The Cerebellum: From Structure to Control

PROGRESS IN BRAIN RESEARCH

VOLUME 114

THE CEREBELLUM: FROM STRUCTURE TO CONTROL

EDITED BY

C.I. DE ZEEUW Department of Anatomy, Erasmw University Rotterdam, Dr. Molewatetplein 50, Rotterdam,

The Netherlands

P. STRATA Human Anatomy and Physiology, University of Turin, Corso Raffaelo 30. Torino, Italy

J. VOOGD

Department of Anatomy, Erasmus University Rotterdam, Dr. Molewateplein 50, Rotterdam, The Netherlands

ELSEVIER AMSTERDAM - LAUSANNE - NEW YORK - OXFORD - SHANNON - TOKYO

1997

0 1997 Elsevier Science B.V. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permis- sion of the publisher, Elsevier Science BV. Copyright and Permissions Department, P.O. Box 521, lo00 AM Amsterdam. The Netherlands.

No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of the rapid advances in the medical sciences, the publisher recommends that independent verification of diagnoses and drug dosages should be made.

Special regulations for readers in the U.S.A.: This publication has been registered with the Copyright Clearance Center Inc. (CCC), 222 Rosewood Drive, Danvers. MA 01923. Information can be obtained from the CCC about conditions under which the photocopying of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside of the U.S.A., should be referred to the copyright owner, Elsevier Science B.V. unless othelwise stated.

ISBN 0-444-82313-1 (volume) ISBN 0-444-80104-9 (series)

Published by: Elsevier Science B.V. P.O. Box 211 lo00 AE Amsterdam The Netherlands

Printed in The Netherlands on acid-free paper

V

List of Contributors

P. Adeleine, HBpital Neurologique and Neuro-chirurgical, Pierre Wertheimer, 59, Boulevard Pinel, B.P. Lyon-montchat 69394, Lyon cedex 03, France

R. Arikan, Department of Otolaryngology, Washington University School of Medicine, 4911 Barnes Hospital Plaza, St. Louis, MO 63110, USA

R. Apps, The Medical School, Department of Physiology, University of Bristol, University Walk, Bristol BS8 lTD, United Kingdom

J.A. Armengol, Departamento de Morfologia, Facultad de Medicina, Avda Reina Mercedes 6, Sevilla 41012, Spain

D.M. Armstrong, The Medical School, Department of Physiology, University of Bristol, Univer- sity Walk, Bristol BS8 ITD, United Kingdom, email: [email protected]

R.G. Baker, Department of Physiology and Neuroscience, New York University Medical Center, 555 1st Ave, New York NY 10016, USA, email: [email protected]

G.A. Bishop, Department of Cell Biology, Neurobiology and Anatomy, Ohio State University, 333 W 10th Ave, Columbus, OH 43221, USA, email: [email protected]

J.G. Bjaalie, University of Oslo, Department of Anatomy, Institute of Basic Medical Sciences, P.O. Box 1105, Blindern N-0317, Oslo 3, Norway 2, email: [email protected]

J.R. Bloedel, Division of Neurobiology, St. Josephs Hospital and Medical Centre, Barrow Neurological Institute, 305 W Thomas Road, Phoenix, AZ 85013, USA, email: jbloedel@m ha.chw.edu

J.M Bower, Division of Biology, California Institute of Technology, Pasadena, CA 91125, USA, email: [email protected]

V. Bracha, Division of Neurobiology, St. Josephs Hospital & Medical Centre, Barrow Neuro- logical Institute, 305 W Thomas Road, Phoenix, AZ 85013, USA, email: vbracha@m ha.chw.edu

M. Bravin, Department of Neuroscience, University of Turin, Corm Raffaelo 30, 10125 Torino, Italy

P. Brodal, University of Oslo, Department of Anatomy, Institute of Basic Medical Sciences, P.O. Box 1105, Blindern N-0317, Oslo 3, Norway 2, email: [email protected]

A. Buffo, Department of Neuroscience, University of Turin, Corso Raffaelo 30, 10125 Torino, Italy

R. Caff6, Department of Anatomy, Erasmus University of Rotterdam, Dr. Molewaterplein 50, P.O. 1738,3000 DR, Rotterdam, The Netherlands

F.A. Chaudhry, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3, Norway

G. Cheron, Laboratory of Neurosciences, University of Mons-Hainaut, Place du Parc 20 - 7000 Mons, Belgium, email: Mcheron@bmsuemll

B. Cohen, Department of Neurophysiology, Mount Sinai School of Medicine, 1 Gustave L Levy Place, New York, NY10029, USA, email: [email protected]

C. Cozzari, Institute for Cell Biology, CNR, Via C. Marx 23,00137 Rome, Italy N.C. Danbolt, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3,

R.R. De la Cruz, Lab. Neurociencia, Universidad de Sevilla, Facultad Biologia, Avda Reina Norway

Mercedes 6, Sevilla 41012, Spain

vi

J.M. Delgado-Garcia, Lab. Neurociencia, Universidad de Sevilla, Facultad Biologia, Avda

E. De Schutter, Departement Geneeskunde, Universiteit van Antwerpen, Universiteitsplein 1,

C.I. De Zeeuw, Department of Anatomy, Erasmus University Rotterdam, Dr. Molewaterplein

J. Dichgans, Neurologische Klinik, Hoppe-Seylerstrasse 3, 72076 Tubingen, Germany M.R. Diiio, Institute of Neuroscience, Northwestern University, 5-305 Ward Building, 303 East

Chicago Avenue, Chicago, IL 60611, USA J.P. Draye, Laboratory of Neurosciences, University of Mons-Hainaut, Place du Parc 20 - 7000

Mons, Belgium M.P. Dufief, Laboratory of Neurosciences, University of Mons-Hainaut, Place du Parc 20 -

7000 Mons, Belgium T.J. Ebner, Department of Neurosurgery, University of Minnesota, Lions Research Building,

2001 Sixth St. SE#421, Minneapolis, MN 55455, USA, email: [email protected] C.F. Ekerot, Department of Physiology, University of Lund, Solvegatan 19, S-223 62 Lund,

Sweden M. Fronte, Department of Neuroscience, University of Turin, Corso Raffaelo 30,10125 Torino,

Italy Q. Fu, Department of Neurosurgery, University of Minnesota, Lions Research Building, 2001

Sixth St. SE#421, Minneapolis, MN 55455, USA T. Futami, Riken Institute, 21 Hirosawa Waksohi, 35101 Saitama, Japan M. Garwicz, Department of Physiology, University of Lund, Solvegatan 19, S-223 62 Lund,

Sweden N.M. Gerrits, Department of Anatomy, Erasmus University Rotterdam, Dr. Molewaterplein 50,

P.O. 1738, 3000DR, Rotterdam, The Netherlands, email: [email protected] M. Glickstein, Department of Anat. Neuroscience and Behaviour, University College of

London, Gower Street, London WClE 6BT, United Kingdom, email: [email protected] E. Godaux, Laboratory of Neurosciences, University of Mons-Hainaut, Place du Parc 20 - 7000

Mons, Belgium A. Gruart, Lab. Neurociencia, Universidad de Sevilla, Facultad Biologia, Avda Reina Mercedes

6, Sevilla 41012, Spain, email: [email protected] R.B. Hawkes, Department of Anatomy, University of Calgary, Health Science Center, 3330

Hospital Drive, New Calgary, T2N4Ni Alberta, Canada, email: [email protected] S.M. Highstein, Department of Otolaryngology, Washington University School of Medicine,

4911 Barnes Hospital Plaza, St. Louis, MO 631 10, USA, email: [email protected] J.C. Houk, Department of Physiology, Northwestern University Medical School, M211 Ward

5315,303 E Chicago Avenue, Chicago, IL 60611, USA, email: [email protected] Y. Izawa, Department of Physiology, School of Medicine, Tokyo Medical and Dental Univer-

sity, 1-5-45, Yushima. Bunkyo-Ku, Tokyo 113, Japan D. Jaarsma, Department of Anatomy, Erasmus University Rotterdam, Dr. Molewaterplein 50,

P.O. 1738,3000 DR, Rotterdam, The Netherlands, email: [email protected] H. Jorntell, Department of Physiology, University of Lund, Solvegatan 19, S-223 62 Lund,

Sweden J.S. King, Department of Cell Biology, Neurobiology and Anatomy, Ohio State University, 333

W 10th Ave, Columbus, OH 43221, USA, email: [email protected] G.A. Kinney, Northwestern University, Department of Physiology, 303 E. Chicago Avenue,

Chicago, IL 60611, USA P.H. Kitzman, Department of Cell Biology, Neurobiology and Anatomy, Ohio State University,

333 W 10th Ave, Columbus, OH 43221, USA

Reina Mercedes 6, Sevilla 41012, Spain, email: [email protected]

2610 Antwerpen, Belgie, email: [email protected]

50, P.O. 1738, 3000DR, Rotterdam, The Netherlands, email: [email protected]

vii

T. Klockgether, Neurologische Klinik, Hoppe-Seylerstrasse 3, 72076 Tubingen, Germany S.K.E. Koekkoek, Department of Anatomy, Erasmus University Rotterdam, Dr. Molewater-

plein 50, P.O. 1738,3000 DR, Rotterdam, The Netherlands, email: [email protected] J.H. hake, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3,

Norway A.I. Levey, Department of Neurology, Emory University School of Medicine, Woodruff

Memorial Building, Suite 6000, Atlanta, GA 30322, USA R. Llinls, Department of Physiology and Neuroscience, New York University Medical Center,

555 1st Ave, New York, NY 10016, USA, email: [email protected] P.C. Madtes Jr., Department of Biology, Mt Vernon Nazarene College, 800 Martinsburg Rd,

Mt Vernon, OH 43050, USA, email: [email protected] D.E. Marple-Horvat, The Medical School, Department of Physiology, University of Bristol,

University Walk, Bristol BS8 lTD, United Kingdom F.A. Middleton, Department of Physiology, SUNY Health Science Center, 766 Irving Avenue,

Syracuse, NY 13210, USA, email: [email protected] M.S. Milak, Division of Neurobiology, St. Josephs Hospital and Medical Centre, Barrow

Neurological Institute, 305 W Thomas Road, Phoenix, AZ 85013, USA, email: [email protected]

E. Mugnaini, Institute of Neuroscience, Northwestern University, 5-305 Ward Building, 303 East Chicago Avenue, Chicago, IL 60611, USA, email: [email protected]

J.H. Nagelhus, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3, Norway

M.F. Nitschke, Department of Neurology, Medical University, Ratzeburger Alee 160, D23538, Luebeck, Germany

Z. Nusser, Medical Research Council, Anatomical and Neuropharmacology Unit, University of Oxford, Mansfield Road, Oxford OX1 3th, United Kingdom, email: zoltan.nusser@ pharm.ox.ac.uk

