The Localization of the Motor Neurons Innervating the Extraocular Muscles in the Oculomotor Nuclei of the Cat and Rabbit, Using Horseradish Peroxidase YOSHIO AKAGI Department of Anatomy, Kyoto Prefectural University of Medicine, Kyoto 602,Japan
ABSTRACT The localization of the motor neurons innervating the extraocular muscles in the oculomotor nuclei of adult cats and rabbits was investigated by means of retrograde labelling with horseradish peroxidase (HRP). The groups consisting of the motor neurons innervating an individual muscle lay in the nucleus as elongated columns extending in a longitudinal direction. The position of each group in the transverse section varied according to the rostro-caudal level of the nucleus. In the cat and rabbit, entire contralateral innervation of t h e superior rectus and entire ipsilateral innervation of three muscles of the inferior rectus, medial rectus and inferior oblique were similarly observed. However, the arrangement of individual motor groups differed considerably in both animals except for the group innervating the inferior rectus which was generally found in the ventral position running through the rostra1 two-thirds of the oculomotor nucleus. In the case of cats, the central caudal nucleus bilaterally innervated the levator palpebrae superioris. The motor neurons innervating this muscle in the rabbit (which lacks the central caudal nucleus) formed a rostro-caudal clubshaped column close to the group innervating the superior rectus. The aberrant cellular mass in the adjoining medial longitudinal fasciculus which belongs to the medial rectus appears to play an important role in the eye movement, because it commonly appears in various animals. Although many schematic diagrams showing the localization of the motor neurons innervating the extraocular muscles in the oculomotor nucleus have been proposed from time to time based on the methods of retrograde neuronal degeneration and electric stimulation, it has been generally accepted that Warwick's work ('53) using the monkey is the most satisfying model. Using retrograde neuronal degeneration, Tarlov proposed a representative model of the kitten oculomotor nucleus in which the medial rectus and inferior rectus groups were the reverse of those in the monkey (Tarlov and Tarlov, '71). It may be supposed t h a t there are considerable differences in the arrangement of the oculomotor nucleus among the species of experimental animals. In 1971, Kristensson and Olsson demonstrated t h a t horseradish peroxidase (HRP), after injection into the muscle of immature animals, was taken up by axons and then J. COMP. NEUR. (1978)181: 745-762.
transported retrogressively t o t h e corresponding motor neurons (Kristensson and Olsson, '71a,b; Kristensson e t al., '71). Utilizing this technique, Gacek ('74) reported localization of the motor neurons innervating extraocular muscles of the kitten. However there has been no study attempting the retrograde transport of HRP to the extraocular muscles of adult animals. The aim of the present work is to clarify whether there is a difference in the organization of t h e oculomotor nucleus between adult cats and rabbits when HRP is injected. MATERIALS A N D METHODS
Thirty-four adult cats (2.0-3.0 kg body weight) and 35 albino rabbits (2.0-2.5 kg body weight) of both sexes were used in this study. Under light nembutal anesthesia, the isolation of the superior rectus, the inferior rectus, the medial rectus and the inferior oblique was performed after a conjunctival incision, but
745
746
YOSHIO AKAGI
the levator palpebrae superioris was exposed after a parallel lid incision as far as t h e upper lid margin. All operative procedures were done only on t h e right side of t h e animals. In order to provide adequate exposure of the extraocular muscles, the partial collapse of t h e eye ball was performed by t h e aspiration of t h e aqueous humor with a syringe. Subsequently, t h e muscle was fully infiltrated with HRP (Sigma type VI) by t h e use of a microsyringe. The volumes used were as follows: a 3-5 p1 solution containing 1.5-2.5mg HRP in t h e cat, and a 7-10 jd solution containing 4-5mg H R P in t h e rabbit. The injected muscle was enveloped by parafilm coated with surgical cement, Alonalpha A (Sankyo Co.), to avoid t h e escape of HRP to any other muscle. After t h e survival time of two days, t h e brain of the animals were fixed by transcardiac perfusion with a solution of 1%glutaraldehyde and 1% paraformaldehyde in phosphate buffer to pH 7.4. The midbrain and brain stem were removed and immersed in t h e same fixative solution overnight. The tissue was then immersed for one day in 10%sucrose buffered with phosphate (pH 7.4). Frozen transverse serial sections, 40 p m in thickness, were cut with a freezing microtome a t right angles to t h e aqueduct, washed in 0.05 M Tris-HC1 buffer and incubated for 20 minutes a t room temperature in a medium containing hydrogen peroxidase and 3,3'-diaminobenzidine tetrahydrochloride (Graham and Karnovsky, '66). Simultaneously, noninjected muscles were also immersed in t h e same solution to ascertain whether or not HRP had escaped from treated muscle to other muscle. Sections were counterstained with 0.1% cresyl-violet. Every third section containing t h e oculomotor nucleus was photographed a t a magnification of 50 x . These sections were examined to compare them with t h e corresponding photograph. The state of t h e neurons was carefully observed under high magnification with a light microscope and the motor neurons showing HRP-positive reaction were marked on t h e photograph with asterisks. RESULTS
At first t h e normal features of t h e cat and rabbit oculomotor nuclei will be described together and then t h e experimental observations obtained from both animals will be separately described as follows.
