- Email: [email protected]
The inferior olivary connections to the cerebellum in the rat studied by retrograde axonal transport of horseradish peroxidase
Brain Resrorch Bulktin,
Vol. 5, pp. 267-275. Printed in the U.S.A.
The Inferior Olivary Connections to the Cerebellum in the Rat Studied by Retrograde Axonal Transport of Horseradish Peroxidase PATRICIA Department
of Anatomy,
A. BROWN
University of Auckland School of Medicine, Auckland,
New Zealand
Received 31 January 1980 BROWN, P. A. Thr inferior olivary connrctioms to the cerc~hellum in the rat studied by retrograde axonal transport of peroxidase. BRAIN RES. BULL. 5(3)267-275, 1980.-Olivocerebellar projections were investigated in the rat using retrograde axonal transport of horseradish peroxidase. Discrete cell groups of the inferior olive were labelled, subsequent to injections in the paravermal region, the vermis, or the caudolateral hemisphere. Injections in the midrostrocaudal third of the paravermal area resulted in labelling of cells in the medial accessory olive (MAO), in cell group “b” at caudal levels, and in its lateral portion at mid-rostrocaudal levels. The rostra1 pole of the principal olive (PO), the dorsal accessory olive (DAO), and the dorsomedial cell column, were heavily labelled. By comparison, caudal paravermal injections resulted in labelling in the medial part of the mid-rostrocaudal levels of the MAO, but not in its caudal portion. The PO lamellae were labelled in their lateral half, excluding the lateral bend connecting them. Injections slightly lateral within this paravermal area gave no caudal MAO labelling, but did label cells in segments of both PO lamellae, medial to those in the previous case. From vermal injections, cell groups “b” and “c” of the caudal MAO were labelled, but no labelled cells were present in the PO. Subsequent to injections in the paramedian lobule, cells in the dorsal lamella of the PO were labelled. No cells of the MAO were labelled. These results are discussed in terms of specific labelling patterns and the general concepts of organization presently held for the olivocerebellar system.
horseradish
Inferior olive Horseradish peroxidase Paramedian Paravermal Vermal Retrograde axonal transport
Rat Crus I
Medial accessory olive Crus II Contralateral
IMPAIRMENTS of the locomotor system, such as those examples seen in Parkinson’s disease [IO], intention tremor, and akinesia, stem from abnormal functioning of the motor systems-the cerebellar system and the interdependent pyramidal and extrapyramidal systems. Functionally, the inferior olive serves as an important correlation center in the medulla, for these latter two systems, relaying impulses which ascend the spinal funiculi from cutaneous mechanoreceptors, joint afferents, Golgi tendon organs, and visceral mechanoreceptors, to the cerebellum. Partial or complete ablation of the inferior olive, in the cat and rabbit, causes disturbances of motor control similar to those seen in cerebellar disease [3]. Anatomically, the inferior olivary complex projects in an organized way to the cerebellum [3,21]. Using retrograde axonal transport of horseradish peroxidase (HRP), Brodal and his colleagues [4, 5, 20, 241 studied projections of the principal olive and the accessory olivary nuclei to various subdivisions of the cat cerebellum. No comparable studies have appeared in the literature, on the rat olivary projections. Using anterograde axonal transport of radioactive amino acids, Chan-Palay ct al. [ 121 have mapped the overall projection of the olive to the cerebellum in the rat. They determined that olivary fiber terminals in the cerebellum end in parasagittal strips, which have no relationship to the boundaries of the transverse fissures (see also
Copyright
0 1980 ANKHO
International
Dorsal accessory olive Ipsilateral
1281). Chan-Palav et al., however, did not provide detail of the-point-to-point organization of the olivocerebellar projections. Preliminary work in the rat indicates that discrete cell groups in the olive project to vermal and hemispheric regions of the cerebellar cortex [8]. The present HRP study in the rat has been undertaken to provide more detailed information on the area-to-area relationship between the inferior olivary complex and the cerebellum. The clearcut labelling of cells with HRP, and possible estimation of area of uptake are advantages well-suited to such a study [27,30]. The present investigation has been carried out with these questions in mind: (1) Do any cell groups within the rat inferior olive project to specific areas of the cerebellar cortex? (2) Are olivocerebellar projections in the rat similar to those in the cat, and if so, are there major differences in olivary projections to such areas as the vermis, the paravermal region, and the hemispheric areas of the rat cerebellum? (3) Is there any evidence of an ipsilateral olivary connection in the rat? METHOD Sixteen
adult male albino rats (Wistar), of 275-325 g, were
obtained from the Animal Laboratories, Auckland Medical School. Two animals received multiple injections, while all others were given single injections of 0.1-0.2 ~1 of a 30-50%
Inc.-0361-9230/80/030267-09$01.40/O
3hH
horseradish peroxidase solution (Sigma Chemical Co., St. Louis, MO-Grade VI) in 0.9% saIine, with a Hamilton-type syringe (Scientific Glass Engineering Pty., Ltd., North Melbourne, Australia). Sodium pentobarbitone was the anaesthetic used (4.5 m&IO0 8). The positions of the injection sites were varied in 0.5 mm increments, lateral from the midline, in conjunction with posterior displacement from lambda. with increments of the same magnitude; using stereotaxic procedures. Animals were allowed to survive for 24 hr. after which they were given an overdose of anaesthetic, perfused via the aorta with 150 ml of Ringer’s solution, to which 0.025 ml of heparin (~la~~~vans Medical, Ltd., Liverpool, EnglanbSOOO units/ml) was added per liter, followed by 500 ml of fixative (0.4% (wiv) paraformaldehyde and 1.25% (v/v) ~utaraldehyd~, in 0.1 M pho~hate buffer. pH 7.35) 1221. The brains were removed and fixed overnight in fresh fixative. The following day, the brains were transferred to three changes of 0.1 M phosphate buffer, pH 7.4. over a period of eight hours, with the last change containing 5% (w/v) sucrose. The brains were left overnight in the lasr solution. Frozen sections 40 urn thick were taken. Brains were sectioned either coronally or sagittally. Sections were reacted with 3,3’ diaminobenzidine (Sigma) and hydrogen peroxide, in a sequential way, as described by Graham and Kamovsky [IT]. Reacted sections were then mounted from 0.5% aqueous gelatin, one series being counterstained lightly with cresyl violet, and one adjacent series left unst~ned. Both series were prepared for darkfield microscopy. In addition, another adjacent series of unreacted sections was stained with cresyl violet at the normal intensity, for brightfield microscopy.
Following injections of small amounts of HRP. by pressure injection, into the cerebellar cortex of sixteen rats. consistent patterns of labelled cells were found in the inferior olive. In 75% of the cases, the area of dense reaction product at the injection site was restricted to the cortex, and did not involve the subcortical white matter. Laterally or subcorticaIly. small areas of faint reaction product were considered to be regions of secondary spread of HRP-doubtfuf as areas of uptake 1301. Projections from the olive were studied in three areas of the rat cerebellum-(l) the Ed-rost~aud~ and caudal paravermal region of the hemisphere; (2) the vermis; and (3) the paramedian lobule of the caudolaterdl hemisphere. Representative cases from each area will be presented.