J.D. Oberdick, Ohio State University, Biotechnology Center, 190 Rightmire Hall, 1060 Car- mack Road, Columbus, OH 43210, USA, email: [email protected]

O.P. Ottersen, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3, Norway, email: [email protected]

T.L. Overbeck, Department of Cell Biology, Neurobiology and Anatomy / The Neuroscience Graduate Program, Ohio State University, 333 W 10th Ave, Columbus, OH 43221, USA

A. Partsalis, Department of Otolaryngology, Washington University School of Medicine, 491 1 Barnes Hospital Plaza, St. Louis, MO 63110, USA, email: [email protected]

A.M. Pastor, Lab. Neurociencia, Universidad de Sevilla, Facultad Biologia, Avda Reina Mercedes 6, Sevilla 41012, Spain, email: [email protected]

T. Raphan, Institute of Neural and Intelligent Systems, Department of Computer and Informa- tion Science, Brooklyn College of CUNY, Brooklyn, NY 11210, USA, email: [email protected]

T.J.H. Ruigrok, Department of Anatomy, Erasmus University Rotterdam, 3 r . Molewaterplein 50, P.O. 1738,3000 DR, Rotterdam, The Netherlands, email: [email protected]

D.J. Rossi, Department of Physiology, University College London, Gower Street, London WClE 6BT, United Kingdom

F. Rossi, Human Anaiomy and Physiology, University of Turin, Corso Raffaelo 30, 10125 Torino, Italy, email: [email protected]

S. Sanlioglu-Crisman, Ohio State University, Biotechnology Center, 190 Rightmire Hall, 1060 Carmack Road, Columbus, OH 43210, USA

T. Savio, Institute of Anatomy, University of Genoa, Via de Toni 14, 1-16132 Genoa, Italy Y. Shimansky, Division of Neurobiology, St. Josephs Hospital and Medical Centre, Barrow

Neurological Institute, 305 W Thomas Road, Phoenix, AZ 85013, USA, email: [email protected]

viii

Y. Shinoda, Department of Physiology, School of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-Ku, Tokyo 113, Japan, email: Yshinoda.phyl@ med.tmd.ac.jp

N.T. Slater, Northwestern University, Department of Physiology, 303 E. Chicago Avenue, Chicago, IL 60611, USA, email: [email protected]

P. Somogyi, Medical Research Council, Anatomical and Neuropharmacology Unit m, University of Oxford, Mansfield Road, Oxford OX13th, United Kingdom, email: [email protected]

J. van der Steen, Department of Physiology, Erasmus University Rotterdam, Dr. Molewater- plein 50, P.O. 1738, 3000 DR, Rotterdam, The Netherlands, email: vandenteen@ fysl.fgg.eur.nl

J. Storm-Mathisen, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3, Norway, email: [email protected]

P. Strata, Department of Neuroscience, University of Turin, Corso Raffaelo 30, 10125 Torino, Italy, email: [email protected]

P.L. Strick, Research Service 151, VA Medical Centre, 800 Irving Avenue, Syracuse, NY 13210, USA,email: [email protected]

Y. Sugiuchi, Department of Physiology, School of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-Ku, Tokyo 113, Japan

H.S. Tan, Department of Physiology, Erasmus University Rotterdam, Dr. Molewaterplein 50, P.O. 1738,3000 DR, Rotterdam, The Netherlands

F. Ternpia, Department of Neuroscience, University of Turin, Corso Raffaelo 30, 10125 Torino, Italy, email: [email protected]

R. Torp, Anatomical Institute, University of Oslo, PO Box 1105 Blindern, 0317 Oslo 3, Norway, email: [email protected]

P. Trouillas, H6pital Neurologique and Neuro-chirurgical, Pierre Wertheimer, 59, Boulevard Pinel, B.P. Lyon-montchat 69394, Lyon cedex 03, France

J. Voogd, Department of Anatomy, Erasmus University Rotterdam, Dr. Molewaterplein 50, P.O. 1738,3000 DR, Rotterdam, The Netherlands, email: [email protected]

W. Waespe, Fachartze, Zentrum Prima am Spial Neumiinster, Trichtenhausenstrasse 12, CH-8125, Zollikerberg, Switzerland

J.P. Welsh, Department of Physiology and Biophysics, New York University Medical Center, 555 1st Ave, New York, NY 10016, USA, email: [email protected]

S. Wearne, Department of Neurophysiology, Mount Sinai School of Medicine, 1 Gustave L Levy Place, New York, NY 10029, USA, email: [email protected]

K. Wessel, Department of Neurology, Medical University, Ratzeburger Allee 160, D23538, Luebeck, Germany, email: [email protected]

J. Xie, H6pital Neurologique and Neuro-chirurgical, Pierre Wertheimer, 59, Boulevard Pinel, B.P. Lyon-montchat 69394, Lyon cedex 03, France

M. Zagrebelsky, Department of Neuroscience, University of Turin, Corso Raffaelo 30, 10125 Torino, Italy

ix

Preface

The present issue of Progress in Bmin Research, The Cerebellum: From Structure to Control, carries the name and contains the proceedings of a Satellite Symposium of the Meeting of European Neuroscience held at the Erasmus University Medical School Rotterdam in 1995. During this Symposium a wide variety of topics was presented ranging from developmental, morphological and physiological, to molecular and clinical aspects of the cerebellum. The multidisciplinary character of the meeting is clearly reflected in the variety of chapters in this book. Many of the findings have been obtained with the use of new molecular tools or novel imaging techniques, or with a combination of new technologies and traditional neuroanatomical and/or electrophysiological methods. Interestingly, these new important findings have not lead to a firm consensus on the issue of cerebellar function. Instead the number of possible functions of the cerebellum has expanded over the past decade and it covers a wide spectrum including not only motor timing and motor learning, but also for example sensory data acquisition and cognition control.

At present we are not just standing at the end of the twentieth century, but also at the end of an era of cerebellar research that has been carried out by a group of distinguished scientists over the past 30 to 40 years. Their work, which started in the sixties at a time when cerebellar research was booming, has turned out to be classical and of fundamental importance. Many of the cerebellar scientists of this established generation have contributed substantially to the quality of this issue. In addition, the book is marked by chapters from the coming generations of scientists who will determine the direction of cerebellar research for the next century. As in other fields of neuroscience, this research will be dominated by molecular neurobiology and new functional imaging techniques. Taken together, we feel that the book is pluriform and unique in that it is multidisciplinary, in that it promotes different views on cerebellar function, and that it is being published on the verge of different eras dominated by different generations of cerebellar scientists. Therefore, we are confident that the wealth of new information and ideas contained in these important papers will stimulate even more intensive research in the twenty-first century leading to a greater understanding of cerebellar function(s1.

C.I. De Zeeuw

xi

Acknowledgements

This volume largely represents the material that was presented at the Cerebellar Satellite Symposium of the Meeting of European Neuroscience from August 31 to September 3, 1995. The Editors acknowledge the generous financial support of the Medical Faculty of the Erasmus University Rotterdam (EUR), Elsevier Science BV, Pharmacia, Foundation for Basic Lens Research, Gemeente Rotterdam, Gerrit Jan Mulder Stichting, International Brain Research Organization (IBRO), Royal Dutch Academy of Sciences (KNAW), Philips, Sigma Tau, SOFAM Beheer BV, and the Vereniging Trustfonds EUR. The organization of the Symposium was made possible through the excellent help of Hans Van Der Burg, Richard Hawkins, Erika Goed- knegt, and Eddie Dalm of the Department of Anatomy (EUR), Yvonne Aberson-Kap of the Department of Physiology (EUR), Mrs. M. Wenckebach and E. Van Dijk of the Hoboken Congress Organization, and Mr. W. Hamming and Mrs. M. Scholder of the Rotterdam Congress Bureau. Finally, we would like to acknowledge S.K.E. Koekkoek for his excellent editorial assistance.

C.I. De Zeeuw, P. Strata and J. Voogd

C.I. dc Zeeuw. P. Strata and J . Vooyd (Eds.) pkgnss in Bmin Rrsrurch. Vol 1 14 0 1Y97 Elsevier Science BV. All rights reserved.

CHAPTER 1

Functional cloning of candidate genes that regulate Purkinje cell-specific gene expression

Salih Sanlioglu-Crisman1i2 and John Oberdi~k'>~*~*

'Neumbiotechnology Center, zMolecular, Cellular and Developmental Biology Graduate Ptvgmm,

'Department of Cell Biology, Neumbiology and Anatomy The Ohio State University, 190 Rightmire Hall, 1060 Carmack Road, Columbus, OH 43210, USA

Introduction

The cerebellum is a useful model system for dissecting the molecular and cellular mechanisms which underlie nervous system patterning. We have previously described the expression of a neuron-specific transgene which faithfully recapit- ulates the pattern of expression of its parent gene, pcp-2(L7). Expression of both genes is not only restricted to cerebellar Purkinje cells in the brain proper, but also initially appears in a pattern of stripes during the early phases of cerebellar development (Oberdick et al., 1990; Smeyne et al., 1991). As was demonstrated for the Purkinje cell-specific marker zebrin (Gravel et al., 19871, it is likely that these stripes are related to the well-characterized functional zonation of the cerebellum (Oberdick et al., 1993). By truncation and mutation of the promoter of a fusion transgene between the mouse pcp-2(L7) gene and the bacterial lacZ gene, alterations in the transgene banding pattern were observed suggesting that similar mechanisms as those controlling segmen- tation in hsophi la apply to compartmentation in the cerebellum (Oberdick et al., 1993; Oberdick,

*Corresponding author. Tel.: + 1 614 2928714; fax: + 1 614 2925379; e-mail: oberdick.1 @osu.edu

1994). In addition, it was shown that the 250 bp most proximal region of the promoter, in combination with intact downstream coding and non- coding regions, was sufficient to maintain Purk- inje cell-specific expression of the transgene (Ob- erdick et al., 1993). Interestingly, within this short promoter region a discrete element was identified by DNAse I footprint analysis that was functio- nally assessed by mutagenesis and analysis of expression in transgenic mice (Oberdick et al., 1993). This element was particularly interesting since over most of its extent it appeared to be related to known homeobox protein binding sites; on one end, however, it was flanked by a perfect consensus E-box of the class A (Ohsako et al., 1994) bHLH protein binding type.