The normal oculomotor nuclei of the cat and rabbit The nucleus (dorso-lateral and ventro-media1 nuclei) consisted of two columns of large multipolar cells of about 40 p m in diameter extending for about 4 mm in t h e cat and about 3 mm in t h e rabbit in a rostro-caudal direction. These two large-celled groups lay close to the dorsal region of the medial longitudinal fasciculus beneath the floor of t h e cerebral aqueduct in t h e central gray matter. The small-celled nucleus of Edinger-Westphal (nucleus accessorius autonomicus) lay in t h e space between t h e two lateral columns in t h e rostral two-thirds of t h e nucleus. In t h e cat, t h e central caudal nucleus, composed of medium-sized multipolar cells, lay anterior to t h e dorsal raphe nucleus. The lateral paired nuclei were almost fused with each other across t h e midline except for t h e rostral third and t h e caudal end of t h e nucleus. On t h e other hand, they were divisible into the ventro-medial and dorso-lateral subnuclei in t h e middle third of t h e nucleus. In the rabbit, t h e central caudal nucleus was never found. The paired nuclei were clearly separated from each other in all levels of t h e oculomotor nucleus and were not divisible into distinct subnuclei. In t h e caudal third of the oculomotor nuclei of both animals, a few multipolar neurons were seen scattered throughout the fibers of t h e medial longitudinal fasciculus which was apposed ventrolaterally to t h e paired nuclei. Although t h e trochlear nucleus was consistently situated in t h e caudal part of t h e oculomotor nucleus in both animals, i t was not difficult to distinguish t h e trochlear nucleus from t h e oculomotor nucleus since their positions were quite different when seen in transverse section. EXPERIMENTAL RESULTS
When H R P was injected into t h e extraocular muscles in both animals, t h e parent motor neurons were clearly labelled (fig. 1) and they formed a compact group corresponding with t h e individual muscle (fig. 2). However, there were some cases in which t h e motor neurons were insufficiently labelled despite the careful injection of HRP. In this study, t h e results from t h e animals whose motor neurons innervating t h e individual extraocular muscle were entirely labelled with t h e reaction products
747
THE LOCALIZATION OF THE OCULOMOTOR NEURONS TABLE 1
Number of experimental and of sufficiently labelled cases Cat Muscle injected
Number of total cases
Rabbit Sufficiently labelled
Number of total cases
Sufficiently labelled
Inferior rectus Medial rectus Superior rectus Inferior oblique Levator palpebrae superioris
were used to create an exact reconstruction of the oculomotor nucleus. The total number of experimental cases and the proportion sufficiently labelled is indicated in table 1. a. Cut It required, using photographs of every third section, an average of 31 photographic plates to make a reconstruction of the entire oculomotor nucleus of the cat. Therefore, the number of the photograph indicated the rostro-caudal level of the transverse section. The results concerned with each muscle innervated by the HRP-positive neurons in the nucleus will be mentioned first and then the representation of the cat oculomotor nucleus will be created by the integration of those observations. Inferior rectus The labelled motor neurons were seen in all levels of ipsilateral oculomotor nucleus except for its caudal end. They were never found in the contralateral side. In the rostral fifth of the nucleus, all nerve cells were the motor neurons innervating this muscle. The area consisting of the labelled cells was ventral in position in the rostral third (fig. 3) and occupied the ventro-lateral position in the caudal two-thirds of t h e nucleus (fig. 4 ) . The nurnber of labelled cells was at a maximum a t the level of the rostral third of the nucleus. Medial rectus The labelled nerve cells were present in all sections except in the rostral fifth of the oculomotor nucleus and were always observed ipsilaterally. In the rostral third, the group formed by them appeared in the most dorsal part of the nucleus although a group of unlabelled cells interposed between these groups and the Edinger-Westphal nucleus
(fig. 3). Their number was at a maximum a t the level of the middle of the oculomotor nucleus and they occupied the dorsolateral position. The number of these cells gradually decreased towards the caudal level and disappeared a t a level somewhat anterior to t h e caudal end. In addition to this group, the aberrant motor neurons within the ipsilateral medial longitudinal fasciculus were also labelled. Superior rectus The labelled cells were always found on the contralateral side. They were always aggregated in the medial position of the nucleus through the caudal two-thirds of the nucleus (fig. 4). This was quite different from the mode of innervation of other muscles. Inferior oblique The labelled cells were demonstrated ipsilaterally in the caudal two-thirds of the oculomotor nucleus and were never seen in the contralateral side. They appeared in the dorsomedial portion of t h e group innervating the medial rectus at t h e level of the rostral third and gradually occupied the narrow oblique area shown in figure 4. At the caudal end, they occupied the most lateral position of the nucleus. Their number was a t a maximum a t the level of t h e middle of the nucleus. However i t was still less than half of those found in the inferior rectus, medial rectus and superior rectus. Levator palpebrae superioris The labelled cells were found dispersed in the central caudal nucleus, distributed uniformly in this unpaired nucleus. Their number was about half the total cell population. The above results obtained by the application of HRP to individual extraocular muscle were summarized in the schematic diagram of
748
YOSHIO AKAGI
the cat oculomotor nucleus by the careful comparison of many photographic plates (figs. 8, 9). The motor columns innervating five muscles appeared to be clearly separated from each other. b. Rabbit The clear pattern of the distribution of the labelled and unlabelled neurons was demonstrated by 22 photographic plates taken a t intervals of 80 pm through the rostro-caudal extremities of the rabbit oculomotor nucleus. At first, their patterns of distribution will be described for the individual muscle injected with HRP and thereafter they will be summarized into a schematic diagram representing the neuron groups innervating each muscle in the nucleus. Inferior rectus The labelled nerve cells were always present ipsilaterally in all transverse sections except in the caudal end of the oculomotor nucleus. The motor neurons in the rostral end of the nucleus were entirely labelled. The group of nerve cells innervating t h i s muscle was located in the ventro-lateral position in the rostral two-thirds of the nucleus (figs. 6, 7). Labelled cells consisting of this group were maximum in number a t the level of the rostral third of the nucleus (fig. 5). Medial rectus The group of nerve cells innervating this muscle were located ipsilaterally in the caudal two-thirds of the oculomotor nucleus. At the level of the rostral third, this group appeared as a narrow vertical band of labelled cells in the most medial part of the nucleus (fig. 5 ) . Advancing towards the caudal end, they formed a broad area extending obliquely from a dorso-lateral to a ventro-medial direction. The labelled neurons were maximum in number a t the level of the middle of the nucleus (fig. 6). On the other hand, the mass of the nerve cells lying among the fibers of the ipsilateral medial longitudinal fasciculus was also labelled after the injection of HRP into this muscle (fig. 7). Superior rectus The group of nerve cells innervating this muscle were always located in the contralatera1 dorsal position through the caudal half of the oculomotor nucleus (figs. 6, 7). The num-
