Journal für Hirnforschung: Band 19, Heft 1 1978 [Reprint 2021 ed.] 9783112526743, 9783112526736

Citation preview

ISSN 0021 - 8359

Journal für Hirnforschung Heft 1-1978-Band 19

internationales Journal für Neurobiologie Begründet von Cécile und Oskar Vogt

Redaktion: J. Anthony, Paris • A. Hopf, Düsseldorf W. Kirsche, Berlin -J.Szentágothai, Budapest Wiss.Sekretär:J. Wenzel, Berlin

Akademie-Verlag • Berlin EVP25-M- 32105

Begründet von Cécile und Oskax Vogt Unter Mitwirkung des Cécile und Oskar Vogt Instituts für Hirnforschung in Düsseldorf und der Arbeitsgemeinschaft für vergleichende Neuroanatomie der Fédération mondiale de Neurologie (World Fédération of Neurology) Herausgeber: H . ADAM, Salzburg—Wien O. S . ADRIANOW, Moskau J . ANTHONY, P a r i s J . ARIÉNS KAPPERS,

Amsterdam

E . CROSBY, A n n Arbor A. DEWULF, C o r b e c k - L o J . ESCOLAR, Zaragoza R . HASSLER, F r a n k f u r t a. M. E . HERZOG, S a n t i a g o A. HOPF, Düsseldorf J . JANSEN, Oslo W . KIRSCHE, B e r l i n J . KONORSKI, W a r s c h a u ST. KÖRNYEY, P é c s

Le J O U R N A L F Ü R HIRNFORSCHUNG publiera des études sur la morphologie normale (anatomie, histologie, cytologie, microscopie électronique, histochimie), sur le développement du système nerveux ainsi que des études anatomiques expérimentales. On acceptera aussi des travaux du caractère de la coopération entre des domaines différents à condition qu'ils contiennent des résultats morphologiques obtenus par les méthodes de la neuromorphologie et de la neurophysiologie ou de la neuropharmacologie et de la neurochimie. Les travaux doivent contenir des acquisitions nouvelles sur l'action réciproque entre la structure et la fonction. Des études neuropathologiques seront seulement acceptées quand elles contribuent à la conaissance des structures normales, des changements structurels ou de leur signification fonctionelle. Des études sur la localisation cérebrale de phénomènes expérimentaux ou cliniques d'excitation ou de déficit (doctrine des localisations) seront également publiées par le JOURNAL F Ü R HIRNFORSCHUNG. Une partie spéciale sera réservée à la neurobiologie comparée.

M. MARIN-PADILLA, Hanover-New Hampshire J . MARÉALA, KoSice H . A. MATZKE, L a w r e n c e

D. MISKOLCZY, Tirgu Mures

G. PILLERI, W a l d a u — B e r n T . OGAWA, T o k y o

B. REXED, Upsala

H . STEPHAN, F r a n k f u r t a. M. J . SZENTÁGOTHAI, B u d a p e s t W . J . C. VERHAART, Leiden M. VOGT, Cambridge F . WALBERG, Oslo K . G . WINGSTRAND, K o p e n h a g e n E . WINKELMANN, Leipzig W . WÜNSCHER, B e r l i n A . D . ZURABASHVILI, Tbilissi

Im J O U R N A L F Ü R HIRNFORSCHUNG werden Arbeiten aus dem Gesamtgebiet der normalen Morphologie (Anatomie, Histologie, Cytologie, Elektronenmikroskopie, Histochemie) und der Entwicklungsgeschichte des Nervensystems unter Einschluß experimentell-anatomischer Arbeiten veröffentlicht. Es werden auch Arbeiten multidisziplinären Charakters aufgenommen, sofern sie morphologische Ergebnisse beinhalten, die mit neuromorphologischen und neurophysiologischen oder neuropharmakologischen bzw. neurochemischen Methoden gewonnen wurden und einen Erkenntnisgewinn hinsichtlich der Wechselwirkung zwischen Struktur und Funktion beinhalten. Neuropathologische Arbeiten werden nur angenommen, wenn sie Beiträge zur normalen Struktur, den Strukturwandlungen oder deren funktionellen Bedeutungen enthalten. Zum Publikationsgebiet des J O U R N A L F Ü R HIRNFORSCHUNG gehören auch Arbeiten, die sich mit der Zuordnung experimenteller Reiz- und Ausfallerscheinungen bzw. klinischen Symptomen zu bestimmten Strukturen des Gehirns („Lokalisationslehre") befassen. Als spezielles Publikationsgebiet ist die vergleichende Neurobiologie vorgesehen. The J O U R N A L F Ü R HIRNFORSCHUNG will publish studies on normal morphology (anatomy, histology, cytology, electron microscopy, histochemistry), on the development of the nervous system, as well experimental anatomical studies. Papers of multidisciplinary character will also be included so far as they contain morphological results which were obtained using neuromorphological and neurophysiological or neuropharmacological and neurochemical methods and provide further information on the interaction between structure and function. Neuropathological studies will only be published of they contribute to the knowledge of normal structures structurals changes or their functional significance. Papers dealing with the cerebral localization of experimental excitation and deficit phenomena or clinical symptoms (localization theory) will also be published by the J O U R N A L F Ü R HIRNFORSCHUNG. A special part of the publication is reserved for comparative neurobiology.

Bezugsmöglichkeiten des ,,Journal für Hirnforschung". Bestellungen sind zu richten — in der D D R an eine Buchhandlung oder an den AkademieVerlag, D D R - 108 Berlin, Leipziger Straße 3—4 — im sozialistischen Ausland an eine Buchhandlung für fremdsprachige Literatur oder an den zuständigen Postzeitungsvertrieb — in der B R D und Westberlin an eine Buchhandlung oder an die Auslieferungsstelle KUNST UND WISSEN, Erich Bieber, 7 Stuttgart 1, Wilhelmstraße 4 - 6 — in Österreich an den Globus-Buchvertrieb, 1201 Wien, Höchstädtplatz 3 — im übrigen Ausland an den Internationalen Buch- und Zeitschriftenhandel; den Buchexport, Volkseigener Außenhandelsbetrieb der Deutschen Demokratischen Republik, D D R - 701 Leipzig, Postfach 160, oder an den Akademie-Verlag, D D R - 108 Berlin, Leipziger Straße 3 bis 4

Journal für Hirnforschung Herausgeber: Im Auftrag des Akademie-Verlages von einem internationalen Wissenschaftlerkollektiv herausgegeben. Verlag: Akademie-Verlag, D D R - 1 0 8 Berlin, Leipziger Straße 3 - 4 ; Fernruf: 2 2 0 0 4 4 1 ; Telex-Nr.: 114420; Postscheckkonto: Berlin 350 21. Bank: Staatsbank der DDR, Berlin, Kto.-Nr.: 6836-26-20712. Chefredaktion: Prof. Dr. J . Anthony, Paris; Prof. Dr. A. Hopf, Düsseldorf; Prof. Dr. W. Kirsche, Berlin; Prof. Dr. J . Szentcigothai, Budapest. Wissenschaftlicher Sekretär: Dr. Jürgen Wenzel, Berlin. Veröffentlicht unter der Lizenznummer 1326 des Presseamtes beim Vorsitzenden des Ministerrates der Deutschen Demokratischen Republik. Gesamtherstellung: V E B Druckhaus „Maxim Gorki", D D R - 74 Altenburg. Erscheinungsweise: Das Journal für Hirnforschung erscheint jährlich in einem Band mit 6 Heften. Bezugspreis eines Bandes 180,— M zuzüglich Versandspesen (Preis für die D D R 1 5 0 , - M); Preis je Heft 3 0 , - M (Preis für die D D R 2 5 , - M). Bestellnummer dieses Heftes 1018/19/1. © 1977 by Akademie-Verlag Berlin. Printed in the German Democratic Republic. AN (EDV) 60315

ISSN 0021 • 8359

Journal

. Himforsch. 19 (1978) 1 —3

Internationales Journal für Neurobiologie

f y y

Hirnforschung

Heft 1 • 1978 • Band 19

tmidernbahn nsibt« phnen %Täshhn}

"ractus Capsula Y

opticus interna?

lì- V. I ' M J j W

In memoriam Richard Arwed Pfeifer In diesem Jahr gedenken wir des 100. Geburtstages von Richard Arwed PFEIFER, dessen Name mit der Entwicklung der Hirnforschung und der Nervenheilkunde eng verbunden ist. Der Lebensweg und der Bildungsgang P F E I F E R S sind für seine Zeit ungewöhnlich gewesen. Als Sohn einer kinderreichen Bergmannsfamilie in Brand bei Freiberg in Sachsen am 21. 11. 1877 geboren, besuchte er nach der Volksschule von 1892 bis 1898 das Lehrerseminar in Nossen. Danach trat P F E I F E R als Hilfslehrer in den Schuldienst ein. E r wurde ohne Abitur auf Grund vorzüglicher Zeugnisse zum Universitätsstudium der Pädagogik im Interesse der Erlangung einer höheren Berufsausbildung zugelassen. Von 1901 bis 1907 studierte P F E I F E R Naturwissenschaften und Mathematik an der Universität Leipzig. Hirnforschung, Bd. 19, Heft 1

Im Jahre 1905 promovierte er bei dem Psychologen W U N D T zum Doktor der Philosophie. Nach Ablegung der Staatsprüfung für das höhere Lehramt wirkte P F E I F E R als Oberlehrer in Rochlitz, Bautzen und Dresden. Sein Studium leitete einen bedeutsamen Schritt für sein weiteres Leben ein. Durch seine Doktorarbeit „Über Tiefenlokalisation von Doppelbildern" kam er mit dem Physiologen H E R I N G und dem Psychiater F L E C H S I G in Berührung. Diese Gelehrten erkannten P F E I F E R S hohe Begabung für Naturwissenschaften und Medizin. Besonders F L E C H S I G regte den jungen Oberlehrer an, Medizin zu studieren. P F E I F E R selbst fühlte sich zum Arzt berufen. Er verließ 1910 den Schuldienst und bestand als 34j ähriger die Reifeprüfung am Realgymnasium Dresden-Blasewitz. An1

2

WÜNSCHER,

W.

schließend erfolgte in Leipzig und München sein Medizinstudium. In dieser Zeit vertiefte PFEIFER, der 1915 sein ärztliches Staatsexamen ablegte, seine Beziehungen zu den Arbeiten FLECHSIGS, dem ersten Direktor der 1882 gegründeten Universitätsnervenklinik Leipzig. FLECHSIG vertrat den besonders von GRIESINGER postulierten Standpunkt „Geisteskrankheiten sind Krankheiten des Gehirns." Entsprechend dieser Auffassung erhielt die Nervenklinik den Vorläufer des späteren Hirnforschungsinstitutes, das hirnanatomische Laboratorium. Hier führte PFEIFER seine mye ogenetisch-anatomischen Untersuchungen für seine Dissertation „Über den feineren Bau des Zentralnervensystems eines Anencephalus" durch, mit der er 1915 zum Doktor der Medizin promovierte. Nach Beendigung des ersten Weltkrieges und des militärärztlichen Dienstes in einem Lazarett für Hirnverletzte trat PFEIFER 1919 in die Universitätsnervenklinik Leipzig ein. Er übernahm 1920 nach dem Ausscheiden von FLECHSIG die Leitung des hirnanatomischen Laboratoriums. Seine Ernennung zum außerplanmäßigen außerordentlichen Professor für Psychiatrie und Neurologie erfolgte im Jahr 1924. Unter dem Einfluß der Lehre FLECHSIGS wandte sich PFEIFER zunächst weiteren myelogenetisch-anatomischen Forschungen zu. Es entstand die myeloarchitektonische Bearbeitung des zentralen Abschnittes der Hörleitung — mit dieser Arbeit habilitierte er sich 1920 - der Seh- (1925) und der Tastleitung (1934). In Erweiterung seines Arbeitsgebietes publizierte PFEIFER eine Pathologie der Seh- (1930) und eine solche der Hörstrahlung (1936) einschließlich der kortikalen Seh- und Hörsphäre. Untrennbar ist sein Name mit der Angioarchitektonik des Gehirns verbunden. Mit einem von ihm erarbeiteten Verfahren gelang es PFEIFER, eine Darstellung der Gehirngefäße in einer Vollständigkeit zu erreichen, die heute noch unübertroffen ist. Seine Angioarchitektonik hat den Ruf PFEIFERS als Hirnforscher in der Welt besonders bekannt gemazht. PFEIFER führte den Nachweis, daß die Gefäße des Gehirns ein enges zusammenhängendes Netz bilden. Damit war die bis dahin allgemein anerkannte Lehre von den Cohnheimschen Endarterien und die daraus gezogenen Schlüsse für die Entstehung der Hirnerweichung unhaltbar geworden. Die ersten Ergebnisse seiner angioarchitektonischen Untersuchungen veröffentlichte PFEIFER 1927, in einem Jahr, das für ihn von hoher Bedeutung war. Er erhielt die erste planmäßige außerordentliche Professur für Hirnforschung in Deutschland und seine Wirkungsstätte hieß seit dem 8.6. 1927 „Hirnforschungsinstitut der Universität Leipzig". PFEIFER vertrat stets die Ansicht, daß die Hirn-

forschung sowohl der „zweckfreien" Grundlagenforschung als auch der Bearbeitung von der Klinik gestellter Probleme zu dienen habe. In dieser Meinung wurde er durch seine jahrzehntelange Tätigkeit als Nervenarzt bestärkt. Unter den Kliniknachfolgern FLECHSIGS, BUMKE u n d SCHRÖTER, l e i t e t e PFEIFER

außerdem die Kinderabteilung der Nervenklinik. Hier wurde seine Auffassung von den psychischen Störungen und Erkrankungen des Menschen und deren Entstehen wesentlich mitbestimmt. PFEIFER warnte seine Schüler vor unzulässiger Vereinfachung im Sinne eines psychophysischen Parallelismus. Er wies auf das Multifaktorielle im Ursachen- und Bedingungsgefüge der psychopathologischen Zustände hin. Für die psychiatrische Forschung forderte er die Berücksichtigung der naturwissenschaftlichen, psychologischen und pädagogischen Erkenntnisse. Seine Vorbildung als vorzüglicher Lehrer und sein warmherziges Wesen, die sich in seinem geschickten Umgang mit psychisch Kranken kundgaben, machten ihn frühzeitig auf die Bedeutung psychosozialer Einflüsse für das Auftreten psychischer Störungen und Erkrankungen aufmerksam. Seine Tätigkeit in der Kinderabteilung veranlaßte PFEIFER, eine Lehre der medizinischen Pädagogik zu schaffen, deren Grundsätze noch heute ihre Gültigkeit besitzen. Ein weiterer Höhepunkt des arbeitsreichen Lebens PFEIFERS lag in der Zeit nach dem zweiten Weltkrieg. A m 4. 12. 1943 zerstörte der britische Luftangriff die Nervenklinik und das Hirnforschungsinstitut der Universität. In neugewonnenen Räumen wurde am 1. 7. 1947 das Hirnforschungsinstitut praktisch wieder gegründet, für dessen Aufbau PFEIFER kein persönliches Opfer scheute. Durch die Errichtung einer neuropathologischen Prosektur wurden auch im Interesse der Klinik die Aufgaben des Institutes erweitert. Die Kräfte von PFEIFER wurden in besonderem Maße durch die Leitung der Universitätsnervenklinik gefordert, die ihm als 69jähriger im Feburar 1946 kommissarisch übertragen wurde und die er bis zum 31. 5. 1952 innehatte. Mit der Rekonstruktion dieser Klinik erwarb sich PFEIFER einen weiteren bleibenden Verdienst. Trotz aller Belastungen widmete er sich seinen wissenschaftlichen Untersuchungen, wie die in diesen Jahren erschienenen Veröffentlichungen belegen. PFEIFER war ein gütiger Lehrer, der seinen Schülern aufgeschlossen begegnete. Sein hohes ärztliches Ethos, sein pädagogisches Geschick und sein Forscherdrang sind für seine Umgebung stetige Eindrücke. Die ihm gebührende Anerkennung fand in den ihn zuteil gewordenen Ehrungen auch äußerlich ihren Niederschlag. A m 17.3. 1932 erfolgte die Wahl PFEIFERS zum Mitglied der Deutschen Akademie der Naturforscher (Leopoldina) und am 8. 11. 1949 zum

In memoriam R.

Mitglied der Sächsischen Akademie der Wissenschaften. Die Verleihung des Nationalpreises der Deutschen Demokratischen Republik im Jahre 1954 stellte die Krönung seiner Wertschätzung dar. Nach langer, schwerer, mit der ihm eigenen Geduld getragenen Krankheit verstarb P F E I F E R am 15. 3.

A. PFEIFER

3

1957. In seinem kurz bemessenen Ruhestand konnte er noch die von ihm angeregte Abtrennung des Hirnforschungsinstitutes von der Nervenklinik und seine Verselbständigung erleben. W . WÜNSCHER

l*

J . Hirnlorsch. 19 (1978) 5 - 2 0

Department of Zoology, University of Udaipur, Udaipur - 313001, India

A comparative histochemical mapping of ATPase and 5-Nucleotidase in the medulla oblongata, spinal cord and cerebellum of mouse. By

P . P . SOOD

and Manjula

LODHA

With 28 figures and 2 tables (Received 19 t h January 1977)

Summary: In the present contribution a comparative study of histochemical mapping of the distribution of adenosine triphosphatase and 5-nucleotidase in the hind brain of mouse has been made. There are many similarities and dissimilarities between the distribution of these two enzymes in various nuclei, tracts and fiber bundles. The noteworthy differences are as follows: - 1. The AP, NDNV, NI, NNH, NO AD, NOAM, NOI, NPC, T S and SG are very intensely positive for ATPase whereas, in 5-NUC study none of these nuclei demonstrates intensity of such degree. 2. Nucleus ambiguus is intensely positive for ATPase and is completely negative for 5-NUC. 3. The nucleus n. facialis is intensely positive for ATPase and is moderately positive for 5-NUC. 4. NC, GN, PN, PCI and TC are completely negative for 5-NUC. In ATPase preparations only GN is negative and the rest of the areas demonstrate intensity of various degrees. Along with these differences, similarities in the intensity of both enzymes in various nuclei, tracts and fibrous bundles also exist. An attempt has been made to correlate all the aforesaid differences and similarities in the distribution of these two enzymes with the functional nature of the various areas of hind brain of mouse.

Acknowledgement

Material and Methods

The authors express their sincere gratitude to Dr. H. B . T E W A R I , Senior Professor, Department of Zoology, University of Udaipur for general supervision and providing the laboratory facilities.

Locally collected young mice were used in this study. After decapitation, the hind brain was dissected out and fixed for 18 hours in 10% neutral formalin chilled at 4°C. 40 ¡x serial sections were cut on a freezing microtome, washed in distilled water and then incubated for l 1 / 2 hours at 37 °C for 5-nucleotidase and ATPase ( W A C H S T E I N and M E I S E L , 1957 techniques). Proper controls were also made simultaneously. The nuclei, tracts and fibrous bundles were identified on the basis of S I D M A N et al. (1971).

Introduction The present work is a continuation of the efforts made by this laboratory to describe and compare the distribution of various oxidative and hydrolytic enzymes in the hind brain of mouse (SOOD and BOHRA, 1 9 7 7 ; a n d SOOD a n d MULCHANDANI,

1977).

Earlier, quite a number of studies have been made on the distribution of ATPase and 5-nucleotidase in the hind brain of rat and squirrel monkey (TEWARI and BOURNE,

1 9 6 3 a,

b;

NANDY

and

BOURNE,

1964;

SHANTHA et al., 1 9 6 7 ; MANOCHA e t al., 1 9 6 7 ; MANO-

CHA, 1970). Comparative histoenzymological mapping of ATPase and 5-nucleotidase is lacking in literature, hence the present authors were prompted to take up study of histochemical mapping of these enzymes in serial sections of medulla oblongata, spinal cord and cerebellum. Furthermore, this study also provides information on the exact location of these enzymes in the various nuclei, tracts and fibrous bundles of the hind brain of mouse.

Results A. Spinal Cord: Much variation in the activities of 5-NUC and ATPase has been observed in spinal cord of the mouse. As for example, the nucleus dorsalis (ND; Fig. 14); Nucleus ventromedialis (NV; Fig. 14); Nucleus n. accessorii (NNA; Fig. 14) and nucleus reticularis (NRS; Fig. 14) are intensely positive for 5-NUC, whereas the same nuclei are moderately positive for ATPase (ND, NV, NNA, NRS; Fig. 1). A reverse condition is seen in substantia gelatinosa which is very intensely positive for ATPase (SG; Fig. 1) and moderately positive for 5-NUC (SG; Fig. 14). The nucleus cornucommissuralis dorsalis (DCN; Figs. 1,14); nucleus proprius cornus dorsalis (DN; Figs. 1, 14) and nucleus cornucommissuralis ventralis (VCN; Figs. 1, 14) are moderately positive for both enzymes. Nucleus posteromarginalis

6

Sood, P. P., and M. Lodha

RS NC NN A

VC N

Fig. 1. Transverse section of spinal cord (ATPase reaction).

