Methods in

Molecular Biology 1082

Simon G. Sprecher Editor

Brain
Development
Methods and Protocols
METHODS IN MOLECULAR BIOLOGY™

Series Editor
John M. Walker
School of Life Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651
 Brain Development

Methods and Protocols

Edited by

Simon G. Sprecher
Department of Biology, University of Fribourg, Fribourg, Switzerland
Editor
Simon G. Sprecher
Department of Biology
University of Fribourg
Fribourg, Switzerland

ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-62703-654-2 ISBN 978-1-62703-655-9 (eBook)
DOI 10.1007/978-1-62703-655-9
Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2013946845

© Springer Science+Business Media, LLC 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for
the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution
under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be
made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Humana Press is a brand of Springer

Springer is part of Springer Science+Business Media (www.springer.com)
Preface

The brain is without any doubt the most complex organ. Particularly, cellular diversity, con-
nectivity among neurons, formation of neuronal networks, the use of distinct neurotrans-
mitter system, and the underlying function for behavior raise the question of how this
highly interconnected organ develops. It is therefore not surprising that the intersection
between developmental biology and neuroscience provides an exceptional field to address
and investigate impacting biological questions. Complementing findings of an array of dis-
tinct animal model systems provide the basis of brain development research. Our current
understanding is based on widely used genetic model systems including the fruit fly, zebra
fish, chicken, and mouse. These animal models are impacting in particular since they allow
elaborate genetic manipulations including transgenic expression systems and conditional
knockout or knockdown of developmental genes. Genetic developmental studies are fur-
ther complemented by several non-genetic animal models, which further substantiate gen-
eral principles and mechanisms in brain development. Questions that can be investigated
often depend on the methodological accessibility. Therefore, progress and developments in
the constantly improving laboratory technologies provide an essential foundation for the
advancement in the field.
This book aims to provide a broad overview and introduction of widely used leading-
edge techniques in genetic model systems as well as some of the complementing animal
models. The main focus lies on two key technical aspects of developmental neurobiology:
Detection of gene expression and functional characterization of developmental control
genes. The basic principle of expression and function studies are shared between different
model systems. This includes in situ hybridization, reporter gene expression, and immuno-
histochemical staining methods, as well as RNA interference, Morpholino, or transgenic
techniques. However the experimental procedure, such as, for instance, tissue treatment,
fixation, dissection, genetic manipulation, and imaging, often differs substantially between
animal models or even distinct stages in the same species. The collection of protocols aims
to provide precise technical protocols but also allows for comparing a wide range of proto-
cols in different tissues and species.

Fribourg, Switzerland Simon G. Sprecher

v
Contents

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

PART I DROSOPHILA PROTOCOLS

1 Immunostaining of the Developing Embryonic

and Larval Drosophila Brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Danielle C. Diaper and Frank Hirth
2 Non-fluorescent RNA In Situ Hybridization Combined with Antibody
Staining to Visualize Multiple Gene Expression Patterns in the Embryonic
Brain of Drosophila . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
David Jussen and Rolf Urbach
3 Analysis of Complete Neuroblast Cell Lineages
in the Drosophila Embryonic Brain via DiI Labeling . . . . . . . . . . . . . . . . . . . . . 37
Karoline F. Kraft and Rolf Urbach
4 Flybow to Dissect Circuit Assembly in the Drosophila Brain . . . . . . . . . . . . . . . . 57
Nana Shimosako, Dafni Hadjieconomou, and Iris Salecker
5 Immunofluorescent Labeling of Neural Stem Cells
in the Drosophila Optic Lobe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Benjamin Perruchoud and Boris Egger
6 Using MARCM to Study Drosophila Brain Development . . . . . . . . . . . . . . . . . 79
Gudrun Viktorin

PART II OTHER ARTHROPODS

7 Dye Coupling and Immunostaining of Astrocyte-Like Glia Following

Intracellular Injection of Fluorochromes in Brain Slices
of the Grasshopper, Schistocerca gregaria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
George Boyan and Yu Liu

PART III MOLLUSCS

8 Methods in Brain Development of Molluscs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Andreas Wanninger and Tim Wollesen

PART IV XENOPUS PROTOCOLS

9 In Situ Hybridization and Immunostaining of Xenopus Brain. . . . . . . . . . . . . . . 129

Kai-li Liu, Xiu-mei Wang, Zi-long Li, Rong-qiao He, and Ying Liu

vii
viii Contents

10 Microinjection Manipulations in the Elucidation of Xenopus

Brain Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Cristine Smoczer, Lara Hooker, Saqib S. Sachani,
and Michael J. Crawford
11 Morpholino Studies in Xenopus Brain Development . . . . . . . . . . . . . . . . . . . . . 155
Jennifer E. Bestman and Hollis T. Cline

PART V ZEBRAFISH PROTOCOLS

12 Sensitive Whole-Mount Fluorescent In Situ Hybridization in Zebrafish

Using Enhanced Tyramide Signal Amplification. . . . . . . . . . . . . . . . . . . . . . . . . 175
Gilbert Lauter, Iris Söll, and Giselbert Hauptmann
13 Dynamic Neuroanatomy at Subcellular
Resolution in the Zebrafish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Adèle Faucherre and Hernán López-Schier
14 Anatomical Dissection of Zebrafish Brain Development . . . . . . . . . . . . . . . . . . . 197
Katherine J. Turner, Thomas G. Bracewell, and Thomas A. Hawkins

PART VI CHICKEN PROTOCOLS

15 Immunohistochemistry and In Situ Hybridization

in the Developing Chicken Brain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Richard P. Tucker and Qizhi Gong
16 Transplantation of Neural Tissue: Quail-Chick Chimeras . . . . . . . . . . . . . . . . . . 235
Andrea Streit and Claudio D. Stern
17 RNAi-Based Gene Silencing in Chicken Brain Development . . . . . . . . . . . . . . . 253
Irwin Andermatt and Esther T. Stoeckli

PART VII MOUSE PROTOCOLS

18 Immunohistochemistry and RNA In Situ Hybridization in Mouse

Brain Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Jinling Liu and Aimin Liu
19 In Utero Electroporation to Study Mouse Brain Development . . . . . . . . . . . . . 285
Emilie Pacary and François Guillemot
20 The Cre/Lox System to Assess the Development
of the Mouse Brain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Claudius F. Kratochwil and Filippo M. Rijli

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Contributors

IRWIN ANDERMATT • Institute of Molecular Life Sciences and Neuroscience Center Zurich,
University of Zurich, Zurich, Switzerland
JENNIFER E. BESTMAN • The Dorris Neuroscience Center, The Scripps Research Institute,
La Jolla, CA, USA
GEORGE BOYAN • Developmental Neurobiology Group, Biocenter, Ludwig-Maximilians-
Universität, Martinsried, Germany
THOMAS G. BRACEWELL • Department of Cell and Developmental Biology, University College
London, London, UK
HOLLIS T. CLINE • The Dorris Neuroscience Center, The Scripps Research Institute, La Jolla,
CA, USA
MICHAEL J. CRAWFORD • Biological Sciences, University of Windsor, Windsor, ON, Canada
DANIELLE C. DIAPER • Department of Neuroscience, MRC Centre for Neurodegeneration
Research, Institute of Psychiatry, King’s College London, London, UK
BORIS EGGER • Zoology Unit, Department of Biology, University of Fribourg, Fribourg,
Switzerland
ADÈLE FAUCHERRE • Département de Physiologie, Institut de Génomique Fonctionnelle,
CNRS UMR 5203, INSERM U661, Universtités Montpellier 1 & 2, Montpellier, France
QIZHI GONG • Department of Cell Biology and Human Anatomy, University of California,
Davis, CA, USA
FRANÇOIS GUILLEMOT • Division of Molecular Neurobiology, MRC National Institute for
Medical Research, London, UK
DAFNI HADJIECONOMOU • Division of Molecular Neurobiology, MRC National Institute for
Medical Research, London, UK
GISELBERT HAUPTMANN • Department of Biosciences and Nutrition, NOVUM, Karolinska
Institutet, Huddinge, Sweden
THOMAS A. HAWKINS • Department of Cell and Developmental Biology, University College
London, London, UK
RONG-QIAO HE • State Key Laboratory of Brain and Cognitive Science, Institute of
Biophysics, Chinese Academy of Sciences, Beijing, China
FRANK HIRTH • Department of Neuroscience, MRC Centre for Neurodegeneration
Research, Institute of Psychiatry, King’s College London, London, UK
LARA HOOKER • Biological Sciences, University of Windsor, Windsor, ON, Canada
DAVID JUSSEN • Institute of Genetics, University of Mainz, Mainz, Germany
KAROLINE F. KRAFT • Institute of Genetics, University of Mainz, Mainz, Germany
CLAUDIUS F. KRATOCHWIL • Friedrich Miescher Institute for Biomedical Research,
Basel, Switzerland
GILBERT LAUTER • Department of Biosciences and Nutrition, NOVUM, Karolinska
Institutet, Huddinge, Sweden
ZI-LONG LI • State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics,
Chinese Academy of Sciences, Beijing, China

ix
x Contributors

AIMIN LIU • Department of Biology, Eberly College of Science, Center for Cellular Dynamics,
Huck Institute of Life Sciences, The Penn State University, University Park, PA, USA
JINLING LIU • Department of Biology, Eberly College of Science, Center for Cellular Dynamics,
Huck Institute of Life Sciences, The Penn State University, University Park, PA, USA
KAI-LI LIU • State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese
Academy of Sciences, Beijing, China
YING LIU • State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese
Academy of Sciences, Beijing, China
YU LIU • Developmental Neurobiology Group, Biocenter, Ludwig-Maximilians-Universität,
Martinsried, Germany
HERNÁN LÓPEZ-SCHIER • Unit of Sensory Biology & Organogenesis, Helmholtz Zentrum
München, Munich, Germany
EMILIE PACARY • Neurocentre Magendie, Physiopathologie de la Plasticité Neuronale,
U862, Institut National de la Santé et de la Recherche Médicale, Université de Bordeaux,
Bordeaux, France
BENJAMIN PERRUCHOUD • Zoology Unit, Department of Biology, University of Fribourg,
Fribourg, Switzerland
FILIPPO M. RIJLI • Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland
SAQIB S. SACHANI • Biological Sciences, University of Windsor, Windsor, ON, Canada
IRIS SALECKER • Division of Molecular Neurobiology, MRC National Institute for Medical
Research, London, UK
NANA SHIMOSAKO • Division of Molecular Neurobiology, MRC National Institute for Medical
Research, London, UK
CRISTINE SMOCZER • Biological Sciences, University of Windsor, Windsor, ON, Canada
IRIS SÖLL • Department of Biosciences and Nutrition, NOVUM, Karolinska Institutet,
Huddinge, Sweden
CLAUDIO D. STERN • Department of Cell & Developmental Biology, University College London,
London, UK
ESTHER T. STOECKLI • Institute of Molecular Life Sciences and Neuroscience Center Zurich,
University of Zurich, Zurich, Switzerland
ANDREA STREIT • Department of Craniofacial Development & Stem Cell Biology, King’s College
London, London, UK
RICHARD P. TUCKER • Department of Cell Biology and Human Anatomy, University of
California, Davis, CA, USA
KATHERINE J. TURNER • Department of Cell and Developmental Biology, University College
London, London, UK
ROLF URBACH • Institute of Genetics, University of Mainz, Mainz, Germany
GUDRUN VIKTORIN • Biozentrum, University of Basel, Basel, Switzerland
XIU-MEI WANG • State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics,
Chinese Academy of Sciences, Beijing, China
ANDREAS WANNINGER • Faculty of Life Sciences, Department of Integrative Zoology, University
of Vienna, Vienna, Austria
TIM WOLLESEN • Department of Integrative Zoology, Faculty of Life Sciences, University of
Vienna, Vienna, Austria
 Part I

Drosophila Protocols
 Chapter 1

Immunostaining of the Developing Embryonic

and Larval Drosophila Brain
Danielle C. Diaper and Frank Hirth

Abstract
Immunostaining is used to visualize the spatiotemporal expression pattern of developmental control genes
that regulate the genesis and specification of the embryonic and larval brain of Drosophila. Immunostaining
uses specific antibodies to mark expressed proteins and allows their localization to be traced throughout
development. This method reveals insights into gene regulation, cell-type specification, neuron and glial
differentiation, and posttranslational protein modifications underlying the patterning and specification of
the maturing brain. Depending on the targeted protein, it is possible to visualize a multitude of regions of
the Drosophila brain, such as small groups of neurons or glia, defined subcomponents of the brain’s axon
scaffold, or pre- and postsynaptic structures of neurons. Thus, antibody probes that recognize defined
tissues, cells, or subcellular structures like axons or synaptic terminals can be used as markers to identify
and analyze phenotypes in mutant embryos and larvae. Several antibodies, combined with different labels,
can be used concurrently to examine protein co-localization. This protocol spans over 3–4 days.

Key words Drosophila, Embryo, Larva, Brain, Immunostaining, Fluorescence immunocytochemistry,

Dissection, Antibody

1 Introduction

Similar to mammalian brain development, the Drosophila brain

derives from a monolayered epithelium called the neuroectoderm.
Subsequent neurogenesis is characterized by two neurogenic
periods: one during embryogenesis and another during larval and
pupal stages. The precursor cells of the developing brain, termed
neuroblasts (NBs), derive from the embryonic procephalic neuro-
genic region to form proliferative clusters; they divide repeatedly
and asymmetrically in a stem cell mode to generate a new NB and
a smaller daughter cell, called ganglion mother cell (GMC). Each
GMC is a transient intermediate progenitor cell that generally
divides once to produce two lineage-specific postmitotic cells,
either neuron or glia. Both neuron and glia subsequently initiate
their differentiation processes, finally shaping neural segments and
circuits along the major body axes of the embryo [1, 2].

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_1, © Springer Science+Business Media, LLC 2014

3
4 Danielle C. Diaper and Frank Hirth

Fig. 1 The embryonic brain of Drosophila. (a–d) Laser confocal microscopy, superimposition of optical sections
of stage 14 embryo immunolabeled with anti-HRP-FITC. (a) Frontal view of brain; (b) ventral view of ventral
nerve cord; (c) lateral view of brain; (d) optical section along the midline of (c); (e) cartoon summarizing major
neural segments and axon scaffolding which together constitute the embryonic Drosophila brain. Abbreviations:
sec supraesophageal commissure, fc frontal connective, cnn circumesophageal connective, sco subesophageal
commissure, an antennal nerve, ac anterior commissure, pc posterior commissure, pcn protocerebral con-
nective, lc longitudinal connective, PC protocerebrum, DC deutocerebrum, TC tritocerebrum, MD mandibular
neuromere, MX maxillary neuromere, LB labral neuromere, es esophagus. Scale bar: 25 μm

The central brain of Drosophila derives from 106 embryonic

brain NBs that can be identified based on their positional relation-
ships and NB-specific gene expression [3]. Towards the end of
embryogenesis, most NBs stop proliferating and enter a period
of quiescence. During the first larval instar stage, quiescent NBs
reenter the cell cycle in a characteristic spatiotemporal pattern to
perform a second round of neurogenesis, resulting in postembry-
onic neural progeny. These two phases of brain development are
followed by an extensive morphological transformation during
metamorphosis that ultimately leads to the adult fly brain [2].
Although approximately 95 % of the 200,000 neurons in the
adult Drosophila brain are generated postembryonically, the main
architecture of the adult brain is already laid down during embryonic
neurogenesis [3, 4]. Thus, the Drosophila brain is composed of an
anterior (supraesophageal) and a posterior (subesophageal) part,
both of which are interconnected by an axon scaffold surrounding
the gut; the anterior part comprises the protocerebral, deutocerebral,
and tritocerebral neuromeres, whereas the posterior part comprises
the mandibular, maxillary, and labial neuromeres (Fig. 1).
Immunostaining of the developing embryonic and larval
Drosophila brain is used for two primary purposes. First, once a
 Immunostaining of Embryonic and Larval Drosophila Brain 5

gene of interest has been cloned and antibodies raised against its
encoded protein product(s), immunostaining can be used to
visualize the spatiotemporal pattern of protein expression during
embryonic and larval brain development. Second, antibody probes
that recognize defined tissues, cells, or subcellular structures like
axons or synaptic terminals can be used as markers to identify and
analyze phenotypes in mutant embryos and larvae. Thus, following
protein localization throughout Drosophila development reveals
insights into gene regulation, cell-type specification, neuron and
glial differentiation, and posttranslational protein modifications
underlying the patterning, specification, and neuronal connectivity
of the maturing brain.
In this respect, whole-mount immunohistochemistry (IHC),
which is the process of using antibody probes to detect antigens
(i.e., proteins), is among the most valuable of tools for analysis.
It is a relatively cheap and reliable way to visualize discrete struc-
tures or cell types within the developing embryonic and larval brain
of Drosophila. With the use of multiple antibodies, or when used in
conjunction with genetic labeling, such as mosaic analysis with a
repressible cell marker (MARCM) [5], it is possible to follow
several different proteins, examining co-localization, or identify
specific cells and follow their neuronal projections from the cell
body to its terminal dendritic arborizations [6–8]. Immunostaining
of Drosophila brain tissue is often used to help deconstruct the
complexities of neural circuit development or neurodegeneration
[9, 10], in many cases with the aim of understanding human disease
pathogenesis [11, 12]. The availability of Drosophila antibodies to
human homologues can pave the way for follow-up studies in
mammals or human tissue [13, 14].
There is a growing supply of Drosophila-specific antibodies
available that bind to all manner of cellular proteins, from organ-
elle components to synaptic vesicle markers [15]. Immunostaining
can be done in basically two ways, which are defined by way of
visualization method. One option visualizes proteins/antigens
with secondary antibodies that carry labels, which, upon enzymatic
reactions, lead to precipitates that are visible under the light micro-
scope. Typical examples are alkaline phosphatase or horseradish
peroxidase reactions that lead to brown, black, or purple precipi-
tates (for details, see ref. 16). These histochemical staining methods
have obvious limitations: enzymatic reactions do not penetrate
well into tissue, and resulting precipitates do not allow 3D recon-
structions unless the tissue is embedded in plastic, microdissected,
and subsequently scanned, which is laborious and time-consuming.
The advent of laser confocal and super-resolution microscopy
together with computer algorithms nowadays allows rapid and
reproducible 3D reconstructions of optical sections derived from
fluorescence immunostaining, which has thus become the method
of choice. It is based on secondary antibodies that are conjugated
6 Danielle C. Diaper and Frank Hirth

with various fluorochromes that, following excitation, emit at specific

wavelengths that are detected by the various optical imaging tech-
niques. Our protocol outlines the basics for this method.
First, the tissue must be carefully dissected from the whole
organism and fixed; the fixation process described here uses para-
formaldehyde; however, other fixation processes may be used
depending on the antibody used or the structures being visualized
(see Subheading 1.1). In the case of the embryo, the two protective
membranes, an outer shell called the chorion and a thick inner
vitelline membrane [17], must first be removed. This vitelline bar-
rier becomes permeable to the fixative agent when treated with a
fixative-heptane mix. The membrane is then removed by methanol
to allow the diffusion of the antibodies. Following fixation, the
tissue is blocked, usually with normal serum. This step reduces
binding of the antibody to nonspecific reactive sites of Drosophila
proteins. Blocking with normal serum from the species used to
generate the secondary antibodies is preferred. Either monoclonal or
polyclonal primary antibodies can be used for immunohistological
staining of Drosophila tissue.
Antibodies are raised against specific antigens, and their most
effective concentration should be determined by first carrying out
a dilution series on embryos. Occasionally, primary antibodies are
already conjugated to a fluorophore, meaning that secondary anti-
bodies are not needed to visualize the antigen’s location. When
choosing fluorophores for multiple antibody stainings, you should
bear in mind what filters/lasers are available for imaging the brains
and any crossover in the excitation or emission wavelengths [18].
Separating each incubation step is a series of washes. Thorough
washing is essential to prevent antibodies interacting with surplus
fixative and to clear the sample of residual antibodies. This immu-
nohistological protocol can also be applied to adult brain tissue as
well as to larval imaginal discs.

1.1 Troubleshooting Although now a relatively standard technique, elements of this

protocol may benefit from tweaking, depending on what you are
hoping to visualize and the antibodies you use. If you have persis-
tent trouble in visualizing your proteins, you may want to try the
following suggestions.

1.1.1 Alter the Fixation Formaldehyde is suitable for deep penetration of the tissue; it
Technique forms strong cross-links between proteins and is suitable for long-
term storage [19] and preserving chromosome morphology [20].
You may want to consider replacing the paraformaldehyde with
freshly prepared formaldehyde solution. Glutaraldehyde may also
be used [21, 22]; however, as it is a larger molecule, it does not
diffuse as well through deep tissue. Its cross-links span a larger
distance, so it can stably fix proteins that are further apart.
Glutaraldehyde fixation is not ideal for immunohistochemistry as
further treatment of the tissue is necessary to avoid aldehyde
 Immunostaining of Embryonic and Larval Drosophila Brain 7

groups binding to the antibody [23]. Embryos may be fixed with

methanol; while suitable for studying embryo morphology, this
method shows poor preservation of cytoplasmic proteins [24].
A more complex technique is cryofixation [25]. With all these
techniques, inadequate or over-fixation may occur, so it may be
useful to slightly increase or decrease the fixation period.

1.1.2 Antibody Too high or low concentrations of antibody may cause either
Optimization antibody aggregation on the surface of the tissue or poor penetra-
tion of the tissue [26]; therefore, carrying out a dilution series will
help to determine the most effective titer. To improve the signal-
to-noise ratio, try using affinity-purified antibodies or altering the
temperature, e.g., room temperature instead of 4 °C, or length of
antibody incubation, e.g., 3 h instead of overnight [22].

2 Materials

Make all stock buffers in a sterile, sealable bottle. Prepare solutions

with distilled H2O (dH2O). To avoid contamination of the stock
solutions, keep individual 50 ml aliquots of PBS, PBL, and PBT
and use these for dissections. These solutions are prone to con-
tamination; check for wispy or cloudy cultures that may form after
around 1 month of storage and dispose of contaminated solutions.
All solutions should be stored at 4 °C unless stated otherwise.

2.1 Solutions 1. PEM: 100 mM PIPES, 2 mM EGTA, 1 mM MgSO4. To 800 ml

for Embryo dH2O, add 34.63 g PIPES, 0.76 g EGTA, and 0.12 g MgSO4.
Preparations Adjust to pH 7 with HCl. Make up to final volume of 1 l
with dH2O.
2. PEM-FA: Add 1 ml 37 % formaldehyde solution to 9 ml PEM.
Make fresh. Do not store.
3. PBT 0.1 %—for embryo and early larval stages (L1 and L2):
Add 0.2 g BSA and 0.2 ml Triton X-100 to 200 ml PBS. Store
at 4 °C (see Note 1).
4. 5 % PBT-NGS: Add 1 ml normal goat serum (Invitrogen) to
19 ml PBT 0.1 %.
5. 50 % sodium hypochlorite.
6. Heptane.
7. Methanol.

2.2 Solutions 1. PBL: Dissolve 1.8 g of lysine HCl in 50 ml dH2O, adjust pH

for Larval Preparations to 7.4 by adding x ml of 0.1 M NaH2PO4 (see Note 2), and
adjust volume to 100 ml 0.1 M PBS. Store at 4 °C.
2. PBS 0.1 M buffer: Make 0.1 M Na2HPO by dissolving 8.52 g
in 600 ml dH2O. Make 0.1 M NaH2PO4 by dissolving 2.4 g
in 200 ml dH2O. Take 500 ml of 0.1 M Na2HPO and adjust
8 Danielle C. Diaper and Frank Hirth

pH to 7.4 by adding x ml of 0.1 M NaH2PO4 (see Note 2).

Store at 4 °C.
3. PBT 0.5 %—late larval stage (L3): Add 1 ml Triton X-100 to
200 ml PBS. Store at 4 °C (see Note 1).
4. 10 % PBT-NGS: Add 1 ml normal goat serum (Invitrogen) to
9 ml PBT 0.5 %.
5. 8 % PFA: Caution—make in a fume hood. In a sealable 50 ml
container, dissolve 1.6 g paraformaldehyde in 20 ml dH2O and
add 140 μl 1 M NaOH. Place in 37 °C water bath, vortexing
occasionally until completely dissolved. Aliquot 400 μl into
2 ml tubes and store at −20 °C (see Note 3).
6. PLP: Mix three parts PBL (1.2 ml) to 1 part 8 % PFA (400 μl).
Make fresh. Do not store (see Note 3).

2.3 Mounting 1. VECTASHIELD mounting medium with or without DAPI (see

Medium and Note 4) (Vector Laboratories, Burlingame, CA).
Antibodies 2. Antibodies: Developmental Studies Hybridoma Bank supplies
a large range of Drosophila-specific primary antibodies.
Secondary antibodies should be selected based on the animal
in which the primary antibody was raised, i.e., if you are using
rat-Elav, then you will need something like a goat anti-rat 488
secondary antibody.

2.4 Equipment 1. Microscope (larval preps).

2. Nylon mesh (embryo preps).
3. Vertical rotator.
4. Glass pipette (embryo preps).
5. Small paintbrush (embryo preps).
6. Rocker.
7. Fine tip forceps and sharpening stone (Dumont no. 5 twee-
zers, superfine, straight tip) (larval preps).
8. Glass microscope slides, SuperFrost (Thermo Scientific
Gerhard Menzel).
9. Cover slips: 22 × 50 mm, 0.13–0.17 thickness (embryo preps);
22 × 22 mm, 0.13–0.17 thickness (larval preps).
10. Silicone grease or petroleum jelly (larval preps).
11. Modeling clay (such as Plasticine) (embryo preps).
12. Dissection watch glass (larval preps).
13. Pin holder and stainless steel minutien pins 0.2 mm diameter
(InterFocus Ltd, Fine Science Tools) (larval preps).
14. Timer.
 Immunostaining of Embryonic and Larval Drosophila Brain 9

3 Methods

To be carried out at room temperature unless otherwise stated.

3.1 Embryo Fixation 1. Recover plates from egg collection and remove any remaining
yeast or dead flies with a spatula or brush being careful not to
damage the agar (see Note 5).
2. Dechorionate the embryos by adding 50 % hypochlorite and
agitate for 2–5 min until dechorionated embryos float to the
surface (see Note 6).
3. Pour the embryos and hypochlorite through the nylon mesh
and rinse thoroughly with dH2O (see Note 7) (see Fig. 2).
4. Transfer embryos to a 2 ml tube containing 1 ml heptane and
1 ml PEM-FA (see Note 8).
5. Agitate on a rotator at high speed for 10–30 min (no longer
than 30 min).
6. Allow embryos to settle—they should be at the interface
between the two phases.
7. Remove the lower phase, then the upper phase (see Note 9).

Fig. 2 Embryo collection apparatus. An open-bottom container (a) has a hole cut
in the lid (b); the hole is closed by a fine mesh (c), which is included into the lid
when screwed on top of the container. A 50 ml plastic container can also be
adapted in this way
10 Danielle C. Diaper and Frank Hirth

8. Replace with another 1 ml heptane. Then add 1 ml 100 %

methanol and shake vigorously for 1–2 min. The devitellinized
embryos will fall to the bottom of the tube.
9. Remove both phases making sure to remove all vitelline debris
(see Note 10).
10. Quickly add 1.5 ml of 100 % methanol and agitate on rotator
for 2 × 5 min, then 1 × 30 min (see Note 11).
11. Replace with fresh methanol and store at −20 °C or rehydrate
for immunostaining.

3.2 Embryo 1. Transfer embryos into a 1.5 ml tube (see Note 12).
Immunostaining 2. Rehydrate the embryos by removing the methanol and washing
2 × 5 min, then 1 × 30 min with PBT 0.1 %.
3. Incubate embryos for 30 min in 5 % PBT-NGS.
4. Remove PBT-NGS and add appropriate amount of primary
antibody diluted in 5 % PBT-NGS to a reaction volume of
100 μl (see Note 13). Incubate overnight at 4 °C.
5. Remove antibody solution and wash 1 × 1 min, 3 × 5 min, and
4 × 30 min with PBT 0.1 % (see Note 11).
6. Incubate embryos for 30 min in 5 % PBT-NGS.
7. Remove PBT-NGS and add appropriate amount of secondary
antibody diluted in 5 % PBT-NGS to a reaction volume of
100 μl. Incubate at 4 °C over night in the dark (see Note 14).
8. Remove antibody solution and wash 3 × 5 min and 4 × 30 min
with PBT 0.1 % (see Note 11).
9. Add 1 drop of VECTASHIELD mounting medium and incu-
bate at 4 °C over night in the dark.
10. Mount on glass slide with 22 × 50 mm cover slip (see Note 15).
11. Your samples are now ready for image acquisition (see Note 16).