ber of labelled cells was a t a maximum a t the level of the caudal fifth of the nucleus. Inferior oblique The labelled nerve cells were seen ipsilaterally through the caudal two-thirds of the oculomotor nucleus. At the level of the rostral third, they lay in the dorso-medial portion of the nucleus, facing the Edinger-Westphal nucleus. Except for its rostral end, they formed a fairly compact group in the intermediate area of the nucleus (figs. 6, 7). In the caudal end of the nucleus, this group gradually diminished, occupying the ventral position. Levator palpebrae superioris A few labelled cells were always found in the contralateral side in the caudal third of the oculomotor nucleus. The group of nerve cells innervating this muscle lay as an elongated column of cells extending in a rostrocaudal direction in the ventro-lateral position adjacent to that of the superior rectus (fig. 7). The above mentioned results obtained for each of five muscles were carefully compared and summarized in the diagram showing the localization of the motor neurons innervating the extraocular muscles in the rabbit oculomotor nucleus (figs. 10, 11). DISCUSSION
Kristensson and Olsson ('71a) showed that HRP, after injection into gastrocnemius muscle of suckling mice, was transported to the corresponding motor neurons. Horseradish peroxidase which has been injected into the region of the peripheral nerve endings is taken up by the terminals and unmyelinated axons and transported retrogradely through the axoplasm to its cell soma (Zacks and Saito, '69; LaVail and LaVail, '74). There were also previous reports that the retrograde transport was greatly suppressed with maturity in the peripheral nerves of mice and rats (Kristensson and Olsson, '71b; Kristensson et al., '71). To explain this fact, they concluded that the perineurium in adult animals play a role as a barrier to penetration by HRP into the endoneurium and to contact with the axon. However numerous recent reports suggested that the HRP technique could also be applied to adult animals (Bunt et al., '74; Kuypers et al., '74; LaVail and LaVail, '74; Nauta et al., '74; Oppenheim and Heaton, '75). LaVail and LaVail ('74) have found the
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
retrograde transport of HRP from the retina to the isthmo-optic nucleus of 63-day-old chickens. Although Gacek ('74),mentioned that such retrograde transport was not present in animals after the age of two to three weeks, in this study when the injection of HRP was performed on the individual extraocular muscle of adult cats and rabbits, the corresponding motor neurons in the oculomotor nucleus were sufficiently and clearly labelled after a survival time of two days. The labelled nerve cells formed a compact group related to the motor column innervating each muscle. It is therefore to be emphasized that the manner of the uptake and retrograde transport of HRP injected into the muscle of adult animals is similar to that of immature animals, except for the difference in the dose for each muscle. Sufficient concentration of HRP on the nerve ending and naked fiber is essential for the detection of the localization of nerve cells innervating extraocular muscles in the oculomotor nucleus. Covering of the injected muscle with parafilm is also useful to get enough concentration of HRP without diffusion into other muscles. However there were also some cases in which only a few parent nerve cells were labelled in spite of the injection of an adequate amount of HRP. The reason for this fact is still unknown, but it may be presumed that the inflammatory response suppressed the uptake and retrograde transport of HRP. The small-sized gamma motor neurons were also labelled with HRP. They dispersed in the individual motor group. Strick et al. ('76) reported that the gamma motor neurons were labelled with HRP more intensely than the alpha ones in the cat gastrocnemius motor nucleus. In the oculomotor nucleus of the cat and rabbit, there were no differences in the intensity of the HRP reaction products between them. In the kitten, Gacek ('74)indicated that the optimal amount is approximately 0.1-0.5 mg of HRP for adequate labelling of nerve cells innervating individual muscles. In the present study using adult cats, however, the amount indispensable for getting the same result was about 1.5-2.5 mg of HRP. Whenever the same amount of HRP was injected into the muscles of the rabbit, the motor neuron of the rabbit oculomotor nucleus was hardly labelled. In the case of the rabbit, therefore, the optimal amount was approximately 4-5 mg of
749
HRP. It appeares t o be due to the differences in the size of the extraocular muscles and in the manner of the terminal bifurcation of the motor neurons between cats and rabbits. Many conflicting schematic diagrams of the oculomotor nucleus have been proposed one after another by several investigators. The classical diagram which received general acceptance for a long time was that of Bernheimer-Brower which was based on both the observations of chromatolysis in the oculomotor nucleus after the extirpation of the extraocular muscles in the monkey and on pathological studies in humans (Bernheimer, 1897; Brouwer, '18). They concluded that the motor columns innervating muscle were rostro-caudally arranged. Abd-el-Malek ('381, Szentagothai ('42) and Danis ('48)supported the theory of this rostro-caudal arrangement in the oculomotor nuclei of cats and dogs although in a quite different order. However, Bach (1899) and van Biervliet (1899) mentioned t h a t the localization of the groups of nerve cells for extraocular muscles was arranged in the dorso-ventral pattern in rabbit. One of the reasons for this confusion appears to depend on the fact that these experiments were performed in different animal species. All of these classical studies are summarized by the very conclusive review of Warwick ('53)on the monkey. These previous reports were mainly performed by the identification of retrograde degeneration after the resection of the nerve innervating individual extraocular muscle. In 1971, Tarlov adapted the retrograde neuronal changes after the nerve section to the oculomotor nucleus of newborn kittens and concluded that the location of the motor neurons of the medial rectus and the inferior rectus appear to be transposed in these two species in comparison with the observation of Warwick on the monkey. In the adult cat, he mentioned t h a t sufficient chromatolytic change of the nerve cells had not occurred even after the complete transection of the oculomotor nerve. It is difficult to observe the oculomotor organization from the appearance of sparse chromatolytic motor neurons. The method utilizing the retrograde transport of HRP, in contrast to nerve transection, is excellent in respect of the basis of the physiological phenomenon and in the ease of determining the survival time. The present model of the oculomotor organization of the adult cat resembles that of Gacek ("74)in the kitten ex-
7 50
YOSHIO AKAGI
cept for t h e rostro-caudal extension of t h e motor columns. In his model of t h e oculomotor nucleus, the inferior rectus and medial rectus neurons show a similar rostro-caudal extension, and the inferior oblique and superior rectus neurons do also. However, in this study, it was revealed that the inferior rectus neurons reached the most rostra1 level, and both t h e medial rectus and superior rectus neurons appeared at a level approximately 0.8 mm caudal to the tip of t h e inferior rectus neurons and t h a t the inferior oblique neurons appeared at t h e most caudal level. It was confirmed t h a t t h e levator palpebrae superioris was innervated bilaterally by t h e cells of t h e central caudal nucleus. The above results a r e in agreement with those of Bienfang ('68) and Naito et al. ('74) who investigated t h e cat oculomotor nucleus by means of t h e marking microelectrode technique. On the other hand, in t h e rabbit, there were no previous reports which sufficiently revealed the distribution of t h e neuron group through i t s oculomotor nucleus using t h e labelling method of HRP. This study proved t h a t the elongated columns extending in t h e longitudinal direction were constituted of t h e nerve cells innervating t h e individual extraocular muscle and their position varied according to the level of t h e oculomotor nucleus. In the present study of t h e rabbit oculomotor nucleus, t h e entire contralateral innervation of t h e superior rectus and t h e levator palpebrae superioris as well as t h e entire ipsilateral innervation of other muscles were investigated. This anatomical relationship, except for the levator palpebrae superioris, is similar to t h a t of both t h e cat and t h e monkey described by Tarlov and Tarlov ('711, Gacek ('74) a n d Warwick ('53). However, t h e localization of t h e motor neurons innervating the extraocular muscles of the rabbit was considerably different from either t h a t of t h e cat and of the monkey. In the cat and t h e monkey, t,he levator palpebrae superioris was bilaterally innervated by t h e central caudal nucleus which was first described by Panegrossi (1898) and Tsuchida ('06). In the rabbit, which lacks this subnucleus, it is interesting t h a t t h e motor neurons innervating t h e levator palpebrae superioris lie contralaterally in t h e small area close t o t h e group innervating t h e superior rectus. Warwick ('53) mentioned t h a t t h e aberrant
