The Synaptic Cluster (Glomerulus) in the Inferior Olivary Nucleus 1 2 3 JAMES S . KING Department of Anatomy, T h e Ohio State University, College of Medicine, 1645 Neil Avenue, Columbus, Ohio 43210
ABSTRACT This report describes the fine structural features and distribution of the synaptic cluster (glomerulus) within the inferior olivary nucleus of the opossum. The postsynaptic elements typically include spiny appendages and small diameter dendrites which exhibit attachment plaques and gap junctions. Profiles presynaptic to the central core of postsynaptic elements were differentiated on the basis of vesicle shape, vesicle size, as measured by a computer system, and junctional characteristics. Three categories of terminals with clear vesicles are present within the synaptic clusters in all nuclear divisions of the olive, whereas a fourth with large dense core vesicles is restricted primarily to the principal nucleus. The groups of pre and postsynaptic elements are surrounded by astrocytic lamellae and are most frequently encountered in the principal and rostra1 portions of the medial accessory nuclei. Possible identification of the sources of the synaptic components is discussed in relation to data available from Golgi impregnations, physiological reports and hodological evidence.
Synaptic clusters, termed glomeruli, endoplasmic reticulum. The central core within the inferior olivary nucleus were of the synaptic cluster is made up of first described by St. NEmeEek and Wolff spines, spiny appendages and small diam('69) in the cat. Previous electron micro- eter dendrites. Desmosome-like contacts scopic studies of the olive (Walberg, '63, are present, depending on the plane of sec'64, '65a,b, '66) had focused on the popu- tion, between all of the various postlations of synaptic endings which were de- synaptic components which are contacted fined as Gray's types I and I1 or as termi- by a peripheral rim of synaptic profiles. nals with elongated vesicles. Junctional The entire area is circumscribed by lamelzones similar to desmosomes were observed lar sheets of astrocytic processes. An addiwithin the glomeruli described by St. tional ultrastructural feature of the gloNEmeCek and Wolff ('69) and between ad- meruli, gap junctions, was described by jacent dendrites by Walberg ('64). The Sotelo et al. ('74) in the cat and subselatter author also emphasized the abun- quently by King et al. ( ' 7 5 ) in restricted dance of astrocytic lamellae interposed be- portions of the dorsal and medial accestween various elements of the neuropil. sory olivary nuclei. In our previous account (Bowman and Potential corollaries for the sources of preand post-synaptic profiles seen in eIectron King, ' 7 3 ) synaptic profiles within glommicrographs are provided by the Golgi eruli were differentiated by vesicle size and descriptions of Scheibel and Scheibel shape. At that time the typical mammalian ( ' 5 5 ) , Scheibel et al. ('56) and Ramon y nuclear subdivisions were not apparent in the opossum. Such divisions (principal, Cajal ('11). More recently, Bowman and King ( ' 7 3 ) dorsal and medial accessory nuclei) have described glomeruli in the inferior olive of been identified recently (Martin et al., '75) the opossum. In that species unique spiny This paper i s dedicated in memory of appendages arise from the dendrites of the my1 Dedication: brother Charles D. King, M.D, (1944-1975). ZThese results were presented in part at the 88th typical olivary neuron and when viewed in Meeting of the American Association of Anatomists, electron micrographs their profiles are seen Los Angeles, California. 3 Acknowledgment of Support: This investigation to contain vesicles and strands of smooth u a s supported by a U.S.P.H.S. Grant NS-08798. J. COMP. N E U R . , 165: 387-400.
387
388
JAMES S. KING
and a rapid, standardized method of determining the “size” of synaptic vesicles is now available in our laboratory. This measurement system was developed in the laboratory of Dr. W. M. Cowan at Washington University, St. Louis, Missouri (Cowan and Wann, ’73) and first utilized for this purpose by Rafols and Fox (’71/ ’72). Thus, more extensive sampling has been directed toward prescribed nuclear subdivisions for the following purposes : ( 1) to determine the uniformity of glomerular distribution, as well as the uniformity of the pre- and post-synaptic elements within the glomeruli, ( 2 ) to assess the populations of synaptic profiles by computer measurement of synaptic vesicle size and ( 3 ) to provide control data for future experiments designed to determine the origin of the resident presynaptic profiles. MATERIALS AND METHODS
The brains used for the present electron microscopic study were taken from nine adult opossums ( Didelphis marsupialis virg i n i a n a ) subsequent to the double aldehyde perfusion method described in detail by Mihailoff and King (’75). The inferior olive from each side of the brainstem was divided into four one millimeter slabs cut in the transverse plane. After osmium postfixation, dehydration and plastic embedding, one micron sections were cut from the face of each of the blocks. Each tissue slab was flat embedded to ensure a transverse plane of section. Thus, each thin section also was cut in the same plane from precise areas throughout the rostra1 to caudal extent of the principal, dorsal and medial accessory olivary nuclei. The reader should consult figures 1-21 in the article by Martin et al. (’75) for illustrations of the nuclear divisions of the opossum inferior olive. Sampling included the dorsal and ventral lamellae of the principal nucleus and the more caudal portion of the same nucleus where it is not divided into individual lamellae. Portions a, b and c of the medial accessory nucleus were examined, but because of its small size the cap of Kooy was not included. Both the direct and indirect (dorsal column) spinal receiving portions of the dorsal accessory nucleus were extensively sampled. Sections
picked u p on copper grids were examined with a Philips 300 electron microscope. In order to measure the size of synaptic vesicles, electron micrographs of glomeruli from both the accessory and principal nuclei were enlarged 100,000 times. The membranes of the vesicles i n each presynaptic profile were traced with a stylus attached to a cybergraphic tablet (Talos Systems Inc., Scottsdale, Arizona) which was interfaced with a PDP-12 computer. The computer program, provided by Dr. W. Maxwell Cowan, Washington University, St. Louis, determines the area of each traced vesicle in nmz and computes the mean, standard deviation, and variance of each sample of vesicles from an ending, as well as providing histograms of the vesicle areas (bin width = 200 nmz). The obvious values of this method of deriving synaptic vesicle size are speed and ease of measurement, as well as standardization of measurement . Five glomeruli were selected from the direct spinal portion of the dorsal accessory nucleus, five from the medial accessory (either part a or b ) and five from the principal nucleus (either the ventral lamellae or the caudal undivided part) for measurement of synaptic vesicles. Vesicles in all endings in each glomerulus were measured. Initial observations based on these measurements and other morphological features of the vesicle populations suggested that there were three categories in the glomeruli of the accessory nuclei (large spherical, small spherical and pleomorphic) and four categories in the principal nucleus (large spherical, small spherical, pleomorphic and dense core). In order to determine: ( 1 ) whether the categories of clear vesicles represent populations which are significantly different in size, as their ranges clearly overlap (figs. 8, IOA), and ( 2 ) whether the categories of terminals common to both the accessory and principal nuclei are from the same populations; four synaptic endings containing each of the vesicle types (large spherical, small spherical and pleomorphic) were randomly selected from glomeruli in the accessory nuclei and four endings of each type were selected from glomeruli in the principal nucleus. The area of thirty vesicles in each ending was measured and these data were