The bHLH family of transcription factors are key developmental modulators controlling cell fate in both vertebrate and invertebrate systems. The proneural genes achaete-scute (AS-C) and daughterless (da), for example, are key regulators of neuronal cell fate in hsophi la (Vaessin et al., 1990; Campos-Ortega, 1993). Likewise, the role of bHLH proteins in vertebrate neurogenesis has been partially elucidated by gene knock out experiments in mice. The mammalian achaete-scute homolog, MASH-1, is expressed in neural crest derivatives (Johnson et al., 1990). Null mutation of MASH-1 impaired proper development of neu-

4

ral crest cell lineages and gave rise to autonomic nervous system defects. This is manifested by the elimination of sympathetic and parasympathetic neurons, and enteric neurons of the foregut (Lo et al., 1994). Persistent expression of HES-1, a mammalian homolog of the Drosophih enhancer of split gene (Preiss et al., 1988), disrupted neuronal and glial differentiation (Ishibashi et al., 1994). Furthermore, targeted disruption of HES-1 led to premature neurogenesis and neural defects in mice (Ishibashi et al., 1995). While some bHLH family members (like E l2 and E47) are expressed ubiquitously, many of the bHLH proteins discussed above show tissue restricted expression patterns. As an extreme example of this, HES-3, another member of the bHLH family, is exclu- sively expressed in cerebellar Purkinje cells (Sasai et al., 1992; Sakagami et al., 1994). This factor was reported to be a negative regulator of transcription (Sasai et al., 1992).

Here we report the isolation of members of diverse transcription factor families by screening a cDNA expression library with an oligonucleotide corresponding to the above-mentioned homeobox + HLH protein-interacting element (L7 promoter AT-rich region + E-box, or L7- ATE). Most notably, a clone wasidentified with identity to ITF2 (Henthorn et al., 1990), or ME2 (Soosaar et al., 19941, encoding a class A bHLH protein which is the functional homolog of human SEE- lb (Corneliussen et al., 1991). mRNA for ME2 is highly enriched in the brain relative to other tissues, is detectable in the cerebellum throughout development, appears to be expressed transiently in cerebellar Purkinje cells, but expression switches predominantly to granule cells postnatally. We show that ME2 can bind specifically to the L7-ATE and that it can activate transcription from tandem multimers of this element attached to an exogenous promoter. DNA binding of ME2 is not inhibited by interaction with HES-3. In addition, ME1, another class A bHLH protein abundantly expressed in developing nervous tissue, can synergistically activate transcription in conjunction with ME2 in culture. MEl, in contrast, is inactive by itself. These data

suggest that ME2 may be involved in the initia- tion of pcp-2(L7) gene expression as opposed to its maintenance and that ME2 may belong to a developmental cascade of genes that control the generation and fate of cerebellar neurons.

Materials and methods

P20 cerebellum cDNA expression library screening

Generation of DNA probes Two complementary 51-mer oligonucleotides

corresponding to an E-box + AT rich sequence of the L7 promoter were synthesized of the fol- lowing sequence:

TCGGCACCTGTAATTGACAAGATTAAT T C A T T T A T A G G G C A T C G C G T G GACA'ITAACI'G'ITCTAATTAAGTAAAT ATCCCGTAGA'ITAATCGTTCG

Phosphorylated oligomers were first annealed to each other, then end filled in the presence of a3*P-dCTP and a3*P-dATP using Klenow fragment. Subsequently these annealed oligomers were ligated to each other to form concatemers.

Screening procedure A P20 mouse cerebellar expression library was

constructed using AEXlox vectors (Novagen) and screened as described previously (Smeyne at al., 1995).

Northern blotting

Cerebellar total RNA was prepared using Trizol Reagent according to manufacturer's instructions (Gibco BRL). RNA was electrophoresed, blotted and probed as reported previously (Maniatis et al., 1982; Oberdick et al., 1988).

Construction of ME2-PET14b qression vector and purification of ME2 protein

761 bp of ME2 cDNA carrying complete bHLH sequence was cloned into the BamHI site of PET 14b prokaryotic expression vector (Novagen).

5

ME2-PET 14b clone was transformed into BL21(DE3)plysS host strain. Transformed culture (100 ml) was grown to an OD6O,, of 0.5. Protein expression was induced by addition of 0.4 mM of IPTG and continued for 3 more hours. Cells were resuspended in binding buffer (5 mM imidazole, 500 mM NaCl, 20 mM Tris-HC1 pH 7.9) and then exposed to sonication using a microtip on ice. Soluble fractions were then recovered after cen- trifugation and loaded onto 50 mM NiSO, charged resin columns. After washing columns with the binding buffer and a wash buffer (60 mM imidazole, 500 mM NaCl, 20 mM Tris-HC1, pH 7.91, ME2 protein was eluted with an elution buffer containing 1 M imidazole, 500 mM NaCl, 20 mM Tris-HC1 (pH 7.9). The sample was dialyzed in PBS and concentrated using Centricon-lO con- centrator (Amicon).

Mobility ship assay

The probe preparation was the same as the one which was described as in cerebellar expression library screening. However, the probe was a monomer not catenated. Purified ME2 protein was incubated in binding buffer which consisted of 10 mM Tris (pH 8.01, 1 mM EDTA, 1 mM D'IT, 5 mM MgCl,, 50 mM NaCl, 5% glycerol, 5% sucrose, 1 p g poly(dIdC), 30 ng aprotinin, 20 ng leupeptin and 0.1 mM PMSF for 15 min at 37C. Then 4.7 nM probe (10 000 cpm) was added into each reaction and samples were incubated at room temperature for another 15 min. Samples were run in 5% native polyacrylamide gels for 4-5 h at 4C. The gel was fixed with 10% acetic acid. After drying the gel was exposed to X-ray film, 6-8 h at -80C. For the competition assays double stranded, end-filled CRE (Som)-2 DNA (kindly provided by Dr. Tsonwin Hai) and PKC-I DNA were used as non-specific competitors.

DNAse I footprinting analysis

Fragment prepamtion A 147 bp promoter fragment of L7 carrying

AT-rich sequence immediately flanked by an E-

box was PCR amplified and cloned into pGem7ZfT + ) (Promega) as BamHI fragment. Then the insert was digested with Hind111 and EcaRV (a blunt end cutter). Agarose purified insert DNA was end-labeled with Klenow in the presence of a-[ 32 Pl-dC". Furthermore, asym- metric end-labeled probe was eluted from Nuc- trap Push Columns (Stratagene). 2.2 nM of the probe (20 000 cpm) was used per reaction.

Footprinting reaction Purified ME2 protein was added to a 25-pl

reaction containing 50 mM Tris-C1 (pH 7.91, 12.5 mM MgCl,, 5 mM D'IT, 100 mM KCl, 0.1% "40 and 20% glycerol. The reaction mixtures were incubated at 37C for 15 min, later placed on ice. Binding reaction was started by addition of 25 pl of the DNA mix (20000 cpm of asym- metric end-labeled probe, 1 pg of poly(d1-dC), 4% polyvinyl alcohol) and the reaction mixtures were held on ice for another 15 min. DNAse I buffer (50 pl) (10 mM MgCl, 5 mM CaC12) was added into the binding reaction and kept at room temperature for 1 min. DNAse I digestion was initiated by addition of 2 p1 of freshly diluted DNAse I (1/500 dilution of 2.062 units/pl; Worthington) and ended by addition of 90 p1 DNAse I stop solution (20 mM EDTA, 1% SDS, 0.2 M NaCl, 250 pg/ml total yeast RNA). The protein was removed by phenol/chloroform extraction. Ethanol precipitated DNA was boiled for 3 min in formamide loading dye and electrophoresed in 6% polyacrylamide-8 M urea gel. Dried gel was autoradiographed with Kodak X- OMAT Film.

In situ hybridization of ME2

Synthesis of antisense riboprobes Template DNAs were linearized by restriction

enzyme digestion. 500 ng of linearized template DNA was added into the in vitro transcription reaction mixture containing 1 x transcription buffer (Promega), 10 mM DTI', 40 U of rRNasin (Promega), 250 fM of each rATP, rGTP, rCTP, 6 p M cold UTP, 16 pCi of [35S]UTP (Amersham),

6

40 U of RNA polymerases (I7 or SP6). In vitro transcription reaction was carried out in 40 p1 of total reaction volume and incubated at 37C for 1 h. Template DNA was removed with RQ1 RNase free DNAse (Promega). After phenol/chloroform extraction the newly synthesized RNA was precipitated with ammonium acetate. Specific activity of the probe was monitored via TCA precipitation.

In situ hybriduations on the tissue sections

previously (Bian et al., 1996). In situ hybridization was performed as reported

Cell culture and DNA tmnsfections

NIH/3T3 cells were obtained from the American type culture collection (ATCC). NIH/3T3 cells were grown in Dulbeccos modified eagle medium with high glucose supplemented with 10% calf serum and antibiotics. NIH/3T3 cells were plated on 60-mm tissue culture dishes and transfected at 50% confluency by calcium phosphate coprecipi- tation technique. Each transfection contained 1 pg of pGL3 promoter vector with or without (as a control) a dimer of L7-ATE sequence in conjunction with ME2 coding expression vectors. The total amount of DNA per transfection was ad- justed to 9 pg appropriately. DNA used in transfection experiments was purified using Qiagen columns. Cells were haxvested 48 h after the transfection. Cells were lysed in 1 X reporter lysis buffer and assayed for luciferase activity according to Promegas luciferase reporter gene assay protocol.

Results

Isolation of a bHLH protein by screening of a P20 cerebellum cDNA expression library

A 51-bp promoter fragment (L7-ATE) of the L7 gene carrying an AT-rich sequence and an E box (Oberdick et al., 1993) was used to screen a postnatal day 20 (P20) cerebellar cDNA expression library. In this manner we targeted transcription factors which can directly interact with this

putative enhancer. Both POU-domain- and E- box-containing clones were isolated as a result of screening 1 X 107 phage plaques (Table 1). One clone in particular, which carried a 761 bp DNA insert, was revealed by DNA sequencing to en- code a complete bHLH domain. Sequence ho- mology analysis indicated that this clone was highly homologous to human SEF2-1B (Corneliussen et al., 19911, also known as ITF-2 or E2-2 (Henthorn et al., 1990). SEF2 clones had been isolated in similar fashion by screening a human thymocyte expression library using an oligonucleotide probe corresponding to the SL3-3 E-box-containing enhancer (Corneliussen et al., 1991). The mouse homolog of this gene, named ME2, has been shown to be highly expressed in the nervous system (Soosaar et al., 1994). The gene designation ME2 is used throughout the remainder of this work.