NDN v

•KOD NCOV

(PN; Fig. 1) and nucleus cervicalis lateralis (NC; Fig. 1) are mildly positive for ATPase and completely negative for 5-NUC (PN, NC, Fig. 14). It is interesting to note that unlike other nuclei, nucleus gracilis is almost negative for both the enzymes (GN; Figs. 1, 14). B.

Cerebellum:

In ATPase preparations, amongst the 3 layers of cerebellum, the most intense reaction is seen in the molecular layer (ML; Fig. 11) and the Purkinje layer (PL; Fig. 11). On the other hand, the granular layer shows moderate activity (GL; Fig. 11). The white matter demonstrates mildly positive fibers (NFL;

Fig. 2. Transverse section of medulla oblongata at the level of area postrema (ATPase reaction).

Fig. 11). Like ATPase preparations the molecular layer is also intensely positive for 5-NUC (ML; Fig. 21) and the granular layer exhibits moderate activity (GL; Fig. 21), whereas the Purkinje layer demonstrates mild activity (PL; Fig. 21). The white matter (NFL) shows negligible activity. The central lobes of cerebellum show stronger activity of 5-NUC than the peripheral lobes (see Figs. 22, 23). C. Medulla

Oblongata:

Nucleus n. hpyoglossi:

In the ATPase preparations the boundaries of this nucleus are very sharp due to its intense activity (NNH; Figs. 2 — 5). The same nucleus is mode-

Histochemical mapping of the CNS

7

NVM

NVS

NSV HO

©

NOA 0 NCOV

NOA M

Fig. 4. Transverse section of medulla oblongata at the region of maximum development of olivary complex (ATPase reaction).

Fig. 5. Transverse section of medulla oblongata at the level of nucleus paragigantocellularis lateralis (ATPase reaction).

NPD

NPH NC L

Fig. 6. Transverse section of medulla oblongata at the level of nucleus prepositus hypoglossi (ATPase reaction).

rately positive for 5-NUC (NNH; Figs. 16, 17). In histological studies, two cell groups — a rostromedial group and a lateral group, have been identified. In both the histoenzymological preparations, however, the reactions are uniformly distributed, hence it has not been possible to differentiate these two cell groups. Nucleus interfascicularis n. hypoglossi:

The neurons of this nucleus are arranged along the intramedullary hypoglossal nerve. The nucleus de-

monstrates intense activity both for 5-NUC (NINH; Fig. 16) and ATPase (NINH; Figs. 2, 3, 12, 13). Further, the boundaries of the nucleus are not distinct in the present study, as all the surrounding nuclei also demonstrate activity of the same nature. Nucleus

intercalatus:

This nucleus is located between the nucleus n. hypoglossi and the nucleus dorsalis n. vagi. In ATPase preparations all these demonstrate intense activity of same nature, therefore, a distinct interca-

8

Sood, P. P., and M. Lodha

Fig. 7. Transverse section of medulla oblongata at the level of substantia grisea centralis medullae oblongatae (ATPase reaction). Fig. 8. Transverse section of medulla oblongata at the level of nucleus cuneatus lateralis (ATPase reaction). Fig. 9. Transverse section of medulla oblongata at the level of facialis nerve (ATPase reaction). Fig. 10. Transverse section of medulla oblongata passing through the nucleus n. facialis (ATPase reaction). Fig. 11. Section of cerebellum. Note intense activity in molecular layer (ML) and Purkinje layer (PL). The granular layer (GL) is moderately positive and nerve fiber layer (NFL) demonstrates mild activity (ATPase reaction).

latus nucleus does not appear in our study (NI; Fig. 2). The nucleus is moderately positive for5-NUC(NI; Fig. 16). Like ATPase, in this enzyme too, nucleus intercalatus does not appear as a distinct nucleus since the other two surrounding nuclei are also moderately positive (i. e. NDNV and NNH; Fig. 16). Nucleus dorsalis n. vagi:

This nucleus is situated above the nucleus intercalatus and consists of a characteristic concentration of the nerve cells. Such concentration of cells is clearly seen in 5-NUC preparations (NDNV; Fig. 28). These cells show intense activity for 5-NUC. In ATPase preparations (NDNV; Fig. 2) the nucleus is characterised by very strong activity in the nerve cells as

Histochemical mapping of the CNS

well as in the neuropil, therefore, even at high magnification the cellular arrangement is not distinct (NDNV; Fig. 13). Area

postrema:

The area postrema, situated on both sides of the 4th ventricle and dorsal to the nucleus tractus solitarius, is very intensely positive for ATPase (AP; Fig. 2). Further, its boundaries with nucleus dorsalis n. vagi are so closely mixed up due to intense reaction in both the nuclei that it is not possible to identify the area postrema as a distinct nucleus (see Fig. 2). The area postrema demonstrates a moderate activity for 5-NUC (AP; Fig. 16). Unlike ATPase, in 5-NUC study the boundaries of area postrema are clearly marked out from nucleus dorsalis n. vagi as the latter shows more intense activity than the former (see Fig. 16). Nucleus

It consists of medium to large multipolar cells with long dendrites placed in the reticular formation. The nucleus ambiguus is intensely positive for ATPase (NA; Figs. 3, 7,13) and completely negative for 5-NUC (NA; Figs. 18, 23). hypoglossi:

The nucleus prepositus is a triangular-shaped nucleus, situated dorsal to the rostral end of nucleus n. hypoglossi in the posterior sections of medulla oblongata. This nucleus is clearly seen in the histoenzymological preparations and is intensely positive both for ATPase (NPH; Fig. 6) and 5-NUC (NPH; Figs. 18, 20). Nucleus

trigemini:

This nucleus has irregular and loose outline and extends throughout the length of medulla oblongata. In the posterior sections of medulla oblongata, the nucleus is very easily identifiable in all histoenzymological preparations. The nucleus is strongly positive for ATPase (NSV; Figs. 2 - 5 , 12) and 5-NUC (NSV; Figs. 15—18, 28). Both these enzymes are mainly concentrated in neuropil. Nucleus tractus solitarius:

This nucleus, lies lateral to the dorsal nucleus of vagus, is intensely positive for ATPase (NTS; Figs. 4, 6, 7) and moderately positive for 5-NUC (NTS; Figs. 19, 20, 23). The boundaries of the nucleus are clearly visible in 5-NUC reactions as the surrounding nuclei show more intense activity (see Figs. 19, 20, 23). Nucleus tractus spinalis n. trigemini

ambiguus:

Nucleus prepositus

Nucleus tractus spinalis

vestibularis:

In the present study only the nuclear vestibularis medialis, lateralis and spinalis are recognised. All these three nuclei are intensely positive for ATPase (NVL, NVM, NVS; Figs. 4, 5, 7 - 1 0 ) as well as for 5-NUC (NVL, NVM, NVS; Figs. 18, 2 3 - 2 5 , 27). Nucleus n. facialis:

This nucleus shows oval to ovoid shape outline with welldefined boundaries. In the histoenzymological preparations the nucleus n. facialis is clearly visible (Figs. 8, 22, 23). I t is strongly positive for ATPase (NNF; Fig. 8) and moderately positive for 5-NUC (NNF; Figs. 22, 23). Further, in histological studies, it is demonstrated that its neuropil is permeated by longitudinal fiber bundles particularly at its lateral part. Such fibers are not seen in 5-NUC study. However, in some sections there appears to be some ATPase positive fibers crossing through the nucleus n. facialis (see Fig. 8).

9

oralis:

This nucleus lies on the ventro-lateral side in some anterior sections of medulla oblongata and like nucleus tractus spinalis trigemini, demonstrates intense activity both for ATPase (NO; Figs. 8,10) and 5-NUC (NO;Figs. 23, 24, 25). Cochlearis complex:

Two cochlear nuclei are identified in present study namely Nucleus cochlearis dorsalis (NCD) and ventralis (NCV). Nucleus cochlearis dorsalis is an elongated and oval-shaped nucleus whereas the ventral nucleus is ovoid and extends to the caudal end of pons. In ATPase preparations both the cochlear nuclei are intensely positive (NCD, NCV; Figs. 7, 9, 10). In contrast to this, the 5-NUC shows a mild activity (NCD, NCV; Figs. 2 2 - 2 5 ) . Nucleus olivaris complex:

The complex consists of nucleus olivaris accessorius dorsalis (NOAD) and medialis (NOAM), nucleus olivaris inferior (NOI) and superior (NOS). These nuclei are composed of relatively small to medium sized cells with numerous short branching dendrites, In this study nucleus olivaris accessorius medialis (NOAM; Figs. 2 - 4 , 1 2 ) , dorsalis (NOAD; Figs. 4 - 7 ) , inferior (NOI; Fig. 4) and superior (NOS; Fig. 9) are very intensely positive for ATPase whereas, these nuclei are mildly positive for 5-NUC (NOAM, NOAD, NOI; Figs. 16—18,28) except nucleus olivaris superior which shows intense reaction (NOS; Figs. 25, 26). The boundaries of these nuclei, in both the cases, are clearly marked out. Nucleus cuneatii medialis and lateralis:

In ATPase preparations the nucleus cuneatus lateralis is moderately positive (NCL; Figs. 6, 8, 9). The neuro-

10

Sood, P. P., and M. Lodha

NCM

NCOD NCOV NINH Fig. 12. High magnification of a section of medulla oblongata passing through nucleus n. hypoglossi. Note intense activity of ATPase in neuropil of various nuclei (ATPase reaction).

NOAM NCM

Fig. 13. High magnification of figure 3. Note intense activity of ATPase in cytoplasm and nucleoli of the neurons of various nuclei (NINH, NNA, NS, NCM). Further note positive activity in the neuropil of all the nuclei.

pil and the neurons show positive reaction. 5-NUC preparations also demonstrate similar activity (NCL; Figs. 19, 20). The nucleus cuneatus medialis lies dorsal to the nucleus parvocellularis compactus and has irregular obundaries. This nucleus is intensely positive for

ATPase (NCM; Figs. 2, 12, 13) and moderately positive for 5-NUC (NCM; Fig. 16). Nuclei

raphe:

In the present study two parts of nuclei raphe — namely nuclei raphe obscurus (NRO) and nuclei raphe

Histochemical mapping of the CNS

N PC

NDNV

11

AP

Fig. 14. Transverse section of spinal cord demonstrating 5-NUC activity. Fig. 15. Transverse section of medulla oblongata through the posterior region (5-NUC reaction). Fig. 16. Transverse section of medulla oblongata at the level of area postrema (5-NUC reaction). Fig. 17. Transverse section of medulla oblongata at the region of maximum development of olivary complex (5-NUC reaction).

magnus (NRM) which lie along the midline of the medulla oblongata, have been identified. The former extends throughout the length of medulla while the latter appears in the anterior sections only. In histo-

enzymological preparations both these nuclei are intensely positive for ATPase (NRO, NRM; Figs. 2 - 1 0 ) and mildly positive for 5-NUC (NRO, NRM; Figs. 1 5 - 1 8 , 22-25). Nucleus

sub-trigeminalis:

The nucleus lies ventral to the nucleus tractus spinalis n. trigemini (NSV) in the medial sections of medulla oblongata. Its boundaries are clearly seen in ATPase (NS; Figs. 2 - 4 , 1 3 ) , as well as in 5-NUC (NS;Figs. 18, 28) reactions since the nucleus demonstrates more intense activity for both the enzymes than the surrounding areas.

12

Sood, P. P., and M. Lodha

Fig. 19. Transverse section of medulla oblongata at the level of nucleus paragigantocellularis lateralis (5-NUC reaction).

Fig. 20. Transverse section of medulla oblongata passing through nucleus prepositus hypoglossi (5-NUC reaction). Nucleus centralis medulla

oblongatae:

Two sub-divisions of this nucleus, generally known as nucleus centralis medullae oblongatae pars dorsalis (NCOD) and pars ventralis (NCOV), have been recognised in the central region of reticular formation (Koikegami, 1957). Like other reticular nuclei the NCOD and NCOV also demonstrate intense activity for both the enzymes. NCOV; Figs. 2 - 5 , 1 2 , 1 5 - 1 7 , 28). Nucleus reticularis

medullae

oblongatae:

Histologically, the reticular formation throughout the medulla oblongata can be differentiated into four main nuclei — namely formatio reticularis, nucleus

reticularis paramedianus, nucleus reticularis gigantocellularis and nucleus reticularis lateralis. a) Nucleus reticularis gigantocellularis is located rostral to the nucleus reticularis paramedianus and has characteristic large cells with elongated dendrites, as well as medium-sized and small cells. This nucleus is intensely positive both for ATPase (NG; Figs. 6, 8) and 5-NUC (NG; Figs. 20, 22, 26). b) Nucleus paragigantocellularis dorsalis: According to OLSZEWSKI and B A X T E R ( 1 9 5 4 ) the dorsal area above the nucleus reticularis gigantocellularis which contains hardly any gigant cells, is known as nucleus

Histochemical mapping of the CNS

13

Fig. 21. Transverse section of cerebellum demonstrating intense activity of 5-NUC in molecular layer (ML), moderate activity in granular layer (GL) and mild activity in nerve fiber layer (NFL) and Purkinje layer (PL).

NRM

Fig. 22. Transverse section of medulla oblongata passing through nucleus raphe magnus (5-NUC activity).

Fig. 23. Transverse section of medulla oblongata passing through the anterior region of medulla oblongata (5-NUC reaction).

paragigantocellularis dorsalis. Like other reticular nuclei it demonstrates intense activity both for 5-NUC (NPD; Figs. 19, 20, 22, 26) and ATPase (NPD; Figs. 6, 8). c) Nucleus paragigantocellularis lateralis is located ventrolaterally and is oval to ovoid-shaped. It is distinguished from the rest of the reticular nuclei by strong ATPase reactions (NPL; Figs. 5, 6, 8). In 5-NUC preparations the nucleus is also intensely positive (NPL; Figs. 19, 20). d) Nucleus reticularis paramedianus is located on either side of nucleus raphe. Its cells are parallely

arranged with the nucleus raphe. The nucleus is moderately positive for ATPase (NRP; Figs. 2, 13) as well as for 5-NUC (NRP; Figs. 15, 16, 28). Substantia grisea centralis medullae oblongatae :

The nucleus lies at the mid-dorsal area of medulla oblongata. In the present study there is no clear-cut demarcation of the boundaries of this nucleus as the nucleus and the surrounding areas are intensely positive for ATPase (SGO ; Fig. 7). In 5-NUC preparations the substantia grisea centralis medullae oblongatae is moderately positive '(SGO; Fig. 23) and similar is the condition of its surrounding nuclei.

14

Sood, P. P., and M. Lodha

NVM „. NCD

Fig. 24. T. S. through nucleus n. facialis (5-NUC reaction) .

nrm

NVM

NRM Nucleus parvocellularis:

The nucleus consists of densely packed cells (NPC; Figs. 2, 12, 16). In the anterior sections the cells of the nucleus are loosely arranged and its position is somewhat lateral (NP; Figs. 6, 8, 20, 22, 23). As far as the intensity of the ATPase is concerned, the nucleus parvocellularis compactus is very intensely positive (NPC; Figs. 2,12,13) as compared to nucleus parvocellularis (NP; Figs. 6, 8). In 5-NUC preparations the intensity is the reverse; here the nucleus parvocellularis demonstrates stronger activity than the nucleus compactus. In other words the nucleus parvocellularis is intensely positive (NP; Figs. 20,

Fig. 25. T. S. through most anterior region of medulla oblongata (5-NUC reaction).

22, 23) and the nucleus parvocellularis compactus is mildly positive for 5-NUC (NPC; Fig. 16). Facial nerve — VII:

This nerve is intensely positive for ATPase (NF; Fig. 9) and moderately positive for 5-NUC (NF; Figs. 22, 24, 26, 27). Tracts:

Though in histological studies a large number of tracts have been identified in the hind brain of mouse (see Sidman et al., 1971) but in our histoenzymological preparations only few tracts are seen. These are

Histochemical mapping of the CNS

15

Fig. 26. High magnification of a section of medulla oblongata a t t h e level of nervus facialis. Note intense activity of 5-NUC in t h e cytoplasm and nucleoli of t h e neurons of various nuclei (5-NUC reaction).

Fig. 27. High magnification of a section of medulla oblongata a t the level of nervus facialis (5-NUC reaction) .

— tractus spinocerebellaris dorsalis (TSD) and ventralis (TSV), tractus spinalis n. trigemini (TST), tractus corticospinalis (TC) and tractus solitarius (TS). Of these five tracts, in ATPase preparations, the most intense reaction is seen in tractus solitarius (TS; Fig. 4), whereas the same tract is moderately positive for 5-NUC (TS; Fig. 17). Tractus spinalis n. trigemini is moderately positive for ATPase (TST; Figs. 2 - 7 , 10) and mildly positive for 5-NUC (TST; Figs. 22, 24, 25, 28). The tractus spinocerebellaris dorsalis and ventralis are moderately positive for ATPase (TSD, TSV; Figs. 2, 13) and intensely posi-

tive for 5-NUC (TSD, TSV; Fig. 28). Finally, the tractus corticospinalis demonstrates moderate activity for ATPase (TC; Figs. 3, 6, 12) while the same tract is completely devoid of 5-NUC activity (TC; Figs. 16, 17, 19, 20, 23, 26). Discussion Evaluating the different patterns of enzymatic localization in the hind brain of mouse, it appears that in general the activity of ATPase is more intense in most of the nuclei and tracts than 5-NUC activity,

16

Sood, P. P., and M. Lodha

NDNV

NCOD

NCOV

Fig. 28. High magnification of a section of medulla oblongata at the level of nucleus olivaris accessories medialis (5-NUC reac-

NOAM T a b l e 1. Showing the various nuclei, figure numbers and the intensity of the enzymes. Nuclei

Area postrema Nucleus ambiguus Nucleus cornucommissuralis dorsalis Nucleus cochlearis dorsalis Nucleus cervicalis lateralis Nucleus cuneatus lateralis Nucleus cuneatus medialis Nucleus centralis medullae oblongatae pars dorsalis Nucleus centralis medullae oblongatae pars ventralis Nucleus cochlearis ventralis Nucleus cornucommissuralis ventralis Nucleus dorsalis Nucleus dorsalis n. vagi Facial nerve V I I Nerve fiber layer Nucleus gracilis Nucleus reticularis gigantocellularis Nucleus intracalatus Nucleus interfascicularis n. hypoglossi Nucleus n. accessorii Nucleus n. facialis Nucleus n. hypoglossi Nucleus tractus spinalis n. trigemini oralis

Abbreviation

ATPase activity Fig. No. Intensity

5-Nucleotidase activity Fig. No. Intensity

2.

3.

4.

5.

6.

AP NA DCN

+++

2 3, 7, 13 1

+ -

16 18, 23 14

NCD NC NCL NCM NCOD

++ + ++ ++

7, 1 6, 2, 2,

NCOV

++

2, 4, 5, 12

NCV VCN

++ + -

9, 10 1

ND NDNV NF NFL GN NG

+ +++ ++

++ ++ + -

++

1 2, 13 9 11 1 6, 8

++

14 28 22, 24, 26, 27 21 14 20, 22, 26

NI NINH

++ + ++

2 2, 3, 12, 13

+ ++

16 16

NNA NNF NNH NO

+ ++ +++ ++

1 8 2-5 8, 10

++

14 22, 23 16, 17 23-25

+ 4-

+ -

+

—b

9, 10 8, 9 12, 13 3 - 5 , 12

+ + + ++

22-25 14 19, 20 16 1 5 - 1 7 , 28

++

1 5 - 1 7 , 28

-

-

+

+

+ ++

24 14

Histochemical mapping of the CNS

17

Continuation T a b l e 1 : Nuclei 1. Nucleus olivaris accessorius medialis Nucleus olivaris inferior Nucleus olivaris superior Nucleus parvocellularis Nucleus parvocellularis compactus Nucleus paragigantocellularis dorsalis Nucleus proprius cornus dorsalis Nucleus prepositus hypoglossi Nucleus posteromarginalis Nucleus paragigantocellularis lateralis Nucleus raphe magnus Nucleus raphe obscurus Nucleus olivaris accessorius dorsalis Nucleus reticularis paramedianus Nucleus reticularis Nucleus sub-trigeminalis Nucleus tractus spinalis trigemini Nucleus tractus solitarii Nucleus ventromedialis Nucleus vestibularis lateralis Nucleus vestibularis medialis Nucleus vestibularis spinalis Granular layer Molecular layer Pedunculus cerebellaris inferior Purkinje layer Substantia gelatinosa Substantia grisea centralis medullae oblongatae Tractus corticospinalis Tractus solitarius Tractus spinocerebellaris dorsalis Tractus spinalis n. trigemini Tractus spinocerebellaris ventralis

Abbreviation 2. NOAM NOI NOS NP NPC NPD DN NPH PN NPL NRM NRO NOAD NRP NRS NS NSV NTS NV NVL NVM NVS GL ML PCI PL SG SGO TC TS TSD TST TSV

ATPase activity Fig. No. Intensity 4. 3.