3.3 Larval 1. Collect larvae and place into a watch glass containing cold PBS.
Immunostaining 2. Fill a 0.5 ml tube with cold PBS and keep on ice with the
lid open.
3. Roughly remove the larval CNS (see Note 17), placing dissected
brains into the PBS-containing 0.5 ml tube.
4. After 30 min of dissection, remove PBS and add 500 μl of PLP.
Agitate on a rotator for 1 h at room temperature.
5. Remove PLP and wash 3 × 10 min in PBT (see Note 18).
6. Block by incubating the brains in 5 % PBT-NGS (for L1 and L2)
or 10 % PBT-NGS (for L3) on the rotator for 15 min.
7. Remove PBT-NGS and add appropriate amount of primary
antibody diluted in 5 or 10 % PBT-NGS to a reaction volume
 Immunostaining of Embryonic and Larval Drosophila Brain 11

Fig. 3 Immunostaining of whole mount embryonic CNS and 3rd instar larval CNS. Top row: The embryonic
nervous system is staining with horseradish peroxidase Cy3 (HRP-cy3; red) and visualized under a standard
fluorescence microscope. The anterior brain is visible (arrow) as well as the ladderlike axon tracts of the
ventral nerve cord (arrowhead); compare to Fig. 1. The nuclei of all embryonic cells are visualized with DAPI.
Merge shows co-labeling of HRP-cy3 and DAPI. Bottom row: 3rd instar larval CNS immunostained with anti-
Bruchpilot counterlabeled with goat anti-mouse 568, and nuclei are highlighted with DAPI; merge shows co-
labeling of nc82 and DAPI. The nc82 antibody recognizes the Bruchpilot protein which is specifically enriched
in the active zone of synapses

of 100 μl (see Note 13). Incubate overnight (unless alternative

duration is specified by antibody manufacturer) at 4 °C.
8. Remove antibody solution and wash 1 × 1 min, then 3 × 20 min
in PBT.
9. Add appropriate amount of secondary antibody diluted in 5 or
10 % PBT-NGS to a reaction volume of 200 μl. Incubate for
2–3 h on a rocker in the dark (see Note 14).
10. Wash brains 2 × 15 min in PBT, then 2 × 15 min in PBS
(see Note 19).
11. Add ~2 drops of VECTASHIELD mounting medium and
incubate overnight at 4 °C in the dark.
12. Fine dissect in mounting medium on microscope slide under
microscope (see Note 20). Remove debris, arrange brains, and
make a small dab of silicone grease or petroleum jelly at four
corners where the cover slip will go. Gently lower the 22 × 22 mm
cover slip in to place and seal with nail varnish (see Note 21).
13. Store in a microscope slide box in the dark at 4 °C (see Note 22).
14. Your samples are now ready for image acquisition (see Note 16).
Examples of successful fixation, labeling, and imaging under a
standard fluorescence microscope are given in Fig. 3.
12 Danielle C. Diaper and Frank Hirth

4 Notes

1. Triton X-100 is very viscous, so pipette slowly to make sure

you take up the right amount. Mix by using a slow setting on
a magnetic stirrer to avoid an abundance of bubbles.
2. Place the solution on a magnetic stirrer while measuring the
pH and use a 10 ml pipette to add small amounts of NaH2PO4
to the solution.
3. Before you start the dissections, place a frozen aliquot of PFA
in a 37 °C water bath. By the time you have finished the dissec-
tion and are ready to fix the tissue, the PFA should be fully
dissolved and you can add the PBL directly to it.
4. DAPI (4,6-diamidino-2-phenylindole) is a nuclear stain that
binds to double-stranded DNA and fluoresces when excited
with a mercury arc lamp or UV. Even when not examining the
nuclei, it is useful to use the consistent DAPI staining pattern
to help orientate around the embryonic and larval brains.
5. Embryos can be collected by placing fruit agar plates (12.5 g
sugar, 21.25 g agar, 750 ml dH2O; autoclave in large flask; add
250 ml apple juice, 2 g nipagen; stir and pour into 55 mm petri
dishes; allow to cool; store at 4 °C for 1 month) that have a
smear of yeast paste in the middle, on top of a bottle of flies.
Keep the bottle, which should have holes pierced in the bottom
for air circulation, inverted in the dark and replace the agar
plate as necessary.
6. You can use 50 % household bleach instead of hypochlorite but
make sure that it hasn’t gone off—yellow bleach is still in date.
7. Wash agar plates with water and pour this through the mesh
× 2. Hold the nylon mesh over a waste beaker and use a water
bottle to rinse thoroughly. You can test for remaining bleach
by dabbing it on a colored paper towels (e.g., blue roll); if any
bleach is present, it will change the paper color. Residual bleach
will interfere with the fixation process. If you plan on doing a
lot of embryo stainings, it is worth trying to make a straining
tool out of a small glass jar with a screw lid and the bottom
removed (see Fig. 2).
8. Hold the mesh with forceps and dip into the 2 ml tube to
transfer the embryos and use the brush to transfer any embryos
stuck to side of straining tool (if you have one).
9. Use a glass pipette as embryos are less likely to adhere to
glass than to a plastic pipette. Remove as much of the
solutions as possible—this may mean pipetting off some of
your embryos too.
10. Remove the layers by pipetting from the interface of the two
phases.
 Immunostaining of Embryonic and Larval Drosophila Brain 13

11. For each wash, remove the old solution and add the same
amount of fresh solution. Washes should be agitated on a rotator
at full speed.
12. Pipette as fast as you can; as soon as the liquid stops being
turbulent, the embryos tend to stick to the sides of the pipette
and are very difficult to remove. This means you should have
your two tubes ready, with lids open and suck up the liquid and
embryos and transfer as quickly as possible.
13. Depending on what you are attempting to visualize, you can
add several primary antibodies at once. Take care to avoid using
similar species that may cross-react, for example, if you use a
mouse and rat primary antibody, the secondary anti-mouse will
bind to the rat antibody and the secondary anti-rat will bind to
the mouse antibody. If you are visualizing an endogenous GFP
signal (e.g., using the GAL4>UAS system), you may get a better
signal by also using an antibody against GFP and a secondary
with a 488 (green) fluorophore. Some primary antibodies are
already conjugated to a fluorophore and do not need a secondary
antibody, such as horseradish peroxidase Cy3 (HRP-cy3), which
specifically labels the Drosophila nervous system.
14. You will need a secondary antibody that targets the animal
species in which the primary antibody was raised. You can use
up to four separate channnels by picking fluorophores that are
unlikely to interfere with each other, e.g., DAPI, UV; 488,
green; 568, red; and 647, far red. Of course, the fluorophore
you use will depend on what filters, lamps, or lasers are avail-
able for image capture.
15. Suck up VECTASHIELD and embryos and pipette out in a
wide zigzag line on the labeled microscope slide (see Fig. 7a).
Roll 4 small balls of modeling clay and place in the corners
where the cover slip will go. Lower one short end of the cover
slip in place and gently lower the cover slip, avoiding air bub-
bles. Gently press cover slip at the 4 corners.
16. To avoid bleaching the brains, first use a fluorescent microscope
to select your best brains or embryos. When imaging using a
laser scanning microscope, scan each channel sequentially to
avoid interference between fluorophores. If you have issues with
the noise-to-signal ratio, check that you are using the antibody
at the appropriate dilution, ensure you are completing all wash
steps, and check the pH of the buffer solutions.
17. This bit takes practice! There are several ways to pull the CNS
from the larvae, but the thing that will help the most is being
able to differentiate the CNS from the imaginal discs and fat
bodies (see Figs. 4 and 5). I find it easier to gently clamp the
body of the larvae about two thirds the way towards the mouth
hooks; then when the larvae extends its mouth hooks, grab
14 Danielle C. Diaper and Frank Hirth

Fig. 4 Cartoon illustrating a rough larval CNS dissection. (1) First grab the
extended mouth hooks and make a gentle tear in the cuticle. (2) Rip the larvae in
half. (3) Holding the mouth hooks again, slowly pull off the remaining cuticle until
(4) you are left with the larval brain still attached to the head, imaginal discs, and
fat bodies

Fig. 5 Identifying the 3rd instar larval brain and CNS. (a) Several imaginal discs
(filled arrow), fat bodies (arrowhead), and digestive components (empty arrow)
will be attached to the 3rd instar larval brain and CNS. Identifying the brain
among these other tissues will aid successful dissections. The larval brain and
CNS are outlined with a black line. (b) DAPI labeled 3rd instar larval brain and
CNS visualized under the fluorescence microscope after successful dissection;
note that wing and leg imaginal discs are still attached

them with the forceps (see Fig. 4). Gently pull the mouth hooks
until the cuticle rips and the internal matter is exposed.
The round optic lobes of the CNS may be visible. Place both
forceps on the body and rip the lower half of the larvae off.
Grab the mouth hooks and body again and gently pull until
the mouth hooks and CNS come away from the rest of the
larvae. Holding onto the mouth hooks, transfer to the 0.5 ml
tube that is sat on ice. If you do not have much luck this way,
or find that the ventral nerve cord is lost using this method,
you can also pinch both forceps just below the mouth hooks
(by the scruff of its neck!) and rip the cuticle open and then bit by
bit remove the lower part of the larvae, the digestive tract, fat,
and some imaginal discs.
 Immunostaining of Embryonic and Larval Drosophila Brain 15

Fig. 6 Cartoon illustrating fine dissection of the larval brain and CNS. (1) Using
the dissecting pins like a knife and fork, remove the mouth hooks from the brain
and then (2) gently remove all other tissues taking care (3) not to damage or
puncture the brain

18. When washing it is best to leave brains to settle for 15 s and then
pipette off the solution. Eject the pipetted solution into a clean
watch glass, check under the microscope, and rescue any brains
by pipetting them back into the tube. If carrying out several
dissections, you can leave samples washing for an extended
period of time until all preparations have reached the same stage.
19. As the secondary antibodies are photosensitive, you should
keep your preparation in the dark as much as possible to avoid
bleaching the fluorophores. Cover the tubes with tinfoil or
repurpose a small container to hold your tubes and attach to
the rotator using Velcro strips.
20. Cut around 2 cm off the end of a P200 pipette tip and transfer
brains and VECTASHIELD to a microscope slide. Label the
microscope slide with a dissection code (for cross-reference
with your records, e.g., Initials_01), genotype, the primary
antibody and secondary fluorophore used, and the date.
Remove the majority of the VECTASHIELD, so that the
brains are no longer floating, and pipette back into your dissec-
tion tube. Use the pin holder tools to remove unwanted imagi-
nal discs, fat bodies, and mouth hooks. You can use them like a
knife and fork to slice away tissue, or pin an imaginal disc into
place with one needle and cut the connecting tissue with the
other (see Fig. 6). Take care not to pin or damage the CNS itself.
21. Arrange brains in concentric circles with the ventral nerve cord
pointing outwards (see Fig. 7b). Orientate the brains in a mix-
ture of dorsal or ventral side up. Add a small amount of
VECTASHIELD in a circle around the brains—not too much
or they will float out of position. Once the cover slip is in place,
gently press on the four corners to flatten the brains slightly, and
be careful as they are easily ruptured by squashing. Add more
VECTASHIELD if necessary by pipetting small amounts at the
edge of the cover slip. Remove excess VECTASHIELD with
cotton buds or tissues dipped in ethanol—again, be very careful
not to dislodge the cover slip as this will destroy your brains!
16 Danielle C. Diaper and Frank Hirth

Fig. 7 Cartoon illustrating slide preparation. (a) Pipette embryos in VECTASHIELD

mounting medium in a wide zigzag line onto labeled microscope slide, place
small balls of modeling clay where the corners of the cover slip will be, and
gently lower the cover slip (arrow) onto the embryos in VECTASHIELD; make sure
to avoid air inclusions. (b) Arrange the larval brains/CNS in a rosette onto labeled
microscope slide, place small dabs of silicone grease (or petroleum jelly) at the
corners of where the cover slip will be, and gently lower the cover slip (arrow);
make sure to avoid air inclusions

22. A standard 4 °C fridge is a good place to store your slide box.

Cold rooms often have a damp atmosphere, which can cause
the nail varnish to peel off.

Acknowledgements

This work was supported by grants from the UK Medical Research

Council (G070149), Royal Society (Hirth2007/R2), Parkinson’s
UK (G-0714), Motor Neurone Disease Association (Hirth/
Mar12/6085, Hirth/Oct07/6233), Alzheimer Research UK
(Hirth/ARUK/2012), and the Fondation Thierry Latran
(DrosALS to F.H.).

References
1. Skeath JB, Thor S (2003) Genetic control of development in Drosophila. Bioessays 26:
Drosophila nerve cord development. Curr Opin 739–751
Neurobiol 13:8–15 4. Hirth F, Reichert H (1999) Conserved genetic
2. Kim DW, Hirth F (2009) Genetic mechanisms programs in insect and mammalian brain devel-
regulating stem cell self-renewal and differen- opment. Bioessays 21:677–684
tiation in the central nervous system of 5. Lee T, Luo L (2001) Mosaic analysis with a
Drosophila. Cell Adh Migr 3:402–411 repressible cell marker (MARCM) for Drosophila
3. Urbach R, Technau GM (2004) Neuroblast neural development. Trends Neurosci 24:
formation and patterning during early brain 251–254
 Immunostaining of Embryonic and Larval Drosophila Brain 17

6. Grueber WB, Ye B, Yang CH, Younger S, New York, NY, For an amended and updated
Borden K, Jan LY, Jan YN (2007) Projections version, follow the link: http://patelweb.
of Drosophila multidendritic neurons in the berkeley.edu/Images/Protocols/pdf%20files/
central nervous system: links with peripheral Antibody%20Methods%202006.pdf
dendrite morphology. Development 134: 17. Ashburner M (1989) Drosophila: a laboratory
55–64 manual. Cold Spring Harbor Laboratory Press,
7. Selcho M, Pauls D, Han KA, Stocker RF, New York, NY
Thum AS (2009) The role of dopamine in 18. Hoffman, G. (2008) Seeing is believing: Use
Drosophila larval classical olfactory condition- of antibodies in immunohistochemistry and in
ing. PLoS One 126:e5897 situ hybridization. In: Short course II of SfN’s
8. White KE, Humphrey DM, Hirth F (2010) 38 annual meeting: 15–19 November 2008;
The dopaminergic system in the aging brain of Washington, DC. Society for Neuroscience
Drosophila. Front Neurosci 4:205 19. Rothwell WF, Sullivan W (2000) Fluorescent
9. Muqit MK, Feany MB (2002) Modelling neu- analysis of Drosophila embryos. In: Sullivan W,
rodegenerative diseases in Drosophila: a fruitful Ashburner M, Hawley RS (eds) Drosophila
approach? Nat Revs Neurosci 3:237–243 protocols. Cold Spring Harbor Laboratory
10. Koizumi K, Higashida H, Yoo S et al (2007) Press, New York, NY, p 141
RNA interference screen to identify genes 20. Bonaccorsi S, Giansanti MG, Cenci G, Gatti M
required for Drosophila embryonic nervous (2012) Formaldehyde fixation of Drosophila
system development. Proc Natl Acad Sci USA testes. Cold Spring Harb Protoc. doi:10.1101
104:5626–5631 21. Heimbeck G, Bugnon V, Gendre N, Häberlin
11. Stochmanski SJ, Therrien M, Laganière J et al C, Stocker RF (1999) Smell and taste percep-
(2012) Expanded ATXN3 frameshifting events tion in Drosophila melanogaster larva: toxin
are toxic in Drosophila and mammalian neuron expression studies in chemosensory neurons.
models. Hum Mol Genet 21:2211–2218 J Neurosci 19:6599–6609
12. Dunlop J, Morin X, Corominas M et al (2004) 22. Stocker RF, Heimbeck G, Gendre N, de Belle
glaikit is essential for the formation of epithelial JS (1997) Neuroblast ablation in Drosophila
polarity and neuronal development. Curr Biol P[GAL4] lines reveals origins of olfactory
14:2039–2045 interneurons. J Neurobiol 32:443–456
13. Tsuji T, Higashida C, Yoshida Y et al (2011) 23. Hassell J, Hand AR (1974) Tissue fixation with
Ect2, an ortholog of Drosophila’s pebble, nega- diimidoesters as an alternative to aldehydes. I.
tively regulates neurite outgrowth in neuroblas- Comparison of cross-linking and ultrastructure
toma × glioma hybrid NG108-15 cells. Cell Mol obtained with dimethylsuberimidate and glu-
Neurobiol 31:663–668 taraldehyde. J Histochem Cytochem 22:
14. Pandey UB, Nichols CD (2011) Human disease 223–229
models in Drosophila melanogaster and the 24. Wieschaus E, Nüsslein-Volhard C (1998)
role of the fly in therapeutic drug discovery. Looking at embryos. In: Roberts DB (ed)
Pharmacol Rev 63:411–436 Drosophila, a practical approach. Oxford
15. Developmental Studies Hybridoma Bank University Press Inc, New York, NY, p 205
developed under the auspices of the NICHD 25. Ripper D, Schwarz H, Stierhof YD (2008)
and maintained by The University of Iowa, Cryo-section immunolabelling of difficult to
Department of Biology, Iowa City, IA 52242 preserve specimens: advantages of cryofixation,
http://dshb.biology.uiowa.edu/ freeze-substitution and rehydration. Biol Cell
16. Patel N (1994) Imaging neuronal subsets and 100:109–123
other cell types in whole mount Drosophila 26. Rebay I, Fehon R (2000) Generating antibod-
embryos and larvae using antibody probes. In: ies against Drosophila proteins. In: Sullivan W,
Goldstein LSB, Fryberg E (eds) Methods in Ashburner M, Hawley RS (eds) Drosophila
Cell Biology, Vol 44. Drosophila melanogaster: protocols. Cold Spring Harbor Laboratory
Practical Uses in Cell Biology. Academic, Press, New York, NY, p 400
 Chapter 2

Non-fluorescent RNA In Situ Hybridization Combined

with Antibody Staining to Visualize Multiple Gene
Expression Patterns in the Embryonic Brain of Drosophila
David Jussen and Rolf Urbach

Abstract
In Drosophila, the brain arises from about 100 neural stem cells (called neuroblasts) per hemisphere which
originate from the neuroectoderm. Products of developmental control genes are expressed in spatially
restricted domains in the neuroectoderm and provide positional cues that determine the formation and
identity of neuroblasts. Here, we present a protocol for non-fluorescent double in situ hybridization com-
bined with antibody staining which allows the simultaneous representation of gene expression patterns in
Drosophila embryos in up to three different colors. Such visible multiple stainings are especially useful to
analyze the expression and regulatory interactions of developmental control genes during early embryonic
brain development. We also provide protocols for whole mount and flat preparations of Drosophila embryos,
which allow a more detailed analysis of gene expression patterns in relation to the cellular context of the early
brain (and facilitate the identification of individual brain neuroblasts) using conventional light microscopy.

Key words Drosophila, Embryonic brain, Neuroectoderm, Neuroblast identification, In situ hybrid-
ization, Antibody staining

1 Introduction

In situ hybridization and immunohistochemistry are fundamental

and widely used methods in the field of developmental biology.
They serve multiple purposes, such as the identification of a
specific cell type or tissue by the visualization of marker genes,
or the study of genetic interactions by the comparative analysis
of gene expression patterns in different genetic backgrounds.
Huge progress has been made in that field in the last decades.
The invention of fluorescence-imaging techniques allows the
detection of gene products in deeper layers of tissue, at consis-
tently increasing resolutions. This also led to the development
of staining protocols involving different fluorescent dyes, which
allow the detection of multiple gene products at once [1].

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_2, © Springer Science+Business Media, LLC 2014

19
20 David Jussen and Rolf Urbach

However, fluorescence-imaging methods have limitations.

Those concern, for example, the relatively low stability of fluo-
rescence dyes hampering the long-term storage of specimens or
the need of costly documentation systems.
Here, we present a protocol for non-fluorescent double in situ
hybridization combined with antibody staining on Drosophila whole
mount embryos, which allows the simultaneous visualization of
transcripts and proteins in up to three different colors. This method
combines the benefits of simultaneously labeling multiple factors at
once, like it is popular in fluorescence microscopy, while allowing
the easy handling of traditional, non-fluorescent labeling tech-
niques. That way, a high amount of specimens can be analyzed
under a conventional light microscope, without any need of fluores-
cent detection systems. Using differential interference contrast (i.e.,
Nomarski), histochemically labeled cells are visualized in context of
the surrounding (uncolored) tissue, which often simplifies interpre-
tation of gene expression patterns and their relation to specific cells
without the need of additional markers (as, e.g., membrane markers
used in fluorescence microscopy to visualize cell boundaries).
This protocol is based on the sequential detection of differen-
tially labeled RNA probes via the alkaline phosphatase (AP)-
catalyzed turnover of different chromogenic compounds. The
precipitates formed in those reactions are insoluble in aqueous
solution, which allows the elution of AP-bound antibodies after
the chromogenic reaction without affecting the staining per se.
By that, the AP-coupled antibody directed against the first probe
can be eluted after the first chromogenic reaction, with the subse-
quent introduction of an AP-coupled antibody directed against the
second probe. Here, we use NBT/BCIP and Vector Red as chro-
mogenic compounds to sequentially detect DIG- and FITC-
labeled probes, in combination with a subsequent antibody staining
which is visualized by the peroxidase-catalyzed turnover of DAB.
That way, three different gene products are visualized by blue, red,
and brown precipitates, respectively.
Such stainings are particularly useful to analyze the expression
and regulatory interactions of developmental control genes during
the early period of embryonic brain development. In organisms as
diverse as insects and mammals, the brain arises from multipotent
neural stem cells (in insects called neuroblasts) which originate
from the neuroectoderm. Products of developmental control genes
are expressed in spatially restricted domains in the neuroectoderm
and provide positional cues that regulate the development of neu-
roblasts [2, 3]. Therefore, the expression of these genes has to be
precisely controlled. Recently, we uncovered a novel gene regula-
tory network in Drosophila, in which “DorsoVentral” and
“AnterioPosterior” patterning genes interact to precisely control
their spatially restricted expression in the brain neuroectoderm,
which is essential for the correct formation and fate specification of
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 21

neuroblasts [4–6]. These patterning genes display not only a high

degree of intersegmental modulation but also considerable changes
in their expression patterns over time. The staining technique pre-
sented here has been used to visualize multiple patterning genes at
once in order to analyze their dynamic expression patterns in rela-
tion to each other and to uncover their genetic interactions and
cellular functions. Non-fluorescent multiple stainings are also suit-
able to identify embryonic brain neuroblasts. The brain of
Drosophila is built up by about 100 neuroblasts, which develop in
a stereotypic pattern during the first third of embryonic develop-
ment [7]. Brain neuroblasts can be identified individually by posi-
tion, time point of formation, morphology, and the specific
combination of marker genes they express [8]. As the identified
marker genes are usually not selective enough, it is often necessary
to stain for a combination of those in order to identify individual
brain neuroblasts. Using this protocol, marker gene expression can
be analyzed using conventional Nomarski optics. That way, labeled
as well as unlabeled neuroblasts are visible alongside each other,
which allows the identification of individual neuroblasts with the
use of only a few molecular markers.

2 Materials

2.1 Reagents 1. Anti-digoxigenin (DIG)-AP, Fab fragments from sheep

(Roche Applied Science).
2. Anti-fluorescein (FITC)-AP, Fab fragments from sheep (Roche
Applied Science).
3. Methanol (≥99.9 %, p.a.).
4. ssDNA (10 mg/μl, MB grade, from fish sperm) (Roche
Applied Science).

2.2 Buffers 1. AP buffer (50 ml): 41.2 ddH2O, 1 ml 5 M NaCl, 2.5 ml 1 M

MgCl2, 5 ml 1 M Tris–HCl. Adjust pH to 9.5. Add 50 μl
Tween®20.
2. Blocking buffer (100 ml): 10 ml 1 M Tris–HCl, 3 ml 5 M NaCl.
Adjust pH to 7.5. Add 0.5 g blocking reagent (delivered with
TSA Biotin System, PerkinElmer). Store aliquots at −20 °C.
3. Glycine buffer (50 ml): 50 ml ddH2O, 0.038 g glycine, 1.46 g
NaCl. Adjust pH to 2.3. Add 0.05 g BSA and 50 μl Triton™
X-100.
4. Hybridization buffer (50 ml): 12.5 ml SSC (20×), 12.5 ml
DEPC-H2O, 25 ml formamide, 50 μl Tween®20.
5. PBS (20× stock solution, 500 ml): 500 ml ddH2O, 75.97 g
NaCl, 9.94 g Na2HPO4, 4.14 g NaH2PO4. Adjust pH to 7.4.
Dilute with ddH2O to obtain 1× working solution.
22 David Jussen and Rolf Urbach

6. PBT (0.3 %, 50 ml): 47.5 ml ddH2O, 2.5 ml PBS (20×),

150 μl Triton™ X-100.
7. PBTween (0.1 %, 50 ml): 2.5 ml PBS (20×), 47.5 ml ddH2O,
50 μl Tween®20 (use DEPC-treated PBS and ddH2O for
PBTween-DEPC).
8. Vector Red buffer (100 ml): 100 ml distilled ddH2O, 1.21 g
Tris–HCl. Adjust pH to 8.2–8.5. Add 100 μl Tween®20.
9. Washing buffer (100 ml): 10 ml 1 M Tris–HCl, 7.5 ml 2 M
NaCl, 82.5 ml ddH2O. Adjust pH to 7.5. Add 50 μl Tween®20.

2.3 Solutions 1. ABC solution (avidin-biotinylated peroxidase complex): 1 ml

and Media PBT, 4 μl solution A, 4 μl solution B (both are delivered with
the Vectastain ABC Kit). Prepare solution 1 h before use and
keep on a shaker (100 rpm) at room temperature (RT).
2. AP staining solution: 1 ml AP buffer, 3 μl NBT stock solution,
1.5 μl BCIP stock solution. Prepare freshly before use.
3. Apple juice agar: 1,000 ml apple (or grape) juice, 28 g agar.
Heat until the apple juice agar solution starts boiling and
becomes clear. Dispense in fly culture vials or petri dishes (the
latter requires additional fly cages). Once the agar has cooled
and hardened, store at −4 °C.
4. BCIP stock solution: 10 ml dimethylformamide (100 %),
500 mg BCIP. Store 1 ml aliquots at −20 °C.
5. Chlorine bleach (6 %): Dilute sodium hypochlorite solution
(12 % Cl) 1:1 with H2O.
6. 3,3′-Diaminobenzidine tetrahydrochloride (DAB) stock solu-
tion: Dissolve 1 DAB tablet (10 mg; Sigma-Aldrich) in 35 ml
PBT. Store 400 μl aliquots at −20 °C.
7. DAB staining solution: 400 μl DAB stock solution, 600 μl
PBT, 2 μl H2O2. Prepare freshly before use.
8. DEPC-H2O: 1,000 ml H2O, 1 ml DEPC. Autoclave before use.
9. Fixative: 450 μl PBT, 70 μl formaldehyde (37 %), 600 μl
n-heptane.
10. Glycerol (70/90 %). Dilute glycerol with PBS to obtain the
particular concentration.
11. NBT stock solution: 10 ml dimethylformamide (70 %),
500 mg NBT. Store 1 ml aliquots at −20 °C.
12. Vector Red (VR) staining solution: 1 ml VR buffer, 16 μl VR
Substrate Solution 1, 16 μl VR Substrate Solution 2, 16 μl VR
Substrate Solution 3 (VR Substrate Solutions are delivered
with the Vector Red Alkaline Substrate Kit I). Vortex ~5 s
after addition of each VR Substrate Solution. Prepare freshly
before use.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 23

2.4 Kits 1. TSA™ Biotin System (PerkinElmer).

2. Vectastain ABC Kit (Standard) (Vector Labs).
3. Vector Red Alkaline Phosphatase Substrate Kit I (Vector Labs).

2.5 Equipment for 1. Aluminum foil (thickness 0.02–0.03 mm).

Staining, Preparation, 2. Preparation forceps (e.g., Dumont no. 5, Fine Science Tools).
and Mounting
3. Concavity slide.
4. Cover slips (18 × 18 mm, 22 × 22 mm, and 24 × 60 mm).
5. Microscope slides.
6. Minutien pins (stainless steel, 0.1 mm diameter, Fine Science
Tools).
7. Nail polish (transparent).
8. 2 nickel-plated pinholders (Fine Science Tools).
9. Nylon mesh (120 μm, e.g., from Merck Millipore).
10. Small scalpel.
11. Spot plate (white).
12. Weighing dishes (white, 41 × 41 mm).
13. Whetstone (e.g., Sharpening stone for Dumont forceps, Fine
Science Tools; alternatively abrasive paper, grain size
600–1,000).

2.6 Equipment 1. Cold light source (halogen or LED) equipped with fiber light
for Microscopy guides (e.g., KL1500 by Schott).
2. Digital microscope camera (e.g., ProgRes® series, Jenoptik).
3. Upright light microscope (equipped with differential interfer-
ence contrast and 40×–100× objectives, e.g., Axioscope Zeiss).
4. Stereo microscope (e.g., MZ series by Leica Microsystems).

3 Methods

General remarks: Wash steps and incubations are carried out at RT

upon shaking (100 rpm) with 1 ml of solution used, unless stated
differently. Reactions are carried out in standard 1.5 ml reaction
tubes (Eppendorf tubes).

3.1 Embryo 1. Place flies on apple juice agar for egg laying (see Note 1).
Collection and 2. Dechorionize embryos after egg laying by covering the apple
Dechorionization juice agar with chlorine bleach (6 %) for 3 min. Slightly rotate
once in a while. Dechorionized embryos will float up.
3. Collect embryos by transferring the chlorine bleach into an
egg basket (see Note 2).
4. Wash embryos with water.
24 David Jussen and Rolf Urbach

3.2 Fixation 1. Open the egg basket. Take embryos from the nylon mesh with
a small scalpel and transfer them into an Eppendorf tube with
fixative.
2. Fix embryos by vigorously shaking for 20 min (900 rpm).
3. Let the two phases of the fixative separate.
4. Remove the lower phase of the fixative without removing too
many embryos.
5. Add 500 μl methanol.
6. Remove the vitelline membrane by vortexing at maximum
speed for 2 min.
7. Let the devitellinized embryos sink down and remove lower
phase of the solution.
8. Add 500 μl methanol.
9. Vortex for 1 min.
10. Let the embryos sink down and remove as much of the solu-
tion as possible without removing too many embryos.
11. Rinse 4× with methanol.
● Embryos may be stored in methanol at −20 °C (see Note 3).

3.3 (Double) In Situ Use gloves, filtered tips, and DEPC-treated PBTween for the
Hybridization following in situ hybridization procedure:
1. Rinse 5× with PBTween.
2. Incubate 5 min in PBTween/hybridization buffer (1:1).
3. Incubate 5 min in hybridization buffer.
4. Perform prehybridization by incubating 1 h in hybridization
buffer + ssDNA (1:100) at 55 °C upon shaking (300 rpm).
5. Perform hybridization by incubating in hybridization buffer +
ssDNA (1:100) containing your probe (both of your probes
for double in situ hybridization) at the appropriate working
dilution(s) (see Note 4).
6. Incubate overnight at 55 °C upon shaking (300 rpm).
7. Incubate 30 min in hybridization buffer at 65 °C upon shaking
(350 rpm).
8. Incubate 30 min in PBTween/hybridization buffer (1:1)
upon shaking (350 rpm).
9. Wash 4 × 20 min with PBTween at 65 °C upon shaking
(350 rpm).
Standard tips and untreated PBTween may be used from
now on.
10. Wash 10 min with PBTween at RT.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 25

3.4 Detection 1. Incubate 1.5 h in anti-DIG-AP or anti-FITC-AP (1:1,000 in

of the First Probe PBTween) (see Note 5).
(via NBT/BCIP) ● The antibody can alternatively be incubated overnight at
4 °C upon slight shaking.
2. Rinse 3× and wash 3 × 10 min with PBTween.
3. Incubate 2 × 5 min in AP buffer.
● Prepare NBT/BCIP staining solution during that step.
4. Incubate in AP staining solution and transfer the solution con-
taining the embryos into a white weighing dish (41 × 41 mm).
Keep it cool and dark.
5. Judge the chromogenic reaction from time to time under a
stereo microscope (see Note 6).
6. When staining intensity has reached the desired level, transfer
the embryos back into the tube (see Note 7).
7. Rinse 3× with PBTween and terminate the reaction by incu-
bating 10 min in methanol.
8. Rinse 3× with PBTween.
● For double in situ hybridization combined with antibody
staining, continue with Subheading 3.5.
● For single in situ hybridization combined with antibody
staining, continue with Subheading 3.7.
● For double in situ hybridization without further antibody
staining, continue with Subheading 3.9.

3.5 Antibody Elution 1. Rinse 1× and incubate 3 × 10 min in glycine buffer.

and Inactivation 2. Rinse 3× with PBT.
of Residual AP
3. Incubate 5 min in PBT/formaldehyde (10:1).