motor neurons in t h e adjoining medial longitudinal fasciculus belong to the medial rectus. In this study using t h e cat and the rabbit, a similar result was observed, namely, t h a t those motor neurons ipsilaterally innervated t h e medial rectus. The similar appearance of this mass of motor neurons in various species of animals suggests t h a t they may be significantly correlated with t h e function of t h e eye movement. LITERATURE CITED Abd-el-Malek, S. 1938 On t h e localization of the nerve centers of the extrinsic ocular muscles in the oculomotor nucleus. J. Anat. (London), 72: 518-523. Bach, L. 1899 Zur Lehre von den Augenmuskellahmungen und den Storungen der Pupillenbewegung. V. Gaefe's Arch. f. Ophthal., 47: 339-386 and 551-630. Berheimer, S. 1897 Experimentelle Studien zur Kenntniss der Innervation der inneren und ausseren vom Oculomotorius versorgten Muskeln des Auges. V. Graefe's Arch. f. Ophthal., 44: 481-525. Bienfang, D. C. 1968 Location of the cell bodies of the superior rectus and inferior oblique motoneurons in the cat. Exp. Neurol., 21: 455-466. van Biervliet, J. 1899 Noyau d'origine du nerf oculomoteur commun du lapin. La Cellule, 16: 7-29. Broiiwer, B. 1918 Klinisch-anatomische Untersuchung uber den Oculomotoriuskern. Ztschr. f. d. ges. Neur. u. Psychiat., 40: 152-193. Bunt, A. H., R. D. Lund and S. J. Lund 1974 Retrograde axonal transport of horseradish peroxidase by ganglion cells of the albino rat retina. Brain Res., 73: 215-228. Danis, P. C. 1948 The functional organization of the third nerve nucleus in t h e cat. Am. J . Ophthal., 31: 1122-1131. Gacek, R. R. 1974 Localization of neurons supplying the extraocular muscles in t h e kitten using horseradish peroxidase. Exp. Neurol., 44: 381-403. Graham, R. C., Jr., and M. J. Karnovsky 1966 The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Kristensson, K., and Y. Olsson 1971a Retrograde axonal transport of protein. Brain Res., 29: 363-365. 1971b The perineurium as a diffusion barrier to protein tracers. Differences between mature and immature animals. Acta Neuropath. (Berlin), 17: 127-138. Kristensson, K., Y. Olsson and J. Sjostrand 1971 Axonal uptake and retrograde transport of exogenous proteins in t h e hypoglossal nerve. Brain Res., 32: 399-406. Kuypers, H. G. J. M., J. Kievit and A. C. Groen-Klevant 1974 Retrograde axonal transport of horseradish peroxidase in rat's forebrain. Brain Res., 67: 211-218. LaVail, J. H., and M. M. LaVail 1974 The retrograde intraaxonal transport of horseradish peroxidase in the chick visual system: A light and electron microscopic study. J. Comp. Neur., 157: 303-358. Naito, H., K. Tanimura, N. Taga andY. Hosoya 1974 Microelectrode study on the subnuclei of the oculomotor nucleus in t h e cat. Brain Res., 81: 215-231. Nauta, H. J . , M. B. Pritz and R. J. Lasek 1974 Afferents t o the rat caudoputamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method. Brain Res., 67: 219-238.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Oppenheim, R., and M. B. Heaton 1975 The retrograde transport of horseradish peroxidase from t he developing limb of the chick embryo. Brain Res., 98: 291-302. Panegrossi, G. 1898 Contributo allo studio anatomafisiologico die centri dei nerve oculomotori dell’uomo. Ric. Lab. Anat. Norm. Univ. Roma., 6: 103.155. Strick, P. L., R. E. Burke, K. Kanda. C. C. Kim and B. Walmsley 1976 Differences between alpha and gamma motoneurons labeled with horseradish peroxidase by retrograde transport. Brain Res., 113: 582-588. Szentagothai, J. 1942 Die innere Gliederung des Oculomotoriuskernes. Arch. f. Psychiat., ZZ5: 127-135. Tarlov, E., and S. R. Tarlov 1971 The representation of ex-
751
traocular muscles in the oculomotor nuclei: Experimental studies in the cat. Brain Res., 34: 37-52. Tsuchida, U. 1906 Uber die Ursprungskerne der Augenbewegungen und uber die mit deisen in Beziehung stehenden Bahnen im Mittel-und Zwischernhirn. Arb. Hirnanat. Inst. in Zurich., 2: 1-205. Warwick, R. 1953 Representation of the extra-ocular muscles in the oculomotor nuclei of the monkey. J. Comp. New., 98: 449-503. Zachs, S. I., and A. Saito 1969 Uptake of exogenous horseradish peroxidase by coated vesicles in mouse neuromuscular junctions. J. Histochem. Cytochem., Z7: 161-170.
PLATE 1 EXPLANATION OF FIGURES
1 A light micrograph of a motor neuron containing HRP-positive granules in its perikaryon and axon. Cresyl violet stain. X 500. 2 A dark field light micrograph of the oculomotor nucleus after the injection of HRP
into extraocular muscles. Many labelled motor neurons form the compact group. X 140. 3 Transverse section of the cat oculomotor nucleus in its rostra1 third. At this level, the nucleus is divisible into the two subnuclei of the ventral and the dorso-lateral. Labelled cells (asterisks) appeared in the contralateral medial portion after careful injection of HRP into the superior rectus. IR, inferior rectus; MR, medial rectus. Cresyl violet stain. x 53.
752
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Yoshio Aka@
PLATE 1
753
PLATE 2 EXPLANATION OF FIGURES
754
4
Transverse section of the cat oculomotor nucleus in its caudal third. Labelled cells (asterisks) appeared in the ipsilateral narrow area after t h e careful injection of HRP into the inferior oblique. IR, inferior rectus; MR, medial rectus; SR, superior rectus. Cresyl violet stain. x 53.
5
Transverse section of the rabbit oculomotor nucleus in its rostra1 third. Labelled cells (asterisks) appeared in the ipsilateral medial portion after the injection of HRP into the medial rectus. IR, inferior rectus. Cresyl violet stain. X 56.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Ynshio Akagi
PLATE 2
755
PLATE 3 EXPLANATION OF FIGURES
6 Transverse section of the rabbit oculomotor nucleus in its half. Labelled cells (asterisks) appeared in the contralateral dorsal position after the injection of HRP into the superior rectus. IR, inferior rectus; MR, medial rectus; 10,inferior oblique. Cresyl violet stain. x 56.
7
756
Transverse section of the rabbit oculomotor nucleus in its caudal third. Labelled cells (asterisks) appeared in the ipsilateral intermediate area after the injection of HRP into t h e inferior oblique. Some motor neurons (arrows) belonging to the medial rectus lie among the fibers of t h e medial longitudinal fasciculus. IR, inferior rectus; MR. medial rectus; SR, superior rectus; LP, levator palpebrae superioris. Cresyl violet stain. X 56.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
PLATE 3
Yoshio Akagi
757
Abbreviations
Inf. rect.. Inferior rectus Med. rect , Medial rectus Sup. rect.. Superior rectus
Inf. obliq., Inferior oblique L.P.s., Levator palpebrae superioris E.W., Edinger Westphal nucleus
PLATE 4 EXPLANATION OF FIGURES
Figs. 8, 9 Schematic diagram showing the localization of the motor neurons innervating the extraocular muscles in the oculomotor nucleus of the cat.
8 Frontal view of the nucleus. The number above each drawing indicates the rostrocaudal level of the nucleus in transverse section. 9 Dorsal view and lateral view of the nucleus. An aberrant cellular mass belonging to the medial rectus is not recorded in the drawing of the lateral view.
758
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
PLATE 4
Yushio Akagi
759
PLATE 5 EXPLANATION OF FIGURES
Figs. 10, 11 Schematic diagram showing the localization of motor neurons innervating the extraocular muscles In the oculomotor nucleus of the rabbit. 10 Frontal view of the nucleus. The number above each drawing indicates the rostra-
caudal level of the nucleus in the transverse section. 11 Dorsal view and lateral view of the nucleus. An aberrant cellular mass belonging
to the medial rectus is not recorded in the drawing of the lateral view.