Mid-rostrocaud~ti third. In case 829 (Fig. I), an injection was made in the paravermal portion of crus I. The cerebellar cortex, which was stained with HRP, atso included the lateral portion of vermal lobule VII. No diffusion of marker across the midline was present. At caudal levels, cell group “b” f t9] of the cont~ate~l medial accessory olive was labelled. Numerous Iabelled cells were seen in the lateral part of the mid-rostrocaudal levels of the same subnucleus. The entire rostra1 two-thirds of both the principal olive. and the dorsal accessory olive, were labelled. as well as the dorsomedial cell column. There was a strong contralateral component to the projection, but a small ipsilateral component was definitely present, as seen from labeiied cells in the rostral pole of the complex (Fig. 1).
BROWN
FIG. IA. Case 8%. Segittal section of. the rat cerebellum, with injection site in black, indicating area of WRP uptake. drawn from camert lucida projection.
FIG. IB. Dorsal view of rat cerebellum, with the injection sate or black. Vermai lobules IV-IX are sbwn. S-lobulus simplex’ PM~~~~rn~ian lobule.
~‘~tuditl rhird. Injections in this area are represented by case 828 (Fig. 2). in which the spread of the marker at the injection site involved the paraverfjlal portions of crus II, and the pa~rn~~ Iobuie. There was no diffusion of HRP over the midline. Labelling in the inferior olive was similar tc the previous case, but with some significant differences. There was no labelling of the caudal medial accessory olive. However, in the mid-rostrocaudaI levels of this nucleus. labelled cells were present in a rather more medial positior (Fig. 23, in com~~son to case 824 (Fig. I). Comparablr levels of the principal olive were not IabeIled compfetelylabelled cells being present only in the lateral half of both the dorsal and the ventral lam&c, but not present in the FRI lateral bend or in the extreme medial potions. Labelling ir the rostral pole of the olivary complex was comparable to tht: previous case. in&ding a few cells in the ipsilateral side. In a similar experiment, case 832 (Fig. 31, the injectior was made 1 mm laterally to the position in case 828 (Fig. 2) having the same posterior displacement from lambda. The injection again involved both the pa~e~an lo&de and CIW 11, and a smctll part of crus I. As irt case 828 (Fig. 2), then was no labelling in the caudal medial accessory olive. Ia belled cells in the principzd olive were rest&ted to a mofi medial segment, in both the dorsal and ventral lamcllae. Thl
OLIVOCEREBELLAR
CAUDAl
CONNECTIONS
269
IN THE RAT BY HRP
I I
FIG. 1D. 2. Inferior olivary complex of the rat near the rostra1 pole. Subsequent to an injection in the same paravermal area as above, HRP-labelled cells are present in the contralateral principal olive, the dorsal accessory olive, and the dorsomedial cell column. Labelled cells are also present in the ipsilateral dorsal accessory olive (cells are circled). Darkfield microscopy. X150.
ROSTRA1 FIG. IC. Findings in case 829. Representative transverse sections through the inferior olive of the rat. Sections are approximately 300 pm apart (levels through extreme poles of complex not included). POd=principal olive, dorsal lamella; POv=principal olive, ventral lamella; MAO=medial accessory olive; DAO=dorsal accessory olive: dmcc=dorsomedial cell column; shaded area=pyramidal tract. Right side is contralateral to the injection site.
FIG. 2A. Case 828. Transverse section of the rat cerebellum, the injection site in black, indicating area of HRP uptake.
with
FIG. ID. 1. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to an injection in the mid-rostrocaudal third of the paravermal region of the hemisphere, HRP-labelled cells are present in the contralateral principal olive, the dorsal accessory olive, and the lateral portion of the medial accessory olive. Darkfield microscopy. X150. FIG. 2B. Dorsal view of rat cerebellum, breviations as in Fig. 1.
injection site in black. Ab-
CAUDAL
I I
FIG. 3A. Case 832. Sagittal section of the rat cerebellum. with In.jection site in black, indicating area of HRP uptake.
ROSTRA1 FIG. 2C. Findings in case 828. See Fig. plex components.
1 for inferior olivary com-
FIG. 3B. Dorsal view of rat cerebellum. injection site in black. Abbreviations as in Fig. I.
FIG. 2D. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to an injection in the caudal third of the paravermal region of the hemisphere. HRP-labelled cells are present in the contralateral principal olive. in the lateral half of both lamellac. Darktield microscopy. X 150.
dorsal accessory olive was labelled only in its rostra1 lateral aspect. A small group of cells in the contralateral rostra1 pole of the olivary complex was labelled, in addition to a few ceils in the ipsilateral side, with each labelled area similar in location to the above case.
Injc*ction.\
in the Vcrmis
tSi.r C‘u.sc~.\ I
As a result of multiple midline injections in case 849 (Fig. 4). 0.25 mm laterally. and I mm apart rostrocaudally (six
injections), involving lobules V-VIII; two cell groups were labelled in the caudal medial accessory olive-one intermediate, and one medial (cell groups “b” and “c”. respectively-[19]). The remaining five cases were cut sagittally. not transversely as in case 849, to determine ,the rostrocaudal extent of the labelling in the entire complex. Labelling was present, in each of these cases. in the medial accessory nucleus.
Subsequent to injections involvmg the lateral paramedian lobule, as in case 827 (Fig. 5). labelled cells appeared only in the dorsal lamella of the principal olive, at mid-roctrocaudal levels. Multiple injections were made in the caudal third ot the hemisphere, in one animal. involving the paramedian lobule (case 850. Fig. 6). Three injecttons were made. each placed 4.5 mm posterior to lambda. and at intervals of 2. 3. and 4 mm from the midline. Labelled cells were present in the dorsal arm of the principal olive. and a few cells were presenl in cell groups “h” and “c” of the caudal medial accessory olive.
OLIVOCEREBELLAR
CAUDAL
CONNECTIONS
IN THE RAT BY HRP
271
I I
FIG. 4A. Case 849. Transverse section of the rat cerebellum, with injection site in black, indicati~ area of HRP uptake.
ROSTRA1 FIG. 3C. Findings in case 832. Refer to Fig. 1 for inferior olivary components.
FIG. 4B. Dorsal view of rat cerebellum, Abbreviations as in Fig. 1,
CAUDAL
injection sites in black.
I I
FIG. 3D. Inferior olivary complex of the rat at mid-rostro~udal levels. Subsequent to an injection 1 mm lateral to that of case 828, HRP-labelled cells are present in the contralateral principal olive, in a position medial to HRP-labelled cells seen in case 828 (Fig. 2). Darkfield microscopy. X 150.
Transport
qf’ HRP to Non-Olivup
Loci
Bilateral labelling in the pons was present from all injections made, both hemispheric and vermai. More cells were
iabelled contralateral to the injection site, but ipsilateral labelling was also substantial. The specific patterns of these connections have been presented in abstract form [l 11, and will be reported, in detail, separately. The reticulotegmental pontine nucleus (Rtp) was bilaterally labelled from all injections, except from those in the far
ROSI’RAL FIG. 4C. Findings in case 849. Refer to Fig. 1 for inferior olivary components.
CAUDAL
I I
FIG. 4D. Caudal medial accessory olive of the rat. Subsequent IO multiple injections in the vermis, HRP-labelled cells are numerous in the contralateral cell group “c”. and some are nresent in cell -erouo* .‘b”. Darkfield micro&ody. X 150.
ROSTRA1 FIG. 5C. Findings in case 827. Refer to Fig. complex components.
I for inferior olivary
FIG. 5A. Case 827. Transverse section of the rat cerebellum, with injection site in black, indicating area of HRP uptake.
