IO, inferior olive; sct, sptV and SPVI, spinocerebellar tract, spinal trigeminal tract and nucleus interpolaris (Shinoda, 1999).
179
c~
:i !
:
~dll
,
f
._.g
......
Fig. 4. Frontal (top) and lateral views (bottom) of a completely-reconstructed single LRN axon originating from the caudal part of the right LRN. This reconstruction was made on 148 serial coronal sections. This fiber entered the cerebellum through the ipsilateral icp, projected to the bilateral Cx (lobules II through VII) and DCN, and formed a multiple longitudinal zonal projection pattern by its cortical arborescent collaterals. Bottom drawings show lateral views of individual collaterals as indicated in the top drawing with lower case letters (a-g). Injection site shown in right inset on the bottom. Scale bars, 0.5 mm (top) and 1 m m (bottom) (from Wu et al., 1999).
180
spread of these terminal branches was not purely in the transverse direction, but also broad in the longitudinal direction which was not shown in the frontal view. Therefore, there seems to be a general tendency for longitudinal spread of terminal branches given off from single cortical primary collaterals. To reveal the nature of the longitudinal distribution of terminals in single LRN axons, terminal distributions of individual primary collaterals were analyzed in more detail, by using the completely reconstructed LRN axons. Figure 5 shows an example of such an analysis on the reconstructed LRN axon in Figure 4. The distances from the midline to individual terminals were measured and plotted on unfolded longitudinal strips of the vermal lobules in the Cx. In Figure 5A, two adjacent primary collaterals terminated in a longitudinal zone, which was labeled as zone c in Figure 4. In this zone, terminals that originated from one primary collateral were distributed in lobule V, and those from the other primary collateral were distributed in lobule VI where two clusters of terminals were slightly separated in lobule Via and VId. The longitudinal spread of these terminals extended over 6.5 mm, whereas the mediolateral spread of each cluster of terminals was localized between 230 and 400 I~m. Even as a whole, these terminals were well aligned in a single longitudinal strip less than 400 t~m wide in the transverse plane. Similarly, Figure 5B shows the distribution of terminals of another primary collateral of the same axon (labeled as zone e in Figure 4) on the unfolded cortical parasagittal strip. This collateral bifurcated into two main branches, of which one branch terminated in lobule III, and the other branch terminated in lobules IV and V. Although terminals were distributed widely along a parasagittal longitudinal strip (11.6 mm), the mediolateral spread o f the terminals was restricted
within 600 Ixm as a whole. As shown in this example, the general feature of the cortical distribution of terminals of a single LRN axon could be summarized as follows. Terminals belonging to one or sometimes two primary collaterals and spreading in a few lobules made a longitudinal zone in the parasagittal plane that was usually less than 500 Ixm wide in the transverse plane, and several such longitudinal zones of terminals were arranged in parallel in the mediolateral direction. (2) Morphology and distribution of collaterals terminating in the DCN and VN All completely reconstructed LRN axons supplied projections to the DCN by way of their collaterals. Axon collaterals terminating in the DCN were given off from transverse stem axons or proximal parts of cortical primary branches, and ran in the parasagittal direction. A single LRN axon had 2-3 (2.5+0.5, n = l l ) primary collaterals to the cerebellar (and/or vestibular) nuclei. Each of these primary collaterals was much thinner (diameter, 0.4-1.0 p,m) than stem axons (diameter 1.5-2.0 Ixm) and cortical branches, and had a localized termination area in the DCN. An arbor of a nuclear collateral could be sorted into a primary axon collateral and terminal branches, although the distinction between them was not definite. A primary axon collateral ran toward the target DCN in a relatively straight path without any branching when the branching point was far from the DCN. Within the DCN, a primary collateral bore several en-passant swellings and had only one or two ramifications to produce several (usually one to four) terminal branches. Each terminal branch was usually about 200-500 Ixm long and bore frequent en-passant swellings and sometimes short branchlets bearing a terminal swelling and occasionally a
Fig. 5. Longitudinal distribution of axon terminals of single primary collaterals of a reconstructed LRN axon in the unfolded Cx. A: Terminals of two primary collaterals plotted on the unfolded Cx (left). Two terminals in lobule VII are not included in the unfolded strip, but they were located in the same parasagittal strip, even though they wer e far away from the other terminals. B: Terminals borne on a proimary collateral plotted on unfolded lobules III-V (right). The rostrocaludal distance of the Cx was measured at the surface of the molecular layer on a reconstructed parasagittal section, and the locations of axon ternimals in the granular layer were projected to the corresponding sites on the surface of the unfolded Cx. Arabic numbers attached to the portions of the folia correspond to those on the unfolded cortical parasagittal strips. Roman numbers indicate names of lobules. Note the narrow mediolateral and wide longitudinal distributions of terminals belonging to individual primary collaterals. Scale bars = 1 mm (from Wu et al., 1999).
181
A
3L
B
J
~
;J T~
]
•
:
]
~l III
i
L
!~iri IV
i"il ~ ~
"
",, '.,': .....
""'-,1 e
° I
_2___r
V
Vl 8
:I i i 1
R 7
,
, ,
I t
Lt+
Rt C
182
few en-passant swelling. Usually, a single primary collateral terminated in only one target DCN, but sometimes sent terminal branches into two cerebellar nuclei, or into both the cerebellar and vestibular nuclei. Projection to the DCN was observed in all reconstructed LRN axons. LRN axons innervated most often the fastigial nucleus (FN) and interpositus nucleus (IP), and occasionally the DN and the IE There seemed to be a rough correspondence in the cortical and nuclear projections of single LRN axons compatible with the zonal arrangement in corticonuclear projection (Voogd et al., 1996). When an axon had a collateral in the FN, the same axon always had terminations in the vermis on the same side as the FN. When an axon had collateral(s) in the IP and the DN, the same axon often innervated the intermediate area including the most lateral vermis and the hemisphere of the Cx, respectively, although there were occasional exceptions.
Summary and discussion The present study has revealed that OC axons gave rise to a number of thin collaterals. Due to the abundance of these non-CF thin collaterals, it seems better to make a distinction between the terms CFs and OC axons, as was done in the present paper. The present findings on the innervation of PC dendrites by CFs are basically similar to those in previous reports (Ramdn y Cajal, 1911; Palay and Chan-Palay, 1974). The number of swellings on a single CF in the present study (n = 250) is comparable to a previously measured value in the rat (n= 288; Rossi et al., 1993) and larger than a value in the frog (n = about 100 beads; Llin~is et al., 1969). The average number of CFs per OC axon in this study was close to the number (n = about 7) inferred in the rat by counting the total number of IO neurons and PCs (Schild, 1970). Contact of interneurons by some swellings of CFs in the molecular layer was emphasized by Scheibel and Scheibel (1954) in their study with Golgi staining. Despite the contact of CF terminals on interneurons, the formation of a synaptic structure between them has been excluded in an electron-microscopic study (Hfimori and Szentgothai, 1980). On the other hand, electrophysiological
studies have demonstrated a weak excitatory effect of CFs on some interneurons (Eccles et al., 1966). Terminals in the granular layer were originated either from thin collaterals of OC axons or from retrograde collaterals of CF terminal arborizations. The former was the main source of swellings in the granular layer. The morphology of the thin collaterals in the present study was consistent with "globose varicosities connected by a fine thread" as described in Golgi preparations and electron micrograms (Chan-Palay and Palay, 1971). Swellings of thin collaterals (about 1.7 % of the total number of swellings per OC axon) were most abundant in the upper portion of the granular layer just underneath the PC layer, in which Golgi cells are usually located. Furthermore, some of these swellings were observed to touch presumed Golgi cells in the present study, which is consistent with electronmicroscopic findings on the innervation of somata of Golgi cells by thin collaterals (H~imori and Szenthothai, 1980; Chan-Palay and Palay, 1971). Inferior olive stimulation has been shown electrophysiologically to have a weak direct excitatory effect on Golgi cells (Eccles et al., 1966). Ninety-one percent of the OC axons examined had nuclear collaterals; since the possibility of insufficient staining could not be excluded, this percentage may be an underestimation. The ratio of swellings in the cerebellar nuclei versus those of CF terminal arborizations was about 0.036 in individual OC axons in the present study. However, since the volume of the cerebellar nuclei is much smaller than that of the cerebellar cortex, and significant convergence of input from OC axons to cerebellar nucleus neurons is present (Sugihara et al., 1996), cerebellar nucleus projection of OC fibers can still be functionally important. Some swellings seemed to make contact with the soma and the proximal portions of dendrites of large neurons in the present study, which is consistent with the steep rising phase of postsynaptic excitatory potentials in cerebellar nucleus neurons following IO stimulation (Kitai et al., 1977; Shinoda et al., 1987). Although intracellular potentials were presumably recorded only from large output neurons in the cerebellar nuclei, the present study suggested that small neurons were also innervated by OC axons.