760
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Yoshio Akagi
PLATE 5
761
ABSTRACT The localization of the motor neurons innervating the extraocular muscles in the oculomotor nuclei of adult cats and rabbits was investigated by means of retrograde labelling with horseradish peroxidase (HRP). The groups consisting of the motor neurons innervating an individual muscle lay in the nucleus as elongated columns extending in a longitudinal direction. The position of each group in the transverse section varied according to the rostro-caudal level of the nucleus. In the cat and rabbit, entire contralateral innervation of t h e superior rectus and entire ipsilateral innervation of three muscles of the inferior rectus, medial rectus and inferior oblique were similarly observed. However, the arrangement of individual motor groups differed considerably in both animals except for the group innervating the inferior rectus which was generally found in the ventral position running through the rostra1 two-thirds of the oculomotor nucleus. In the case of cats, the central caudal nucleus bilaterally innervated the levator palpebrae superioris. The motor neurons innervating this muscle in the rabbit (which lacks the central caudal nucleus) formed a rostro-caudal clubshaped column close to the group innervating the superior rectus. The aberrant cellular mass in the adjoining medial longitudinal fasciculus which belongs to the medial rectus appears to play an important role in the eye movement, because it commonly appears in various animals. Although many schematic diagrams showing the localization of the motor neurons innervating the extraocular muscles in the oculomotor nucleus have been proposed from time to time based on the methods of retrograde neuronal degeneration and electric stimulation, it has been generally accepted that Warwick's work ('53) using the monkey is the most satisfying model. Using retrograde neuronal degeneration, Tarlov proposed a representative model of the kitten oculomotor nucleus in which the medial rectus and inferior rectus groups were the reverse of those in the monkey (Tarlov and Tarlov, '71). It may be supposed t h a t there are considerable differences in the arrangement of the oculomotor nucleus among the species of experimental animals. In 1971, Kristensson and Olsson demonstrated t h a t horseradish peroxidase (HRP), after injection into the muscle of immature animals, was taken up by axons and then J. COMP. NEUR. (1978)181: 745-762.
transported retrogressively t o t h e corresponding motor neurons (Kristensson and Olsson, '71a,b; Kristensson e t al., '71). Utilizing this technique, Gacek ('74) reported localization of the motor neurons innervating extraocular muscles of the kitten. However there has been no study attempting the retrograde transport of HRP to the extraocular muscles of adult animals. The aim of the present work is to clarify whether there is a difference in the organization of t h e oculomotor nucleus between adult cats and rabbits when HRP is injected. MATERIALS A N D METHODS
Thirty-four adult cats (2.0-3.0 kg body weight) and 35 albino rabbits (2.0-2.5 kg body weight) of both sexes were used in this study. Under light nembutal anesthesia, the isolation of the superior rectus, the inferior rectus, the medial rectus and the inferior oblique was performed after a conjunctival incision, but
745
746
YOSHIO AKAGI
the levator palpebrae superioris was exposed after a parallel lid incision as far as t h e upper lid margin. All operative procedures were done only on t h e right side of t h e animals. In order to provide adequate exposure of the extraocular muscles, the partial collapse of t h e eye ball was performed by t h e aspiration of t h e aqueous humor with a syringe. Subsequently, t h e muscle was fully infiltrated with HRP (Sigma type VI) by t h e use of a microsyringe. The volumes used were as follows: a 3-5 p1 solution containing 1.5-2.5mg HRP in t h e cat, and a 7-10 jd solution containing 4-5mg H R P in t h e rabbit. The injected muscle was enveloped by parafilm coated with surgical cement, Alonalpha A (Sankyo Co.), to avoid t h e escape of HRP to any other muscle. After t h e survival time of two days, t h e brain of the animals were fixed by transcardiac perfusion with a solution of 1%glutaraldehyde and 1% paraformaldehyde in phosphate buffer to pH 7.4. The midbrain and brain stem were removed and immersed in t h e same fixative solution overnight. The tissue was then immersed for one day in 10%sucrose buffered with phosphate (pH 7.4). Frozen transverse serial sections, 40 p m in thickness, were cut with a freezing microtome a t right angles to t h e aqueduct, washed in 0.05 M Tris-HC1 buffer and incubated for 20 minutes a t room temperature in a medium containing hydrogen peroxidase and 3,3'-diaminobenzidine tetrahydrochloride (Graham and Karnovsky, '66). Simultaneously, noninjected muscles were also immersed in t h e same solution to ascertain whether or not HRP had escaped from treated muscle to other muscle. Sections were counterstained with 0.1% cresyl-violet. Every third section containing t h e oculomotor nucleus was photographed a t a magnification of 50 x . These sections were examined to compare them with t h e corresponding photograph. The state of t h e neurons was carefully observed under high magnification with a light microscope and the motor neurons showing HRP-positive reaction were marked on t h e photograph with asterisks. RESULTS
At first t h e normal features of t h e cat and rabbit oculomotor nuclei will be described together and then t h e experimental observations obtained from both animals will be separately described as follows.
The normal oculomotor nuclei of the cat and rabbit The nucleus (dorso-lateral and ventro-media1 nuclei) consisted of two columns of large multipolar cells of about 40 p m in diameter extending for about 4 mm in t h e cat and about 3 mm in t h e rabbit in a rostro-caudal direction. These two large-celled groups lay close to the dorsal region of the medial longitudinal fasciculus beneath the floor of t h e cerebral aqueduct in t h e central gray matter. The small-celled nucleus of Edinger-Westphal (nucleus accessorius autonomicus) lay in t h e space between t h e two lateral columns in t h e rostral two-thirds of t h e nucleus. In t h e cat, t h e central caudal nucleus, composed of medium-sized multipolar cells, lay anterior to t h e dorsal raphe nucleus. The lateral paired nuclei were almost fused with each other across t h e midline except for t h e rostral third and t h e caudal end of t h e nucleus. On t h e other hand, they were divisible into the ventro-medial and dorso-lateral subnuclei in t h e middle third of t h e nucleus. In the rabbit, t h e central caudal nucleus was never found. The paired nuclei were clearly separated from each other in all levels of t h e oculomotor nucleus and were not divisible into distinct subnuclei. In t h e caudal third of the oculomotor nuclei of both animals, a few multipolar neurons were seen scattered throughout the fibers of t h e medial longitudinal fasciculus which was apposed ventrolaterally to t h e paired nuclei. Although t h e trochlear nucleus was consistently situated in t h e caudal part of t h e oculomotor nucleus in both animals, i t was not difficult to distinguish t h e trochlear nucleus from t h e oculomotor nucleus since their positions were quite different when seen in transverse section. EXPERIMENTAL RESULTS