Determination of the spatial and temporal distribution of ME2 RNA by Northern blotting

Northern blotting using the complete bHLH domain as probe identified a single 6 kb mRNA. It also revealed that among the tissues we tested, ME2 was predominantly expressed in tissues of the nervous system as shown in Fig. 1B. All brain areas tested showed ME2 expression. Among the non-brain tissues that were tested only heart, lung and ovaries were positive for ME2. Within the cerebellum, the level of ME2 expression is highest in neonates, but decreases to a slightly lower steady-state level which is maintained into adulthood (Fig. 1A). We also looked at the level of expression in the cerebellum of another homologous bHLH protein-encoding mRNA known as ME1 (Neumann et al., 1993). This alternatively spliced mRNA follows roughly the same time course of expression as ME2 in the cerebellum (Fig. 1A). ME1 and ME2 were previously described to have different binding preferences and thus each may control the expression of unique sets of downstream genes (Chiaramello et al., 1995).

7

Fig. 1. Spatial and temporal distribution of ME2 mRNA. (A) Northern blot analysis of the developmental expression pattern of ME2 (top panel) and ME1 (middle panel) in the cerebellum. 10 pg of PO (lane l), P8 (lane 2). PI4 (lane 3). P27 (lane 4). P60 (lane 5) and 1 year old (lane 6) mouse cerebellar RNA were analysed. The bottom panel is the EtBr stained control revealing the 28s and 18s ribosomal RNAs in each sample. (B) Northern blot analysis of the tissue distribution of ME2 mRNA. Samples include 10 pg of total RNA from PI cerebellum (lane 1). adult cerebellum (lane 2), whole adult brain (lane 3), cerebral cortex (lane 4). midbrain (lane 5). brain stem (lane 6) , spinal cord (lane 7). olfactory bulb (lane 8), retina (lane 9). rat pituitary (lane 10). skeletal muscle (lane 11). heart (lane 12). liver (lane 13). kidney (lane 141, thymus (lane 15). lung (lane 16). ovaries (lane 17) and testis (lane 18). The bottom panel is the EtBr stained control. (C) Localization of ME2 in situ using "S-labeled antisense riboprobe. The section is from an adult mouse brain cut in the sagittal plane; 12 p m thickness.

In situ hybriduation analysis of ME2 expression

ME2 is rather ubiquitously expressed in all regions of the brain at all ages examined. However, it shows several areas of relative enrichment as reported previously (Soosaar et al., 1994 and Fig. 1 ). Namely, the hippocampus, cerebral cortex, olfactoly bulb and cerebellum show high levels of ME2 mRNA (Fig. 1C). If ME2 is involved in the positive regulation of the L7 gene, it must be expressed in cerebellar Purkinje cells. Thus, the grain distribution in the cerebellum at various time points was examined by in situ hybridization with [35Sl-labeled riboprobe. The superficial re-

gion from El5 to PO (Fig. 2A, arrows), and the EGL at P10 (see Fig. 5 below), show the heaviest level of hybridization. Deep to the superficial region at PO is a narrow layer of relatively low grain density, and deeper still grain densities appear higher yet again (Fig. 2C, arrows and Fig. 3). The intermediate ME2 low grain density layer overlaps with the region containing the highest density of Purkinje cells as revealed by L7 and CaBP hybridization signals in immediately serial sections, as does the deep layer of ME2 expression. CaBP probe reveals a rather broad gradient of hybridization which appears highest where ME2 is at its minimum (immediately deep to the EGL)

8

but clearly remains high in the deeper region of ME2 expression (Figs. 2 and 3). The majority of cells in the deep layer at PO appear positive for ME2 and the same is true for CaBP. Thus, it is likely that many Purkinje cells express ME2 at birth.

In contrast, at El5 there is very little evidence of overlap between the high ME2 and high CaBP-positive regions (Figs. 2 and 4); likewise, in P10 and adult animals, it is difficult to tell whether

there are more than background levels of ME2 mRNA in Purkinje cells since the bulk of ME2 expression is concentrated in granule cells (Fig. 5). Together with the quantitative analysis of ME2 mRNA levels in postnatal cerebellum which shows peak levels at PO (Fig. 11, these data suggest that there is a brief window of ME2 upregulation in Purkinje cells at birth. Before and after this time, ME2 is expressed in these cells at relatively low or basal levels.

Fig. 2. Comparison of ME2 and CaBP mRNA expression patterns during early cerebellar development. In situ hybridiLation was performed on serial coronal sections using "S-labeled riboprobes for ME2 o r CaBP. (A) ME2 expression at E15. Arrows point to the superlicial layer of highest grain density, at the medio-latcral position that was used to produce the high-mag view i n Fig. 4. (B) CaBP expression at E1.5. Grains are conccritrated in two broad patches that lie deep to tlic superlicial laycr in which ME:! is concentratcd. For the most part, CaBf' + and ME2 + populations are non-overlapping. Section is serial to that in (A). (C) ME2 expression at PO. Arrows point to the deep layer of high ME2 grain density. (D) CaBP expression at PO. There is extensive overl;ip between the ME2 and CaUP patterns.

9

Fig. 3. Comparison of ME2 and CaBP probe grain densities in PO cerebellum. (A) The grain density with ME2 probe is low in the inferior colliculus (ic, top of panel). I t is high in the egl. low again immediately deep to the egl, then high again in what will become the inner granule cell layer, or igl. The egl and deep regions have roughly equivalent grain densities. (B) There are no detectable CaBP probe grains in the ic 01- egl. Immediately deep t o the cgl the grain density is quite high and is maintained deep into the presumptive igl. Both panels are photo montage with a 100 x oil objective, produced from the dorsal surface of the cerebellum at roughly the position shown by arrows in Fig. 2C.

10

Fig. 4. Comparison of ME2 and CaBP probe grain densities in El5 cerebellum. (A) ME2 grain density is high in the superficial layer (sl, or presumptive egl) but low in the deeper regions. (B) CaBP grain density is low in the choroid plexus (cp) and in the sl. but is high deep to the sl. Both panels are photo montage with a 100 X oil objective, produced from the ventral surface of the cerebellum at roughly the position shown by arrows in Fig. 2A.

At P10, ME2 is highest in the EGL (Fig. 5A, arrow) where grains appear to be distributed evenly throughout the matrix (proliferative) and mantle (postmitotic) layers. Another marker, CBFA (see Table 11, that was isolated in the same expression library screen was previously shown to be preferentially expressed in the matrix layer (Smeyne et al., 1995). Thus, ME2 function cannot be restricted to cell proliferation, but may certainly be related to neuronal maturation and/or

plasticity. Interestingly, at P10 some Purkinje cells are clearly positive for ME2 while others show background grain densities (Fig. 5B). By adulthood, when the EGL is no longer present, ME2 expression is highest in the granule cell layer (Fig. 5C, asterisk) with only basal levels in Purkinje cells; as at P10, there is a general sense that some individual Purkinje cells express more ME2 than others (not shown). It also appears to be expressed in the molecular layer (Fig. 5C, arrow)

11

Fig. 5. Expression of ME2 mRNA in P10 and adult cerebellum. (A) Sagittal view of a P10 cerebellum section probed with "S-labeled ME2 riboprobe. Arrow points to the egl which is the zone of highest grain density. Grains are also quite high in the igl. 2.5 x obj. (B) High mag view of a Purkinje cell (arrows) showing expression of ME2. The molecular layer is to the right and top, the igl to the left and bottom. Note that all Purkinje cells are not clearly positive for ME2. 100 X oil obj., cresyl violet counterstain. (C) Sagittal view of an adult cerebellum section probed with ME2 riboprobe. The grain density is highest in the igl (asterisk), but is also quite high in cells within the molecular layer (arrow). and in cells of the deep nuclear region. 2.5 X obj. (D) Higher mag view of image from (C) showing ME2-positive cells in the deep nuclear region (arrows). 10 X obj.

and in deep nuclear neurons (Fig. 5D, arrows), but is low to undetectable in the white matter.

DNA binding analysis of ME2 The sequence "specificity of ME2 was de-

termined by a competitive mobility shift assay using purified ME2 protein. Briefly, an oligonucleotide corresponding to the AT-rich plus E-box motif (L7-ATE) in the L7 promoter was used as probe. When mixed with potential binding pro-

teins, the mobility of the oligonucleotide through an acrylamide gel is retarded if it is recognized and bound by any proteins. As seen in Fig. 6, the L7-ATE oligonucleotide is shifted (arrow), and only L7-ATE-specific competitor competed for binding to the probe whereas CRE and PKC-I non-specific competitors did not compete. This means that there is a sequence within the probe specifically recognized by ME2 protein.

Sasai et al. (1992) have reported that rat HES-3,

12

TABLE 1

Results of the 17 affinity screening assay

0 OCT-1 B POU (isolated 22 times) 0 ME& m-ITF2 bHLH 0 CBF A(CArG Binding Factor A)

(isolated 3 times) 0 m-PO-GA (Lu et al., 1993) 0 clone 17 (unknown) 0 Mitochondria1 Single Stranded

0 (Tiranti et al., 1993) 0 Tubulin

(Suzuki et al., 1993) (Soosaar et al., 1994) (Kamad et al., 1992)

DNA Binding P.

a Purkinje cell-specific bHLH protein, repressed E47 mediated transcription activation. Since they failed to express HES-3 in E. coli, these investiga- tors could not prove the possible interaction between HES-3 and E47 in vitro as they did for HES-1 and E47. Therefore, the effect of HES-3 on ME2 binding to the L7-ATE was examined.

mHES-3 was cloned by RT-PCR, expressed in E. coli, and purified as described in methods. HES-3 did not recognize L7-ATE by itself (Fig. 6B, lanes 8-12) nor did it block the binding of ME2 to the E-box On the contrary, the binding affinity of ME2 to the E-box seemed to be enhanced in the presence of HES-3 (Fig. 6B, lanes 3-71. This is not simply a carrier effect, since enhancement was not evident when a non-specific protein such as BSA was added to the binding reaction (Fig. 6B, lanes 13-16). The enhancement was also observed in the DNAse I footprinting analysis (see below). We have not directly determined, however, if ME2 and HES-3 can form heterodimers.

DNAse I footprinting assay with ME2 As was shown previously, the entire L7-ATE

region (AT-rich plus E-box) could be protected by nuclear protein extracts prepared from whole cerebellum (Oberdick et al., 1993). A purified POU-domain protein (Pit-11, however, only protected the AT-rich region. This has been repeated here with Pit-1 alone (Fig. 7, lanes 3-6) or in

combination with ME2 (Fig. 7, lanes 7-10). The latter pattern appears to be simply the additive of individual Pit-1 (lanes 3-6) and ME2 (lanes 11-14) patterns. Oct-lB, however, which was cloned during the current expression library screen (see Table 11, protects a region which extends into the adjoining E-box (Fig. 7, lanes 15-18). Thus, the protection of the full AT-rich plus E-box site by cerebellar extracts as reported previously (Oberdick et al., 1993) could be due to POU/homeodomain protein binding alone or in combination with bHLH proteins. In addition, binding of either one of HLH or POU proteins could in principle affect binding of the other by steric constraints associated with the close prox- imity of the AT-rich and E-box subsites. This in fact may explain the slight decrease in protection over the entire L7-ATE when Pit-1 and ME2 proteins are footprinted together relative to the extent of protection observed in the individual footprints. The same effect is observed when Oct- 1B and ME2 are combined (see Fig. 7).