+++ +++ +++ ++ +++ ++ +++ ++ ++ ++ +++ ++++ ++ ++ + ++ ++ ++ +++ ++ ++ +++ ++ + +++ + ++-

+

2 - 4 , 12 4 9 6, 8 2, 12, 13 6, 8 1 6 1 5, 6, 8 8-10 2 - 7 , 9, 10 4, 7 2, 13 1 3, 4, 13 2 - 5 , 12 4, 6, 7 1 7, 9, 10 4, 5, 7 — 10 4, 5 11 11 6, 7 11 1 7 3, 6, 12 4 2, 13 2, 3, 5 - 7 , 10, 13 2

5-Nucleotidase activity Intensity Fig. No. 5. 6. -

+

-

+

-

+

++ ++ ++ + ++ ++ —h -

+

-

+

+++ ++ ++ + ++ ++ ++ ++ +++

1 6 - 1 8 , 28 17, 18 25, 26 20, 22, 23 16 19, 20, 22, 26 14 18, 20 14 19, 20 22, 24, 25 1 5 - 1 8 , 23, 25 16-18 15, 16, 28 14 18, 28 1 5 - 1 8 , 28 19, 20, 23 14 2 3 - 2 5 , 27 18. 2 3 - 2 5 , 27 18 21 21 17, 20, 22

++ -

21 14 23

+ ++

16, 17, 19, 20, 17 28

-

-

+

+

++

22, 24, 25, 28 28

Note: For abbreviations see table II.

as for example few nuclei (AP, NDNV, NI, NNH, NOAD, NOAM, NOI, NPC, TS), are very intensely positive and in such nuclei it has not been possible to distinguish the various elements of the area. It is interesting to note that such a condition is not seen in 5-NUC preparations. In all we have studied fifty-six nuclei, tracts and fibrous bundles (seeTable 1), out of which eleven areas are very intensely positive Hirnforschung, Bd. 19, Heft 1

for ATPase and none for 5-NUC; twenty-seven are intensely positive for ATPase and twenty three for 5-NUC; fourteen nuclei are moderately positive while three are mildly positive for ATPase. In 5-NUC preparations sixteen nuclei are moderately positive and eleven are mildly positive. One nucleus of spinal cord, i.e., nucleus gracilis, is negative for ATPase while the number of negative nuclei is six in 5-NUC 2

18

Sood, P.,P. and M. Lodha

study. Thus an overall picture makes it clear that intensity of ATPase in various nuclei, tracts and fibrous bundles is stronger than 5-NUC (see Table 1). The ATPase activity appears in glial cells, blood vessels, nerve cells, neuropil and nerve fibers. In other words most of the structural elements of the medulla oblongata are positive for ATPase. In nerve cells the activity of ATPase is restricted mainly to neuronal membranes and nucleoli. There appears to be little activity in the cytoplasm while the nuclei are generally free from enzymatic activity. Further, the neuropil is more strongly positive as compared to neurons. Earlier BARRON and TUNCBAY (1962) and NANDY and BOURNE (1964) h a v e m a d e similar obser-

vations. In ATPase preparations few nerve fibers and tracts demonstrate moderate to intense activity. Most of the nerve fibers, fibrous bundles and tracts could not be identified in the present series of sections as either they are so intense that they have been mixed up with intensely positive nuclei or are completely devoid of enzymatic activity. Though detailed studies about the presence of ATPase and 5-NUC in tracts and fibre bundles of hind brain of mammals are not available in literature, yet few attempts have been made to demonstrate these enzymes in fibrous bundles and tracts of the central nervous system of other

olfactory bulb of mouse. However, TEWARI and RAJBANSHI (1972) found negligible a c t i v i t y of 5 - N U C

and ATPase in the tracts of hind brain of Saccobranchus fossilis, though they found intense activity of 5-NUC and mild activity for ATPase in the 10th cranial nerve. In literature there are different opinions about the presence of 5-NUC and ATPase in the white and grey matter. In biochemical studies, great species variability in the regional distribution of 5-NUC in brain tissue is suggested by the ratio of the activity in white matter to that of grey matter. REIS (1937) found this ratio is 1.0 to 4.2 for man, 1.0 to 0.9 for rabbit and 1.0 to 0.4 for calves. In histochemical studies of 5-NUC and ATPase

vertebrates. Earlier, TEWARI and SOOD (1974) and SOOD and TEWARI (1972 a; 1976) h a v e demon-

strated intense activity of ATPase and 5-NUC in various olfactory tracts of frog and toad. SOOD and TEWARI (1972 b) have also seen intense activity of 5-NUC in lateral and medial olfactory tracts in the

NAIDOO

(1962)

found

enzyme

activity

exclusively in myelinated fibers in mice. TEWARI and BORNE (1963 a) showed intense a c t i v i t y of 5 - N U C

in the cerebellar white matter of rat. BARRON and TUNCBAY (1964) f o u n d 5 - N U C in the glia of white

matter of cats, with much higher activity in the neuropil of grey matter. Intense activity of 5-NUC in the white matter of mouse brain is also seen by (SCOTT 1967). Similarly, intense activtiy of A T P a s e

in the white matter of rat has been reported by ToRACK (1965) and MANOCHA (1970). Further, BONTING et al. (1961) demonstrated a large amount of

Na—K ATPase in the grey matter of nervous tissue. Since the reticular formation is formed by intermingling of grey and white matter and all above data including present study support our view that both the enzymes are present in the grey and white matter of the brain, intense activity of 5-NUC and ATPase in reticular nuclei is expected. Further, the reticular

T a b l e II Showing the differences of the distribution of ATPase and 5-NUC in the various layers of cerebella of different mammals. Name of Authors

Molecular layer

Purkinje layer

Granular layer

Nerve fiber layer

ATPase

5-NUC

ATPase

5-NUC

ATPase

5-NUC

ATPase

5-NUC

NA NA

++ ++

NA NA

NA

NA NA

NA

+ -

+ -

NA NA + +

Becker et al. (1960) Rat. ++ Tewari and Bourne (1963a, b) + Rat. Shantha et al. (1967) Squirrel + monkey NA Scott (1967) Mouse

+ -

+

+ ++

Manocha (1970) Squirrel monkey Present study Mouse

+ -

++

++ = = = =

-

Low to NA moderate NA + to + -

Very Intense NA

+++

+ -

to + + Low to NA moderate activity

NA

-|— —|NA

+

to + — to + + Show -five NA & — ive bands

Scott (1967) Cat.

Abbreviations :

-

+ -

NA

-

+

+ -

to + + —ive except nerve fibers Negative NA

+ -

Very intensely positive ; + + = Intensely positive Moderately positive; Mildly positive; —— = Negative; Not attempted.

NA

Moderately strong

NA

Negative

+ -

NA

-

+

-

+

Histochemical mapping of the CNS nuclei a r e b o t h s e n s o r y a n d m o t o r in n a t u r e (CAJAL, 1 9 1 1 ) , therefore, t h e m o t o r a n d sensory functions of 5 N U C a n d A T P a s e in t h e r e t i c u l a r f o r m a t i o n seems t o b e negligible,

instead the

e n z y m e s in t h e

mouse

b r a i n m a y b e c o n c e r n e d w i t h general m e t a b o l i s m . I n cerebellum, intense a c t i v i t y of A T P a s e a n d 5 N U C is seen in m o l e c u l a r l a y e r , m o d e r a t e a c t i v i t y in g r a n u l a r l a y e r a n d m i l d t o negligible a c t i v i t y in w h i t e m a t t e r . H o w e v e r , t h e a c t i v i t y of 5 - N U C a n d A T P a s e in P u r k i n j e l a y e r is quite different.

The

P u r k i n j e cells d e m o n s t r a t e intense a c t i v i t y for A T P a s e a n d m i l d a c t i v i t y for 5 - N U C . W h i l e c o m p a r i n g t h e a c t i v i t y of 5 - N U C a n d A T P a s e of m o u s e cerebell u m w i t h o t h e r m a m m a l s like squirrel m o n k e y (SHANTHA e t al. 1 9 6 7 ; MANOCHA, 1 9 7 0 ) ; r a t (TEWARI a n d BOURNE, 1 9 6 3 a , b) a n d c a t (SCOTT, 1 9 6 7 ) , large n u m b e r of differences in t h e i n t e n s i t y of b o t h t h e e n z y m e s in v a r i o u s l a y e r s of t h e cerebellum a r e d e t e c t e d (see T a b l e I I ) , which suggest t h a t t h e r e is a species v a r i a tion of t h e distribution of t h e s e e n z y m e s in t h e differ e n t l a y e r s of cerebellum of m a m m a l s . T h e f u n c t i o n a l significance of 5 - N U C in t h e cerebellum o r in a n y o t h e r n e r v o u s tissue is n o t well known. T h e role of 5 - N U C in inhibition a n d e x c i t a t i o n o r in afferent o r efferent functions h a s been e x p e c t e d by

SCOTT ( 1 9 6 7 ) ,

though

he

doubts

the

relation

of this e n z y m e in e i t h e r of these pairs of functions. T h e o t h e r role of 5 - N U C , like nucleic a c i d c a t a b o l i s m a n d b r e a k d o w n of N A D P (DIXON a n d W E B B , HARDONK, (REIS,

1968),

1951),

in

the

in t r a n s p o r t

regulation

of

mechanism

1964;

glycolysis (HARDONK,

1 9 6 8 ) , in cell g r o w t h (EKER, 1 9 6 5 , 1 9 6 6 ; FRITZSON, 1 9 6 7 ; HARDONK a n d KOUDSIAAL, 1 9 6 8 ) , in different tissues h a s been r e p o r t e d . B u t h o w far 5 - N U C in t h e cerebellum is c o n c e r n e d w i t h t h e s e functions or w i t h e x c i t a t i o n a n d inhibition, is n o t y e t k n o w n .

References K. D. and T. O. T U N C B A Y : Phosphatase in cuneate nuclei after brachial plexectomy. Arch. Neurol. (Chic). 7, 2 0 3 - 2 1 0 (1962). Phosphatase histoB A R R O N , K . D. and T. O. T U N C B A Y : chemistry of feline cervical spinal cord after brachial plexectomy. J . Neuropath, and Exper. Neurol. 23, 368 (1964). BARRON,

BECKER, N. H.,

S . GOLDFISCHER,

W . Y . SHIN

and

A.

B.

The localization of enzyme activities in the rat brain. J . Biophys. Biochem. Cytol. 8, 649—663 (1960). B O N T I N G , S . L., K . A. S I M O N and M. H . H A W K I N S : Studies on sodium-potassium activated adenosine triphosphatase. I. Quantitative distribution in several tissues of the cat. Arch. Biochem. 95, 4 1 6 - 4 2 3 (1961). C A J A L , S. R . Y . : Histologic du systeme nerveux de l'homme NOVIKOFF:

et des vertebres. Norbert Maloine, Paris, 2 vols (1911). and E . C . W E B B : Enzymes, 2nd edition. Longmans, Green and Co. Ltd., London (1964). EKER, P . : Activities of thymidine kinase and thymidine DIXON, M.,

19

deoxyribonucleotide phosphatase during growth of cells in tissue culture. J . biol. Chem. 240, 2607 — 2611 (1965). EKER, P . : Studies on thymidine kinase of human liver cells in culture. J . biol. Chem. 241, 6 5 9 - 6 6 2 (1966). F R I T Z S O N , P . : Dephosphorylation of pyrimidine nucleotides in the soluble fraction of homogenates from normal and regenerating rat liver. Eur. J . Biochem. 1, 12 — 20 (1967). H A R D O N K , M. J . : I. Distribution of 5'-nucleotidase in tissue of rat and mouse. Histochemie. 12, 1 — 17 (1968). H A R D O N K , M. J . and J . K O U D S T A A L : I I . The significance of 5'nucleotidase in the metabolism of nucleotides studied by histochemical and biochemical methods. Histochemie. 12, 1 8 - 2 8 (1968). KOIKEGAMI, H . : On the correlation between cellular and fibrous patterns of the human brain stem reticular formation with some cytoarchitectonic remarks on the other mammals. Acta Medica et Biologica 5, 21 — 72 (1957). M A N O C H A , S . L . , T . R . S H A N T H A and G . H. B O U R N E : Histochemical studies on the spinal cord of squirrel monkey (Saimiri seiweus). E x p . Brain Res. 3, 25 — 39 (1967). M A N O C H A , S . L . : Histochemical distribution of alkaline and acid phosphatase and adenosine triphosphatase in the brain of squirrel monkey (Saimiri sciureus). Histochemie. 21, 2 2 1 - 2 3 5 (1970). N A N D Y , K . and G . H. B O U R N E : Adenosine triphosphatase and 5-nucleotidase in spinal cord — Histochemical study of localization in rat spinal cord. Archives of Neurology. 11, 5 4 7 - 5 5 3 (1964). N A I D O O , D . : The activity of 5-nucleotidase determined histochemically in the developing rat brain. J . Histochem. Cytochem. 10, 4 2 1 - 4 3 4 (1962). O L S Z E W S K I , J . and D . B A X T E R : Cytoarchitecture of the human brain stem. J . B . Lippincott Company, Philadelphia (1954). R E I S , J . L . : Über die spezifische Phosphatase der Nervengewebe. Enzymologia. 2, 1 1 0 - 1 1 6 (1937). REIS, J . L . : The specificity of phosphomonoesterases in human tissue. Biochem. J . 48, 548 — 551 (1951). S C O T T , T. G . : The distribution of 5'-nucleotidase in the brain of the mouse. J . Comp. Neurol. 129, 97 — 108 (1967). S H A N T A , T . R . , K . I I J I M A and G . H. B O U R N E : Histochemical studies on the cerebellum of squirrel monkey. Acta Histochem. 27, 1 2 9 - 1 6 2 (1967). S I D M A N , R . L., J . B . A N G E V I N E and E . T. P I E R C E : Atlas of the mouse brain and spinal cord. A Commenwealth Fund Book. Harvard Univ. Press, Cambridge, Massachusetts. pp. 1 - 2 6 1 (1971). S O O D , P . P . and M . H. B O H R A : Histochemical mapping on the distribution of simple esterase and acetylcholinesterase in the medulla oblongata, cerebellum and spinal cord of mouse. J . Hirnforschung 18,75—87 (1977). S O O D , P. P. and M . M U L C H A N D A N I : A comparative histoenzymological mapping on the distribution of acid and alkaline phosphatases and succinic dehydrogenase in the spinal cord, and medulla oblongata of mouse. Acta Histochem. 60, 1 8 0 - 2 0 3 (1977). S O O D , P. P. and H. B . T E W A R I : Histochemical mapping of the distribution of acid phophatase, 5-nucleotidase and nonspecific esterase in the forebrain of the toad. Brain Res. 38, 4 0 7 - 4 2 0 (1972a). S O O D , P. P., and H. B . T E W A R I : Study on the distribution of some hydrolytic enzymes and their functional significance in the olfactory bulb of mouse. Acta Histochem. 43, 2 1 8 - 2 3 4 (1972b). S O O D , P. P., and H. B . T E W A R I : Histochemical mapping of the distribution of acid phosphatase and 5-nucleotidase in 2*

20

Sood, P. P., and M. Lodha

the forebrain of frog (Rana tigrina). J . Hirnforschung. 17, 289-303 (1976).

TEWARI, H. B., and G. H. BOURNE: Histochemical studies

on the distribution of alkaline and acid phosphatases and 5-nucleotidase in the cerebellum of rat. J . Anat. (London) 97, 6 5 - 7 2 (1963a).

TEWARI, H . B . , a n d G. H . BOURNE : H i s t o c h e m i c a l

studies

on the localization of adenosine triphosphatase in the cerebellum of the rat. J . Histochem. Cytochem 11, 246—256 (1963 b).

TEWARI, H . B . ,

and

V . K . RAJBANSHI:

Histological

and

TORACK, R. M. : The relationship between adenosine triphosphatase activity and triethyltin toxicity in the production cerebral edema of the rat. The American J . of Pathol. 46, 245-261 (1965). WACHSTEIN, M., and E. M E I S E L : Histochemistry of hepatic phosphatases at the physiologic pH with special reference to the demonstration of bile canaliculi. Amer. J . Clin. Path. 27, 1 3 - 2 3 (1957).

histochemical studies in the hind brain of a freshwater teleost, Saccobranchus fossilis. Ann. Histochem. 17, 1 — 25 Authors address: (1972). TEWARI, H. B., and P. P. SOOD: Histochemical mapping of D r . P . P . SOOD, the distribution of adenosine triphosphatase, succinic deDepartment of Zoology, hydrogenase and non-specific esterase in the fore-brain of University of Udaipur, frog (Rana tigrina). J . Hirnforschung. 15,129 — 142(1974). Udaipur-313001, India.

J . Hirnforsch. 19 {1978) 21 - 4 3

Department of Anatomy, University of Alabama in Birmingham

An Experimental Study of the Central Gustatory Pathways in the Monkey, Macaca mulatta and Cercopithecus aethiops B y Francis Cleveland KINNEY, Ph. D. 1 With 19 figures (Received 20 t h January 1977)

Summary: 1. The cortex at the base of the central fissure, that is the fronto-parietal operculum, represents primary sensory receptive cortex for gustatory modalities. 2. This primary receptive cortex for taste is linked to the anterior Island of Reil by short inter- and intracortical association fibers. Thus the anterior island is a gustatory association area involved in the subjective recognition of gustatory modalities. 3. The nucleus ventralis posteromedials pars parvocellularis of the dorsal thalamus is the thalamic receptive nucleus for gustatory impulses in the macaque. 4. Fibers which originate at rostral glossopharyngeal levels from the dorsal visceral gray terminate in the contralateral nucleus ventralis posteromedialis pars parvocellularis.

Acknowledgements I am especially grateful and appreciative for the constant guidance and direction so freely given by my teacher, Dr. Elizabeth C . C R O S B Y . I am indebted to Dr. Earl G. H A M E L , J r . , chairman of my graduate committee, for his continual help and support. For his contributing to the Department of Anatomy of the University of Alabama in Birmingham all the rhesus monkeys used in this study, I owe a special thanks to Dr. Leon H. SCHMIDT, of the Southern Research Institute.

Introduction Relatively little research has been completed to determine and document the central (neural) gustatory pathways in primates and man. This is surprising in view of the obvious importance of taste. The loss of taste or one of the gustatory sensations (such as the ability to recognize salty, sweet, sour or bitter substances) may, if no peripheral lesion can be located, be indicative of a lesion within the central nervous system. If the pathways involved in transmitting gustatory impulses to higher centers, including the cerebral receptive and association cortices for the cortical recognition of taste, are known, then the loss of taste or better still the loss of one of the 1

This investigation was completed as partial fulfillment for the Doctor of Philosophy Degree in Anatomy at the University of Alabama in Birmingham.

modalities of taste will give the clinician a clue as to the possible level of the lesion. The cell bodies of gustatory fibers are found in the geniculate ganglion of the facial nerve, the inferior (opetrosal) ganglion of the glossopharyngeal nerve, and the inferior (or nodose) ganglion of the vagus nerve. The central processes of these fibers enter the brain stem (at pons and medulla levels) and synapse, at or near their level of entrance, in the nucleus (dorsal visceral gray) of the tractus solitarius (Schwartz, Roulhac, Lam and O'Leary, 1951, man). The secondary projections to higher centers from the nuclei of the tractus solitarius which together receive both special and general visceral afferent fibers are not well documented. It has been reported that some secondary fibers concerned with taste relay by way of the contralateral medial lemiscus, or adjacent fibers, in guinea pig (ALLEN, 1923) to the " . . . nucleus ventralis et lateralis thalami" and thence by thalamic sensory radiations to the cerebral cortex (von B E C H T E R E W , 1 9 0 8 — 1 9 1 1 ) . P E N F I E L D and RASMUSSEN ( 1 9 5 0 ) traced gustatory impulses to the fronto-parietal operculum at the base of the central fissure for the cortical recognition of taste. GORSCHKOW ( 1 9 0 1 ) , a student of VON BECHTEREW, carried gustatory impulses to the potential island region in the dog. Terminal degeneration within the contralateral nucleus ventralis posteromedialis pars parvocellularis resulting from an electrolytic stereotaxic lesion of the nucleus of the tractus

22

Kinney, F. C.

solitarius at the level of the glossopharyngeal nerve in monkey NR 1 (of this study) confirms the work of others that this is the thalamic receptive nucleus in regard to gustatory sensibilities. The question also arises as to the identification of specific levels within the central nervous system at which gustatory impulses come into consciousness. BRADLEY (1963) performed a number of experiments on monkey (Macaca mulatto) which involved the removal of cortical areas and the placing of lesions in thalamic and superior collicular levels. Preoperatively and postoperatively monkeys were given banana sprinkled with quinine. Following removal of the cortex at the base of the central fissure and its opercular extension one monkey continued to respond to the quinine but did so more slowly than observed preoperatively. With lesions involving the superior colliculus of the midbrain there was a great reduction in response but again the animal rejected the quinine. This would indicate, perhaps, that awareness of bitter gustatory sensation comes into consciousness at some level other than those ablated by Bradley. There are few reported clinical cases which support the available anatomical evidence for the central gustatory pathways in man, two of which are reported below. ADLER (1934) reported a case in which a patient had a glioblastoma of the third ventricle which initially involved the most rostral medial poition of the nucleus ventralis posteromedialis of the dorsal thalamus. Contralateral to the side of the lesion the patient had a loss of all gustatory sensation except a slight recognition of bitter. SHENKIN and LEWEY (1943) reported a case in which the patient had epileptic seizures preceded by an aura in which he experienced a sour-bitter taste. At operation a vascular anomaly was discovered in which the vessels located in the lateral fissure were greatly dilated, the point of greatest dilatation being over the most inferior portion of the postcentral gyrus. The patient had a loss of sweet perception over the entire side of the tongue contralateral to the lesion. While the above two clinical cases concur with the available information on the central connections of secondary and tertiary gustatory pathways, it is apparent from the literature that further documentation for these pathways is necessary. A more complete review of the pertinent literature is included in the Discussion. Materials and Methods

Normal Material A large portion of the brain stem of one monkey was used to study the normal configuration of the fiber connections and nuclear groups extending from caudal medullary levels to rostral inferior collicular levels. The brain stem was embedded

in paraffin and cut transversely on a rotary microtome at twenty microns. Alternating sections were stained by the Weil-Weigert technique for myelinated fibers and with thionin for the nuclear groups.