INFERIOR OLIVE SYNAPTIC CLUSTER
subjected to a tweway analysis of variance. The results of the statistical analysis revealed that there was a significant main effect for type of vesicle (F = 53.42, df = 2, p < 0.001), but no significant main effect for locus, i.e., principal vs. accessory, ( F = 3.74, df = 1, p > 0.05 and no significant interaction (F = 0.79, df = 2, p > 0.05). Post-hoc individual comparisons were made using the Newman-Kuels method and the results will be reported subsequently in the observations section. The statistical analysis was performed by the Biometrics Laboratory, Department of Preventive Medicine, The Ohio State University. OBSERVATIONS
The frequency with which olivary synaptic clusters are encountered in electron micrographs from the various nuclear divisions is markedly different. The impression one derives is that the principal nucleus including the dorsal and ventral lamellae and the more caudal undivided portion has many more glomeruli per unit area than any of the accessory nuclei. The sole exception is the rostral portion of the medial accessory nucleus (fig. 17A, Martin et al., '75), where many synaptic clusters also are present. Throughout the principal, dorsal and medial accessory olivary nuclei aggregates of pre- and post-synaptic profiles are partially or totally circumscribed by astrocytic Jamellae. In figure 1, which is taken from the principal olive, three such areas are demarcated by outlining the astrocytic processes in ink. The greatest and least dimensions of the synaptic cluster average 3.63 and 2.46 respectively, with little variation in size between the parts of the nucleus that were sampled. The central core of the synaptic cluster is made up of an average of six profiles (range 3-1 1 ). These post-synaptic elements (fig. 2 ) include small dendrites (dd), spines and excresences which have been termed spiny appendages (SA). When viewed in Golgi preparations (fig. 3, arrows), these dendritic appendages arise from intermediate and distal dendrites, and comparable profiles (fig. 2, SA) are readily identified in electron micrographs. The spiny appendages contain dense core vesicles, mitochondria, a flocculent ma-
389
terial, clear vesicles (fig. 2, block arrows), and strands of smooth endoplasmic reticulum (fig, 2, block arrows) enabling their identification (fig. 2, asterisks) even when they are not attached to their parent dendrite in the plane of section. Attachment plaques are frequently encountered (57% of the glomeruli) between the dendritic elements of the synaptic cluster with no predilection for a specific nuclear division (figs. 2, 11, circles). In addition, gap junctions are present (fig. 4, open block arrows; fig. 7). Figure 7 is a 100,000 x enlargement of the junction marked by block arrows in figure 4. Gap junctions are encountered most frequently in the various subdivisions of the accessory nuclei. In the accessory nuclei, the profiles that are presynaptic to the central core of postsynaptic elements are divisible into three categories based on the following criteria: ( 1 ) vesicle size, ( 2 ) vesicle shape and ( 3 ) junctional features. Synaptic profiles with spherical vesicles and Gray's type I active sites are divisible into two groups based on vesicle size (fig. 8). Small spherical vesicles (fig. 4, endings labelled A and B ) have a mean area of 908.78 nm', -C 222.18 nm'. If this area was considered to be the area of a circle the mean diameter would be 340 A. The synaptic endings with small spherical vesicles typically contact one post-synaptic profile via a single junctional area. They are infrequently presynaptic to a spiny appendage that displays a gap junction. The population of larger spherical vesicles (fig. 4, endings labelled C and D ) has a mean area of 1248.81 nm', -t 303.10 nm2.If this area were considered to be that of a circle it would have a diameter of 399 A. Terminals with large spherical vesicles may be presynaptic to a single post-synaptic element (fig. 4, ending labelled D) or to multiple dendritic processes some of which also exhibit gap junctions (fig. 4, ending labelled C ) . The presynaptic profiles with spherical vesicles number three or four per synaptic cluster (in a single section) with varying proportions of the large and small varieties. The third category of synaptic ending can be differentiated by its pleomorphic vesicles and intermediate type of synaptic junction (fig. 4, ending labelled E). The vesicles have a mean area of 897.89 nm',
390
JAMES S. KING
229.99 nm2, with an estimated diameter of 338 A. This axon terminal typically is presynaptic to at least three post-synaptic profiles, two of which often display a gap junction (fig. 4, ending labelled E). No more than two of these endings appear in a synaptic cluster and frequently only a single profile with pleomorphic vesicles is encountered. The differences in shape and size of the three vesicle populations can be better appreciated when the micrographs are enlarged 100,000 x (compare (figs. 5, 6, 9 ) as they were for all measurements with the computer system. The post-hoc comparisons revealed no significant difference between the size of the pleomorphic and small spherical vesicles (fig. 8 ) , however the large spherical vesicles were significantly different in size from the other two populations of vesicles at the 0.01 level. In the principal nucleus the same three types of presynaptic profiles are evident. The mean area for the small spherical vesicles is 910.49 nm2, +- 213.53 nm2, for the large spherical vesicles is 1336.04 nm2, 2 336.14 nmz, and for the pleomorphic 332.50 nm2. A vesicles is 995.66 nm2, frequency histogram (fig. 10A) depicts their distribution. If each of the three areal measurements were considered to be areas of circles the vesicles would have diameters of 340 A, 412 A and 356 A respectively. The two factor analysis of variance is supportive of the statement that the three vesicle population are similar within the accessory and principal nuclei, as no significant difference was found between the matched categories of vesicles in the principal and accessory nuclei (compare figs. 8 and 10A). The post-hoc analysis indicated that the significant difference in size was between the large spherical vesicles and the small spherical vesicles ( p < O.Ol), and between the large spherical and pleomorphic ( p < 0.01) vesicle populations. No significant size differential was evident between the small spherical vesicles and the pleomorphic vesicles. A feature unique to the synaptic clusters within the principal nucleus is a presynaptic profile with large dense core vesicles (fig. 11, enlarged in fig. 12). One such profile is typically present and presynaptic to as many as three of the dendritic or
*
spiny elements found in the center of the glomerulus. At 100,000 x magnification (fig. 1 2 ) the dense core vesicles, which average 22 per synaptic profile can be seen to better advantage and compared with clear spherical vesicles that also are present. The mean area of each of these vesicle populations is 4773.57 nm2, 1254.58 nm2 and 1097.03 nm2, 268.78 nmz respectively. If each area was considered to be that of a circle, their diameters in Angstroms would be 780 and 374. Their size distributions are illustrated by frequency histograms in figures 10B, C. In the example of this type of ending (figs. 11, 12) the dense core vesicles make up only 29% of the total vesicle population. As is the case in the accessory nuclei, terminals with spherical vesicles average three or four per synaptic cluster and contact one or two post-synaptic elements. A single synaptic profile with pleomorphic vesicles is typically seen although two may be present. Thus, the only presynaptic terminal within the glomeruli that is restricted in its nuclear distribution is the ending with dense core vesicles. Although characteristic of the principal nucleus in general, it is most frequently seen in the more caudal portion of the nucleus (fig. 13, Martin et al., ’75).