In support of this, ME2 by itself protected a region of sequence GCACCTGTAATY which includes a canonical A type E-box (CANNTG) and one of the three TAAT repeat motifs within the adjacent AT-rich region of the L7-ATE (Fig. 7, lanes 11-14). These repeat motifs are typical core recognition sequences of several known homeodomain proteins (Desplan et al., 1988; Ru- vkun and Finney, 19911, which would explain the overlap into the E-box of Oct-1B binding described above. Thus, like Oct-lB, ME2 binding to its recognition site may sterically hinder binding of other recognition factors. This suggests that the transcriptional activity mediated by the L7- ATE could be dependent upon the competing influences and levels of bHLH and homeodamain proteins. The sequences involved in binding and the extent of overlap between recognition sequences of the diverse proteins tested are sum- marized in schematic form in Fig. 8.

mHES-3, in contrast, did not protect any region on this promoter fragment (Fig. 7, lanes 27-30). Failure to efficiently recognize any type of E-box

13

Fig. 6. Mobility shift assays. (A) Competitive mobility shift assay using the ME2 protein. L7-ATE probe was incubated with 50 ng of the ME2 protein with or without increasing concentrations of L7-ATE oligo (lanes 3-5). CRE oligo (lanes 6-8) and PKC-I oligo (lanes 9-11) cold competitors. Lanes 3.6.9 contain a 10 fold excess of cold competitor (47 nM). Lanes 4.7.10 contain a 50 fold excess of cold competitor (235 nM). Lanes 5, 8, 11 contain 100 fold excess of cold competitor (470 nM). (B) Mobility shift assay using ME2 and mHES3 proteins. L7-ATE probe was incubated with 10 ng (lanes 3.8). 20 ng (lanes 4,9), 40 ng (lanes 5, 10). 100 ng (lanes 6, 11) or 200 ng (lanes 7, 12) of mHES3 with (lanes 3-7) or without (lanes 8-12) 50 ng of ME2 protein. As a control, the probe was incubated with 50 ng (lane 14) and 100 ng (lane 15) of BSA and 50 ng of ME2 protein. The mobility shift using BSA is a different gel with a different exposure time.

14

Fig. 7. DNAse I footprinting assay. The labeled 147 bp promoter fragment was incubated with various amounts of purified transcription factors. 0.1, 0.2,0.5 or 1 pg of Pit-1 (lanes 3-6). 0.05, 0.1.0.25, or 0.5 pg of ME2 (lanes 11-14), 0.1.0.2,0.5 or 1 p g of Octl-B (lanes 15-18), 0.09, 0.18, 0.27 or 0.06 pg of mHES3 (lanes 27-30) were used in the assay. In addition, increasing concentrations of ME2 were incubated with increasing concentrations of Pit-I (lanes 7-10), Octl-B (lanes 19-22) or mHES3 (lanes 23-26). The concentration range of each protein in the latter combination lanes is the same as the concentration range when each was tested singly.

(even an N box) by HES-3 can be attributed to association with ME2, however, certainly did not the large deletion within its basic domain (Sasai abrogate ME2s ability to protect the E-box (Fig. et al., 1992). Addition of HES-3 to the reaction in 7, lanes 23-26). On the contrary, protection

15

TCGGCACCTGTAATTGACAAGATTAATTCATTTATAGGGCATCTAATTAGCMGC

ME2 PROTECTED REGION

W 1 - B PROTETED REGION

0 PIT 1 PROTECTED REGION Fig. 8. Schematic representation of the data from the DNAse I footprinting assay. The ME2 protected region is shown as a black bar right underneath the sequence. The textured bar right above the sequence represents Octl-B protected region whereas Pit-1 protected region is shown as a white bar. Weakly protected regions in the assay are portrayed as thin bars. The sequence residues shown in bold type are the E-box (bHLH protein binding) and the TAAT repeats (putative POU/homeodomain binding).

seemed to be enhanced (compare lanes 23-26 to lanes 11-14) as was shown by the band shift experiments (Fig. 6). Interestingly, a similar enhancement was observed between HES-1 and MyoD (Sasai et al., 1992). This suggests that loss of DNA binding of heterodimers is not the only possible mechanism whereby HES-3 negatively regulates transcription. Another possibility, for example, is that HES-3 may act as a negative regulator by forming non-functional heterodimers which retain DNA binding capability; the low activity could result from the formation of a relatively poor transcriptional activation domain.

Activation by ME2 of a reporter gene linked to the L 7ATE

The preceding experiments have established that ME2 can in fact recognize and bind to the E-box contained within the L7-ATE. It is critical, however, to show that this factor can activate gene expression. A dimer of the L7-ATE was linked to a standard luciferase reporter vector, which was then co-transfected into NIH/3T3 cells along with a full-length ME2 expression plasmid (the latter courtesy of Dr. Toomas Neumann). The data from these assays are presented in Figs. 9 and 10. Each value in Fig. 9A represents the

mean + SEM of three independent trials each run in duplicate ( n = 6); those in Fig. 1OA represent the mean + SEM of two independent trials (n = 4).

The data in Fig. 9 illustrate the cooperative association of ME2 with the L7-ATE and the resultant activation of the exogenous promoter linked to the reporter. That is, doubling of the ME2 concentration in these cells results in a 5-fold enhancement of the activity over what would be expected if the effect were simply additive (Fig. 9B). Luciferase expression is only marginally increased when ME2 is transfected in conjunction with the basal promoter with no L7 enhancer (pGL3pro). This illustrates the transfer- able nature of gene activation through the L7- ATE enhancer.

Fig. 10 illustrates that ME2 can cooperatively interact with another bHLH protein of the A type, ME1. That is, the activity of ME1 and ME2 together is 10-fold greater than the additive of each alone. In fact, ME1 has little or no activity by itself in this assay.

Although ME1, like ME2, is clearly expressed at high levels in the cerebellum (Fig. 11, we have not determined whether it can be detected in Purkinje cells. Nevertheless, these expression data illustrate the role that combinatorial action of

16

L7-ATE 4ME2 8ME2 pGL3pro 4ME2 8ME2

CONSTRUCTS

50

40

20

10

0 L7-ATE 4ME2 8ME2 pGL3pro 4ME2 8ME2

CONSTRUCTS Fig. 9. Luciferase assay of ME2. (A) ME2 can activate L7-ATE enhancers alone. NIH/3T3 cells were transiently cotransfected with 1 pg of L7-ATE-luciferase vector (a pGL3pro vector carrying a dirner of L7-ATE enhancers with tail to tail orientation) alone or with 4 pg (4ME2) and 8 pg (8ME2) of ME2 coding vectors. As a control transfections were performed using 1 pg of pGL3prc vector alone or with either 4 pg or 8 pg of ME2 coding vectors. Data is presented as activity in relative light units (RLU) per pg of total protein vs the construct name and amount. Assays were repeated three times in duplicate ( n = 6) . (B) Graphic representation of the fold activation of L7-ATE enhancers by ME2. The fold activation was plotted against the constructs as explained in (A).

17

A 3 x 1 0 - 1 1 1 1 ! I I ~ ~ ! ~ 1 ~ 1 ! ~ ~ 1 ~ 4

4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................ I .................... ..................................... f" ..........._... r.""" - 2.5~10 ~

L7-ATE 4 ME1 4 ME2 4 ME1+4 ME2

CONSTRUCTS Fig. 10. Synergistic activation of the L7-ATE enhancer by ME2 and MEl. (A) 1 pg of L7-ATE-luciferase vector was transfected into N I H / 3 n cells alone or cotransfected with either 4 pg of ME1 (4ME1) or 4 pg of ME2 (4ME2) coding vectors. In addition, 4 pg of ME1 and 4 pg of ME2 together (4ME1 + 4ME2) were cotransfected with 1 pg of L7-ATE-luciferase vector into NIH/3T3 cells (n = 4). (B) The synergistic interaction of ME2 and ME1 on the L7-ATE enhancer is plotted as fold activation.

18

transcription factors may play in cell type-specific gene regulation.

Discussion

In summav, these data describe the isolation of transcription factors which putatively regulate cerebellar genesis. Although this latter process, and the cell-specific regulation of genes like pcp- 2(L7), are clearly complex, involving the concerted action of both ubiquitously expressed and restricted control factors, some generalizations can be made. For example, well-defined developmental stages are delineated by the up- and down-regulation of these proteins. Like the En-2 gene whose pattern of early cerebellar expression is highly reminiscent of patterns of expression of several Purkinje cell markers (Millen et al., 1993, ME2 is transiently expressed in immature cerebellar Purkinje cells; both genes, however, ulti- mately have restricted expression in granule cells during postnatal development. In contrast, several Hox genes, although most noted for their control of early embryonic patterning in caudal hindbrain neurons and spinal cordTassociated somitic tissues, are clearly activated in the brain and Purk- inje cells during the postnatal period and continu- ing into adulthood (Sanlioglu-Crisman et al. (1995) ASCB Abstract; to be presented elsewhere in published form). The same late activation is true as well for the restricted expression of HES-3, a bHLH protein which is not activated in Purkinje cells until around postnatal day 6 (Sasai et al., 1992).

Another suggestion from these data is that transcription factors such as ME1 and ME2, which have relatively restricted patterns of expression compared to ubiquitous bHLH proteins such as E l2 and E47, may contribute in a combinatorial fashion to the cell type-specific expression of genes like L7. This is evidenced by the synergistic activation of gene expression by ME1 and ME2 in the cotransfection experiments. Although these experiments cannot prove that L7 is a target gene of ME1 or ME2, that is a distinct possibility that one can test.

In this regard, it is clear that normal develop ment of Purkinje cells requires that genes such as En-2 be appropriately down-regulated, since overexpression of the En-2 protein in late embryonic and early postnatal Purkinje cells causes a dra- matic decrease in overall cerebellar size and in the number of Purkinje cells (Sanlioglu-Crisman and Oberdick (1996) ASN Abstracts (Sanlioglu- Crisman and Oberdick, 1996); to be presented elsewhere in published form). It remains to be seen whether either overexpression or cell-selective elimination of similarly staged transcription factors, such as ME2, can have related effects on Purkinje cell development, or in any way interfere with L7 gene expression.

Acknowledgements

The authors wish to thank Dr. Toomas Neuman for providing the full-length ME1 and ME2 expression vectors. This work was supported by NSF Grant IBN-9309611.