3.6 Detection 1. Rinse 3× and wash 3 × 10 min with PBT.

of the Second Probe 2. Incubate overnight in anti-FITC-AP or anti-DIG-AP (1:1,000
(via Vector Red) in PBT; depending on the choice of the first antibody) at 4 °C.
3. Rinse 3× and wash 3 × 10 min with PBT.
4. Incubate 10 min in Vector Red buffer. Prepare Vector Red
staining solution during that step.
5. Incubate in Vector Red staining solution. Transfer the solu-
tion containing the embryos into a white weighing dish
(41 × 41 mm). Keep it cool and dark. Judge the chromogenic
reaction (in this case, a red colored staining) from time to time
under a stereo microscope (see Note 8).
6. Transfer embryos back into the tube.
7. Rinse 3× and wash 3 × 10 min with PBT.
8. Incubate 5 min in PBT/formaldehyde (10:1) (see Note 9).
26 David Jussen and Rolf Urbach

3.7 Incubation with 1. Rinse and wash 3 × 10 min in PBT.

Primary Antibody 2. Incubate overnight with primary antibody (at the appropriate
working dilution in PBT) at 4 °C (see Notes 10 and 11).
Depending on the quality of the antiserum, 2–4 h incubation
at RT may be suitable, too.
3. Rinse 3× and wash 3 × 10 min with PBT.

3.8 Detection 1. Incubate 2–3 h with biotin-coupled secondary antibody

of Primary Antibody (1:500 in PBT) RT.
(via DAB) 2. Optional: Continue with Subheading 3.10 for Tyramide
Signal Amplification (see Note 12). If not, prepare ABC before
the following step. ABC has to incubate for at least 1 h before
use upon shaking (100 rpm).
3. Rinse 3× and wash 3× 10 min with PBT.
4. Incubate 1 h in ABC.
5. Rinse 3× and wash 3 × 10 min with PBT. Prepare DAB stain-
ing solution during the last wash step.
6. Incubate in DAB staining solution. Transfer the solution con-
taining the embryos into a white weighing dish (41 × 41 mm).
Judge the chromogenic reaction under a stereo microscope.
The staining should emerge quickly (~ 1–15 min).
7. When the intensity of the brown DAB staining has reached the
desired level, transfer the embryos back into the tube.

3.9 Final 8. Rinse 3× and wash 10 min with PBT.

Washing Steps 9. Rinse 2× with PBS.
10. Store in 70 % glycerol.

3.10 Optional: 1. Rinse 1× and wash 1 × 10 min with PBT.

Tyramide Signal 2. Incubate 1 × 5 min and 1 × 30 min in blocking buffer.
Amplification
3. Incubate 30 min in Streptavidin-HRP (1:500 in blocking
buffer).
4. Wash 3 × 10 min with TSA washing buffer. Prepare ABC before
the following step. ABC has to incubate for at least 1 h upon
shaking (100 rpm) before use (see step 4 of Subheading 3.8).
5. Incubate 5 min in 140 μl TSA reagent (1:70 in amplification
diluent) (see Note 13).
6. Continue with step 3 of Subheading 3.8.

3.11 Manufacturing Two different kinds of needles are needed for the preparation of
of Preparation Needles embryos. One will be used for handling the embryos (i.e., moving in
glycerol, transferring onto microscope slide, holding during prepara-
tion; termed “handling needle”), while the other is used for cutting
the tissue during preparation (termed “cutting needle”) (Fig. 1a).
For that, the pinpoint of both needles will be processed differently.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 27

Fig. 1 Filet preparation of Drosophila embryos. (a) The cutting needle has a flattened tip with sharp edges,
while the tip of the handling needle is rounded. (b) Main steps performing filet preparation of embryos at dif-
ferent stages. The posterior part of the abdomen is removed by cutting the embryo transversally. Then, the
remaining ectoderm is cut along the dorsal midline towards the anterior tip of the embryo to open the head
capsule. Finally, the ectoderm has to be flattened. (c–e) Instructions for flat preparations. Black and white dots
indicate corresponding positions of the head capsule. Arrowheads indicate how to move the upper (small)
cover slip. Arrows indicate the flow of glycerol caused by moving the cover slip. (c) An embryo with opened
head capsule is positioned in glycerol between a large (24 × 60 mm) and a small (18 × 18 mm) cover slip
(separated by aluminum spacers) with anterior facing forwards and ventral facing down. (d) The small cover
slip is carefully moved in posterior direction (regarding the body axis of the embryo), which causes the hemi-
spheres to erect. (e) The small cover slip is carefully moved back until the hemispheres are unfurled flatly on
the cover slip. a anterior, p posterior, d dorsal, vML ventral midline, pNE procephalic neuroectoderm
28 David Jussen and Rolf Urbach

1. Lock the dull end of a minutien pin tightly in a pinholder.

2. Process the tip of the needle under a stereo microscope.
3. Place the needle horizontally on a solid steel ground (e.g., on
the flat end of a forceps).
4. Flatten the tip of the needle by placing the flat end of another
forceps on the tip of the needle, slowly streaking away from it
with some pressure (see Note 14).
5. Adjust both needles according to Fig. 1a with the aid of a
whetting stone or fine abrasive paper (see Note 15).

3.12 Initial 1. Transfer embryos into a well of a white spot plate.

Examination 2. Examine embryos under a stereo microscope using maximum
of Embryos magnification.
3. Sort embryos with the handling needle according to staining
quality, stage, and genotype of interest.
4. Examine selected embryos in more detail by making
whole mounts (Subheading 3.13) or filet preparations
(Subheading 3.14) for microscopy.

3.13 Whole Mount 1. Fix two cover slips (22 × 22 mm) side by side on a microscope
Preparation of slide with a drop of nail polish. The distance between both
Embryos should be approx. 15 mm.
2. Transfer selected embryos between the cover slips with a
pipette.
3. Place another cover slip (22 × 22 mm) on the embryos. Make
sure that it stays on top of both spacing cover slips.
4. Examine under a microscope (see Note 16; Fig. 2a, d)
5. Seal for documentation and storage by applying nail polish to
the edges of the cover slip

3.14 Filet The following filet preparation is recommended for embryos until
Preparation of Early to developmental stage 12 and is performed under a stereo micro-
Mid-Stage Embryos scope (see Note 17).
1. Fill the pit of a concavity slide with 70 % glycerol.
2. Transfer the selected embryo into the pit with the handling
needle.
3. Place the embryo on the lateral side and hold it carefully with
the handling needle.
4. Remove the posterior part of the abdomen and cut along the
dorsal midline from posterior to anterior with a cutting needle
to open the head capsule (Fig. 1b).
5. Spread the hemispheres slightly with both needles or a Dumont
forceps.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 29

Fig. 2 Non-fluorescent multiple stainings of the head neuroectoderm and brain neuroblasts in whole mounts
and flat preparations. (a) Ventral view on the head region of a stage 8 whole mount embryo (anterior is up)
stained for Nkx6-mRNA (via NBT/BCIP, blue) and single-minded protein (via DAB, brown). (b) Schematic repre-
sentation of the details presented in C/D′. (c) View on the head region (left hemisphere) of a filet prepared stage
6 embryo with almost all neuroectodermal cells in one plane: blue is ind-mRNA (via NBT/BCIP), magenta is
msh-mRNA (via Vector Red) and brown is Vnd-Protein (via DAB). (D/D′) Early brain neuroblast pattern of a late
stage 9 embryo as whole mount (d) and after filet preparation of the same embryo (D′), stained against Engrailed
(via NBT/BCIP, blue) and svp-LacZ (via DAB, brown). Note that filet preparation allows a more clear view, and
thus the identification of individual neuroblasts by marker expression and relative position (as illustrated in D″)
according to [7, 8]. a anterior, p posterior, v ventral, d dorsal, CF cephalic furrow, vML ventral midline, hs
engrailed head spot, as engrailed antennal stripe, is engrailed intercalary stripe. D-D″ Adapted from [7]

6. Remove excess yolk carefully with the cutting needle while

holding the embryo with the handling needle.
7. Arrange two spacers (i.e., strips of aluminum foil, approx.
2 × 15 mm) in parallel at a distance of 10–12 mm in the middle
of a large cover slip (24 × 60 mm).
8. Add 45 μl glycerol (90 %) between the aluminum spacers.
9. Transfer the prepared embryo with the handling needle into
the glycerol drop and push it carefully to the ground, with
anterior forwards and ventral downwards.
10. Place a small cover slip (18 × 18 mm) on the glycerol drop.
And wait until the glycerol is spread between the two cover
slips (Fig. 1c).
30 David Jussen and Rolf Urbach

11. Carefully pull the small cover slip (e.g., with a forceps) back-
wards (i.e., towards yourself). This will cause the hemispheres
to erect (Fig. 1d).
12. Push the small cover slip forwards (i.e., away from yourself).
By that, glycerol will float towards the erected hemispheres
causing them to unfurl flatly on the cover slip (Fig. 1e; see
Note 18).
13. When the embryo is appropriately flattened, fix the small cover
slip by applying a drop of nail polish to every corner. Let the
nail polish dry, remove excess glycerol (e.g., with small stripes
of tissue paper), and seal the cover slip by applying nail polish
to the edges (two times).
14. Clean the cover slip from glycerol and fix it on a microscope
slide with small stickers.
15. Under a microscope, the filet preparation can be examined
from both sides (with a view from ventral or dorsal) by turning
the cover slides including the filet preparation (Fig. 2c, D′;
see Note 19).

3.15 CNS 1. Fill the pit of a concavity slide with 70 % glycerol.

Preparation of Stage 2. Transfer the selected embryo into the pit.
17 Embryos
3. Place the embryo with the dorsal side up and hold it carefully
at the anterior end with the handling needle.
4. Cut the epidermis from the posterior tip of the embryo along
the dorsal midline towards the anterior tip with the cutting
needle.
5. Open the embryo by moving the epidermis aside with the
preparation needles or a Dumont forceps.
6. Remove the non-neural tissue that lies on top of the CNS (i.e.,
guts and fat body).
7. Carefully detach the CNS from the epidermis.
8. Continue as in steps 7–10 of Subheading 3.14.
9. Push the small cover slip slightly forwards. The glycerol flow
should push the brain hemispheres forwards in front of the
ventral nerve cord (see Note 20).
10. When the CNS is appropriately aligned, fix the small cover slip
by applying a drop of nail polish to each corner. Let the nail
polish dry and seal the cover slip by applying nail polish to the
edges (two times).
11. Clean the cover slip from glycerol and fix it on a microscope
slide with small stickers.
12. Under a microscope, the CNS can be examined from both
sides (with a view from ventral or dorsal) by turning the cover
slips including the filet preparation.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 31

3.16 Analysis of the The development of brain neuroblasts occurs in a stereotypic spa-
Developing Pattern of tiotemporal pattern, with every neuroblast expressing a unique
Brain Neuroblasts combination of marker genes. Both have been described in detail
for Drosophila [7, 8], which allows the identification of individual
brain neuroblasts. The brain neuroblast pattern is best examined in
filet preparation of embryos from late stage 8 to late stage 11 using
100× magnification and differential interference contrast. We rec-
ommend (combinatorial) antibody stainings for distinct molecular
markers (particularly against Deadpan, svp-LacZ, and Engrailed
protein) to analyze the patterns of brain neuroblasts in more detail.
Deadpan (Dpn) is a general neuroblast marker. In early develop-
mental stages (i.e., late stage 8–stage 9), Dpn antibody may even
be sufficient to identify single neuroblasts according to their posi-
tion within the entire brain neuroblast pattern [7]. From stage 9
onwards, identification of single brain neuroblasts becomes increas-
ingly difficult with general markers. We recommend the use of svp-
LacZ enhancer trap line (available at Bloomington Stock Center,
Bloomington, Indiana, USA). The expression pattern of svp-LacZ
lines is stable, well described, and persists throughout embryonic
brain development. svp-LacZ marks a subset of (about 40) brain
neuroblasts in a characteristic pattern, which covers all three main
subdivisions of the brain (i.e., proto-, deuto-, and tritocerebrum).
By that, it can be used to identify the svp-LacZ-labeled neuroblasts
within the entire pattern of brain neuroblasts but also the adjacent
(unlabeled) ones (Fig. 2D-D″) according to ref. 7. In any case, it is
helpful to combine stainings with Engrailed, which is segmentally
expressed in the posterior part of each subdivision of the brain
neuroectoderm (engrailed head spot—protocerebrum, engrailed
antennal stripe—deutocerebrum, engrailed intercalary stripe—
tritocerebrum) and in distinct subsets of brain neuroblasts that
emerge from the corresponding Engrailed neuroectodermal
domains (Fig. 2D-D″). These Engrailed-positive brain neuroblasts
additionally serve as reference points for orientation within the
Dpn- or svp-LacZ-labeled neuroblast patterns [7, 8]. The (combi-
natorial) use of the markers mentioned above may in most cases be
sufficient to identify single brain neuroblasts. However, further
markers are listed in [8].

4 Notes

1. The desired embryonic stage, and thus time window of egg

collections, depends on the process to be examined and has to
be chosen according to [9]. For example, the analysis of pat-
terning in the neuroectoderm requires early developmental
stages (starting at stage 5), while the analysis of NB formation
requires stages of mid-embryogenesis (stages 8–11). For the
32 David Jussen and Rolf Urbach

former, we suggest egg collections of 3–4 h time windows,

while for the latter, 8–16 h time windows are suitable
(at 25 °C).
2. Egg baskets can easily be made by cutting a 50 ml falcon tube
in half, cutting a hole into the cap and screwing a piece of
nylon mesh between tube and cap (see ref. [10]).
3. Note that long-term storage of fixed embryos may affect the
staining quality. Some antibodies yield best stainings with
freshly fixed embryos.
4. For double in situ hybridization, one probe has to be labeled
with DIG and the other one with FITC (the use of different
labels like biotin and dinitrophenol may be suitable, too). The
dilution at which the probes work best must be tested before.
We suggest testing probes by performing a standard in situ
hybridization procedure (as described above) for a serial dilu-
tion of probes (e.g., 1:50, 1:500, 1:2,500) on wild-type
embryos. Probes should be detected via NBT/BCIP, since this
method is very sensitive and easy. The optimal dilution is reached
when the specific signal develops clearly without background
staining (which may take up to 2 h or longer). If this is not the
case, dilute the probe further. Higher dilutions may prolong the
staining reaction. However, a clear signal should be preferred,
since background staining of the different colorimetric reactions
adds up and can decrease the overall staining quality signifi-
cantly. Probes may be reused multiple times when stored at
−20 °C. Often, the best results are reached when probes have
been used before, since less unspecific bindings occur.
5. The choice of the first antibody (anti-DIG-AP or anti-FITC-
AP) depends on the probe to be detected first. The first detec-
tion step is the generally less demanding and more sensitive
NBT/BCIP reaction. It results in a deep blue precipitate,
which is per se richer in contrast alongside the whitish tissue of
the embryo. For those reasons, the inferior probe should be
detected first. Here, only AP-dependent reactions are used for
probe detection. Other enzymes used for colorimetric detec-
tion are peroxidases (using DAB as chromogen) or beta-
galactosidase (using X-beta-D-Gal, Salmon-beta-D-Gal, or
Magenta-beta-D-Gal as chromogen). However, we found that
the peroxidase-dependent reactions produce relatively poor
signal/background ratio with this protocol, while the beta-
Gal-dependent reactions have been reported to be little sensi-
tive (see ref. [11]); this report also contains information on the
use of further AP substrates.
6. The AP reaction may take from ~5 min up to several hours,
depending on the probe used and the amount of transcript pres-
ent. The staining solution should be replaced every hour.
Usually, staining patterns start to emerge during the first 15 min.
Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 33

The reaction should be terminated as soon as a) the relative

amount of background vs. specific staining rises markedly or b)
the specific signal becomes too strong. The optimal staining
intensity also depends on the context. When co-labeling of
certain cells is expected (e.g., when a general NB marker is
combined with an NB subset marker), the staining should be
terminated earlier, since strong stainings may mask the pres-
ence of further dyes. Apart from that, it is recommended to
terminate stainings relatively late to assure the representation
of low transcript levels. Note that the NBT/BCIP staining will
appear slightly lighter after the incubation in methanol and
changes to a blue color since the red component of the initially
purple reaction product becomes washed out.
7. Embryos can be transferred into the tube by slightly tilting the
dish and flushing the embryos into one corner with PBTween.
From there, they can be sucked up with a pipette.
8. Vector Red staining reactions may take up to several hours.
The staining solution should be replaced every hour. The
expression pattern usually starts to emerge during the first
30 min. The reaction can be slowed down by incubating at
4 °C. This may allow overnight incubation. Note that none of
the dye will be washed away during subsequent steps of this
protocol. Therefore, the staining reaction should be termi-
nated as soon as the staining intensity is judged optimal.
9. This fixation step inactivates AP. Methanol is not suitable for
terminating the Vector Red reaction, because the red precipi-
tate is (at least partially) soluble in methanol. Efficient AP
inactivation is especially important if further NBT/BCIP
stainings are planned (e.g., for the detection of balancer
expression) to prevent cross-reactions.
10. Many primary antibodies can be used in solutions containing
1 % sodium azide; this prevents contamination with fungi or
bacteria and is advantageous as an antibody can be reused sev-
eral times often with increasingly better staining results.
However, keep in mind that some antibodies can be impaired
or do not work at all when using sodium azide; therefore we
recommend to compare stainings in which the antibody has
been incubated in PBT and in PBT with sodium azide.
11. Additional antibodies for balancer detection can be added to
the primary antibody solution. Balancers can be detected via
DAB in parallel to the actual primary antibody. However, we
recommend detecting balancers via AP and NBT/BCIP, since
this method is more sensitive.
12. Tyramide Signal Amplification increases the staining intensity
of DAB stainings. This amplification step may not be neces-
sary with good antisera.
34 David Jussen and Rolf Urbach

13. The parameters of this signal amplification step are particularly

critical and may have to be adapted. Prolonged incubation and
inappropriately high concentrations of TSA reagent quickly
lead to high background levels during the DAB staining
reaction.
14. The cutting needle has to be flattened more intensely than the
handling needle to obtain a sharp edge for preparation.
15. Add a drop of glycerol on the whetstone/abrasive paper.
Prepare the needle under the stereo microscope by gently
streaking it over the glycerol-covered surface of the stone/
paper. The cutting needle has to be sharpened to obtain an
even blade with somewhat rectangular edges, while the han-
dling needle should be whetted in a way to obtain a thicker tip
with rounded edges to prevent damaging of embryos during
preparation.
16. Embryos can be examined from different angles by carefully
moving the upper cover slide over the spacers, which will cause
the embryos to rotate in the glycerol. By that, different regions
of the embryonic brain can be examined in one specimen.
17. Flat preparations, though time consuming in production,
allow the most detailed examination of gene expression pat-
tern, even on the level of single cells (e.g., neuroblasts) [7, 8].
Yolk and other tissue which often accumulate unspecific stain-
ing are removed, and the whole neuroectoderm is presented
in one plane, which significantly improves differential interfer-
ence contrast. That way, neuroectodermal cells and neuro-
blasts, including the unstained, can be easily detected.
18. This step may need some exercise. The position of the embryo
may be corrected by sliding the small cover slip in different
directions. Avoid squeezing the embryo by pushing the cover
slip down.
19. We recommend 40× or 63× objectives for the examination of
gene expression in the neuroectoderm and 63× or 100× objec-
tives for the analysis of the NB pattern. Stainings can be docu-
mented on a microscope equipped with a standard digital
(CCD/CMOS) camera. For digital documentation, we rec-
ommend recording selected regions of flat preparations.
Depending on the magnification used, a stack of 5–10 pictures
along the z-axis is usually enough to cover all cells of the
embryonic brain (i.e., neuroectoderm and all underlying neu-
roblasts of one hemisphere).
20. The hemispheres of the late embryonic brain lie on top of the
ventral nerve cord, which hampers the view during micros-
copy. After this step, the brain hemispheres should lie in front
of the ventral nerve cord, so that the whole embryonic CNS is
arranged in a straight line.
 Non-fluorescent Multiple Stainings in the Embryonic Brain of Drosophila 35

Acknowledgments

The authors thank Janina Seibert and Dagmar Volland for sharing
protocols and considerable expertise, Janina Seibert for Fig. 2c,
and Karoline F. Kraft for critically reading the manuscript. This
work was supported by grants from the Deutsche
Forschungsgemeinschaft (UR163/2-1 and UR163/3-1).

References
1. Kosman D, Mizutani CM, Lemons D, Cox brain and ventral nerve cord. Dev Biol 346:
WG, McGinnis W, Bier E (2004) Multiplex 332–345
Detection of RNA Expression in Drosophila 7. Urbach R, Schnabel R, Technau GM (2003)
Embryos. Science 305:846 The pattern of neuroblast formation, mitotic
2. Skeath JB, Thor S (2003) Genetic control of domains and proneural gene expression during
Drosophila nerve cord development. Curr early brain development in Drosophila.
Opin Neurobiol 13:8–15 Development 103:3589–3606
3. Urbach R, Technau GM (2004) Neuroblast 8. Urbach R, Technau GM (2003) Molecular
formation and patterning during early brain markers for identified neuroblasts in the devel-
development in Drosophila. Bioessays 26: oping brain of Drosophila. Development 103:
739–751 3621–3637
4. Urbach R, Volland D, Seibert J, Technau GM 9. Campos-Ortega J, Hartenstein V (1997) The
(2006) Segment-specific requirements for dor- embryonic development of Drosophila mela-
soventral patterning genes during early brain nogaster. Springer, Berlin Heidelberg
development in Drosophila. Development 10. Rothwell WF, Sullivan W (2000) Fluorescent
133:4315–4330 analysis of Drosophila embryos. In: Sullivan
5. Seibert J, Volland D, Urbach R (2009) Ems W, Ashburner M, Hawley RS (eds) Drosophila
and Nkx6 are central regulators in dorsoven- protocols, 1st edn. CSHL Press, Cold Spring
tral patterning of the Drosophila brain. Harbor, New York, pp 143–145
Development 136:3937–3947 11. Hauptmann G (2001) One-, two-, and
6. Seibert J, Urbach R (2010) Role of en and three-color whole-mount in situ hybridiza-
novel interactions between msh, ind, and vnd tion to Drosophila embryos. Methods
in dorsoventral patterning of the Drosophila 23:359–372
 Chapter 3

Analysis of Complete Neuroblast Cell Lineages

in the Drosophila Embryonic Brain via DiI Labeling
Karoline F. Kraft and Rolf Urbach

Abstract
Proper functioning of the brain relies on an enormous diversity of neural cells generated by neural stem
cell-like neuroblasts (NBs). Each of the about 100 NBs in each side of brain generates a nearly invariant
and unique cell lineage, consisting of specific neural cell types that develop in defined time periods. In this
chapter we describe a method that labels entire NB lineages in the embryonic brain. Clonal DiI labeling
allows us to follow the development of a NB lineage starting from the neuroectodermal precursor cell up
to the fully developed cell clone in the first larval instar brain. We also show how to ablate individual cells
within a NB clone, which reveals information about the temporal succession in which daughter cells are
generated. Finally, we describe how to combine clonal DiI labeling with fluorescent antibody staining that
permits relating protein expression to individual cells within a labeled NB lineage. These protocols make
it feasible to uncover precise lineage relationships between a brain NB and its daughter cells, and to assign
gene expression to individual clonal cells. Such lineage-based information is a critical key for understanding
the cellular and molecular mechanisms that underlie specification of cell fates in spatial and temporal
dimension in the embryonic brain.

Key words Drosophila, Embryonic brain, Neural stem cell, Neuroblast lineage, DiI labeling, Antibody
staining

1 Introduction

Proper functioning of the brain relies on an enormous diversity of

neural cells generated by neural stem cells, called neuroblasts (NBs)
in insects. In Drosophila, the central nervous system (CNS) includ-
ing the brain develops from a bilaterally symmetrical sheet of neu-
roectodermal cells. It gives rise to a fixed number of NBs that
segregate to the interior of the embryo. Upon segregation, a NB
typically undergoes rounds of asymmetric cell division, budding off
smaller intermediate precursor cells, which usually divide once to
produce two postmitotic cells. Each NB produces specific neural
cell types in defined time periods and, by that, generates a nearly
invariant and unique cell lineage, suggesting stereotyped patterns
of lineage development [1]. The fate of an individual NB and,

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_3, © Springer Science+Business Media, LLC 2014

37
38 Karoline F. Kraft and Rolf Urbach

accordingly, the main features of its developing cell lineage are

specified by positional cues within the neuroectoderm [2, 3] and,
thereafter, by temporal cues as well as by a combination of further
developmental control genes that it expresses [1, 4–7]. Many stud-
ies have shown that knowledge about the precise lineage relation-
ships between a NB and its daughter cells is a critical key to
investigate the molecular mechanisms that underlie specification of
cell fates in spatial and temporal dimension [5, 8–13]. Such lineage-
based information on the developmental origin of a NB and the
neuron types it develops will not only decipher the sophisticated
circuitry of the brain but also elucidate how a complex brain devel-
ops. Therefore, analysis of NB lineages with high resolution is
needed, permitting a systematic identification of neural cell types
by resolving the development of each single cell in a NB lineage.
To unravel NB lineages in the fly brain, especially in the post-
embryonic period of development, very straightforward genetic
clonal labeling techniques have been established, such as FLP/
FRT-based (e.g., [14, 15]), Gal4-based G-TRACE [16], or
MARCM-based methods [17–20]. However, due to technical lim-
itations, these genetic labeling techniques seem to be less applica-
ble to the elucidation of the entire NB lineages in the embryo.
As the production of FLP/FRT clones depends on heat shock-flippase
necessary to induce recombination, during embryogenesis critical
levels of heat shock-flippase may not become enriched before the
early-born part of a lineage has developed [21]. Also, MARCM
fails to disclose the entire NB lineages in the embryo, since after
recombination (which depends also on a critical level of heat shock-
flippase), the clonal reporter expression additionally relies on the
loss of the GAL80 repressor, which seems to persist over the entire
embryonic development [22]. To our knowledge, only clonal DiI
labeling undoubtedly reveals the entire lineage of embryonic brain
NBs. Clonal DiI labeling is highly selective and noninvasive. It is
based on the application of a lipophilic fluorescent dye onto early
neuroectodermal cells. The dye is easily absorbed by the mem-
brane of the cell and diffuses through the lipids quickly without
being transmitted to neighboring cells (see also ref. [23]). There
are several different versions of carbocyanine dyes available (e.g.,
the most commonly used DiI, DiO, and DiD) with different fluo-
rescent properties. DiI, for example, has similar excitation proper-
ties to rhodamine; excited by green light, it fluoresces red.
Depending on requirement, a number of alternative versions of the
classical lipophilic carbocyanine dyes have been developed over the
last few years, e.g., a chloromethylbenzamino derivate of DiI
(CM-DiI), which shows a better staining persistence after standard
tissue fixation procedures. As lipophilic carbocyanine dyes
are entirely atoxic [24], the labeled cell can develop completely
unaffected by the dye according to its position-dependent fate.
 DiI-labeling in the Drosophila Embryonic Brain 39

Clonal DiI labeling allows in vivo to study the development of all

the progeny cells derived from a single-labeled neuroectodermal
precursor cell (e.g., processes of cell proliferation and differentia-
tion) and, after fixation, to analyze the cellular composition (i.e.,
glial or neuronal cells) and overall morphology of the fully differ-
entiated cell clone. By using this technique, the embryonic lineage
of all neuroblasts in the ventral nerve cord (VNC) has been
described previously [25–27]. Recently, clonal DiI labeling in
combination with molecular markers and cell ablation experiments
allowed us to analyze the embryonic development of a prominent
central brain structure, the mushroom bodies (MBs). We unrav-
eled the origin of the four mushroom body neuroblasts (MBNBs),
their mode of formation, and could clarify the spatiotemporal
development and individual cellular composition of their embry-
onic lineages [28].
Here we describe a protocol that was basically developed by
Bossing and Technau [29] and modified by us to allow targeted
DiI labeling of brain NB clones. Our methodical modifications
describe also how to selectively ablate cells within a labeled NB
clone, which, for example, reveals informations about the temporal
succession at which daughter cells are generated. Finally, we
describe how to combine clonal DiI labeling with fluorescent anti-
body staining that permits relating protein expression to individual
cells within a DiI labeled NB clone.

2 Materials

2.1 Reagents 1. DiI (1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine

perchlorate; DiI, CM-DiI, DiD, and other carbocyanine trac-
ers can be obtained at Molecular Probes).
2. DiI/oil (5 mg/ml).
3. CM-DiI/oil (3 mg/ml).
4. DiD/oil (10 mg/ml).
5. Apple juice agar: Boil 29 g agar in 1 l apple juice until mixture
is translucent. Decant into Petri dishes as long as the mixture
is hot; let cool.
6. PBS (1 l 20× stock solution): Solve 151.92 g NaCl, 19.88 g
Na2HPO4 (dehydrated), and 8.28 g NaH2PO4 in ddH2O, pH 7.4.
7. PBTween (0.1 % Tween20 in PBS).
8. Heptane glue: Mix Cello 31.39 (Tesa) with heptane in a 2:1
ratio.
9. DAB solution: Dissolve 2–3 mg/ml DAB in 100 mM Tris–
HCl, pH 7.4.
40 Karoline F. Kraft and Rolf Urbach

10. Fixatives: For antibody staining after DiI labeling, mix 200 μl
37 % formaldehyde in 900 μl 1× PBS. For photoconversion,
mix 260 μl 37 % formaldehyde in 900 μl 1× PBS.
11. Glycerol 70 %: Dissolve 70 ml 100 % glycerol in 30 ml 1× PBS.
12. Fluorocarbon oil Voltalef 10S.
13. Fetal calf serum.

2.2 Equipment 1. Inverse microscope (e.g., Leitz Fluovert FU) equipped with
different fluorescent filters (Fluorescein FITC, Rhodamin, etc.).
2. 100 W halogen lamp.
3. 100 W mercury lamp.
4. Stereomicroscope.
5. Cold light source.
6. Capillary grinder (e.g., Bachofer Type 462).
7. Capillary puller (e.g., Sutter P97).
8. Micromanipulator (e.g., Leitz M).
9. Bunsen burner.
10. Cover slips (22 × 22 mm, 24 × 60 mm).
11. Glass slide (76 × 26 × 1 mm).
12. Plasticine.
13. Borosilicate glass microcapillaries (Hilgenberg normal
“Meterware”).
14. Glass cutter.
15. Wet chamber (e.g., Petri dish containing wet pieces of filter
paper).
16. Petri dish (60 × 20 mm).
17. Scalpel.
18. Preparation needle.
19. Fine forceps (e.g., Dumont 5SF).
20. Black block dish.
21. Single-use syringe (5 or 10 ml).
22. Polyethylene tube (6 × 4 × 1 mm).
23. Transparent nail polish.
24. Tape (single sided, double sided).
25. Dissecting knife.
26. Cage (60 × 70 mm).
27. Weighing tray (41 × 41 × 8 mm).
 DiI-labeling in the Drosophila Embryonic Brain 41

3 Methods

3.1 The Workplace 1. We recommend performing clonal DiI labelings, if possible, on

a microscopic stage with controllable thermocouple element, or
alternatively, in a temperature-controlled room (18–20 °C), so
that development of the embryos is decelerated. This prolongs
the time period in which successful labeling can be obtained. At
warmer temperatures, the viability of treated embryos declines
and precise timing of DiI labeling becomes more difficult.
2. Labeling and observation are carried out on an inverted
microscope.
3. Ideally, to minimize vibration, the microscope and microma-
nipulator carrying the DiI-filled capillary are posed on a heavy-
weight balance table or comparable equipment.
4. The microscope needs to be equipped with exchangeable halogen
and mercury lamps and appropriate filter settings capable of
exciting the respective dye (for detailed informations on different
dyes and filter settings, see “Molecular Probes Handbook of
Fluorescent Probes and Research Chemicals”) and GFP.

