シンポジウム発表の要旨(金子):

大脳皮質内で行なわれる情報処理過程を理解するために、皮質の局所神経回路の構成原則を明らかにしたい。そこで、情報伝達の骨格とも言うべき興奮性ニューロン間の連絡を、特に各層間のニューロンの結合を形態学的に調べる手法を紹介し、得られた若干の知見を解説する。まず、Tetramethylrhodamine-Dextran amine (TMR-DA) の逆行性取込みにより大脳皮質 V 層あるいは VI 層の出力ニューロンを標識したラット脳スライスを用意する。微小電極を用いて細胞内記録を取り、Biocytin を細胞内注入する。 固定後、多重蛍光染色法を用いて、細胞内染色されたニューロンがグルタミン酸作動性であるか、GABA 作動性であるか決定し、その後そのニューロンを黒色に染色する。一方、逆行性に標識された大脳皮質出力ニューロンを抗 TMR 抗体を用いて Golgi 染色様に赤く染色する。このようにして細胞内染色されたニューロンの皮質内軸索側枝が出力ニューロンにどの様に入力するか検討した。運動野。層のグルタミン酸作動性錐体ニューロンの軸索側枝は、V層の皮質脊髄投射ニューロンに大量に入力するのに対し、VI 層の皮質視床投射ニューロンには小量入力するだけであった。 (さらに、体性感覚野 VI 層の皮質視床投射ニューロンに対するII からVI 各層の錐体ニューロンからの入力についての結果を報告した)。

To reveal the mechanism of information processing in the cerebral cortex, we focused our efforts on the morphological analysis of its intrinsic circuitry. Recently, we developed a retrograde neuronal tracing method with dextran amines, which labeled the soma and dendrites of cortical projection neurons in a Golgi stain-like fashion (Kaneko et al., J. Neurosci. Methods, 1996, in press). By combining this method with the intracellular staining method, local connections from layer III pyramidal cells to corticospinal and corticothalamic projection neurons were examined using the cortical slices of rat primary motor area. Almost all the intracellularly labeled pyramidal cells in layer III were glutaminase-immunoreactive, thus probably glutamatergic, and had similar electrical and morphological characteristics. Twelve to nineteen percent (mean ア S.D. = 15.2 ア 3.9%, n = 4) of the axon boutons of the local axon collaterals of layer III pyramidal cells were closely apposed to the dendrites and somata of the corticospinal projection neurons. Much fewer were the close appositions of the axon boutons of layer III pyramidal cell axon collaterals on the corticothalamic neurons (3.8 ア 1.6% of axon boutons, n = 5). These results suggest that information in superficial cortical layers is transferred mainly to corticospinal neurons within the motor cortex.


大脳皮質の構築に関する研究論文のリスト:

1. Kaneko T. and Mizuno N., Immunohistochemical study of glutaminase-containing neurons in the cerebral cortex and thalamus of the rat. J. Comp. Neurol. 267: 590-602, 1988.

2. Akiyama H., McGeer P. L., Itagaki S., McGeer E. G. and Kaneko T., Loss of glutaminase-positive cortical neurons in Alzheimer's disease. Neurochem. Res. 14: 353-358, 1989.

3. Akiyama H., Kaneko T., Mizuno N. and McGeer P. L., Distribution of phosphate-activated glutaminase in the human cerebral cortex. J. Comp. Neurol. 297: 239-252, 1990.

4. Aoki C., Kaneko T., Starr A. and Pickel V. M., Identification of mitochondrial and non-mitochondrial glutaminase within select neurons and glia of rat forebrain by electron microscopic immunocytochemistry. J. Neurosci. Res. 28: 531-548, 1991.