183
The present study revealed that virtually all reconstructed LRN axons projected not only to the Cx as mossy fibers, but also to the DCN including the VN by their axon collaterals. None of the LRN neurons specifically projected to the DCN without projecting to the Cx, namely all axon terminals of LRN neurons in the DCN and VN belonged to axon collaterals of mossy fibers projecting to the Cx. In general, axon collaterals in the DCN and VN had a simple branching pattern and a narrow projection area, whereas axon collaterals in the Cx had a complex branching pattern, and a wide mediolateral spread, usually bilaterally, with multiple longitudinal zones in multiple lobules. Cerebellar cortical projection of single LRN axons is characterized by broadness of the termination area. Single LRN axons simultaneously projected to areas different in mediolateral locations in several lobules, often bilaterally. A significant overlap of termination area of single axons originating from within an injection site was suggested. Similar broadness of cortical innervation has been suggested in axons of other mossy fiber systems such as cerebellar projection from the pontine nucleus (Shinoda et al., 1992). This broadness of cerebellar cortical innervation in single LRN axons makes a good contrast to a single zonal projection of olivocerebellar axons. Virtually all LRN axons reconstructed in the present study had collaterals terminating in the DCN and also in the VN in some cases. This is the first mossy fiber system in which all cerebellar cortical projection neurons have axon collaterals to the DCN. However, high incidence of nuclear projection via axon collaterals can not be generalized to all mossy fiber systems, since many mossy fibers originating from the pontine nucleus lack nuclear collaterals (Shinoda et al., 1992). We confirmed that virtually all cerebellar nuclear projection from the LRN was via thin collaterals given off from stem axons or primary cortical collaterals. Previous anterograde or retrograde studies have not identified whether nuclear projections of LRN neurons are via collaterals of mossy fibers or there exist specific nuclear projecting neurons. It is very clear that the topographical relationship in the LRN-nuclear projection was much weaker and vaguer than in the olivonuclear
projection (Van der Want et al., 1989; Sugihara et al., 1996), in which neurons in a subdivision of the IO project to a relatively localized area in one of the cerebellar nuclei. This is partly because single LRN axons had multiple nuclear collaterals projecting to more than one cerebellar nuclei, and multiple cortical collaterals spreading rather widely in the transverse plane, while single IO neurons had multiple collaterals spreading in a very narrow longitudinal zone in the Cx (Sugihara et al., 1997). Large injections of BDA into the LRN visualized longitudinal multiple zones of mossy fiber terminals in the Cx in the present study. This finding is consistent with the results obtained by Ktinzle (1975) and Chan-Palay et al. (1977) using an autoradiographic method in the rat, and by Ruigrok and Cella (1995) using PHA-L. One of the most important findings in this study is the existence of multiple zonal projection of single LRN axons in the Cx. This projection made a sharp contrast with the projection of the IO, which also shows multiple longitudinal zonal pattern, when labeled by a mass injection of anterograde tracer (Chan-Patay et al., 1977; Van der Want et al., 1989). However, at the level of single axons, the cortical projection is generally localized within a single longitudinal zone (Sugihara et al., 1997; Sugihara et al., 1999). The multiple zonal projection may not necessarily contradict with the patchy representation of receptive fields of the facial area (Shambes et al., 1978), since terminals of single axons seemed to be arranged into clusters within each zone. However, analysis of single mossy fiber projection from the trigeminal system would be necessary to clarify the projection pattern. One of the important remaining questions is how longitudinal arrangements of the LRN projection are related to those zonal components named with A, B, C1, C2, C3, DO, D1, D2, which have been identified with olivocerebellar and corticonuclear projections (Groenewegen and Voogd, 1977; Buissert-Delmas and Angaut, 1993; Voogd, 1995). We attempted in vain to correlate these zones with the present zones of LRN terminals by combining acetylcholine esterase staining and the present staining, since the zones visualized by acetylcholine esterase staining appeared in the posterior vermis. The relationship between LRN projecting
184
multiple zones and zebrin-defined zones which has been addressed by Ruigrok and Cella (1995) and those zones defined by the different methods seem to correspond well to each other. Thus, functional significance of cerebellar compartmentation, which has been so far studied from the viewpoints of labeling with molecular markers, olivocerebellar and corticonuclear projections, will be also considered from the viewpoint of mossy fiber projections. Excitatory mossy fiber inputs to the Cx is relayed by granule cells and parallel fibers to reach Purkinje cells. Although a parallel fiber spreads in the transverse direction, the most effective input from the parallel fiber is given to Purkinje cells just above the given granule cell through synapses formed by the parallel fiber (Eccles et al., 1967). Therefore, the zonal projection of a single LRN axon may be functionally significant in conveying an excitatory input to Purkinje cells in that specific zone. Individual LRN axons projected to multiple lobules in both the hemisphere and the vermis, and even often bilaterally. Therefore, cerebellar cortical areas innervated by a single mossy fiber from the LRN may not necessarily be related to each other in terms of their functional roles. Then, the question is what is the role of LRN axons sending presumably specific information to different cortical areas. Concerning the nuclear projection, it is obvious that the cortical and nuclear projections of any single reconstructed LRN axon did not exactly follow the so-called corticonuclear topographical relationship, since the cortical projection was generally more widely spread in the transverse direction than the nuclear projection. Then, this situation does not exactly fit the classical micro-complex scheme of Ito (1984), in which the nuclear inputs by a mossy fiber collateral and by a Purkinje cell that receives the same mossy fiber input through granule cells converge onto the same target nuclear output neuron. The nuclear collaterals of mossy fibers are important in activating nuclear neurons as a source for excitation of their target neurons (Shinoda et al., 1992, 1997). To understand the functional interactions at nuclear efferent neurons between Purkinje cell inputs in widely-distributed cortical areas innervated by a single mossy fiber and nuclear inputs by a collateral of the same mossy
fiber, we need more detailed information as to how single Purkinje cells spread in the DCN, how Purkinje cells in different cortical areas send their informations to the DCN, and how those informations are integrated as an output in the DCN. Furthermore, it will be also useful for understanding the functional mechanism of whole mossy fiber systems to specify the similarity and the difference in mossy fibers in other systems such as spinocerebellar and cuneocerebellar systems.
Acknowledgements This research was supported by a research grant from CREST (Core Research for Evolutional Science and Technology) of the Japan Science and Technology Corporation to Y.S., and by Grants-inAid for Scientific Research from the Ministry of Education, Science and Culture of Japan to I.S.
References Andersson, G. and Oscarsson, O. (1978) Climbing fiber microzones in cerebellar vermis and their projection to different groups of cells in the lateral vestibular nucleus. Exp. Brain Res., 32: 564-579. Armstrong, D.M., Harvey, R.J. and Schild, R.E (1973) The spatial organization of climbing fibre branching in the cat cerebellum. Exp. Brain Res., 18: 40-58. Brodal, A., Walberg, E, Berkley, K.J. and Pelt, A. (1980) Anatomical demonstration of branching olivocerebellar fibres by means of a double retrograde labelling technique. Neuroscience, 5: 2193-2202. Buisseret-Delmas, C. and Angaut, P. (1993) The cerebellar olivo-corticonuclear connections in the rat. Prog. Neurobiol., 40: 63-87. Chan-Palay, V. and Palay, S.L. (1971) Tendril and glomerular collaterals of climbing fibers in the granular layer of the rat's cerebellar cortex. Z. Anat. Entwickl-Gesch., 133: 247-273. Chan-Palay, V., Palay, D.L., Brown, J.T. and Van Itallie, C. (1977) Sagittal organization of olivocerebellar and reticulocerebellar projections: autoradiographic studies with 35S-methionine. Exp. Brain Res., 30: 561-576. Dietrichs, E., Bjaalie, J.G. and Brodal, P. (1983) Do pontocerebellar fibers send collaterals to the cerebellar nuclei? Brain Res., 259: 127-131. Eccles, J.C., Ito, M. and Szent~igothai, J. (1967) The Cerebellum as a Neuronal Machine. Springer-Verlag, Berlin-Heidelberg-New York, pp. 227-261. Eccles, J.C., Llin~is, R. and Sasaki, K. (1966) The inhibitory interneurones within the cerebellar cortex. Exp. Brain Res., 1: 1-16. Ekerot, C-.E and Larson, B. (1982) Branching of olivary axons to innervate pairs of sagittal zones in the cerebellar anterior lobe of the cat. Exp. Brain Res., 48: 185-198.