When H R P was injected into t h e extraocular muscles in both animals, t h e parent motor neurons were clearly labelled (fig. 1) and they formed a compact group corresponding with t h e individual muscle (fig. 2). However, there were some cases in which t h e motor neurons were insufficiently labelled despite the careful injection of HRP. In this study, t h e results from t h e animals whose motor neurons innervating t h e individual extraocular muscle were entirely labelled with t h e reaction products
747
THE LOCALIZATION OF THE OCULOMOTOR NEURONS TABLE 1
Number of experimental and of sufficiently labelled cases Cat Muscle injected
Number of total cases
Rabbit Sufficiently labelled
Number of total cases
Sufficiently labelled
Inferior rectus Medial rectus Superior rectus Inferior oblique Levator palpebrae superioris
were used to create an exact reconstruction of the oculomotor nucleus. The total number of experimental cases and the proportion sufficiently labelled is indicated in table 1. a. Cut It required, using photographs of every third section, an average of 31 photographic plates to make a reconstruction of the entire oculomotor nucleus of the cat. Therefore, the number of the photograph indicated the rostro-caudal level of the transverse section. The results concerned with each muscle innervated by the HRP-positive neurons in the nucleus will be mentioned first and then the representation of the cat oculomotor nucleus will be created by the integration of those observations. Inferior rectus The labelled motor neurons were seen in all levels of ipsilateral oculomotor nucleus except for its caudal end. They were never found in the contralateral side. In the rostral fifth of the nucleus, all nerve cells were the motor neurons innervating this muscle. The area consisting of the labelled cells was ventral in position in the rostral third (fig. 3) and occupied the ventro-lateral position in the caudal two-thirds of t h e nucleus (fig. 4 ) . The nurnber of labelled cells was at a maximum a t the level of the rostral third of the nucleus. Medial rectus The labelled nerve cells were present in all sections except in the rostral fifth of the oculomotor nucleus and were always observed ipsilaterally. In the rostral third, the group formed by them appeared in the most dorsal part of the nucleus although a group of unlabelled cells interposed between these groups and the Edinger-Westphal nucleus
(fig. 3). Their number was at a maximum a t the level of the middle of the oculomotor nucleus and they occupied the dorsolateral position. The number of these cells gradually decreased towards the caudal level and disappeared a t a level somewhat anterior to t h e caudal end. In addition to this group, the aberrant motor neurons within the ipsilateral medial longitudinal fasciculus were also labelled. Superior rectus The labelled cells were always found on the contralateral side. They were always aggregated in the medial position of the nucleus through the caudal two-thirds of the nucleus (fig. 4). This was quite different from the mode of innervation of other muscles. Inferior oblique The labelled cells were demonstrated ipsilaterally in the caudal two-thirds of the oculomotor nucleus and were never seen in the contralateral side. They appeared in the dorsomedial portion of t h e group innervating the medial rectus at t h e level of the rostral third and gradually occupied the narrow oblique area shown in figure 4. At the caudal end, they occupied the most lateral position of the nucleus. Their number was a t a maximum a t the level of t h e middle of the nucleus. However i t was still less than half of those found in the inferior rectus, medial rectus and superior rectus. Levator palpebrae superioris The labelled cells were found dispersed in the central caudal nucleus, distributed uniformly in this unpaired nucleus. Their number was about half the total cell population. The above results obtained by the application of HRP to individual extraocular muscle were summarized in the schematic diagram of
748
YOSHIO AKAGI
the cat oculomotor nucleus by the careful comparison of many photographic plates (figs. 8, 9). The motor columns innervating five muscles appeared to be clearly separated from each other. b. Rabbit The clear pattern of the distribution of the labelled and unlabelled neurons was demonstrated by 22 photographic plates taken a t intervals of 80 pm through the rostro-caudal extremities of the rabbit oculomotor nucleus. At first, their patterns of distribution will be described for the individual muscle injected with HRP and thereafter they will be summarized into a schematic diagram representing the neuron groups innervating each muscle in the nucleus. Inferior rectus The labelled nerve cells were always present ipsilaterally in all transverse sections except in the caudal end of the oculomotor nucleus. The motor neurons in the rostral end of the nucleus were entirely labelled. The group of nerve cells innervating t h i s muscle was located in the ventro-lateral position in the rostral two-thirds of the nucleus (figs. 6, 7). Labelled cells consisting of this group were maximum in number a t the level of the rostral third of the nucleus (fig. 5). Medial rectus The group of nerve cells innervating this muscle were located ipsilaterally in the caudal two-thirds of the oculomotor nucleus. At the level of the rostral third, this group appeared as a narrow vertical band of labelled cells in the most medial part of the nucleus (fig. 5 ) . Advancing towards the caudal end, they formed a broad area extending obliquely from a dorso-lateral to a ventro-medial direction. The labelled neurons were maximum in number a t the level of the middle of the nucleus (fig. 6). On the other hand, the mass of the nerve cells lying among the fibers of the ipsilateral medial longitudinal fasciculus was also labelled after the injection of HRP into this muscle (fig. 7). Superior rectus The group of nerve cells innervating this muscle were always located in the contralatera1 dorsal position through the caudal half of the oculomotor nucleus (figs. 6, 7). The num-
ber of labelled cells was a t a maximum a t the level of the caudal fifth of the nucleus. Inferior oblique The labelled nerve cells were seen ipsilaterally through the caudal two-thirds of the oculomotor nucleus. At the level of the rostral third, they lay in the dorso-medial portion of the nucleus, facing the Edinger-Westphal nucleus. Except for its rostral end, they formed a fairly compact group in the intermediate area of the nucleus (figs. 6, 7). In the caudal end of the nucleus, this group gradually diminished, occupying the ventral position. Levator palpebrae superioris A few labelled cells were always found in the contralateral side in the caudal third of the oculomotor nucleus. The group of nerve cells innervating this muscle lay as an elongated column of cells extending in a rostrocaudal direction in the ventro-lateral position adjacent to that of the superior rectus (fig. 7). The above mentioned results obtained for each of five muscles were carefully compared and summarized in the diagram showing the localization of the motor neurons innervating the extraocular muscles in the rabbit oculomotor nucleus (figs. 10, 11). DISCUSSION
Kristensson and Olsson ('71a) showed that HRP, after injection into gastrocnemius muscle of suckling mice, was transported to the corresponding motor neurons. Horseradish peroxidase which has been injected into the region of the peripheral nerve endings is taken up by the terminals and unmyelinated axons and transported retrogradely through the axoplasm to its cell soma (Zacks and Saito, '69; LaVail and LaVail, '74). There were also previous reports that the retrograde transport was greatly suppressed with maturity in the peripheral nerves of mice and rats (Kristensson and Olsson, '71b; Kristensson et al., '71). To explain this fact, they concluded that the perineurium in adult animals play a role as a barrier to penetration by HRP into the endoneurium and to contact with the axon. However numerous recent reports suggested that the HRP technique could also be applied to adult animals (Bunt et al., '74; Kuypers et al., '74; LaVail and LaVail, '74; Nauta et al., '74; Oppenheim and Heaton, '75). LaVail and LaVail ('74) have found the