3.2 Preparation of 1. The fluorescent, lipophilic dye is mixed at a given concentra-

Fluorescent Dye/Oil tion (see Subheading 2.1) with any commercially available veg-
Mixture etable oil (e.g., canola or soya oil).
2. Shake slowly (ca. 200 rpm) for 3–5 h in the dark at room
temperature. The aliquots (~10 μl) can be stored at −20 °C for
several years.

3.3 Preparation 1. Take a piece of polyethylene tube (ca. 8 cm) and heat it briefly
of Polyethylene Tube over a small flame of a Bunsen burner (until the polyethylene
starts “bubbling”).
2. Pull apart slowly from both ends until the tube reaches a
length of about 50 cm.
3. Cut one end of the tube at appropriate position so that the
inner diameter fits the capillary (Fig. 1c).

3.4 Manufacturing 1. Capillaries most suitable for DiI labeling have the following
Labeling Capillaries characteristics: (a) a relatively long shank and (b) a relatively
small tip diameter. The outer diameter at the distal tip of the
capillary should be around 3–5 μm (Fig. 1d) (see Note 1).
2. Grind the capillary on a commercial capillary grinder in an
angle of ca. 30°; the resulting capillary will be sharp enough to
easily penetrate the vitelline membrane. For best results, the
capillary should be wet-ground. In the following, this capillary
type will be called “labeling capillary,” whereas a pulled but
unground capillary, needed for transfer and preparation of
embryos, will be called “flat capillary.”
42 Karoline F. Kraft and Rolf Urbach

Fig. 1 Set up for DiI labeling. (a) Gastrula-stage embryo (at stage 7) glued to the
cover slip for DiI labeling in the brain neuroectoderm. cf cephalic furrow; dML
dorsal midline. (b) Schematic cross section through the head region boxed in
(a); the embryo is fixed on a cover slip in appropriate orientation to label cells in
the ventral or intermediate neuroectoderm (NE) under an inverse microscope.
vML ventral midline. Modified from [29]. (c) Micromanipulator equipped with a
labeling capillary coupled with a polyethylene tube which is connected to a
syringe. (d) The angle of the bevel of the labeling capillary should be around 30°,
and the outer diameter at the distalmost tip between 3 and 5 μm. (e) Labeling of
two neighboring neuroectodermal cells with DiI (magenta) and DiD (green).
Dashed line indicates the adhesive border (color figure online)

3. Under visual control through a stereomicroscope, slowly

lower the capillary until its tip touches the abrasive wheel of
the grinder. The tip is open when you see water climbing up
into the capillary. Note, grinding does not take longer than a
few seconds.
4. Mark the untreated, upper end of the labeling capillary with a
water-resistant pen while it is still clamped in the grinder
holder. This will help you later to fix the capillary in the micro-
manipulator in the right orientation.
5. To evaluate the inner diameter of the capillary tip, couple a poly-
ethylene tube (as prepared above, see Subheading 3.3) to the
capillary. Insert the syringe at the other end of the tube and apply
pressure while the tip of the capillary is under distillated water
 DiI-labeling in the Drosophila Embryonic Brain 43

filled in a black block dish: the capillary is useful for labeling if

you can hardly detect the escaping small air bubbles under
intense light. For ablation of cells (see Subheading 3.16), we rec-
ommend using capillaries with a slightly larger tip opening.
6. Using the syringe, draw acetone into the capillary and eject it.
The labeling capillary is then dry and ready for storage.
7. Store the labeling capillaries dust-free in a Petri dish; fix each
capillary on a thin strip of plasticine. After labeling or ablation,
the capillary can be restored when before rinsed with acetone.

3.5 Preparing 1. In order to prepare a glue-coated cover slip (24 × 60 mm),

Heptane Glue-Covered spread a drop of heptane glue with the help of a small cover
Cover Slips slip (22 × 22 mm) so that the cover slip becomes coated
homogenously with a thin film of glue, and let it dry for at
least 10 min (Fig. 2a) (see Note 2).
2. Cut frames out of single-sided tape (width 25 mm). The
border of a frame should be ~5 mm and the opening ~15 mm
wide (Fig. 2a). Stick the frame onto the glue-coated cover slip
(see Note 3).

3.6 Egg Collection 1. At least 2 days before you start collecting embryos for DiI
and Staging labelings: Place the flies (between 4 and 14 days old) in a fly
cage on an apple juice plate with yeast. Change the plate every
24 h (see Note 4).
2. For DiI labeling, precisely staged embryos within a range of
1 h can be obtained when changing the plate every hour at
25 °C.

3.7 Dechorionization 1. Fix a piece of double-sided tape to a cover slip (22 × 22 mm)
and Mounting (Fig. 2b).
of Embryos for 2. Take cooled, fresh apple juice agar plates, and cut the agar into
DiI Labeling blocks of 2 × 2 cm. Place two blocks of agar in a distance of
about 1.5 cm on a glass slide (76 × 26 mm), and the cover slip
on top, leaving a part of each agar block exposed (Fig. 2b1).
3. Make 3 lines of 5–7 holes each, by pressing a preparation nee-
dle into the agar blocks at an angle of about 20° (Fig. 2b2).
4. Transfer and scatter a few embryos onto the tape with a dis-
secting knife (Fig. 2b3).
5. Wait about 3 min to let the chorion dry.
6. Collect eggs at the blastoderm stage and dechorionate by
rotating the embryo slightly along the dorsoventral axis with
the tip of a preparation needle. The chorion breaks and releases
the embryo (see Note 5).
7. Carefully pick up each dechorionated embryo and transfer it
to one of the agar blocks to prevent further drying (Fig. 2b4).
44 Karoline F. Kraft and Rolf Urbach

Fig. 2 Preparation of embryos for DiI labeling. (a) Spread a drop of heptane glue on
a cover slip. Let the heptane glue dry for at least 10 min. Stick a single-sided tape
frame on the glue-coated cover slip. (b) (1) Fix a piece of double-sided tape on a
cover slip. Place the cover slip on top of two blocks of agar. (2) Make holes into the
agar. (3) Transfer embryos onto the tape. Dechorionate embryos. (4) Transfer each
embryo on an agar block and orientate them. (5) Cut up the agar blocks so that
every embryo is on an individual block. (6) Check for optimal orientation. (7) Fix the
embryo on the prepared cover slip. (8) When embryos are properly dried, cover
them with Voltalef 10S oil. a anterior, p posterior, d dorsal, v ventral

8. For DiI labelings in the brain neuroectoderm, the

anteroposterior axis of the embryo should be orientated per-
pendicular to the injection capillary. Transfer embryos into the
prepared holes in the agar block, so that the posterior half
of the embryo is sticking in the hole, and the procephalon is
 DiI-labeling in the Drosophila Embryonic Brain 45

facing you (Fig. 2b6). Each embryo has to be orientated accu-

rately with respect to the dorsoventral axis depending on the
area of the brain neuroectoderm you want to label in (Fig. 2a)
(see Note 6).
9. Using a scalpel, separate embryos on smaller agar blocks, each
embryo on a single block (Fig. 1b5).
10. Now, stick the part of the procephalon facing you on the cover
slip. Carefully lower a prepared cover slip (see below and
Subheading 3.5) onto a single block and press slightly in order
to fix the embryo in proper orientation (Fig. 1b7) (see Note 7).
11. Repeat this process until all embryos are stuck next to each
other (head-to-tail) to the prepared cover slip.
12. Let embryos desiccate for 5–12 min at room temperature
(see Note 8).
13. As soon as the embryos are dried properly, cover them with a
few drops of fluorocarbon oil Voltalef 10S (Fig. 2b8).
Fluorocarbon oil (like halocarbon oil) allows oxygen to per-
meate to the embryo but prevents water from evaporating.
14. Until the beginning of injection, the cover slip is kept in a wet
chamber at room temperature.

3.8 Filling the 1. Put the unbeveled end of the capillary (as indicated by the pen
Capillary with Dye mark, see Subheading 3.4) into the polyethylene tube, and a
Solution single-use syringe on the other end of the tube.
2. Fix the capillary with the topside up (as indicated by the pen
mark) in the holder of a micromanipulator (Fig. 1c).
3. Place a drop of 1 μl dye/oil on a cover slip (24 × 60 mm) and
bring both the dye drop and the capillary tip under the micro-
scope in focus using a 10× objective.
4. Now change to a 50× objective, and slowly draw up the dye
into the capillary.
5. As soon as the tip is filled with dye, the syringe has to be
removed from the polyethylene tube, the tip of the capillary to
be lifted up, and the syringe, with the plunger pulled out, to
be inserted again into the tube.
6. To reuse the drop of dye later, you can store the cover slip in
a dark, dry, and dust-free box.

3.9 Labeling 1. Place the cover slip with the embryos under the microscope.
of Individual 2. If the embryos are properly desiccated, there should be a rela-
Neuroectodermal Cells tively sharp adhesive border around the area where each
and Determination embryo is stacked to the glue and several rows of neuroecto-
of Their Position dermal cells should lie in one focal plane (Fig. 1b).
46 Karoline F. Kraft and Rolf Urbach

3. To identify the optimal time point for labeling and the posi-
tion of the cell to be labeled, it is useful to have at least a 600×
or higher magnification (e.g., using a 50× immersion oil objec-
tive and a 12.5× ocular magnification). Optionally, the label-
ing can be controlled on a monitor using a standard digital
camera (CCD/CMOS).
4. Check under the microscope the developmental stage of the
embryos. By stage 7 (staging according to [30]) you can
detect, for example, the dorsal and ventral midline, the cephalic
furrow (see Note 9). The latest time point for successful label-
ing seems to be around stage 9.
5. Determine the neuroectodermal cell to be labeled by making
use of the above mentioned, morphological landmarks that
help to recognize the position of individual cells (see Note 9).
6. Via the micromanipulator, lower the tip of the capillary until it
is in focus, but still in a certain distance to the embryo. From
now on, do not work with the micromanipulator anymore, as
all movements should be effected via the microscope table.
7. Penetrate the vitelline membrane slowly and carefully with the
tip of the capillary in the position where the neuroectodermal
cell of interest is located, and approach the tip as closely as
possible towards its cell membrane (see Note 10).
8. By slowly pushing the plunger of the syringe, deposit a small
drop of dye next to this cell. The diameter of the applied drop
is critical for successful labeling: about 1/2–1/5 of the diam-
eter of a neuroectodermal cell seems to be most promising
(Fig. 1e). Immediately after depositing the drop, the dye dif-
fuses into the cell membrane leaving behind the nonfluores-
cent drop of oil solvent.
9. Slowly, pull the capillary out of the embryo.
10. For double labeling in the same embryo, place two capillaries
filled with different dyes (each on a different micromanipula-
tor) as close as possible to the embryo before labeling. If no
additional micromanipulator is available, you can also quickly
exchange the two capillaries between the labeling.
11. After labeling, briefly inspect the quality of the labeling under
fluorescent light using a halogen lamp (see Note 11).
12. Finally, control the position of the labeled cell via DIC optics.
13. Take notes of each labeling concerning position and behavior
of the labeled cell (see Subheading 3.10).
14. Store the labeled embryos on the cover slip in a wet chamber
at 18–20 °C.
 DiI-labeling in the Drosophila Embryonic Brain 47

3.10 Control of Cell 1. About 1 h after labeling (around late stage 8/early stage 9),
Behavior After judge the behavior of the labeled cell using the same fluores-
Labeling cence and magnification settings (see Note 12).
2. If the neuroectodermal cell has delaminated, the NB was suc-
cessfully labeled. To follow the development of the NB lineage
in more detail, it is possible to inspect the labeling in shorter
time intervals (e.g., every 30 min) without disturbing the
labeled cells.
3. The identification of a labeled NB clone is facilitated when
labeling is done in embryos of appropriate Gal4 strains (most
can be ordered in the Bloomington or Kyoto Stock Center) in
which the Gal4 pattern is visualized via a GFP reporter.
Alternatively, you can combine your labeling with an antibody
staining (see Subheading 3.15).
4. For further development, the embryos are kept under fluoro-
carbon oil Voltalef 10S at 18 °C until they reach the desired
stage. To obtain fully developed stage 17 NB clones, embryos
are kept at 18 °C overnight.
5. Before further preparation, monitor the embryos and eliminate
all those which do not show properly labeled cell clones.

3.11 Preparation 1. For better handling, we recommend removing the parts of the
of Embryos at cover slip outside of the single-sided tape frame using a glass
Developmental Stage cutter.
17 (See Fig. 3) 2. Tip the cover slip at an ~80° angle for several minutes so that
the fluorocarbon oil can drain off.
3. Then, place the cover slip in a weighing tray (41 × 41 × 8 mm)
and spread about 1 ml heptane on the leftover fluorocarbon
oil in order to remove it.
4. Rock the cover slip slowly for about 10–20 s. Do not rock too
long since the heptane will also solve the glue and the embryos
will detach.
5. Quickly place the cover slip on a glass slide for better
handling.
6. Pour away the remaining heptane carefully. Wait a few seconds
until the heptane is almost completely evaporated and cover
the embryos with PBS immediately to prevent drying.
7. Get the embryos, one by one, out of the vitelline membrane
using a flat capillary.
8. Dissect the CNS out of the embryo by placing one fine forceps
at the anterior end of the embryo and another one in its
middle/posterior part. Carefully lacerate the embryo to make
the CNS visible. Remove all remaining tissue from the CNS.
48 Karoline F. Kraft and Rolf Urbach

Fig. 3 Different applications of clonal DiI labeling. (a) Combined confocal images
of a DiI labeled mushroom body NB clone (a) in a Ok107 (ey)-Gal4; UAS-CD8::GFP
brain (a′) at mid/late stage 17. Note that some cells of the DiI labeled clone
express GFP, indicating that this is a mushroom body NB clone. (b) DiI labeled
mushroom body NB clone at late stage 17 after ablation of the NB at stage 11.
Accordingly, only the early-born part of its lineage has developed. (c) 3D recon-
struction of a DiI labeled mushroom body NB clone (magenta) in the early first
larval instar in comparison to the structures of the entire mushroom body (green)
as revealed by Ok107 (ey)-Gal4; UAS-CD8::GFP expression. (d) A mushroom
body NB clone (magenta) and another protocerebral NB clone (green) at late
stage 17 obtained after labeling of two neighboring cells with DiI (magenta) and
DiD (green) (see inset). A–D Adapted from [28]. (e) Photoconverted protocerebral
NB clone at early stage 17. (f) DiI labeled protocerebral NB clone (at early first
larval instar) before (f) and after (f′) fluorescent antibody staining against the
gene products Eyes absent (f″a) and Retinal homeobox (f″b). Note that antibody
staining leads to partial loss of DiI signal. (f″) Merge

9. Use the flat capillary to transfer each CNS to a glue-coated

cover slip (24 × 60 mm) prepared with a tape frame
(as explained in Subheading 3.5) that is filled with PBS.
10. Stick the ventral nerve cord with its ventral side on the glue so
that the brain hemispheres are orientated upwards.
11. The preparation can be stored in a wet chamber at 4 °C until
fixation and documentation.
 DiI-labeling in the Drosophila Embryonic Brain 49

3.12 Preparation After about 30 h at 18 °C, embryos are checked for eclosion under
of First Instar Larvae a stereomicroscope in intervals of about 15 min. Take notes on
which larva hatches at which position on the cover slip, as this
information is necessary to make correlations to the notes you took
during the labeling (see Note 13).
1. Immediately after eclosion, take the larvae out of the oil by
using a flat capillary. Transfer each larva separately to a glue-
coated cover slip prepared with a tape frame (as explained in
Subheading 3.5) filled in with PBS.
2. To prepare the CNS, tear the larva in half with a pair of fine
forceps. Then turn the cuticle “inside out” and the CNS will
appear attached to the cuticle. Remove the CNS entirely from
the tissue, transfer it with a capillary to another glue-coated
cover slip prepared with a tape frame (as explained in
Subheading 3.5) filled with PBS, and stick the ventral nerve
cord with its ventral side on the glue.
3. The preparation can be stored in a wet chamber at 4 °C.
If you want to photoconvert the clone continue as described
in Subheading 3.17, otherwise go on with fixation
(see Subheading 3.13).

3.13 Fixation 1. Remove the PBS under visual control using a stereomicro-
scope, and replace it immediately with fixation solution
(see Subheading 2.1). Note, as fixatives are toxic, take precautions
(e.g., place the stereomicroscope on a portable fume hood).
2. Fix 20 min for subsequent antibody staining (as described in
Subheading 3.15), otherwise 10 min.
3. Thereafter, carefully replace the fixative with PBS.
4. Change the PBS 4–5 times.

3.14 Documentation 1. Fix the cover slip (24 × 60 mm) carrying the preparation on a
glass slide with tape for better handling.
2. Remove excessive PBS so that the border of the liquid slightly
overtops the single-sided tape frame (see Note 14).
3. Now carefully place a cover slip (22 × 22 mm) onto the single-
sided tape frame.
4. By slightly moving the upper cover slip, the brain hemispheres
that get in contact with this cover slip (as explained in
Subheadings 3.11 and 3.12) can be brought into the best pos-
sible position for exposure.
5. Fix the cover slip to the underlying cover slip by applying nail
polish at the corners.
50 Karoline F. Kraft and Rolf Urbach

6. Immediately document the labeled NB clone using a confocal

microscope. Transport the specimen in a wet chamber (see
Note 15). Optionally, 3D reconstructions can be generated
from the LSM stacks using, for example, Amira or comparable
software.

3.15 Optional: As far it was only possible to combine photoconverted DiI labelings
DiI Labeling Combined with nonfluorescent histochemical antibody stainings (visualized,
with Fluorescent e.g., via the alkaline phosphatase reaction) with the disadvantage
Antibody Stainings that the photoconverted cells themselves, which have accumulated
strong levels of diaminobenzidine (DAB), could not be antibody
stained. Now this limitation can be avoided by combining labeling,
performed with the improved fixable DiI derivatives (e.g., CM-DiI),
with fluorescent antibody stainings. To carry out fluorescent anti-
body stainings with fluorescent DiI labeled clones, another critical
issue is the proper usage of detergents, needed for antibodies to
penetrate the membranes of cell tissues. As the carbocyanine dyes
are lipophilic, conventional detergents in widely used concentra-
tions (such as 0.1–0.3 % Triton X-100 or 0.05 % saponin) result in
extensive diffusion of the DiI label out of the tissue (own observa-
tions; see also ref. [31]). Presented below is a protocol in which we
apply a low concentration of 0.1 % Tween20, and in addition,
mechanically perforate the neurilemma that envelops the brain.
1. After fixation, to facilitate the penetration of the antibodies
into the brain tissue, carefully perforate the neurilemma with a
flat capillary. Perforate generously but do not damage the area
containing the labeled clones.
2. Thereafter, cover the specimen for 48 h with primary anti-
body/antibodies dissolved in 0.1 % PBTween (~200 μl). Put
the cover slip in a wet closed chamber at 4 °C.
3. Then take off the primary antibody/antibodies and replace
with 0.1 % PBTween twice.
4. Cover three times for 20 min with 0.1 % PBTween.
5. Dissolve the secondary antibody/antibodies (take in consider-
ation that you use appropriate fluorescent conjugates which
do not interfere with applicated carbocyanine dyes) in 0.1 %
PBTween and incubate for 18–20 h at 4 °C. In case of weak
antibody stainings, a Tyramide Signal Amplification System
(TSA) is recommended (according to the manufacturers’ pro-
tocol of PerkinElmer, Waltham, USA).
6. After taking off the secondary antibody/antibodies, replace
with 0.1 % PBTween twice and then cover for 15 min with
0.1 % PBTween twice.
7. Subsequently, wash twice with PBS.
8. Finally, put a cover slip on the tape frame, fix with nail polish,
and follow up with documentation immediately as the DiI sig-
nal fades relatively fast (see Subheading 3.14).
 DiI-labeling in the Drosophila Embryonic Brain 51

3.16 Optional: 1. For cell ablations, use embryos that have been labeled as
Ablation of DiI Labeled described in Subheading 3.9 and have developed into the
NBs and/or Daughter desired stage. Use a microscope equipped with a 100 W halo-
Cells gen lamp and appropriate filter settings. Bring the specimen in
which cell ablations will be done under a sufficient magnifica-
tion (600× or higher) in focus. Connect the ablation capillary
(see Subheading 3.4) with the tube, then clamp the capillary
into the micromanipulator and lower it into the Voltalef oil.
Finally, connect the syringe to the other end of the tube to
prevent Voltalef oil entering the tip.
2. Identify the labeled cell(s) to be ablated under the halogen
lamp.
3. Using the micromanipulator, move the capillary towards the
embryo under transmitted light.
4. Slowly and carefully move the cross table of the microscope
towards the capillary tip until the capillary penetrates the vitel-
line membrane, and then carefully direct it to the cell(s) desig-
nated for ablation.
5. Once the tip of the capillary reaches the cell to be ablated, cau-
tiously pull the syringe plunger in order to suck off the cell(s)
(see Note 16).
6. Remove the capillary from the embryo by moving the cross
table.
7. Store the treated embryos in a wet chamber at 18 °C until
they have reached the desired developmental stage.
8. Then prepare, fix, and document the specimen as described in
Subheadings 3.11–3.14.

3.17 Optional: Photoconversion transforms the DiI label into a permanent reac-
Photoconversion tion product via the Maranto reaction in which illumination
of DiI Labeled releases a singlet oxygen that oxidizes DAB [32] (see also ref. [33]).
Preparations DiI labeled photoconverted clones can be investigated using con-
ventional light microscopy and preserved for many years.
1. Incubate for 1 h in calf serum.
2. Exchange the drop of PBS with a drop of DAB solution (that
has been briefly centrifuged before). Note, DAB is a hazard-
ous chemical—handle with appropriate precautions!
3. Photoconvert the labeled NB clone under the microscope
with a 100 W mercury lamp and rhodamine filter using a water
or oil immersion objective with about 50× magnification.
Photoconversion is completed when all fluorescent signal is
gone, this takes usually 10–20 min. Check regularly to make
sure that the preparation is not over dyed.
4. After photoconversion, remove the DAB.
5. Rinse the preparation with PBS 3–4 times.
52 Karoline F. Kraft and Rolf Urbach

6. Fix in formaldehyde solution for 15 min.

7. Rinse 3–4 times with PBS.
8. Place a drop (20–25 μl) of 70 % glycerol on another cover slip
(24 × 60 mm).
9. Using a flat capillary, transfer the preparation to the glycerol
drop and bring it in appropriate orientation.
10. Carefully cover the glycerol drop containing the CNS with a
small cover slip (18 × 18 mm). Control the orientation of the
preparation under a stereomicroscope while the glycerol
spreads between both glass slides. Excess glycerol can be
removed with tissue paper.
11. Fix the cover slip by applying a drop of nail polish to all four
corners. Let those drops dry and seal all sides two times with
nail polish. The photoconverted NB clones can now be docu-
mented using Nomarski optics and a 63× (or 100×) immer-
sion oil objective, as camera lucida drawings, or can be
digitized. For example, with the help of a standard digital
camera (CCD/CMOS), a motorized microscope table, and
appropriate documentation software, it is useful to take a
sequence of individual pictures in z-axis (the distance between
different focal planes should be 1 μm). Saving each focal plane
in Tiff format, you can merge all files to a film sequence using,
e.g., QuickTime Player; this often facilitates understanding
the spatial complexity of brain NB clones.

4 Notes

1. Glass capillaries for DiI labeling can be produced by any com-

mercial pipette puller, e.g., Sutter P97 which allows both the
heat and the strength to be varied. Change the settings of the
puller according to the operating manual until you get ideal
needles (see, e.g., the comprehensive guide to pulling capillar-
ies on the website of Sutter instruments http://www.sutter.
com/contact/faqs/pipette_cookbook.pdf).
2. Spread the heptane glue out carefully so that you have one thin
but continuous layer of glue. If there is not enough glue, the
embryos will not stick to the cover slip. However, if you apply
too much glue, the embryo sinks into the glue and does also not
stick properly. In addition, the capillary will be clogged by glue.
3. Make sure the frame is big enough so that there is a distance
of at least 4 mm between the embryos and the inner edges of
the frame. Otherwise the capillary cannot be placed near
enough to the embryo. Fix the frame carefully without pro-
ducing air bubbles. Otherwise the oil or PBS will later flow off
the cover slip. After all embryos are fixed on the cover slip,
place it onto a glass slide for better handling.
 DiI-labeling in the Drosophila Embryonic Brain 53

4. You do not need a vast amount of individuals, but as you will

only work with embryos from a 1 h collection at 25 °C, you
need to make sure that your flies lay enough eggs. In order to
guarantee efficient labeling, it is important that you have suf-
ficient numbers of precisely staged embryos. Of course, the
collection schedule is temperature dependent.
5. We do not recommend chemical dechorionization as accord-
ing to our experience this seems to cause misdevelopment or
damage of embryos at higher rates.
6. The orientation of the embryo on the agar block is critical for
the accessibility of the capillary to the neuroectodermal area
one wants to label in. Orientate the embryos on the agar block
by slightly moving them with the tip of the preparation nee-
dle. Consider that the adhesive border of the attached part of
the neuroectoderm is close to the cell you want to label when
you stick the embryo to the cover slip (Fig. 1b). For example,
if you plan to label cells in the ventral neuroectoderm of the
brain, the embryo should be placed sagittally so that the adhe-
sive border is close to the ventral head midline (Fig. 1a).
Accordingly, if you want to label in the dorsal neuroectoderm,
orientate the embryo in the opposite direction. It is also
important for the viability of the embryo that the ectodermal
region that sticks to the glass surface is not too large, as other-
wise the embryo does not develop any further.
7. While mounting the embryos, on the one hand, it is important
to press carefully and vertically so that the embryos are neither
damaged nor altered in position, on the other hand, you need
to press slightly onto the cover slip so that several rows of neu-
roectodermal cells come to lie in one focal plane.
8. The average time for desiccation depends highly on tempera-
ture and humidity. Control the drying of the embryos under a
stereomicroscope. Observe the embryos under a cold light
source until you note two reflecting stripes on each side of the
ventral midline. These stripes will disclose the level of desicca-
tion: at the beginning, these stripes are well defined and
smooth, while after a few minutes, they show tiny folds; once
they are visible, the right level of desiccation has been achieved.
9. To perform targeted labeling of neuroectodermal cells in the
brain, it is necessary to make use of morphological landmarks
which help to recognize the position of individual neuroecto-
dermal cells: easily identifiable are the cephalic furrow, the dor-
sal, and ventral midline. It is also helpful to use the stereotypical
pattern of procephalic mitotic domains as morphological land-
marks, in addition to the abovementioned landmarks. The pro-
cephalic ectoderm can be subdivided into several mitotic
domains, which are characterized as discrete groups of cells that
synchronously enter mitosis [34]. Each mitotic domain has a
distinct time of entry into mitosis and discloses a specific shape.
54 Karoline F. Kraft and Rolf Urbach

Mitotic domains (except domain B) are only recognizable during

their period of mitosis but not before or thereafter. Almost all
procephalic mitotic domains have already completed mitosis
(by early/mid-stage 8) before they give rise to first NBs.
Recently, we were able to establish mitotic domain B, posi-
tioned in the central area of the brain neuroectoderm, as a use-
ful landmark for a systematical DiI labeling of neuroectodermal
cells, and could show that, among others, the NBs of the mush-
room body originate there [28]. Mitotic domains can be visual-
ized in vivo, for example, in ubi-GFPnls embryos: cells of the
mitotically inactive domain B (continuously showing nuclear
GFP) can be distinguished from the surrounding mitotically
active domains which transiently loose nuclear GFP [28].
10. Once the tip of the dye-filled capillary has entered the embryo,
you must completely avoid movements along the anteroposte-
rior axis of the embryo. Otherwise you will destroy tissue or
the labeled cell and risk misdevelopment of the cell clone.
11. You must use a halogen lamp for the inspection. Using a mer-
cury lamp, due to the problem of phototoxicity, will lead to
apoptosis of the labeled cells!
12. Check the behavior of the cell after labeling. The mode of
brain NB formation reveals some differences depending on the
neuroectodermal region a NB originates. For example, except
in mitotic domain B, cells in all other parts of the brain neuro-
ectoderm undergo mitosis before first NBs emerge. If a neuro-
ectodermal cell divides in parallel to the ectodermal surface
(typical for mitotic domain 1 or 5), then it may produce a NB
and an epidermoblast, each generating a subclone, one becom-
ing located in the later brain, the other in the epidermis (per-
haps mapping in a sensory organ). If the neuroectodermal cell
divides perpendicular to the ectodermal surface, then it may
belong to mitotic domain 9. If the cell divides perpendicular
(and asymmetrical) but in subectodermal position, then this is
likely to be a NB producing a ganglion mother cell. It is also
helpful to make notes on the time point of NB formation
(indicating if it is an early or late developing NBs). For further
details on different modes of brain NB formation, see [35].
13. According to our experience, DiI labelings can be followed
until the first hours of the first larval instar since the lipophilic
dye gradually dilutes, and thus fades, in the growing mem-
brane of developing neuronal/glial cells.
14. Do not embed the preparations in pure glycerol nor in other
glycerol-based mounting mediums (e.g., Vectashield).
Lipophilic dyes, such as DiI, dissolve in glycerol. Embedding
can only be done in PBS without glycerol. However, due to
the evaporation of PBS, every supplement should be prepared
individually and immediately prior to the exposure under
microscope.
 DiI-labeling in the Drosophila Embryonic Brain 55

15. As DiI signal may become weaker over time, it is often useful
to document a labeled NB clone before and after the antibody
staining. Because of the relatively long procedure, you risk to
lose the signal.
16. For cell ablations, halogen lamps with rhodamine filters and
transmitted light at once are recommended. In order to recog-
nize the DiI fluorescent signal of the cell(s) to be ablated, and
simultaneously, the surrounding tissue, reduce the intensity of
the transmitted light to an appropriate low level.