FIG. SD. Inferior olivary complex of the rat at mid-rostrocaudai levels. Subsequent to an injection in the lateral paramedian lobule. HRP-labelled cells are present in the dorsal lamella of the principal olive. Darkfield microscopy. X 150.
FIG. 53. Dorsal view of rat cerebelJum, injection site in black. Abbreviations as in Fig. 1.
lateral hemispherefrom which there was no labelling; in contrast to the results noted in the pons. Injections specifically in the vermis resulted in labelling in the medial and dorsolateral aspects of Rtp. bilaterally. Retrograde cell degeneration of Rtp after vermal lesions confirms this pattern of connections [9]. The lateral reticular nucleus of the reticular formation was heavily labelled, ipsilaterally; and fewer labelled cells were seen in the corresponding contraiaterai nucleus, as a result of iqjections in the rostral portion of the rostrocaudal extent of the paravermal area of the cerebellum. However.
OLIVOCEREBELLAR
CONNECTIONS
273
IN THE RAT BY HRP
FIG. 6A. Case 850. Transverse section of the rat cerebellum, injection sites in black, indicating HRP uptake.
with
FIG. 6D. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to multiple injections in the paramedian lobule, HRP-labelled cells are present in the dorsal lamella of the principal
olive. Darkfield microscopy. X150.
FIG. 6B. Dorsal view of rat cerebellum, Abbreviations as in Fig. 1.
injection sites in black.
I
CAUDAL
I
subsequent to injections in the posterior portion of the paravermal strip, and the far lateral hemisphere, no labelled cells were seen in this nucleus. Injections in the vermis tended to result in bilateral labelling in the rostra1 pole of the lateral reticular nucleus. Cells in the principal vestibular nuclei of both hemispheres were labelled as a result of injections in the vermal and paravermal areas, but no vestibular labelling was present in cases of injections in the far lateral hemisphere. Heavy labelling was seen in the ipsilateral sensory nucleus V, when injections were made in the posterior region of crus II. A few contralateral cells were labelled as well. No cells in this nucleus were labelled from injections made in the anterior paravermal area, or in the vermis. Bilateral labelling was found in the gigantocellular nucleus of the reticular formation, resulting from injections in the vermis. DISCUSSION
I
~
:<$$$.:.:.:A..‘.*
*.
I i *
...
>..’
.*.+:
>.>.&.:X..
I .:.:.:+:.:
. . . . . . . . . .. . . . .. .
. ‘~::::::~:~::::::.:.’. .
ROSTRA1 FIG. 6C. Findings in case 8.50. Refer to Fig. 1 for inferior olivary complex components.
Portions of the inferior olivary complex have been shown, in this horseradish peroxidase study, to project to particular regions of the cerebellar cortex. Three general patterns have been noted in these projections. The vermis has been shown to receive projections basically from the accessory nuclei, while the lateral hemisphere receives projections only from the principal olive. The intermediate paravermal area, in contrast, receives projections from both principal and accessory nuclei. This overlap is not as extensive as it is in the cat [4, 5, 20, 241. Early studies reported a pattern of olive connections specific for either the vermis or hemisphere, based on clinical evidence [6,21], and comparative anatomical findings [23]. Recent neuroanatomical evidence, by Groenewegen and coworkers [17,181, has shown the organization of the cerebellum to consist of longitudinal, or parasagittal bands, based on fiber caliber, and has shown that the olivocerebellar terminations fall within these bands. They found that the vermal band was comprised of two subunits, and was the end point for fibers from the accessory olivary nuclei. Projections to the vermis have been considered among the phylogenetically more primitive connections
B KOWN
274
of the inferior olive [6,23]. Although the present study indicates that three zones can be described in the cerebellum of the rat, on the basis of olivary connections, more zones than this have been described in the cat and ferret [ 17,181, and numerous bands of terminals are seen in the rat cerebellum when the entire olive is injected with radioactive materials [ 121. These are most likely subdivisions of the three main zones. It is very probable that it is these subdivisions that are reflected in the present study, particularly in the instance of the projections of cells in adjacent segments of the dorsal and ventral lamellae of the principal olive, to adjacent strips of cortex in the paravermal area. Parcrvermal
Injections
The fact that the areas bordering the vermis were found to receive projections from the rostra1 pole of the inferior olivary complex-specifically, from the principal olive, dorsal accessory nucleus, and the dorsomedial cell column, is consistent with the idea that the rostra1 part of the complex is more “primitive”. For, as Brouwer [6] observed, it is this frontal portion of the olive which connects with the more primitive part of the cerebellum-the vermal area. The exclusive labelling of cell group “b”, in the caudal medial accessory olive, represents a specific projection from this portion of the olive, to more anterior parts of the paravermal area, and is consistent with results for the cat [ 1, 4, 201, and opossum 1251. Some differences between the origin of projections to the mid-rostrocaudaland those to the caudal-thirds of the paravermal strip were found. The caudal medial accessory nucleus does not appear to project to the caudal portions of the paravermal area. This could be the only paravermal area not receiving such fibers. Labelling of the medial portion of the medial accessory olive at midrostrocaudal levels, after caudal paravermal injections. and labelling of the lateral part, subsequent to injections in the more rostra1 part of the paravermal strip, may indicate a further subdivision of the olivocerebellar terminations in this area.
Labelling solely in the medial two-thirds of the caudal portion of the medial accessory olive, subsequent to vermal injections, may be due to the fact that the injections were restricted to lobules V-VIII. There has been some evidence that the anterior vermis in the rat receives from the lateral parts of the caudal medial accessory olive [26]. It has been demonstrated in the cat [20] that the dorsal accessory olive. as well as the medial accessory olive, connects with vernal lobule VI. It is important to note, however, that in the same study, labelling in the medial accessory olive occurred solely in the caudal portion. Information from all of the samples, in the present study in the rat, indicates that the medial-most part of the caudal medial accessory olive projects to the caudal part of the vermal cortex, and that the intermediate
part of the caudal medial accessory olive projects lo the mid-rostrocaudal portion of the cercbellar vermis (Fig. 4).
A highly specific projection from the dorsal lamella of the principal olive, to the lateral paramedian lobule. exists in the rat; as revealed by labelled cells in this lamella, subsequent to injections in the caudolateral hemisphere. The verity of this result is substantiated from both single and multiple injections in the hemisphere (Figs. 5 and 6), and is consistent with the results of studies in the cat [I.S]. Evidence from corticonuclear studies also indicates that it is the dorsal arm of the principal olive that receive\ projections from the part of the lateral deep cerebellar nucleus, which receives fibers from the cortex of the lateral cerebellar hemisphere 12.291. It is probable that the specificity of this anatomical connection is also characteristic of the physiological relationships involved in cerebellar control of olivary inputs. No labelling was present in the caudal medial accessory olive, from single injections in the lateral paramedian lobule. However, in the multiple injection case, which involved paravermal cortex as well as the lateral portion of the pammedian lobule (Fig. 6). heavy labelling in the caudal medial accessory olive was observed. This is ascribed to transport of HRP from the paravermal area. It appears possible for adjacent cortical strips to receive projections from unique portions of the olivary complex. Once again, this possibly reflects the division of the more “primitive” (vermal-paravermal) from the more “recent” (lateral hemisphere) connections of the olivocerebellar system.
lpsilateral labelling has been observed for cases involving injections in the paravermal region. The labelled cells. on the side ipsilateral to the injection. were located in the rostra1 pole of the inferior olivary complex. in the dorsal accessory olive. Since no HRP spread over the midline. it is assumed that these labelled cells represent an ipsilateral connection. The concept of an ipsilateral component for the olivocerebellar system is not new 1211. However. there has been little evidence to support this idea, especially in studies in the cat 1I. 13. 161. The fact that cerebella-olivary connections appear to be entirely contralateral in the cat [16], has given strength to the case against any ipsilateral reciprocals. However. ipsilateral connections of both the cerebella-olivary system (71 and olivo-cerebellar system I121 have been reported for the rat. Also, electrophysiological studies in the rat point to the existence of ipsilateral
present horseradish peroxidase of ipsilateral components.
components
1,141. The
studies support the concept
ACKNOWl.ELXEMtN’l’S This work was supported by the Me&cat Research Council New Zealand. I wish to thank Drs. John B. Carman and Richard M. Faull for their helpful comments on the manuscript.
