The Laminar and Size Distribution of Commissural Efferent Neurons in the Cat Visual Cortex*

Arch. histol. jap., Vol. 42, No. 2 (1979) p. 119-128 The Laminar and Size Distribution of Commissural Efferent Neurons in the Cat Visual Cortex* Kazuhiko SHOUMURA Department of Anatomy (Prof. S. DEURA), Gifu University School of Medicine, Gifu, Japan Received July 14, 1978 Summary. The laminar and size distribution of commissural efferent neurons were studied in the cat visual cortex including area 17, 18, 19 and the lateral suprasylvian cortex (the lateral wall of the middle suprasylvian gyrus) by means of retrograde transport of horseradish peroxidase. Single or multiple injections of the enzyme were made unilaterally along the medial part of the lateral gyrus (the adjoining parts of area 17 and 18) or along the lateral wall of the middle suprasylvian gyrus. In the contralateral cerebral cortex (area 17, 18, 19 and lateral suprasylvian cortex) the majority of peroxidase labeled commissural efferent neurons was identified in layer III with a smaller population in other layers except layer I. The size distribution of labeled cells of layer III showed that all sizes of neurons contribute to interhemispheric connections, ranging from small cells which are comparable to those predominating in layer III of the acommissural medial area 17 to the characteristic large pyramids of layer III at the area 17/18 boundary. There was a certain difference in the pattern of size distribution of commissural efferent neurons of layer III between area 17, 18, 19 and the lateral suprasylvian cortex. Labeled neurons in area 17, 18, 19 showed a dominant peak in the histograms of cell size, while in the lateral suprasylvian cortex they were more evenly distributed in a wide size range. Commissural connections of the cerebral cortex, especially the visual and somatic sensory cortices, show considerable topical changes in distribution and density (see review by DOTY and NEGRAO, 1973). It is presumed that the differences in cytoarchitecture are reflected upon the differences in fiber connections. In the visual cortex of cats, we have shown that the regional differences in commissural connections were paralleled by the regional differences in cytoarchitecture; the commissural dense area 17/18 boundary contained a recognizable group of large pyramids in layer III, while in the acommissural medial area 17 layer III was composed mostly of small neurons (SHOUMURA, 1974). In monkeys and man, there is a cluster of unusually large pyramids in layer III of area 18 bordering area 17, and this narrow band of the cortex has been considered or OBg (BONIN, 1942). Degeneration studies in the monkey (CRAGG, 1969; ZEKI, 1970; *This work was supported in part by grants (No. 067062, 1975; No. 267005, 1977) from the Ministry of Education. 119

120 K. SHOUMURA: dense commissural terminals. GLICKSTEIN and WHITTERIDGE (1976) noted degeneration or atrophy of these large pyramids following a large lesion of the occipital lobe of the opposite side or section of the corpus callosum. In agreement with this observation, layer III of OBg of the human brains lacking in the corpus callosum did not contain any characteristic large pyramids (SHOUMURA, ANDO and KATO, 1975). Thus, it is very probable that large pyramids of layer III at the area 17/18 boundary play some role in interhemispheric communication through the corpus callosum. It is of particular importance that an increase in the size of the pyramids of layer III is a strikingly common feature of the commissurally connected regions of the cerebral cortex in different species. The aim of this study is to determine whether or not commissural efferent neurons of cat visual cortical areas comprising area 17, 18, 19 and the lateral suprasylvian cortex can be distinguished in terms of cell size or cortical layer where they reside. For this purpose the retrograde transport of horseradish peroxidase was used (LAVAIL and LAVAIL, 1972). MATERIALS AND METHODS In 5 young adult cats (2.0-2.5kg) anesthetized with 25mg/kg of ketamine hydrochloride, 50% horseradish peroxidase (HRP; Sigma Type VI) in saline was injected unilaterally into 1-13 places along the medial part of the lateral gyrus (the adjoining parts of area 17 and 18) and the lateral wall of the middle suprasylvian gyrus (lateral fused with 0.9% saline followed by 3% paraformaldehyde and 1% glutaraldehyde in and every fifth section was incubated in Tris-HCl buffer (ph 7.6) containing 3,3'- diaminobenzidine 4HCl and hydrogen peroxide following the method of GRAHAM and KARNOVSKY (1966). Mounted sections were lightly counterstained with cresyl violet. Neurons containing HRP-positive granules were identified under both bright and dark field illumination. The distribution of retrogradely labeled neurons was cortical layer were identified under bright field illumination. In order to estimate the cell size of labeled neurons, outline drawings were made of all somata containing HRP reaction products with the aid of a drawing tube at a magnification of were found only in the axon hillock and their complete outline drawings could be obtained only by Nissl stain. The cell outline drawings were, then, photographed and printed at the same magnification with the original drawings. Outlined cell profiles printed on photopapers were cut out carefully and weighed using a balance (Mettler, Type H6). Cell sizes were calculated from the weights of outlined cell profiles and several reference squares of 1,225 mm2.