Acknowledgments

The authors thank Thomas Kunz, Martin Steimel, and Christof

Rickert for sharing protocols and considerable expertise and David
Jussen for critically reading the manuscript. We are grateful to
Gerd Technau for his general support. This work was supported by
grants from the Deutsche Forschungsgemeinschaft (UR163/2-1
and UR163/3-1) and by a research stipend to K.F.K. from the
FTN of the University Mainz.

References

1. Technau GM, Berger C, Urbach R (2006) 9. Jefferis GS, Marin EC, Stocker RF et al (2001)
Generation of cell diversity and segmental pat- Target neuron prespecification in the olfactory
tern in the embryonic central nervous system map of Drosophila. Nature 414:204–208
of Drosophila. Dev Dyn 235:861–869 10. Lai S-L, Awasaki T, Ito K et al (2008)
2. Skeath JB (1999) At the nexus between pat- Clonal analysis of Drosophila antennal lobe
tern formation and cell-type specification: the neurons: diverse neuronal architectures in the
generation of individual neuroblast fates in the lateral neuroblast lineage. Development 135:
Drosophila embryonic central nervous system. 2883–2893
Bioessays 21:922–931 11. Pearson BJ, Doe CQ (2004) Specification of
3. Urbach R, Technau GM (2004) Neuroblast for- temporal identity in the developing nervous
mation and patterning during early brain devel- system. Annu Rev Cell Dev Biol 20:619–647
opment in Drosophila. Bioessays 26:739–751 12. Yu H-H, Lee T (2007) Neuronal temporal
4. Jacob J, Maurange C, Gould AP (2008) identity in post-embryonic Drosophila brain.
Temporal control of neuronal diversity: com- Trends Neurosci 30:520–526
mon regulatory principles in insects and verte- 13. Yu H-H, Kao C-F, He Y et al (2010) A com-
brates? Development 135:3481–3489 plete developmental sequence of a Drosophila
5. Kao C-F, Lee T (2010) Birth time/order- neuronal lineage as revealed by twin-spot
dependent neuron type specification. Curr MARCM. PLoS Biol 8
Opin Neurobiol 20:14–21 14. Golic KG, Lindquist S (1989) The FLP recom-
6. Lin S, Lee T (2012) Generating neuronal binase of yeast catalyzes site-specific recombi-
diversity in the Drosophila central nervous nation in the Drosophila genome. Cell 59:
system. Dev Dyn 241:57–68 499–509
7. Skeath JB, Thor S (2003) Genetic control of 15. Xu T, Rubin GM (1993) Analysis of genetic
Drosophila nerve cord development. Curr mosaics in developing and adult Drosophila
Opin Neurobiol 13:8–15 tissues. Development 117:1223–1237
8. Baumgardt M, Karlsson D, Terriente J et al 16. Evans CJ, Olson JM, Ngo KT et al (2009)
(2009) Neuronal subtype specification within G-TRACE: rapid Gal4-based cell lineage
a lineage by opposing temporal feed-forward analysis in Drosophila. Nat Methods 6:
loops. Cell 139:969–982 603–605
56 Karoline F. Kraft and Rolf Urbach

17. Lai S-L, Lee T (2006) Genetic mosaic with of Drosophila melanogaster. II. Neuroblast
dual binary transcriptional systems in lineages derived from the dorsal part of the
Drosophila. Nat Neurosci 9:703–709 neuroectoderm. Dev Biol 189:186–204
18. Lee T, Luo L (1999) Mosaic analysis with a 27. Schmid A, Chiba A, Doe CQ (1999) Clonal
repressible cell marker for studies of gene func- analysis of Drosophila embryonic neuroblasts:
tion in neuronal morphogenesis. Neuron neural cell types, axon projections and muscle
22:451–461 targets. Development 126:4653–4689
19. Lee T (2009) New genetic tools for cell lin- 28. Kunz T, Kraft KF, Technau GM, Urbach R
eage analysis in Drosophila. Nat Methods 6: (2012) Origin of Drosophila mushroom
566–568 body neuroblasts and generation of divergent
20. Yu HH, Chen CH, Shi L et al (2009) Twin- embryonic lineages. Development 139:
spot MARCM to reveal the developmental ori- 2510–2522
gin and identity of neurons. Nat Neurosci 29. Bossing T, Technau GM (1994) The fate of
12:947–953 the CNS midline progenitors in Drosophila as
21. Larsen C, Shy D, Spindler SR et al (2009) revealed by a new method for single cell label-
Patterns of growth, axonal extension and axo- ling. Development 120:1895–1906
nal arborization of neuronal lineages in the 30. Campos-Ortega JA, Hartenstein V (1997)
developing Drosophila brain. Dev Biol The embryonic development of Drosophila
335:289–304 melanogaster. Springer, Berlin
22. Luo L (2005) A practical guide: single-neuron 31. Elberger AJ, Honig MG (1990) Double-
labeling using genetic methods. In: Yuste R, labeling of tissue containing the carbocyanine
Konnerth A (eds) Imaging in neuroscience dye DiI for immunocytochemistry. J Histochem
and development. A laboratory manual. Cold Cytochem 38:735–739
Spring Harbor Laboratory, New York, NY, 32. Maranto AR (1982) Neuronal mapping: a
pp 99–110 photooxidation reaction makes Lucifer yellow
23. Honig MG, Hume RI (1986) Fluorescent car- useful for electron microscopy. Science
bocyanine dyes allow living neurons of identi- 217:953–955
fied origin to be studied in long-term cultures. 33. von Bartheld CS, Cunningham DE, Rubel EW
J Cell Biol 103:171–187 (1990) Neuronal tracing with DiI: decalcifica-
24. Honig MG, Hume RI (1989) Dil and DiO: ver- tion, cryosectioning, and photoconversion
satile fluorescent dyes for neuronal labelling and for light and electron microscopic analysis.
pathway tracing. Trends Neurosci 12:333–340 J Histochem Cytochem 38:725–733
25. Bossing T, Udolph G, Doe CQ et al (1996) 34. Foe VE (1989) Mitotic domains reveal early
The embryonic central nervous system lin- commitment of cells in Drosophila embryos.
eages of Drosophila melanogaster. I. Development 107:1–22
Neuroblast lineages derived from the ventral 35. Urbach R, Schnabel R, Technau GM (2003)
half of the neuroectoderm. Dev Biol 179: The pattern of neuroblast formation, mitotic
41–64 domains and proneural gene expression during
26. Schmidt H, Rickert C, Bossing T et al (1997) early brain development in Drosophila.
The embryonic central nervous system lineages Development 130:3589–3606
 Chapter 4

Flybow to Dissect Circuit Assembly in the Drosophila Brain

Nana Shimosako, Dafni Hadjieconomou, and Iris Salecker

Abstract
Visualization of single neurons within their complex environment is a pivotal step towards uncovering the
mechanisms that control neural circuit development and function. This chapter provides detailed technical
information on how to use Drosophila variants of the mouse Brainbow-2 system, called Flybow, for sto-
chastic labeling of cells with different fluorescent proteins in one sample. We first describe the genetic
strategies and the heat shock regime required for induction of recombination events. This is followed by a
detailed protocol as to how to prepare samples for imaging. Finally, we provide specifications to facilitate
multichannel image acquisition using confocal microscopy.

Key words Drosophila, Brainbow, Multicolor cell labeling, Genetics, Immunostaining, Confocal laser
scanning microscopy

1 Introduction

Different neuron subtypes have elaborate axonal and dendritic pro-

cesses with distinct shapes, which are indicative of their specific
functions in neural circuits. Techniques labeling single neurons
within the context of their complex surroundings are thus much
needed additions to the neurobiologist’s toolbox. In 2007, Livet
et al. [1] pioneered the Brainbow system, a genetic multicolor cell
labeling approach for mice, which makes it possible to visualize
neurons and glia in different hues by the stochastic and combinato-
rial expression of three spectral variants of fluorescent proteins
(FP). The Drosophila Brainbow [2] and Flybow [3] systems are
adaptations of this approach for use in flies, containing distinct fea-
tures that take advantage of genetic techniques available in this
model organism. Here, we provide detailed information on how to
use one of these approaches, Flybow, for studies in the brain.
Flybow (FB) is based on the Brainbow-2 strategy (see Fig. 1a). It
relies on the Gal4/UAS system [4] to control transgene expression,
while a FLP recombinase with altered specificity (mFLP5) mediates
inversions and excisions of cassettes flanked by mFRT71 sites [5].

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_4, © Springer Science+Business Media, LLC 2014

57
58 Nana Shimosako et al.

Fig. 1 Schematic representation of Flybow transgenes and color outcomes. Expression of the modified FLP
recombinase mFLP5 induces inversions and excisions of cassettes each flanked by inward-facing mFRT71
recognition sites (black triangles). Cassettes consist of two FP cDNAs with opposite orientations. In FB2.0, an
additional FLP-out cassette flanked by canonical FRT sites facing in the same orientation (grey triangles) pre-
cedes the FP-containing cassettes. (a) A detailed description of FB1.0, FB1.1, and FB2.0 transgenes is provided
in [3]. (b) In FB1.0B, FB1.1B, and FB2.0B transgenes, Cerulean-V5 has been replaced by mTurquoise (mTq)

Each cassette contains two FP-encoding cDNAs in opposing

orientations. FPs are membrane tethered by either Cd8a (cd8) [6]
or the myr/palm (mp) sequence of Lyn kinase [7]. FP sequences
are followed by SV40 or hsp70Ab polyadenylation (pA) signals.
mFLP5 is expressed under the control of the heat shock promoter
(hs-mFLP5) and induces inversions of DNA cassettes by recombin-
ing mFRT71 sites in opposing orientations, or excisions (FLP-out)
by recombining mFRT71 sites in the same orientation. Cassettes
have been subcloned into a vector containing 10 UAS sites [8]; the
FP closest to these sites will be expressed. Flybow 1.0 consists of one
cassette encoding two FPs (mCherry [9] and Cerulean-V5 [10]).
Flybow 1.1 contains two cassettes, each encoding two FPs (EGFP
[11] and mCitrine [12]; mCherry and Cerulean-V5). Flybow 2.0
features an additional stop cassette, flanked by canonical FRT sites
facing in the same orientation, which can be excised by the canoni-
cal FLP recombinase. The stop cassette consists of lamin cDNA,
followed by two HA tag sequences and hsp70Aa and hsp27 pA sig-
nals. Since expression is only induced in cells, in which Gal4 and
canonical FLP expression overlap, sparse labeling is achieved, thus
facilitating the identification of single cells.
The first set of FB constructs relied on the cyan FP Cerulean,
which requires detection with a V5 antibody because of its weak
endogenous fluorescent signal in flies. To eliminate the necessity
 Flybow Approach 59

Fig. 2 Expression of B set of FB transgenes in the adult Drosophila optic lobe. Examples for original transgenes
can be found in [3]. Photoreceptor cells (R1–R8) visualized with mAb24B10 (blue) extend axons into the optic
lobe: R1–R6 axons innervate the lamina; R8 and R7 axons terminate in the medulla [15]. NP4151-Gal4, an
enhancer trap insertion into the Netrin B locus [16, 17], drives expression of FB transgenes in lamina neurons
L3 and medulla neuron subtypes, which extend axons into the lobula and lobula plate. (a, a′) Heat shock appli-
cation induces expression of mFLP5 to trigger inversion events in FB1.0B260b (n = 14), labeling individual neu-
rons with mTurquoise instead of mCherry. (b, b′) In FB1.1B260b (n = 19), excision and inversion events lead to
expression of mCitrine, mCherry, or mTurquoise instead of the default marker EGFP. (c, c′) FB2.0B260b (n = 29)
facilitates sparse labeling of single neurons in a population because it relies on heat shock-induced expression
of FLP for excision of the stop cassette flanked by FRT sites, and of mFLP5 for stochastic expression of the four
fluorescent proteins. Confocal images represent single optic sections. Images in b, b′, c, and c′ were pro-
cessed using channel separation software. Heat shock exposure: 3 × 45 min (a–b′) and 3 × 30 min (c, c′) at
48, 72, and 96 h after puparium formation. Scale bars: 20 μm

for antibody detection, we generated a second set of transgenic

lines, i.e., FB1.0B, FB1.1B, and FB2.0B, in which cd8-tethered
Cerulean was replaced by myr/palm-tethered mTurquoise, a
brighter cyan FP variant [13] (see Figs. 1b and 2). As a further
improvement, we generated additional hs-mFLP5 stocks for our
toolkit. Three are homozygous viable transgene insertions on the
X, second, and third chromosomes, while one line contains a
homozygous lethal insertion on the second chromosome showing
higher recombination efficiency (see Table 1).
Because Flybow is readily compatible with available genetic
loss- and gain-of-function approaches, it can support anatomical
and functional studies of any genetically accessible cell population
60 Nana Shimosako et al.

Table 1
Basic Flybow transgene toolkit

Flybow a Chr. Insertion Related parental stocks Reference

FB1.0 260b 2 Viable [3]
FB1.0 49b 3 Viable [3]
260b
FB1.0B 2 Viable This study
49b
FB1.0B 3 Viable This study
FB1.1 260b 2 Viable [3]
FB1.1 49b 3 Viable [3]
260b
FB1.1B 2 Viable This study
FB1.1B 49b 3 Viable This study
FB2.0 260b 2 Viable hs-FLP 1; FB2.0260b [3]
49b 1 49b
FB2.0 3 Viable hs-FLP ; FB2.0 [3]
260b 1 260b
FB2.0B 2 Viable hs-FLP ; FB2.0B This study
FB2.0B 49b 3 Viable hs-FLP 1; FB2.0B49b This study
hs-mFLP Chr. Insertion Related parental stocks Efficiencyb
hs-mFLP5 2 Lethal hs-mFLP5/Gla Bc; TM2/TM6B ** [3]
hs-mFLP5 3 Lethal Gla Bc/CyO; hs-mFLP5/TM2 ** [3]
hs-mFLP5 MH15 X Viable ** This study
hs-mFLP5 MH1 2 Lethal *** This study
MH12
hs-mFLP5 2 Viable ** This study
hs-mFLP5 MH3 3 Viable ** This study
a
260b and 49b indicate the attP site-containing loci used for FB transgene insertion on the second and third chromo-
somes, respectively [3]
b
Recombination efficiencies were estimated by monitoring excision events of a stop cassette flanked by mFRT71 sites
in 3rd instar larval eye imaginal discs
**
good efficiency
***
very good efficiency

in the nervous system and other tissues [3]. While Flybow also can
be applied to studies in embryos, in this chapter, we focus on its
use in the larval, pupal, and adult nervous system.

2 Materials

Prepare all solutions with ultrapure deionized water (18.2 MΩ cm

resistivity) from a water purification system.

2.1 Genetic Crosses 1. Plastic vials containing standard cornmeal/agar medium.

and Clone Induction 2. Virgin and male flies with the proper genotype (see Table 1).
3. Water bath at 37 °C.
 Flybow Approach 61

2.2 Buffers 1. Phosphate-buffered saline (PBS, 130 mM NaCl, 7 mM

Na2HPO4⋅7H2O, 3 mM NaH2PO4⋅H2O, pH 7.4). Prepare
10× PBS stock solution: 75.97 g NaCl, 18.76 g Na2HPO4⋅7H2O,
4.14 g NaH2PO4⋅H2O. Add water to a volume of 990 mL.
Dissolve crystals by stirring with a magnetic stirrer and adjust
pH with drops of concentrated 12 N HCl and 10 M NaOH,
and subsequently 1 M dilutions. Fill up to 1 L with water.
Store this solution at room temperature in a glass bottle (see
Note 1). To prepare 1 L PBS solution, mix 100 mL 10× PBS
stock solution with 900 mL water and readjust pH using 1 N
HCl or 1 M NaOH if necessary. Store at 4 °C.
2. Phosphate-buffered saline with Triton (PBT, 0.5 %): 100 mL
PBS, 0.5 mL Triton®-X-100 (Sigma). Add Triton®-X-100 by
using a 1 mL syringe and mix vigorously with a magnetic stir-
rer to dissolve the detergent. Store solution at 4 °C.
3. Phosphate buffer (PB, 0.1 M, pH 7.4): Weigh 1.73 g
NaH2PO4⋅H2O into a glass beaker, add 125 mL distilled
water, and stir until crystals are dissolved to make a 0.1 M
NaH2PO4 solution. Weigh 13.4 g Na2HPO4⋅7H2O into a sec-
ond larger glass beaker, add 500 mL water, and stir to make a
0.1 M Na2HPO4 solution. Store 100 mL of 0.1 M Na2HPO4
solution in a separate glass bottle. Pour the 0.1 M NaH2PO4
solution into the remaining 0.1 M Na2HPO4 solution until
pH reaches pH 7.4 to make 0.1 M PB. Store solutions at 4 °C.
4. Phosphate buffer with lysine (PBL): Dissolve 3.6 g L-Lysine-
HCl (Sigma) in 100 mL water. Add approximately 40 mL
0.1 M Na2HPO4 solution until pH 7.4 is reached. Add 0.1 M
PB to make up 200 mL. Filter-sterilize using a 250 mL filter
unit (0.2 μm pore size, 50 mm membrane diameter, Nalgene,
Cat. # 126-0020). Store solution at 4 °C for up to 3 months.
5. Blocking buffer (10 % NGS in PBT): 1 mL normal goat serum
(NGS, Sigma), 9 mL PBT (see Note 2).
2.3 Fixative 1. Paraformaldehyde stock solution (8 %): 0.8 g paraformalde-
hyde (PFA, EM grade powder), 10 mL water, 70 μL of 1 M
NaOH. Weigh PFA powder in 50 mL self-standing centrifuge
tube (Corning). Wear a mask and gloves when weighing para-
formaldehyde to minimize exposure, as the powder is classi-
fied as carcinogenic. Add 10 mL water and 70 μL of 1 M
NaOH. Shake gently and microwave for 10 s with the cap
slightly loosened. Fully dissolve in a 37 °C water bath for
30–60 min with occasional shaking of the tube. Filter into a
fresh 15 mL centrifuge tube using a sterile hydrophilic syringe
filter with 0.2 μm pore size (Sartorius). Store solution for up
to 7 days at 4 °C (see Note 3).
2. Phosphate buffer with lysine and paraformaldehyde (PLP, 2 %
PFA): 3 mL PBL, 1 mL 8 % PFA stock solution. Make up
freshly prior to use.
62 Nana Shimosako et al.

2.4 Antibodies 1. Primary antibodies: mouse monoclonal antibody anti-V5

(1:500 dilution in blocking buffer; Invitrogen) to visualize
Cerulean expression (see Note 4).
2. Secondary antibodies: goat anti-mouse F(ab′)2 fragments
conjugated with Cy5 or Alexa Fluor®647 (Jackson
ImmunoResearch Laboratories) used as a dilution of 1:200 in
blocking buffer (see Note 5).

2.5 Dissections, 1. Stainless steel No. 5 forceps.

Immunostaining 2. Mesh baskets.
and Mounting
3. 24-well multidish (e.g., Nunclon™ Δ surface).
4. Terasaki plate (e.g., Nunc 60-well MicroWell™ MiniTray).
5. Parafilm®.
6. Sarstedt microtubes with O-ring screw caps.
7. Fine nylon mesh.
8. Rotating titer plate shaker (e.g., Thermo Scientific MiniMix
microplate shaker).
9. Aluminum foil.
10. Vectashield (Vector Laboratories).
11. Microscope slide (SuperFrost®).
12. Cover slip (18 × 18 mm, No. 1 or 1.5).
13. Soft modeling mass (e.g., Fimo).
14. Nail polish.
15. Medical wipes.
16. Glas pipettes.
17. Sylgard (Dow Corning, 184 silicone elastomer kit).
18. Dissecting and fluorescence dissecting microscopes.

2.6 Image 1. Confocal laser scanning microscope.

Acquisition 2. Long-range objectives (e.g., Leica 20× (0.7 NA) air, 40×
and Analysis (1.25 NA), and 100× (1.46 NA) oil objectives).
3. Cargille immersion oil Type DF.
4. Confocal image acquisition software: (e.g. Leica LAS
software).
5. Image analysis software (e.g. Volocity Improvision PerkinElmer;
Image J; Fiji including Simple Neurite Tracer plug-in).

3 Methods

3.1 Genetic Crosses 1. Build and expand the driver stock, which contains both an hs-
and Clone Induction mFLP5 and a Gal4 transgene active in a cell type, brain area,
Protocol or other tissue of interest (see Note 6).
 Flybow Approach 63

Fig. 3 Example of experimental strategy to induce recombination events. After setting up or transferring the
cross into a fresh vial (t = 0), flies are left to lay eggs for 1 day (t = 24 h after egg laying (AEL)). Subsequently,
parents are transferred into a fresh vial or removed. Progeny in vials are heat shocked up to three times at 48,
72, and 96 h AEL in a 37 °C water bath. Brains can be dissected from larvae, staged pupae, or adults. APF, after
puparium formation

2. Collect male and unfertilized female flies from parental stocks

for about 4 days (see Note 7). Set up crosses in plastic vials con-
taining standard cornmeal/agar medium with about 10–12
females and 5–6 males per vial. The crosses should be designed
so that progeny will have one Gal4, one hs-mFLP5, and one
FB1.0 or FB1.1 transgene. Note that for FB2.0 an additional
transgene is required that expresses canonical FLP recombinase,
such as hs-FLP1 (see Fig. 3). FB and hs-mFLP5 transgenes are
available as insertions on different chromosomes to facilitate
genetic crosses (see Table 1). We usually set up three to five par-
allel crosses for a given experiment. After a period of 24 h of egg
laying, transfer parents into fresh vials. Repeat this process for
maximally 1 week. All vials are kept in an incubator at 25 °C.
3. To induce recombination events (see Fig. 3), progeny of
crosses are heat shocked by placing the vials into a water bath
with the temperature set to 37 °C (see Note 8). The water
level should be sufficiently high, so that vials, when weighed
down, are immersed well above the food level to ensure that
also larvae leaving the medium are exposed to heat.

3.2 Immunolabeling 1. Dissect 3rd instar larval, pupal, or adult brains (see Note 9) in
drops of cold PBS using forceps on a dissecting pad. To maxi-
mize the number of dissected samples with recombination
events, some Gal4 drivers make it possible to preselect flies with
clones under a fluorescence-dissecting microscope. Transfer
the brains into a glass embryo dish filled with PBS using forceps
or a glass pipette. Keep the dish on ice until all samples of a
given genotype are dissected (see Notes 10 and 11).
2. Transfer the brains into a mesh basket (see Note 12) posi-
tioned in one well of a 24-well multidish filled with 1 mL PLP.
Fix for 1 h at room temperature (see Note 13).
64 Nana Shimosako et al.

3. Wash the brains by transferring the mesh baskets sequentially

into three wells filled with 1 ml PBT each. Keep baskets in the
first two wells for a few seconds each (short washes) and in the
third well for at least 15 min (long wash) (see Note 14).
4. Block tissue for at least 15 min by transferring the baskets into
a new well containing blocking buffer (10 % NGS in PBT).
5. For incubation in primary antibody (see Note 15), transfer
brains into the wells of a Terasaki plate, each filled with 10 μL
of antibody diluted in blocking buffer. Distribute the brains
over multiple wells, with each well not containing more than 10
brains (see Note 16). Add a stripe of moist filter paper on one
side of the plate, close and seal tightly with Parafilm®. Place the
plate into a humid chamber, such as a small closable box with a
moist paper towel at the bottom. Keep on a gently horizontally
rotating titer plate shaker overnight at 4 °C in a cold room.
6. Return the brains into the mesh baskets placed into 24-well
plates filled with PBT. Wash for 15 min by transferring the
baskets three times into the fresh wells for two short and one
long washes.
7. For incubation in secondary antibodies, transfer the baskets
into a new well containing the antibody diluted in blocking
buffer. Wrap the 24-well plate with aluminum foil and keep on
a titer plate shaker for 2.5 h at room temperature.
8. Transfer the baskets into the fresh wells: two filled with PBT
for two short washes, and two filled with PBS for one short
and one long wash. In total, wash at least for one hour.
Immunolabeled brains should be stored in PBS at 4 °C, not in
mounting media, to avoid softening of the tissue prior to
mounting on slides.

3.3 Mounting 1. Take immunolabeled brains with clones individually out of

Samples PBS and place in a drop of about 30 μL cold Vectashield on a
pre-cleaned microscope slide.
2. Take a cover slip (18 × 18 mm, No. 1 or 1.5), place small pieces
of soft modeling mass on the four corners of the cover slip to
prevent squashing of the brains.
3. Using forceps, position the cover slip with the modeling mass
pieces facing down on the sample. Gently press the corners.
Then place the slide under a fluorescence dissecting
microscope.
4. By pressing the forceps against the cover slip edges, move the
cover slip gently up and down or left and right to roll at least
one brain hemisphere into the correct orientation.
Subsequently, brains can be gently flattened by pressing down
the cover slip corners. Seal slide with clear nail polish. Store
slides at 4 °C in the dark.
 Flybow Approach 65

Table 2
Example of a confocal microscopy scanning method for imaging FB transgenes

Fluorescent PMTb Excitation Emission Excitation Emission range

Scan proteina/dye (e.g.) maximum (nm) maximum (nm) laser line (nm) imaged (nm)
1 mCitrine 2 516 529 514 525–565
Alexa Fluor®647 4 651 667 633 675–735
2 mCherry 3 (HyD) 587 610 561 572–630
3 EGFP 1 488 507 488 497–515
4 mTurquoise 3 (HyD) 434 475 458 460–495
a
Detailed information about the FPs used in the Flybow approach can be found in these references: EGFP [11],
mCitrine [12], mCherry [9], and mTurquoise [13]
b
PMT, photomultiplier tube

3.4 Image 1. Collect images using a confocal laser scanning microscope

Acquisition equipped with high-quality long-range objectives. For oil
and Analysis objectives, use immersion oil suitable for immunofluorescence
microscopy recommended by the confocal microscope pro-
vider. Set up a confocal microscope imaging method that
combines sequential and simultaneous scan modes (see Note
17). For each FP or dye, select a laser line suitable for optimal
excitation, as well as detection windows or acousto-optical
beam splitter [AOBS] settings for collecting specific emission
signals. An example of such a method is provided in Table 2.
For the original set of FB transgenes, mCitrine and Cy5/Alexa
Fluor®647 signals are collected using a simultaneous scan
mode, followed by sequential scans of EGFP and mCherry
signals. For set B of FB transgenes, a fourth sequential scan is
added to image mTurquoise. To obtain optimal signals and
minimal cross talk between channels, adjust the power of indi-
vidual laser lines for each sample. The detection windows may
also require fine-tuning depending on the experiment or
microscope used. Assign a color to each channel, e.g., EGFP,
green; mCitrine, yellow; mCherry, red; Cerulean-V5/Cy5,
medium blue; and mTurquoise, cyan.
2. For samples displaying extensive bleed-through between chan-
nels, images can be processed using channel separation soft-
ware. We use the Leica LAS AF suite channel separation tool
for this purpose. In representative images of each detection
channel, regions of interest with unambiguous strong but not
saturated signals are manually selected for each FP or dye.
Using these obtained values, the software algorithms subtract
unspecific proportions of detected signals.
66 Nana Shimosako et al.

3. Confocal images can be further analyzed and processed using

Volocity (Improvision PerkinElmer) and ImageJ (Fiji) soft-
ware to perform z projections of selected sections. Fiji can be
used for brightness and contrast adjustments of the four or five
channels in color composite images, as well as for subsequent
conversion into RGB images. The Simple Neurite Tracer plug-
in is highly useful for the reconstruction of individual neurons
from stacks.

4 Notes

1. We store most of our solutions in autoclaved standard screw

cap laboratory glass bottles.
2. We use higher-quality goat serum to extend the lifetime of
solutions.
3. For optimal fixation of tissue, we use EM grade paraformalde-
hyde powder (Polysciences, Cat. # 0030) in our protocol.
Some investigators freeze paraformaldehyde aliquots.
However, we prepare fresh solutions at a weekly basis to obtain
best staining results.
4. Although Cerulean had been described as the best CFP deriv-
ative with respect to brightness, quantum yield, and oligomer-
ization properties [10], consistent with other studies [2], we
observed that this FP is less suitable for studies in Drosophila
because of low endogenous emission signals. Cerulean is
therefore visualized using anti-V5 primary and Cy5- or Alexa
Fluor®647-conjugated secondary antisera.
5. F(ab′)2 fragment-based secondary antibodies ensure even
immunolabeling throughout the entire brain, because their
smaller size facilitates penetration deeper into the tissue.
6. To build stable stocks, we prefer to combine hs-mFLP5 trans-
genes with the Gal4 driver instead of FB transgenes, as con-
tinuous low levels of recombinase expression may cause
transmittable recombination events in the germ line. Stocks
containing both hs-FLP1 and FB2.0 transgenes (see Table 1)
should be continuously monitored. Note that the two hs-
mFLP5 transgenes described in [3] are homozygous lethal
insertions and are therefore kept over a balancer.
7. As the occurrence of hs-mFLP5-induced recombination events
decreases with the parental age of flies, maximally 4-day-old
adult flies should be used for genetic crosses. The first emerg-
ing progeny of a cross should be prepared for analysis to maxi-
mize the number of samples with recombination events. In a
24-h egg collection, these correspond most closely to the
progeny, at which the heat shock protocols were aimed.
 Flybow Approach 67

8. Color outcomes are influenced by the time points, duration,

and number of heat shocks. These parameters can be adjusted
for each Gal4 line, cell or brain area of interest, and experi-
mental aim. While early heat shocks lead to recombination
events in dividing progenitors and thus tend to label larger
lineage-related groups of cells with the same FP, later or
shorter heat shocks facilitate labeling of single cells. Repeated
heat shocks further promote labeling of neurons or glia with
different FPs. We observed that lineage-related cells born in a
narrow time window are difficult to separate by the expression
of different FPs using the FB approach. For studies in the
optic lobe, the earliest time point, at which larvae are sub-
jected to heat shock is 48 h after egg laying. The time point,
at which crosses are transferred into a fresh vial is defined as
t=0. In a regime requiring multiple heat shocks, this proce-
dure was repeated at 72 and 96 h. The length of heat shocks
ranges from 30 to 45 min. When combining the FB approach
with mosaic analysis with a repressible cell marker (MARCM,
[14]), it is useful to extend heat shocks to 90 min [3].
9. For the visual system, large late third instar larvae, which
stopped wandering on the side of vials, are selected for dissec-
tions. To stage pupae, white pupae are collected from vials at
1-h intervals, placed on grape-juice agar plates to avoid desic-
cation, stored in a 25 °C incubator and dissected at specific
time points after puparium formation. Pharate adult flies are
dissected shortly before eclosion to avoid tracheal filling.
10. Brains are dissected on plates coated with Sylgard® (Dow
Corning, 184 silicone elastomer kit). We use the lids of Terasaki
plates for this purpose. When transferring the brains with a
glass or plastic pipette, keep samples in the tip of the pipette,
filled with PBS, to avoid losing them during the process.
11. Keep unfixed brains for not longer than 30 min on ice, as
extended storage can cause connectivity defects. Fixed brains can
be stored in PBT while completing all dissections for the day.
12. Mesh baskets are handmade using conical 1.5 mL Sarstedt
microtubes with O-Ring Screw Caps and a fine nylon mesh.
Cut microtubes at the bottom border of the ridged part using a
pair of sharp scissors, so that the 24-well plate can still be closed
with the lid. Carefully, cut off the top of the screw cap below
the O-Ring using a sharp razor blade. Cut out a 2 × 2 cm square
of nylon mesh. Place the mesh on top of the cut tube, and screw
the cap ring in place while tightening the mesh. Remove all
fabric on the outside of the baskets using a razor blade. Newly
made baskets should be extensively washed in PBT followed by
deionized water to avoid sticking of brains. After each use, bas-
kets should be thoroughly washed by soaking and rinsing them
in deionized water. They can be reused for years.
68 Nana Shimosako et al.