References

Bian F., Chu T., Schilling K. and Oberdick J. (1996) Differen- tial mRNA transport and the regulation of protein synthesis: Selective sensitivity of Purkinje cell dentritic mRNAs to translational inhibition. Mol. Cel. Neuroscience, 7: 116- 133.

Campos-Ortega J.A. (1993) Mechanism of early neurogenesis in Drosophila melanogaster. J. Neurobiology, 24: 1305-1 327.

Chiramello A., Soosaar A., Neuman T. and Zuber M.X. (1995) Differential expression and distinct DNA-binding specificity of MEla and ME2 suggest a unique role during differentiation and neuronal plasticity. Mol. Brain Research, 29:

Corneliussen B.. Thornell A., Hallberg B. and Grundstrom T. (1991) Helix-loop-helix transcriptional activators bind to a sequence in glucocorticoid response elements of Retrovirus enhancers. J. Virology, 65: 6084-6093.

Desplan C., Theis J. and OFarrell P.H. (1988) The sequence specificity of homeodomain-DNA interaction. Cell, 54: 1081 - 1090.

Gravel C., Eisenman L.M., Sasseville R. and Hawkes R. (1987) Parasagittal organization of the rat cerebellar cortex: direct correlation between antigenic Purkinje cell bands revealed by mab Q113 and the organization of the olivocerebellar projection. J. Comp. Neurol. 265: 294-310.

Henthorn P., Kiledjian M. and Kadesch T. (1990) Two distinct

107- 1 18.

19

transcription factors that bind the immunoglobulin enhancer E ~ / K E2 motif. Science, 247: 467-470.

Hawkes R. and Leclerc N. (1989) Purkinje cell axon collateral distributions reflect the chemical compartmentation of the rat cerebellar cortex. Brain Reseach, 476: 279-290.

Ishibashi M., Moriyoshi K., Sasai Y., Shiota K., Nakanishi S. and Kageyama R. (1994) Persistent expression of helix- loop-helix factor HES-1 prevents mammalian neural differentiation in the central nervous system. EMBO J., 13:

Ishibashi M., Ang S.L, Shiota K., Nakanishi S., Kageyama R. and Guillemot F. (1995) Targeted disruption of mammalian hairy and Enhancer of Split homolog-1 (Hes-1) leads to upregulation of neural helix-loop-helix factors, premature neurogenesis, and severe neural tube defects. Genes and Development, 9: 3136-3148.

Johnson J.E., Birren S.J. and Anderson D.J. (1990) Two rat homologues of Drosophila achaete-scute specifically expressed in neuronal precursors. Nature, 346: 858-861.

Kamada S. and Miwa T. (1992) A protein binding to CArG box motifs and to single-stranded DNA functions as a transcriptional repressor. Gene, 119: 229-236.

b L., Guillemot F., Joyner A.L. and Anderson D.J. (1994) Mash-1: a marker and a mutation for the mammalian neural crest development. Persp. Deu. Neurobiology, 2: 191-201.

Lw Y.. Zeft A.S. and Riegel A.T. (1993) Cloning and expression of a novel human DNA binding protein, PO-GA. Biochim. Biophy. Research Corn., 193: 779-786.

Millen K.J., Hui C-C. and Joyner A.L. (1995) A role for En-2 and other murine homologues of Drosophila segment polarity genes in regulating positional information in the developing cerebellum. Deuelopmenf, 121: 3935-3945.

Neuman T., Keen A., Knapik E., Shain D., Ross M., Nornes H. 0. and Zuber M.X. (1993) ME1 and GE1: basic helix- loop-helix transcription factors expressed at high levels in the developing nervous system and in morphogenetically active regions. European J. Neuroscience, 5: 311-318.

Oberdick J., Smeyne R.J., Mann J.R., Zackson S., Morgan J.1. (1990) A promoter that drives transgene expression in cerebellar Purkinje and retinal bipolar neurons. Science, 248: 223-226.

Oberdick J., Schilling K., Smeyne R.J., Corbin J.G., Bocchiaro C. and Morgan J.I. (1993) Control of segment-like patterns of gene expression in the mouse cerebellum. Neuron, 10:

Oberdick J. (1994) Evidence for a genetically encoded map of functional development in the cerebellum. Hisrochemistry, 102: 1-14.

Ohsako S., Hyer J., Panganiban G., Oliver I. and Caudy M.

1799- 1805.

1007-1018.

(1994) Hairy function as a DNA-binding helix-loop-helix repressor of Drosophila sensory organ formation. Genes and Deuelopment, 8: 2743-2755.

Preiss A., Hartley D.A. and Artavanis-Tsakonas S. (1988) The molecular genetics of Enhancer of split, a gene required for embryonic neural development in Drosophila. EMBO J.,

Ruvkun G. and Finney M. (1991) Regulation of transcription and cell identity by POU domain proteins. Cell, 64: 475-478.

Sakagami T., Sakurada K., Sakai Y., Watanabe T., Nakanishi S. and Kageyama R. (1994) Structure and chromosomal locus of the mouse gene encoding a cerebellar Purkinje cell-specific helix-loop-helix factor HES-3. Biochim. Bio- phys. Reseach Com., 203: 594-601.

Sambrook J.. Fritsch E. f. and Maniatis T. (1989) Molecular cloning. A laboratory manual.

Sanlioglu-Crisman S. and Oberdick J. (1995) Activation of a Purkinje cell specific gene by homeobox proteins. Mof. Bio. of the Cell 6 Supplement: 344a.

Sanlioglu-Crisman and Oberdick J. (1 996) Effect of ectopic En-2 expression in Purkinje cells on development of the cerebellum. Ann. SOC. Neurosci. Abstracts, 26: 121.2.

Sasai Y., Kageyama R., Tagawa Y., Shigemoto R. and Nakan- ishi S. (1992) Two mammalian helix-loop-helix factors structurally related to Drosophila hairy and Enhancer of split. Genes and Development, 6: 2620-2630.

Smeyne R.J., Oberdick J., Schilling K., Berrebi AS., Mugnaini E. and Morgan J.I. (1991) Dynamic organization of developing Purkinje cells revealed by transgene expression. Sci- ence, 254: 719-721.

Smeyne R.J. Chu T., Lewin A., Bian F., S-Crisman S., Kunsch C.. Lira S.A. and Oberdick J. (1995) Local control of granule cell generation by cerebellar Purkinje cells. Mol Cell Neurosci, 6: 230-251.

Soosaar A., Chiaramello A., Zuber M. X. and Neuman T. (1994) Expression of basic-helix-loop-helix transcription factor ME2 during brain development and in the regions of neuronal plasticity in the adult brain. Mol. Brain Research, 25: 176-180.

Suzuki N., Peter W., Ciesiolka T., Gruss P. and Schoeler H.R. (1993) Mouse Oct-1 contains a composite homeodomain of human Oct-1 and Oct-2. Nucleic Acids Res., 21: 245-252.

Tiranti V., Rocchi M., Donato S.D. and Zeviani M. (1993) Cloning of human and rat cDNAs encoding the mitochon- drial single-stranded-DNA binding protein. Gene, 126:

Vaessin H., Caudy M., Bier E., Jan L.Y. and Jan Y.N. (1990) Role of helix-loop-helix proteins in Drosophila neurogenesis. Cold Spring Harbor Symp. on Quant. Biol., LV: 239-245

7: 3911-3927.

21 9-225.

C.I. dc Zccuw, P. Strata and J . Voogd (Eds.) Progress in Brain Research. Vol 114 0 1997 Elscvicr Scicncc BV. All rights rcserved.

CHAPTER 2

Transverse and longitudinal patterns in the mammalian cerebellum

J. Voogd and T.J.H. Ruigrok

Department of Anatomy, Emsmus University Rotterdam, P.O. Box 1738,3000 DR Rotterdam, The Netherlands

Transverse and longitudinal patterns of organization can be recognized in the gross anatomy of the cerebellum, in the perpendicular arrangement of the parallel fibers with respect to the Purkinje, basket and stellate cells of the cerebellar cortex (Braitenberg and Atwood 1958: the lattice structure of the cerebellar cortex), in its chemoarchi- tecture and in the distribution of the afferent and efferent connections of the cerebellar cortex.

The stage for functional interpretations of the morphology of the mammalian cerebellum was set by the comparative anatomical studies of Bradley (1903, 19041, Ellioth Smith (1900, 1902, 1903) and Bolk (1906). Bradley and Ellioth Smith stressed the early appearance and the constancy of certain transverse fissures. This position was also adopted by Larsell(1937), who advocated the subdivision of the cerebellum into transverse lobules and who considered the distinction of vermis and hemispheres of secondary importance. In Larsells (1936) words the lobules of the hemisphere are merely lateral extensions of the medial portion. Bolk (1906) accentuated the relative independance of vermis and hemispheres. In the caudal part of the posterior lobe (Bolks lobulus complicatus) the deep paramedian sulcus sepa- rates the vermis from the hemisphere and the transverse fissures of vermis and hemisphere gen- erally are discontinuous. The anterior lobe and the lobulus simplex, where the paramedian sulcus

is either absent or indistinct and does not inter- rupt the continuity of the transverse fissures, gen- erally lack a subdivision into vermis and hemispheres. Bolks own ideas on functional localization in the cerebellum were not based on the study of its connections, but on a correlation of the relative size of its parts with the degree of independance of movements of the limbs and axial structures in different species. He considered that the coordination of axial movements of head and neck was localized in the unpaired anterior lobe and lobulus simplex, whereas the independancy of limb and axial movements found its expression in their localization in the hemisphere lobule and the vermis of the lobulus complicatus respectively (see Glickstein and Voogd, 1995 for a discussion of this topic).

The cerebellum has been subdivided into larger and smaller functional units, that perform the same operations, but differ in their connections. Transverse and longitudinal patterns of organization also were recognized at the level of these functional subunits but, in most descriptions, either transverse or longitudinal principles prevail. The vestibular, spinal and corticopontine divisions of the mammalian cerebellum of Ingvar (19191, Larsell (1937) and Dow (1942) are arranged as a rostro-caudal series of transverse lobules (Fig. 1). Longitudinal functional units were distinguished by Edinger (1910) and Comolli

22

Fig. 1 . Diagram of the transverse lobular distribution of evoked potentials on stimulation of spino-, ponto- and vestibulo-cerebellar afferent systems. The position of the medial, intermediate and lateral corticonuclear projection zones of Jansen and Brodal (1940) is indicated in the anterior lobe and the lobulus simplex. From Dow (1942).

(1910) as their palaeocerebellar vermis and the neocerebellar hemisphere, by Jansen and Brodal (1940) and Chambers and Sprague (1955a,b) in their partition of the cerebellum into vermis, pars intermedia and hemisphere, by Oscarsson (1980) as narrow sagittal zones that control a particular motor mechanism and by Anderson and Os- carsson (1978b) as the even narrower microzones.