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
retrograde transport of HRP from the retina to the isthmo-optic nucleus of 63-day-old chickens. Although Gacek ('74),mentioned that such retrograde transport was not present in animals after the age of two to three weeks, in this study when the injection of HRP was performed on the individual extraocular muscle of adult cats and rabbits, the corresponding motor neurons in the oculomotor nucleus were sufficiently and clearly labelled after a survival time of two days. The labelled nerve cells formed a compact group related to the motor column innervating each muscle. It is therefore to be emphasized that the manner of the uptake and retrograde transport of HRP injected into the muscle of adult animals is similar to that of immature animals, except for the difference in the dose for each muscle. Sufficient concentration of HRP on the nerve ending and naked fiber is essential for the detection of the localization of nerve cells innervating extraocular muscles in the oculomotor nucleus. Covering of the injected muscle with parafilm is also useful to get enough concentration of HRP without diffusion into other muscles. However there were also some cases in which only a few parent nerve cells were labelled in spite of the injection of an adequate amount of HRP. The reason for this fact is still unknown, but it may be presumed that the inflammatory response suppressed the uptake and retrograde transport of HRP. The small-sized gamma motor neurons were also labelled with HRP. They dispersed in the individual motor group. Strick et al. ('76) reported that the gamma motor neurons were labelled with HRP more intensely than the alpha ones in the cat gastrocnemius motor nucleus. In the oculomotor nucleus of the cat and rabbit, there were no differences in the intensity of the HRP reaction products between them. In the kitten, Gacek ('74)indicated that the optimal amount is approximately 0.1-0.5 mg of HRP for adequate labelling of nerve cells innervating individual muscles. In the present study using adult cats, however, the amount indispensable for getting the same result was about 1.5-2.5 mg of HRP. Whenever the same amount of HRP was injected into the muscles of the rabbit, the motor neuron of the rabbit oculomotor nucleus was hardly labelled. In the case of the rabbit, therefore, the optimal amount was approximately 4-5 mg of
749
HRP. It appeares t o be due to the differences in the size of the extraocular muscles and in the manner of the terminal bifurcation of the motor neurons between cats and rabbits. Many conflicting schematic diagrams of the oculomotor nucleus have been proposed one after another by several investigators. The classical diagram which received general acceptance for a long time was that of Bernheimer-Brower which was based on both the observations of chromatolysis in the oculomotor nucleus after the extirpation of the extraocular muscles in the monkey and on pathological studies in humans (Bernheimer, 1897; Brouwer, '18). They concluded that the motor columns innervating muscle were rostro-caudally arranged. Abd-el-Malek ('381, Szentagothai ('42) and Danis ('48)supported the theory of this rostro-caudal arrangement in the oculomotor nuclei of cats and dogs although in a quite different order. However, Bach (1899) and van Biervliet (1899) mentioned t h a t the localization of the groups of nerve cells for extraocular muscles was arranged in the dorso-ventral pattern in rabbit. One of the reasons for this confusion appears to depend on the fact that these experiments were performed in different animal species. All of these classical studies are summarized by the very conclusive review of Warwick ('53)on the monkey. These previous reports were mainly performed by the identification of retrograde degeneration after the resection of the nerve innervating individual extraocular muscle. In 1971, Tarlov adapted the retrograde neuronal changes after the nerve section to the oculomotor nucleus of newborn kittens and concluded that the location of the motor neurons of the medial rectus and the inferior rectus appear to be transposed in these two species in comparison with the observation of Warwick on the monkey. In the adult cat, he mentioned t h a t sufficient chromatolytic change of the nerve cells had not occurred even after the complete transection of the oculomotor nerve. It is difficult to observe the oculomotor organization from the appearance of sparse chromatolytic motor neurons. The method utilizing the retrograde transport of HRP, in contrast to nerve transection, is excellent in respect of the basis of the physiological phenomenon and in the ease of determining the survival time. The present model of the oculomotor organization of the adult cat resembles that of Gacek ("74)in the kitten ex-
7 50
YOSHIO AKAGI
cept for t h e rostro-caudal extension of t h e motor columns. In his model of t h e oculomotor nucleus, the inferior rectus and medial rectus neurons show a similar rostro-caudal extension, and the inferior oblique and superior rectus neurons do also. However, in this study, it was revealed that the inferior rectus neurons reached the most rostra1 level, and both t h e medial rectus and superior rectus neurons appeared at a level approximately 0.8 mm caudal to the tip of t h e inferior rectus neurons and t h a t the inferior oblique neurons appeared at t h e most caudal level. It was confirmed t h a t t h e levator palpebrae superioris was innervated bilaterally by t h e cells of t h e central caudal nucleus. The above results a r e in agreement with those of Bienfang ('68) and Naito et al. ('74) who investigated t h e cat oculomotor nucleus by means of t h e marking microelectrode technique. On the other hand, in t h e rabbit, there were no previous reports which sufficiently revealed the distribution of t h e neuron group through i t s oculomotor nucleus using t h e labelling method of HRP. This study proved t h a t the elongated columns extending in t h e longitudinal direction were constituted of t h e nerve cells innervating t h e individual extraocular muscle and their position varied according to the level of t h e oculomotor nucleus. In the present study of t h e rabbit oculomotor nucleus, t h e entire contralateral innervation of t h e superior rectus and t h e levator palpebrae superioris as well as t h e entire ipsilateral innervation of other muscles were investigated. This anatomical relationship, except for the levator palpebrae superioris, is similar to t h a t of both t h e cat and t h e monkey described by Tarlov and Tarlov ('711, Gacek ('74) a n d Warwick ('53). However, t h e localization of t h e motor neurons innervating the extraocular muscles of the rabbit was considerably different from either t h a t of t h e cat and of the monkey. In the cat and t h e monkey, t,he levator palpebrae superioris was bilaterally innervated by t h e central caudal nucleus which was first described by Panegrossi (1898) and Tsuchida ('06). In the rabbit, which lacks this subnucleus, it is interesting t h a t t h e motor neurons innervating t h e levator palpebrae superioris lie contralaterally in t h e small area close t o t h e group innervating t h e superior rectus. Warwick ('53) mentioned t h a t t h e aberrant