5. Kaneko T., Nakaya Y. and Mizuno N., Paucity of glutaminase-immunoreactive nonpyramidal neurons in the rat cerebral cortex. J. Comp. Neurol. 322: 181-190, 1992.
[Abstract] Glutaminase has been considered to be a synthesizing enzyme of transmitter glutamate in pyramidal neurons of the cerebral cortex. In the present study an attempt was made to examine with a double immunofluorescence method whether non-pyramidal neurons of the cerebral cortex are immunoreactive for glutaminase or not. Glutaminase was stained with mouse anti-glutaminase IgM and FITC-labeled anti-[mouse IgM] antibody. In the same section, parvalbumin (PA), calbindin (CB), choline acetyltransferase (CAT), vasoactive intestinal polypeptide (VIP), corticotropin releasing factor (CRF), cholecystokinin (CCK), somatostatin (SS) or neuropeptide Y (NPY) was visualized as a marker for non-pyramidal neurons with an antibody to each substance, biotinylated secondary antibody and Texas Red-labeled avidin. Virtually no glutaminase-immunoreactivity was seen in PA-, CB-, CAT-, VIP-, CRF-, CCK-, SS- or NPY-immunoreactive neuronal perikarya in the neocortex and mesocortex (cingulate and retrosplenial cortices), although it was detected in a few PA-, CB-, VIP-, CCK-, SS- or NPY-immunoreactive non-pyramidal neurons in the piriform, entorhinal and hippocampal cortices. PA- and CB-positive neurons have been reported to constitute the major population of GABAergic neurons in the cerebral cortex (Celio, '86; Hendry et al., '89; Van Brederode et al., '90; Hendry and Jones, '91). Thus, the present results, together with the previous reports, suggest that most GABAergic, cholinergic and peptidergic non-pyramidal neurons in the neo- and mesocortex do not contain glutaminase.

6. Kaneko T., Shigemoto R., Nakanishi S. and Mizuno N., Morphological and chemical characteristics of substance P receptor-immunoreactive neurons in the rat neocortex. Neuroscience 60: 199-211, 1994.

7. Kang Y., Kaneko T., Ohishi H., Endo K. and Araki T., Spatiotemporally differential inhibition of pyramidal cells in the rat motor cortex. J. Neurophysiol. 71: 280-293, 1994.

8. Kaneko T., Caria M. A. and Asanuma H., Information processing within the motor cortex. I. Responses of morphologically identified motor cortical cells to stimulation of the somatosensory cortex. J. Comp. Neurol. 345: 161-171, 1994.
[Abstract] Neurons in layers II-III of the cat motor cortex were classified into 3 groups on the basis of synaptic input elicited by intracortical microstimulation (ICMS) of the somatosensory cortex (area 2). ICMS was delivered through 7 electrodes implanted in area 2. When ICMS through one of the 7 sites produced the response which was greater than 50% of the response produced by stimulating the 7 sites at a time, the site was called a ヤdominantユ site. Type I cells were those which had a dominant stimulating site in area 2 and showed a constant response latency when examined by a double shock test. Type II cells were those which had a dominant site, but displayed a variable latency with the double shock test. Type III cells had no dominant site and showed responses with variable latencies. Latency of type I responses to stimulation of area 2 was 1.2 to 2.6 msec, which was much shorter than that of type II and type III responses.
Seventy-nine neurons in layers II-III of the motor cortex, which responded to ICMS in area 2, were stained by intracellular injection of biocytin. Based on the presence of an apical dendrite and rich spines on the dendrites, 23 type I, 21 type II, and 15 type III cells were classified as pyramidal cells. Type II pyramidal cells were located more superficially than type I and type III pyramidal cells. On the basis of the absence or sparseness of dendritic spines, 3 type I and 4 type II cells in layers II-III were classified as non-pyramidal cells. These cells consisted of 5 small multipolar cells in layer II, and a large multipolar cell and a small bitufted cell in layer III. The remaining 11 cells were not classified because of insufficient staining.
Since type I and type II cells were considered to represent monosynaptic and polysynaptic responses to stimulation of area 2, respectively, the results suggested information flow in layers II-III from type I cells to more superficially-located type II cells. Type III polysynaptic responses suggest the presence of a convergent flow of of impulses inside of and/or between areas 2 and 4.