Experimental Material One green monkey (Cercopithecus aethiops) and four rhesus monkeys (Macaca mulatto) were used in the experimental studies. Sour and bitter were the taste modalities tested. Each monkey was tested several times both preoperatively and postoperatively after having fasted for a minimum of twelve hours. Initially, when testing for the ability to recognize sour substances, a piece of banana which had been soaked for a minimum of twelve hours in reconstituted lemon juice ("ReaLemon") was offered to each animal. Subsequently a piece of lemon-treated banana or a slice of lemon was given to each animal when testing his responses to sour substances. A piece of banana with quinine sulfate powder sprinkled within its center was offered to each monkey to test his ability to recognize bitter substances. The response to the sour and bitter substances of each monkey were consistent with those reported on particular days in the Results section. Records were kept of the responses of the monkeys to these two substances and, whenever possible, responses were documented photographically with an eight or sixteen millimeter movie camera. All surgical procedures were performed under sterile technique. An electrolytic stereotaxic lesion, based on the f i g u r e s o f ATLAS a n d INGRAM ( 1 9 3 7 ) , w a s p l a c e d i n t h e n u c -

leus of the tractus solitarius at the level of the glossopharyngeal nerve in one monkey so that degeneration studies could confirm the rostral projection of those secondary gustatory fibers from the nucleus of the tractus solitarius to the nucleus ventralis posteromedialis pars parvocellularis of the dorsal thalamus. Either left or right craniotomies or sequentially both were performed in the remaining monkeys and cortical lesions using either surgical suction or cauterization were placed at the base of the central fissure in the fronto-parietal operculum. Following surgery, 300,000 units of sterile procaine penicillin G suspension were administered intramuscularly for a period of five days to minimize further the possibility of infection. Postoperatively the monkeys were not tested for their responses to the sour and bitter substances until the period of antibiotic therapy was completed. Immediately upon recovery from surgery, and thereafter until the time of sacrifice, any behavioral changes or neurological deficits present were recorded. Each monkey was not sacrificed for a minimum of fourteen days to assure a sufficient amount of degeneration of the pathways affedted. Prior to perfusion-fixation each animal was anesthetized with ketamine hydrochloride and his anterior chest wall incised to expose the pericardial cavity. The left ventricle was cannulated and 0.9% saline followed by 10% formalin buffered with calcium carbonate or cacodylate was gravity fed through the cardiovascular system. Immediately following perfusion-fixation each brain was removed and placed in 10% formalin buffered with calcium carbonate or cacodylate. The gross locations of cortical lesions were determined on postmortem. The location and extent of the electrolytic stereotaxic lesion in the monkey (monkey NR 1) in which such a lesion was placed and the connections from this lesion and from the lesion of the fronto-parietal operculum of monkey N R 2 were determined histologically by

Central Gustatory Pathways in the Monkey t h e FINK and HEIMER silver i m p r e g n a t i o n technique f o r degenerating axons (1967), the D E OLMOS and INGRAM silver

impregnation technique (1971) and the use of thionin stain.

Results The protocols and histological examination of the lesions and resulting degeneration where available are presented. Monkey NR 1 Preoperative Testing: 20th, 21st and 22nd of May 1975. The monkey was slower to reject the quinine banana than was observed in other monkeys prior to their operative procedures. She did not hesitate to eat a slice of lemon. Operative Protocol: Female. Weight — 4.5 kg. 23 May 1975. The skin and underlying connective tissue were reflected from the superior medial aspect of the left side of the head. A burr hole was drilled through the bone to the dura at the stereotaxic coordinates: A 0.7 and L5. The electrode was inserted through the cortex of the left cerebral hemisphere at the stereotaxic coordinates: A 0.7, L 5 and H —12. A stimulus of 0.1 volt produced ispilateral movement of the lower face at the corner of the mouth confirming that the tip of the electrode was in the motor nucleus of the facial nerve. In attempting to locate the nucleus of the tractus solitarius, it was considered essential as an important anatomical landmark to locate stereotaxically the motor nucleus of the facial nerve. Once stereotaxic placement of the electrode in this nucleus was verified by appropriate response to electrical stimulation the electrode was moved laterally (0.5—1.0 millimeters) and superiorly (1 — 2 millimeters) to the approximate anatomical location of the nucleus of the tractus solitarius (dorsal visceral gray) that receives those primary special visceral afferent fibers for taste which enter the brain stem in the facial nerve. Using a unipolar electrode, an electrolytic stereotaxic lesion was placed at this point. This area is continuous longitudinally with those parts of the nucleus of the tractus solitarius which receive primary gustatory fibers entering the brain stem by way of the glossopharyngeal and vagus nerves. These coordinates were then transposed caudally and longitudinally within the pons and the medulla to coincide with the known anatomical relationships of the nucleus of the tractus solitarius at these levels. A lesion was then placed in the nucleus of the tractus solitarius at the level of entrance of the glossopharyngeal nerve to the medulla. Four electrolytic stereotaxic lesions with a current of 3.5 ma for a period of 30 seconds were placed in monkey N R 1 at the following coordinates: Lesion 1, A 0.7, L 4.5 and H —10; Lesion 2,

23

A 0.7, L 5.0, and H - 1 0 ; Lesion 3, P 2.6, L 3.5 and H - 1 3 and Lesion 4, P 2.6, L 4.5 and H - 1 2 and H -13. Postoperative Notes: Immediately upon recovery from anesthesia the monkey exhibited a rotary nystagmus which lasted for approximately twentyfour hours. She held her head at an angle toward the side of the lesion until the time of sacrifice. The monkey continually held on to the side of the cage from the time of recovery from the anesthesia until her demise. Postoperative Testing: The monkey responded exactly as she had done preoperatively. She responded slowly to the quinine-treated banana, finally rejecting it. The monkey did not hesitate to eat the slice of lemon. Sacrificed: 6 June 1975. Description of stereotaxic lesions and resulting degeneration: The brain stem of this monkey, from caudal medullary levels through rostral thalamic levels, was frozen with solid carbon dioxide and sliced on a sliding microtome at 25 microns. Every fifth section was stained w i t h the DE OLMOS and INGRAM (1971) silver

impregnation method for degenerating axons. Adjacent sections through the levels of the lesions and through the thalamus were stained with thionin. Light microscopy revealed that lesions 1 and 2 (Figure 1) were placed at the level of the entering rootlets of the facial nerve in the caudal on-third of the pons. The lesions involved the superior vestibular nucleus at this level and extended from the floor of the fourth ventricle superiorly to the middle of the lateral reticular formation inferiorly. The lesion did not directly involve the nucleus of the tractus solitarius at this level; however, the lesion was so extensive that many secondary gustatory fibers which may have crossed to the opposite side were most probably interrupted in their contralateral passage and indeed the secondary gustatory tract which had been traced from medullary levels was observed to become larger and more discrete at this level (Figure 2). Lesions 3 and 4 were placed at rostral glossopharyngeal levels in the medulle. Lesion 3 (Figure 1) involved the deep white matter of the cerebellum and degenerating axons were traced into the left superior cerebellar peduncle. Lesion 4 (Figure 3) involved the medial, lateral and inferior vestibular nuclei as well as the nucleus of the tractus solitarius (dorsal visceral gray) at the level of the most rostral portion of the glossopharyngeal nerve. At the level of the lesion in the medulla a small, degenerating finely medullated bundle of fibers is present just dorsal to the medial lemniscus. At caudal pontine levels this same bundle becomes larger and more discrete (Figure 1). It is difficult to follow the

24

Kinney, F. C.

secondary gustatory tract in its rostral projection to the contralateral side of the brain since it crosses at an oblique angle to reach its destination just dorsal to the medial lemniscus, for only a few fine fibers can be traced directly across the midline. However, as the medial lemniscus assumes its characteristic horizontal plane in the rostral pons, the secondary gustatory fibers are seen to maintain their medial and dorsal position in relationship to the medial border of the medial lemniscus. At rostral inferior collicular levels degenerating axons of the secondary gustatory tract are still dorsal to the medial border of the medial lemniscus in close relationship to the ventral portion of the central tegmental bundle within which there is a small degenerating descending component which represents a path resulting from the lesion of the left brachium conjunctivum. At the junction between the inferior and superior colliculi (Figure 4) secondary gustatory fibers interdigitate with the most lateral portion of the decussating degenerating ascending axons of the brachium conjunctivum (primarily dentatorubrothalamic tract). These secondary gustatory fibers cannot be clearly distinguished (Figure 5) from this last mentioned path until their respective terminations within the dorsal thalamus. Many fibers of the ascending component of the brachium conjunctivum terminate within the rostrally located small-celled portion of the red nucleus, whereas other components within this bundle project directly to the nucleus ventralis lateralis of the dorsal thalamus. Within the medially and rostrally located parvocellular portion of the nucleus ventralis posteromedialis (arcuate nucleus of many authors) of the dorsal thalamus terminal degeneration of the secondary ascending gustatory fibers was observed. The level of termination within the dorsal thalamus of ascending degenerating axons of the secondary gustatory tract is shown in Figure 6. Terminal degeneration within the most rostral portion of the nucleus ventralis posteromedialis pars parvocellularis is seen in Figure 7; and terminal degeneration of the ascending component of the brachium conjunctivum in the nucleus ventralis lateralis can be observed in Figure 8. Monkey NR 2 Operation NR One

Preoperative Testing: 27 April 1975 and 28 April 1975. Monkey N R 2 ate the lemon-soaked banana without hesitation. The animal bit into but did not continue to eat a piece of lemon. The monkey immediately rejected the quinine-treated banana upon tasting the quinine.

Operative Protocol: Male. Weight — 3.0 kg. 29 April 1975. The skin and temporalis muscle were reflected on the left side of the head. A left craniotomy was performed and the bone and dura overlying the base of the central fissure and the superior and middle temporal gyri were removed. A large lesion which involved the base of the precentral gyrus, the base of the postcentral gyrus, the adjacent superior temporal gyrus and a small portion of the middle temporal gyrus (Figure 9) was placed with surgical suction. Massive bleeding of the middle cerebral artery resulted. The bleeding was controlled with absorbable gelatin sponge (Gelfoam, Upjohn) and as soon as the surgical area was dry the wound was closed. Postoperative Notes and Testing: The monkey showed a marked paralysis of the entire right side of the body. However, by 29 July 1975, this monkey had apparently fully recovered from his paralysis. He was tested for his responses to quinine and bitter substances on this date. The monkey rejected the quinine-treated banana more slowly than prior to surgery and he no longer found the slice of lemon objectionable, for he ate the entire slice without hesitation. Operation NR Two

Operative Protocol: Male. Weight — 2.9 kg. 19 August 1975. A right craniotomy was performed and the bone overlying the base of the central fissure and the adjacent superior temporal gyrus was removed. The dura was incised with a cutting needle and reflected. Using a micropipette for surgical suction, the cortex at the base of the central fissure, its opercular surface and the adjacent superior temporal gyrus were removed. Little bleeding occurred and the wound was closed. Postoperative Notes and Testing: The monkey fully recovered from this operation with no apparent neurological deficits. Following this second operation little difference could be detected in the taste responses of this monkey. The response to the quinine-treated banana was the same as had been observed following the first operation in that the monkey still rejected the quinine more slowly than had been observed prior to any surgical procedure and he still ate the slice of lemon. The location and extent of the lesions in both hemispheres are seen in Figures 9 and 10. Sacrificed: 1 September 1975. Description of resulting degeneration from the lesion in the right cerebral hemisphere: The right cerebral hemisphere of monkey N R 2 was sliced in a coronal plane on a sliding microtome at 30 microns after having been frozen with solid carbon dioxide. Every fifth section was stained by the DE OLMOS and INGRAM (1971) silver method for impregnating degenerat-

Central Gustatory Pathways in the Monkey

ing axons. Adjacent sections were stained with thionin. The level of the lesion at the base of the central fissure can be observed in Figure 11. The base and the inferior portion of the opercular surface were ablated as well as the adjacent portion of the superior temporal gyrus. From the lesion in the frontal and parietal opercula fine degenerating axons can be traced which course through the extreme capsule to end in the dorsal anterior island (Figure 12) thus linking the bases and opercular surfaces of the pre- and postcentral gyri to the island by short association bundles. Degenerating axons from neurons in the frontal and parietal lesions can also be traced to the nucleus ventralis posteromedialis of the dorsal thalamus (Figures 13 and 14). These axons course dorsal to the claustrum, and, accompanying the thalamocortical sensory radiations (Figure 15), pass through the posterior limb of the internal capsule to gain the thalamus. Terminal degeneration is abundant in the nucleus ventralis posteromedialis with only a few fine fibers terminating in the nucleus ventralis posteromedialis pars parvocellularis. Cytolysis of the cells of the nucleus ventralis posteromedialis as well as massive gliosis is clearly evident thus indicating retrograde degeneration from the cortical lesions. Even though degenerating axons could be traced to the parvocellular portion of the nucleus ventralis posteromedialis little if any retrograde degeneration is observable in this portion of the nucleus. Monkey NR 3 Preoperative Testing: 3 January 1975. Monkey NR 3 was tested for taste. He hesitantly rejected the lemon-soaked banana after he tasted it. He dropped the piece of lemon as soon as he tasted it. He expectorated the quinine-treated banana as soon as he had tasted the quinine. Operative Protocol: Male. Weight — 7.0 kg. 4 January 1975. An incision of approximately six centimeters slanting posterior superiorly to anterior interiorly was made above the left ear. It was necessary to excise a large portion of the temporalis muscle in order to have an adequate visual field. As much bone as possible was removed overlying the left temporal lobe and the inferior aspects of the rostral parietal lobe and the caudal frontal lobe. The dura was incised with a cutting needle and reflected with a scapel to expose the lateral fissure and the superior, middle and part of the inferior temporal gyri of the left temporal lobe. In an attempt to remove the island and the gustatory association areas a large portion of the temporal lobe was removed with suction. Little bleeding oc-

25

curred. The wound was packed with Gelfoam and was closed as soon as the surgical area was dry. Postoperative Notes and Testing: On 5 January, 1975 the monkey was eating, taking water and sitting erect. By 31 January 1975, the monkey's incision was completely healed and he appeared to have no neurological deficits. On 2 February 1975 monkey NR 3 was tested for his responses to sour and bitter substances. The animal did not, as had been previously observed, find the slice of lemon objectionable for he ate the entire slice. He rejected the quininetreated banana but did so slightly more slowly and with less vehemence than had been observed prior to any surgical procedure. The location and extent of the lesion was confirmed on postmortem and can be observed in Figure 16. Sacrificed: 12 February 1975. Monkey NR 4 Cercopithecus aethiops Operation NR One

Preoperative Testing: 12 August 1974. Monkey NR 4 was offered a piece of quinine-treated banana which he expectorated as soon as he tasted the quinine. Following his rejection of the quinine banana the animal did not hesitate to eat a piece of untreated banana. He was then offered a piece of lemon-treated banana which he smelled, dropped, then tasted and finally rejected. Operative Protocol: Male. Weight — 5.6 kg. 13 August 1974. The skin and temporalis muscle were reflected on the left. A left craniotomy was performed in which the bone and dura overlying the base of the central fissure were removed. The cortex at the base of the precentral gyrus and at the base of the postcentral gyrus and the opercular surfaces of these two cortical areas were removed by cauterization in the left cerebral hemisphere. Postoperative Notes: Monkey NR 4 fully recovered from this operation with no permanent neurological deficit. For the first two weeks following surgery a slight paresis of the right upper extremity was evident. Operation NR Two

Preoperative Testing: 26 November 1974. Monkey NR 4 was given a piece of lemon-soaked banana which he sniffed, tasted, then dropped and finally ate. The animal was then offered a piece of quinine-treated banana which he expectorated immediately upon tasting the quinine. Operative Protocol: 27 November 1974. The skin and temporalis muscle were reflected on the right. A right craniotomy was performed in which the bone and dura

2 6

Kinney, F. C.

overlying the base of the central fissure were removed. By cauterization a large lesion was placed in the cortex at the base of the precentral gyrus and at the base of the postcentral gyrus and also the opercular surfaces of these two gyri. There occurred a large amount of hemorrhage due to involvement of branches of the middle cerebral artery. The bleeding was controlled with Gelfoam, cotton and suction. The wound was packed with Gelfoam and as soon as the surgical area was dry the wound was closed. Postoperative Notes and Testing: 10 December 1974. Monkey NR 4 was tested for taste. He did not hesitate, as had been previously observed, to eat the lemon-treated banana. The monkey rejected the quinine-treated banana, but did so more slowly than had been previously observed. His behavior indicated that there had been a loss of some taste discrimination following the removal of the cortices at the base of the central fissure and of the opercular surface of this area bilaterally. The locations of the cortical lesions were confirmed on postmortem and are shown in Figure 17 and Figure 18. Sacrificed:

11 December 1974.

Monkey NR 5 Preoperative Testing: 27 December 1974. Rhesus monkey NR 5 was offered a piece of quinine-treated banana which he expectorated immediately upon tasting the quinine. He bit into but did not continue to eat a slice of lemon. Unlike Monkey NR 4 (Cercopithecus aethiops) and the older rhesus monkey NR 3, both of whom had initially rejected, rather hesitantly, the banana which had been soaked for twelve hours in reconstituted lemon juice, the young rhesus did not hesitate to eat the lemon-treated banana. Operative Protocol: Male. Weight — 2.9 kg. 30 December 1974. The skin and temporalis muscle on the left were reflected. A left craniotomy was performed in which the bone and dura overlying the base of the left central fissure were removed. Cauterization was utilized to remove the cortex at the base of the central fissure and the opercular surface of this area. Little bleeding occurred. As soon as the surgical area was dry the wound was closed. Postoperative Notes: Monkey NR 5 recovered from this operation with no apparent neurological deficits. On 2 February 1975 monkey NR 5 was tested for his responses to sour and bitter substances. He expectorated the quinine-treated banana as soon as he tasted the quinine. He bit into but did not continue to eat a piece of lemon. He did not hesitate to eat a piece of lemon-treated banana. On 5 February 1975 monkey NR 5 was stricken with shigellosis and on 14 February 1975 died from

complications arising from this disease. His brain was removed and placed in 10% formalin buffered with calcium carbonate. Figure 19 shows the location and extent of the cortical lesion in the left hemisphere. Discussion and review of the literature Many previous investigators (ALLEN, 1 9 2 3 ; A D L E R , 1934;

WALKER,

1938;

BLUM, WALKER a n d

RUCH,

1 9 4 3 ; PATTON, RUCH a n d W A L K E R , 1 9 4 4 a n d A B L E S

and BENJAMIN, 1 9 6 0 ) have concluded that the nucleus ventralis posteromedialis, and/or its most medial portion (pars parvocellularis), of the dorsal thalamus is responsible for the thalamic reception of gustatory stimuli. Thalamocortical connections between the medial portion of the nucleus ventralis posteromedialis (i. e. parvocellular portion of this nucleus as described by OLSZEWSKI, 1 9 5 2 ) and the fronto-parietal operculum have frequently been demonstrated in the macaque ( W A L K E R , 1 9 3 4 ; L E GROS CLARK, 1 9 3 7 a n d LOCKE, 1 9 6 7 ) . CHOW and PRIBRAM ( 1 9 5 6 ) observed, based on retrograde degeneration studies following the placing of lesions at various cortical locations in the macaque, that the nucleus ventralis posteromedialis projected to the operculum and to the anterior island. Based on experimental evidence in the macaque, R O B E R T S and A K E R T ( 1 9 6 3 ) reported that it was the anterior island which received primary afferent connections from the nucleus ventralis posteromedialis pars parvocellularis -and not the fronto-parietal operculum. BENJAMIN, EMMERS and BLOMQUIST ( 1 9 6 8 ) and BENJAMIN and BURTON ( 1 9 6 8 ) reached the conclusion that there are only sustaining projections of gustatory modalities to somatic sensory area I in the squirrel monkey and that both the anterior island and the cortical area in somatic sensory area I which receives gustatory modalities must be removed surgically in order to result in retrograde degeneration in the nucleus ventralis posteromedialis pars parvocellularis of the dorsal thalamus. Using evoked potential techniques (BLOMQUIST, BENJAMIN and EMMERS, 1 9 6 2 ) the dorsal thalamus in the squirrel monkey was explored bilaterally with recording electrodes for responses to electrical stimulation of the chorda tympani nerve, the lingualtonsilar branch of the glossopharyngeal nerve and the lingual nerve. The responses of the two nerves relaying gustatory modalities were reported to be largely ipsilateral with meager contralateral projections, whereas the lingual nerve was reported to have a mainly contralateral projection. Considering the available literature, there can be little question that the nucleus ventralis posterome-