It appears as though transverse, lobular subdi- visions of the cerebellum are mainly based on the distribution of the mossy fiber input, whereas longitudinal divisions represent the output systems of the cerebellar cortex and consist of the Purkinje cell zones and their cerebellar and vestibular target nuclei. Climbing fibers are distributed according to the same parasagittal pat-

tern and, therefore, could subserve a function in the modification or selection of the cerebellar output. The transverse arrangement of the mossy fibers is still enhanced by the transverse orientation of the parallel fibers, that link the mossy fiber input to the granule cells with the Purkinje cells. At the border of vermis and hemispheres the molecular layer is often attenuated or com- pletely interrupted (Voogd, 1975) thus, limiting particular mossy fiber-parallel fiber inputs to the lobules of vermis or hemisphere. Certain mossy fiber systems do not terminate as continuous, transversely oriented fields but as a series of discontinuous patches (Welkers fractured soma- totopy: see Welker, 1987, for a review) or longitudinal concentrations of mossy fiber rosettes in

23

vermis and hemispheres (Voogd, 1969; Van Rossum, 1969; Gravel and Hawkes, 1990; Ji and Hawkes, 1994).

Zonal and modular organization of the anterior lobe in the cat

The output of the cerebellar cortex, through the Purkinje cells and the cerebellar and vestibular nuclei, is organized in a zona or modular fashion. Jansen and Brodal(1940) were among the first to subdivide the cerebellum in longitudinal zones that project to different target nuclei (Fig. 1). Their medial zone corresponds to the vermis and projects to the medial cerebellar and vestibular nuclei. Their intermediate zone projects to the nucleus interpositus and their lateral zone to the lateral or dentate nucleus. They noticed a correspondence between these corticonuclear projection zones and the longitudinal arrangement in the olivocerebellar projection, with the accessory olives projecting to the medial and intermediate zones, and the principal olive to the lateral zone of the anterior lobe (Brodal, 1940). The three corticonuclear projection zones of Brodal and Jansen were further subdivided by Voogd (1964, 1969; Voogd and Bigar6, 1980), on the basis of the observation that the axons of the Purkinje cells of a particular longitudinal zone on the way to their target nucleus occupy a specific compart- ment in the cerebellar white matter. The borders of these compartments, where Purkinje cell axons are absent, can be positively stained with acetyl- cholinesterase (Hess and Voogd, 1986; Voogd et al., 1987, 1996a,b; Tan et al., 1995a). Climbing fibers from certain subnuclei of the inferior olive use the same compartments to distribute to particular Purkinje cell zones (Voogd, 1969; Groenewegen and Voogd, 1977; Groenewegen et al., 1979; Voogd and Bigare, 1980; Tan et al., 1995b). Six zones (A,B,C,,C,,C,,D, and D,), each with its own cerebellar or vestibular target nucleus and its private climbing fiber afferents, orig- inally were distinguished. The X zone (Fig. 2) is a

later addition, prompted by the electro-physiological studies of Oscarsson and his group.

The electrophysiological mapping studies of spino-olivocerebellar climbing fiber paths (SOCPs) of Oscarsson (1980) and Armstrong et al. (1974; Armstrong, 1974) confirmed and greatly extended our knowledge of the zonal topography. Their criteria to distinguish SOCPs included the localization of these pathways in the spinal cord and their pre-olivary relay nuclei in the brain stem and their latency, laterally and somatotopic organization. The topography of the projection zones of the SOCPs closely resembles the disposi- tion of the anatomical zones. To distinguish the electrophysiological zones they are indicated with lower case characters (Fig. 3). In certain zones the climbing fibers that share the same receptive field are arranged in long and narrow (1-4 Purk- inje cells wide) microzones. The somatotopical organization of these zones is represented by the ensemble of microzones. Antidromic stimulation proved to be a powerfull technique to study syste- matic branching of olivocerebellar fibers within, or between sets of parasagittal climbing fiber zones (Armstrong et al., 1973). Branching in the transverse plane revealed three sets of zones in the anterior lobe of the cat (Fig. 3) that receive collaterals from the same olivary neurons (Ekerot and Larson, 1982). The x zone and the c, zone (indicated as the lateral part of c,) receive branches from a region of the medial accessory olive, located at the junction of its caudal and rostral halves (Campbell and Armstrong, 1985). The pairs of zones (medial c, and the medial c3, and lateral c3 and a zone in the hemisphere variously known as zone y or d,) are innervated by branching fibers originating from the rostral dorsal accessory olive. The branching fibers syste- matically skip one or two intervening zones that are innervated by other subnuclei of the inferior olive.

The electrophysiological climbing fiber zones a, b, x, c,, c2 and c3 zones correspond to their anatomical namesakes and were found to project to the same cerebellar or vestibular target nuclei

24

dorsal

princip

medial accessory olive

medial accessory principal olive olive

dorsal accessory olive

Fig. 2. Diagram of the corticonuclear and olivocerebellar projection, based on experimental studies in the cat. A transverse section through the inferior olive and a diagram of the unfolded inferior olive are represented in the left panels. The cerebellar cortex as represented in the right panel, similarly, is unfolded. Regions that are interconnected are indicated with the same symbols. Abbreviations: A = A zone; B = B zone; /3 = group beta; C,-C, = C,-C3 zones; D = D zone; Dent = dentate nucleus; dmcc = dorsomedial cell column; Dt = Lateral vestibular nucleus of Deiters; F = fastigial nucleus; IA = anterior interposed nucleus; IP = posterior interposed nucleus; MV = medial vestibular nucleus; X = X zone; VI-X = lobules VI-X.

(Trott and Armstrong, 1987a,b, 1990). The c, (or lateral c , ) and y (or d,) zones and the partition of the c3 zone in medial and lateral parts have not been definitely identified in the anatomical studies. The inclusion of the c, (lateral c , ) zone with the c , zone by Ekerot and Larson (1982, Apps et al., 1991; Trott and Apps, 1991) was based on the similarity in the electrophysiological properties of the dorsal funiculus spino-olivocerebellar climbing fiber paths terminating in both zones. Ana- tomically the c, zone should be included with the C, zone, because it receives its climbing fibers from the medial accessory olive (Fig. 4). Gener- ally zones with a common climbing fiber input project to the same target nucleus. This is true for the projection of the anatomical and electrophysiological C,/medial c1 and C3/c3 zones in the cat to the anterior interposed nucleus. However, the

projection of the c, (lateral c , ) zone was located in the anterior interposed nucleus, whereas the x zone, which receives branches from the same climbing fibers from the medial accessory olive, projects to the medial part of the posterior interposed nucleus (Trott and Armstrong, 1987a,b; Voogd et al., 1991a).

Efferent modules and the mossy fiber afferent projection to the anterior lobe of the rat: partial congruence with the zebrin pattern

The earlier anatomical and electrophysiological studies, that were reviewed in the previous section of this chapter, were mostly conducted in the cat. The discovery of the zebrins, molecular markers for longitudinally arranged subpopulations of Purkinje cells in mouse and rat (Hawkes and

25

vermis pars intermedia la zOne tera 1

0 - d ~ C1-C3 X-CI Fig. 3. Termination areas in the anterior lobe of forelimb activated olivary neurons with transversely branching axons. Areas innervated by x - cI(cx) neurons are indicated by stip- pling, areas innelvated by c , - c j neurones by sparse hatching and areas innervated by c j - d2(y) neurones by dense hatching. Branching of olivary axons is indicated by forked arrows. Hindlimb areas of the c,. c3 and d, zones are indicated by light shading. Continuous lines indicate borders between sagittal zones and lines paravermal groove and midline. Borders between the c3, d , and d, zones in the rostra1 part of the anterior lobe are based on unpublished observations by Ekerot and Larson. Vermal, intermediate and lateral cortices according to Jansen and Brodal (1940). Abbreviations: small letters and indices indicate sagittal zones according to Ekerot and Larson (1979). Roman numerals refer to lobules according to Larsell. PF = primary fissure; 10 = inferior olive; LAT = pars lateralis; lntermed = pars intermedia; H. Vermis = hemivermis. From Ekerot and Larson (1982).

k l e r c , 1987; Hawkes, this volume), and the de- scription of the A, X, B, C and D zones in the

Fig. 4. Correspondence between the classical three-zonal subdivision of the cerebellum of Jansen and Brodal (1940: upper panel), the sagittal projection zones of the spino-olivo-cerebellar climbing fiber paths of Oscarsson C.S. (middle panel) and the anatomical zones of Voogd (lower panel). Arrows indicate the transverse branching of climbing fibers between zones (Ekerot and Larson, 1982). Hatched zones receive short-latency input from the DF-SOCP, Ekerot and Larson (1979).

cerebellum of the rat (Buisseret-Delmas, 1988a,b; Buisseret-Delmas and Angaut, 1993; Buisseret- Delmas et al., 19931, made it possible to study the modular organization of the cerebellum in more detail. The nucleo-olivary projection (Ruigrok and Voogd, 1990) and the olivocerebellar projection to the flocculus and the paraflocculus (Ruigrok et al. 1992) were studied in the rat and investiga- tions of the collateral projection of the inferior olive to the cerebellar and vestibular nuclei are under way (Ruigrok, this volume). The main con- clusion of these studies is that the nucleo-olivary and olivonuclear connections in the rat, are strictly reciprocal. A large number of experiments in the rat, with small injections of the anterograde axonal tracer Phaseolus vulgaris lectin, or the retrograde tracer wheatgerm-coupled horseradish per- oxidase (WGA-HRP) is available. In most of these experiments the labeled structures were stabilized with cobalt (Lemann et al., 1985) and the sections were counterstained with an antibody to zebrin I (Hawkes and Leclerc, 1987). Some of these experiments are reported in this chapter (Voogd et al., 1991b).

The zebrin pattern in the anterior lobe of the rat consists of a midline and 5 parasagittal bands of zebrin-positive Purkinje cells, indicated as the P1 + /P6 + bands by Hawkes and Leclerc (1987). They are separated by wide bands of zebrin-nega-

26

tive Purkinje cells indicated with the same number as the zebrin-positive zones on their medial side (P1 - /P5 - ). The zebrin pattern is highly reproducible and can be used to compare the localization of labeled Purkinje cells, mossy and climbing fibers in different experiments. The questions that will be studied are: (1) is it possible to identify a similar connectivity of the A, X and B zones in the anterior vermis of the rat as has been reported for the cat ?, (2) do the A, X and B zones and their climbing fiber afferents coincide with zebrin-positive, or zebrin-negative Purkinje cell zones ?, (3) what is the topographical relation between these zones and the distribution of spino- and cuneocerebellar mossy fibers, as reported in the literature?