01 I..
REFt?RENCE:S 1. Armstrong, D. M., R. J. Harvey and R. F. Schild. Topographical localization in the olive-cerebellar projection: An electrophysiological study in the cat. ./. camp. Nrrtrol. 154: 287-302. 1974.
2. Beitz, A. J. The topographical and dentate-olivary pathways 31.5. 1976.
organization of the olivo-dentate in the cat. Rrtrin Rrs. 115: 31 I-
OLIVOCEREBELLAR
CONNECTIONS
IN THE RAT BY HRP
3. Brodal, A. Expe~mentelie cerebellare 1940. 4. Brodal, A. cat studied horseradish
Untersuchungen iiber die olivoLokalisation. Z. ges. Neural. Psychiut. 169: l-153,
and F. Walberg. The olivocerebellar projection in the with the method of retrograde axonal transport of peroxidase. IV. The projection to the anterior lobe. J. camp. Neurol. 172: 85-108, 1977. 5. Brodal, A. and F. Walberg. The olivocerebellar projection in the cat studied with the method of retrograde axonal transport of horseradish peroxidase. VI. The projection onto longitudinal zones of the paramedian lobule. J. camp. Neurol. 176: 281-294, 1977. 6. Brouwer, B. Anatomische Untersuchung iiber das Kleinhim des Menschen. P.~v~~i~~~. neurof. 31. Amsf. 1: lw149, 1915. 7. Brown, J. T.,.V. Chan-Palay and S. L. Palay. A study of afferent input to the inferior olivary complex in the rat by retrograde axonal transport of horseradish peroxidase. J. camp. Neural. 176: l-22, 1977. 8. Brown, P. A. A horseradish peroxidase study of inferior olivary projections to the cerebellar hemisphere in the rat. (Abstract)f. Gnat. 130: 201, 1980. 9. Brown, P. A. and J. B. Carman. Cerebellar decussation offibres from the nucleus reticularis tegmenti pontis in the brain of the albino rat. Experienfia 34: 10391041, 1978. 10. Carman, J. B. Anatomic basis of surgical treatment of Parkinson’s disease. Nerr Enwl. J. Med. 279: 919-930. 1%8. 11. Carman, J. B. and P. A~:Brown. A horseradish peroxidase study of the projection upon the cerebellum of the nuclei pontis and nucleus reticularis tegmenti pontis in the rat. (Abstract)J. Anaf. 128: 658-659, 1979. 12. Chan-Palay, V., S. L. Palay, J. T. Brown and C. Van Itallie. Sagittal organization of olivocerebellar and reticulocerebellar projections: Autoradiographic studies with “SS-methionine. Exp. Brait: Res. 30: 561-576, 1977. 13. Courville, J. and F. Faraco-Cantin. On the origin of the climbing fibers of the cerebellum. An experimental study in the cat with an autoradiographic tracing method. Neuroscience 3: 797-809, 1978.
14. Goodman, D. C. and J. T. Simpson, Jr. Functional localization in the cerebellum of the albino rat. Expr Neural. 3: 174188, 1961. 1.5. Graham, R. C., Jr. and M. J. Karnovsky. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Hisfochem. Cyfochem. 14: 291-302, 1966. 16. Graybiel, A. M., H. J. W. Nauta, R. J. Lasek and W. J. H. Nauta. A cerebeilo-olivary pathway in the cat. An expe~mental study using autoradiographic tracing techniques. Brain Res. 58: 205-211, 1973. 17. Groenewegen, H. J. and J. Voogd. The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of cat cerebellum. J. camp. Neural. 174: 417-488. 1977.
275
18. Groenewegen, H. J., J. Voogd and S. L. Freedman. The parasagittal zonation within the olivocerebellar projection. II. Climbing fiber distribution in the intermediate and hemispheric parts of cat cerebellum. J. camp. Neural. 183: 551-602, 1979. 19. Gwyn, D. G., G. P. Nicholson and B. A. Flumerfelt. The inferior olivary nucleus of the rat: A light and electron microscopic study. J. camp. Neural. t74: 489-520, 1977. 20. Hoddevik. G. H.. A. Brodal and F. Walbere. The olivocerebellar projection in the cat studied with the m;thod of retrograde axonal transport of horseradish peroxidase. III. The projection to the vermal visual area. J. romp. Neural. 169: 155-170, 1976. 21. Holmes, G. and T. G. Stewart. On the connection of the inferior olives with the cerebellum in man. Bruin 31: 125-137, 1908. 22. Kamovsky, M. J. A fo~aldehyde-glutar~dehyde fixative of high osmolality for use in electron microscopy. (Abstract) i. Cell Biol. 27: l37A, 1965. 23. Kooy, F. H. The inferior olive in vertebrates. Folk Neuro-hiol. 10: 205-369, 1917. 24. Kotchabhakdi, N., F. Walberg and A. Brodal. The olivocerebellar projection in the cat studied with the method of retrograde axonal transport of horseradish peroxidase. VII. The projection to lobulus simplex, CNS I and 1I.J. camp. Ncrrrol. 182: 293-314, 1978. 25. Linauts, M. and G. F. Martin. The organization of olivocerebellar projections in the opossum, Didelphis virginiunu, as revealed by the retrograde transport of horseradish peroxidase. J. camp. Neurd. 179: 355382, 1978. 26. McGrane, M. K., D. J. Woodward, M. A. Eriksson, R. A. Bume and J. A. Saint-Cyr. The inferior olivary complex in the rat: Gross nuclear organization and topography of olivocerebellar projections. Sot. Neurosci. Ahsfr. 3: 59, 1977. 27. Nauta, H. J. W., M. B. Pritz and R. J. Lasek. Afferents to the rat caudoputamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method. Brain Res. 67: 219-238,
1974.
28. Oscarsson, 0. The sagittal organization of the cerebellar anterior lobe as revealed by the projection patterns of the climbing fiber system. In: Neurobiology qf Cerehellar Evolution and Devetopmenf, edited by R. Llinas. Chicago: Am. Med. Ass., 1969, pp. 525-537. 29. Tolbert, D. L., L. C. Massopust, M. G. Murphy and P. A. Young. The anatomical organization of the cerebella-olivary projection in the cat. J. camp. Neural. 170: 525-544, 1976. 30. Walberg, F., A. Brodal and G. H. Hoddevik. A note on the method of retrograde transport of horseradish peroxidase as a tool in studies of afferent cerebellar connections, particularly those from the inferior olive; with comments on the orthograde transport in Purkinje cell axons. Expi Bruin Res. 24: 383-401, 1976.