Central Gustatory Pathways in the Monkey

dialis pars parvocellularis receives secondary gustatory projections from that part of the nucleus of the tractus solitarius (dorsal visceral gray) which is found at the level of the facial and glossopharyngeal nerves in the pons and medulla. In the present study such projections were confirmed by the rostral projection of secondary gustatory fibers to the contralateral nucleus ventralis posteromedialis pars parvocellularis in monkey NR 1. The neurophysiological evidence for the ispilaterality, contralaterality or bilaterality of the secondary projections of the dorsal visceral gray is conflicting. The evidence from monkey NR 1 is in accord with the observations that the lingual-tonsilar nerve has a primarily contralateral cortical (and therefore thalamic) projection (BENJAMIN a n d PFAFFMANN, 1 9 5 5 ; YAMAMOTO a n d KAWAMURA, 1 9 7 5 ) . Certainly the clinical evidence (which will be discussed later in this report) overwhelmingly favors thalamic and cortical projections of gustatory modalities which are primarily contralateral in termination. During the latter part of the nineteenth century and throughout the first part of this century, it was believed by most observers that the cortical localization for gustatory impulses was in the temporal lobe in close association with afferent olfactory impulses. F E R R I E R (1886) concluded in experimental procedures on two monkeys that bilateral lesions of the lower temporo-sphenoidal lobe resulted in abolition of smell and taste over the entire tongue. F E R R I E R thus felt ... the gustatory centres are situated at the lower extremity of the temporosphenoidal lobes in close relation with those of smell. Ferrier failed to mention, however, that the adjacent parietal lobe had also been damaged in his experiments, as is shown in Figure 101 of his book. The significance of this fact was also noted by BORNSTEIN (1940a). K E N N E D Y (1911) considered gustatory auras and olfactory hallucinations to be important tools in diagnosing uncinate fits and temporal lobe lesions, linking gustatory impulses with olfactory impulses in the temporal lobe. CUSHING (1922) discussed ten of fifty-nine clinical cases in which there were temporal lobe lesions. Of these only two patients had gustatory auras; one patient thought he had smelled and tasted peaches on one occasion and insisted that he smelled and tasted roasted peanuts on another, and another patient complained of gustatory sensations which were peppermint in nature. V I L L I G E R (1925) stated " . . . the gustatory center has not been definitely located but probably adjoins that for smell," and G R I N K E R (1937) considered the hippocampal formation to have both olfactory and gustatory functions. BORNSTEIN ( 1 9 4 0 a) in an extensive paper, reviewed

27

the clinical and experimental evidence for the temporal lobe theory for the cortical localization of taste and concluded that there was insufficient evidence to support this theory. Three clinical cases, each one of which was the result of a bullet wound to the left parietal bone, were discussed by BORNSTEIN (1940b). The first case resulted in a lesion of the middle third of the pre- and postcentral gyri. The patient had an ageusia for sweet, sour, salty and bitter on the contralateral side of the tongue except for hypogeusia for salty substances on the contralateral tip. Taste was intact or only slightly impaired on the ipsilateral side of the tongue. The second case resulted in a lesion which was located in the face and tongue areas of the pre- and postcentral gyri, i. e., the upper border of the posterior frontal operculum and the parietal operculum. The patient exhibited an ageusia for sweet, sour and weak bitter solutions on the contralateral border of the tongue with slight hypogeusia for salty. On the contralateral tip there was a hypogeusia for sweet, salty and sour and on the contralateral base only sour and sweet were impaired. The third patient developed a parietal opercular syndrome not from the bullet wound itself but rather from the transplantation of a bone graft over the pre- and postcentral gyri which resulted in a subdural hematoma over this area. There was a hypogeusia for sweet, salty and bitter and a slight hypogeusia for sour on the contralateral side of the tongue. The border of the tongue on the contralateral side was chiefly involved. Gustatory disturbances were slight on the ipsilateral side of the tongue. SHENKIN and L E W E Y ' S clinical case (1943) in which there was an involvement of the most inferior portion of the postcentral gyrus also indicated this area as a possible site for the cortical localization of gustatory sensations. GORSCHKOW ( 1 9 0 1 ) postulated that gustatory impulses were localized in the potential island region in the dog. A D L E R ( 1 9 3 5 ) gave support to the possibility that the island may be the cortical area responsible for the cortical recognition of gustatory impulses. She discussed a case in which a tumor grew into the island from the superior temporal lobe. The patient had a hypogeusia on the tongue contralateral to the cortical lesion. P E N F I E L D and ERICKSON ( 1 9 4 1 ) cited a clinical case in which the patient had seizures preceded by a gustatory aura of an indescribably nature. "This patient had a well circumscribed slowly growing astrocytoma of the island of Reil." P E N F I E L D and ERICKSON concluded that taste was probably represented cortically at the base of the pre- and postcentral gyri in close association with the cortical areas for the jaws and the tongue. P E N F I E L D and J A S P E R ( 1 9 5 4 ) , based on their observations of clinical cases, stated " . . . gustatory sensation has a representation

28

Kinney, F. C.

in the cortex which seems to be closely related to salivation, to the alimentary system, and to second sensory representation. It is located beneath the fissure of Sylvius and within the circular sulcus." Gustatory impulses have been localized in the opercular insular junction and in the floor of the anterior island by P E N F I E L D and RASMUSSEN (1950). PATTON and R U C H (1946) concluded in a series of cortical ablation experiments on monkeys and chimpanzees that taste deficits only occurred when the buried cortex of the fronto-parietal operculum was invaded. After recognizing that Gerhardt had described an area of granular cortex of sensory type in the chimpanze which was located deep in the frontal operculum bordering on the circular fissure, R U C H and PATTON (1946) attempted destruction of this area. Successful ablation was achieved in only one monkey resulting in a severe taste impairment in regard to bitter solutions. These investigators expanded their previous conclusions by stating that taste is localized in the parainsular operculum and not in the free cortex of the operculum. BAGSHAW and PRIBRAM (1953) observed a lowering in quinine acceptance thresholds in monkeys in which the anterior insula or both the anterior island and the anterior supratemporal plane had been surgically ablated. Lesions of the anterior island plus operculum resulted in moderate decreases in quinine thresholds. Lesions of the insula, the operculum and the anterior supratemporal plane together resulted in marked and prolonged ageusia. B R A D L E Y ( 1 9 6 3 ) , in his experiments on macaques, considered the cortex at the base of the central fissure and the opercular surface of this area to be important for normal taste discrimination. He, furthermore, suggests from his experiments that taste may have a representation at mid-brain levels. Bilateral cortical representation in the squirrel monkey for all three nerves innervating the tongue, that is, the chorda tympani branch of the facial nerve, the lingual-tonsilar branch of the glossopharyngeal nerve, and the lingual branch of the trigeminal nerve, was demonstrated by B E N J A M I N et al. (1968) by the utilization of evoked potential techniques. The cortical area found to be responsive to electrical stimulation was in somatic sensory area I at the base of the pre- and postcentral gyri. The major projection, however, of the chorda tympani and lingual-tonsilar nerves was found to be ipsilateral. In a related study which also utilized evoked potential techniques, a responsive locus for taste was located in the most anterior opercular-insular cortex (BENJAMIN and BURTON, 1968). Only stimulation of the ipsilateral chorda tympani nerve and the ipsilateral lingual-tonsilar nerve was effective.

Chemical stimulation of the tongue and/or electrical stimulation of the ipsilateral chorda tympani nerve produced evoked potentials in neurons of the nucleus ventralis posteromedialis pars parvocellularis in the rat and marmoset (GANCHROW and ERICKSON, 1 9 7 2 ) . Antidromic stimulation of these same cells by electrical stimulation of the cortical projection area of the chorda tympani nerve indicated direct thalamocortical projections of taste neurons and direct projections from the cortex to the cells of the thalamus which receive gustatory impulses. Based on cytoarchitectural studies in the squirrel monkey, SANIDES ( 1 9 6 8 ) concluded that there was a surface cortical gustatory area in somatic sensory area I and also a deep pure gustatory area located in the frontal operculum and bordering on the insula. In cat, PATTON and AMASSIAN ( 1 9 5 2 ) reported a bilateral cortically responsive area to electrical stimulation of the chorda tympani nerve which was superior to the rhinal fissure and rostral to the anterior ectosylvian fissure. The insula as well as other cortical areas were unresponsive to the stimulation. RUDERMAN et al. ( 1 9 7 2 ) obtained responses of neurons in the nucleus ventralis posteromedialis pars parvocellularis of the cat while stimulating the anterior portion of the tongue with citric acid solutions. Light brush strokes of various facial regions resulted in evoked potentials which were recorded from electrodes in the nucleus ventralis posteromedialis. Electrolytic lesions of the nucleus ventralis posteromedialis pars parvocellularis resulted in cortical degeneration in either the lateral or both blanks of the presylvian sulcus. Electrolytic lesions that were placed in the nucleus ventralis posteromedialis resulted in cortical degeneration in the coronal gyrus. Thus separate cortical projection areas for taste and facial tactile impulses were delineated. The cortical area responsive to mechanical stimulation of the ipsilateral chorda tympani nerve in the cat was located by BURTON and E A R L S ( 1 9 6 9 ) in an area which extended from the coronal gyrus anteriorly to the orbital sulcus posteriorly and lay dorsal to the rhinal fissure. This active surface cortex was located within somatic sensory area I for the tongue. ZOTTERMAN (1958, cat) reported evoked potentials from cortical cells which responded only to taste stimuli. The responses were not to specific taste modalities. COHEN et al. ( 1 9 5 7 ) explored the cortical projection area of the tongue (orbital surface of the cerebral hemisphere) in cat, with recording electrodes. This area was responsive to electrical stimulation of the chorda tympani and lingual nerves and also to thermal, mechanical and gustatory stimulation of the tongue. Some cortical cells were responsive to

Central Gustatory Pathways i n the Monkey

electrical stimulation of both chorda tympani and lingual nerves. Cortical cells which responded to mechanical stimulation did not respond to gustatory or thermal stimulation. Evoked potentials were recorded from five cortical cells which responded only to gustatory modalities, but again the cells were not specific in their response to particular gustatory sensations. LANDGREN (1957) reported 2 7 single cortical cells (out of 101 cortical cells from which evoked potentials were recorded) which responded to more than one type of sensory stimulation. Cells were found which responded to combinations of touch, stretch, cooling, warming and various gustatory stimuli. Various sensory modalities including taste, therefore, were seen to converge on single cortical cells. Evoked potential studies were utilized by BENJAMIN and PFAFFMANN (1955) to locate the cortical receptive zones for the glossopharyngeal and chorda tympani nerves in the rat. The chorda tympani nerve was found to be represented bilaterally in the cerebral cortex whereas the glossopharyngeal nerve only had a contralateral cortical representation. Bilateral ablations of this cortical area resulted in permanent taste deficits. The authors stated that the boundaries of the lesions included the anterior insular cortex as delimited by Krieg in the rat. Bilateral removal of a small area of cortex in the rat just dorsal to the rhinal fissure in somatic sensory area I produced partial taste deficits (BENJAMIN and AKERT, 1959). Inclusion of an adjacent fringe area in bilateral ablations caused severe gustatory impairments. This cortical area corresponded to the cortex which was responsive to stimulation of the glossopharyngeal and chorda tympani nerves as determined by evoked potential techniques. Ablation of all neocortex with the exception of this focal gustatory area had no effect upon the taste discrimination of the animals. Electrical stimulation of the chorda tympani, glossopharyngeal and lingual nerves in rabbits (YAMAMOTO and KAWAMURA, 1975) demonstrated a primarily ipislateral responsive area in the insular cortex for the chorda tympani nerve, a primarily contralateral response in the same cortical area for the glossopharyngeal nerve and predominantly contralateral projection in somatic sensory area I for the lingual nerve. All projections, however, were bilateral. The cortical projection area for the chorda tympani nerve was found rostral to the projection for the glossopharyngeal nerve and partially overlapped it and the lingual nerve projection area partially overlapped the dorsal aspect of the chorda tympani projection area. The possible existence of a subcortical and even subthalamic localization of taste responses in carni-

29

vores was demonstrated by MACHT ( 1 9 5 1 ) . Mesencephalic and bulbospinal cat preparations rejected quinine, saline and acetic treated substances at the same concentrations as normal rejection thresholds. In the present study lesions of the cortex at the base of the central fissure and its opercular surface resulted in taste deficits. A unilateral destructive lesion of this area resulted in a definite slowing of the taste responses of one monkey and a loss of some discrimination in regard to sour substances in two out of three monkeys. Destructive lesions placed bilaterally in the cerebral cortex at the base fo the central fissure and extending into the opercular surfaces of the precentral and postcentral gyri resulted in a slowing of the response to quinine in one monkey. More importantly the loss of a more delicate type of discrimination, in regard to sour substances, was observed with monkey NR 4 (Cercopithecus aethiops) following bilateral ablation of this area. It is evident from work of others and from the present study that the cortex at the base of the central fissure extending into the fronto-parietal operculum is of particular importance in the reception of gustatory modalities and, being directly continuous with primary sensory cortex, represents primary receptive sensory cortex for taste. The functional significance of this area as it relates to the anterior island has, however, remained somewhat obscure. Experimental observations have repeatedly shown that the anterior island in primates and man is involved in the recognition of gustatory modalities. The coronal sections of monkey NR 2 indicate that the cortex at the base of the central fissure and continuing into the fronto-parietal operculum is linked to the dorsal anterior island by short association bundles which course through the extreme capsule. These short association bundles, then, relate gustatory modalities from their primary receptive cortex to the association cortical areas involved in the recognition of gustatory impulses, that is the anterior dorsal part of the Island of Reil. ELLIOT SMITH ( 1 9 0 7 ) traced, in man, fibers from the lateral olfactory tract into the anterior island. CROSBY, HUMPHREY and LAUER ( 1 9 6 2 ) stated that "The olfactory stalk has been traced into the frontal end of the island on the pyriform side. This relation is seen in the His embryological models." Thus, the anterior island, which receives both olfactory and gustatory impulses, is a correlative cortical area for special visceral afferent sensations and allows for subjective discrimination in regard to these sensations. Special visceral sensations are in turn correlated with other known general visceral functions of the island. P E N F I E L D and RASMUSSEN ( 1 9 5 0 ) elicited a number

30

Kinney, F. C.

of visceral sensations upon electrical stimulation of the insula in awake patients. Upon stimulation of various locations within the insula, patients reported a curious disagreeable taste, the sensation of nausea, vomiting, a feeling of illness, an aura of attack with swallowing and mouth opening and the desire to defecate. KAHN et al. (1969) discussed several clinical cases involving the insula. One patient complained of a bitter taste in his mouth which preceded convulsions. The patient had a tumor of the anterior island. Another patient had a four-year history of convulsions during which he complained of nausea and often vomited. At operation an arteriovenous malformation was discovered which had a focal point in the parietal operculum and the posterior island. A third patient had auras of abdominal pain which frequently preceded convulsions. A right temporo-parietal cyst was found at operation. The location of the lesion

suggested that stimulation in the posterior island region resulted in abdominal pain. KAHN et al. (1969) suggested that there is some evidence that primary gustatory cortex is connected to the uncus region by fascicles which pass through the island. Such connections could account for reported cases (KENNEDY, 1911 and CUSHING, 1922) in which gustatory auras preceded convulsions which were the result of lesions located primarily in the temporal lobe. Uncinate fits which are preceded by auras of an olfactory nature are the result of lesions located primarily in the region of the amygdala; or along the lateral olfactory tract and perhaps the anterior perforated space. However, uncinate fits which are preceded by gustatory auras might result from lesions located largely in the anterior island but which involve those fascicles which connect the island with the uncus region.

Iff, rlitlPS seam s s n m r

GNVII y

J

»r'5 ' • wSSL

MIF

ft

m '^mme^

• 3 HHB .S i M i f c S M " BRSia-i-irSSB sin.:-'»

SGTr I

t . Ijfc 1

Ml

©

Inf 0 Pyi

Fig. 1. Photomicrograph of a transverse section through the caudal one third of the pons of Monkey NR 1. Stereotaxic electrolytic lesions 1 and 2 at the level of the entering rootlets of the facial nerve on the left side are clearly evident. The arrow in the upper left quadrant of the photomicrograph indicates the position of stereotaxic electrolytic lesion number 3 which involved the deep white matter of the cerebellum. The area ecnlosed in black (shown at a higher magnification in Figure 2) indicates the position of the secondary ascending gustatory tract just dorsal to the medial lemniscus. De Olmos and Ingram preparation, x 6.3

List of Abbreviations Used in Figs. 1, 2 and 3 GN VII InfO ML MLF Pyr

Genu of Seventh (Facial) Cranial Nerve Inferior Olive Medial Lemniscus Medial Longitudinal Fasciculus Pyramid

SGTr NTS NIX IV V

Secondary Gustatory Tract Nucleus of the tractus solitarius Ninth (Glossopharyngeal) Nerve Fourth Ventricle

Central Gustatory Pathways in the Monkey

31

Fig. 2. Photomicrograph of the area outlined in black in Figure 11 (Monkey NR1). The arrows point to some of the degenerating axons of the secondary ascending gustatory tract. The secondary ascending gustatory tract passes in a rostral longitudinal direction through the brain stem and the degenerating axons are therefore cut transversely. De Olmos and Ingram preparation, x 32

Fig. 3. Photomicrograph of a transverse section through the rostral medulla of Monkey NR 1. The size, relationships and extent of the fourth stereotaxic electrolytic lesion placed in Monkey NR 1 are clearly evident. De Olmos and Ingram preparation, x 6.3

32

Kinney, F. C.

Fig. 4. Photomicrograph of a transverse section through the junction between the inferior and superior colliculi of Monkey NR. 1 at the level of the decussation of the brachium conjunctivum. The area outlined in black indicates the position of the secondary gustatory tract at this level. De Olmos and Ingram preparation, x 6.3 Fig. 5. Photomicrograph at a higher magnification of the area enclosed in black in Figure 4 (Monkey NR 1). The secondary gustatory tract interdigitates with degenerating fibers of the decussating brachium conjunctivum and cannot be distinguished from them until their respective levels of termination within the dorsal thalamus. The arrows indicate some of the fibers of the secondary gustatory tract as they swing superiorly to interdigitate with the fibers of the brachium conjunctivum. De Olmos and Ingram preparation, x 100 List of Abbreviations Used in Figs. 4 and 5 CA DBC

Cerebral Aqueduct Decussation of the Brachium Conjunctivum

MLF SGTr

Medial Longitudinal Fasciculus Secondary Gustatory Tract

Central Gustatory Pathways in the Monkey

Hirnforschung, Bd. 19, Heft 1

3

33

34

Kinney, F. C.

Fig. 6. Photomicrograph of a transverse section through the dorsal thalamus at the most rostral level of the nucleus ventralis posteromedialis pars parvocellularis of Monkey NR 1. Terminal degeneration within the nucleus ventralis posteromedialis pars parvocellularis (outlined in black) of the secondary gustatory tract is shown at a higher magnification in Figure 7. Terminal degeneration of the brachium conjunctivum (dentatorubrothalamic tract) in the nucleus ventralis lateralis is outlined in black and shown at a higher magnification in Figure 22. De Olmos and Ingram preparation, x 6.3 Fig. 7. Photomicrograph of the nucleus ventralis posteromedialis pars parvocellularis (outlined in Figure 6, Monkey N R 1) demonstrating terminal degeneration of the secondary gustatory tract. De Olmos and Ingram preparation, x 32 Fig. 8. Photomicrograph of the nucleus ventralis lateralis (outlined in Figure 6, Monkey NR 1) of the dorsal thalamus illustrating terminal degeneration of the ascending component of the brachium conjunctivum. De Olmos and Ingram preparation, x 32 List of Abbreviations Used in Figs. 6, 7 and 8 CN VL

Centromedian Nucleus Nucleus Ventralis Lateralis

VPMpc III V

Nucleus Ventralis Posteromedialis pars parvocellularis Third Ventricle

Central Gustatory Pathways in the Monkey

3*

35

36

Kinney, F. C.

Fig. 9. Left cerebral hemisphere of Monkey N R 2. The large cortical lesion which involved the bases of the pre- and postcentral gyri, at the base of the central fissure (CF), and the adjacent superior and middle temporal gyri is clearly apparent. Fig. 10. Right cerebral hemisphere of Monkey N R 2 illustrating the cortical lesion which involved the fronto-parietal operculum at the base of the central fissure (CF), its opercular surface and the adjacent portion of the superior temporal gyrus.