*
*
DISCUSSION
The definition of a glomerulus as “a circumscribed structural arrangement between a number of specific axon terminals and one or several dendrites” which “is separated from the environment by a capsule of glial processes” has been suggested by Szentagothai (’70). Moreover, various groups of synaptic elements, possessing similar and certain distinctive features, have been termed glomeruli by numerous authors. Descriptions include the : olfactory bulb (Pinching and Powell, ’71), medial geniculate body (Morest, ’71), dorsal cochlear nucleus (Kane, ’74), dorsal lateral geniculate nucleus (Famiglietti and Peters, ’72), pulvinar (Majorossy et al., ’65), nucleus gracilis (Rustioni and Sotelo, ’74), main sensory nucleus of V (Gobel and Dubner, ’69), spinal trigeminal nucleus (Gobel, ’74), substantia gelatinosa (Ralston, ’65), ventral basal thalamus (Ralston and Herman, ’69), lateral ves-
INFERIOR OLIVE SYNAPTIC CLUSTER
tibular nucleus (Sotelo and Palay, '70) and cerebellar cortex (Palay and ChanPalay, '73). To this growing list should be added the glomerulus of the inferior olivary nucleus based on the present account and the descriptions of St. NEmeEek and Wolff, '69; Bowman and King, '73 and Sotelo et al., '74. However, as serial sections have not been utilized to establish that the synaptic area under consideration is a diminutive ball (glomerulus), reference to the synaptic cluster of the inferior olive is perhaps more appropriate. Moreover, in any given plane of section the axonal and dendritic elements are not always encircled totally by astrocytic lamellae. Szentagothai ('70) suggests that the latter feature when observed warrants the use of this term. From the present results it appears that the synaptic cluster of the inferior olive is ubiquitous, although most frequently encountered in the principal and rostral part ( a ) of the medial accessory nuclei. Due to its limited size the cap of Kooy was not sampled. Mizuno et al. ('74) did not report synaptic clusters in their study, however, their primary goal was to identify pretecto-olivary axon terminals within the same region. It is clear that the central core of the synaptic cluster is dendritic and primarily derived from spiny appendages. The number of these spinous protrusions that contribute to one complex is not certain, but in favorable planes of section (fig. 1: dd) two dendrites can be seen to give rise to individual spiny appendages at the periphery of the cluster. Although the ultrastructur a1 basis of dendrodendritic transmission via chemical transmitters has been conclusively demonstrated in other glomeruli, gap junctions, the morphological corollary of electrotonic transmission, have not been reported. Gap junctions are more typical within synaptic clusters of the accessory olivary nuclei than in the principal nucleus. Electrophysiological evidence supportive of their suggested role in the synchronous firing of olivary neurons has been presented by Llinas et al. ('74). The axonal systems that influence the dendritic profiles within glomeruli are the subject of current investigation and in a preliminary report (King and Andrezik.
39 1
'75) destruction of the deep cerebellar nuclei has been shown to result in degenerating axon terminals within the olivary synaptic islands. Their precise nuclear origin and their distribution upon post-synaptic elements will be presented in a more detailed report (King et al., '76, submitted for publication). Aside from the input from the deep cerebellar nuclei (King et al.,'75), no experimental ultrastructural evidence is available that would suggest possible sources for the four distinct populations of presynaptic profiles identified in the present account. However, the multiple afferents to select parts of the opossum inferior olive (Martin et al., '75) provide a number of potential sources. Axon terminals of an inhibitory neuron activated by recurrent collaterals have been suggested as one possibiIity (LlinLs et al., '74). It is tempting to speculate, but by no means certain, that the synaptic profiles with pleomorphic vesicles would be inhibitory in function since they are clearly presynaptic to spiny appendages coupled by gap junctions. The terminals with large dense core vesicles are similar to those described in other areas of the central nervous system that contain fluorescent fibers when viewed with the light microscope (Chan-Palay, '73). They are most comparable to the CAT, endings of Chan-Palay ('73). Moreover, fluorescent fibers have been reported in different portions of the inferior olive in a number of species (Hoffman and Sladek, '73). The present terminals also are strikingly similar to those depicted and described in micrographs of the nucleus parasolitarius of the rat (W. H. Mehler, personal communication), an area demonstrating optimal formaldehyde-induced fluorescence. Thus, the present results can be interpreted to support the presence of a catacholaminergic axon system that influences synaptic events within the principal nucleus of the inferior olive via the synaptic cluster. When electron micrographs are enlarged 100,000 x it is apparent that the synaptic vesicles present in a single profile vary in both size and shape. The term pleomorphic was used in the present study to reflect these dieerences which are most apparent in profiles that contain flattened vesicles.
392
JAMES S . KING
It was these variations that prompted the search €or a method to measure a large number of vesicles quickly and the computer system described by Cowan and Wann (’73) was chosen. It is clear from the present results that no single criterion (e.g., vesicle size) is sufficient to differentiate populations of pre-synaptic profiles. For example, vesicles that are markedly different in shape (small spherical and pleomorphic) are similar in size. It remains to be established if differences in multiple ultrastructural features of synaptic terminals reflect different axon systems, transmitters or some yet undetermined variant. One of the purposes of the present study is to provide baseline measurements €or future experiments designed to define the origin of the synaptic profiles herein characterized by terminal size, vesicle shape and size, and junctional features. In particular the size of the spherical vesicles may prove of value in determining if either the large or small variety remain unaffected after a lesion to an afferent system that reaches the inferior olive. The pathological changes that signify axonal degeneration include alterations in vesicle size and, over time, shape. Measurements of vesicles in “normal” terminals that remain subsequent to a lesion may be of help in the determination of which category of axon terminal is undergoing degenerative changes. In conclusion, it is apparent that like many other nuclear areas currently under investigation, the inferior olive has the synaptic substrate for the “integration” of afferent volleys prior to activation of the principal projection neuron. An understanding of the synaptic organization within the olivary synaptic cluster will hopefully provide detail needed for further investigation of the physiological nature of that integration. ACKNOWLEDGMENTS
Thanks are expressed to Ms. Malinda Amspaugh for typing the manuscript, Mr. Gabriel Palkuti for photographic assistance and Ms. Barbara Diener for her excellent technical assistance. The frequent discussions of this material with Dr. George Martin are greatly appreciated and the proofreading by Dr. Jacqueline Bresnahan
was most helpful. The use of the Department of Physiology’s PDP-12 computer also is gratefully acknowledged. LITERATURE CITED Bowman, M. H., and J. S. King 1973 The conformation, cytology and synaptology of the opossum inferior olivary nucleus. J. Comp. Neur., 148: 491-524. Chan-Palay, V. 1973 On certain fluorescent axon terminals containing granular synaptic vesicles in the cerebellar nucleus lateralis. Z. Anat. Entwick1.-Gesch., 142: 239-258. Famiglietti, E. V., and A. Peters 1972 The synaptic glomerulus and the intrinsic neuron in the dorsal lateral geniculate nucleus of the cat. J. Comp. Neur., 144: 285-334. Fox, C. A., M. Ubeda-Purkiss, H. P. Ihrig and D. Biagioli 1951 Zinc chromate modification of the Golgi techniques. Stain Tech., 26: 109114. Gobel, S. 1974 Synaptic organization of the substantia gelatinosa glomeruli in the spinal trigeminal nucleus of the adult cat. J. Neurocytology, 3: 219-243. Gobel, S . , and R . Dubner 1969 Fine structural studies of the main sensory trigeminal nucleus i n cat and rat. J. Comp. Neur., 137: 459-494. Hoffman, D. L., and J. R. Sladek 1973 The distribution of catecholamine within the inferior olivary complex of the gerbil and rabbit. J. Comp. Neur., 151: 101-112. Kane, E. C. 1974 Synaptic organization i n the dorsal cochlear nucleus of the cat: A light and electron microscopic study. J. Comp. Neur., 155: 301-330. King, J. S . , and J. A. Andrezik 1975 The fine structure of the synaptic complex (glomerulus) in the inferior olivary nucleus. Anat. Rec., 181: 394-395. King, J. S., J. A. Andrezik, W. M. Falls and G. F. Martin (1976, submitted for publication) The synaptic organization of the cerebello-olivary circuit. Exp. Br. Res. King, J. S . , G. F. Martin and M. H. Bowman 1975 The direct spinal receiving area of the inferior olivary nucleus: A n electron microscopic study. Exp. Br. Res., 22: 13-24. LlinLs, R., R. Baker and C. Sotelo 1974 Electrotonic coupling between neurons in cat inferior olive. J. Neurophysiol., 37: 560-571. Majorossy, K., M. Rethelyi and J. Szentagothai 1965 The large glomerular synapse of the pulvinar. J. Hirnforsch., 7: 415-432. Martin, G. F., R. Dom, J. S . King, M. RoBards and C. R. R. Watson 1975 The inferior olivary nucleus of the opossum (Didelphis mmsupialis virginiana), its organization and connections. J. Comp. Neur., 160: 507-534. Mihailoff, G. A., and J. S. King 1975 The basilar pontine gray of the opossum: A correlated light and electron microscopic analysis. J. Comp. Neur., 159: 521-552. Palay, S., and V. Chan-Palay 1974 Cerebellar Cortex Cytology and Organization. SpringerVerlag, New York.