Commissural Efferent Neurons in Cat Visual Cortex 121 RESULTS In three examples which received single (Cat 16) or multiple (Cat 15, 26) peroxidase injections along the medial part of the lateral gyrus, the distribution of labeled neurons in the opposite cortex fell approximately within the region where degenerated commissural fibers appeared following a large lesion affecting area 17, 18 and 19 (SHOUMURA, 1974). Although the area of strong HRP reaction at the injection sites was extensive enough to involve most of the splenial, suprasplenial and lateral gyri, labeled neurons were identified in area 19 in a small number and less consistently than in other areas. Area 19 contained no labeled neurons in Cat 16 (Fig. 1). In one example (Cat 26, Fig. 5), labeled neurons were also observed in the lateral suprasylvian cortex contralateral to injections and this confirms the results of previous degeneration (SHOUMURA and ITOH, 1972) and HRP (SHATZ, 1977) studies, which have demonstrated heterotopical commissural efferents from the lateral suprasylvian cortex to the contralateral lateral gyrus. The number of labeled neurons in the lateral suprasylvian cortax was far smaller than that in area 17 and 18. This may be partly due to smallness in the number of heterotopically oriented commissural fibers. Cat 16 part of the lateral gyrus. This example contained the smallest number of labeled neurons, but their distribution within the medial wall of the lateral gyrus (area 17) was most extensive. In this and subsequent figures, each dot represents one HRP-positive neuron. The solid or stippled area indicates strong or weak HRP reaction at the injection site respectively. Arrowhead indicates area 17/18 boundary. lat Lateral gyrus, ssy suprasylvian gyrus, IV layer IV.

122 K. SHOUMURA: In addition, the distribution of labeled neurons within area 17 was not limited to its extreme lateral part, but was more widely scattered extending well down into the medial wall of the lateral gyrus (Fig. 1). However, the splenial gyrus (medial area 17) contained no such neurons. INNOCENTI and FIORE (1976) confirmed that the contribution of area 17 to the corpus callosum was not so small as to be negligible. SHATZ (1977) also found several neurons filled with HRP positive granules in the medial wall of the lateral gyrus of an ordinary cat. In two examples (Cat 19 and 22) in which HRP was injected along the lateral suprasylvian cortex, many labeled neurons appeared in the corresponding portions of the opposite cortex, and a small number of them were found heterotopically in Cat 19 Fig. 2. An example in which labeled neurons were found in both homotopical and heterotopical portions of the cortex contralateral to injections. Arrowhead indicates area 17/18 boundary. lat Lateral gyrus, ssy suprasylvian gyrus, IV layer IV. Cat 22 Fig. 3. An example to show that labeled neurons were distributed throughout most of the mediolateral extent of the banks of the suprasylvian sulcus (arrowheads). lat Lateral gyrus, ssy suprasylvian gyrus, IV layer IV.