185
Futami, T., Shinoda, Y. and Yokota, J. (1979) Spinal axon collaterals of corticospinal neurons identified by intracellular injection of horseradish peroxidase. Brain Res., 164: 279-284. Gerrits, N.M. and Voogd, J. (1987) The projection of the nucleus reticularis tegmenti pontis and adjacent regions of the pontine nuclei to the central cerebellar nuclei in the cat. J. Comp. Neurol., 258: 52-62. Groenewegen, H.J. and Voogd, J. (1977) The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of the cat cerebellum. J. Comp. NeuroL, 174: 417-488. H~,mori, J. and Szentfiothai, .1. (1980) Lack of evidence of synaptic contacts by climbing fibre collaterals to basket and stellate cells in developing rat cerebellar cortex. Brain Res., 186: 454-457. Ito, M., Orlov, I. and Yamamoto, M. (1982) Topographical representation of vestibulo-ocular reflex in rabbit cerebellar flocculus. Neuroscience, 7: 1657-1664. Ito, M. (1984) The Cerebellum and Neural Control. New York: Raven Press. Kitai, S.T., McCrea, R.A., Preston, R.J. and Bishop, G.A. (1977) Electrophysiological and horseradish peroxidase studies of precerebellar afferents to the nucleus interpositus anterior. I. Climbing fiber system. Brain Res., 122: 197-214. Krieger, C., Shinoda, Y. and Smith, A.M. (1985) Labelling of cerebellar mossy fiber afferents with intra-axonal horseradish peroxidase. Exp. Brain Res., 59: 414-417. K[inzle, H. (1975) Autoradiographic tracing of the cerebellar projections from the lateral reticular nucleus in the cat. Exp. Brain Res., 22: 255-266. Llin~s, R., Bloedel, J.R. and Hillman, D.E. (1969) Functional characterization of neuronal circuitry of frog cerebellar cortex. J. Neurophysiol., 32: 847-870. Matsushita, M. and Yaginuma, H. (1995) Projections from the central cervical nucleus to the cerebellar nuclei in the rat, studied by anterograde axonal tracing. J. Comp. Neurol., 353: 234-246. Mihailoff, G.A. (1993) Cerebellar nuclear projections from the basilar pontine nuclei and nucleus reticularis tegmenti pontis as demonstrated with PHA-L tracing in the rat. J. Comp. Neurol., 330: 130-146. Oscarsson, O. (1976) Spatial distribution of climbing and mossy fibre inputs into the cerebeUar cortex. In: O. Creutzfeldt (Ed.), Experimental Brain Research SuppL 1: Afferent and Intrinsic Organization of Laminated Structures in the Brain, Berlin: Springer Verlag, pp. 36-42. Palay, S.L. and Chan-Palay, V. (1974) Cerebellar Cortex. Cytology and Organization. New York: Springer-Verlag. Ram6n y Cajal S. (1911) Histologie du Systeme Nerveux de l'Homme et des VertbrC Vol. II. Paris: Maloine. Rosina, A. and Provini, L. (1983) Somatotopy of climbing fiber branching to the cerebellar cortex in cat. Brain Res., 289: 45-63. Rossi, E, Borsello, T., Vaudano, E. and Strata, P. (1993) Regressive modifications of climbing fibres following Put-
kinje cell degeneration in the cerebellar cortex of the adult rat. Neuroscience, 53: 759-778. Ruigrok, T.J.H. and Cella, E (1995) Precerebellar nuclei and red nucleus. In: G. Paxinos (Ed.), The Rat Nervous System, Vol. I11. Brain Stem and Cerebellum, Sydney: Academic Press, pp. 281-286. Sato, Y., Kawasaki, T. and Ikarashi, K. (1983) Afferent projections from the brainstem to the floccular three zones in cats. II. Mossy fiber projections. Brain Res., 272: 37-48. Scheibel, M.E. and Scheibel, A.B. (1954) Observations on the intracortical relations of the climbing fibers of the cerebellum. A Golgi study. J. Comp. Neurol., 101: 733-763. Schild, R.E (1970) On the inferior olive of the albino rat. J. Comp. Neurol., 140: 255-260. Shambes, G.M., Beermann, D. H. and Welker, W. (1978) Multiple tactile areas in cerebellar cortex: another patchy cutaneous projection to granule cell columns in rats. Brain Res., 157: 123-128. Shinoda, Y. (1999) Visualization of the entire trajectory of long axons of single mammalian CNS neurons. Brain Res. Bull., 50: 387-388. Shinoda, Y., Sugiuchi, Y. and Futami, T. (1993) Organization of excitatory inputs from the cerebral cortex to the cerebellar dentate nucleus. Can. J. Neurol. Sei., 20 (Suppl. 3): S19$28. Shinoda, Y., Izawa, Y., Sugiuchi Y. and Futami, T. (1997) Functional significance of excitatory projections from the precerebellar nuclei to interpositus and dentate nucleus neurons for mediating motor, premotor and parietal cortical inputs. In: C.I. deZeeuw, P. Strata, and J. Voogd (Eds), The Cerebellum: from Structure to Control Progress in Brain Research, Vol. 114, Amsterdam: Elsevier, pp. 193-207. Shinoda, yr., Ohgaki, T. and Futami, T. (1986) The morphology of single lateral vestibulospinal tract axons in the lower cervical spinal cord of the cat. J. Comp. NeuroL, 249: 226-241. Shinoda, Y., Yokota, J. and Futami, T. (1981) Divergent projection of individual corticospinal axons to motoneurons of multiple muscles in the monkey. Neurosci. Lett., 23: 7-12. Shinoda, Y., Sugiuchi, Y. and Futami, T. (1987) Excitatory inputs to cerebellar dentate nucleus neurons from the cerebral cortex in the cat. Exp. Brain Res., 67: 299-315. Shinoda, Y., Sugiuchi, Y., Futami, T. and Izawa, R. (1992) Axon collaterals of mossy fibers from the pontine nucleus in the cerebellar dentate nucleus. J. Neurophysiol., 67: 547-560. Sugihara, I., Wu, H. and Shinoda, Y. (1996) Morphology of axon collaterals of single climbing fibers in the deep cerebellar nuclei of the rat. Neurosci. Lett., 217: 33-36. Sugihara, I., Wu, H.-S. and Shinoda, Y. (1997) Projection of climbing fibers originating from single olivocerebellar neurons in the rat cerebellum. Soc. Neurosci. Abstr., 23: 1830. Sugihara, I., Wu, H. and Shinoda, Y. (1999) Morphology of single olivocerebellar axons labeled with biotinylated dextran amine in the rat. J. Comp. NeuroL, 414: 131-148. Van der Want, J.J.L. and Voogd, J. (1987) Ultrastructural identification and localization of climbing fiber terminals in
186 the fastigial nucleus of the cat. J. Comp. NeuroI., 258: 81-90. Van der Want, J.J.L., Wiklund, L., Guegan, M., Ruigrok, T. and J. Voogd (1989) Anterograde tracing of the rat olivocerebellar system with Phaseolus vulgaris leucoagglutinin (PHA-L). Demonstration of climbing fiber collateral innervation of the cerebellar nuclei. J. Comp. NeuroL, 288: 1-18. Voogd, J. (1995) Cerebellum. In: G. Paxinos (Ed.), The Rat Nervous System, 2nd edn. Sydney: Academic Press, pp. 309-350.