INFERIOR OLIVE SYNAPTIC CLUSTER Pinching, A. J., and T. P. S. Powell 1971 The neuropil of the glomeruli of the olfactory bulb. J. Cell Sci., 9: 347-377. Rafols, J. A., and C. A. Fox 1971/72 Further observations on the spiny neurons and synaptic endings in the striatum of the monkey. J. fur Hirnforschung, 13: 299-308. Ralston, H. J. 1965 The organization of the substantia gelatinosa Rolandi in the cat lumbosacral spinal cord. Z. Zellforsch. Mikroskop. Anat., 67: 1-23. Ralston, H. J., and M. M. Herman 1969 The fine structure of neurons and synapses in the ventrobasal thalamus of the cat. Brain Res., 14: 77-97. Ram6n y Cajal, S. 1911 Histologie due Systeme Nerveux de I'homme et des Vertebres. Vol. 11. Trans. by L. Azoulay. Maloine, Paris. Rustioni, A., and C. Sotelo 1974 Synaptic organization of the nucleus gracilis of the cat. Experimental identification of dorsal root fibers and cortical afferents. J. Comp. Neur., 155: 44 1-468. Scheibel. M. E.. and A. B. Scheibel 1955 The inferior olive: A Golgi study. J. Comp. Neur., 102; 77-132. Scheibel, M. E., A. B. Scheibel, F. Walberg and A. Brodal 1956 Areal distribution of axonal and dendritic patterns on inferior olive. J. Comp. Neur., 106: 21-49. Sotelo, C., R. Llinks and R. Baker 1974 Structural study of inferior olivary nucleus of the
393
cat. morphological correlates of electrotonic coupling. J. Neurophysiology, 37: 541-559. Sotelo, C., and S. Palay 1970 The fine structure of the lateral vestibular nucleus i n the rat. 11. Synaptic Organization. Brain Res., 18: 93-115. St. NBmeEek, and J. Wolff 1969 Light and electron microscopic evidence of complex synapses (glomeruli) in oliva inferior (cat). Experientia, 25: 634-635. Szentagothai, J. 1970 Glomerular synapses, complex synaptic arrangements and their operational significance. In: The Neurosciences Second Study Program. F. 0. Schmitt, ed. Rockefeller Press, New York, pp. 4 2 7 4 4 3 . Walberg, F. 1963 A n electron microscopic study of the inferior olive of the cat. J. Comp. Neur., 120: 1-17. 1964 Further electron microscopical investigations of the inferior olive of the cat. Prog. Brain Res., 6 ; 59-75. 1965a A special type of synaptic vesicle i n boutons i n the inferior olive. J. Ultrastruct. Res., 12: 237. 1965b An electron microscopic study of terminal degeneration i n the olive of the cat. J. Comp. Neur., 125: 205-222. 1966 Elongated vesicles in terminal boutons of the central nervous system, a result of aldehyde fixation. Acta Anat., 65: 229-235. 1970 Light and electron microscopic studies of two cerebellar relay nuclei. In: The Cerebellum i n Health and Disease. W. S. Fields and W. D. Willis, eds. W. H. Green Inc., St. Louis, pp. 39-62.
PLATE 1 EXPLANATION OF FIGURES
394
1
The astrocytic lamellae partially circumscribing three synaptic clusters are outlined in ink. The labelled dendrites ( d d ) give rise to spiny appendages which enter the synaptic cluster a t the upper right. This low magnification electron micrograph is taken from the principal nucleus and also illustrates a typical olivary neuron.
2
A n electron micrograph that illustrates a spiny appendage (SA) arising from a distal dendrite (dd). The vesicles and strands of smooth endoplasmic reticulum (solid black arrows) that characterize this appendage are labelled i n adjacent profiles (asterisks) not attached to their parent dendrite.
3
Typical neuron from the principal nucleus drawn from Golgi preparations with a 54 x oil immersion objective. The arrows indicate spiny appendages.
INFERIOR OLIVE SYNAPTIC CLUSTER James S. King
PLATE 1
PLATE 2 EXPLANATION O F FIGURES
4
The astrocytic lamellae surrounding a synaptic cluster are labelled with solid block arrows. The spiny appendages (asterisks) and a dendrite (dd) are post-synaptic to three populations of synaptic profiles which contain either large spherical ( C and D ) , small spherical ( A and B ) or pleomorphic ( E ) synaptic vesicles. The two open block arrows indicate a gap junction which is enlarged 100,000 x in figure 7. The three rectangular areas which include portions of the synaptic profiles labelled D, A and E are enlarged 100,000 x i n figures 5, 6 and 9 respectively. Electron micrograph from the medial accessory nucleus.
5
100,000 x enlargement from the area shown as a rectangle i n the synaptic ending labelled D i n figure 4.
6
100,000 x enlargement from the area shown as a rectangle i n the synaptic profile labelled A in figure 4.
7
100,000
x enlargement of the gap junctions labelled by block arrows
i n figure 4.
396
8
A frequency histogram which compares the mean area of synaptic vesicles within the synaptic clusters of the accessory nuclei.
9
100,000 x enlargement from the a.rea shown as a rectangle i n the synaptic profile labelled E in figure 4.
INFERIOR OLIVE SYNAPTIC CLUSTER James S. King
PLATE 2
397
PLATE 3 EXPLANATION OF FIGURES
10 Frequency histograms which compare the mean area of synaptic vesicles within the synaptic clusters of the principal nucleus. 11 The astrocytic lamellae surrounding a synaptic cluster are labelled with solid block arrows. The spiny appendages (asterisks) and a dendrite (dd) are labelled. The rectangular area i n the synaptic profile with dense core vesicles is enlarged 100,000 X in figure 12. Electron micrograph from the principal nucleus. 12
398
100,000 x enlargement from the area shown as a rectangle in the synaptic ending with dense core vesicles in figure 11.