Vol. 5, pp. 267-275. Printed in the U.S.A.
The Inferior Olivary Connections to the Cerebellum in the Rat Studied by Retrograde Axonal Transport of Horseradish Peroxidase PATRICIA Department
of Anatomy,
A. BROWN
University of Auckland School of Medicine, Auckland,
New Zealand
Received 31 January 1980 BROWN, P. A. Thr inferior olivary connrctioms to the cerc~hellum in the rat studied by retrograde axonal transport of peroxidase. BRAIN RES. BULL. 5(3)267-275, 1980.-Olivocerebellar projections were investigated in the rat using retrograde axonal transport of horseradish peroxidase. Discrete cell groups of the inferior olive were labelled, subsequent to injections in the paravermal region, the vermis, or the caudolateral hemisphere. Injections in the midrostrocaudal third of the paravermal area resulted in labelling of cells in the medial accessory olive (MAO), in cell group “b” at caudal levels, and in its lateral portion at mid-rostrocaudal levels. The rostra1 pole of the principal olive (PO), the dorsal accessory olive (DAO), and the dorsomedial cell column, were heavily labelled. By comparison, caudal paravermal injections resulted in labelling in the medial part of the mid-rostrocaudal levels of the MAO, but not in its caudal portion. The PO lamellae were labelled in their lateral half, excluding the lateral bend connecting them. Injections slightly lateral within this paravermal area gave no caudal MAO labelling, but did label cells in segments of both PO lamellae, medial to those in the previous case. From vermal injections, cell groups “b” and “c” of the caudal MAO were labelled, but no labelled cells were present in the PO. Subsequent to injections in the paramedian lobule, cells in the dorsal lamella of the PO were labelled. No cells of the MAO were labelled. These results are discussed in terms of specific labelling patterns and the general concepts of organization presently held for the olivocerebellar system.
horseradish
Inferior olive Horseradish peroxidase Paramedian Paravermal Vermal Retrograde axonal transport
Rat Crus I
Medial accessory olive Crus II Contralateral
IMPAIRMENTS of the locomotor system, such as those examples seen in Parkinson’s disease [IO], intention tremor, and akinesia, stem from abnormal functioning of the motor systems-the cerebellar system and the interdependent pyramidal and extrapyramidal systems. Functionally, the inferior olive serves as an important correlation center in the medulla, for these latter two systems, relaying impulses which ascend the spinal funiculi from cutaneous mechanoreceptors, joint afferents, Golgi tendon organs, and visceral mechanoreceptors, to the cerebellum. Partial or complete ablation of the inferior olive, in the cat and rabbit, causes disturbances of motor control similar to those seen in cerebellar disease [3]. Anatomically, the inferior olivary complex projects in an organized way to the cerebellum [3,21]. Using retrograde axonal transport of horseradish peroxidase (HRP), Brodal and his colleagues [4, 5, 20, 241 studied projections of the principal olive and the accessory olivary nuclei to various subdivisions of the cat cerebellum. No comparable studies have appeared in the literature, on the rat olivary projections. Using anterograde axonal transport of radioactive amino acids, Chan-Palay ct al. [ 121 have mapped the overall projection of the olive to the cerebellum in the rat. They determined that olivary fiber terminals in the cerebellum end in parasagittal strips, which have no relationship to the boundaries of the transverse fissures (see also
Copyright
0 1980 ANKHO
International
Dorsal accessory olive Ipsilateral
1281). Chan-Palav et al., however, did not provide detail of the-point-to-point organization of the olivocerebellar projections. Preliminary work in the rat indicates that discrete cell groups in the olive project to vermal and hemispheric regions of the cerebellar cortex [8]. The present HRP study in the rat has been undertaken to provide more detailed information on the area-to-area relationship between the inferior olivary complex and the cerebellum. The clearcut labelling of cells with HRP, and possible estimation of area of uptake are advantages well-suited to such a study [27,30]. The present investigation has been carried out with these questions in mind: (1) Do any cell groups within the rat inferior olive project to specific areas of the cerebellar cortex? (2) Are olivocerebellar projections in the rat similar to those in the cat, and if so, are there major differences in olivary projections to such areas as the vermis, the paravermal region, and the hemispheric areas of the rat cerebellum? (3) Is there any evidence of an ipsilateral olivary connection in the rat? METHOD Sixteen
adult male albino rats (Wistar), of 275-325 g, were
obtained from the Animal Laboratories, Auckland Medical School. Two animals received multiple injections, while all others were given single injections of 0.1-0.2 ~1 of a 30-50%
Inc.-0361-9230/80/030267-09$01.40/O
3hH
horseradish peroxidase solution (Sigma Chemical Co., St. Louis, MO-Grade VI) in 0.9% saIine, with a Hamilton-type syringe (Scientific Glass Engineering Pty., Ltd., North Melbourne, Australia). Sodium pentobarbitone was the anaesthetic used (4.5 m&IO0 8). The positions of the injection sites were varied in 0.5 mm increments, lateral from the midline, in conjunction with posterior displacement from lambda. with increments of the same magnitude; using stereotaxic procedures. Animals were allowed to survive for 24 hr. after which they were given an overdose of anaesthetic, perfused via the aorta with 150 ml of Ringer’s solution, to which 0.025 ml of heparin (~la~~~vans Medical, Ltd., Liverpool, EnglanbSOOO units/ml) was added per liter, followed by 500 ml of fixative (0.4% (wiv) paraformaldehyde and 1.25% (v/v) ~utaraldehyd~, in 0.1 M pho~hate buffer. pH 7.35) 1221. The brains were removed and fixed overnight in fresh fixative. The following day, the brains were transferred to three changes of 0.1 M phosphate buffer, pH 7.4. over a period of eight hours, with the last change containing 5% (w/v) sucrose. The brains were left overnight in the lasr solution. Frozen sections 40 urn thick were taken. Brains were sectioned either coronally or sagittally. Sections were reacted with 3,3’ diaminobenzidine (Sigma) and hydrogen peroxide, in a sequential way, as described by Graham and Kamovsky [IT]. Reacted sections were then mounted from 0.5% aqueous gelatin, one series being counterstained lightly with cresyl violet, and one adjacent series left unst~ned. Both series were prepared for darkfield microscopy. In addition, another adjacent series of unreacted sections was stained with cresyl violet at the normal intensity, for brightfield microscopy.
Following injections of small amounts of HRP. by pressure injection, into the cerebellar cortex of sixteen rats. consistent patterns of labelled cells were found in the inferior olive. In 75% of the cases, the area of dense reaction product at the injection site was restricted to the cortex, and did not involve the subcortical white matter. Laterally or subcorticaIly. small areas of faint reaction product were considered to be regions of secondary spread of HRP-doubtfuf as areas of uptake 1301. Projections from the olive were studied in three areas of the rat cerebellum-(l) the Ed-rost~aud~ and caudal paravermal region of the hemisphere; (2) the vermis; and (3) the paramedian lobule of the caudolaterdl hemisphere. Representative cases from each area will be presented.