9. Kaneko T., Caria M. A. and Asanuma H., Information processing within the motor cortex. II. Intracortical connections between neurons receiving somatosensory cortical input and motor output neurons of the cortex. J. Comp. Neurol. 345: 172-184, 1994.
[Abstract] Connections between motor cortical neurons receiving somatosensory inputs from area 2 and large pyramidal cells in layer V were examined in the cat using intracellular injection of biocytin and immunohistochemistry of non-phosphorylated neurofilament proteins (npNFP). Biocytin was injected into pyramidal cells in layers II-III of the motor cortex which responded monosynaptically and polysynaptically to microstimulation of the somatosensory cortex, and subsequently stained black by the ABC method with diaminobenzidine (DAB) and nickel. By using a monoclonal antibody SMI-32 and the modified PAP method with Tris-aminophenylmethane (TAPM) and p-cresol as a chromogen, pyramidal cells in layers III and V of the motor cortex were stained red for npNFP. In particular, all the large pyramidal cells in layer V, Betz cells, displayed intense npNFP immunoreactivity not only in the perikarya but also in the dendrites.
Double staining with DAB/nickel and TAPM/p-cresol showed that biocytin-filled axon varicosities of the pyramidal cells which received monosynaptic inputs from area 2 made contacts with npNFP-positive dendrites in layers I-III around the biocytin-injected cell and in layers V-VI beneath the cell. The present results suggest that the cortico-cortical input from area 2 to pyramidal cells in layers II-III of the motor cortex is transferred to layer V pyramidal cells including Betz cells as well as to neighboring layer II-III pyramidal cells. Since tetanic stimulation of the somatosensory cortex can produce long-term potentiation in layer II-III cells of the motor cortex (Sakamoto et al., ユ87; Iriki et al., ユ91), it seems reasonable to assume that a given area of the somatosensory cortex can produce a long-lasting change in the activity of a given group of output cells in the motor cortex.

10. Kaneko T. and Mizuno N., Glutamate-synthesizing enzymes in GABAergic neurons of the neocortex: a double immunofluorescence study in the rat. Neuroscience 61: 839-849, 1994.

11. Kaneko T., Kang Y. and Mizuno N., Glutaminase-positive and glutaminase-negative pyramidal cells in layer VI of the primary motor and somatosensory cortices: A combined analysis by intracellular staining and immunocytochemistry in the rat. J. Neuorsci. 15: 8362-8377.
[Abstract] Pyramidal neurons in layer VI of the primary motor and somatosensory cortices were examined by a combined method of intracellular recording, biocytin injection and immunocytochemistry using in vitro slice preparations of rat brain. Immunofluorescence staining revealed that biocytin-injected pyramidal cells in layer VI were separated into glutaminase (PAG)-immunopositive and PAG-immunonegative cells. Although the two groups of pyramidal cells showed no statistically significant differences in passive membrane properties and spike characteristics, a clear difference was found in spike afterpotentials. Ten of twelve PAG-positive pyramidal cells showed no or a small fast afterhyperpolarization (fAHP), whereas ten of eleven PAG-negative pyramidal cells displayed a large fAHP. Depolarizing afterpotentials were observed only in PAG-positive pyramidal cells. Peroxidase staining for biocytin revealed that apical dendrites were shorter in PAG-positive pyramidal cells than in PAG-negative cells. In contrast, the arborization of basal dendrites was more developed in PAG-positive pyramidal cells than in PAG-negative cells. The main axons of all the pyramidal cells entered the subcortical white matter. The local axon collaterals of PAG-positive pyramidal cells were widely spread in the horizontal direction, whereas those of PAG-negative cells were distributed vertically along the dendritic tree. Since PAG is considered to be a marker of glutamatergic neurons in the cerebral cortex, the present results indicate that layer VI pyramidal cells are separated into glutamatergic and non-glutamatergic neurons that have different electrical properties and input-output organizations. Thus, cortical outputs from layer VI are suggested to use at least two distinct systems.


Symposium Index Page