Voogd, J., Jaarsma, D. and Marani, E. (1996) The cerebellum, chemoarchitecture and anatomy. In: L.W. Swanson, A. Bjsrklund, and T. Hskfelt (Eds), Integrated Systems of the
CNS, Part 111. Cerebellum, Basal Ganglia, Olfactory System. Handbook of Chemical Neuroanatomy Vol. 12, Amsterdam: Elsevier, pp. 1-369. Wu, H., Sugihara, I. and Shinoda, Y. (1999) Projection patterns of single mossy fibers originating from the lateral reticular nucleus in the rat cerebellar cortex and nuclei. J. Comp. Neurol., 411:97-118.
CHAPTER
12
The entire trajectory of single climbing and mossy fibers in the cerebellar nuclei and cortex Y. Shinoda*, I. Sugihara, H.-S. Wu and Y. Sugiuchi Department of Systems Neurophysiology, School of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo, 113-8519, Japan
Introduction
Afferents of the cerebellum from many different origins, including brain stem precerebellar nuclei except the inferior olive (IO), take the form of mossy fibers in the cerebellar cortex (Cx), while olivocerebellar (OC) axons take the form of climbing fibers (CFs). Since the discovery of CFs in the cerebellar cortex (Ram6n y Cajal, 1911), the morphology of these fibers has been extensively studied by Golgi staining, electron-microscopy, and recently anterograde labeling. Despite abundant anatomical and physiological studies on the OC system, there is little information available on the organization of OC projection at the level of single axons. For example, the exact number of CFs that a single OC axon gives rise to is not known, although the average number of CFs per OC axon (about seven in the rat) was inferred by counting the total numbers of PCs and IO neurons (Schild, 1970). Electrophysiological studies (Armstrong et al., 1973; Ekerot and Larson, 1982) and retrograde labeling studies (Brodal et al., 1980; Rosina and Provini, 1983) have shown that single OC axons project to separate sites in the Cx, but the general extent of the spread of single OC axons in the cerebellar cortex remains unknown. Due to uniformity of the local circuit of the Cx, the *Corresponding author. Tel.: 81-3-5803-5153; Fax: 81-3-5803-5155; e-mail: [email protected]
projection pattern of afferents to the Cx including the mossy fiber system is significant in determining functional compartmentation and organized action of the cerebellum (Eccles et al., 1967; Palay and Chan-Palay, 1974; Ito, 1984). Recent studies of mapping receptive fields through the olivocerebellar climbing fiber projection have suggested that the Cx is composed of many longitudinal functional compartments (Andersson and Oscarsson, 1978; Ito et al, 1982; Ito, 1984). This electrophysiologically defined organization is consistent with the anatomically defined longitudinal compartments inthe Cx (Groenewegen and Voogd, 1977; Voogd et al., 1996). Some mossy fiber projections also seem to show longitudinal organization in the vestibulofloccular system (Sato et al., 1983) and the spinocerebellar system (Oscarsson, 1976). Whether or not there is a common principle in the cortical organization of the projection of mossy fibers conveying somatosensory information is still elusive. Besides the projection to the Cx, climbing and mossy fiber projections to the deep cerebellar nucleus (DCN) have not been fully investigated. The DCN gives rise to cerebellar output to target structures outside the cerebellum, and efferent neurons in the DCN receive inhibitory input from cortical Purkinje cells (Eccles et al., 1967; Ito, 1984). Therefore, the existence or absence of excitatory inputs to the DCN from the brainstem must be important for our understanding of the
174
cerebellar function (Shinoda et al., 1993, 1997). It has been generally assumed that excitatory inputs to the DCN are supplied through collaterals of mossy and climbing fiber afferents to the Cx. In spite of a wealth of anatomical reports on the afferent pathways to the Cx, there have been far fewer studies on afferent projections to the DCN until recently. The existence of the projections of mossy fibers to the DCN had been controversial (Chan-Palay, 1977; Dietriches et al., 1983; Ito, 1984), and reliable anatomical data on the projections from the precerebellar nuclei to the DCN had not been available until recently. (see Dietrichs et al. (1983) for the details of the problems). Using modem reliable anatomical staining methods, the projections from the pontine nucleus, nucleus reticularis tegmenti pontis, IO and spinal cord to the DCN have been definitely confirmed (Gerritis and Voogd, 1987; Van der Want et al., 1989; Shinoda et al., 1992; Mihailoff, 1993; Matsushita and Yaginuma, 1995). Intra-axonal staining with horseradish peroxidase of electrophysiologically identified neurons has been successfully used by our group to analyze the trajectories of single axons of various neurons (Futami et al., 1979; Shinoda et al., 1981, 1986, 1992). With this method, we have revealed the detailed morphology of single mossy fibers in the Cx (Krieger et al., 1985) and the existence of collateral projection of mossy fibers of the pontine nucleus and the nucleus reticularis tegmenti pontis to the DCN (Shinoda et al., t992, 1997). In the pontine nucleus and the nucleus reticularis tegmenti pontis, only a portion of mossy fibers projecting to the Cx have axon collaterals to the DCN, and many mossy fibers project only to the Cx. However, a question still remains as to whether there exist fibers that only project to the DCN without projecting to the Cx. In addition, it is essential for understanding the neural mechanism of the cerebellar function to determine the exact percentage of fibers that have axon collaterals to the DCN in different cerebellar afferent fiber systems. The purpose of the present study was to analyze the entire trajectories of single OC axons and single axons of the lateral reticular nucleus (LRN) that had been labeled by the injection of biotinylated dextran amine into the IO, and the LRN in the rat, respectively. An anterograde tracer, biotinylated
dextran amine dissolved in physiological saline to give a concentration of 10-15% was injected either with pressure (0.001-0.05 Ixl) or electrophoresis (2 IxA positive current pulses of 1 sec duration at 0.5 Hz for 5-20 min to the IO or the LRN). By reconstructing entire axons on serial sections, we examined the detailed morphological characteristics of single OC axons and mossy fibers of the LRN (Wu et al., 1999; Sugihara et al., 1999).
Morphology of single olivocerebellar axons The trajectory of 16 completely reconstructed OC axons and 46 partially reconstructed axons terminating in the vermis, intermediate area, hemisphere and flocculus were used for the present analysis. Axons left an injection site in the IO toward the contralateral side, crossed the midline, ran transversely above or through the contralateral IO, and entered the white matter of the inferior cerebellar peduncle (ICP) (Fig. 1). Then they ran longitudinally through the dorsolateral ICP to enter the cerebellum. After entering the cerebellum, stem axons frequently ramified into many branches in the deep cerebellar white matter; these ramifications often occurred when the stem axons reached the semiparasagittal plane in which branches terminated as CFs. Stem axons and branches usually ran in the white matter rostral and dorsal to the cerebellar nuclei, but sometimes they ran through or beneath the cerebellar nuclei. Branches entered the folial white matter, ramified and ran along each other in the semi-parasagittal plane to reach the areas for their termination in the cerebellar cortex, in which they further ramified. The branches of OC axons were grouped into thick branches (0.7-1.4 Ixm in diameter) and thin collaterals (0.2-0.5 ixm in diameter). A single OC axon generated 2-17 final thick branches (6.1 _ 3.7, mean _+S.D., n = 16), each of which innervated a PC as a CE We observed thin retrograde collaterals (Scheibel and Scheibel, 1954) originating from a CF terminal arborization and reentering the PC and granular layers, and some swellings of the retrograde collaterals in the PC layer apparently touched a PC which was not the target of the CF terminal arborization.
175
n Z
o.~= \
Z
\
\ \\
E E
c~
(..)
/
/
/
-J
.....
\
~,.;>
/
¢)
© .-
J '
"
:
~i'i ¸
•. . . .
0
.
•
, ~.U .-~ ~ ,.
.~.~ v/
/
.
....