Central Gustatory Pathways in the Monkey 37

E x t i C a p — » • Hp--. OT l '- tii'f?-

aafi"-» "1 ¿Si kê^y'" *

Ê J Ê È * \"Jf

;

• •

f P *

Cl

F P O 1 V i ~"

LF

I s '»

— E x t - C a p

- ¿/*.ätfi ^T^*: 'y^ËVU--¡r vf^j. "¿rfF1 » ¿m:. Vt»

j t

• ' -1 ©

S

T

\

f

\

; '"'••it

W^'^kÊ^^l-M'' •

r

Fig. 11. Photomicrograph of a coronal section through the level of the right cortical lesion of Monky NR 2. As can be observed, the deep white matter of this cortical area as well as the fronto-parietal operculum was involved in the lesion. Fine degenerating axons can be traced through the extreme capsule into the dorsal anterior part of the Island of Reil. De Olmos and Ingram preparation, x 6.3 Fig. 12. Photomicrograph at a higher magnification of the area outlined in black in Figure 11 (Monkey NR 2). Preterminal and terminal degeneration can be observed throughout the field. De Olmos and Ingram preparation, x 250 List of Abbreviations Used in Figs. 11 and 12 CI Claustrum FPO Ext Cap External Capsule Extr Cap Extreme Capsule LF

Fronto-Parietal operculum

Is

Island of Reil

Lateral Fissure

STG

Superior Temporal Gyrus

38

Kinney, F. C.

Fig. 13. Photomicrograph of a coronal section through the dorsal thalamus of Monkey NR 2. Terminal degeneration from the lesion in the fronto-parietal operculum can be observed within the nucleus ventralis posteromedialis and a few fine fibers can be traced to the parvocellular portion of this nucleus. De Olmos and Ingram preparation. X 6.3 Fig. 14. Photomicrograph at a higher magnification of the area outlined in black in Figure 27 (Monkey NR 2). Massive terminal degeneration can be seen in the nucleus ventralis posteromedialis. The arrows indicate the terminal degeneration of a few fibers within the parvocellular portion of the nucleus ventralis posteromedialis. De Olmos and Ingram preparation. X 21.6 Fig. 15. Photomicrograph of degenerating corticothalamic fibers which accompany the sensory radiations (thalamocortical fibers) as they pass through the posterior limb of the internal capsule to gain entrance to or exit from the dorsal thalamus (Monkey NR 2). De Olmos and Ingram preparation. X 32 List of Abbreviations Used in Figs. 13, 14 and 15 Cn Centromedian nucleus CT Corticothalamic fibers ICapPl Internal Capsule Posterior Limb

VPM VPMpc

Nucleus Ventralis Posteromedialis Nucleus Ventralis Posteromedialis pars parvocellularis

Central Gustatory Pathways in the Monkey

Wm \

ICapPL

>

39

40

Kinney, F. C.

Fig. 16. The left cerebral hemisphere of Monkey N R 3. The arrow indicates the lesion of the postcentral gyrus and its opercular surface. Fig. 17. The left cerebral hemisphere of Cercopithecus aethiops. The cortical lesion at the base of the pre- and postcentral gyri is clearly evident. Arrow indicates the central fissure (CF).

Central Gustatory Pathways in the Monkey

41

Fig. 18. The right cerebral hemisphere of Cercopithecus caethiops. The cortical lesion at the base of the pre- and postcentral gyri, which also include the adjacent superior temporal gyrus, is clearly evident. Fig. 19. The left cerebral hemisphere of Monkey N R 5. The cortical lesion a t the base of the central fissure and the adjacent superior temporal gyrus is clearly shown.

42

Kinney, F. C.

References and R . M . B E N J A M I N : Thalamic relay nucleus for taste in albino rat. J. Neurophysiol., 23, 376 — 382 (1960). A D L E R , A . : Zur Topik des Verlaufes der Geschmackssinnsfasern und anderer afferenter Bahnen in Thalamus. Ztschr. ges. Neurol. Psychiat., 149, 208-220 (1934). A D L E R , A . : Zur Topik der corticalen Geschmackssphäre. Ztschr. ges. Neurol. Psyciat., 152, 25 — 33 (1935). A L L E N , W . F . : Origin and distribution of the secondary visceral fibers in the guinea pig. J. Comp. Neurol., 35, 2 7 5 - 3 1 1 (1923). A T L A S , D . , and W . R . I N G R A M : Topography of the brain stem of the rhesus monkey with special reference to the diencephalon. J. Comp. Neurol., 66, 2 6 3 - 2 8 9 (1937). B A G S H A W , M . H . , and K . H . P R I B R A M : Cortical organization in gustation (Macaca mulatto). J. Neurophysiol., 16, 4 9 9 - 5 0 8 (1953). B E N J A M I N , R . M . , and K . A K E R T : Cortical and thalamic areas involved in taste discrimination in the albino rat. J. Comp. Neurol., I l l , 2 3 1 - 2 5 9 (1959). B E N J A M I N , R. M., and H. B U R T O N : Projection of taste nerve afferents to anterior opercular-insular cortex in squirrel monkey (Saimiri sciureus). Brain Res., 7, 221 — 231 (1968). ABLES, M . F . ,

BENJAMIN, R . M.,

R . EMMERS,

and

A . J . BLOMQUIST:

Pro-

jection of tongue afferents to somatic sensory area I in squirrel monkey (Saimiri sciureus). Brain Res., 7, 208— 220 (1968). B E N J A M I N , R . M . , and C. P F A F F M A N N : Cortical localization of taste in albino rat. J. Neurophysiol., 18, 56 — 64 (1955). B L O M Q U I S T , A . J . , R . M . B E N J A M I N , and R . E M M E R S : Thalamic localization of afferents from the tongue in squirrel monkey (Saimiri sciureus). J. Comp. Neurol., 118, 77 — 81 (1962). B L U M , M., A. E. W A L K E R , and T. C. R U C H : Localization of taste in the thalamus of Macaca mulatta. Yale J. Biol. Med., 16, 1 7 5 - 1 9 2 (1943). B Ö R N S T E I N , W. S.: Cortical representation of taste in man and monkey. I. Functional and anatomical relations of taste, olfaction and somatic sensibility. Yale J. Biol. Med., 12, 7 1 9 - 7 3 6 (1940a). B Ö R N S T E I N , W. S . : Cortical representation of taste in man and monkey. II. The localization of the cortical taste area in man and a method of measuring impairment of taste in man. Yale J. Biol. Med., 13, 1 3 3 - 1 5 6 (1940b). B R A D L E Y , W . H . : Central localization of gustatory perception: an experimental study. J. Comp. Neurol., 121, 4 1 7 - 4 2 4 (1963). B U R T O N , H . , and F . E A R L S : Cortical representation of the ipsilateral chorda tympani in the cat. Brain Res., 16, 5 2 0 - 5 2 3 (1969). C H O W , K . , and K . P R I B R A M : Cortical projection of the thalamic ventrolateral nuclear group in monkeys. J. Comp. Neurol., 104, 5 7 - 7 5 (1956). C O H E N , M . J . , S . L A N D G R E N , L . STRÖM, a n d Y . Z O T T E R M A N :

Cortical reception of touch and taste in the cat. Acta Physiol. Scand., Supp., 135, 40, 1 - 5 0 (1957). C R O S B Y , E . C . , T . H U M P H R E Y , and E . W . L A U E R : Correlative Anatomy of the Nervous System. Macmillan, New York (1962). C U S H I N G , H.: Distortions of the visual fields in cases of brain tumour, (sixth paper). The field defects produced by temporal lobe lesions. Brain, 44, 341 — 396. D E O L M O S , J . , and W . R . I N G R A M : An improved cupric-

silver method for impregnation of axonal and terminal degeneration. Brain Res., 33, 523 — 529 (1971). E L L I O T S M I T H , G.: A new topographical survey of the human cerebral cortex, being an account of the distribution of the anatomically distinct cortical areas and their relationship of the cerebral sulci. J. Anat. (Lond.), 41, 2 3 7 - 2 5 4 (1907). F E R R I E R , D.: The Functions of the Brain. Smith, Elder and Co., London (1886). F I N K , R. P., and L. H E I M E R : T W O methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system. Brain Res., 4, 3 6 9 - 3 7 4 (1967). G A N C H R O W , D., and R. P. E R I C K S O N : Thalamocortical relations in gustation. Brain Res., 36, 2 8 9 - 3 0 5 (1972). G O R S C H K O W , J . P.: Über Geschmacks- und Geruchszentren in der Hirnrinde Inaug.-Diss., St. Petersburg (Russian) Abstract in Neurol. Zentralblatt, 20, 1092-1093 (1901). G R I N K E R , R. R.: Neurology. Charles C Thomas, Springfield (1937). KAHN, E . A.,

E . C. CROSBY,

R . C. SCHNEIDER,

and

J. A.

Correlative Neurosurgery. (2nd Ed.) Charles C Thomas, Springfield (1969). K E N N E D Y , F.: The symptomatology of temporosphenoidal tumors. Arch. Int. Med., 8, 3 1 7 - 3 5 0 (1911). L A N D G R E N , S.: Convergence of tactile, thermal and gustatory impulses on single cortical cells. Acta Physiol. Scand., 110, 2 1 0 - 2 2 1 (1957). L E G R O S C L A R K , W. E.: The connections of the arcuate nucleus of the thalamus. Proc. Roy. Soc. London, Ser B, 123, 1 6 6 - 1 7 9 (1937). L O C K E , S.: Thalamic connections to insular and opercular cortex of monkey. J. Comp. Neurol., 129, 219—240 (1967). M A C H T , M . B.: Subcortical localization of certain "taste" responses in the cat. Fed. Proc., 10, 88 (1951). O L S Z E W S K I , J.: The Thalamus of the Macaca mulatta. Karger, Basel (1952). P A T T O N , H. D., and V. E. A M A S S I A N : Cortical projection of chorda tympani nerve in cat. J. Neurophysiol., 15, 245— 250 (1952). P A T T O N , H. D., and T. C. R U C H : The relation of the foot of the pre and postcentral gyrus to taste in the monkey and chimpanzee. Fed. Proc., 5, 79 (1946). P A T T O N , H. D., T . C . R U C H , and A . E . W A L K E R : Experimental hypogeusia from Horsley-Clarke lesions of the thalamus in Macaca mulatta. J. Neurophysiol., 7,171 — 184 (1944). P E N F I E L D , W., and T. C. E R I C K S O N : Epilepsy and cerebral localization. Charles C Thomas, Springfield (1941). P E N F I E L D , W., and H. J A S P E R : Epilepsy and the Functional Anatomy of the Human Brain. Little, Brown and Co., Boston (1954). P E N F I E L D , W., and T. R A S M U S S E N : The Cerebral Cortex of Man. Macmillan, New York (1950). R O B E R T S , T., and K. A K E R T : Insular and opercular cortex and its thalamic projection in Macaca mulatta. Schweiz. Arch. Neurol. Neurochir. Psychiat., 92, 1 — 43 (1963). R U C H , T. C., and H. D. P A T T O N : The relation of the deep opercular cortex to taste. Fed. Proc., 5, 89 — 90 (1946). TAREN:

RUDERMAN, M . I.,

A.R.MORRISON,

and

P.J.HAND:

A

solution to the problem of cerebral cortical localization of taste in the cat. Exp. Neurol., 37, 5 2 2 - 5 3 7 (1972). S A N I D E S , F.: The architecture of the cortical taste nerve areas in squirrel monkey (Saimiri sciureus) and their relationship to insular, sensorimotor and prefrontal regions. Brain Res., 8, 87 — 124 (1968).

Centrai Gustatory Pathways in the Monkey SCHWARTZ, H . G . ,

G . E . ROULHAC, R . L . L A M ,

and

J. L.

O'LEARY: Organization of the fasciculus solitarius in man. J . Comp. Neurol., 94, 2 2 1 - 2 3 9 (1951). S H E N K I N , H . A., and F . H . L E W E Y : Aura of taste preceding convulsions associated with a lesion of the parietal operculum: Report of a case. Arch. Neurol. Psychiat., 50, 3 7 5 - 3 7 8 (1943). VILLIGER, E . : Brain and Spinal Cord. (3rd Ed., W . H. F . Addison, Editor) J . B . Lippincott Co., Philadelphia. 4 9 2 5 . VON B E C H T E R E W , W . : Die Functionen der Nervencentra. Fischer, J e n a (1908 — 1911). WALKER, A. E . : The thalamic projection to the central gyri in Macacus rhesus. J . Comp. Neurol., 60: 161 — 184 (1934). WALKER, A. E . : The Primate Thalamus. The University of Chicago Press, Chicago (1938).

43

and Y . K A W A M U R A : Cortical responses to electrical and gustatory stimuli in the rabbit. Brain Res., 94, 4 4 7 - 4 6 3 (1975). ZOTTERMAN, Y . : Studies in the nervous mechanism of taste. Exp. Cell Res., Supp 5, 5 2 0 - 5 2 6 (1958). YAMAMOTO, T . ,

Author's

address:

Francis Cleveland KINNEY, Ph. D. Department of Anatomy University of Alabama in Birmingham B o x 317, University Station Birmingham, Alabama 35294

J . Hiniforsch. 19 (1978) 45 - 73

Anatomisches Institut des Bereiches Medizin (Charité) der Humboldt-Universität Berlin (Director: Prof. Dr. sc. med. W . K I R S C H E )

Experimental alterations in number and length of different membrane complexes on axosomatic contacts in the trout (Salmo irideus, Gibbons 1855) By Thomas

SCHUSTER

With 10 figures and 7 tables (Received 12 th April 1977) Summary: 1. Subjects of this research are the mixed synapses between secondary vestibular axon terminals and perikarya of the oculomotor nucleus of the trout (Salmo irideus, GIBBONS 1855). These mixed synapses of characteristic shape and extraordinary size show three different types of special membrane complexes: active zones, gap junctions and desmosome-like structures. 2. Quantitative estimations regarding the plasticity of these mixed synapses have been performed on ultrastructural level. Test condition was a different labyrinthine receptor stimulation. For this purpose the fish had to swim for a period of 5 or 14 days a) in violantly streaming and edding water or b) in steadily running water (a: "stressed fish", b: "immobile fish"). After counting the number and measuring the length of the different membrane complexes the "portion" of the membrane complexes (product of number and length) has been estimated. 3. The obtained results are in the control group as follows: portion of active zones at the contact zone — 15.8%, portion of gap junctions — 6.4%, portion of desmosome-like structures — 6.6%, total portion of all membrane complexes at the contact zone — 28.8%. The mean length of the different membrane complexes are: active zones — 0.33 [im, gap junctions — 0.23 (jtm, desmosome-like structures — 0.22 (jtm. 4. The changes of the different membrane complexes in relation to test condition and time are the following: The portion of active zones of 14-d-immobile fish covers 8.8% of the contact zone. This value increases to 26% in the 14-d-stressed fish. By contrast, the portion of desmosome-like structures and gap junctions falls from 13.2% or 17%, respectively (immobile fish) to 6.2% or 3.6%, respectively (stressed fish). Comparing the total portions of all membrane complexes it is seen that the 14-d-groups (stressed and immobile fish) show a higher proportion of membrane specialization than the control group (39% in immobile fish, 3 5 . 8 % in stressed fish, in contrast to 28.8% in the control group). 5. Regarding the positioning of the three different membrane specializations, as a rule, a close vicinity of gap junctions and desmosome-like structures is striking. This fact correlates with the quantitative changes of the portions (see above) and is in accordance with the literature. 6. With regard to these results the experimental conditions, especially the influence of stress factors are discussed. Difficulties in differentiating the three types of membrane specializations are mentioned. 7. Changes in active zone portions are discussed in relation to a neuromodulating function of ACh, the likely transmitter at these chemical synapses. The active zone increase in stressed fish may be interpreted as the effect of an functional "overloading". 8. Comparing the metabolic dependence of both chemical and electrotonic transmission gap junctions are supposed to be more independent transmission structures, their changes may represent a compensatory mechanism in metabolic transport processes. 9. Juxtaposition of desmosome-like structures and gap junctions as well as the simultaneous changes of both structures support the idea of a functional relationship between them.

Zusammenfassung: 1. Untersuchungsobjekt dieser Arbeit waren die „gemischten" Synapsen, die zwischen sekundären vestibulären Axon-Terminalen und Perikarya des Okulomotorius-Kerns der Forelle (Salmo irideus, GIBBONS 1855) ausgebildet werden. Diese außerordentlich großen und charakteristisch gestalteten Kontakte besitzen drei unterschiedliche Typen von Membranspezialisierungen: Aktivzonen, "gap junctions" und desmosomen-artige Strukturen. 2. Die quantitativ-elektronenmikroskopischen Untersuchungen galten der Plastizität dieser gemischten Synapsen, die mittels unterschiedlich starker Reizung der Labyrinth-Rezeptoren getestet werden sollte. Zu diesem Zweck hatten die Fische 5 bzw. 14 Tage a) in stark strömenden und wirbelndem Wasser („Belastungstiere") oder b) in ruhig und gleichmäßig fließendem Wasser zu schwimmen („Ruhetiere"). Nach Zählung und Messung der speziellen Membrankontakte wurde aus dem Produkt beider der „Anteil" des entsprechenden Membrankomplexes an der Gesamtlänge der Kontaktzone bestimmt.

46

Schuster, Th. 3. Inder Kontrollgruppe wurden folgende Werte ermittelt: Anteil der Aktivzonen an der Kontaktzone — 15,8%, gap junctions — 6,4%, desmosomen-artige Strukturen — 6,6%, Gesamtanteil aller speziellen Membrankomplexe — 28,8%. Durchschnittliche Länge der verschiedenen Membrankomplexe: Aktivzonen — 0,33 ¡¿m, gap junctions — 0,23 |j.m, desmosomen-artige Strukturen — 0,22 ¡¿m. 4. Die Änderung der Anteile der Membrankomplexe in Abhängigkeit von den Versuchsbedingungen: Der Anteil der Aktivzonen beträgt bei 14-d-Ruhetieren 8,8%. Er steigt bei 14-d-Belastungstieren auf 26% an. Im Gegensatz dazu sinkt der Anteil der desmosomen-artigen Strukturen und gap junctions von 13,2 bzw. 17% bei Ruhetieren auf 6,2 bzw. 3,6% bei 14-d-Belastungstieren. Der Gesamtanteil aller speziellen Membrankomplexe liegt sowohl bei 14-d-Belastungstieren (35,8%) als auch bei 14-d-Ruhetieren (39%) über dem Wert der Kontrolltiere (28,8%). 5. Im Hinblick auf die Lage-Beziehungen der drei unterschiedlichen Membrankomplexe zueinander war die häufige, enge räumliche Nachbarschaft von gap junctions und desmosomen-artigen Strukturen auffällig. Dieser Befund stimmt mit den quantitativen Veränderungen des Anteils beider Strukturen, die simultan erfolgen, überein und steht mit den Angaben aus der Literatur in Einklang. 6. Die Ergebnisse werden zunächst in Hinblick auf mögliche Einflüsse von Streß-Faktoren diskutiert. Weiterhin wird auf die Schwierigkeiten bei der Differenzierung der drei speziellen Membrankomplexe eingegangen. 7. Die Veränderungen des Anteils der Aktivzonen werden in Zusammenhang mit einer neuromodulatorischen Funktion des wahrscheinlichen Transmitters ACH betrachtet. Der Anstieg des Aktivzonen-Anteils bei Belastungstieren wird als das Resultat einer Reizüberlastung ("overloading") angesehen und umgekehrt. 8. Beim Vergleich der Stoffwechsel-Abhängigkeit der chemischen und der elektrotonischen Transmission scheinen gap-junction-Strukturen weniger abhängig zu sein; die entgegengesetzt zu den Aktivzonen erfolgende Änderung des Anteils der gap junctions könnte als ein Kompensationsmechanismus beim interneuronalen Stofftransport angesehen werden. 9. Enge Nachbarschaft von gap junctions und desmosomen-artigen Strukturen sowie simultane Änderung des Anteils beider Strukturen unter Belastung sprechen für einen funktionellen Zusammenhang zwischen beiden.