Mid-rostrocaud~ti third. In case 829 (Fig. I), an injection was made in the paravermal portion of crus I. The cerebellar cortex, which was stained with HRP, atso included the lateral portion of vermal lobule VII. No diffusion of marker across the midline was present. At caudal levels, cell group “b” f t9] of the cont~ate~l medial accessory olive was labelled. Numerous Iabelled cells were seen in the lateral part of the mid-rostrocaudal levels of the same subnucleus. The entire rostra1 two-thirds of both the principal olive. and the dorsal accessory olive, were labelled. as well as the dorsomedial cell column. There was a strong contralateral component to the projection, but a small ipsilateral component was definitely present, as seen from labeiied cells in the rostral pole of the complex (Fig. 1).
BROWN
FIG. IA. Case 8%. Segittal section of. the rat cerebellum, with injection site in black, indicating area of WRP uptake. drawn from camert lucida projection.
FIG. IB. Dorsal view of rat cerebellum, with the injection sate or black. Vermai lobules IV-IX are sbwn. S-lobulus simplex’ PM~~~~rn~ian lobule.
~‘~tuditl rhird. Injections in this area are represented by case 828 (Fig. 2). in which the spread of the marker at the injection site involved the paraverfjlal portions of crus II, and the pa~rn~~ Iobuie. There was no diffusion of HRP over the midline. Labelling in the inferior olive was similar tc the previous case, but with some significant differences. There was no labelling of the caudal medial accessory olive. However, in the mid-rostrocaudaI levels of this nucleus. labelled cells were present in a rather more medial positior (Fig. 23, in com~~son to case 824 (Fig. I). Comparablr levels of the principal olive were not IabeIled compfetelylabelled cells being present only in the lateral half of both the dorsal and the ventral lam&c, but not present in the FRI lateral bend or in the extreme medial potions. Labelling ir the rostral pole of the olivary complex was comparable to tht: previous case. in&ding a few cells in the ipsilateral side. In a similar experiment, case 832 (Fig. 31, the injectior was made 1 mm laterally to the position in case 828 (Fig. 2) having the same posterior displacement from lambda. The injection again involved both the pa~e~an lo&de and CIW 11, and a smctll part of crus I. As irt case 828 (Fig. 2), then was no labelling in the caudal medial accessory olive. Ia belled cells in the principzd olive were rest&ted to a mofi medial segment, in both the dorsal and ventral lamcllae. Thl
OLIVOCEREBELLAR
CAUDAl
CONNECTIONS
269
IN THE RAT BY HRP
I I
FIG. 1D. 2. Inferior olivary complex of the rat near the rostra1 pole. Subsequent to an injection in the same paravermal area as above, HRP-labelled cells are present in the contralateral principal olive, the dorsal accessory olive, and the dorsomedial cell column. Labelled cells are also present in the ipsilateral dorsal accessory olive (cells are circled). Darkfield microscopy. X150.
ROSTRA1 FIG. IC. Findings in case 829. Representative transverse sections through the inferior olive of the rat. Sections are approximately 300 pm apart (levels through extreme poles of complex not included). POd=principal olive, dorsal lamella; POv=principal olive, ventral lamella; MAO=medial accessory olive; DAO=dorsal accessory olive: dmcc=dorsomedial cell column; shaded area=pyramidal tract. Right side is contralateral to the injection site.
FIG. 2A. Case 828. Transverse section of the rat cerebellum, the injection site in black, indicating area of HRP uptake.
with
FIG. ID. 1. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to an injection in the mid-rostrocaudal third of the paravermal region of the hemisphere, HRP-labelled cells are present in the contralateral principal olive, the dorsal accessory olive, and the lateral portion of the medial accessory olive. Darkfield microscopy. X150. FIG. 2B. Dorsal view of rat cerebellum, breviations as in Fig. 1.
injection site in black. Ab-
CAUDAL
I I
FIG. 3A. Case 832. Sagittal section of the rat cerebellum. with In.jection site in black, indicating area of HRP uptake.
ROSTRA1 FIG. 2C. Findings in case 828. See Fig. plex components.
1 for inferior olivary com-
FIG. 3B. Dorsal view of rat cerebellum. injection site in black. Abbreviations as in Fig. I.
FIG. 2D. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to an injection in the caudal third of the paravermal region of the hemisphere. HRP-labelled cells are present in the contralateral principal olive. in the lateral half of both lamellac. Darktield microscopy. X 150.
dorsal accessory olive was labelled only in its rostra1 lateral aspect. A small group of cells in the contralateral rostra1 pole of the olivary complex was labelled, in addition to a few ceils in the ipsilateral side, with each labelled area similar in location to the above case.
Injc*ction.\
in the Vcrmis
tSi.r C‘u.sc~.\ I
As a result of multiple midline injections in case 849 (Fig. 4). 0.25 mm laterally. and I mm apart rostrocaudally (six
injections), involving lobules V-VIII; two cell groups were labelled in the caudal medial accessory olive-one intermediate, and one medial (cell groups “b” and “c”. respectively-[19]). The remaining five cases were cut sagittally. not transversely as in case 849, to determine ,the rostrocaudal extent of the labelling in the entire complex. Labelling was present, in each of these cases. in the medial accessory nucleus.
Subsequent to injections involvmg the lateral paramedian lobule, as in case 827 (Fig. 5). labelled cells appeared only in the dorsal lamella of the principal olive, at mid-roctrocaudal levels. Multiple injections were made in the caudal third ot the hemisphere, in one animal. involving the paramedian lobule (case 850. Fig. 6). Three injecttons were made. each placed 4.5 mm posterior to lambda. and at intervals of 2. 3. and 4 mm from the midline. Labelled cells were present in the dorsal arm of the principal olive. and a few cells were presenl in cell groups “h” and “c” of the caudal medial accessory olive.
OLIVOCEREBELLAR
CAUDAL
CONNECTIONS
IN THE RAT BY HRP
271
I I
FIG. 4A. Case 849. Transverse section of the rat cerebellum, with injection site in black, indicati~ area of HRP uptake.
ROSTRA1 FIG. 3C. Findings in case 832. Refer to Fig. 1 for inferior olivary components.
FIG. 4B. Dorsal view of rat cerebellum, Abbreviations as in Fig. 1,
CAUDAL
injection sites in black.
I I
FIG. 3D. Inferior olivary complex of the rat at mid-rostro~udal levels. Subsequent to an injection 1 mm lateral to that of case 828, HRP-labelled cells are present in the contralateral principal olive, in a position medial to HRP-labelled cells seen in case 828 (Fig. 2). Darkfield microscopy. X 150.
Transport
qf’ HRP to Non-Olivup
Loci
Bilateral labelling in the pons was present from all injections made, both hemispheric and vermai. More cells were
iabelled contralateral to the injection site, but ipsilateral labelling was also substantial. The specific patterns of these connections have been presented in abstract form [l 11, and will be reported, in detail, separately. The reticulotegmental pontine nucleus (Rtp) was bilaterally labelled from all injections, except from those in the far
ROSI’RAL FIG. 4C. Findings in case 849. Refer to Fig. 1 for inferior olivary components.
CAUDAL
I I
FIG. 4D. Caudal medial accessory olive of the rat. Subsequent IO multiple injections in the vermis, HRP-labelled cells are numerous in the contralateral cell group “c”. and some are nresent in cell -erouo* .‘b”. Darkfield micro&ody. X 150.
ROSTRA1 FIG. 5C. Findings in case 827. Refer to Fig. complex components.
I for inferior olivary
FIG. 5A. Case 827. Transverse section of the rat cerebellum, with injection site in black, indicating area of HRP uptake.
FIG. SD. Inferior olivary complex of the rat at mid-rostrocaudai levels. Subsequent to an injection in the lateral paramedian lobule. HRP-labelled cells are present in the dorsal lamella of the principal olive. Darkfield microscopy. X 150.