motor neurons in t h e adjoining medial longitudinal fasciculus belong to the medial rectus. In this study using t h e cat and the rabbit, a similar result was observed, namely, t h a t those motor neurons ipsilaterally innervated t h e medial rectus. The similar appearance of this mass of motor neurons in various species of animals suggests t h a t they may be significantly correlated with t h e function of t h e eye movement. LITERATURE CITED Abd-el-Malek, S. 1938 On t h e localization of the nerve centers of the extrinsic ocular muscles in the oculomotor nucleus. J. Anat. (London), 72: 518-523. Bach, L. 1899 Zur Lehre von den Augenmuskellahmungen und den Storungen der Pupillenbewegung. V. Gaefe's Arch. f. Ophthal., 47: 339-386 and 551-630. Berheimer, S. 1897 Experimentelle Studien zur Kenntniss der Innervation der inneren und ausseren vom Oculomotorius versorgten Muskeln des Auges. V. Graefe's Arch. f. Ophthal., 44: 481-525. Bienfang, D. C. 1968 Location of the cell bodies of the superior rectus and inferior oblique motoneurons in the cat. Exp. Neurol., 21: 455-466. van Biervliet, J. 1899 Noyau d'origine du nerf oculomoteur commun du lapin. La Cellule, 16: 7-29. Broiiwer, B. 1918 Klinisch-anatomische Untersuchung uber den Oculomotoriuskern. Ztschr. f. d. ges. Neur. u. Psychiat., 40: 152-193. Bunt, A. H., R. D. Lund and S. J. Lund 1974 Retrograde axonal transport of horseradish peroxidase by ganglion cells of the albino rat retina. Brain Res., 73: 215-228. Danis, P. C. 1948 The functional organization of the third nerve nucleus in t h e cat. Am. J . Ophthal., 31: 1122-1131. Gacek, R. R. 1974 Localization of neurons supplying the extraocular muscles in t h e kitten using horseradish peroxidase. Exp. Neurol., 44: 381-403. Graham, R. C., Jr., and M. J. Karnovsky 1966 The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Kristensson, K., and Y. Olsson 1971a Retrograde axonal transport of protein. Brain Res., 29: 363-365. 1971b The perineurium as a diffusion barrier to protein tracers. Differences between mature and immature animals. Acta Neuropath. (Berlin), 17: 127-138. Kristensson, K., Y. Olsson and J. Sjostrand 1971 Axonal uptake and retrograde transport of exogenous proteins in t h e hypoglossal nerve. Brain Res., 32: 399-406. Kuypers, H. G. J. M., J. Kievit and A. C. Groen-Klevant 1974 Retrograde axonal transport of horseradish peroxidase in rat's forebrain. Brain Res., 67: 211-218. LaVail, J. H., and M. M. LaVail 1974 The retrograde intraaxonal transport of horseradish peroxidase in the chick visual system: A light and electron microscopic study. J. Comp. Neur., 157: 303-358. Naito, H., K. Tanimura, N. Taga andY. Hosoya 1974 Microelectrode study on the subnuclei of the oculomotor nucleus in t h e cat. Brain Res., 81: 215-231. Nauta, H. J . , M. B. Pritz and R. J. Lasek 1974 Afferents t o the rat caudoputamen studied with horseradish peroxidase. An evaluation of a retrograde neuroanatomical research method. Brain Res., 67: 219-238.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Oppenheim, R., and M. B. Heaton 1975 The retrograde transport of horseradish peroxidase from t he developing limb of the chick embryo. Brain Res., 98: 291-302. Panegrossi, G. 1898 Contributo allo studio anatomafisiologico die centri dei nerve oculomotori dell’uomo. Ric. Lab. Anat. Norm. Univ. Roma., 6: 103.155. Strick, P. L., R. E. Burke, K. Kanda. C. C. Kim and B. Walmsley 1976 Differences between alpha and gamma motoneurons labeled with horseradish peroxidase by retrograde transport. Brain Res., 113: 582-588. Szentagothai, J. 1942 Die innere Gliederung des Oculomotoriuskernes. Arch. f. Psychiat., ZZ5: 127-135. Tarlov, E., and S. R. Tarlov 1971 The representation of ex-
751
traocular muscles in the oculomotor nuclei: Experimental studies in the cat. Brain Res., 34: 37-52. Tsuchida, U. 1906 Uber die Ursprungskerne der Augenbewegungen und uber die mit deisen in Beziehung stehenden Bahnen im Mittel-und Zwischernhirn. Arb. Hirnanat. Inst. in Zurich., 2: 1-205. Warwick, R. 1953 Representation of the extra-ocular muscles in the oculomotor nuclei of the monkey. J. Comp. New., 98: 449-503. Zachs, S. I., and A. Saito 1969 Uptake of exogenous horseradish peroxidase by coated vesicles in mouse neuromuscular junctions. J. Histochem. Cytochem., Z7: 161-170.
PLATE 1 EXPLANATION OF FIGURES
1 A light micrograph of a motor neuron containing HRP-positive granules in its perikaryon and axon. Cresyl violet stain. X 500. 2 A dark field light micrograph of the oculomotor nucleus after the injection of HRP
into extraocular muscles. Many labelled motor neurons form the compact group. X 140. 3 Transverse section of the cat oculomotor nucleus in its rostra1 third. At this level, the nucleus is divisible into the two subnuclei of the ventral and the dorso-lateral. Labelled cells (asterisks) appeared in the contralateral medial portion after careful injection of HRP into the superior rectus. IR, inferior rectus; MR, medial rectus. Cresyl violet stain. x 53.
752
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Yoshio Aka@
PLATE 1
753
PLATE 2 EXPLANATION OF FIGURES
754
4
Transverse section of the cat oculomotor nucleus in its caudal third. Labelled cells (asterisks) appeared in the ipsilateral narrow area after t h e careful injection of HRP into the inferior oblique. IR, inferior rectus; MR, medial rectus; SR, superior rectus. Cresyl violet stain. x 53.
5
Transverse section of the rabbit oculomotor nucleus in its rostra1 third. Labelled cells (asterisks) appeared in the ipsilateral medial portion after the injection of HRP into the medial rectus. IR, inferior rectus. Cresyl violet stain. X 56.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Ynshio Akagi
PLATE 2
755
PLATE 3 EXPLANATION OF FIGURES
6 Transverse section of the rabbit oculomotor nucleus in its half. Labelled cells (asterisks) appeared in the contralateral dorsal position after the injection of HRP into the superior rectus. IR, inferior rectus; MR, medial rectus; 10,inferior oblique. Cresyl violet stain. x 56.
7
756
Transverse section of the rabbit oculomotor nucleus in its caudal third. Labelled cells (asterisks) appeared in the ipsilateral intermediate area after the injection of HRP into t h e inferior oblique. Some motor neurons (arrows) belonging to the medial rectus lie among the fibers of t h e medial longitudinal fasciculus. IR, inferior rectus; MR. medial rectus; SR, superior rectus; LP, levator palpebrae superioris. Cresyl violet stain. X 56.
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
PLATE 3
Yoshio Akagi
757
Abbreviations
Inf. rect.. Inferior rectus Med. rect , Medial rectus Sup. rect.. Superior rectus
Inf. obliq., Inferior oblique L.P.s., Levator palpebrae superioris E.W., Edinger Westphal nucleus
PLATE 4 EXPLANATION OF FIGURES
Figs. 8, 9 Schematic diagram showing the localization of the motor neurons innervating the extraocular muscles in the oculomotor nucleus of the cat.
8 Frontal view of the nucleus. The number above each drawing indicates the rostrocaudal level of the nucleus in transverse section. 9 Dorsal view and lateral view of the nucleus. An aberrant cellular mass belonging to the medial rectus is not recorded in the drawing of the lateral view.
758
THE LOCALIZATION OF THE OCULOMOTOR NEURONS
PLATE 4
Yushio Akagi
759
PLATE 5 EXPLANATION OF FIGURES
Figs. 10, 11 Schematic diagram showing the localization of motor neurons innervating the extraocular muscles In the oculomotor nucleus of the rabbit. 10 Frontal view of the nucleus. The number above each drawing indicates the rostra-
caudal level of the nucleus in the transverse section. 11 Dorsal view and lateral view of the nucleus. An aberrant cellular mass belonging
to the medial rectus is not recorded in the drawing of the lateral view.
760
THE LOCALIZATION OF THE OCULOMOTOR NEURONS Yoshio Akagi
PLATE 5
761