Commissural Efferent Neurons in Cat Visual Cortex 123 area 17/18 boundary (Fig. 2, 3). All labeled neurons at area 17/18 boundary were large to medium sized pyramids identified in layer III. These two examples demonstrated that any portion of the lateral suprasylvian cortex gives rise to commissural Fig. 4. Histograms of the sizes of labeled cells of layer III (left) and of laminar distribution of all labeled neurons (right) in three examples. 17, 18, 19 Cortical areas 17, 18 and 19, LS lateral suprasylvian cortex, 2, 2/3, 3, 3/4, 4, 5, 6 cortical layers; 2/3 or 3/4 means that it was uncertain which layer the labeled neurons belonged to. n Number of labeled neurons of layer III whose cell sizes were measured. N total number of labeled neurons observed in each example. Note that in some labeled neurons of layer III cell sizes were not measured because in Nissl stain they did not contain nuclei and appeared as fragments of cells.

124 K. SHOUMURA: fibers as shown by degeneration studies (HEATH and JONES, 1970; SHOUMURA and ITOH, 1972). Of all labeled neurons in area 17, 18 and 19 (Cat 15 and 26), 93.7-94.4% were identified in layer III and a much smaller number of neurons were located in other layers except layer I (Fig. 4). Within layer III the lower part contained more numerous labeled neurons. Figure 4 indicates that commissural efferent neurons were distributed in a considerably wide size range with the largest population at characterize the cytoarchitecture of area 17/18 boundary appeared to take up the enzyme. The major population of labeled neurons in the lateral suprasylvian cortex was located again in layer III (88.8-96.3%), where they were found preferentially in the deeper half (Fig. 2-4). Unlike labeled neurons in area 17, 18 and 19, these of the lateral suprasylvian cortex did not show any dominant peak in the histograms of cell sizes (Fig. 4). It appears from this result that the contribution to the corpus callosum is more evenly shared by a wide size range of neurons of layer III in the lateral suprasylvian cortex than by those in area 17, 18 and 19. In all examples the majority of labeled cells were pyramidal in shape (Fig. 5A) and a small number of somata were round to fusiform. In a few cells, some radially projecting dendrites were stippled with HRP-positive granules (Fig. 5B). A B Fig. 5. Two HRP-labeled neurons of approximately the same cell size. A. A pyramidal cell of layer III at area 17/18 boundary. B. A cell whose dendrites are projecting radially; this neuron was identified in layer III of medial area 18. DISCUSSION In agreement with HRP studies of other authors (MACIEWICZ, 1974; INNOCENTI and FIORE, 1976; SHATZ, 1977), commissural efferent neurons of area 17, 18, 19 and the lateral suprasylvian cortex were located preferentially in layer III with a smaller population in other layers except layer I. However, with respect to the laminar distribution of the cells of origin of commissural fibers oriented heterotopically from the lateral suprasylvian cortex to the contralateral lateral gyrus, there is a certain