FIG. 53. Dorsal view of rat cerebelJum, injection site in black. Abbreviations as in Fig. 1.
lateral hemispherefrom which there was no labelling; in contrast to the results noted in the pons. Injections specifically in the vermis resulted in labelling in the medial and dorsolateral aspects of Rtp. bilaterally. Retrograde cell degeneration of Rtp after vermal lesions confirms this pattern of connections [9]. The lateral reticular nucleus of the reticular formation was heavily labelled, ipsilaterally; and fewer labelled cells were seen in the corresponding contraiaterai nucleus, as a result of iqjections in the rostral portion of the rostrocaudal extent of the paravermal area of the cerebellum. However.
OLIVOCEREBELLAR
CONNECTIONS
273
IN THE RAT BY HRP
FIG. 6A. Case 850. Transverse section of the rat cerebellum, injection sites in black, indicating HRP uptake.
with
FIG. 6D. Inferior olivary complex of the rat at mid-rostrocaudal levels. Subsequent to multiple injections in the paramedian lobule, HRP-labelled cells are present in the dorsal lamella of the principal
olive. Darkfield microscopy. X150.
FIG. 6B. Dorsal view of rat cerebellum, Abbreviations as in Fig. 1.
injection sites in black.
I
CAUDAL
I
subsequent to injections in the posterior portion of the paravermal strip, and the far lateral hemisphere, no labelled cells were seen in this nucleus. Injections in the vermis tended to result in bilateral labelling in the rostra1 pole of the lateral reticular nucleus. Cells in the principal vestibular nuclei of both hemispheres were labelled as a result of injections in the vermal and paravermal areas, but no vestibular labelling was present in cases of injections in the far lateral hemisphere. Heavy labelling was seen in the ipsilateral sensory nucleus V, when injections were made in the posterior region of crus II. A few contralateral cells were labelled as well. No cells in this nucleus were labelled from injections made in the anterior paravermal area, or in the vermis. Bilateral labelling was found in the gigantocellular nucleus of the reticular formation, resulting from injections in the vermis. DISCUSSION
I
~
:<$$$.:.:.:A..‘.*
*.
I i *
...
>..’
.*.+:
>.>.&.:X..
I .:.:.:+:.:
. . . . . . . . . .. . . . .. .
. ‘~::::::~:~::::::.:.’. .
ROSTRA1 FIG. 6C. Findings in case 8.50. Refer to Fig. 1 for inferior olivary complex components.
Portions of the inferior olivary complex have been shown, in this horseradish peroxidase study, to project to particular regions of the cerebellar cortex. Three general patterns have been noted in these projections. The vermis has been shown to receive projections basically from the accessory nuclei, while the lateral hemisphere receives projections only from the principal olive. The intermediate paravermal area, in contrast, receives projections from both principal and accessory nuclei. This overlap is not as extensive as it is in the cat [4, 5, 20, 241. Early studies reported a pattern of olive connections specific for either the vermis or hemisphere, based on clinical evidence [6,21], and comparative anatomical findings [23]. Recent neuroanatomical evidence, by Groenewegen and coworkers [17,181, has shown the organization of the cerebellum to consist of longitudinal, or parasagittal bands, based on fiber caliber, and has shown that the olivocerebellar terminations fall within these bands. They found that the vermal band was comprised of two subunits, and was the end point for fibers from the accessory olivary nuclei. Projections to the vermis have been considered among the phylogenetically more primitive connections
B KOWN
274
of the inferior olive [6,23]. Although the present study indicates that three zones can be described in the cerebellum of the rat, on the basis of olivary connections, more zones than this have been described in the cat and ferret [ 17,181, and numerous bands of terminals are seen in the rat cerebellum when the entire olive is injected with radioactive materials [ 121. These are most likely subdivisions of the three main zones. It is very probable that it is these subdivisions that are reflected in the present study, particularly in the instance of the projections of cells in adjacent segments of the dorsal and ventral lamellae of the principal olive, to adjacent strips of cortex in the paravermal area. Parcrvermal
Injections
The fact that the areas bordering the vermis were found to receive projections from the rostra1 pole of the inferior olivary complex-specifically, from the principal olive, dorsal accessory nucleus, and the dorsomedial cell column, is consistent with the idea that the rostra1 part of the complex is more “primitive”. For, as Brouwer [6] observed, it is this frontal portion of the olive which connects with the more primitive part of the cerebellum-the vermal area. The exclusive labelling of cell group “b”, in the caudal medial accessory olive, represents a specific projection from this portion of the olive, to more anterior parts of the paravermal area, and is consistent with results for the cat [ 1, 4, 201, and opossum 1251. Some differences between the origin of projections to the mid-rostrocaudaland those to the caudal-thirds of the paravermal strip were found. The caudal medial accessory nucleus does not appear to project to the caudal portions of the paravermal area. This could be the only paravermal area not receiving such fibers. Labelling of the medial portion of the medial accessory olive at midrostrocaudal levels, after caudal paravermal injections. and labelling of the lateral part, subsequent to injections in the more rostra1 part of the paravermal strip, may indicate a further subdivision of the olivocerebellar terminations in this area.
Labelling solely in the medial two-thirds of the caudal portion of the medial accessory olive, subsequent to vermal injections, may be due to the fact that the injections were restricted to lobules V-VIII. There has been some evidence that the anterior vermis in the rat receives from the lateral parts of the caudal medial accessory olive [26]. It has been demonstrated in the cat [20] that the dorsal accessory olive. as well as the medial accessory olive, connects with vernal lobule VI. It is important to note, however, that in the same study, labelling in the medial accessory olive occurred solely in the caudal portion. Information from all of the samples, in the present study in the rat, indicates that the medial-most part of the caudal medial accessory olive projects to the caudal part of the vermal cortex, and that the intermediate
part of the caudal medial accessory olive projects lo the mid-rostrocaudal portion of the cercbellar vermis (Fig. 4).
A highly specific projection from the dorsal lamella of the principal olive, to the lateral paramedian lobule. exists in the rat; as revealed by labelled cells in this lamella, subsequent to injections in the caudolateral hemisphere. The verity of this result is substantiated from both single and multiple injections in the hemisphere (Figs. 5 and 6), and is consistent with the results of studies in the cat [I.S]. Evidence from corticonuclear studies also indicates that it is the dorsal arm of the principal olive that receive\ projections from the part of the lateral deep cerebellar nucleus, which receives fibers from the cortex of the lateral cerebellar hemisphere 12.291. It is probable that the specificity of this anatomical connection is also characteristic of the physiological relationships involved in cerebellar control of olivary inputs. No labelling was present in the caudal medial accessory olive, from single injections in the lateral paramedian lobule. However, in the multiple injection case, which involved paravermal cortex as well as the lateral portion of the pammedian lobule (Fig. 6). heavy labelling in the caudal medial accessory olive was observed. This is ascribed to transport of HRP from the paravermal area. It appears possible for adjacent cortical strips to receive projections from unique portions of the olivary complex. Once again, this possibly reflects the division of the more “primitive” (vermal-paravermal) from the more “recent” (lateral hemisphere) connections of the olivocerebellar system.
lpsilateral labelling has been observed for cases involving injections in the paravermal region. The labelled cells. on the side ipsilateral to the injection. were located in the rostra1 pole of the inferior olivary complex. in the dorsal accessory olive. Since no HRP spread over the midline. it is assumed that these labelled cells represent an ipsilateral connection. The concept of an ipsilateral component for the olivocerebellar system is not new 1211. However. there has been little evidence to support this idea, especially in studies in the cat 1I. 13. 161. The fact that cerebella-olivary connections appear to be entirely contralateral in the cat [16], has given strength to the case against any ipsilateral reciprocals. However. ipsilateral connections of both the cerebella-olivary system (71 and olivo-cerebellar system I121 have been reported for the rat. Also, electrophysiological studies in the rat point to the existence of ipsilateral
present horseradish peroxidase of ipsilateral components.
components
1,141. The
studies support the concept
ACKNOWl.ELXEMtN’l’S This work was supported by the Me&cat Research Council New Zealand. I wish to thank Drs. John B. Carman and Richard M. Faull for their helpful comments on the manuscript.