Corticonuclear projection and climbing fiber afferents of the anterior vermis of rat cerebellum

It has been known for some time that Purkinje cells in the anterior vermis that project to the vestibular nuclei are localized in two zones: A and B, separated by a wedge-shaped area that contains Purkinje cells that do not project to the vestibular nuclei, corresponding to the X zone. This configuration has been found in cat, rabbit, rat and monkey (Voogd and Bigare, 1980; Bala- ban, 1984; Epema, 1990; Voogd et al., 1991a 1. In the rat the Purkinje cells that are retrogradely labelled from injections of WGA-HRP in the vestibular nuclei, including the lateral vestibular nucleus, occupy the zebrin-negative P1 - and P2 - zones (Fig. 4, case 16). In lobules 1-111 they are separated by one or two zebrin-positive Purk- inje cells of P2 + , In lobules IV and V the labeled Purkinje cells in P1 - are present in its lateral half; medially they are bordered by a narrow zebrin-positive satellite band, that is always present in the anterior lobe and the simple lobule (Figs. 4 and 5). Purkinje cells in P2 - in these lobules remain separated from P2 + by a gap containing unlabeled zebrin-negative Purkinje cells (Fig. 4, open arrow). Labeling of Purkinje cells of P2 + and of the satellite band in P1 - is found with injections of retrograde tracers in the

border region of the fastigial and posterior interposed nuclei (the interstitial cell groups of Buis- seret-Delmas et al. 1993) (Fig. 5, case 29L). The labeling is confined to lobules IV and V, but is not restricted to P2 + , some zebrin-negative Purkinje cells in medial P2 - are also labeled. With injections more laterally in the posterior interposed nucleus, few labeled Purkinje cells are present in P2 + , more are found in medial P2 - (Fig. 4, open arrow) and in P4 + , extending into the zebrin-negative P3 - band (Fig. 4, case 29R).

The same pattern can be recognized in the distribution of climbing fibers from small injections of Phaseoh vulgaris lectin in the contralat- era1 inferior olive. Injections in the dorsal fold of the dorsal accessory olive label a climbing fiber band in P2 - , located next to P2 + in the lobules 1-111, but separated from it by a gap in lobules IV and V. Climbing fibers from the dorsal fold emit collaterals to the lateral vestibular nucleus (Fig. 6, case 415). Climbing fiber labeling in the lateral P1 - strip, corresponding in position with the Purkinje cells that can be labeled from the vestibular nuclei, were present in a single case with an injection of the caudal medial accessory olive (Fig. 6, case 548). Collateral labeling was present in the fastigial nucleus and in the lateral vestibular nucleus at its border with the medial vestibular nucleus. Climbing fibers in P2 + and the adjoining medial strip of P2 - are labeled with injections of the medial accessory olive, located at the junction of its caudal and rostra1 halves or more rostromedially (Figs. 5 and 6, cases 428R, 427R and 459). Collateral labeling in these cases is found in the posterior interposed nucleus at its junction with the fastigial nucleus (Fig. 6, case 459).

It can be concluded that the correspondence between the pattern of zebrin-positive and -negative Purkinje cells in the rat anterior vermis, and the zonal distribution of Purkinje cells with specific efferent and afferent connections is only partial. Purkinje cell of the B zone, defined by a projection to the lateral vestibular nucleus and climbing fiber afferents from the caudal pole (the dorsal fold) of the dorsal accessory olive, occupy

27

V

IV

I I I

II

\

V

IV

111

II

23 L

m J

@ ///. '// '

2 9 R

m I

Fig. 5. Graphical reconstructions in anterior view, of the anterior lobe of rat cerebellum, prepared from serial sections with retrograde labeling of Purkinje cells from injection sites in the vestibular or cerebellar nuclei, counterstained with an antibody against zebrin I. Zebrin-positive Purkinje cell zones are indicated in black, 1-6 denote the zebrin-positive Purkinje cell zones P1 + /P6 + of Hawkes and Leclerc (1987). Labeled Purkinje cells are indicated with empty circles. The WGA-HRP injections are indicated in the diagrams of the cerebellar nuclei at the bottom of the figure. Open arrow indicates medial part of zone that Purkinje cells projecting to the interstitial cell groups and/or the posterior interposed nucleus. Filled arrow indicates lateral part of PI - that contains Purkinje cells projecting to vestibular nuclei. Abbreviations: I-V = lobules I-V; AIN = anterior nucleus; DLH = dorsolateral hump; DLP = dorsolateral protuberance; icg = interstitial cell groups of Buissere-Delmas et al. (1993); LVN = lateral vestibular nucleus; MCN = medial cerebellar nucleus; PIN = posterior interposed nucleus. Voogd and Ruigrok, unpublished.

the entire P2 - zone, in lobules 1-111, but only its lateral half, in the lobules IV and V. Purkinje cells in the lobules lV and V of P2 + and the medial half of P2 - project to the junction of the posterior interposed and fastigial nucleus, and receive climbing fibers from the medial accessory olive, that give off collaterals to the same region of the cerebellar nuclei. These Purkinje cells dis- play some of the characteristics of the X zone. The X zone of the rat, therefore, would correspond to a zebrin positive (P2 + and an adjoining zebrin-negative (P2 - ) region in lobules IV and V.

The identification of this region as the X zone

is supported by the observation that climbing fiber labeling in the X zone is always found together with climbing fiber labeling in P4 + or the medially adjoining strip of P3 - . Branching of climbing fibers between the X and the C, zone (variously interpreted as part of C, or of C,) has been discussed in the previous section. The present data are insufficient to decide whether the labeling in P3 - and/or P4 + represents a C, or a C, zone. Moreover, the question whether the Purkinje cells of P2 + in the lobules 1-111, which were not labeled and did not receive labeled climbing fibers in our experiments, belong to the X zone, remains unanswered. Experiments de-

28

Fig. 6. Graphical reconstruction in anterior view, of the anterior lobe of rat cerebellum, prepared from serial sections with anterograde labeling of climbing fibers from injection sites of fhaseolus vulgaris lectin in the inferior olive, counterstained with an antibody against zebrin I. Zebrin-positive Purkinje cell zones are indicated in black, 1-6 denote the zebrin-positive Purkinje cell zones P1 + /P6 + of Hawkes and Leclerc (1987). Labeled climbing fibers are indicated with empty lines. The injections in the inferior olive are indicated in the diagrams of the unfolded medial and dorsal accessory olives at the bottom of the figure. Open arrows indicate medial part of P2 - zone that receives climbing fibers from the medial accessory olive. Abbreviations: p = group beta; DAO = dorsal accessory olive; dmcc = dorsomedial cell column; M A 0 = medial accessory olive. Voogd and Ruigrok, unpublished.

signed to study the branching of climbing fibers between identified zones may solve these questions.

Injections of retrograde tracers in the vestibular nuclei, including the lateral fastigial nucleus, always label the two strips of Purkinje cells in the A and B zones. It has been suggested on the basis of smaller injections in the vestibular nuclei of the cat and the rabbit (Voogd and Bigart, 1980; Epema, 1990) that the B zone projects to Deiters nucleus, and the A zone to the rostra1 medial vestibular nucleus. This idea was tested in experiments with injections of Phaseolus vulgaris lectin in the anterior vermis of the rat. Purkinje cells axons from injections of P1 - , pass through the fastigial nucleus, where they appear to terminate,

into the ventral part of the lateral vestibular nucleus. They are scarce or absent in the medial and descending vestibular nuclei (Fig. 7). Injec- tions in P2 - label Purkinje cells axons that show some varicosities in the medialmost part of the anterior interposed nucleus, and terminate in the dorsal and central parts of the lateral vestibular nucleus.

It can be concluded from these, still prelimi- nary, observations that Purkinje cells from the A (P1 - ) zone project to both the fastigial and the lateral vestibular nucleus. It is not clear whether the single projection of the B (P2 - ) zone to the lateral vestibular nucleus overlaps with the projection from the A zone. These projections are closely matched by the collateral projections to

29

ventral -

fold dorsal fold

Fig. 7. Distribution of olivocerebellar fibers, their nuclear collaterals and their climbing fiber terminals in three transverse sections through the cerebellum of the rat in cases with injections of Phaseoh uulguris lectin in subnuclei of the inferior olive (diagrams on the left). Abbreviations: 1-4 = zebrin-positive Purkinje cell zones P1 + /P4 + ; 7 = genu of the facial nerve; ANS = ansiform lobule: p = group beta; CO = cochlear nuclei; cr = restiform body; DAO = dorsal accessory olive; DLP = dorsolateral protuberance; dmcc = dorsomedial cell column; DV = descending vestibular nucleus; F = fastigial nucleus; FLO = flocculus; I = interstitial cell groups of Buisseret-Delmas et al. (1993); IA = anterior interposed nucleus; L = lateral cerebellar nucleus; lob ant = anterior lobe; LV = lateral vestibular nucleus; M A 0 = medial accessory olive; MV = medial vestibular nucleus; PFL = paraflocculus; SI = lobulus simplex; V = spinal tract of the trigeminal nerve. Voogd and Ruigrok, unpublished.

30

the cerebellar and vestibular nuclei of the medial accessory olive to the A zone and of the dorsal accessory olive to the B zone (Fig. 6). The demon- stration of a projection of Purkinje cells of the A zone that are innervated by the medial accessory olive, to the lateral vestibular nucleus confirms earlier observations in electrophysiological experiments of Andersson and Oscarsson (1978a) in the cat.

Our observations confirm the presence and the afferent and efferent projections of the X and B zones in the anterior vermis of the rat, as determined by Buisseret-Delmas (1988a) and Buis- seret-Delmas et al. (1993) from small injections with the bidirectionally transported axonal tracer WGA-HRP in the cortex With respect to the zebrin pattern, our experiments show a partial congruence with the modular organization of the output systems of the cerebellar cortex Cer- tain borders of the modules at the level of the cortex i.e. the border between X and B in lobule IV and V, are located in the middle of the zebrin-negative zone.

Spino- and cuneocerebellar mossy fiber projections to the anterior lobe

Spinocerebellar and cuneocerebellar fibers in the anterior lobe of different species terminate in parasagittal aggregates of mossy fibers and their rosettes (Voogd, 1967, 1969; Van Rossum, 1969; Hazzlett et al., 1971; Watson et al., 1976; Gerrits et al., 1984). Matsushita, who studied different spinocerebellar systems in great detail, pointed out that these systems differ in their termination, with respect to their lobular distribution, the me- diolateral extent of the terminal field, their baso-apical d

The Cerebellum: From Structure to Control

Documents

university of bristol

department of physiology

university walk

university of oslo

university of turin

erasmw university rotterdam

erasmus university rotterdam

department of otolaryngology