:,
E
~
.~
-~
_~~ •
"~
~
176
(1) Climbing fibers Most of the reconstructed OC axons terminated on the side contralateral to the IO from which they originated (Fig. 1B). The most distal portion of a thick branch of an OC axon, which was equivalent to a CF, reached a target PC usually at its soma. The number of swellings on the terminal arborization of a single CF was 154-321 (249.8_+42.2, n = 3 2 terminal arborizations). The diameter of swellings ranged from 0.5 to 2.0 p.m, mostly 1.2-1.6 p.m. Assuming 6.1 CFs per OC axon, 249.8 swellings per CF corresponded to 1524 swellings on CF terminal arborizations per OC axon, on average. To examine whether a terminal arborization on a single PC is always formed by a single thick branch of an OC axon, we observed 523 CFs in three experiments. A terminal arborization always originated from a single thick branch of an OC axon. Spatial distribution of climbing fibers originating from a single olivocerebellar axon had a characteristic feature as expected from the anatomical analysis of the olivocerebellar mass projection. All CFs of a single olivocerebellar axon terminated in a single strip-shaped area in the cerebellar cortex. This area was narrow in the medio-lateral direction (usually about 200 p.m) and wide in the rostrocaudal direction, covering single or multiple lobules (Fig. 2). Strip shaped termination areas of
Via Vlb
~~J//ll ' " ~ ~~~ VII 1ram½
~
Fig. 2. Lateral view of the entire trajectory of a single OC axon in the cerebellum. The mediolateal extent of this axon in the Cx was less than 300 ~ m (unpublished data).
single axons originating from olivary neurons located in adjacency in the inferior olive tended to have significant overlap in the cerebellar cortex. These projection patterns of olivocerebellar axons stood in sharp contrast to those of LRN neurons described in the next section. (2) Thin collaterals of OC axons Thin collaterals, which were given off from stem axons and thick branches of OC axons before they formed CF terminal arborizations, were more abundant in terms of the number per OC axon than thick branches. For convenience of description, they were classified into three types according to their main termination sites: (1) white matter in the ICP, (2) the cerebellar nuclei, and (3) the cerebellar granular layer. When labeled thin axons bearing swellings in the ICP (n=23), cerebellar nuclei (n=42) and the granular layer (n = 81) were traced proximally, they were always found to be collaterals of OC axons which gave rise to some CFs. This indicated that all of the axon terminals in the ICE the cerebellar nuclei and the granular layer were derived from OC axons terminating as CFs in the cerebellar cortex, and none of the OC axons specifically terminated in the ICE the cerebellar nuclei or the granular layer without sending CFs to the cerebellar cortex. About half of the OC axons reconstructed throughout the ICP had a single or occasionally a few thin collaterals in the ICP. Thin collaterals to the cerebellar nuclei were observed in 20 out of 22 OC axons that were reconstructed throughout the deep cerebellar white matter. Each of these 20 axons had one (n= 15), two (n=3), three (n= 1), or six collaterals (n= 1) (1.4 + 1.2 collaterals per OC axon). Each OC axon innervated only a single cerebellar nucleus. Since thin collaterals terminating in the granular layer were also sometimes given off at this portion of OC axons, collaterals could not be identified as terminating in the cerebellar nuclei without tracing them to their terminals. The diameters of nuclear collaterals were 0.2-0.3 p,m. Terminal branches of nuclear collaterals bore several en-passant swellings and terminal swellings. The average number of swellings per OC axon was 54.0 +_66.0 (n = 22).
177
In counterstained sections, some swellings touched somata and proximal dendrites of both large and small neurons. We observed that some OC axons gave rise to thin collaterals terminating in the i vestibular nucleus. Thin collaterals terminating mainly in the granular layer were given off from stem or thick branches of OC axons in the deep white matter and in the folial white matter (Fig. 1A). Some of these thin collaterals terminated in the white matter with a solitary terminal swelling or a few en-passant and. a terminal swelling, while the rest of these thin collaterals entered the granular layer. Swellings were most dense in the upper portion of the granular layer, but were seen at all depths in that layer. In counterstained preparations, some swellings in the granular layer seemed to make a contact with a soma of a presumed Golgi cell. Other swellings were located among a dense aggregation of blue-colored granule cells. Thin collaterals usually terminated in the same lobule and in the same thin parasagittal zone as CFs originating from the same OC axons. The number of granular layer thin collaterals ranged from 3 to 16 (average, 8.5) axons per OC axon.
Morphology of single mossy fibers of lateral reticular nucleus neurons Complete reconstruction of the whole trajectory of single LRN axons could be made from an injection site to terminations of every branch in 11 identified LRN axons, and nearly- complete reconstruction except for distal portions of some branches in the cerebellum was made in 18 identified LRN axons. A majority of reconstructed axons of LRN neurons passed through the ipsilateral ICP to the cerebellum (n=25 out of 29) (Fig. 3), but some axons (n =4) passed the midline, ran through or under the contralateral IO, and entered the cerebellum through the contralateral ICE Generally, an arbor of an LRN axon within the cerebellum could be regarded as consisting of a thick stem axon running transversely, cortical branches of various diameters, and very thin nuclear branches. The stem axons ran medially toward the midline in the deep white matter rostral and dorsal to the DCN. Some stem axons did not cross the midline in
the cerebellum and made projections only on the same side as the ICP which they passed through (n = 4 of 29), but the other stem axons crossed the midline within the cerebellum to make bilateral projections. All completely reconstructed axons had a larger number of terminals in the Cx on the same side as the ICP through which they ran. While running transversely in the cerebellar white matter towards the contralateral side, stem axons gave rise to several primary collaterals to the Cx and DCN. Individual primary collaterals arose at an almost right angle from the stem axon, and ran largely in the parasagittal plane in the Cx. On their way, individual primary collaterals widely ramified mainly in a dorsoventral direction, but did not spread so much in a mediolateral direction, so that each collateral terminated as mossy fiber rosettes in the granular layer in a relatively narrow longitudinal zonal area covering one to four lobules (Fig. 4). The number of cortical primary collaterals given off from a transverse stem axon in the deep cerebellar white matter was 5-9 (7.0_+ 1.0, n--29). In each of bilaterally projecting neurons, the number of ipsilateral primary collaterals was 3-5 (3.8 _+0.8, n--25) and the number of contralateral primary collaterals was 2-6 (3.7_+ 1.4, n =25). In each of unilateral projecting neurons, the number of primary collaterals was 5-8 (6.0_+ 1.2, n=4). These primary collaterals further ramified to innervate 1-4 lobules. Single longitudinal arrangement of terminals in the Cx was originated usually from a single primary collateral, but occasionally from multiple primary collaterals. (1) Distribution of terminals of single LRN axons in the cerebellar cortex Each terminal branch of a single LRN axon always terminated as rosette-type terminals. They were borne either at the end (terminal type) or in the middle of the fiber (en-passant type) in the granular layer of the Cx. The number of rosettes per a single LRN axon ranged from 84 to 219 (154.0+37.0, n = 11). In each of the bilaterally projecting axons, the ratio of axon terminals in the ipsilateral Cx to those in the contralateral Cx ranged from 1.3 to 2.3 (1.7_+0.4, n=7). In the Cx, the terminations of reconstructed axons were mostly seen in the vermis
178
in the anterior lobe and often in lobule VI, especially Via. Cortical primary collaterals sent terminal branches sometimes to a single lobule but often to multiple lobules. Furthermore, terminal branches given off from each cortical primary collateral generally spread rather widely in the semi-parasagittal plane (width, 400-2000 ~m, usually more than 1000 Ixm) within each lobule, and extended to adjacent one to four lobules (Fig. 4, lower draw-
//
~
ings). In contrast, the spread of the terminal branches belonging to a single primary collateral was relatively restricted in the transverse direction (mediolateral width, usually 300-650 p~m). As a result, the distribution of terminals of a single LRN axon plotted in the coronal plane roughly showed a pattern of multiple longitudinal bands. Some cortical collaterals appeared to give rise to terminal branches which were distributed rather widely up to about 1 mm in the transverse plane. However, the
"
1mm
Fig. 3. Reconstruction of the entire trajectory of an axon originating from a lateral reticular nucleus neuron in the cerebellum and brain stem of the rat. The reconstruction of the axon labeled with extracellular iontophoretic application of biotinylated dextran amine was made from 158 serial sections of 50 I~m thickness. DN, IE FN and LRN, dentate, posterior interpositus, fastigial and lateral reticular nucleus; icp, inferior cerebellar peduncle; IO, inferior olive; sct, sptV and SPVI, spinocerebellar tract, spinal trigeminal tract and nucleus interpolaris (Shinoda, 1999).
179
c~
:i !
:
~dll
,
f
._.g
......