13. Fixation for 1 h in PLP at room temperature is central to obtain-

ing good signals both from FPs and fluorophore-conjugated
secondary antibodies. We observed that fixation using 4 % PFA
even for 30 min significantly quenched FP signals.
14. Even if solely endogenous FP signals are collected and immuno-
labeling steps are not required, brains should still be washed
extensively in PBT after fixation before storage in PBS. Otherwise
the tissue will shrink when transferred into mounting media.
15. For the second set of FB transgenes, immunolabeling with
anti-V5 is not required. Instead, additional antibody stainings,
e.g., mAb24B10 (1:75 dilution in blocking buffer;
Developmental Studies Hybridoma Bank) can be performed
and visualized with a Cy5 or Alexa Fluor®647-coupled sec-
ondary antibody.
16. To avoid damaging brains during the transfer into the Terasaki
plate wells, position brains in the remaining liquid between
slightly held apart and curved forceps tips. To minimize the
transfer of excessive PBT, gently blot up the liquid held
between the forceps arms using a rolled up corner of a medical
wipe (e.g., Kimberly-Clark Professional).
17. We optimized imaging conditions for a Leica TCS SP5 II
upright laser confocal microscope equipped with a resonant
scanner and four photomultiplier tubes (PMTs). However, all
confocal microscope models are suitable, which have the five
listed laser lines and whose detection windows can be adjusted.
We typically collect image stacks using 1,024 × 1,024 pixels
image size, 200 Hz line speed, and 5-line or 4-frame averages.
To accelerate image acquisition, a bidirectional scan mode is
used. Although not essential, the use of the resonant scanner
provides an alternative way to increase the speed of image acqui-
sition of large z stacks and thus to minimize photobleaching.
When using the resonant scanner, images are collected at the
fixed speed of 8 kHz and averaged 96 times. If the microscope
is equipped with hybrid detection (HyD) GaAsP technology,
scan the weakest fluorescent signals using the HyD PMT(s).

Acknowledgements

We thank J. Goedhart for sharing the mTurquoise cDNA. D.H.

developed the original FB transgenes, while N.S. generated the
second set of FB transgenes and validated the new hs-mFLP5
insertions. The original Flybow approach was developed in
collaboration with B.J. Dickson, S. Rotkopf, C. Alexandre, and D.M.
Bell. We are grateful to H. Apitz, B. Richier, F. Rodrigues, and
S. Schrettenbrunner for critical reading of this manuscript. This
work is supported by the Medical Research Council (U117581332).
 Flybow Approach 69

References
1. Livet J, Weissman TA, Kang H, Draft RW, Lu J, Improved monomeric red, orange and yellow
Bennis RA et al (2007) Transgenic strategies for fluorescent proteins derived from Discosoma
combinatorial expression of fluorescent pro- sp. red fluorescent protein. Nat Biotechnol
teins in the nervous system. Nature 450:56–62 22:1567–1572
2. Hampel S, Chung P, McKellar CE, Hall D, 10. Rizzo MA, Springer GH, Granada B, Piston
Looger LL, Simpson JH (2011) Drosophila DW (2004) An improved cyan fluorescent
Brainbow: a recombinase-based fluorescence protein variant useful for FRET. Nat
labeling technique to subdivide neural expres- Biotechnol 22:445–449
sion patterns. Nat Methods 8:253–259 11. Shaner NC, Steinbach PA, Tsien RY (2005) A
3. Hadjieconomou D, Rotkopf S, Alexandre C, guide to choosing fluorescent proteins. Nat
Bell DM, Dickson BJ, Salecker I (2011) Methods 2:905–909
Flybow: genetic multicolor cell labeling for 12. Griesbeck O, Baird GS, Campbell RE,
neural circuit analysis in Drosophila melano- Zacharias DA, Tsien RY (2001) Reducing the
gaster. Nat Methods 8:260–266 environmental sensitivity of yellow fluorescent
4. Brand AH, Perrimon N (1993) Targeted gene protein. Mechanism and applications. J Biol
expression as a means of altering cell fates and Chem 276:29188–29194
generating dominant phenotypes. Development 13. Goedhart J, van Weeren L, Hink MA, Vischer
118:401–415 NO, Jalink K, Gadella TW Jr (2010) Bright
5. Voziyanov Y, Konieczka JH, Stewart AF, cyan fluorescent protein variants identified by
Jayaram M (2003) Stepwise manipulation of fluorescence lifetime screening. Nat Methods
DNA specificity in Flp recombinase: progres- 7:137–139
sively adapting Flp to individual and combina- 14. Lee T, Luo L (1999) Mosaic analysis with a
torial mutations in its target site. J Mol Biol repressible cell marker for studies of gene func-
326:65–76 tion in neuronal morphogenesis. Neuron
6. Liaw CW, Zamoyska R, Parnes JR (1986) 22:451–461
Structure, sequence, and polymorphism of the 15. Hadjieconomou D, Timofeev K, Salecker I
Lyt-2 T cell differentiation antigen gene. J (2011) A step-by-step guide to visual circuit
Immunol 137:1037–1043 assembly in Drosophila. Curr Opin Neurobiol
7. Zacharias DA, Violin JD, Newton AC, Tsien 21:76–84
RY (2002) Partitioning of lipid-modified 16. Hayashi S, Ito K, Sado Y, Taniguchi M,
monomeric GFPs into membrane microdo- Akimoto A, Takeuchi H et al (2002) GETDB,
mains of live cells. Science 296:913–916 a database compiling expression patterns and
8. Dietzl G, Chen D, Schnorrer F, Su KC, molecular locations of a collection of Gal4
Barinova Y, Fellner M et al (2007) A genome- enhancer traps. Genesis 34:58–61
wide transgenic RNAi library for conditional 17. Timofeev K, Joly W, Hadjieconomou D,
gene inactivation in Drosophila. Nature 448: Salecker I (2012) Localized netrins act as posi-
151–156 tional cues to control layer-specific targeting of
9. Shaner NC, Campbell RE, Steinbach PA, photoreceptor axons in Drosophila. Neuron
Giepmans BN, Palmer AE, Tsien RY (2004) 75:80–93
 Chapter 5

Immunofluorescent Labeling of Neural Stem Cells

in the Drosophila Optic Lobe
Benjamin Perruchoud and Boris Egger

Abstract
The Drosophila visual system is an excellent model system to study the switch from proliferating to differ-
entiating neural stem cells. In the developing larval optic lobe, symmetrically dividing neuroepithelial cells
transform to asymmetrically dividing neuroblasts in a highly ordered and sequential manner. This chapter
presents a protocol to visualize neural stem cell types in the Drosophila optic lobe by fluorescence confocal
microscopy. A main focus is given on how to dissect, fix, immunolabel, and mount brains to reveal cellular
morphology during early larval brain development.

Key words Neural stem cell, Drosophila, Brain, Optic lobe, Immunofluorescent labeling

1 Introduction

The developing Drosophila visual system or optic lobe is an attrac-

tive model system to study the transition from symmetric to asym-
metric neural stem cell division. The larval optic lobe is comprised
of an outer proliferation center that gives rise to the lamina and
the outer medulla and an inner proliferation center that generates
the inner medulla, lobula, and lobula plate [1]. In recent years the
work of many research groups has shown that neurogenesis in the
optic lobe resembles neural development in the mammalian brain
and the retina [2–10]. In the optic lobe symmetrically dividing
neuroepithelial cells progressively transform to asymmetrically
dividing neuroblasts. This cell fate transformation is reminiscent of
the transition of neuroepithelial cells to radial glial cells in the
mammalian brain [11].
Here we present a detailed protocol to visualize optic lobe
development by immunolabeling for fluorescent confocal micros-
copy. We specifically focus on early larval stages and describe how
to dissect, fix, immunolabel, and mount 1st, 2nd, and early 3rd
instar brains. In the described method, the dissected brains are
attached to a cover slip during the entire procedure, which has

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_5, © Springer Science+Business Media, LLC 2014

71
72 Benjamin Perruchoud and Boris Egger

several advantages. Firstly, it prevents loosing the rather small

brains during immunolabeling and washing procedures. Secondly,
the brains can be orientated on the cover slips in a fixed position,
which facilitates the imaging procedure. Thirdly, the method
allows simultaneously assessing brains of various different stages
and genotypes in rapid and efficient manner.

2 Materials

2.1 Staged Larval 1. Fly cages.

Collection 2. Apple juice embryo collection plates with wet yeast.
3. Petri dishes 35 mm filled with fly food with wet yeast.

2.2 Preparation 1. Hot plate.

of Coated Cover Slips 2. 22 × 22 cover slips thickness # 1.
3. 0.1 % w/v Poly-L-Lysine in water (Sigma).

2.3 Larval Brain 1. Fine forceps (Dumont # 5, Fine Science Tools).

Dissection 2. Pipettes and sterile tips.
3. Paper tissue.
4. Two needles (25G or 30G) on cotton buds.
5. Petri dishes (plastic).
6. Phosphate-buffered saline, 1× PBS: 10 mM Na2HPO4,
2.68 mM KCl, 140 mM NaCl (PBS Tablets, Gibco,
Invitrogen). pH is 7.45. Sterilize by autoclaving.

2.4 Fixation 1. Fixative: 4 % formaldehyde, 0.5 mM EGTA, 5 mM MgCl2 in

and Immunolabeling 1× PBS.
2. PBST: 0.3 % Triton X-100 in 1× PBS. Store at 4 °C.
3. Primary antibodies (e.g., anti-Discs large 4F3 (1:100),
Developmental Studies Hybridoma Bank (http://dshb.biology.
uiowa.edu/)).
4. Secondary antibodies (e.g., Goat anti Mouse Alexa488,
Molecular Probes).
5. Columbia staining jars.
6. Rotating platform (shaker).
7. Humid chamber (e.g., empty Vectashield box).

2.5 Mounting 1. Microscopy slides.

and Visualization 2. Vaseline.
3. Syringe with 18G needle.
 Immunofluorescent Labelling on Drosophila Optic Lobes 73

4. Mounting media (Vectashield).

5. Transparent nail varnish.
6. Fluorescent (confocal) microscope.

3 Methods

3.1 Preparing 1. Prepare 0.05 % w/v Poly-L-Lysine in ddH2O in a microfuge

Poly-L-Lysine-Coated tube.
Cover Slips 2. Place cover slips on a hot plate.
3. Add 10 μl of 0.05 % Poly-L-Lysine diluted in ddH2O to each
slide and let the drop dry (see Note 1).
4. Immediately after the first drop is dry, add another 10 μl of
0.05 % Poly-L-Lysine on top of the first drop and let it dry again.
5. Store slides in a dust-free environment (e.g., in a closed petri
dish or cover slip holder).

3.2 Staging 1. Collect embryos on apple juice plates for 4–6 h and let embryos
of Larvae develop at 25 °C (see Note 2).
2. 24 h after midpoint of egg collection, pick about 80–100
freshly hatched larvae and place them in a food plate contain-
ing a drop of wet yeast.
3. Let larvae develop at 25 °C to the appropriate larval stage
(e.g., 24 h ALH, after larval hatching or 48 h ALH).

3.3 Dissection 1. Pick the larvae from food plate and place them on a paper tis-
of Larval Brains sue soaked with PBS (see Note 3).
2. Place several drops of 1× PBS in a circle on the inside of a
plastic petri dish lid. Add one larva to each drop (see Fig. 1a).
3. To dissect small larvae (0–48 h ALH), use two syringe needles
on cotton buds (see Fig. 1a). Carefully hold the larva in place
with one needle pressing down the middle of the larval body.
Stick the tip of the other needle firmly into the very anterior
part of the larva where the mouth hooks are located. Draw the body
slowly away from the sticked down mouthpart. As the anterior
head part breaks away from the rest of the body, the brain spills
out. The brain might be still attached to the mouth hooks.
Dissect off the mouth hooks and imaginal discs as necessary
(see Note 4). To dissect larger larvae (72–96 h ALH), use fine
forceps and the so-called “inverted sock” technique. Pull the
larva in the middle apart and discard posterior part. Grab the
larval mouth hooks at the anterior half with one forceps and
invert the larval cuticle over the tip of the forceps in a manner
like to invert a sock. The interior organs of the larva are now
on the outside and the brain can be further dissected.
74 Benjamin Perruchoud and Boris Egger

a b

c d

1 2 3
f

g
h

Fig. 1 (a) Larvae are placed in a drop of 1× PBS in a petri dish lid and dissected by using needles on cotton
buds or fine forceps. (b) Dissected brains are transferred by using a pipette (small brains) or forceps (larger
brains) and oriented in desired position on Poly-L-Lysine-coated cover slips. (c) Columbia jar holds maximal
four cover slips. (d) Cover slips with attached brains are transferred to Columbia jars for fixation and washing
steps. (e) A humid chamber is used for primary and secondary antibody incubation. (f) After the final washing
step, a strip of Vaseline is added to each cover slip corner as a spacer. (g) Cover slip is carefully placed on a
slide containing a drop of Vectashield and corners are gently pushed down. (h) Samples can be sealed with
transparent nail varnish and stored at 4 °C
 Immunofluorescent Labelling on Drosophila Optic Lobes 75

4. Place a drop of PBS onto the coated cover slip. For small
brains (0–48 h ALH), cut off the very tip of a 200 μl pipette
tip and transfer the brains to the cover slip (see Note 5 and
Fig. 1b). For larger brains (72–96 h ALH) use the forceps to
transfer the brains.
5. Place brains to the Poly-L-Lysine-coated area in the desired
orientation by using cotton bud needles or forceps (see Note 6).

3.4 Fixation 1. Rinse cover slips with attached brains in a Columbia jar
and Staining (see Fig. 1c) containing 1× PBS.
2. Fix brains on cover slips in a Columbia jar containing 4 ml
fixative by rotating on a platform for 18 min (see Fig. 1d).
3. Replace fixative by PBST ~6 ml and wash the brains 1 × 1 min,
2 × 5 min, and 1 × 15 min on a rotating platform.
4. Place the cover slips into a humid chamber and add primary
antibodies (see Note 7) diluted in PBST (100 μl total volume)
on top of slide (see Fig. 1e). Close lid of humid chamber and
incubate overnight at 4 °C (see Note 8).
5. Wash the cover slips in a Columbia jar with PBST 1 × 1 min,
3 × 5 min, and 2 × 15 min.
6. Place the cover slips in a humid chamber and add secondary
antibodies diluted in PBST (100 μl total volume) on top of the
slide. Close the lid of humid chamber and incubate overnight
at 4 °C (see Note 9).
7. Wash cover slips in a Columbia jar with PBST 1 × 1 min,
3 × 5 min, and 3 × 15 min.

3.5 Mounting 1. Transfer cover slips using a forceps from the Columbia jar and
rinse off PBST with ddH2O from top of cover slips (see Note 10).
2. Remove excess liquid on a paper tissue and add a strip of
Vaseline to each corner as a space holder by using the syringe
(see Fig. 1f).
3. Add a drop of Vectashield on a clean microscopy slide.
4. Place cover slips with attached brains onto microscopy slides
and tap or gently press down each corner using cotton buds
(see Note 11 and Fig. 1g).
5. Seal cover slips with transparent nail varnish (see Fig. 1h).
6. Samples can be stored in slide folder at 4 °C.
76 Benjamin Perruchoud and Boris Egger

4 Notes

1. In order to use smaller volumes of fixative (4 ml), add the Poly-

L-Lysine drop towards one side of the cover slips (see Fig. 1b).

2. For more accurate staging, you may want to make a 1 h pre-

collection for eggs that have been retained and developed in
the female abdomen before starting the experimental
collection.
3. Larvae clean themselves of yeast and fly food by moving
around on a wet paper tissue.
4. Leaving some imaginal discs attached can help to stick the
brains onto cover slips in the desired orientation.
5. Small brains often stick to the pipette tip. A trick is to coat the
tip with cheek cells of your own saliva. Pipette saliva up and
down and rinse once in 1× PBS before transferring the brains.
6. At this stage it helps to draw a brain orientation map of each
cover slip. So the samples can be distinguished at a later time
point. Alternatively, remember the position of each cover slip
in the Columbia jar (e.g., wild type, position 1, mutant, posi-
tion 2). Marker pen writings are washed off in PBST and are
therefore not suitable to label cover slips.
7. We usually use a monoclonal mouse antibody against the
protein Discs large (1:100), which is available from the
Developmental Studies Hybridoma Bank (http://dshb.biology.
uiowa.edu/). Discs large is localized to cell cortices in neuro-
blasts and septate junctions in neuroepithelial cells and is very
suitable to visualize cell morphology in the larval optic lobe. To
visualize optic lobe precursor cells, GAL4 lines such as GAL4c855a
can be used to drive UAS-GFP constructs (see Fig. 2).
8. For most primary antibodies, best results are obtained with
overnight incubation; however, primary antibody incubation
for 3 h at room temperature can also give good results.
Different conditions should be tested in cases where immuno-
labelings are unsatisfactory.
9. For better tissue penetration, secondary antibodies are also
applied overnight at 4 °C.
10. PBST should be quickly washed off the top of the cover slip
(not the side on which brains are attached) with ddH2O
because dried out PBST will leave a salt crust on cover slips.
11. Do press down gently only to leave morphology and shape of
brains intact.
Fig. 2 Morphology and proliferation centers of the developing optic lobe revealed by immunofluorescent label-
ing. (a) A late third instar larval central nervous system (CNS): ventral nerve cord (VNC), central brain (CB), and
optic lobes (OL). Subsequent images show frontal confocal sections, as shown in the inset diagram (OPC in
green). Anterior and posterior refer to the neuraxis of the larval CNS. (b) A frontal section through a brain lobe
at mid-third instar: the OPC (green), the inner proliferation center (IPC, yellow), and the medulla cortex (me).
Discs large (Dlg; gray) outlines all cell cortices in the larval brain and highlights the morphology of the two optic
lobe proliferation centers. (c) GAL4c855a begins to drive expression of UAS-pon-gfp (green; Dlg in red) at first
instar. At late first/early second instar (24 h ALH; after hatching), the OPC and the IPC can be distinguished as
two closely associated epithelia. The cells belonging to the proliferation centers (green) are clearly distinguish-
able by their columnar shape, in contrast to the round, isolated central brain cells. (d) At the end of second/
early third instar (48 h ALH), the epithelia of the OPC and IPC separate from each other and smaller progeny
cells are located between the two epithelia. (e) As development progresses during second to mid-third instar
(72 h ALH), the OPC cells at the medial edge of the epithelium lose their columnar shape (to the left of the
arrowheads). (f) At late third instar (96 h ALH), the OPC epithelium decreases in size while the number of round
neuroblast-like cells increases at the medial edges (to the left of the arrowheads). All images are single confo-
cal sections, with anterior on top and lateral to the right. Scale bar is 50 μm (a) and 20 μm (b–f) (Fig. 2 is
reproduced from ref. [2] originally published in Neural Development, BioMed Central)
78 Benjamin Perruchoud and Boris Egger

Acknowledgments

We thank Mike Bate for his advice on how to dissect brains of 1st
and 2nd instar larvae. B.P. and B.E. are funded by the Swiss
University Conference (SUK/CUS) Project P-01 Bio-BEFRI.

References
1. Hofbauer A, Campos-Ortega JA (1990) negatively regulated by JAK/STAT.
Proliferation pattern and early differentiation Development 135:1471–1480
of the optic lobes in Drosophila melanogaster. 7. Reddy BV, Rauskolb C, Irvine KD (2010)
Roux’s Arch Dev Biol 198:264–274 Influence of fat-hippo and notch signaling on
2. Egger B, Boone JQ, Stevens NR et al (2007) the proliferation and differentiation of
Regulation of spindle orientation and neural Drosophila optic neuroepithelia. Development
stem cell fate in the Drosophila optic lobe. 137:2397–2408
Neural Develop 2:1 8. Ngo KT, Wang J, Junker M et al (2010)
3. Egger B, Gold KS, Brand AH (2010) Notch Concomitant requirement for Notch and Jak/
regulates the switch from symmetric to asym- Stat signaling during neuro-epithelial differen-
metric neural stem cell division in the tiation in the Drosophila optic lobe. Dev Biol
Drosophila optic lobe. Development 137: 346:284–295
2981–2987 9. Orihara-Ono M, Toriya M, Nakao K et al
4. Egger B, Gold KS, Brand AH (2011) (2011) Downregulation of Notch mediates
Regulating the balance between symmetric the seamless transition of individual Drosophila
and asymmetric stem cell division in the devel- neuroepithelial progenitors into optic medul-
oping brain. Fly 5(3):237–241 lar neuroblasts during prolonged G1. Dev Biol
5. Yasugi T, Sugie A, Umetsu D et al (2010) 351:163–175
Coordinated sequential action of EGFR and 10. Wang W, Liu W, Wang Y et al (2010) Notch sig-
Notch signaling pathways regulates proneural naling regulates neuroepithelial stem cell mainte-
wave progression in the Drosophila optic lobe. nance and neuroblast formation in Drosophila
Development 137:3193–3203 optic lobe development. Dev Biol 350:414–428
6. Yasugi T, Umetsu D, Murakami S et al (2008) 11. Gotz M, Huttner WB (2005) The cell biology
Drosophila optic lobe neuroblasts triggered by of neurogenesis. Nat Rev Mol Cell Biol
a wave of proneural gene expression that is 6:777–788
 Chapter 6

Using MARCM to Study Drosophila Brain Development

Gudrun Viktorin

Abstract
Mosaic analysis with a repressible cell marker (MARCM) generates positively labeled, wild-type or mutant
mitotic clones by unequally distributing a repressor of a cell lineage marker, originally tubP-driven GAL80
repressing the GAL4/UAS system. Variations of the technique include labeling of both sister clones (twin
spot MARCM), the simultaneous use of two different drivers within the same clone (dual MARCM), as
well as the use of different repressible transcription systems (Q-MARCM). MARCM can be combined
with any UAS-based construct, such as localized GFP fusions to visualize subcellular compartments, genes
for rescue and ectopic expression, and modifiers of neural activity. A related technique, the twin spot gen-
erator, generates positively labeled clones without the use of a repressor, thus minimizing the lag time
between clone induction and appearance of label. The present protocol provides a detailed description of a
standard MARCM analysis of brain development that includes generation of MARCM stocks and crosses,
induction of clones, brain dissection at various stages of development, immunohistochemistry, and confo-
cal microscopy, and can be modified for similar experiments involving mitotic clones.

Key words Drosophila, MARCM, Twin spot generator, Flp-out, Clones, Somatic recombination,
Protocol, Brain development, Cell labeling, Neuroblast lineage

1 Introduction

The generation of clones, composed of cells that are genetically

different from the rest of the animal, is the genetic equivalent to
the single-cell labeling and transplantation techniques of classical
embryology that are possible to a limited extent in Drosophila.
The resulting mosaic animals are essential in studying later functions
of early lethal mutations, local tissue-specific effects of mutations,
and non-cell-autonomous effects (reviewed in [1]). In Drosophila,
mitotic clones are readily induced using the yeast Flp/FRT recom-
binase system [2, 3]. Expressing Flp recombinase from a heat-
shock-inducible promoter or a tissue-specific enhancer in mitotic
cells effects a crossing over at FRT sites that are located in the same
position on homologous chromosome arms. The following cell
division can segregate the recombined chromosomes into the two
daughter cells such that each daughter cell is homozygous for one

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_6, © Springer Science+Business Media, LLC 2014

79
80 Gudrun Viktorin

a b
FRT (m)
(m) (m) 80 80 (m) 80 (m) 80 80
(m)
FLP (m) 80 tubP
GAL80
FRT

tubP
GAL4

UAS
gfp

UAS
gene

hs
FLP

Fig. 1 (a) Schematic representation of unequal repressor (orange) distribution in MARCM by induced somatic
recombination between homologous FRT sites, leading to positively labeled clones (green) and unlabeled sister
clones. Only the recombining chromosome pair is shown, containing Flp recombinase target recognition (FRT)
sites (black triangles) and repressor transgene (orange). Inset is a maximum projection of a confocal stack of
a type II neuroblast clone that was induced around hatching, with wide secondary arborizations (Leica SP5,
63×/1.30 glycerol immersion objective). (b) Complete set of elements in cells of heterozygous offspring of a
MARCM cross prior to recombination, and its interactions. Flp/FRT system, black; GAL4/UAS system, white;
GAL80 repressor driven by ubiquitous promoter, orange; GFP marker, green. Some optional elements are
included in parentheses: a mutation on the GAL80-free FRT chromosome (m), and an additional UAS-driven,
GAL80-repressible transgene for rescue or ectopic expression of genes within the clone

of the recombined chromosome arms. Thus, if one chromosome

arm contains a mutation, the mutation will become homozygous
in one of the daughter cells and all its progeny, and the effects of
the mutation can be studied in a heterozygous, phenotypically
wild-type surrounding.
Positive labeling of clones using a genetically encoded marker
such as GFP is not straightforward because any marker will be
expressed throughout the heterozygous animal. Mosaic analysis
using a repressible cell marker (MARCM) [4, 5] overcomes this
limitation by asymmetrically distributing a repressor of marker
expression, rather than the marker itself. Thus, within an animal
that is heterozygous for the repressor transgene and therefore
unlabeled, only clones that lost the repressor will express the
marker (Fig. 1a). If a mutation is present on the other, homolo-
gous FRT-bearing chromosome, the labeled clone of cells will in
addition be homozygous for that mutation. Therefore, MARCM
generates positively labeled, wild-type or mutant clones of mitotic
cells that allow the investigation of wild-type and mutant develop-
ment with high temporal and spatial resolution.
 MARCM in Drosophila Brain Development 81

As inducible, stochastic method of labeling single cells and

their progeny, MARCM has enabled detailed morphological birth
date analyses of single neurons as well as tracing of neuroblast lin-
eages. For example, MARCM experiments identified neuronal
subtypes produced from mushroom body and antennal lobe projec-
tion neurons, and revealed that these are produced by the corre-
sponding neuroblasts in a strict temporal order [6, 7]. Comprehensive
analyses of postembryonic neuroblast lineages [8, 9] and glial cells
[10, 11] have also been performed. MARCM experiments identi-
fied a unique proliferation mode in type II lineages via neuroblast-
like transit amplifying cells [12–14], and identified them as source
of many larval brain glia [10, 15, 16]. MARCM experiments also
revealed the existence of hemilineages that are formed as the
two sibling neurons of ganglion mother cells adapt either of the
two different fates, a choice mediated by Notch signaling [17, 18].
To the clonal analysis of mutant phenotypes, MARCM added not
only a positive label but also the possibility for clone-specific rescue
and ectopic expression of genes, for example the switching of cell
fates by switching developmental control genes [19–21]. Using a
behavioral assay together with clonal stimulation of neuronal activity
via the thermosensitive TrpA1 channel, MARCM analysis has iden-
tified neuronal clusters that trigger male courtship behavior [22].
For clones that appear at higher frequency, or with directed expres-
sion of Flp [23], phenotypic screens for mutations on the FRT-
bearing chromosome arm are feasible, and have for example led to
the discovery of genes controlling the temporal identity of mush-
room body neurons, branching of their axons, and cell type-specific
axon pruning during metamorphosis [24–26].
The initially employed and most frequently used repressor
in MARCM is the yeast GAL80 protein, driven by a ubiquitous tubP
promoter, that can silence expression from a plethora of GAL4
driver and UAS responder stocks available for Drosophila, as well
as the GAL80-repressible LexA-GAD fusion protein in dual
MARCM [27]. Alternatively, other repressible transcription sys-
tems (Q MARCM, [28]) as well as UAS-marker-RNAi (twin spot
MARCM [27]) have been adapted as MARCM systems and used
for high-resolution lineage analysis [17, 29]. A limitation of
MARCM is the long decay time of the repressor, 24–48 h for
GAL80 [4]; this precludes its use for the embryo [30]. The repres-
sor-free twin spot generator is more immediately visible as it
recombines the 5′- and 3′- halves of two fluorescent marker genes,
separated by an FRT-containing intron, to yield two different fluo-
rescent proteins in sister clones. Thus, the time between clone
induction and appearance of label is only limited by expression and
fluorophore formation in the marker proteins [31]. Alternatively,
flp-out clones, albeit not limited to mitotic cells, can be generated
using UAS>CD2,y>mCD8GFP (G. Struhl in [32]; for neuroblast
clones in the embryo, see [30, 33]), Actin > CD2 > GAL4 [34–36],
or Actin > Draf + > nlacZ [37].
82 Gudrun Viktorin

This protocol provides a detailed procedure of a standard

MARCM experiment in brain development (see also [38–43]) that
covers the generation of MARCM stocks, crosses and egg collec-
tion, clone induction, dissection of brains, immunohistochemistry,
and confocal microscopy. The described tools and procedures can
be adapted for related clonal techniques and other tissues.

2 Materials

2.1 Fly Stocks A pair of MARCM stocks contains the following genetic elements
(Fig. 1b), available through Bloomington [44]:
1. FRT chromosome with distally located tubP-GAL80 trans-
gene (see Note 1).
2. Homologous FRT chromosome, with or without distally
located mutation.
3. GAL4 driver (see Note 2).
4. UAS marker (see Note 3).
5. hs-flp or promoter-driven flp transgene (see Note 4).
6. Balancer chromosomes with markers visible at the time of
dissection (see Note 5).

2.2 Preparing 1. Maintenance: Incubator at 25 °C (and 18 °C, optional).

Crosses, Egg 2. Standard fly food [45] with a drop of fresh yeast paste (com-
Collection, Heat-Shock mercial live baker’s yeast suspended in little sugar water), dried
Induction of Clones under a fan for a few minutes until the surface appears matte.
3. Egg collection: Grape juice plates [45] and matching contain-
ers covered with mesh, or food bottles, both with fresh yeast
paste (see Note 6).
4. Heat shock: Water bath at 37 °C.
5. Parafilm to seal plates.