01 I..
REFt?RENCE:S 1. Armstrong, D. M., R. J. Harvey and R. F. Schild. Topographical localization in the olive-cerebellar projection: An electrophysiological study in the cat. ./. camp. Nrrtrol. 154: 287-302. 1974.
2. Beitz, A. J. The topographical and dentate-olivary pathways 31.5. 1976.
organization of the olivo-dentate in the cat. Rrtrin Rrs. 115: 31 I-
OLIVOCEREBELLAR
CONNECTIONS
IN THE RAT BY HRP
3. Brodal, A. Expe~mentelie cerebellare 1940. 4. Brodal, A. cat studied horseradish
Untersuchungen iiber die olivoLokalisation. Z. ges. Neural. Psychiut. 169: l-153,
and F. Walberg. The olivocerebellar projection in the with the method of retrograde axonal transport of peroxidase. IV. The projection to the anterior lobe. J. camp. Neurol. 172: 85-108, 1977. 5. Brodal, A. and F. Walberg. The olivocerebellar projection in the cat studied with the method of retrograde axonal transport of horseradish peroxidase. VI. The projection onto longitudinal zones of the paramedian lobule. J. camp. Neurol. 176: 281-294, 1977. 6. Brouwer, B. Anatomische Untersuchung iiber das Kleinhim des Menschen. P.~v~~i~~~. neurof. 31. Amsf. 1: lw149, 1915. 7. Brown, J. T.,.V. Chan-Palay and S. L. Palay. A study of afferent input to the inferior olivary complex in the rat by retrograde axonal transport of horseradish peroxidase. J. camp. Neural. 176: l-22, 1977. 8. Brown, P. A. A horseradish peroxidase study of inferior olivary projections to the cerebellar hemisphere in the rat. (Abstract)f. Gnat. 130: 201, 1980. 9. Brown, P. A. and J. B. Carman. Cerebellar decussation offibres from the nucleus reticularis tegmenti pontis in the brain of the albino rat. Experienfia 34: 10391041, 1978. 10. Carman, J. B. Anatomic basis of surgical treatment of Parkinson’s disease. Nerr Enwl. J. Med. 279: 919-930. 1%8. 11. Carman, J. B. and P. A~:Brown. A horseradish peroxidase study of the projection upon the cerebellum of the nuclei pontis and nucleus reticularis tegmenti pontis in the rat. (Abstract)J. Anaf. 128: 658-659, 1979. 12. Chan-Palay, V., S. L. Palay, J. T. Brown and C. Van Itallie. Sagittal organization of olivocerebellar and reticulocerebellar projections: Autoradiographic studies with “SS-methionine. Exp. Brait: Res. 30: 561-576, 1977. 13. Courville, J. and F. Faraco-Cantin. On the origin of the climbing fibers of the cerebellum. An experimental study in the cat with an autoradiographic tracing method. Neuroscience 3: 797-809, 1978.
14. Goodman, D. C. and J. T. Simpson, Jr. Functional localization in the cerebellum of the albino rat. Expr Neural. 3: 174188, 1961. 1.5. Graham, R. C., Jr. and M. J. Karnovsky. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Hisfochem. Cyfochem. 14: 291-302, 1966. 16. Graybiel, A. M., H. J. W. Nauta, R. J. Lasek and W. J. H. Nauta. A cerebeilo-olivary pathway in the cat. An expe~mental study using autoradiographic tracing techniques. Brain Res. 58: 205-211, 1973. 17. Groenewegen, H. J. and J. Voogd. The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of cat cerebellum. J. camp. Neural. 174: 417-488. 1977.
275
18. Groenewegen, H. J., J. Voogd and S. L. Freedman. The parasagittal zonation within the olivocerebellar projection. II. Climbing fiber distribution in the intermediate and hemispheric parts of cat cerebellum. J. camp. Neural. 183: 551-602, 1979. 19. Gwyn, D. G., G. P. Nicholson and B. A. Flumerfelt. The inferior olivary nucleus of the rat: A light and electron microscopic study. J. camp. Neural. t74: 489-520, 1977. 20. Hoddevik. G. H.. A. Brodal and F. Walbere. The olivocerebellar projection in the cat studied with the m;thod of retrograde axonal transport of horseradish peroxidase. III. The projection to the vermal visual area. J. romp. Neural. 169: 155-170, 1976. 21. Holmes, G. and T. G. Stewart. On the connection of the inferior olives with the cerebellum in man. Bruin 31: 125-137, 1908. 22. Kamovsky, M. J. A fo~aldehyde-glutar~dehyde fixative of high osmolality for use in electron microscopy. (Abstract) i. Cell Biol. 27: l37A, 1965. 23. Kooy, F. H. The inferior olive in vertebrates. Folk Neuro-hiol. 10: 205-369, 1917. 24. Kotchabhakdi, N., F. Walberg and A. Brodal. The olivocerebellar projection in the cat studied with the method of retrograde axonal transport of horseradish peroxidase. VII. The projection to lobulus simplex, CNS I and 1I.J. camp. Ncrrrol. 182: 293-314, 1978. 25. Linauts, M. and G. F. Martin. The organization of olivocerebellar projections in the opossum, Didelphis virginiunu, as revealed by the retrograde transport of horseradish peroxidase. J. camp. Neurd. 179: 355382, 1978. 26. McGrane, M. K., D. J. Woodward, M. A. Eriksson, R. A. Bume and J. A. Saint-Cyr. The inferior olivary complex in the rat: Gross nuclear organization and topography of olivocerebellar projections. Sot. Neurosci. Ahsfr. 3: 59, 1977. 27. Nauta, H. J. W., M. B. Pritz and R. J. Lasek. Afferents to the rat caudoputamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method. Brain Res. 67: 219-238,
1974.
28. Oscarsson, 0. The sagittal organization of the cerebellar anterior lobe as revealed by the projection patterns of the climbing fiber system. In: Neurobiology qf Cerehellar Evolution and Devetopmenf, edited by R. Llinas. Chicago: Am. Med. Ass., 1969, pp. 525-537. 29. Tolbert, D. L., L. C. Massopust, M. G. Murphy and P. A. Young. The anatomical organization of the cerebella-olivary projection in the cat. J. camp. Neural. 170: 525-544, 1976. 30. Walberg, F., A. Brodal and G. H. Hoddevik. A note on the method of retrograde transport of horseradish peroxidase as a tool in studies of afferent cerebellar connections, particularly those from the inferior olive; with comments on the orthograde transport in Purkinje cell axons. Expi Bruin Res. 24: 383-401, 1976.