INFERIOR OLIVE SYNAPTIC CLUSTER James S . King
PLATE 3
399
ABSTRACT This report describes the fine structural features and distribution of the synaptic cluster (glomerulus) within the inferior olivary nucleus of the opossum. The postsynaptic elements typically include spiny appendages and small diameter dendrites which exhibit attachment plaques and gap junctions. Profiles presynaptic to the central core of postsynaptic elements were differentiated on the basis of vesicle shape, vesicle size, as measured by a computer system, and junctional characteristics. Three categories of terminals with clear vesicles are present within the synaptic clusters in all nuclear divisions of the olive, whereas a fourth with large dense core vesicles is restricted primarily to the principal nucleus. The groups of pre and postsynaptic elements are surrounded by astrocytic lamellae and are most frequently encountered in the principal and rostra1 portions of the medial accessory nuclei. Possible identification of the sources of the synaptic components is discussed in relation to data available from Golgi impregnations, physiological reports and hodological evidence.
Synaptic clusters, termed glomeruli, endoplasmic reticulum. The central core within the inferior olivary nucleus were of the synaptic cluster is made up of first described by St. NEmeEek and Wolff spines, spiny appendages and small diam('69) in the cat. Previous electron micro- eter dendrites. Desmosome-like contacts scopic studies of the olive (Walberg, '63, are present, depending on the plane of sec'64, '65a,b, '66) had focused on the popu- tion, between all of the various postlations of synaptic endings which were de- synaptic components which are contacted fined as Gray's types I and I1 or as termi- by a peripheral rim of synaptic profiles. nals with elongated vesicles. Junctional The entire area is circumscribed by lamelzones similar to desmosomes were observed lar sheets of astrocytic processes. An addiwithin the glomeruli described by St. tional ultrastructural feature of the gloNEmeCek and Wolff ('69) and between ad- meruli, gap junctions, was described by jacent dendrites by Walberg ('64). The Sotelo et al. ('74) in the cat and subselatter author also emphasized the abun- quently by King et al. ( ' 7 5 ) in restricted dance of astrocytic lamellae interposed be- portions of the dorsal and medial accestween various elements of the neuropil. sory olivary nuclei. In our previous account (Bowman and Potential corollaries for the sources of preand post-synaptic profiles seen in eIectron King, ' 7 3 ) synaptic profiles within glommicrographs are provided by the Golgi eruli were differentiated by vesicle size and descriptions of Scheibel and Scheibel shape. At that time the typical mammalian ( ' 5 5 ) , Scheibel et al. ('56) and Ramon y nuclear subdivisions were not apparent in the opossum. Such divisions (principal, Cajal ('11). More recently, Bowman and King ( ' 7 3 ) dorsal and medial accessory nuclei) have described glomeruli in the inferior olive of been identified recently (Martin et al., '75) the opossum. In that species unique spiny This paper i s dedicated in memory of appendages arise from the dendrites of the my1 Dedication: brother Charles D. King, M.D, (1944-1975). ZThese results were presented in part at the 88th typical olivary neuron and when viewed in Meeting of the American Association of Anatomists, electron micrographs their profiles are seen Los Angeles, California. 3 Acknowledgment of Support: This investigation to contain vesicles and strands of smooth u a s supported by a U.S.P.H.S. Grant NS-08798. J. COMP. N E U R . , 165: 387-400.
387
388
JAMES S. KING
and a rapid, standardized method of determining the “size” of synaptic vesicles is now available in our laboratory. This measurement system was developed in the laboratory of Dr. W. M. Cowan at Washington University, St. Louis, Missouri (Cowan and Wann, ’73) and first utilized for this purpose by Rafols and Fox (’71/ ’72). Thus, more extensive sampling has been directed toward prescribed nuclear subdivisions for the following purposes : ( 1) to determine the uniformity of glomerular distribution, as well as the uniformity of the pre- and post-synaptic elements within the glomeruli, ( 2 ) to assess the populations of synaptic profiles by computer measurement of synaptic vesicle size and ( 3 ) to provide control data for future experiments designed to determine the origin of the resident presynaptic profiles. MATERIALS AND METHODS
The brains used for the present electron microscopic study were taken from nine adult opossums ( Didelphis marsupialis virg i n i a n a ) subsequent to the double aldehyde perfusion method described in detail by Mihailoff and King (’75). The inferior olive from each side of the brainstem was divided into four one millimeter slabs cut in the transverse plane. After osmium postfixation, dehydration and plastic embedding, one micron sections were cut from the face of each of the blocks. Each tissue slab was flat embedded to ensure a transverse plane of section. Thus, each thin section also was cut in the same plane from precise areas throughout the rostra1 to caudal extent of the principal, dorsal and medial accessory olivary nuclei. The reader should consult figures 1-21 in the article by Martin et al. (’75) for illustrations of the nuclear divisions of the opossum inferior olive. Sampling included the dorsal and ventral lamellae of the principal nucleus and the more caudal portion of the same nucleus where it is not divided into individual lamellae. Portions a, b and c of the medial accessory nucleus were examined, but because of its small size the cap of Kooy was not included. Both the direct and indirect (dorsal column) spinal receiving portions of the dorsal accessory nucleus were extensively sampled. Sections
picked u p on copper grids were examined with a Philips 300 electron microscope. In order to measure the size of synaptic vesicles, electron micrographs of glomeruli from both the accessory and principal nuclei were enlarged 100,000 times. The membranes of the vesicles i n each presynaptic profile were traced with a stylus attached to a cybergraphic tablet (Talos Systems Inc., Scottsdale, Arizona) which was interfaced with a PDP-12 computer. The computer program, provided by Dr. W. Maxwell Cowan, Washington University, St. Louis, determines the area of each traced vesicle in nmz and computes the mean, standard deviation, and variance of each sample of vesicles from an ending, as well as providing histograms of the vesicle areas (bin width = 200 nmz). The obvious values of this method of deriving synaptic vesicle size are speed and ease of measurement, as well as standardization of measurement . Five glomeruli were selected from the direct spinal portion of the dorsal accessory nucleus, five from the medial accessory (either part a or b ) and five from the principal nucleus (either the ventral lamellae or the caudal undivided part) for measurement of synaptic vesicles. Vesicles in all endings in each glomerulus were measured. Initial observations based on these measurements and other morphological features of the vesicle populations suggested that there were three categories in the glomeruli of the accessory nuclei (large spherical, small spherical and pleomorphic) and four categories in the principal nucleus (large spherical, small spherical, pleomorphic and dense core). In order to determine: ( 1 ) whether the categories of clear vesicles represent populations which are significantly different in size, as their ranges clearly overlap (figs. 8, IOA), and ( 2 ) whether the categories of terminals common to both the accessory and principal nuclei are from the same populations; four synaptic endings containing each of the vesicle types (large spherical, small spherical and pleomorphic) were randomly selected from glomeruli in the accessory nuclei and four endings of each type were selected from glomeruli in the principal nucleus. The area of thirty vesicles in each ending was measured and these data were