2.3 Dissection 1. Handling animals at different stages: Forceps or a wet brush

for wandering larvae and pupae; 30 % glycerol and spoon for
floating and skimming off pre-wandering larvae; CO2 fly sta-
tion for anesthetizing adults.
2. Ice for keeping animals immobilized and for arresting devel-
opment at exact stages.
3. Stereomicroscope with black working surface and adjustable
side illumination, equipped with fluorescence for selecting
genotypes prior to dissection.
4. Transparent glass or acrylic dishes for dissection and immuno-
histochemistry (see Note 7).
5. Dumont forceps (tip diameters 0.05 × 0.005–0.02 mm, Fine
Science Tools).
 MARCM in Drosophila Brain Development 83

6. 20 μl micropipette, or wide bore glass Pasteur pipette with

10 ml standard pipette pump (Bel-Art) for transferring brains
(see Note 8).

2.4 Solutions for 1. PBS: Tablets (Sigma) or diluted from 10× PBS: 2.56 g/l
Dissection and NaH2PO4, 11.94 g/l Na2HPO4, 102.2 g/l NaCl, pH at
Immunohisto- 7.4 [46].
chemistry 2. PBT: 0.1–0.5 % Triton X-100 in PBS.
3. PBL: [47] (see Note 9) (0.1 M lysine, 0.05 M phosphate buffer,
pH 7.4—dissolve lysine in half the volume of water, adjust pH
to 7.4 with 0.1 M Na2HPO4, adjust volume to final concen-
trations with 0.1 M NaH2PO4/Na2HPO4 buffer pH 7.4).
4. 8 % paraformaldehyde stock solution: 0.8 g paraformaldehyde
dissolved in 10 ml water with 70 μl 1 M NaOH, incubate in
37 °C water bath with occasional shaking until just dissolved,
prepare fresh or freeze aliquots at −20 °C.
5. Methanol.
6. BT-NGS: PBT with 10 % normal goat serum that had been
heat inactivated for 30 min at 56 °C and frozen in aliquots.
7. Antibody stocks (see Note 9).

2.5 Microscopy and 1. Microscope slides.

Image Analysis 2. Coverslips that match the objective to be used.
3. Mounting medium, e.g., Vectashield H-1000 (Vector Labs,
Burlingame, CA, USA).
4. Spacers (see Note 10).
5. Nail polish to seal slides (see Note 10).
6. Confocal scanning microscope with suitable objectives
(see Note 11).
7. Fiji/ImageJ or other image processing software [48].

3 Methods

3.1 Generation 1. MARCM offspring for clone induction is produced from a

of MARCM Stocks final cross of two stocks to achieve heterozygosity of the two
and Crosses FRT chromosomes (e.g., FRT82B, tubP-Gal80 × FRT82B).
Distribute the remaining MARCM elements into the two
stocks as convenient (see Note 12), using standard genetic
techniques [49, 50].
2. Grow enough flies of the pair of stocks to be able to collect
a sufficient number of eggs within the desired time period
(see Note 13).
84 Gudrun Viktorin

3. Collect virgins and males, combine them in bottles or tubes

with freshly yeasted food, and keep them at 25 °C for 4–5 days
to mate (see Note 14).

3.2 Egg Collection, 1. Transfer MARCM crosses into egg collectors, cover with
Heat-Shock Induction yeasted grape juice plates or fly bottles, and collect at 25 °C in
of Clones, Brain the dark for the desired time period.
Dissection 2. Keep offspring at 25 °C until the desired stage for heat shock
(or dissection, if flp is not induced by heat shock).
3. Immerse sealed grape juice plates or fly bottles in a 37 °C water
bath for the desired duration (see Note 15).
4. Raise larvae at 25 °C on fly food until dissection (see Note 16).
Grape juice plates can be cut into pieces and transferred into fly
food bottles.
5. Harvest animals as larvae, pupae, or adults: For pre-wandering
larvae, fill bottle with 30 % glycerol, skim larvae off the surface
with spoon, wash with tap water, and immobilize on ice for
genotyping; take wandering third-instar larvae or pupae off the
bottle wall with forceps or a wet brush; Anesthetize adults with
CO2 and keep immobilized on ice.
6. Select offspring of the correct genotype, and discard any off-
spring with balancer chromosomes.
7. Dissect brains of larvae, pupae, or adults (see Note 17) in PBS.
For immunostaining, keep dissected tissue on ice and fix after
20–30 min of dissection. For subsequent live imaging, dissect
at room temperature in live imaging media [14, 51].

3.3 Fixation, 1. Fix freshly dissected tissue for 30–60 min in 2 or 4 % parafor-
Immunostaining, and maldehyde in PBL or PBS (see Note 18).
Microscopy 2. Wash with PBT four times within 45 min.
3. Preincubate in PBT–NGS solution for 15 min.
4. Incubate in primary antibody in PBT–NGS overnight at 4 °C.
5. Wash with PBT four times within 1 h.
6. Incubate in secondary antibody for 2–3 h at room temperature
or, preferably, overnight at 4 °C in the dark.
7. Wash with PBT ≥ 4 times within 1 h.
8. Take off liquid with a pipette, and place a drop of Vectashield
onto the brains. Without stirring incubate for several hours to
overnight, to slowly diffuse in glycerol without osmotically
shocking the tissue. Brains should be imaged soon but can be
kept in Vectashield at −20 °C for later processing.
9. Place brains onto the slide, remove Vectashield as much as
possible so that surface tension holds the brains in place,
and arrange them in rows and columns for microscopy, using
an eyelash tool or forceps. Larval brains that are still attached
 MARCM in Drosophila Brain Development 85

to discs and cuticle can be dissected on the slide, and clones

can be preselected under a stereomicroscope equipped with
GFP fluorescence (see Note 19).
10. Place spacers onto the slide or the coverslip (see Note 10).
11. Place a cleaned coverslip on top of the brains. Ideally, each brain
sits in a drop of Vectashield that just touches the coverslip.
12. Let Vectashield seep in from one corner, slowly to avoid bubbles,
until the entire space under the coverslip is just filled.
13. Fix the coverslip to the slide with nail polish (see Note 20).
14. Image as soon as possible for optimal results. Slides can be
stored flat at −20 °C for several months with some deteriora-
tion of signal.

4 Notes

1. FRT chromosomes with distally located tubP-GAL80 transgenes

are available for all major chromosome arms [41]. Viability of
modular MARCM driver stocks that contain all standard elements
in the FRT, tubP-GAL80 stock (i.e., hs-flp, enhancer-GAL4, and
UAS-marker) is usually excellent for virgin collection (Table 1,
and [41]). The males then only provide the wild-type or mutant
FRT chromosome, and optional UAS-gene constructs.

Table 1
Sample MARCM driver stocks that have been used routinely and remained stable over years
(constructed by B. Bello)

FRT chr. Chr1 Chr2 Chr3 Reference

1 FRT19A, tubP- tubP-GAL4, UAS- [19]
GAL80LL1, mCD8::GFPLL5/CyO,
1
hsFLP , w* ActGFPJMR1
2L y,w,hsFLP122 FRT40A, tubP-GAL80LL10/ tubP-GAL4LL7, UAS- [14]
CyO,ActGFPJMR1 mCD8::GFPLL6/TM6,Tb,Hu
2R y,w,hsFLP1 FRTG13, tubP-GAL80LL2/ tubP-GAL4LL7, UAS- B. Bello
(CyO,actin-GFPJMR1) mCD8::GFPLL6/TM6,Tb,Hu unpub
2R FRTG13, tub-Gal80LL2, hs-Flp38/ tubP-GAL4LL7, UAS- B. Bello
CyO, actin-gfpJMR1 mCD8::GFPLL6/TM6,Tb,Hu unpub
3L y,w,hsFLP122 tubP-GAL4, UAS-mCD8::GFPLL5/ FRT2A, tubP-GAL80LL3/ [82]
CyO,ActGFPJMR1 (TM6,Tb,Hu)
3R y,w,hsFLP122 tubP-GAL4, UAS-mCD8::GFPLL5/ FRT82B, tubP-GAL80LL3/ [83]
CyO,ActGFPJMR1 (TM6,Tb,Hu)
They can be crossed with matching, wild-type FRT chromosomes for labeling, or may include other components for
functional analysis. For simple labeling, the FRT82B and FRT40A tubulin drivers yield many eggs and controllable
clone frequencies. For mutant analysis, the FRT chromosome used will be dictated by the location of the mutation
86 Gudrun Viktorin

2. A ubiquitous driver such as tubP-GAL4 ([4] or J-P Vincent in

[19]) or Actin-GAL4 [9] will label the whole clone. elavC155-
GAL4 [52] can be useful when studying secondary neuronal
projections because it leaves the widely arborized primary
neurons [9] or neuroblast-derived glia (own observations)
largely unlabeled. Non-ubiquitous GAL4 drivers can be used
to label only a subset of neuroblast clones [6], or only certain
cell types [11], although most GAL4 drivers are not specific
[53]. However, lineage relationships can be either missed if not
all cells of a clone are labeled, or wrongly assumed if clones are
not as sparse as it seems. In addition, the presence of unde-
tected neighboring clones or clone parts can yield ambiguous
results if, for example, non-cell-autonomous effects of muta-
tions are involved. An alternative approach is dual MARCM
[27], where in addition to the specific GAL4 driver the
GAL80-repressible tubP-lexA-GAD simultaneously labels the
full clone with a different marker driven by the lexA operator.
If the GAL4 driver of choice is weak, it can be useful to recom-
bine the GAL4 driver and/or the UAS-marker transgene onto
the free FRT chromosome so that they become homozygous
within clones. Drivers homozygous throughout the animal
may fail to be repressed by the one copy of tubP-Gal80.
3. Any combination of UAS-based expression constructs can be
used, many available through Bloomington [44]. The
membrane-bound UAS-mCD8GFP reporter [4, 53] labels cell
bodies and axonal and dendritic processes, while UAS-
mCD4GFP labels finer cell processes but labels cell bodies in
the brain less strongly [54]. Pre- and postsynaptic markers can
highlight axonal and dendritic regions within clones [55–58].
Adding a nuclear marker such as UAS-nlacZ20b (Y. Hiromi and
S. West in [44]) is useful for coexpression studies of nuclear
factors [19], and to characterize cells with complicated mor-
phology. UAS-Brainbow [59] and UAS-Flybow [47] can label
different clones in different colors. Flybow can also further sub-
divide clones in later recombination events. In addition to a
marker, UAS-driven gene constructs for overexpression, rescue,
RNAi knockdown, and modifiers of neural activity can be used.
4. With some hs-flp insertions, clones are rare. The more efficient
and less leaky hs-flp122 [37, 39], or hs-flp38 [60], together with a
short heat shock of 6–7 min at 37 °C, yields more brains with
sparse clones and fewer empty brains than hs-flp1 [61] with a
longer heat shock of 30–60 min at 37 °C. Enhancer-driven flp
such as ey-flp and repo-flp produces mosaics as well [43, 62, 63].
5. Balancers bearing either Tb 1 (e.g., TM6B Tb 1 Antp Hu or
[ 64 , 65]) or GFP/RFP-expressing transgenes such as CyO,
actin-gfpJMR1 (Reichart, J.-M. in [44]) can be selected for by
eye, or under a stereomicroscope equipped with fluorescence
 MARCM in Drosophila Brain Development 87

[44, 50, 66]. Absence of the FRT, tubP-Gal80 chromosome

will cause expression of the GAL4/UAS transgenes used;
therefore this chromosome may be selectable under a stereo-
microscope without the use of a visible balancer.
6. For fast heat distribution during short heat shocks, grape
juice plates or fly bottles with an equally thin layer of fly food
work well.
7. Contrast for dissection of brains is best in a transparent dish
placed on a black surface and illuminated from the side or the
back [66, 67]. Dissection can be performed on ice to arrest
development at specific stages, to preserve antigens, and to
immobilize and anesthetize animals. A metal computer-cooling
element, cut to the size of a flat Styrofoam box and eloxated in
black, can be placed into the ice-filled box to provide cold,
black surface. The cooling element rests on something other
than ice to avoid constant change in focus as the ice melts.
A square, flat lid of a glass staining jar or a glass petri plate can
hold many larvae, while debris can be pushed away to one side
during dissection. The bottom of the dissection dish can option-
ally be coated with Sylgard 184 silicone polymer (Dow Corning)
to aid in delicate dissections. A custom-made hand rest, such as
a piece of firm foam that is fit around the ice box, can help to
keep hands relaxed and still, and provides a soft surface to place
dissection tools. Fixation and staining can be done in clear
Eppendorf or PCR tubes; we prefer custom-made multi-well
dishes cut from 10 mm thick acrylic glass, covered with a micro-
scope slide. Wells of 15 mm diameter and 5 mm depth allow
antibody staining in as little as 70 μl, and visual control of brains
under the stereomicroscope when changing solutions.
8. Additional dissection tools: Spring scissors (tip diameter
0.05 mm, Fine Science Tools) for quick dissections of pupae
and larvae (see Note 17); 1 ml syringe with a short, 27 G needle
for pupal dissection (see Note 17); O-rings (Fine Science
Tools) or pieces of silicone tubing to adjust the maximal open-
ing angle of forceps and use less force to grab tissue; Eyelash
tool (Ted Pella, or custom made; lighter colors are visible
under fluorescence illumination) to dissect and orient stained
brains on the microscope slide (see Note 19), and to dislodge
brains that adhere to the wall of a pipette; sharpening stone
made of True Hard Arkansas rock (Dan’s Whetstone Co. or
Fine Science Tools) to sharpen damaged tips of forceps, and to
some degree scissors, under the stereomicroscope. Tools are
most frequently damaged by inadverted contact with parts of the
stereomicroscope; thus watch from outside while positioning
tools into the field of view, and then locate them through
oculars. Minimal force is needed to sharpen or re-bend the
tips of tools.
88 Gudrun Viktorin

9. Some labs routinely fix in PBL [47], not older than 3 months,
and others in PBS or other buffers [46]. PBL is derived from a
periodate–lysine solution used to fix glycoproteins [68]; lysine
alone may help to preserve certain antigens. Commonly used
antibodies for postembryonic brains label either synaptic neu-
ropil or axon tracts, described in [69]: BP106/mouse anti-
neurotactin (1:20, DSHB [70]) labels larval and pupal
secondary lineages and axon tracts. BP106 requires a 5–10-
min methanol incubation after fixation (Kathy Ngo and Volker
Hartenstein, pers. comm.) that also makes larval and pupal
brains less fragile and does not interfere with many other anti-
body stainings; BP104/mouse anti-neuroglian (1:20, DSHB;
[71]) labels secondary axon tracts in the adult brain; nc82/
mouse anti-bruchpilot (1:10; DSHB; [72]) is used for the
Drosophila standard brain [73] and labels synaptic neuropil. nc82
staining in adult brains greatly improves when incubated over two
nights and in 0.5 % Triton X-100 [40, 74]. Chicken anti-GFP
(1:1,000, Abcam, Cambridge, UK), rabbit anti-RFP (1:200,
Abcam), and rabbit anti-beta-Galactosidase (1:1,000, Cappel/
MP Biomedicals) can be used for transgenic markers, and Alexa
488/568/647 (1:300, Molecular Probes/Life Technologies)
secondary antibodies work well for triple labeling.
10. For third-instar larvae, pupae, and adult brains, a spacer slightly
thinner than a coverslip will hold brains in place without
squashing them, e.g., two layers of scotch tape. For smaller
brains, use one layer of tape, or apply a flexible spacer such as
Vaseline to the edges of the coverslip, and gently push down.
Minor-quality nail polish will not seal well and may interfere with
fluorescence. Hardening mounting media are also in use [9].
11. At 20× magnification, the whole brain can be fully scanned; for
single-cell resolution (a neuronal cell body from a postembry-
onic lineage has a diameter around 4–5 μm), 40× or higher
magnification is necessary. At 63× magnification, a fixed brain
hemisphere from a wandering third-instar larva just fits into the
field of view. For whole-hemisphere scans of brains mounted in
glycerol-based Vectashield, a glycerol immersion objective min-
imizes spherical aberrations [75]. In addition, a long working
distance is useful, since working distance of higher magnifying
oil objectives can be shorter than the thickness of the brain plus
coverslip. Valuable aspects of confocal microscopy and image
processing artifacts are discussed in [53].
12. If a transgenic chromosome for a planned stock is difficult to bal-
ance for introducing the other elements, consider redistributing
the MARCM elements in case the desired stock is not viable. It is
not infrequent to obtain around 15 flies or less of the correct
genotype from a bottle during the final crossing generations.
13. A bottle with 100–120 virgins and 10–30 males, left to lay eggs
within 1–2 h at 25 °C, can give rise to 100–200 live larvae depend-
 MARCM in Drosophila Brain Development 89

ing on the fecundity of the stock used, and the time of egg collec-
tion, highest during the flies’ afternoon/evening [50]. Flies that
had been raised under crowded conditions are small and lay few
eggs; split up bottles with too many larvae into 2–3 bottles prior
to virgin collection. If very large amounts of virgins need to be
collected, it can save time to cross a virginator such as hs-hid (Y)
(M. Van Doren in [76]) into the MARCM driver that can kill off
most males in a 1-h-long heat shock at the third instar prior to
virgin collection (see descriptions of Bloomington stocks 8846,
24638). Each element that is homozygous in the parent stock
doubles the amount of larvae of the correct genotype in the off-
spring; thus, if homozygous viable and fertile, avoid males con-
taining a balancer, and allow a day longer for mating to occur with
fewer males.
14. Virgin collections and crosses can be lost instantly when shaken
into a bottle with sticky food, and females need to remain well
fed with yeast paste. Between egg collections, shake MARCM
crosses into fresh bottles with dry surface as soon as the drop
of yeast is eaten. If bottles tend to be humid, keep flies less
densely, and place a rolled-up paper tissue in the food to absorb
moisture when not collecting eggs. Crosses can be kept at
18 °C when not in use to prolong the fertile period. Females
lay best during the week following mating, but can be used for
3–4 weeks until the amount of eggs declines.
15. Larvae will crawl rapidly to escape the heat; submerge sealed
plates completely, or push down the stopper of the fly food
bottle below the surface of the water bath. Larvae crawl onto
mite-proof stoppers but not cheap foam ones. The optimal
time and duration of heat shock need to be determined experi-
mentally to reach the desired clone density. As an estimate,
using hs-flp122, a heat shock of 6 min at 37 °C around hatching
most frequently induces 0–2 neuroblast clones per brain, while
a heat shock of 8 min most frequently induces 1–5 neuroblast
clones per brain (own observations). Background clone induc-
tion is usually negligible, but single experiments can show con-
siderable clone frequencies without heat shock (own
observations). Certain neuroblast clones are only recovered
upon heat shock during certain instars, and in smaller develop-
mental time windows than may be expected from other experi-
ments (B. Bello, personal communication).
16. The density at which larvae are reared greatly affects their
developmental rate [38, 50]. In my hands, a maximum of 120
larvae in a bottle of 6 cm diameter will develop rather synchro-
nously, but tends to dry out; add water as needed. If larvae
are too dense, split them up into several bottles. Larval and
pupal development at 18 °C takes quite exactly half as long as
development at 25 °C [50], which can be used to avoid nighttime
dissections, unless staging needs to be exact.
90 Gudrun Viktorin

Fig. 2 Dissection of larvae (a, b), prepupae (c, d), and pupae (e, f) using scissors and eyelash tool. Scissors in
(a) are not positioned at the proposed cut site

17. Dissection is the most time-consuming step in a MARCM

experiment; techniques are best collected from several sources
and optimized individually [40, 66, 67, 77–79]. If a collection
of animals needs to be dissected at a defined developmental
stage, some of the intact animals may remain on ice for 2–3 h
without greatly affecting the quality of antibody labeling; how-
ever, attempt to keep this time as short as possible, for example by
preparing a second collection that is 2 h younger than the first.
Larvae: One quick larval brain dissection technique is to
gently tear out the mouth hooks with discs and brain attached
(e.g., [78]); the cuticle needs to tear caudal to the anterior
spiracles. However, the ventral nerve cord will often break off
during the procedure, making later sample orientation more
difficult, and the proventriculus can destroy the commissure.
Another quick and distortion-free dissection technique is to
hold the larva ventral side up at the mouth hooks, locate the
ventral nerve cord and neighboring salivary glands through
the cuticle, and make an oblique cut with spring scissors at the
end of the ventral nerve cord to expose the brain (Fig. 2a). If
the cut is placed well, only small amounts of fat body and gut
will still be attached. Use the open scissors to invert the ante-
rior end onto the closed forceps, strip the remaining fat body
tissue away posteriorly, and remove with one or two cuts. Take
care not to cut into the forceps. If the antibody to be used is
not sensitive to the presence of fat body, it does not need to be
 MARCM in Drosophila Brain Development 91

removed completely, which saves time. Young larvae that are

too small to be inverted can be opened further with another
longitudinal cut, if the brain is not exposed immediately. Still
holding the mouth hooks, lift the tissue out of the dissection
dish and release into a well filled with PBS and placed on ice.
Changing solutions is easier with larger pieces of tissue attached
to the brain. Larval brains can be dissected free during mount-
ing in Vectashield, when the tissue is less fragile (Fig. 2b).
Pupae (see also [79]): One possibility to dissect pupal brains
is to hold the pupa submerged at the anterior spiracles, stab in
the middle with one blade of spring scissors to relieve pressure,
make one large cut, and then cut into half without squeezing.
In prepupae, before head eversion [50, 80], make two long cuts
along the sides to the anterior end, lift off the dorsal part like a
lid, and remove the dorsal half with an anterior transverse cut
between the dorsal and ventral halves (Fig. 2c). An additional
cut may be needed to detach the gut. The brain will sit in the
ventral half of the puparium case and can be fixed like this
(Fig. 2d), since it is fragile especially between 6 and 12 h after
puparium formation (APF), and dissected out after fixation and
before antibody staining. In older pupae, the anterior half of
the pupa can be pulled out from the puparium case with forceps
right after cutting the pupa into half. Use forceps to slide a scis-
sor blade underneath the dorsal pupal cuticle all the way to the
level of the antennae (Fig. 2e), and make a cut. Open the ante-
rior cuticle further with another mediolateral cut, and free the
brain from fat body (Fig. 2f) with a gentle jet of PBS from a
1 ml syringe with short 27 G needle. The brain will adhere to
the cuticle only at the retinae. Remove tracheae, and pull the
cuticle away from the brain, or if adhesion to the retina is
strong, gently tear the cuticle overlying the retina. The water jet
can help in dissecting, but if too strong can tear off the optic
lobes before the retina is released. Remove the ventral nerve
cord if not needed. Transfer pupal brains in a pipette, or in a
layer of PBS between almost closed forceps.
Brains from adults, and from older pupae in which the
adult cuticle has already formed, can be dissected either from
ventral, holding the mouthparts and removing the eye cups
one by one [77], or by breaking the cuticle open above the
antennae and gently pulling the head capsule apart [67]. If the
eyecup is taken off in one piece, the lamina will be removed as
well, which greatly improves antibody penetration into the
medulla and lobula [78]. If the lamina is needed, the eyecup
can be removed in pieces, while removing the red fluorescent
pigment as much as possible [81]. The adult brain is covered
by large tracheal sacs that need to be removed because they
will fill with air during staining and washing. It can be useful to
reserve fine forceps for adult brain dissection.
92 Gudrun Viktorin

Brightly GFP-labeled clones in freshly dissected or stained

tissue are visible under a stereomicroscope equipped with fluo-
rescence; however intensity increases greatly with anti-GFP
antibody staining.
18. Avoid removing all liquid when changing solutions to not
expose the soft tissue to surface tension, and preserve
morphology.
19. For final dissection of larval tissue in Vectashield, line up larval
tissue into the tip of a wide-bore Pasteur pipette below the
tapered part, place all larvae directly onto the slide, and remove
as much Vectashield as possible. Orient a larva so that the brain
is visible in side view, place the thin end of an eyelash tool
between discs and brain, hold the disc–brain connections against
the slide with the eyelash, and pull away the discs with forceps
against the elastic resistance of the eyelash (Fig. 2b). Grab the
remaining discs with forceps and remove the same way. If the
brain is bent over the ventral nerve cord, orient the brain ventral
side up, crush the ventral nerve cord against the slide at the site
of the bend, and then orient as needed. For best imaging quality,
place the most important parts closest to the coverslip.
20. Nail polish will not stick to glass with traces of glycerol; there-
fore cover all the areas that had Vectashield on them with nail
polish. Store slides flat at −20 °C to minimize leaking in case
the seal is not perfect.

Acknowledgments

I thank Bruno Bello for constructing and optimizing MARCM

stocks, and Bruno Bello, Beate Hartmann, Nadia Riebli, Philipp
Kuert, Yanrui Jiang, Susanne Flister, Angela Giangrande, and
Heinrich Reichert for sharing material and techniques, and for
critical reading of the manuscript. I also thank the imaging facility
(Markus Dürrenberger, Oliver Biehlmaier, Alexia Ferrand) and
the workshop of the Biozentrum for technical solutions and advice.
I acknowledge support via Swiss NSF grant 31003A 140607 to
Heinrich Reichert.

References

1. Blair SS (2003) Genetic mosaic techniques for 3. Xu T, Rubin GM (1993) Analysis of genetic
studying Drosophila development. Development mosaics in developing and adult Drosophila tis-
130(21):5065–5072. doi:10.1242/dev.00774 sues. Development 117(4):1223–1237
130/21/5065 [pii] 4. Lee T, Luo L (1999) Mosaic analysis with a
2. Golic KG (1991) Site-specific recombination repressible cell marker for studies of gene function
between homologous chromosomes in in neuronal morphogenesis. Neuron 22(3):451–
Drosophila. Science 252(5008):958–961 461. doi:doi:S0896-6273(00)80701-1 [pii]
 MARCM in Drosophila Brain Development 93

5. Lee T, Luo L (2001) Mosaic analysis with a 16. Viktorin G, Riebli N, Popkova A, Giangrande
repressible cell marker (MARCM) for A, Reichert H (2011) Multipotent neural
Drosophila neural development. Trends stem cells generate glial cells of the central
Neurosci 24(5):251–254. doi:S0166- complex through transit amplifying interme-
2236(00)01791-4 [pii] diate progenitors in Drosophila brain devel-
6. Jefferis GS, Marin EC, Stocker RF, Luo L opment. Dev Biol 356(2):553–565.
(2001) Target neuron prespecification in the doi:S0012-1606(11)01035-9 [pii] 10.1016/
olfactory map of Drosophila. Nature j.ydbio.2011.06.013
414(6860):204–208. doi:10.1038/35102574 17. Lin S, Lai SL, Yu HH, Chihara T, Luo L, Lee
7. Lee T, Lee A, Luo L (1999) Development of T (2010) Lineage-specific effects of Notch/
the Drosophila mushroom bodies: sequential Numb signaling in post-embryonic develop-
generation of three distinct types of neurons ment of the Drosophila brain. Development
from a neuroblast. Development 126(18): 137(1):43–51. doi:137/1/43 [pii] 10.1242/
4065–4076 dev.041699
8. Pereanu W, Hartenstein V (2006) Neural lin- 18. Truman JW, Moats W, Altman J, Marin EC,
eages of the Drosophila brain: a three- Williams DW (2010) Role of Notch signaling
dimensional digital atlas of the pattern of lineage in establishing the hemilineages of secondary
location and projection at the late larval stage. neurons in Drosophila melanogaster.
J Neurosci 26(20):5534–5553. doi:26/20/5534 Development 137(1):53–61. doi:137/1/53
[pii] 10.1523/JNEUROSCI.4708-05.2006 [pii] 10.1242/dev.041749
9. Truman JW, Schuppe H, Shepherd D, Williams 19. Bello BC, Hirth F, Gould AP (2003) A pulse
DW (2004) Developmental architecture of of the Drosophila Hox protein Abdominal-A
adult-specific lineages in the ventral CNS of schedules the end of neural proliferation via
Drosophila. Development 131(20):5167–5184 neuroblast apoptosis. Neuron 37(2):209–219.
10. Pereanu W, Shy D, Hartenstein V (2005) doi:S0896627302011819 [pii]
Morphogenesis and proliferation of the larval 20. Komiyama T, Johnson WA, Luo L, Jefferis
brain glia in Drosophila. Dev Biol 283(1):191– GS (2003) From lineage to wiring specificity
203. doi:S0012-1606(05)00228-9 [pii] POU domain transcription factors control
10.1016/j.ydbio.2005.04.024 precise connections of Drosophila olfactory
11. Awasaki T, Lai SL, Ito K, Lee T (2008) projection neurons. Cell 112(2):157–167.
Organization and postembryonic development doi:S0092867403000308 [pii]
of glial cells in the adult central brain of 21. Liu Z, Steward R, Luo L (2000) Drosophila Lis1
Drosophila. J Neurosci 28(51):13742–13753. is required for neuroblast proliferation, dendritic
doi:28/51/13742 [pii] 10.1523/ elaboration and axonal transport. Nat Cell Biol
JNEUROSCI.4844-08.2008 2(11):776–783. doi:10.1038/35041011
12. Boone JQ, Doe CQ (2008) Identification of 22. Kohatsu S, Koganezawa M, Yamamoto D
Drosophila type II neuroblast lineages contain- (2011) Female contact activates male-specific
ing transit amplifying ganglion mother cells. interneurons that trigger stereotypic courtship
Dev Neurobiol 68(9):1185–1195. behavior in Drosophila. Neuron 69(3):498–
doi:10.1002/dneu.20648 508. doi:S0896-6273(10)01040-8 [pii]
13. Bowman SK, Rolland V, Betschinger J, Kinsey 10.1016/j.neuron.2010.12.017
KA, Emery G, Knoblich JA (2008) The tumor 23. Amanai K, Jiang J (2001) Distinct roles of cen-
suppressors Brat and Numb regulate transit- tral missing and dispatched in sending the
amplifying neuroblast lineages in Drosophila. Dev Hedgehog signal. Development 128(24):
Cell 14(4):535–546. doi:10.1016/j.dev- 5119–5127
cel.2008.03.004; S1534-5807(08)00112-3 [pii] 24. Wang J, Zugates CT, Liang IH, Lee CH, Lee T
14. Bello BC, Izergina N, Caussinus E, Reichert (2002) Drosophila Dscam is required for diver-
H (2008) Amplification of neural stem cell gent segregation of sister branches and sup-
proliferation by intermediate progenitor cells presses ectopic bifurcation of axons. Neuron
in Drosophila brain development. Neural Dev 33(4):559–571. doi:S0896627302005706
3:5. doi:1749-8104-3-5 [pii] 10.1186/1749 [pii]
-8104-3-5 25. Zheng X, Wang J, Haerry TE, Wu AY, Martin
15. Izergina N, Balmer J, Bello B, Reichert H J, O’Connor MB, Lee CH, Lee T (2003)
(2009) Postembryonic development of transit TGF-beta signaling activates steroid hormone
amplifying neuroblast lineages in the receptor expression during neuronal remodeling
Drosophila brain. Neural Dev 4:44. doi:1749- in the Drosophila brain. Cell 112(3):303–315.
8104-4-44 [pii] 10.1186/1749-8104-4-44 doi:S0092867403000722 [pii]
94 Gudrun Viktorin