Introduction Two levels exist at which synapses influenced by experience may be examined morphologically: 1st — new formation or involution of synapses (changes in number), 2nd — modulation of specialized structures of individual synapses. If both changes in the number of synapses and changes in the structure of existing synapses are regarded as changes in efficiency of transmission we are able to examine what has been termed "synaptic plasticity" (BLOOM1970; CRAGG 1972a; STREIT et al. 1972). In this research only changes in the number of synapses have been controlled. Whereas quantitative data on chemical synapses (active zones) are available, until now the plasticity of gap junctions or "mixed synapses" (contacts showing both active zones and gap junctions) has not yet been investigated electron microscopically. This fact contrasts with the occurence of these special synaptic contacts. Gap junctions — the morphological correlate of electrotonic transmission — are not only common in lower vertebrate nervous tissue, they have been found in mammalian C. N . S . (SOTELO 1 9 7 5 ) , in primate neocortex (SLOPER 1 9 7 2 ) and in man (MOLLGARD 1 9 7 5 ) . Mixed synapses are also present throughout the vertebrate phylum (Table V, see below). These findings are not in agreement with the assumption that gap junctions may be important only in lower vertebrates. Because of their extraordinary size and characteristic shape the mixed synapses in trout oculomotor nucleus ("giant synapses" — KIRSCHE 1967) are especially suitable for studying quantitative changes

in synaptic activity electron microscopically. Three types of special membrane complexes occur at these axosomatic contacts between secondary vestibular axon terminals and oculomotor perikarya: active zones, gap junctions and desmosome-like structures. To examine the plasticity of these complex mixed synapses the question should be answered in what way a functional "overloading" or "underloading" of the labyrinthine receptors changes number or size of the three different membrane complexes. Some of these results have been partly reported (SCHUSTER 1976, 1977). Materials and methods Material

and experimental

conditions:

Rainbow trouts (Salmo irideus, GIBBONS 1855) ranging from 11 to 14 inches in length were used. No special selection was made in relation to sex. In order to achieve relatively natural stimulation conditions for the stimulation of the labyrinthine receptors the fish had to swim for two longer periods (5 days or 14 days) in water moved to a different degree. A high and long-lasting stimulation of the labyrinthine receptor cells should be accomplished by swimming in violantly streaming and edding water. To put this into effect the trouts were held in a wire net cage (22 x 24 inches wide and a water depth of about 22 inches) which was placed in a sluice. The success of feeding was douptful under the conditions of the violantly streaming water in the small cage. In the following these fish will be named "stressed" animals. A low stimulation (with few and weak stimuli per time unit) should be achieved by swimming in steadily running water. For this purpose the trouts were held in a channel about 2 m long and 50 cm wide. The water depth here was

Experimental alteration in mixed synapses about 10 cm. These fish will be named "immobile" ones in the following. All other conditions like water composition, O a -content etc. were the same for all groups examined. The mean temperature was 7°C. One point of uncertainty was the feeding of the trouts. In spite of regular feeding it may be, that the stressed group (partly) did not get enough feed because of the restricted food intake. Therefore a difference in nutritional condition of stressed fish as against the immobile ones may exist. The food was the same for all groups and the same as before the beginning of the experiment. Five groups of fish were examined: control group (normally held fish in the pools), 5 days and 14 days stressed animals, 5 days and 14 days immobile animals. All tests were carried out within a period of three weeks in February/March. I t should be mentioned, that in the group " 1 4 days stressed fish", which covered 12 animals at the beginning of the experiment, 4 animals died within this period. This fact will be discussed below.

The total number of fish available for the electron microscopic analysis was 33 (Table I). In contrast to the stressed fish all trouts swimming in the steadily running water survived, but their behaviour was changed: the. fish were swimming slowly and they could be seized without difficulty.

T a b l e I. Number of animals in the experiment. group

below the ventricular surface. For further investigation blocks were used only from the dorsomedial region of the oculomotor nucleus i. e. the region of the so-called claw-shaped giant synapses (KIRSCHE 1967, SCHUSTER 1 9 7 1 ) .

Fixation, staining and microscopy: Blocks were fixed in cold glutaraldehyde (2.5%) solution for 12 hrs. The buffer used was 0.1 M cacodylate adjusted to a pH of 7.4. After an extensive washing procedure in cold buffer the material was postfixed with osmium tetroxide in cacodylate buffer for one hour and again washed repeatedly. The fixed blocks of tissue were then dehydrated in seven 5 to 20 min steps in graded acetone. Embedding was in micropal. The fixation effect varied greatly. Some pieces were well fixed, whereas others showed heavy destructions of the tissue. Thick sections (1 ¡¿m) for light microscopy were stained with hot 1% methylene blue in 1% borax. Thin sections showing silver interference colors were cut with glas knives by a L K B ultramicrotome (Ultrotome III). Sections were stained with lead citrate and uranyl acetate (REYNOLDS 1963). Tissue was examined with a TESLA BS 613 microscope.

number of trouts at the begin- number of ning of exfish which periment died within test period

control 5 days 14 days 5 days 14 days

47

at the end of experiment

6

immob. immob. stressed stressed

5 6 9 12

Preparation: The preparation was performed without anesthetizing and perfusion to largely exclude short-term effects on the morphological substrate. The fish were sacrificed by decapitation. The time for removing the brain and excising blocks of tissue was within one minute. The procedure of preparation was as follows: The upper part of the skull was removed to expose the brain. Then the tectum was removed and an aqueous glutaraldehyde solution (2.5%) was applied rapidly for a first fixation by injecting cautiously into the midbrain ventricle. After amputating the nerves and blood vessels in the base of the brain it was put into an ice-cooled Petri dish filled with fixing solution. A razor blade was used to obtain two slices each 1 mm thick of the tegmental oculomotor region. 4 cubes each 1 mm 3 were excised from the region just

Criticism of preparation and fixation method: The advantages of the preparation method used here are the absence of influences caused by narcotics or by stress (operation) which may occur during the perfusion of animals. The risks are perhaps the occurence of hypoxia during preparation and the distorsion of the brain mass by cutting the tissue. However, glutaraldehyde used as primary fixative has a relatively fast penetration rate (PETERS 1970), and the nervous structures which have to be preserved lie just below the ventricle surface. The results of this preparation method were quite different. According to the definition of a "good fixation" by PALAY et al. (1962) only "good fixated" material has bsen used. Shrinkage of tissue: For the quantitative estimations (see below) the influence of shrinkage produced by fixation, dehydratation and embedding has not been taken into account. But because of the equal treatment of tissue in all groups tested the shrinkage is the same and its possible influence is negligible from this reason.

Counting and measuring Electron micrographs with a magnification of 100.000 x were used. After measuring the length of the contact zone number and lenght of active zones

48

Schuster, Th.

gap junctions and desmosome-like junctions were estimated. Because of tangential or oblique sections of the contact zones a clear differentiation between the different membrane complexes was not always possible. After counting the different membrane specializations their number has been correlated with a standardized contact zone of 5 JI.m by transformation in order to compare all data from contact zones of length differing to a great extend. To get a clear picture of the true relations the term "portion of membrane complexes" has been introduced. This "portion" has been estimated as the product of length and number (mean values). The portion therefore means the t o t a l length of a certain membrane complex along the contact zone either in |xm per 5 ¡Jim contact zone or in %. Statistics Five animals per group were analyzed. Only blocks of the dorsomedial region of the oculomotor nucleus were used (by identifying the claw-shaped giant synapses). The axosomatic contacts of about 3—4 cells per block were investigated, at least contacts of 2 0 — 2 5 cells per group. No special random-sampled method was used. After a first estimation of mean values and errors a runaway-test (HULTZSCH 1 9 6 1 ) was applied to eliminate single values indicating a striking deviation from the mean. This single value xt was eliminated if the following condition had been fullfilled: \x, - x| > A|S.D.| where x is the mean value, k a factor depending on n and significance, and S.D. standard deviation. Mean values and errors (standard deviation S.D. and standard error of mean S.E.M.) were then calculated without using transformation methods. The test for significance was made by means of variance analysis. Results Qualitative characteristics Membrane

complexes

In the trout oculomotor nucleus lightmicroscopically large endings of characteristic size and shape are visible (HORSTMANN 1 9 5 3 / 5 4 - 1 9 5 6 ; KIRSCHE 1 9 6 7 ) . The so-called claw-shaped type of these "giant synapses" (KIRSCHE 1 9 6 7 ) occurs only in the dorsomedial region of the nucleus (KIRSCHE 1 9 6 7 ; SCHUSTER 1 9 7 1 ) . The claw-shaped end-apparatus consists of many large end-knobs, arising by branching of the secondary

vestibular axon terminal. As a rule, the knobs are in contact with the large oculomotor somata or with the dendritic main trunk (for details see SCHUSTER 1971, 1973). Only this type of ending was investigated. The contacts between these large terminals and the oculomotor perikarya form the following three specialized membrane complexes: active zones, gap junctions and desmosome-like structures. These three types are evident in both perfused and non-perfused material (normally kept fish) (Figs. 1 and 2). Likewise they can be found in stressed and immobile fish (Figs. 3 and 4). The three membrane complexes have been differentiated by means of the following morphological ciriteria: 1) Active zones: Pre- and postsynaptic thickening of varying degree, gap widening and local accumulation of clear synaptic vesicles are typical features. But the individual active zones show many variations: The membranes may be curved (Figs. 2 b, 3g, 5 a) or straight (Figs, l b , c, 3d), or they may be folded (Figs. 5 c, f). The asymmetrical membrane thickening may be well visible (Figs. 5 a, d) or less well (Figs, lb—d). Clearly symmetrical synaptic contacts (GRAY-type II, GRAY 1959) have not been observed at these axosomatic contacts; they have been rarely seen in the neighbourhood (Fig. 3 b). The concentration of round clear vesicles varies between a local accumulation in close contact with the presynaptic membrane including omega-figures (Figs, lc, 3f, 5 a, b, d) and a loose distribution of synaptic vesicles among the the presynaptic axoplasm (Fig. 2d). Large granular vesicles of different degrees of filling occur in a small amount (Figs, l b , e, 3c, d, 5a, b, e) or they are absent (Figs, lc, 2e, 5d). Regularly spaced dense projections according to GRAY and GUILLERY (1966) most clearly visible after PTA-treatment, were also noted. Sometimes the dense projections were seen extremly clear (Fig. 3d). They were also discernible in tangentially sectioned synapses (Fig. lg). But the presentation of these dense projections varied considerably. In some special membrane complexes which seem to be desmosomelike structures, "dense projection"-like particles have also been observed (Figs, le, f, 2e). 2) Gap junctions: The neuronal gap junctions observed here are characterized by the apposition of the two membranes to within 20—40 A (in agreement with R E V E L and KARNOVSKY 1967); SOTELO (1975) measured 15—20 A). The overall thickness of the apposed membranes falls in the range of 150 to 170 A. Because of the fixation method used here instead of a gap a median dense line is visible in perpendicular sections (BRIGHTMAN and R E E S E 1969). Therefore two main features are useful in identifying

mfim

^ J g S g i

Fig. 1. Perfused fish. Axosomatic contact zones between secondary vestibular axon terminals and oculomotor somata. a) The contact zone shows from right to left an active zone (A), gap junctions (arrows) and desmosome-like structures (D). 46 000 x . b) and c) Desmosome-like structures between two (peripheral) active zones, b) 78 000 x , c) 95 000 x . d) Two gap junctions (arrows) in close vicinity to a desmosome-like structure, to the right hand an active zone. 120 000 X . e) A desmosome-like structure (in the centre of the contact zone) showing "dense projections" which seem to be spaced regularly. 70 000 X. f) and g) Similar dense projections at a partly tangentially sectioned membrane contact. See Fig. 2e and f. 6 5 0 0 0 x . Hirnforschung, Bd. 19, Heft 1

4

50

Schuster, Th.

Fig. 2. Normally kept fish, decapitated, non-perfused. a) b) c) d) e)

Desmosome-like structures, in the centre of the contact zone arranged in juxtaposition to a gap junction. 82000X. An active zone and one large gap junction (with clear "synaptic" vesicles), partly rotating into a tangential plane. 57000X Detail from b). 160000 x . Desmosome-like structures and "omega"-figures (arrow). 77000X. and f) Desmosome-like structures showing "dense projections" in a regular pattern (arrows). e) 90000 X. f) 76 000 X.

Experimental alteration in mixed synapses

4*

51

52

Schuster, Th.

Fig. 3. Stressed fish. a) b) c) d) e) f)

From left to right: desmosome-like structure, gap junction and active zone. 106000x. GRAY-type-II-synapse (to the upper right). 9 0 0 0 0 x . The three different types of membrane specializations. 45 000 x . Clearly visible dense projections at an active zone. 54 000 X. The section of the contact zone often shows one membrane specialization following the other. 42 000 x . Gap junctions in' close vicinity with desmosome-like structures. The fine structure of the gap junctions is partly visible. 136000X. g) An active zone and an extensive perpendicularly sectioned gap junction contacted by clear synaptic vesicles. 135 000 X.

Experimental alteration in mixed synapses

53

54

Schuster, Th.

Fig. 4. Immobilized fish. a) All special membrane complexes. From left to right: both small gap junctions and desmosome-like structures alternatively arranged. A large gap junction and a desmosome-like structure in the middle of the contact zone: see inset (b). a) 38000X. b) 8 0 0 0 0 x . c) and d) gap junctions are often observed, c) 41000 x d) 45000 x . e) The figure illustrates the difficulties in differentiating clearly between active zones and desmosome-like structures. 62 000 x . f) An arrangement of desmosome-like structures without the presence of other types of membrane specializations. 65 000 x .

Experimental alteration in mixed synapses

ivr?* -. -i

55

56

Schuster, Th.

Fig. 5. Evaginations and indentations. a) A somatic evagination. The contacting membranes are forming active zones. 5-d-immobile fish. 53 000 x . b) A somatic evagination with active zone and gap junction. 14-d-immobile fish. 79000 x . c) Evagination, totally surrounded by the axon terminal. The membranes are forming several active zones and gap junctions. 14-d-immobile fish. 32 000 x . d) An axonic invagination into the soma. The contact zone shows a number of active zones which are partly obliquely sectioned. 14-d-stressed fish. 40 000 x . e) A "finger-like" protrusion into the axon terminal. 14-d-stressed fish. 41000X. f) Invagination of an axon terminal into the soma of an oculomotor neuron showing active zones with a high accumulation of clear synaptic vesicles, 14-d-stressed fish. 4 0 0 0 0 x ,

Experimental alteration in mixed synapses

lifr

57

Experimental alteration in mixed synapses

NUMBER

59

per 5/um contact zone

ACTIVE ZONES significant differences:

u3

H"d"im.» Control

2

5-d-str.«1/,d-str.

1

Control" K " d - s t r . U d 5d immobile fish

Control

5d Ud stressed fish

DESMOSOME-UKE STRUCTURES significant differences: U - d - i m . " Control 5 T H m "Control

Ud 5d immobilefish

Control

5d 14d stressedfish

GAP JUNCTIONS significant differences: U _ d " i m . " Control H-d-im. "5-d

lm

5~d_im." Control

Control" U - d - s t r . Kd 5d immobile fish

Control

Fig. 6. Number of the three different membrane complexes per 5 |xm contact zone in the five test groups. Errors are expressed as S.E.M. Values see T a b l e II.

5d Kd stressed fish

Indentations and evaginations without specializations have never been seen.

synaptic

membrane complexes (product of lenght and number, see material and methods) at the contact zones yields the following results (Table III): The "total portion" Quantitative estimations of the three membrane complexes at the contact zone The results of the quantitative analysis are summa- (control group) is about 30%. This means: 30% of the rized in tables II—IV and figures 6—10. The mean length of each contact zone are covered by membrane values of the l e n g h t of the three membrane com- specializations. Active zones cover 16%; gap juncplexes (control group) are (Table II) : active zones tions and desmosome-like structures each covers — 0.33 (im (n = 116), desmosome-like structures — about 6.5% of the lenght of the contact zone. There0.22 ¡xm (n = 113), gap junctions — 0.23 ¡xm (n = 91). fore the relation between the three types in the The variability is high. The longest diameters of active control group is: active zones : desmosome-like zones (measured in stressed fish) were 1.10 (im and structures : gap junctions = 2.5 : 1 : 1. These values 1.20 (I,m. The longest measured gap junction (control deviate from those cited by SCHUSTER (1973) (active group) was 1.50 |I.M. The mean values of the n u m - zones : gap junctions = 4 : 1 ) . This latter relation b e r s of the three different membrane complexes per had been obtained from 7 analyzed contact zones 5 [¿m contact zone are in the control group: active (of anaesthetized and perfused fish) as compared zones — 2.42, desmosome-like structures — 1.52, with about 60 ones now. Moreover at that time no gap junctions — 1.60 (of about 60 analyzed contact differentiation had been made between active zones and desmosome-like structures. Adding the two zones). The estimation of the "portion" of the specialized values of active zones and desmosome-like structures

58

Schuster, Th.

gap junctions: First, an association of electron dense or semidense material lies in the cytoplasm on either side of the two apposed membranes (cytoplasmic fuzz). Second, the presence of a 100 A hexagonal lattice is an important feature of the junction. In perpendicular sections therefore bands occur at an interval of about 85 —100 A (probably the center-tocenter distances of the interplasmic channels) (Figs. 2 c, 3f, g). The identification of the hexagonal arrays in tangential sections ("honeycomb" pattern — ROBERTSON 1970) was difficult with this preparation method. Only if the gap junction first sectioned perpendicularly is continuously rotating into a tangential plane one may be sure to call the whole complex a gap junction. Commonly on the axoplasmic side of the gap junction vesicles are present, but in a small number (Figs. 2a, b, f, 3f, g, 4b). Not only clear synaptic vesicles but also large dense core vesicles occured above these junctions (Figs, l a , 3e, 4a, c). Pedicle-like connections between the clear synaptic vesicles and the "presynaptic" membrane were observed (Figs. 3f, g, 4 a). 3) Desmosome-like structures: The gap widening occurs in the range of the active zone gap with little amount of electrondense material within the gap. Behind the membrane on each side electron-semidense material of a different degree is visible in the cytoplasm (Figs, lb—d, 2a, d—f, 4a—b, f). Clear spherical ("synaptic") vesicles do not occur in a high accumulation, but they may be present in a small number (Figs. 2 a, d, 4f). Omegafigures ( A K E R T 1973) sometimes are visible at these desmosome-like structures (Fig. 2d). These figures seem to be formed by spiny vesicles (ANDRES 1964; GRAY and W I L L I S 1970; SCHUSTER 1974). Two points should be stressed here: A) Because of the often similar picture of active zones and desmosome-like structures, at least at these contacts — gap widening, slow and unclear membrane thickening — only one characteristic difference exists: a high accumulation of vesicles at the chemical synapses in contrast to the absence or low accumulation of vesicles at desmosome-like structures. For this reason K R I E B E L et al ( 1 9 6 9 ) speak of desmosome-like complexes both "with closely associated presynaptic vesicles" and "desmosome-like complexes that lacked any accumulation of synaptic vesicles". B) It should be remembered, that clear spherical vesicles occur at all membrane complexes mentioned above, but in a different number. Positioning

of the membrane

complexes

The three membrane complexes occur in a different manner along the sectioned contact zone between

axon terminals and perikaryon. In the electron micrograph they may lie separated and independent from one another. On the other hand these structures seem to be organized in ensembles. So for instance desmosome-like structures sometimes occur in a high number alongside the contact zone, only interrupted by short intervals at which the membranes show their normal picture, whereas other specializations are absent (Fig. 4f). But as a rule, one may say that there exists a close vicinity between gap junctions and desmosome-like structures regardless of a single or multiple presence of the latter one (Figs. Id, 2a, f, 3e, f). In such cases the gap-junction-like regions lie between the gap-widened desmosome-like complexes (Fig. 4b). In the majority these gap junction regions are very small. The fact of the close local relations of the two membrane specializations is supported by the quantitative estimations (see below). Evaginations

and

indentations

As noted by previous investigators the perikaryal surface of the fish oculomotor neuron sometimes show pockets or indentations into which terminals protrude (WAXMAN and PAPPAS 1971) (Figs. 5d, f). Axons deeply invaginating into the soma are also reported by B E N N E T T et al (1967 d) in the relay nucleus of a gymnotid fish. On the other hand evaginations of the soma surface similar to those in the chick ciliary ganglion (SZENTAGOTHAI 1964; TAKAHASHI and HAMA 1965; TAKAHASHI 1967; CANTINO andMuGNAiNi 1975) are visible, surrounded by terminals (Figs. 5 a, b, c, e). But these formations are scarcely seen in control animals. They were found in stressed fish (Figs. 5 a, d, e, f) as well as in immobile ones (Figs. 5b, c). In contrast to the "soma-spikes" in chicks (TAKAHASHI 1967) shape and size of the evaginations are irregular. Besides they are too scarce to speak of genuine somatic thorns or spines described for instance in the goldfish tangential nucleus (HINOJOSA 1973) or in the cerebellar central nuclei of the cat (ANGAUD and SOTELO 1973). Unlike short dendritic expansions mitochondria do not occur in the cytoplasm of the evaginations. Both soma indentations and evaginations form synaptic contacts with the adjacent terminals. Active zones and gap junctions are visible in the electron micrographs. In the indentations active zones seem to be in the majority. The distances between the single synaptic contacts are very small (Figs. 5d, f). The somatic evaginations reveal active zones and gap junctions (Figs. 5 a, b, c). In contrast to WAXMAN and PAPPAS (1971) desmosome-like structures were not observed; this is in agreement with the chick ciliary neuron evaginations (CANTINO and MUGNAINI 1975).

60

Schuster, Th.