Fig. 4. Frontal (top) and lateral views (bottom) of a completely-reconstructed single LRN axon originating from the caudal part of the right LRN. This reconstruction was made on 148 serial coronal sections. This fiber entered the cerebellum through the ipsilateral icp, projected to the bilateral Cx (lobules II through VII) and DCN, and formed a multiple longitudinal zonal projection pattern by its cortical arborescent collaterals. Bottom drawings show lateral views of individual collaterals as indicated in the top drawing with lower case letters (a-g). Injection site shown in right inset on the bottom. Scale bars, 0.5 mm (top) and 1 m m (bottom) (from Wu et al., 1999).
180
spread of these terminal branches was not purely in the transverse direction, but also broad in the longitudinal direction which was not shown in the frontal view. Therefore, there seems to be a general tendency for longitudinal spread of terminal branches given off from single cortical primary collaterals. To reveal the nature of the longitudinal distribution of terminals in single LRN axons, terminal distributions of individual primary collaterals were analyzed in more detail, by using the completely reconstructed LRN axons. Figure 5 shows an example of such an analysis on the reconstructed LRN axon in Figure 4. The distances from the midline to individual terminals were measured and plotted on unfolded longitudinal strips of the vermal lobules in the Cx. In Figure 5A, two adjacent primary collaterals terminated in a longitudinal zone, which was labeled as zone c in Figure 4. In this zone, terminals that originated from one primary collateral were distributed in lobule V, and those from the other primary collateral were distributed in lobule VI where two clusters of terminals were slightly separated in lobule Via and VId. The longitudinal spread of these terminals extended over 6.5 mm, whereas the mediolateral spread of each cluster of terminals was localized between 230 and 400 I~m. Even as a whole, these terminals were well aligned in a single longitudinal strip less than 400 t~m wide in the transverse plane. Similarly, Figure 5B shows the distribution of terminals of another primary collateral of the same axon (labeled as zone e in Figure 4) on the unfolded cortical parasagittal strip. This collateral bifurcated into two main branches, of which one branch terminated in lobule III, and the other branch terminated in lobules IV and V. Although terminals were distributed widely along a parasagittal longitudinal strip (11.6 mm), the mediolateral spread o f the terminals was restricted
within 600 Ixm as a whole. As shown in this example, the general feature of the cortical distribution of terminals of a single LRN axon could be summarized as follows. Terminals belonging to one or sometimes two primary collaterals and spreading in a few lobules made a longitudinal zone in the parasagittal plane that was usually less than 500 Ixm wide in the transverse plane, and several such longitudinal zones of terminals were arranged in parallel in the mediolateral direction. (2) Morphology and distribution of collaterals terminating in the DCN and VN All completely reconstructed LRN axons supplied projections to the DCN by way of their collaterals. Axon collaterals terminating in the DCN were given off from transverse stem axons or proximal parts of cortical primary branches, and ran in the parasagittal direction. A single LRN axon had 2-3 (2.5+0.5, n = l l ) primary collaterals to the cerebellar (and/or vestibular) nuclei. Each of these primary collaterals was much thinner (diameter, 0.4-1.0 p,m) than stem axons (diameter 1.5-2.0 Ixm) and cortical branches, and had a localized termination area in the DCN. An arbor of a nuclear collateral could be sorted into a primary axon collateral and terminal branches, although the distinction between them was not definite. A primary axon collateral ran toward the target DCN in a relatively straight path without any branching when the branching point was far from the DCN. Within the DCN, a primary collateral bore several en-passant swellings and had only one or two ramifications to produce several (usually one to four) terminal branches. Each terminal branch was usually about 200-500 Ixm long and bore frequent en-passant swellings and sometimes short branchlets bearing a terminal swelling and occasionally a
Fig. 5. Longitudinal distribution of axon terminals of single primary collaterals of a reconstructed LRN axon in the unfolded Cx. A: Terminals of two primary collaterals plotted on the unfolded Cx (left). Two terminals in lobule VII are not included in the unfolded strip, but they were located in the same parasagittal strip, even though they wer e far away from the other terminals. B: Terminals borne on a proimary collateral plotted on unfolded lobules III-V (right). The rostrocaludal distance of the Cx was measured at the surface of the molecular layer on a reconstructed parasagittal section, and the locations of axon ternimals in the granular layer were projected to the corresponding sites on the surface of the unfolded Cx. Arabic numbers attached to the portions of the folia correspond to those on the unfolded cortical parasagittal strips. Roman numbers indicate names of lobules. Note the narrow mediolateral and wide longitudinal distributions of terminals belonging to individual primary collaterals. Scale bars = 1 mm (from Wu et al., 1999).
181
A
3L
B
J
~
;J T~
]
•
:
]
~l III
i
L
!~iri IV
i"il ~ ~
"
",, '.,': .....
""'-,1 e
° I
_2___r
V
Vl 8
:I i i 1
R 7
,
, ,
I t
Lt+
Rt C
182
few en-passant swelling. Usually, a single primary collateral terminated in only one target DCN, but sometimes sent terminal branches into two cerebellar nuclei, or into both the cerebellar and vestibular nuclei. Projection to the DCN was observed in all reconstructed LRN axons. LRN axons innervated most often the fastigial nucleus (FN) and interpositus nucleus (IP), and occasionally the DN and the IE There seemed to be a rough correspondence in the cortical and nuclear projections of single LRN axons compatible with the zonal arrangement in corticonuclear projection (Voogd et al., 1996). When an axon had a collateral in the FN, the same axon always had terminations in the vermis on the same side as the FN. When an axon had collateral(s) in the IP and the DN, the same axon often innervated the intermediate area including the most lateral vermis and the hemisphere of the Cx, respectively, although there were occasional exceptions.
Summary and discussion The present study has revealed that OC axons gave rise to a number of thin collaterals. Due to the abundance of these non-CF thin collaterals, it seems better to make a distinction between the terms CFs and OC axons, as was done in the present paper. The present findings on the innervation of PC dendrites by CFs are basically similar to those in previous reports (Ramdn y Cajal, 1911; Palay and Chan-Palay, 1974). The number of swellings on a single CF in the present study (n = 250) is comparable to a previously measured value in the rat (n= 288; Rossi et al., 1993) and larger than a value in the frog (n = about 100 beads; Llin~is et al., 1969). The average number of CFs per OC axon in this study was close to the number (n = about 7) inferred in the rat by counting the total number of IO neurons and PCs (Schild, 1970). Contact of interneurons by some swellings of CFs in the molecular layer was emphasized by Scheibel and Scheibel (1954) in their study with Golgi staining. Despite the contact of CF terminals on interneurons, the formation of a synaptic structure between them has been excluded in an electron-microscopic study (Hfimori and Szentgothai, 1980). On the other hand, electrophysiological
studies have demonstrated a weak excitatory effect of CFs on some interneurons (Eccles et al., 1966). Terminals in the granular layer were originated either from thin collaterals of OC axons or from retrograde collaterals of CF terminal arborizations. The former was the main source of swellings in the granular layer. The morphology of the thin collaterals in the present study was consistent with "globose varicosities connected by a fine thread" as described in Golgi preparations and electron micrograms (Chan-Palay and Palay, 1971). Swellings of thin collaterals (about 1.7 % of the total number of swellings per OC axon) were most abundant in the upper portion of the granular layer just underneath the PC layer, in which Golgi cells are usually located. Furthermore, some of these swellings were observed to touch presumed Golgi cells in the present study, which is consistent with electronmicroscopic findings on the innervation of somata of Golgi cells by thin collaterals (H~imori and Szenthothai, 1980; Chan-Palay and Palay, 1971). Inferior olive stimulation has been shown electrophysiologically to have a weak direct excitatory effect on Golgi cells (Eccles et al., 1966). Ninety-one percent of the OC axons examined had nuclear collaterals; since the possibility of insufficient staining could not be excluded, this percentage may be an underestimation. The ratio of swellings in the cerebellar nuclei versus those of CF terminal arborizations was about 0.036 in individual OC axons in the present study. However, since the volume of the cerebellar nuclei is much smaller than that of the cerebellar cortex, and significant convergence of input from OC axons to cerebellar nucleus neurons is present (Sugihara et al., 1996), cerebellar nucleus projection of OC fibers can still be functionally important. Some swellings seemed to make contact with the soma and the proximal portions of dendrites of large neurons in the present study, which is consistent with the steep rising phase of postsynaptic excitatory potentials in cerebellar nucleus neurons following IO stimulation (Kitai et al., 1977; Shinoda et al., 1987). Although intracellular potentials were presumably recorded only from large output neurons in the cerebellar nuclei, the present study suggested that small neurons were also innervated by OC axons.