INFERIOR OLIVE SYNAPTIC CLUSTER
subjected to a tweway analysis of variance. The results of the statistical analysis revealed that there was a significant main effect for type of vesicle (F = 53.42, df = 2, p < 0.001), but no significant main effect for locus, i.e., principal vs. accessory, ( F = 3.74, df = 1, p > 0.05 and no significant interaction (F = 0.79, df = 2, p > 0.05). Post-hoc individual comparisons were made using the Newman-Kuels method and the results will be reported subsequently in the observations section. The statistical analysis was performed by the Biometrics Laboratory, Department of Preventive Medicine, The Ohio State University. OBSERVATIONS
The frequency with which olivary synaptic clusters are encountered in electron micrographs from the various nuclear divisions is markedly different. The impression one derives is that the principal nucleus including the dorsal and ventral lamellae and the more caudal undivided portion has many more glomeruli per unit area than any of the accessory nuclei. The sole exception is the rostral portion of the medial accessory nucleus (fig. 17A, Martin et al., '75), where many synaptic clusters also are present. Throughout the principal, dorsal and medial accessory olivary nuclei aggregates of pre- and post-synaptic profiles are partially or totally circumscribed by astrocytic Jamellae. In figure 1, which is taken from the principal olive, three such areas are demarcated by outlining the astrocytic processes in ink. The greatest and least dimensions of the synaptic cluster average 3.63 and 2.46 respectively, with little variation in size between the parts of the nucleus that were sampled. The central core of the synaptic cluster is made up of an average of six profiles (range 3-1 1 ). These post-synaptic elements (fig. 2 ) include small dendrites (dd), spines and excresences which have been termed spiny appendages (SA). When viewed in Golgi preparations (fig. 3, arrows), these dendritic appendages arise from intermediate and distal dendrites, and comparable profiles (fig. 2, SA) are readily identified in electron micrographs. The spiny appendages contain dense core vesicles, mitochondria, a flocculent ma-
389
terial, clear vesicles (fig. 2, block arrows), and strands of smooth endoplasmic reticulum (fig, 2, block arrows) enabling their identification (fig. 2, asterisks) even when they are not attached to their parent dendrite in the plane of section. Attachment plaques are frequently encountered (57% of the glomeruli) between the dendritic elements of the synaptic cluster with no predilection for a specific nuclear division (figs. 2, 11, circles). In addition, gap junctions are present (fig. 4, open block arrows; fig. 7). Figure 7 is a 100,000 x enlargement of the junction marked by block arrows in figure 4. Gap junctions are encountered most frequently in the various subdivisions of the accessory nuclei. In the accessory nuclei, the profiles that are presynaptic to the central core of postsynaptic elements are divisible into three categories based on the following criteria: ( 1 ) vesicle size, ( 2 ) vesicle shape and ( 3 ) junctional features. Synaptic profiles with spherical vesicles and Gray's type I active sites are divisible into two groups based on vesicle size (fig. 8). Small spherical vesicles (fig. 4, endings labelled A and B ) have a mean area of 908.78 nm', -C 222.18 nm'. If this area was considered to be the area of a circle the mean diameter would be 340 A. The synaptic endings with small spherical vesicles typically contact one post-synaptic profile via a single junctional area. They are infrequently presynaptic to a spiny appendage that displays a gap junction. The population of larger spherical vesicles (fig. 4, endings labelled C and D ) has a mean area of 1248.81 nm', -t 303.10 nm2.If this area were considered to be that of a circle it would have a diameter of 399 A. Terminals with large spherical vesicles may be presynaptic to a single post-synaptic element (fig. 4, ending labelled D) or to multiple dendritic processes some of which also exhibit gap junctions (fig. 4, ending labelled C ) . The presynaptic profiles with spherical vesicles number three or four per synaptic cluster (in a single section) with varying proportions of the large and small varieties. The third category of synaptic ending can be differentiated by its pleomorphic vesicles and intermediate type of synaptic junction (fig. 4, ending labelled E). The vesicles have a mean area of 897.89 nm',
390
JAMES S. KING
229.99 nm2, with an estimated diameter of 338 A. This axon terminal typically is presynaptic to at least three post-synaptic profiles, two of which often display a gap junction (fig. 4, ending labelled E). No more than two of these endings appear in a synaptic cluster and frequently only a single profile with pleomorphic vesicles is encountered. The differences in shape and size of the three vesicle populations can be better appreciated when the micrographs are enlarged 100,000 x (compare (figs. 5, 6, 9 ) as they were for all measurements with the computer system. The post-hoc comparisons revealed no significant difference between the size of the pleomorphic and small spherical vesicles (fig. 8 ) , however the large spherical vesicles were significantly different in size from the other two populations of vesicles at the 0.01 level. In the principal nucleus the same three types of presynaptic profiles are evident. The mean area for the small spherical vesicles is 910.49 nm2, +- 213.53 nm2, for the large spherical vesicles is 1336.04 nm2, 2 336.14 nmz, and for the pleomorphic 332.50 nm2. A vesicles is 995.66 nm2, frequency histogram (fig. 10A) depicts their distribution. If each of the three areal measurements were considered to be areas of circles the vesicles would have diameters of 340 A, 412 A and 356 A respectively. The two factor analysis of variance is supportive of the statement that the three vesicle population are similar within the accessory and principal nuclei, as no significant difference was found between the matched categories of vesicles in the principal and accessory nuclei (compare figs. 8 and 10A). The post-hoc analysis indicated that the significant difference in size was between the large spherical vesicles and the small spherical vesicles ( p < O.Ol), and between the large spherical and pleomorphic ( p < 0.01) vesicle populations. No significant size differential was evident between the small spherical vesicles and the pleomorphic vesicles. A feature unique to the synaptic clusters within the principal nucleus is a presynaptic profile with large dense core vesicles (fig. 11, enlarged in fig. 12). One such profile is typically present and presynaptic to as many as three of the dendritic or
*
spiny elements found in the center of the glomerulus. At 100,000 x magnification (fig. 1 2 ) the dense core vesicles, which average 22 per synaptic profile can be seen to better advantage and compared with clear spherical vesicles that also are present. The mean area of each of these vesicle populations is 4773.57 nm2, 1254.58 nm2 and 1097.03 nm2, 268.78 nmz respectively. If each area was considered to be that of a circle, their diameters in Angstroms would be 780 and 374. Their size distributions are illustrated by frequency histograms in figures 10B, C. In the example of this type of ending (figs. 11, 12) the dense core vesicles make up only 29% of the total vesicle population. As is the case in the accessory nuclei, terminals with spherical vesicles average three or four per synaptic cluster and contact one or two post-synaptic elements. A single synaptic profile with pleomorphic vesicles is typically seen although two may be present. Thus, the only presynaptic terminal within the glomeruli that is restricted in its nuclear distribution is the ending with dense core vesicles. Although characteristic of the principal nucleus in general, it is most frequently seen in the more caudal portion of the nucleus (fig. 13, Martin et al., ’75).