Table II. Number and length of the different membrane specializations in the different test groups. Results and significances (p). Values are expressed as means ± S. E. M. group

feature active zones

desmosome-like structures

gap junctions

number

length

number

length

number

length

P 1 [%]

5

1

1

5

1

NS

14-d-im. fish

1.85±0.16

0.24 ± 0.01

3.31 ± 0.20

0.20 ± 0.01

4.05 ± 0.28

0.21 ± 0.01

P [%]

NS

NS

NS

NS

1

1

5-d-im. fish

1.99 ± 0.22

0.24 ± 0.01

2.64 ± 0.26

0.18 ± 0.01

2.86 ± 0.26

0.15 ± 0.01

P [%]

NS

1

1

1

1

1

control

2.42 ± 0.16

0.32 ± 0.01

1.52 ± 0.18

0.22 ± 0.01

1.60 ± 0.17

0.23 ± 0.01

P [%]

NS

NS

NS

NS

NS

NS

5-d-str. fish

2.81 ± 0.19

0.32 ± 0.01

1.23 ± 0.20

0.21 ± 0.01

1.66 ± 0.16

0.23 ± 0.02

P [%]

1

NS

NS

NS

1

5

14-d-str. fish

3.54 ± 0.13

0.37 ± 0.01

1.91 ± 0.14

0.22 ± 0.01

1.02 ± 0.13

0.18 ± 0.01

p [%]

1

5

NS

NS

1

1

2

number = number per 5 |im contact zone. p 1 = p between 14-d-immobile fish and control. p 2 = p between 14-d-stressed fish and control. NS = non significant.

L E N G T H injum ACTIVE ZONES % Q3 02 i

0,1

• •III

significant differences:

14-d'im.»Control 5-d-im.«Control ControM4"d"str

14 d 5d immobilefish

Control

5d 14 d stressedfish

DESMOSOME-LIKE STRUCTURES signicant differences:

Q4-

Q3-

14"d-im."Control 5 - d - i m . "Control

Q20,1" 5d 14 d immobilefish

Control

5d 14 d stressedfish

GAP JUNCTIONS significant differences:

Q4 0,3

14-d"im. " 5 - d - i m . 5-d"im.»Control 5-d-str.» 14-d-str. Control» 14-d-str.

0,2-

Q1

- >H 14 d 5d immobilefish

Control

5d 14 d stressed fish

published here we come to a ratio of 3.5:1. This ratio is not much different from that of 4:1. The changes of the different membrane complexes in relation to test conditions and time are as follows (Tables I I I and IV): Clear changes are visible in the numbers of membrane complexes (Fig. 6). Whereas not all groups show significant differences from one another, a distinct trend seems to be outlined for all membrane complexes: The number of active zones decreases in immobile fish and increases in stressed fish. By contrast, the numbers of desmosome-like complexes and gap junctions increase in immobile fish. In stressed fish the numbers of both structures do not change markedly. The values of lengths alter as follows (Fig. 7): The length of active zones is decreased in immobile fish but increased in stressed fish. The lenght of both desmosome-like structures and gap junctions changes only to a small extend. In contrast to the changes in their numbers the picture of these alterations does not show any clear trend. A better representation of the "plasticity" of the different membrane specializations is evident by introducing again the "portion" of membrane specializations into the quantitative analysis (Figs. 8 a—c; Table III). Accordingly the portion of active zones Fig. 7. Length of the three different membrane complexes in (xm. Errors are expressed as S.E.M. Values see Table II.

Experimental alteration in mixed synapses

I total portion of all special membrane

61

complexes in the length of the contact zone [%]

Ik d immobile

5 d immobile

E l

control

5 d

stressed

14 d stressed

H

8a

Fig. 8. Portions (number x length) of the different membrane complexes in % (at the left corner of each column). Values see Table I I I . For explanation see text.

JT portion of special membrane

%d immobile

5d

complexes

in the length of the contact

ione[%]

30

u?

Ï7

immobile

control

5d

stressed

M d

stressed ¡active

8b

zones

H gap junctions®

desmosome -1,

struct.

M portion of the different

U d immobile

[ J ^ w l

5d

a

immobile

control

5d

U d

stressed

stressed

membrane

complexes

zone/%/

WL

In

Inlill

m

H

111

m

h

131 Iei

m ¡active

in the length of the contact

zones

11 I desmosome-1,

8c

junctions

struct.

Portion of the different membrane complexes at the contact zone [ % ]

Ud

immobile fish Active zones,

5d

control

5d

stressed fish

M H H Desmosome - like structures.

14d

Gap junctions

Fig. 9. The percentual values of the different membrane complexes in the five test groups. The curves clearly show the alterations. See text. Values see Table I I I .

Experimental alteration in mixed synapses

63

Changes of the portion of the different membrane complexes in % Values of control = 0 %

••••Active zoneSj

mamm Desmosome-like structures;

Gap junctions

g. 10. Percentual changes of the portions of the different membrane complexes (values of the control group = 0). Values see Table IV. For further explanation of figs. 9 and 10 see text.

T a b l e III. Portion of the three special membrane complexes (number x length) of the contact zone length in [im per 5 (im contact zone and in %. type of membrane complex

portion 14-d-immobile fish

active zones desmosome-like str. gap junctions desmosome-like str. a n d gap junctions active zones, gap junctions and desmosome-like str.

5-d-immobile fish

control

|im

%

%

0.44 0.66 0.85 1.51

8.8 13.2 17.0 30.2

0.47 * 0.47 0.42 0.89

9.4 9.4 8.4 17.8

1.95

39.0

1.36

27.2.

(i.m

of 14-d-immobile fish covers 8.8% of the contact zone. This value increases to 26% in the group of stressed fish (Fig. 8 a). By contrast, the portions of desmosome-like structures and gap junctions fall from 13.2% or 17%, respectively (immobile fish) to 6.2% or 3.6%, respectively (stressed fish) (Figs. 8c, 9, 10).

5-d-stressed fish

14-d-stressed fish

%

(im

%

¡im

%

0.79 0.33 0.32 0.65

15.8 6.6 6.4 13.0

0.89 0.25 0.33 0.58

17.8 5.0 6.6 11.6

1.30 0.31 0.18 0.49

26.0 6.2 3.6 9.8

1.44

28.8

1.47

29.4

1.79

35.8

¡un

At this point we should briefly return to the local relations (positioning) of desmosome-like structures and gap junctions mentioned above. The quantitative changes of both specialized membrane structures seem to correlate with what has been said previously because their changes occur fairly simultaneously (Fig. 8 c). Summarizing both the percentual values

64

Schuster, Th.

T a b l e IV. Changes in the portion of the three membrane complexes in % (control values = 0). type of

changes in %

membrane complex

fish

14

_d_im_

5

_d_i m .

control

figh

5-d-str. figh

14-d-str. feh

active zones

-44.3

-40.6

0

+13.6

+64.5

desmosomelike str.

+100.0

+42.5

0

-28.8

-6.1

gap junctions

+165.5

+31.2

0

+3.1

-43.7

of desmosome-like structures and gap junctions it is obvious that an active-zone-increase from 8.8 to 26% (from 14-d-immobile fish to 14-d-stressed fish) is contrasted with a decrease of desmosome-like structures and gap junctions from 30.2 to 9.8% (Fig. 8b). Comparing now the total portions of all membrane complexes (Fig. 8 a ; Table III) it is seen that the 14-dgroups (stressed and immobile fish) show a higher proportion of membrane specializations than the control group (39% in 14-d-immobile fish, 35.8% in 14-d-stressed fish, in contrast to 28.8% in the control group). This means that there is a difference of about 10% between the 14-d-groups and the control. Figure 10 and table IV show these changes on the basis that the values of the control group are = 0. From these curves we can see changes in an enormous range especially with regard to the values of desmosomelike structures and gap junctions. Discussion Experimental

conditions

According to GRONOW ( 1 9 7 4 ) , the experimental conditions to which both stressed and immobilized fish were exposed, have to be regarded as stress situations. This means a general mobilization of the fish organism by psychic excitation, increased muscle activity, hunger, infection and so on. Like in mammals an increased brain activity may be expected (HAMBERGER e t al. 1 9 6 9 ) .

The possible known stress factors in fish are (GRONOW 1 9 7 4 ; JACOB 1 9 7 4 ) :

a) stressed fish: — hunger, produced by a restricted food supply — the keeping in the narrow net cage (the small size of the reservoir) — wounds on mouth, fins and skin caused by injuring at the sides and corners of the wire net cage — intensive muscle activity

— an increased susceptibility to infections resulting from the stress conditions, wounds etc. b) immobilized fish: — decreased 0 2 -content of the water in the channels because of its slow and steady flowing — the keeping in the relatively small channel — an increased susceptibility to infections produced by the stress conditions. Total exhaustion (by long-lasting muscle activity) is likely one reason for the death of some 14-d-stressed fish (Doi 1932). Another reason may be the increased susceptibility to bacterial attack or virus infection caused by the stress conditions. It is interesting to note in this connection that the lysozyme level of the blood in so-called "hunger-stress-trouts" reaches its maximum after a period of 14 days (ZARSKE 1975). The lysozyme level of the blood is an expression of the bacterial loading. It may well be that there exist other but unknown stress factors. We should bear in mind that a number of "stressors" (stress factors — S E L E Y E 1 9 3 6 , 1 9 5 0 ) alter the hormonal situation of the fish and the tone of the sympathetic nervous system which influences the metabolic conditions, the oxygen uptake, the protein biosynthesis and so on (BRETT 1964; RANDALL and STEVENS 1 9 6 7 ; GRONOW 1 9 7 4 ) . Hormonal effects on the growth of synapses for instance are discussed by CRAGG ( 1 9 7 2 a). Therefore the following fact should be taken into consideration: The plasticity of the different membrane specializations at these vestibulo-ocular "giant synapses" may be not only the result of varied labyrinthine influences. Because of the different stress factors of the experimental conditions the results are perhaps influenced in an unknown way. Discussion of the qualitative results Regardless of the difficulties in differentiating the membrane specializations three membrane complexes have been distinguished at these contacts: active zones, gap junctions and desmosome-like structures. Concerning the gap junctions it would perhaps be better to speak of gap-j unction-like complexes because it is not always certain that "true" gap junctions have been counted. So for instance very short sections of gap junctions (within the range of about 0.01—0.03 ¡Jim) between adjacent desmosomelike structures are not always to be clearly identified (BENNETT 1973 b). From a functional point of view it seems possible that there exist gap junctions not only of different shape and size (spots, belts, etc.) but also of differing structural principles (STAEHELIN 1974). BENNETT (1973 a,b) introduced the term "class of gap junctions". If the main function of the

Experimental alteration in mixed synapses

gap junctions is the electrotonic transmission, this function would be safeguarded by the presence of the small gap widening to prevent lateral current leakage (ROBERTSON 1 9 7 0 ) . The wide-spread labile membrane appositions ( B E N N E T T et al. 1 9 6 7 d; BRIGHTMAN and R E E S E 1 9 6 9 ) as a result of fixation artifacts do not show any characteristic fine structural elements. Occulding junctions (true "tight junctions") have so far rarely been observed in nervous tissue. They have been found in addition to desmosome-like junctions in the Nc. oliv. inf. ( K I N G and A N D R E Z I K 1 9 7 5 ) , at the receptor cells of the acoustolateralis system ( B E N N E T T 1 9 7 1 ) and at the olfactory epithelium (BRIGHTMAN and R E E S E 1 9 6 9 ) . Neither they have been seen in mixed synapses nor in the special case of the oculomotor region. Another point of uncertainty is the desmosomelike structure and their distinction from active zones lacking high vesicle accumulation, respectively. These difficulties have been mentioned previously (PALAY 1 9 5 6 ; and

ROBERTSON e t a l .

1963;

G U I L L E R Y 1 9 6 4 ; TAKAHASHI a n d

GRAY a n d

GUILLERY 1 9 6 6 ;

COLONNIER HAMA

1965;

BENNETT e t al. 1967 c ;

K R I E B E L et al. 1 9 6 8 ) . This problem is of special interest in studying embryonic, immature and cultured nervous tissue ( B U N G E et al. 1 9 6 7 ; AGHAJANIAN and

BLOOM

1967;

BLOOM a n d

AGHAJANIAN

H A M O R Y 1 9 6 9 ; LARRAMENDI 1 9 6 9 ; MUGNAINI

65

ture along the same contact zone has been repeatedly described in the vertebrate C.N.S. For the desmosome-like structure several terms have been used ("desmosomoid differentiations" — ROBERTSON 1965; "attachment plaques" — W A X M A N and P A P P A S 1968; SOTELO and P A L A Y 1970; "puncta adhaerentia" — N A K A J I M A 1974; "desmosome-like plaques" — ROSEMARIE D U N N 1975). Contact zones with the combination of the two complexes active zone and gap junction — the so-called mixed synapses ( P E T E R S e t a l . 1 9 7 0 ; SOTELO a n d P A L A Y 1 9 7 0 ; SOTELO

and

1970; SOTELO and ILINAS 1972) — nearly always show desmosome-like complexes (Table V). The combination of active zones and desmosome-like structures with the exclusion of gap junctions also exists (ANGAUT and SOTELO 1973: cat cerebellar central nuclei — active zones and attachment plates; CORDULA N I T S C H and B A K 1974: rabbit hippocampal mossy fiber giant boutons — active zones and "desmosomes or attachment plates"; ROSEMARIE D U N N 1975: desmosome-like plaques flanking the synapses of the outer hair cells in cat cochlea). By contrast, gap junctions combined with desmosome-like structures without active zones on the same contact zone occur much more frequently in the vertebrate C.N.S. TAXI

( H I N O J O S A a n d ROBERTSON 1 9 6 7 ; H I N R I C H S E N

1968;

LARRAMENDI

1969,

YASHIKI

1968;

1969b;

KRIEBEL

HINOJOSA

et

1973;

al.

1968,

SOTELO

and 1969;

et al.

1 9 7 0 ; K A W A N A e t a l . 1 9 7 1 ; D E L CERRO a n d S N I D E R

1 9 7 4 ; CANTINO a n d M U G N A I N I 1 9 7 5 ; P A P P A S e t

et al. 1 9 7 3 ) . In the lateral geniculate nucleus of cats and monkeys G U I L L E R Y and COLONN I E R furthermore differentiated between desmosomelike and filamentous contacts (COLONNIER and G U I L -

1975; SOTELO 1975; SOTELO et al. 1976). Whether the latter type of contact is concerned or the apparently much more common mixed synapse including desmosome-like structure — the close vicinity of desmosome-like complexes and gap junctions is striking ( B E N N E T T et al. 1 9 6 7 b and d; H I N O -

1972; WENZEL

LERY

1964;

GUILLERY

1967,

1969;

GUILLERY

and

According to the authors the filamentous contacts unlike the desmosome-like complexes show fine filaments and tubular profiles either on both sides or together with granular dense bodies on the postsynaptic component. They can often be seen adjacent to active zone contacts. For this reason it has been suggested that these contacts are either stages of formation or destruction of active zones, or both active zones and filamentous contacts are distinct structural specializations with different aspects of synaptic function. Since 1 9 7 0 G U I L L E R Y and COLONNIER as well as B R A U E R and coworkers ( 1 9 7 4 ) have only made a distinction between active zones and symmetrical filamentous contacts. In other regions of the C.N.S. no further subdivisions have been made among the "non-synaptic" junctional complexes. In the case of the trout oculomotor nucleus a further differentiation among the desmosome-like complexes was not possible either. The combined occurrence of the three complexes active zone, gap junction and desmosome-like strucCOLONNIER 1 9 7 0 ) .

Hirnforschung, Bd. 19, Heft I

JOSA a n d

ROBERTSON

1967;

KRIEBEL et al.

al.

1968,

1 9 6 9 ; SOTELO a n d T A X I 1 9 7 0 ; W A X M A N a n d P A P P A S 1 9 6 9 , 1 9 7 1 ; PAPPAS a n d WAXMAN 1 9 7 2 ;

NAKAJIMA

1 9 7 4 ; SOTELO e t a l . 1 9 7 4 ; CANTINO a n d

MUGNAINI

1975;

SOTELO

1975).

This agrees with the results

presented here. In some electron micrographs such a juxtaposition also occurs between active zones and desmosome-like structures ( N A K A J I M A 1974) as mentioned in the case of the filamentous contacts (see above). In Fig. 14 of N A K A J I M A ' S paper (1974) for instance along the contact zone the following complexes occur without intervals of "normally" non-specialized membranes: active zone — punctum adhaerens — gap junction — punctum adhaerens — active zone. But all these combinations are quite in the minority compared with the combination of gap junctions and desmosome-like structures. The fact of the close vicinity of the two membrane specializations gap junction and desmosome-like structure which has been observed so 5

66

Schuster, Th.

T a b l e V. Contact areas with three different types of specialized membrane complexes (active zones, gap junctions, desmosome-like structures) in the vertebrate C.N.S. — data from the literature. animal

region

ammocoetes of the lamprey vestibular nuclei (Lampetra planeri)

mode of contact

reference

axosomatic

S T E F A N E L L I a n d CARAVITA 1970

electric fish (Gymnotus carabo)

magnocellular mesencephalis nucl.

axosomatic

SOTELO e t al. 1 9 7 5

spiny boxfish (Chilomycterus schoepfi)

oculomotor nucleus

axodendritic

PAPPAS a n d WAXMAN 1 9 7 2

Salvelinus pluvius (teleost)

oculomotor nucleus

axosomatic

YASHIKI

goldfish (Carassius aur.)

tangential nucleus

axosomatic

HINOJOSA 1 9 7 3

axosomatic, axodendritic

K R I E B E L e t al. 1 9 6 9

Spheroides mac., Chilomyct. oculomotor nucleus sch., Myoxocephalus (teleosts)

1969a

goldfish (Carassius aur.)

Mauthner cell

axosomatic, axodentritic axoaxonic

NAKAJIMA 1 9 7 4

frog

spinal cord intermediate gray matter

axodendritic

SOTELO a n d T A X I

chick chick

ciliary ganglion, ciliary neurons ciliary ganglion, ciliary neurons

CANTINO a n d MUGNAINI

rat rat

lateral vestibular nucleus descending vestibular nucleus

axosomatic axosomatic axosomatic axosomatic

SOTELO 1 9 7 5

1970

BRIGHTMAN a n d R E E S E

1969 1975

SOTELO a n d P A L A Y 1 9 7 0

T a b l e V I a. Lengths of active zones — some data from the literature. animal

region

length in [im

reference

Formica rufa, Camponodus ligniperdus (invertebrates)

synaptic glomeruli in the neuropil of the corpora pedunc.

0.1 — 0.25

STEIGER 1 9 6 7

trout

oculomotor nucleus, axosomatic giant synapses

0.35

SCHUSTER 1 9 7 3

cat

cerebellar central nuclei, mediumsized boutons on large neurons

0.15 — 0.3

ANGAUT a n d S O T E L O 1 9 7 3

T a b l e V I b. Lengths of gap junctions in nervous tissue — some data from the literature. animal

region

Carassius auratus

laterale dendrite of the Mauthner cell synaptic discs of the club endings

0.2-0.5

Gymnotus carabo

cerebellum, gap junctions between mossy fiber terminals and granular cell dendrites

0.08-0.2

Carassius auratus

Nc. vest, tang., 1st) caliciform endings 2nd) trilobated endings 3rd) club endings

chick

rat rat cat

Nc. vest, tang., gap junctions of the spoon synaptic endings ventral cochl. nc. vestibular nuclei inf. oliv. nc., dendrodendritic gap junctions

length in [im

reference ROBERTSON 1 9 6 3 , ROBERTSON

et al.

1963

SOTELO a n d LLINAS 1 9 7 2

HINOJOSA 1 9 7 3 0.14-0.33 0.2-0.36

0.08-0.22 0.04-0.47

HINOJOSA a n d ROBERTSON 1967

0.1-0.2

SOTELO e t a l . 1 9 7 6

0.05-0.4

SOTELO 1 9 7 5

0.05-0.2

SOTELO e t a l . 1 9 7 4

Experimental alteration in mixed synapses

often support the idea of a functional relation between them. This is in agreement with the quantitative results. Discussion of the quantitative results Lenght and number of special membrane complexes

The mean values of the length of the different membrane complexes measured are within the range of figures cited in the literature (Table Via, b). A problem in this connection is whether the measured length always represents a diameter of the sectioned complex. Especially large figures suggest that the frequent disc-like shapes of the active zones or gap junctions are replaced by other forms, for instance by horseshoe-like configurations in the case of active zones (ANDRES 1975) or belts in the case of gap junctions (BRIGHTMAN and R E E S E 1 9 6 9 ; FRIEND and GILULA 1 9 7 2 ; ZAMPIGHI a n d ROBERTSON 1 9 7 3 ) .

The longest synaptic structure measured was a 1.5 /. U

+ o o

co O

+ +

+

IM O

Í - +

co L> a ü CS I u

4 4 4 4

O

s X>

44 + + + + 4 + 4 + 4 4

4 o

griseum pontis

©

.tí ft co

®

4

nucí, cuneatus laterali! nucí, cuneatus mediali

e 3

«p O

4

í *

1 .2 ffí § c 'S l i g"g fe -S "S 0 > fl

g, tí u 'S3 » a •S a 5 S ^a