183
The present study revealed that virtually all reconstructed LRN axons projected not only to the Cx as mossy fibers, but also to the DCN including the VN by their axon collaterals. None of the LRN neurons specifically projected to the DCN without projecting to the Cx, namely all axon terminals of LRN neurons in the DCN and VN belonged to axon collaterals of mossy fibers projecting to the Cx. In general, axon collaterals in the DCN and VN had a simple branching pattern and a narrow projection area, whereas axon collaterals in the Cx had a complex branching pattern, and a wide mediolateral spread, usually bilaterally, with multiple longitudinal zones in multiple lobules. Cerebellar cortical projection of single LRN axons is characterized by broadness of the termination area. Single LRN axons simultaneously projected to areas different in mediolateral locations in several lobules, often bilaterally. A significant overlap of termination area of single axons originating from within an injection site was suggested. Similar broadness of cortical innervation has been suggested in axons of other mossy fiber systems such as cerebellar projection from the pontine nucleus (Shinoda et al., 1992). This broadness of cerebellar cortical innervation in single LRN axons makes a good contrast to a single zonal projection of olivocerebellar axons. Virtually all LRN axons reconstructed in the present study had collaterals terminating in the DCN and also in the VN in some cases. This is the first mossy fiber system in which all cerebellar cortical projection neurons have axon collaterals to the DCN. However, high incidence of nuclear projection via axon collaterals can not be generalized to all mossy fiber systems, since many mossy fibers originating from the pontine nucleus lack nuclear collaterals (Shinoda et al., 1992). We confirmed that virtually all cerebellar nuclear projection from the LRN was via thin collaterals given off from stem axons or primary cortical collaterals. Previous anterograde or retrograde studies have not identified whether nuclear projections of LRN neurons are via collaterals of mossy fibers or there exist specific nuclear projecting neurons. It is very clear that the topographical relationship in the LRN-nuclear projection was much weaker and vaguer than in the olivonuclear
projection (Van der Want et al., 1989; Sugihara et al., 1996), in which neurons in a subdivision of the IO project to a relatively localized area in one of the cerebellar nuclei. This is partly because single LRN axons had multiple nuclear collaterals projecting to more than one cerebellar nuclei, and multiple cortical collaterals spreading rather widely in the transverse plane, while single IO neurons had multiple collaterals spreading in a very narrow longitudinal zone in the Cx (Sugihara et al., 1997). Large injections of BDA into the LRN visualized longitudinal multiple zones of mossy fiber terminals in the Cx in the present study. This finding is consistent with the results obtained by Ktinzle (1975) and Chan-Palay et al. (1977) using an autoradiographic method in the rat, and by Ruigrok and Cella (1995) using PHA-L. One of the most important findings in this study is the existence of multiple zonal projection of single LRN axons in the Cx. This projection made a sharp contrast with the projection of the IO, which also shows multiple longitudinal zonal pattern, when labeled by a mass injection of anterograde tracer (Chan-Patay et al., 1977; Van der Want et al., 1989). However, at the level of single axons, the cortical projection is generally localized within a single longitudinal zone (Sugihara et al., 1997; Sugihara et al., 1999). The multiple zonal projection may not necessarily contradict with the patchy representation of receptive fields of the facial area (Shambes et al., 1978), since terminals of single axons seemed to be arranged into clusters within each zone. However, analysis of single mossy fiber projection from the trigeminal system would be necessary to clarify the projection pattern. One of the important remaining questions is how longitudinal arrangements of the LRN projection are related to those zonal components named with A, B, C1, C2, C3, DO, D1, D2, which have been identified with olivocerebellar and corticonuclear projections (Groenewegen and Voogd, 1977; Buissert-Delmas and Angaut, 1993; Voogd, 1995). We attempted in vain to correlate these zones with the present zones of LRN terminals by combining acetylcholine esterase staining and the present staining, since the zones visualized by acetylcholine esterase staining appeared in the posterior vermis. The relationship between LRN projecting
184
multiple zones and zebrin-defined zones which has been addressed by Ruigrok and Cella (1995) and those zones defined by the different methods seem to correspond well to each other. Thus, functional significance of cerebellar compartmentation, which has been so far studied from the viewpoints of labeling with molecular markers, olivocerebellar and corticonuclear projections, will be also considered from the viewpoint of mossy fiber projections. Excitatory mossy fiber inputs to the Cx is relayed by granule cells and parallel fibers to reach Purkinje cells. Although a parallel fiber spreads in the transverse direction, the most effective input from the parallel fiber is given to Purkinje cells just above the given granule cell through synapses formed by the parallel fiber (Eccles et al., 1967). Therefore, the zonal projection of a single LRN axon may be functionally significant in conveying an excitatory input to Purkinje cells in that specific zone. Individual LRN axons projected to multiple lobules in both the hemisphere and the vermis, and even often bilaterally. Therefore, cerebellar cortical areas innervated by a single mossy fiber from the LRN may not necessarily be related to each other in terms of their functional roles. Then, the question is what is the role of LRN axons sending presumably specific information to different cortical areas. Concerning the nuclear projection, it is obvious that the cortical and nuclear projections of any single reconstructed LRN axon did not exactly follow the so-called corticonuclear topographical relationship, since the cortical projection was generally more widely spread in the transverse direction than the nuclear projection. Then, this situation does not exactly fit the classical micro-complex scheme of Ito (1984), in which the nuclear inputs by a mossy fiber collateral and by a Purkinje cell that receives the same mossy fiber input through granule cells converge onto the same target nuclear output neuron. The nuclear collaterals of mossy fibers are important in activating nuclear neurons as a source for excitation of their target neurons (Shinoda et al., 1992, 1997). To understand the functional interactions at nuclear efferent neurons between Purkinje cell inputs in widely-distributed cortical areas innervated by a single mossy fiber and nuclear inputs by a collateral of the same mossy
fiber, we need more detailed information as to how single Purkinje cells spread in the DCN, how Purkinje cells in different cortical areas send their informations to the DCN, and how those informations are integrated as an output in the DCN. Furthermore, it will be also useful for understanding the functional mechanism of whole mossy fiber systems to specify the similarity and the difference in mossy fibers in other systems such as spinocerebellar and cuneocerebellar systems.
Acknowledgements This research was supported by a research grant from CREST (Core Research for Evolutional Science and Technology) of the Japan Science and Technology Corporation to Y.S., and by Grants-inAid for Scientific Research from the Ministry of Education, Science and Culture of Japan to I.S.
References Andersson, G. and Oscarsson, O. (1978) Climbing fiber microzones in cerebellar vermis and their projection to different groups of cells in the lateral vestibular nucleus. Exp. Brain Res., 32: 564-579. Armstrong, D.M., Harvey, R.J. and Schild, R.E (1973) The spatial organization of climbing fibre branching in the cat cerebellum. Exp. Brain Res., 18: 40-58. Brodal, A., Walberg, E, Berkley, K.J. and Pelt, A. (1980) Anatomical demonstration of branching olivocerebellar fibres by means of a double retrograde labelling technique. Neuroscience, 5: 2193-2202. Buisseret-Delmas, C. and Angaut, P. (1993) The cerebellar olivo-corticonuclear connections in the rat. Prog. Neurobiol., 40: 63-87. Chan-Palay, V. and Palay, S.L. (1971) Tendril and glomerular collaterals of climbing fibers in the granular layer of the rat's cerebellar cortex. Z. Anat. Entwickl-Gesch., 133: 247-273. Chan-Palay, V., Palay, D.L., Brown, J.T. and Van Itallie, C. (1977) Sagittal organization of olivocerebellar and reticulocerebellar projections: autoradiographic studies with 35S-methionine. Exp. Brain Res., 30: 561-576. Dietrichs, E., Bjaalie, J.G. and Brodal, P. (1983) Do pontocerebellar fibers send collaterals to the cerebellar nuclei? Brain Res., 259: 127-131. Eccles, J.C., Ito, M. and Szent~igothai, J. (1967) The Cerebellum as a Neuronal Machine. Springer-Verlag, Berlin-Heidelberg-New York, pp. 227-261. Eccles, J.C., Llin~is, R. and Sasaki, K. (1966) The inhibitory interneurones within the cerebellar cortex. Exp. Brain Res., 1: 1-16. Ekerot, C-.E and Larson, B. (1982) Branching of olivary axons to innervate pairs of sagittal zones in the cerebellar anterior lobe of the cat. Exp. Brain Res., 48: 185-198.