*
*
DISCUSSION
The definition of a glomerulus as “a circumscribed structural arrangement between a number of specific axon terminals and one or several dendrites” which “is separated from the environment by a capsule of glial processes” has been suggested by Szentagothai (’70). Moreover, various groups of synaptic elements, possessing similar and certain distinctive features, have been termed glomeruli by numerous authors. Descriptions include the : olfactory bulb (Pinching and Powell, ’71), medial geniculate body (Morest, ’71), dorsal cochlear nucleus (Kane, ’74), dorsal lateral geniculate nucleus (Famiglietti and Peters, ’72), pulvinar (Majorossy et al., ’65), nucleus gracilis (Rustioni and Sotelo, ’74), main sensory nucleus of V (Gobel and Dubner, ’69), spinal trigeminal nucleus (Gobel, ’74), substantia gelatinosa (Ralston, ’65), ventral basal thalamus (Ralston and Herman, ’69), lateral ves-
INFERIOR OLIVE SYNAPTIC CLUSTER
tibular nucleus (Sotelo and Palay, '70) and cerebellar cortex (Palay and ChanPalay, '73). To this growing list should be added the glomerulus of the inferior olivary nucleus based on the present account and the descriptions of St. NEmeEek and Wolff, '69; Bowman and King, '73 and Sotelo et al., '74. However, as serial sections have not been utilized to establish that the synaptic area under consideration is a diminutive ball (glomerulus), reference to the synaptic cluster of the inferior olive is perhaps more appropriate. Moreover, in any given plane of section the axonal and dendritic elements are not always encircled totally by astrocytic lamellae. Szentagothai ('70) suggests that the latter feature when observed warrants the use of this term. From the present results it appears that the synaptic cluster of the inferior olive is ubiquitous, although most frequently encountered in the principal and rostral part ( a ) of the medial accessory nuclei. Due to its limited size the cap of Kooy was not sampled. Mizuno et al. ('74) did not report synaptic clusters in their study, however, their primary goal was to identify pretecto-olivary axon terminals within the same region. It is clear that the central core of the synaptic cluster is dendritic and primarily derived from spiny appendages. The number of these spinous protrusions that contribute to one complex is not certain, but in favorable planes of section (fig. 1: dd) two dendrites can be seen to give rise to individual spiny appendages at the periphery of the cluster. Although the ultrastructur a1 basis of dendrodendritic transmission via chemical transmitters has been conclusively demonstrated in other glomeruli, gap junctions, the morphological corollary of electrotonic transmission, have not been reported. Gap junctions are more typical within synaptic clusters of the accessory olivary nuclei than in the principal nucleus. Electrophysiological evidence supportive of their suggested role in the synchronous firing of olivary neurons has been presented by Llinas et al. ('74). The axonal systems that influence the dendritic profiles within glomeruli are the subject of current investigation and in a preliminary report (King and Andrezik.
39 1
'75) destruction of the deep cerebellar nuclei has been shown to result in degenerating axon terminals within the olivary synaptic islands. Their precise nuclear origin and their distribution upon post-synaptic elements will be presented in a more detailed report (King et al., '76, submitted for publication). Aside from the input from the deep cerebellar nuclei (King et al.,'75), no experimental ultrastructural evidence is available that would suggest possible sources for the four distinct populations of presynaptic profiles identified in the present account. However, the multiple afferents to select parts of the opossum inferior olive (Martin et al., '75) provide a number of potential sources. Axon terminals of an inhibitory neuron activated by recurrent collaterals have been suggested as one possibiIity (LlinLs et al., '74). It is tempting to speculate, but by no means certain, that the synaptic profiles with pleomorphic vesicles would be inhibitory in function since they are clearly presynaptic to spiny appendages coupled by gap junctions. The terminals with large dense core vesicles are similar to those described in other areas of the central nervous system that contain fluorescent fibers when viewed with the light microscope (Chan-Palay, '73). They are most comparable to the CAT, endings of Chan-Palay ('73). Moreover, fluorescent fibers have been reported in different portions of the inferior olive in a number of species (Hoffman and Sladek, '73). The present terminals also are strikingly similar to those depicted and described in micrographs of the nucleus parasolitarius of the rat (W. H. Mehler, personal communication), an area demonstrating optimal formaldehyde-induced fluorescence. Thus, the present results can be interpreted to support the presence of a catacholaminergic axon system that influences synaptic events within the principal nucleus of the inferior olive via the synaptic cluster. When electron micrographs are enlarged 100,000 x it is apparent that the synaptic vesicles present in a single profile vary in both size and shape. The term pleomorphic was used in the present study to reflect these dieerences which are most apparent in profiles that contain flattened vesicles.
392
JAMES S . KING
It was these variations that prompted the search €or a method to measure a large number of vesicles quickly and the computer system described by Cowan and Wann (’73) was chosen. It is clear from the present results that no single criterion (e.g., vesicle size) is sufficient to differentiate populations of pre-synaptic profiles. For example, vesicles that are markedly different in shape (small spherical and pleomorphic) are similar in size. It remains to be established if differences in multiple ultrastructural features of synaptic terminals reflect different axon systems, transmitters or some yet undetermined variant. One of the purposes of the present study is to provide baseline measurements €or future experiments designed to define the origin of the synaptic profiles herein characterized by terminal size, vesicle shape and size, and junctional features. In particular the size of the spherical vesicles may prove of value in determining if either the large or small variety remain unaffected after a lesion to an afferent system that reaches the inferior olive. The pathological changes that signify axonal degeneration include alterations in vesicle size and, over time, shape. Measurements of vesicles in “normal” terminals that remain subsequent to a lesion may be of help in the determination of which category of axon terminal is undergoing degenerative changes. In conclusion, it is apparent that like many other nuclear areas currently under investigation, the inferior olive has the synaptic substrate for the “integration” of afferent volleys prior to activation of the principal projection neuron. An understanding of the synaptic organization within the olivary synaptic cluster will hopefully provide detail needed for further investigation of the physiological nature of that integration. ACKNOWLEDGMENTS
Thanks are expressed to Ms. Malinda Amspaugh for typing the manuscript, Mr. Gabriel Palkuti for photographic assistance and Ms. Barbara Diener for her excellent technical assistance. The frequent discussions of this material with Dr. George Martin are greatly appreciated and the proofreading by Dr. Jacqueline Bresnahan
was most helpful. The use of the Department of Physiology’s PDP-12 computer also is gratefully acknowledged. LITERATURE CITED Bowman, M. H., and J. S. King 1973 The conformation, cytology and synaptology of the opossum inferior olivary nucleus. J. Comp. Neur., 148: 491-524. Chan-Palay, V. 1973 On certain fluorescent axon terminals containing granular synaptic vesicles in the cerebellar nucleus lateralis. Z. Anat. Entwick1.-Gesch., 142: 239-258. Famiglietti, E. V., and A. Peters 1972 The synaptic glomerulus and the intrinsic neuron in the dorsal lateral geniculate nucleus of the cat. J. Comp. Neur., 144: 285-334. Fox, C. A., M. Ubeda-Purkiss, H. P. Ihrig and D. Biagioli 1951 Zinc chromate modification of the Golgi techniques. Stain Tech., 26: 109114. Gobel, S. 1974 Synaptic organization of the substantia gelatinosa glomeruli in the spinal trigeminal nucleus of the adult cat. J. Neurocytology, 3: 219-243. Gobel, S . , and R . Dubner 1969 Fine structural studies of the main sensory trigeminal nucleus i n cat and rat. J. Comp. Neur., 137: 459-494. Hoffman, D. L., and J. R. Sladek 1973 The distribution of catecholamine within the inferior olivary complex of the gerbil and rabbit. J. Comp. Neur., 151: 101-112. Kane, E. C. 1974 Synaptic organization i n the dorsal cochlear nucleus of the cat: A light and electron microscopic study. J. Comp. Neur., 155: 301-330. King, J. S . , and J. A. Andrezik 1975 The fine structure of the synaptic complex (glomerulus) in the inferior olivary nucleus. Anat. Rec., 181: 394-395. King, J. S., J. A. Andrezik, W. M. Falls and G. F. Martin (1976, submitted for publication) The synaptic organization of the cerebello-olivary circuit. Exp. Br. Res. King, J. S . , G. F. Martin and M. H. Bowman 1975 The direct spinal receiving area of the inferior olivary nucleus: A n electron microscopic study. Exp. Br. Res., 22: 13-24. LlinLs, R., R. Baker and C. Sotelo 1974 Electrotonic coupling between neurons in cat inferior olive. J. Neurophysiol., 37: 560-571. Majorossy, K., M. Rethelyi and J. Szentagothai 1965 The large glomerular synapse of the pulvinar. J. Hirnforsch., 7: 415-432. Martin, G. F., R. Dom, J. S . King, M. RoBards and C. R. R. Watson 1975 The inferior olivary nucleus of the opossum (Didelphis mmsupialis virginiana), its organization and connections. J. Comp. Neur., 160: 507-534. Mihailoff, G. A., and J. S. King 1975 The basilar pontine gray of the opossum: A correlated light and electron microscopic analysis. J. Comp. Neur., 159: 521-552. Palay, S., and V. Chan-Palay 1974 Cerebellar Cortex Cytology and Organization. SpringerVerlag, New York.