Commissural Efferent Neurons in Cat Visual Cortex 125 difference between the present results and the observation of SHATZ (1977), who stated that a preponderance of them was located in layers V and VI. The size distribution of labeled cells of layer III indicates that commissural efferent neurons have a larger variation in cell size than the previous study (SHOUMURA, 1974) has suggested on the ground that the distribution and density of commissural connections are closely paralleled by an increase in the size of pyramids of layer III. The size range of commissural efferent neurons varies from the largest pyramids of layer III at area 17/18 boundary to small neurons which are comparable to those predominating in layer III of acommissural medial area 17. However, of layer III whose population increases and declines in a manner that follows quite closely the changes in the density of commissural connections participate in interhemispheric connections. Taking these results altogether, it can be concluded that the commissurally connected cortex is expressed cytoarchitectonically by an increase in the number of large to medium sized pyramids of layer III, whereas they are not the only origin of commissural fibers, but the contribution to the corpus callosum is shared by a very wide size range of neurons in layer III and by some neurons of other layers as well. This seems to hold also for the somatic sensory cortex of cats, where the differences in commissural connections between distal limb representations and the face and trunk regions are paralleled by differences in cytoarchitecture (SHANKS, ROCKEL and POWELL, 1975); in the hand representation layer III is composed of small neurons and toward the trunk and face regions large to medium sized pyramids appear in layer III. Our HRP study in the cat somatic sensory cortex (unpublished observations) indicates that the cells of origin of commissural fibers are not confined to the category of large and medium size. In the monkey visual cortex, commissural efferent neurons in area 18 bordering area 17 are restricted to layer IIIB and they are large to medium sized pyramids (WONG-RILEY, 1974; LUND et al., 1975; WINFIELD, GATTER and POWELL, 1975). JONES, BURTON and PORTER (1975) and JONES and WISE (1977) noted in the monkey somatic sensory cortex that only the large pyramids of layer III contained the HRP reaction products after multiple peroxidase injections in the opposite cortex. The difference in the laminar distribution and size rage of cells sending fibers to the opposite cortex between cats and monkeys is, thus, considerable and commissural efferent neurons appear more specialized in the latter. In the cat retina and dorsal lateral geniculate nucleus, it has been demonstrated that there is a correlation between the size of neurons and the site it projects (KELLY and GILBERT, 1975; GILBERT and KELLY, 1975); a large ganglion cell and a large geniculate neuron project to more than one structure-the superior colliculus and lateral geniculate nucleus for a large ganglion cell and area 17 and 18 for a large geniculate neuron. Then, what is the significance of an increase in the size of neurons of layer III at commissurally connected regions of the cerebral cortex? One possibility is that larger pyramids of layer III give rise to fibers, by way of collaterals, ending in both homotopically and heterotopically situated portions of the opposite cortex. In Cats 19 and 26, labeled neurons were found in heterotopical portions of the cortex (Fig. 2, 4). However, the number of labeled neurons is too small (13 neurons in Cat 19 and 36 neurons in Cat 26) to come to a definite conclusion. It was only noted that in Cat 19 labeled neurons at area 17/18 boundary were all large to me-

126 K. SHOUMURA: dium sized pyramids and layer III of the lateral suprasylvian cortex of Cat 26 con- An alternative possibility is that an axon from a large pyramid bifurcates beneath the cortex and a branch enters the corpus callosum and the other ends in a subcortical structure (projection fiber) or in other cortical fields of the same hemisphere (association fiber). At present, however, the possibility of small neurons of layer III giving rise to both commissural and association fibers cannot be ruled out because every size of labeled neuron appeared in layer III after HRP injections into either side of the cerebral cortex. Furthermore, labeled neurons have not been observed in layer III of area 17, 18, 19 and the lateral suprasylvian cortex after peroxidase injections into the subcortical structures including the lateral geniculate nucleus (GILBERT and KELLY, 1975), superior colliculus (HOLLANDER, 1974; GILBERT and KELLY, 1975; MAGALHAES-CASTRO, SARAIVA and MAGALHAES-CASTRO, 1975) and pontine nucleus (KAWAMURA, KONNO and CHIBA, personal communication). REFERENCES Bonin, G. von: The striate area of primates. J. comp. Neurol. 77: 405-429 (1942). Cragg, B. G.: The topography of the afferent projections in the circumstriate visual cortex of the monkey studied by the Nauta method. Vision Res. 9: 733-747 (1969).