185
Futami, T., Shinoda, Y. and Yokota, J. (1979) Spinal axon collaterals of corticospinal neurons identified by intracellular injection of horseradish peroxidase. Brain Res., 164: 279-284. Gerrits, N.M. and Voogd, J. (1987) The projection of the nucleus reticularis tegmenti pontis and adjacent regions of the pontine nuclei to the central cerebellar nuclei in the cat. J. Comp. Neurol., 258: 52-62. Groenewegen, H.J. and Voogd, J. (1977) The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of the cat cerebellum. J. Comp. NeuroL, 174: 417-488. H~,mori, J. and Szentfiothai, .1. (1980) Lack of evidence of synaptic contacts by climbing fibre collaterals to basket and stellate cells in developing rat cerebellar cortex. Brain Res., 186: 454-457. Ito, M., Orlov, I. and Yamamoto, M. (1982) Topographical representation of vestibulo-ocular reflex in rabbit cerebellar flocculus. Neuroscience, 7: 1657-1664. Ito, M. (1984) The Cerebellum and Neural Control. New York: Raven Press. Kitai, S.T., McCrea, R.A., Preston, R.J. and Bishop, G.A. (1977) Electrophysiological and horseradish peroxidase studies of precerebellar afferents to the nucleus interpositus anterior. I. Climbing fiber system. Brain Res., 122: 197-214. Krieger, C., Shinoda, Y. and Smith, A.M. (1985) Labelling of cerebellar mossy fiber afferents with intra-axonal horseradish peroxidase. Exp. Brain Res., 59: 414-417. K[inzle, H. (1975) Autoradiographic tracing of the cerebellar projections from the lateral reticular nucleus in the cat. Exp. Brain Res., 22: 255-266. Llin~s, R., Bloedel, J.R. and Hillman, D.E. (1969) Functional characterization of neuronal circuitry of frog cerebellar cortex. J. Neurophysiol., 32: 847-870. Matsushita, M. and Yaginuma, H. (1995) Projections from the central cervical nucleus to the cerebellar nuclei in the rat, studied by anterograde axonal tracing. J. Comp. Neurol., 353: 234-246. Mihailoff, G.A. (1993) Cerebellar nuclear projections from the basilar pontine nuclei and nucleus reticularis tegmenti pontis as demonstrated with PHA-L tracing in the rat. J. Comp. Neurol., 330: 130-146. Oscarsson, O. (1976) Spatial distribution of climbing and mossy fibre inputs into the cerebeUar cortex. In: O. Creutzfeldt (Ed.), Experimental Brain Research SuppL 1: Afferent and Intrinsic Organization of Laminated Structures in the Brain, Berlin: Springer Verlag, pp. 36-42. Palay, S.L. and Chan-Palay, V. (1974) Cerebellar Cortex. Cytology and Organization. New York: Springer-Verlag. Ram6n y Cajal S. (1911) Histologie du Systeme Nerveux de l'Homme et des VertbrC Vol. II. Paris: Maloine. Rosina, A. and Provini, L. (1983) Somatotopy of climbing fiber branching to the cerebellar cortex in cat. Brain Res., 289: 45-63. Rossi, E, Borsello, T., Vaudano, E. and Strata, P. (1993) Regressive modifications of climbing fibres following Put-
kinje cell degeneration in the cerebellar cortex of the adult rat. Neuroscience, 53: 759-778. Ruigrok, T.J.H. and Cella, E (1995) Precerebellar nuclei and red nucleus. In: G. Paxinos (Ed.), The Rat Nervous System, Vol. I11. Brain Stem and Cerebellum, Sydney: Academic Press, pp. 281-286. Sato, Y., Kawasaki, T. and Ikarashi, K. (1983) Afferent projections from the brainstem to the floccular three zones in cats. II. Mossy fiber projections. Brain Res., 272: 37-48. Scheibel, M.E. and Scheibel, A.B. (1954) Observations on the intracortical relations of the climbing fibers of the cerebellum. A Golgi study. J. Comp. Neurol., 101: 733-763. Schild, R.E (1970) On the inferior olive of the albino rat. J. Comp. Neurol., 140: 255-260. Shambes, G.M., Beermann, D. H. and Welker, W. (1978) Multiple tactile areas in cerebellar cortex: another patchy cutaneous projection to granule cell columns in rats. Brain Res., 157: 123-128. Shinoda, Y. (1999) Visualization of the entire trajectory of long axons of single mammalian CNS neurons. Brain Res. Bull., 50: 387-388. Shinoda, Y., Sugiuchi, Y. and Futami, T. (1993) Organization of excitatory inputs from the cerebral cortex to the cerebellar dentate nucleus. Can. J. Neurol. Sei., 20 (Suppl. 3): S19$28. Shinoda, Y., Izawa, Y., Sugiuchi Y. and Futami, T. (1997) Functional significance of excitatory projections from the precerebellar nuclei to interpositus and dentate nucleus neurons for mediating motor, premotor and parietal cortical inputs. In: C.I. deZeeuw, P. Strata, and J. Voogd (Eds), The Cerebellum: from Structure to Control Progress in Brain Research, Vol. 114, Amsterdam: Elsevier, pp. 193-207. Shinoda, yr., Ohgaki, T. and Futami, T. (1986) The morphology of single lateral vestibulospinal tract axons in the lower cervical spinal cord of the cat. J. Comp. NeuroL, 249: 226-241. Shinoda, Y., Yokota, J. and Futami, T. (1981) Divergent projection of individual corticospinal axons to motoneurons of multiple muscles in the monkey. Neurosci. Lett., 23: 7-12. Shinoda, Y., Sugiuchi, Y. and Futami, T. (1987) Excitatory inputs to cerebellar dentate nucleus neurons from the cerebral cortex in the cat. Exp. Brain Res., 67: 299-315. Shinoda, Y., Sugiuchi, Y., Futami, T. and Izawa, R. (1992) Axon collaterals of mossy fibers from the pontine nucleus in the cerebellar dentate nucleus. J. Neurophysiol., 67: 547-560. Sugihara, I., Wu, H. and Shinoda, Y. (1996) Morphology of axon collaterals of single climbing fibers in the deep cerebellar nuclei of the rat. Neurosci. Lett., 217: 33-36. Sugihara, I., Wu, H.-S. and Shinoda, Y. (1997) Projection of climbing fibers originating from single olivocerebellar neurons in the rat cerebellum. Soc. Neurosci. Abstr., 23: 1830. Sugihara, I., Wu, H. and Shinoda, Y. (1999) Morphology of single olivocerebellar axons labeled with biotinylated dextran amine in the rat. J. Comp. NeuroL, 414: 131-148. Van der Want, J.J.L. and Voogd, J. (1987) Ultrastructural identification and localization of climbing fiber terminals in
186 the fastigial nucleus of the cat. J. Comp. NeuroI., 258: 81-90. Van der Want, J.J.L., Wiklund, L., Guegan, M., Ruigrok, T. and J. Voogd (1989) Anterograde tracing of the rat olivocerebellar system with Phaseolus vulgaris leucoagglutinin (PHA-L). Demonstration of climbing fiber collateral innervation of the cerebellar nuclei. J. Comp. NeuroL, 288: 1-18. Voogd, J. (1995) Cerebellum. In: G. Paxinos (Ed.), The Rat Nervous System, 2nd edn. Sydney: Academic Press, pp. 309-350.
Voogd, J., Jaarsma, D. and Marani, E. (1996) The cerebellum, chemoarchitecture and anatomy. In: L.W. Swanson, A. Bjsrklund, and T. Hskfelt (Eds), Integrated Systems of the
CNS, Part 111. Cerebellum, Basal Ganglia, Olfactory System. Handbook of Chemical Neuroanatomy Vol. 12, Amsterdam: Elsevier, pp. 1-369. Wu, H., Sugihara, I. and Shinoda, Y. (1999) Projection patterns of single mossy fibers originating from the lateral reticular nucleus in the rat cerebellar cortex and nuclei. J. Comp. Neurol., 411:97-118.