INFERIOR OLIVE SYNAPTIC CLUSTER Pinching, A. J., and T. P. S. Powell 1971 The neuropil of the glomeruli of the olfactory bulb. J. Cell Sci., 9: 347-377. Rafols, J. A., and C. A. Fox 1971/72 Further observations on the spiny neurons and synaptic endings in the striatum of the monkey. J. fur Hirnforschung, 13: 299-308. Ralston, H. J. 1965 The organization of the substantia gelatinosa Rolandi in the cat lumbosacral spinal cord. Z. Zellforsch. Mikroskop. Anat., 67: 1-23. Ralston, H. J., and M. M. Herman 1969 The fine structure of neurons and synapses in the ventrobasal thalamus of the cat. Brain Res., 14: 77-97. Ram6n y Cajal, S. 1911 Histologie due Systeme Nerveux de I'homme et des Vertebres. Vol. 11. Trans. by L. Azoulay. Maloine, Paris. Rustioni, A., and C. Sotelo 1974 Synaptic organization of the nucleus gracilis of the cat. Experimental identification of dorsal root fibers and cortical afferents. J. Comp. Neur., 155: 44 1-468. Scheibel. M. E.. and A. B. Scheibel 1955 The inferior olive: A Golgi study. J. Comp. Neur., 102; 77-132. Scheibel, M. E., A. B. Scheibel, F. Walberg and A. Brodal 1956 Areal distribution of axonal and dendritic patterns on inferior olive. J. Comp. Neur., 106: 21-49. Sotelo, C., R. Llinks and R. Baker 1974 Structural study of inferior olivary nucleus of the
393
cat. morphological correlates of electrotonic coupling. J. Neurophysiology, 37: 541-559. Sotelo, C., and S. Palay 1970 The fine structure of the lateral vestibular nucleus i n the rat. 11. Synaptic Organization. Brain Res., 18: 93-115. St. NBmeEek, and J. Wolff 1969 Light and electron microscopic evidence of complex synapses (glomeruli) in oliva inferior (cat). Experientia, 25: 634-635. Szentagothai, J. 1970 Glomerular synapses, complex synaptic arrangements and their operational significance. In: The Neurosciences Second Study Program. F. 0. Schmitt, ed. Rockefeller Press, New York, pp. 4 2 7 4 4 3 . Walberg, F. 1963 A n electron microscopic study of the inferior olive of the cat. J. Comp. Neur., 120: 1-17. 1964 Further electron microscopical investigations of the inferior olive of the cat. Prog. Brain Res., 6 ; 59-75. 1965a A special type of synaptic vesicle i n boutons i n the inferior olive. J. Ultrastruct. Res., 12: 237. 1965b An electron microscopic study of terminal degeneration i n the olive of the cat. J. Comp. Neur., 125: 205-222. 1966 Elongated vesicles in terminal boutons of the central nervous system, a result of aldehyde fixation. Acta Anat., 65: 229-235. 1970 Light and electron microscopic studies of two cerebellar relay nuclei. In: The Cerebellum i n Health and Disease. W. S. Fields and W. D. Willis, eds. W. H. Green Inc., St. Louis, pp. 39-62.
PLATE 1 EXPLANATION OF FIGURES
394
1
The astrocytic lamellae partially circumscribing three synaptic clusters are outlined in ink. The labelled dendrites ( d d ) give rise to spiny appendages which enter the synaptic cluster a t the upper right. This low magnification electron micrograph is taken from the principal nucleus and also illustrates a typical olivary neuron.
2
A n electron micrograph that illustrates a spiny appendage (SA) arising from a distal dendrite (dd). The vesicles and strands of smooth endoplasmic reticulum (solid black arrows) that characterize this appendage are labelled i n adjacent profiles (asterisks) not attached to their parent dendrite.
3
Typical neuron from the principal nucleus drawn from Golgi preparations with a 54 x oil immersion objective. The arrows indicate spiny appendages.
INFERIOR OLIVE SYNAPTIC CLUSTER James S. King
PLATE 1
PLATE 2 EXPLANATION O F FIGURES
4
The astrocytic lamellae surrounding a synaptic cluster are labelled with solid block arrows. The spiny appendages (asterisks) and a dendrite (dd) are post-synaptic to three populations of synaptic profiles which contain either large spherical ( C and D ) , small spherical ( A and B ) or pleomorphic ( E ) synaptic vesicles. The two open block arrows indicate a gap junction which is enlarged 100,000 x in figure 7. The three rectangular areas which include portions of the synaptic profiles labelled D, A and E are enlarged 100,000 x i n figures 5, 6 and 9 respectively. Electron micrograph from the medial accessory nucleus.
5
100,000 x enlargement from the area shown as a rectangle i n the synaptic ending labelled D i n figure 4.
6
100,000 x enlargement from the area shown as a rectangle i n the synaptic profile labelled A in figure 4.
7
100,000
x enlargement of the gap junctions labelled by block arrows
i n figure 4.
396
8
A frequency histogram which compares the mean area of synaptic vesicles within the synaptic clusters of the accessory nuclei.
9
100,000 x enlargement from the a.rea shown as a rectangle i n the synaptic profile labelled E in figure 4.
INFERIOR OLIVE SYNAPTIC CLUSTER James S. King
PLATE 2
397
PLATE 3 EXPLANATION OF FIGURES
10 Frequency histograms which compare the mean area of synaptic vesicles within the synaptic clusters of the principal nucleus. 11 The astrocytic lamellae surrounding a synaptic cluster are labelled with solid block arrows. The spiny appendages (asterisks) and a dendrite (dd) are labelled. The rectangular area i n the synaptic profile with dense core vesicles is enlarged 100,000 X in figure 12. Electron micrograph from the principal nucleus. 12
398
100,000 x enlargement from the area shown as a rectangle in the synaptic ending with dense core vesicles in figure 11.
INFERIOR OLIVE SYNAPTIC CLUSTER James S . King
PLATE 3
399