Commissural Efferent Neurons in Cat Visual Cortex 127 Doty, R. W. and N. Negrao: Forebrain commissures and vision. In: (ed. by) R. Jung: Handbook of sensory physiology. Vol. VII/3B. Springer, Berlin, 1973 (p. 543-582). Economo, C. von: Zellaufbau der Grosshirnrinde des Menschen. Springer, Berlin, 1927. (p. 84-95). Gilbert, C. D. and J. P. Kelly: The projections of cells in different layers of the cat's visual cortex. J. comp. Neurol. 163: 81-106 (1975). Glickstein, M. and D. Whitteridge: Degeneration of layer III pyramidal cells in area 18 following destruction of callosal input. Brain Res. 104: 148-151 (1976). Graham, R. C. and M. J. Karnovsky: The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14: 291-302 (1966). Heath, C. J. and E. G. Jones: Connecxions of area 19 and lateral suprasylvian area of the visual cortex of the cat. Brain Res. 19: 302-305 (1970). Hollander, H.: On the origin of the corticotectal projections in the cat. Exp. Brain Res. 21: 433-439 (1974). Innocenti, G. M. and L. Fiore: Morphological correlates of visual field transformation in the corpus callosum. Neurosci. Letters 2: 245-252 (1976). somatic sensory cortex of primates. Science 190: 572-574 (1975). Jones, E. G. and S. P. Wise: Size, laminar and columnar distribution of efferent cells in the sensory-motor cortex of monkeys. J. comp. Neurol. 175: 391-438 (1977). Karol, E. A. and D. N. Pandya: The distribution of the corpus callosum in the rhesus monkey. Brain 94: 471-486 (1971). Kelly, J. P. and C. D. Gilbert: The projections of different morphological types of ganglion cells in the cat retina. J. comp. Neurol. 163: 65-80 (1975). LaVail, J. H. and M. M. LaVail: Retrograde axonal transport in the central nervous system. Science 176: 1416-1417 (1972). Lund, J. S., R. D. Lund, A. E. Hendrickson, A. Bunt and A. F. Fuchs: The origin of efferent pathways from the primary visual cortex, area 17, of the macaque monkey as shown by retrograde transport of horseradish peroxidase. J. com. Neurol. 164: 287-304 (1975). Maciewiez, R. J.: Afferents to the lateral suprasylvian gyrus of the cat traced with horseradish peroxidase. Brain Res. 78: 139-143 (1974). Magalhaes-Castro, H. H., P. E. S. Saraiva and B. Magalhaes-Castro: Identification of corticotectal cells of the visual cortex of cats by means of horseradish peroxidase. Brain Res. 83: 474-479 (1975). Shanks, M. F., A. J. Rockel and T. P. S. Powell: The commissural fibre connections of the primary somatic sensory cortex. Brain Res. 98: 166-171 (1975). Shatz, C. J.: Anatomy of interhemispheric connections in the visual system of Boston Siamese and ordinary cats. J. comp. Neurol. 173: 497-518 (1977). Shoumura, K.: An attempt to relate the origin and distribution of commissural fibers to the presence of large and medium pyramids in layer III in the cat's visual cortex. Brain Res. 67: 13-25 (1974). callosum agenesis. Brain Res. 93: 241-252 (1975). Shoumura, K. and K. Itoh: Intercortical projections from the lateral wall of the suprasylvian gyrus, the Clare-Bishop area, of the cat. Brain Res. 39: 536-539 (1972). Winfield, D. A., K. C. Gatter and T. P. S. Powell: Certain connections of the visual cortex of the monkey shown by the use of horseradish peroxidase. Brain Res. 92: 456-462 (1975). Wong-Riley, M. T. T.: Demonstration of geniculocortical and callosal projection neurons in the squirrel monkey by means of retrograde axonal transport of horseradish peroxidase. Brain Res. 79: 267-272 (1974).

128 K. SHOUMURA Zeki, S. M.: Interhemispheric connections of prestriate cortex in monkey. Brain Res. 19: 63-75 (1970). Dr. Kazuhiko SHOUMURA Department of Anatomy Gifu University School of Medicine Tsukasa-machi 40 Gifu, 500 Japan