26. Zhu S, Lin S, Kao CF, Awasaki T, Chiang AS, 72(4):527–540. doi:0092-8674(93)90072-X
Lee T (2006) Gradients of the Drosophila [pii]
Chinmo BTB-zinc finger protein govern neu- 38. Rooke JE, Theodosiou NA, Xu T (2000)
ronal temporal identity. Cell 127(2):409–422. Clonal analysis in the examination of gene
doi:S0092-8674(06)01230-X [pii] 10.1016/j. function in Drosophila. Methods Mol Biol
cell.2006.08.045 137:15–22. doi:1-59259-066-7-15 [pii]
27. Lai SL, Lee T (2006) Genetic mosaic with dual 10.1385/1-59259-066-7:15
binary transcriptional systems in Drosophila. 39. Theodosiou NA, Xu T (1998) Use of FLP/
Nat Neurosci 9(5):703–709. doi:nn1681 [pii] FRT system to study Drosophila development.
10.1038/nn1681 Methods 14(4):355–365. doi:S1046-2023
28. Potter CJ, Luo L (2011) Using the Q system (98)90591-6 [pii] 10.1006/meth.1998.0591
in Drosophila melanogaster. Nat Protoc 40. Wu JS, Luo L (2006) A protocol for dissecting
6(8):1105–1120. doi:nprot.2011.347 [pii] Drosophila melanogaster brains for live imag-
10.1038/nprot.2011.347 ing or immunostaining. Nat Protoc 1(4):2110–
29. Yu HH, Kao CF, He Y, Ding P, Kao JC, Lee T 2115. doi:nprot.2006.336 [pii] 10.1038/
(2010) A complete developmental sequence of nprot.2006.336
a Drosophila neuronal lineage as revealed by 41. Wu JS, Luo L (2006) A protocol for mosaic
twin-spot MARCM. PLoS Biol 8 (8) analysis with a repressible cell marker
doi:10.1371/journal.pbio.1000461 (MARCM) in Drosophila. Nat Protoc
30. Kunz T, Kraft KF, Technau, GM, Urbach R 1(6):2583–2589. doi:nprot.2006.320 [pii]
(2012) Origin of Drosophila mushroom body 10.1038/nprot.2006.320
neuroblasts and generation of divergent embry- 42. Kao CF, Lee T (2010) Genetic mosaic analysis
onic lineages. Development 139:2510–2522. of Drosophila brain by MARCM. In: Zhang B,
doi:10.1242/dev.077883 Freeman MR, Waddell S (eds) Drosphila neu-
31. Griffin R, Sustar A, Bonvin M, Binari R, del robiology. Cold Spring Harbor Laboratory
Valle RA, Hohl AM, Bateman JR, Villalta C, Press, Cold Spring Harbor, NY, pp 125–140
Heffern E, Grunwald D, Bakal C, Desplan C, 43. Stork T, Bernardos R, Freeman MR (2012)
Schubiger G, Wu CT, Perrimon N (2009) The Analysis of glial cell development and function
twin spot generator for differential Drosophila in Drosophila. Cold Spring Harb Protoc
lineage analysis. Nat Methods 6(8):600–602. 2012(1):1–17. doi:2012/1/pdb.top067587
doi:nmeth.1349 [pii] 10.1038/nmeth.1349
[pii] 10.1101/pdb.top067587
32. Wong AM, Wang JW, Axel R (2002) Spatial
44. BDSC http://flystocks.bio.indiana.edu/
representation of the glomerular map in the
Drosophila protocerebrum. Cell 109(2):229– 45. Ashburner M, Roote J (2000) Laboratory cul-
241. doi:S0092867402007079 [pii] ture of Drosophila. In: Sullivan W, Ashburner
33. Larsen C, Shy D, Spindler SR, Fung S, Pereanu M, Hawley S (eds) Drosophila protocols. Cold
W, Younossi-Hartenstein A, Hartenstein V Spring Harbor Laboratory Press, Cold Spring
(2009) Patterns of growth, axonal extension and Harbor, pp 585–599
axonal arborization of neuronal lineages in the 46. Patel NH (2006) http://www.patellab.org/ >
developing Drosophila brain. Dev Biol Protocols
335(2):289–304. doi:S0012-1606(09)00926-9 47. Hadjieconomou D, Rotkopf S, Alexandre C,
[pii] 10.1016/j.ydbio.2009.06.015 Bell DM, Dickson BJ, Salecker I (2011)
34. Pignoni F, Zipursky SL (1997) Induction of Flybow: genetic multicolor cell labeling for
Drosophila eye development by decapentaple- neural circuit analysis in Drosophila melano-
gic. Development 124(2):271–278 gaster. Nat Methods 8(3):260–266.
35. Ito K, Awano W, Suzuki K, Hiromi Y, doi:nmeth.1567 [pii] 10.1038/nmeth.1567
Yamamoto D (1997) The Drosophila mush- 48. Eliceiri KW, Berthold MR, Goldberg IG,
room body is a quadruple structure of clonal Ibanez L, Manjunath BS, Martone ME,
units each of which contains a virtually identi- Murphy RF, Peng H, Plant AL, Roysam B,
cal set of neurones and glial cells. Development Stuurmann N, Swedlow JR, Tomancak P,
124(4):761–771 Carpenter AE (2012) Biological imaging soft-
36. de Celis JF, Bray S (1997) Feed-back mecha- ware tools. Nat Methods 9(7):697–710.
nisms affecting Notch activation at the dorso- doi:nmeth.2084 [pii] 10.1038/nmeth.2084
ventral boundary in the Drosophila wing. 49. Greenspan RJ (1997) Fly pushing: the theory
Development 124(17):3241–3251 and practice of drosophila genetics. Cold
37. Struhl G, Basler K (1993) Organizing activ- Spring Harbor Laboratory Press, Cold Spring
ity of wingless protein in Drosophila. Cell Harbor, NY
 MARCM in Drosophila Brain Development 95

50. Ashburner M, Golic KG, Hawley RS (eds) 60. Chou TB, Perrimon N (1992) Use of a yeast
(2005) Drosophila: a laboratory manual, 2nd site-specific recombinase to produce female
edn. Cold Spring Harbor Laboratory Press, germline chimeras in Drosophila. Genetics
Cold Spring Harbor, NY 131(3):643–653
51. Cabernard C, Doe CQ (2011) Live imaging of 61. Golic KG, Lindquist S (1989) The FLP recom-
neuroblast lineages within intact larval brains binase of yeast catalyzes site-specific recombi-
in Drosophila. In: Sharpe J, Wong R, Yuste R nation in the Drosophila genome. Cell
(eds) Imaging in developmental biology: a 59(3):499–509. doi:0092-8674(89)90033-0
laboratory manual. Cold Spring Harbor [pii]
Laboratory Press, Cold Sprint Harbor, NY, pp 62. Newsome TP, Asling B, Dickson BJ (2000)
217–227 Analysis of Drosophila photoreceptor axon
52. Lin DM, Fetter RD, Kopczynski C, guidance in eye-specific mosaics. Development
Grenningloh G, Goodman CS (1994) 127(4):851–860
Genetic analysis of Fasciclin II in Drosophila: 63. Silies M, Yuva Y, Engelen D, Aho A, Stork T,
defasciculation, refasciculation, and altered Klambt C (2007) Glial cell migration in the eye
fasciculation. Neuron 13(5):1055–1069. disc. J Neurosci 27(48):13130–13139.
doi:doi:0896-6273(94)90045-0 [pii] doi:27/48/13130 [pii] 10.1523/
53. Ito K, Okada R, Tanaka NK, Awasaki T (2003) JNEUROSCI.3583-07.2007
Cautionary observations on preparing and 64. Lattao R, Bonaccorsi S, Guan X, Wasserman
interpreting brain images using molecular biol- SS, Gatti M (2011) Tubby-tagged balancers
ogy-based staining techniques. Microsc Res for the Drosophila X and second chromo-
Tech 62(2):170–186. doi:10.1002/ somes. Fly (Austin) 5(4). doi:17283 [pii]
jemt.10369 65. Pina C, Pignoni F (2012) Tubby-RFP balanc-
54. Han C, Jan LY, Jan YN (2011) Enhancer- ers for developmental analysis: FM7c 2xTb-
driven membrane markers for analysis of non- RFP, CyO 2xTb-RFP, and TM3 2xTb-RFP.
autonomous mechanisms reveal neuron-glia Genesis 50(2):119–123. doi:10.1002/
interactions in Drosophila. Proc Natl Acad Sci dvg.20801
U S A 108(23):9673–9678. doi:1106386108 66. Blair SS (2000) Imaginal discs. In: Sullivan W,
[pii] 10.1073/pnas.1106386108 Ashburner M, Hawley S (eds) Drosophila pro-
55. Zhang YQ, Rodesch CK, Broadie K (2002) tocols. Cold Spring Harbor Laboratory Press,
Living synaptic vesicle marker: synaptotagmin- Cold Spring Harbor, NY, pp 159–173
GFP. Genesis 34(1–2):142–145. doi:10.1002/ 67. Ito K (1999) How to dissect Drosophila brains.
gene.10144 http://jfly.iam.u-tokyo.ac.jp/html/movie/
56. Nicolaï LJ, Ramaekers A, Raemaekers T, index.html
Drozdzecki A, Mauss AS, Yan J, Landgraf 68. Van Vactor DL Jr, Cagan RL, Kramer H,
M, Annaert W, Hassan BA (2011) Zipursky SL (1991) Induction in the develop-
Genetically encoded dendritic marker sheds ing compound eye of Drosophila: multiple
light on neuronal connectivity in Drosophila. mechanisms restrict R7 induction to a single
Proc Natl Acad Sci U S A 107(47):20553– retinal precursor cell. Cell 67(6):1145–1155.
20558. doi: 1010198107 [pii] 10.1073/ doi:0092-8674(91)90291-6 [pii]
pnas.1010198107
69. Pereanu W, Kumar A, Jennett A, Reichert H,
57. Jefferis GS, Potter CJ, Chan AM, Marin EC, Hartenstein V (2010) Development-based
Rohlfing T, Maurer CR Jr, Luo L (2007) compartmentalization of the Drosophila cen-
Comprehensive maps of Drosophila higher tral brain. J Comp Neurol 518(15):2996–
olfactory centers: spatially segregated fruit and 3023. doi:10.1002/cne.22376
pheromone representation. Cell 128(6):1187–
1203. doi:S0092-8674(07)00204-8 [pii] 70. Hortsch M, Patel NH, Bieber AJ, Traquina
10.1016/j.cell.2007.01.040 ZR, Goodman CS (1990) Drosophila neuro-
tactin, a surface glycoprotein with homology
58. Simpson JH (2009) Mapping and manipulat- to serine esterases, is dynamically expressed
ing neural circuits in the fly brain. Adv Genet during embryogenesis. Development 110(4):
65:79–143. doi:S0065-2660(09)65003-3 1327–1340
[pii] 10.1016/S0065-2660(09)65003-3
71. Bieber AJ, Snow PM, Hortsch M, Patel NH,
59. Hampel S, Chung P, McKellar CE, Hall D, Jacobs JR, Traquina ZR, Schilling J, Goodman
Looger LL, Simpson JH (2011) Drosophila CS (1989) Drosophila neuroglian: a member
brainbow: a recombinase-based fluorescence of the immunoglobulin superfamily with
labeling technique to subdivide neural expres- extensive homology to the vertebrate neural
sion patterns. Nat Methods 8(3):253–259. adhesion molecule L1. Cell 59(3):447–460.
doi:nmeth.1566 [pii] 10.1038/nmeth.1566 doi:0092-8674(89)90029-9 [pii]
96 Gudrun Viktorin

72. Wagh DA, Rasse TM, Asan E, Hofbauer A, easy. http://wwwstanfordedu/group/luolab/

Schwenkert I, Durrbeck H, Buchner S, Linksshtml
Dabauvalle MC, Schmidt M, Qin G, Wichmann 78. Morante J, Desplan C (2010) Dissecting and
C, Kittel R, Sigrist SJ, Buchner E (2006) staining Drosophila optic lobes. In: Zhang B,
Bruchpilot, a protein with homology to ELKS/ Freeman MR, Waddell S (eds) Drosphila neu-
CAST, is required for structural integrity and robiology. Cold Spring Harbor Laboratory
function of synaptic active zones in Drosophila. Press, Cold Spring Harbor, NY, pp 157–166
Neuron 49(6):833–844. doi:S0896- 79. Wolff T (2000) Histological techniques for the
6273(06)00124-3 [pii] 10.1016/j. Drosophila eye part I: larva and pupa. In: Sullivan
neuron.2006.02.008 W, Ashburner M, Hawley S (eds) Drosophila
73. Rein K, Zockler M, Mader MT, Grubel C, protocols. Cold Spring Harbor Laboratory
Heisenberg M (2002) The Drosophila stan- Press, Cold Spring Harbor, NY, pp 201–227
dard brain. Curr Biol 12(3):227–231. 80. Bainbridge SP, Bownes M (1981) Staging
doi:S0960982202006565 [pii] the metamorphosis of Drosophila melanogaster.
74. Jiang Y, Reichert H (2011) Programmed cell J Embryol Exp Morphol 66:57–80
death in type II neuroblast lineages is required 81. Williamson WR, Hiesinger PR (2010)
for central complex development in the Preparation of developing and adult Drosophila
Drosophila brain. Neural Dev 7:3. doi:1749- brains and retinae for live imaging. J Vis Exp
8104-7-3 [pii] 10.1186/1749-8104-7-3 37:e1936, doi: 1910.3791/1936
75. North AJ (2006) Seeing is believing? A begin- 82. Kumar A, Bello B, Reichert H (2009) Lineage-
ners’ guide to practical pitfalls in image acqui- specific cell death in postembryonic brain
sition. J Cell Biol 172(1):9–18. development of Drosophila. Development
doi:jcb.200507103 [pii] 10.1083/ 136(20):3433–3442. doi:dev.037226 [pii]
jcb.200507103 10.1242/dev.037226
76. Starz-Gaiano M, Cho NK, Forbes A, Lehmann 83. Lichtneckert R, Bello B, Reichert H (2007)
R (2001) Spatially restricted activity of a Cell lineage-specific expression and func-
Drosophila lipid phosphatase guides migrating tion of the empty spiracles gene in adult
germ cells. Development 128(6):983–991 brain development of Drosophila melano-
77. Yu XM, Wu, J., Sweeney, L., Berdnik, D. gaster. Development 134(7):1291–1300.
(2009) Drosophila adult brain dissection made doi: dev.02814 [pii] 10.1242/dev.02814
 Part II

Other Arthropods
 Chapter 7

Dye Coupling and Immunostaining of Astrocyte-Like Glia

Following Intracellular Injection of Fluorochromes in Brain
Slices of the Grasshopper, Schistocerca gregaria
George Boyan and Yu Liu

Abstract
Injection of fluorochromes such as Alexa Fluor® 568 into single cells in brain slices reveals a network of
dye-coupled cells to be associated with the central complex. Subsequent immunolabeling shows these cells
to be repo positive/glutamine synthetase positive/horseradish peroxidase negative, thus identifying them
as astrocyte-like glia. Dye coupling fails in the presence of n-heptanol indicating that dye spreads from cell
to cell via gap junctions. A cellular network of dye-coupled, astrocyte-like, glia surrounds and infiltrates
developing central complex neuropils. Intracellular dye injection techniques complement current molecular
approaches in analyzing the functional properties of such networks.

Key words Insect, Brain slices, Dye injection, Immunolabeling, Glia, Gap junctions

1 Introduction

Glia have been shown to play a major regulatory role in establishing

neuroarchitecture throughout the insect nervous system [1–4],
and those with extensive dendritic projections (or gliopodia) envel-
oping and infiltrating neuropils such as the central complex are of
special interest to studies of brain development [5–9]. Glial/glial
and glial/neuronal communication in both invertebrate [10–15]
and vertebrate [16–22] nervous systems has been shown to involve
gap junctions. In insects such as the grasshopper, gap junctions
assemble in glial cells during the latter half of embryogenesis [11]
resulting in networks of electrically coupled cells [10].
One accepted indicator of gap junctional communication is
the presence of dye coupling as revealed by injected fluorescent
dyes [15, 23–25]. In this chapter we describe a combined intracel-
lular dye injection/immunolabeling method for demonstrating
dye coupling between cells in unfixed, frozen, brain slices of the
embryonic grasshopper Schistocerca gregaria and then identifying

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_7, © Springer Science+Business Media, LLC 2014

99
100 George Boyan and Yu Liu

these cells as being astrocyte-like glia associated with the central

complex [26]. Dye coupling is completely blocked in the presence
of n-heptanol, an acknowledged gap junctional blocker [27–30],
in the bathing medium. Intracellular dye injection and immunola-
beling are also successful in agarose-embedded brains but the qual-
ity of dye coupling is affected by the necessary fixation procedure
(see ref. 30). Our method using frozen, unfixed, brain slices reveals
a network of dye-coupled glia to surround developing central com-
plex modules [26]. Such intracellular studies complement existing
molecular approaches for examining glial interactions in brain
development [6, 8].

2 Materials

2.1 Frozen 1. Tissue Freezing Medium® (125 mL dripper bottle, Jung).

Brain Slices 2. 2-methylbutane maintained at −20 °C for shock freezing tissue.
3. Make a set of glass pipettes, which have been broken back via a
diamond knife and the tips flamed blunt in a Bunsen burner.
The opening should be large enough to accommodate via suction
the heads/brains of all embryo stages being examined. These
pipettes will henceforth be referred to as “modified pipettes.”
4. Plastic wells (Pelco) filled with tissue freezing medium into the
brains are placed for freezing. These wells automatically shape
and size the frozen blocks for mounting in the Cryostat (Leica
cm3050s).
5. Freeze an appropriate number of Superfrost® Plus (Menzel-
Gläser) microscope slides.
6. Pap pen (G. Kisker) to draw a fluid-repelling boundary line
around the brain slice on the slide. Store at 4 °C.
7. Indelible blue lab marker (Sigma) for labeling the slide and
marking brain slice location (on the glass underside).
8. Prefrozen glass slide container (cuvette) with vertical slots for
storing up to ten glass slides.
9. Freezer held at −18 to −20 °C for storing containers with sec-
tioned tissue.
10. Cover slips (17 μm thickness).
11. Fast-drying nail varnish.

2.2 Dissection and 1. Fine forceps, sizes #3, #5 (both Dumont, Switzerland).
Associated Supplies 2. Iridectomy scissors (Martin, Germany).
3. Micropipette (Eppendorf) with exchangeable tips.
4. Packages of exchangeable 1 μL tips for antibody delivery,
and 50 μL tips for medium exchange (we use Ultratip Greiner
Bio-one).
 Dye Coupling Between Glia in Brain Slices 101

5. Obtain a fine (30 G) needle and matching syringe for injecting

the 200 mM KCl into the glass electrode from its blunt end.
6. Binocular microscope (×10 oculars) with adequate zoom
capabilities (×10) for dissection of embryos and brains.

2.3 Solutions, Media 1. 0.1 M PBS: 2.6 g/L NaH2PO4, 22.5 g/L Na2HPO4,
102.2 g/L NaCl, adjust to pH 7.4 with NaOH. Store at 4 °C.
Take 100 mL PBS and store over ice on the day of making
frozen sections.
2. Prepare an ice-cold 1 % methylene blue solution in 0.1 M PBS
and place an aliquot in each of several ice-cold Eppendorf
(100 μL) containers. Embryonic grasshopper heads are small
(up to 1 mm) so several heads can be placed in one Eppendorf.
3. Physiological saline: 8.8 g/L NaCl, 0.2 g/L KCl, 0.3 g/L
CaCl2, 2.5 g/L MgSO4, 0.9 g/L TES buffer. Adjust to pH 7.0
with the buffer. Store at 4 °C. Decant 10 mL for each experi-
mental sitting and allow to reach room temperature prior to
the experiment.
4. Fetal Calf Serum (FCS), store frozen (−18 °C). Maintain an
aliquot of 10 mL in an Eppendorf at 4 °C. This fluid will be
used for dipping the electrode tip into and for internally coating
the glass pipettes with which heads/brains are transferred.
5. Aqueous bouin: picric acid 71 %, formaldehyde 24 %, glacial
acetic acid 5 %. Fixative for immunolabeling with both anti-
glutamine synthetase (for glia) and horseradish peroxidase (for
neurons). Store under a fume hood at room temperature.
6. 3.7 % paraformaldehyde (PFA) as the fixative for double gluta-
mine synthetase/repo immunolabeling. Store in a fume hood
at room temperature.
7. PBT containing NHS: 1 % normal horse serum (NHS), 0.1 %
bovine serum albumin (BSA), PBT (1 % Triton X-100 in 0.1 M
PBS). Adjust to pH 7.4 with buffer. Store at 4 °C in the dark.
8. After dye injection and immunolabeling, we use Vectashield®
(Vector laboratories) to cover brain slices and minimize bleach-
ing during fluorescence microscopy.
9. The fluorochrome we use here is Alexa Fluor® 568 (Invitrogen,
emission maximum 600 nm), but we also used Alexa Fluor®
488 (Invitrogen, emission maximum 517 nm) and Lucifer
Yellow CH (Aldrich, emission maximum 528 nm) (see ref. 30
and Note 1). Alexa Fluor® 568 as a hydrazine has a low molec-
ular weight (790 Da) and so flows readily throughout cells on
injection. Store the stock at −18 °C. Make up a 0.1 mL aliquot
of 10 mM Alexa Fluor® 568 in 200 mM KCl and store in a
glass ampule (3-1801, neoLab, Heidelberg) at 4 °C in the
dark. Maintain 10 mL of 200 mM KCl at room temperature
for backfilling each micropipette.
102 George Boyan and Yu Liu

10. Make up 50 mL of 1 × 10−5 M n-heptanol in TES buffer.

This concentration is based on experiments in the embryonic
grasshopper brain [26, 30] and in intact, unfixed, Hydra [31].
Store at 4 °C.

2.4 Primary 1. Anti-glutamine synthetase (BD Transduction Laboratories,

Antibodies mouse anti-6/GS, Nr. 610518). Glutamine synthetase catalyzes
the conversion of ammonia and glutamate to glutamine [32].
It is found in glial cells of the vertebrate brain [33, 34] and
particularly in astrocytes [35]. The glutamine synthetase anti-
body recognizes an octamer of identical 45 kDa subunits.
2. Anti-horseradish peroxidase (rabbit, Dianova, 323-005-021).
HRP belongs to the cell surface glycoproteins which include
cell adhesion and signal molecules [36]. In insects the epitope
is neuron-specific [37, 38].
3. Anti-Repo. The expression of the glial-specific homeobox gene
reversed polarity (repo) is revealed by using the anti-Repo primary
antibody (Mab 4a3) (see ref. 39).

2.5 Secondary 1. For anti-glutamine synthetase we used donkey anti-mouse

Antibodies (DAM)-Alexa Fluor® 488 (Invitrogen, emission 519 nm).
2. For anti-HRP and anti-repo, we used either goat anti-rabbit
(GAR)-Cy5 (emission 670 nm, Dianova) or goat anti-rabbit
(GAR)-Cy3 (emission 570 nm, Dianova). Choose as appropri-
ate to avoid spectral confusion in double-labeling experiments
(see Note 3).

2.6 Electrodes 1. We used micropipettes made of thin-walled borosilicate glass

for Dye Injection (GC 100TF-10, Clark Electromedical Instruments), with
1.0 mm outer diameter (O.D.) and 0.78 mm inner diameter
(I.D.). An inner filament assists fluid transfer into the tip.
Pipettes are packaged into lots of 100 and are precut to a
length of 10 cm.
2. To manufacture electrodes you will need an automated electrode
puller (e.g., Sutter Instruments, USA) equipped with nitrogen
gas cooling of the filament to regulate electrode-tip length.
Microelectrodes must suit the immediate purpose and are
therefore very individual, but if you have an electrode puller
similar to ours, try the following settings: heat 612, pull 186,
velocity 100, time 80, nitrogen 1.05, and pressure 5 bar, to
make microelectrodes of 5 cm overall length, 7 mm tip length
with resistances of 30–40 MΩ when filled with the dye solu-
tion (10 mM Alexa Fluor® 568, Invitrogen, in 200 mM KCl),
and suitable for filling cell bodies of about 12 μm diameter.
The electrode should not be too long otherwise bending and
high resistances present problems; if the electrode is too short,
it will be blunt and make too large a hole in the cell with
 Dye Coupling Between Glia in Brain Slices 103

consequent leaking of dye. Our puller makes pairs of electrodes

of equal length and quality so make a batch of maximally 20
electrodes at one sitting but use these within 2–3 days. The
tips are very fine and since glass is a fluid crystal it will deform
with time, making dye injection more difficult.
3. Take a glass or plastic petri dish (10 cm diameter, with cover)
and mold fresh malleable plasticine to a flattened block, which
is then placed to fit across the diameter of the petri dish.
The block should be about 5 cm broad and about 1 cm thick.
Take one microelectrode and make repeated impressions in the
plasticine block at evenly spaced intervals so that about 6–8
microelectrodes can be stored side by side in the one petri dish.
Cover the dish to prevent dust access (glass carries electrostatic
charge which attracts dust) and store in a fume hood.
4. A perspex electrode holder for 1.0 mm O.D. glass micropi-
pettes (WPI Instruments, FL, USA). Make sure that the elec-
trode holder has a pressure release opening otherwise the
electrode may move during penetration as internal fluid pres-
sures change. The electrode holder contains an internal
Ag/AgCl metal site for contact with the electrolyte in the elec-
trode and a pin to which a shielded cable from the head stage
(positive pole) of a DC amplifier (see below) is connected.
5. You will need a 4–5 cm length of chlorided silver wire to act as
a reference electrode. The simplest method for chloriding is to
place about 5 g of silver chloride (AgCl) in a small fireproof
container over a Bunsen burner and heat to melting. Repeatedly
dip the end (approximately 2 cm) of the silver wire (held by
wooden forceps) into the melted AgCl until an even coating
has been obtained. This coated length dips into the physiologi-
cal saline covering the brain slice and the chloride coating will
prevent polarization voltages developing between reference
and recording electrode. Solder the non-chlorided end of the
silver wire to a shielded cable (shielding connects to ground)
and attach to the negative pole of the head stage of the DC
amplifier. Over time the chloride coating will develop cracks
and necessitate repeating the coating procedure.
6. Micromanipulator (e.g., Leitz) equipped with fine adjustment
(manual or motorized) of X, Y, Z axes. This guides the electrode
holder with its micropipette to the target cell.

2.7 Electronics 1. DC amplifier (we use a Getting 5) equipped with a head stage
that is positioned as near to the preparation as possible to
reduce electrical interference, and with current passing facility
via a virtual ground circuit.
2. A simple 2-channel oscilloscope for monitoring current injection
and electrode resistance. Use the electrode resistance-testing
capability of the amplifier to check that the electrode has a
104 George Boyan and Yu Liu

resistance of 30–40 MΩ when filled with the fluorochrome

used (10 mM Alexa Fluor® 568, Invitrogen, in 200 mM KCl).
Resistances can be calculated by injecting repeated 1 nA current
pulses (of 100 ms duration) across the bridge circuit of the
amplifier and cancelling the evoked voltage shift being viewed
on the oscilloscope screen with an internally calibrated current
of opposite polarity.

2.8 Fixed-Stage 1. In a fixed-stage microscope, the optics not the stage are raised
Compound Microscope and lowered, allowing focusing of tissues with intracellular
and Camera electrode in position. You will need a fixed-stage compound
microscope (e.g., Zeiss Axioskop 2) equipped with filters for
both epifluorescence illumination (tetramethyl rhodamine iso-
thiocyanate, TRITC, emission 580 nm; fluorescein isothiocya-
nate, FITC, emission 528 nm) and differential interference
contrast (DIC)/transmitted light optics. The microscope
should stand inside an earthed Faraday cage to minimize elec-
trical interference from lamps and circuits in the building, and
on a vibration-free table (preferably supported by compressed
air) for stability so as to optimize microelectrode positioning.
2. The microscope should be equipped with a low-power (×10)
objective (e.g., Zeiss Plan-Neofluar, 440330, n.A. 0.30) for
centering the brain slice and electrode tip and a high-power
(×63) water-immersion objective with long working distance
(2.1 mm) for monitoring the actual intracellular penetration
(e.g., Zeiss Achroplan ceramic-coated ×63 objective, 440067-
9901, n.A. 0.95 W). The ceramic renders the objective elec-
trically inert when placed in the saline solution covering the
brain slice.
3. A color CCD camera (1.3 MP is adequate) mounted on the
microscope, driver software for a computer, a flatscreen monitor
(at least 19 in. diagonal and 1,640 × 1,050 pixel resolution is
recommended). A micrometer scale etched into a glass slide
(Zeiss 473390-9901) is used for checking that the microscope/
camera adapter provides a 1:1 image size on the monitor.

2.9 Imaging 1. To reveal fine structures such as gliopodia in 3D, you need
to acquire optical sections (a so-called Z stack) of prepara-
tions using a confocal laser scanning microscope. We used a
Leica TCS SP5 equipped with Leica Hc Pl Apo CS ×20 and
Leica Hcx Pl Apo lambda blue ×63 oil immersion objectives
(see Note 4).
2. Fluorochromes were visualized using an argon laser with exci-
tation wavelengths of 488 nm for Alexa Fluor® 488, 578 nm
for both Alexa Fluor® 568 and Cy3, and 633 nm for Cy5.
3. Z-stacks of confocal images were processed using public
domain software (ImageJ).
 Dye Coupling Between Glia in Brain Slices 105

3 Methods

3.1 Preparing Frozen 1. Using the #5 forceps and iridectomy scissors, which have
Brain Slices previously been dipped in FCS to prevent tissue adhesion, dis-
sect embryos out of the egg into ice-cold 0.1 M PBS and free
embryos from surrounding membranes. For grasshoppers,
embryos are staged according to percentage of development
(see ref. 40).
2. For younger embryos (<60 %), decapitate embryos with a single
snip of the iridectomy scissors. For older embryos (>70 %), the
cuticle becomes an obstacle during cryosection. Therefore
open the head capsule by cutting away the front with the iri-
dectomy scissors, cut through the left and right optic stalks,
and extract the brain carefully from the head capsule.
3. Take a modified glass pipette and suck up FCS repeatedly to
coat the inside. This prevents tissue adhering to the glass.
Use the modified pipette to now suck up single heads/brains
and transfer to an ice-cold 1 % methylene blue solution in
0.1 M PBS for 5 min. to stain the tissue surface and thus facili-
tate visual orientation of tissue within the frozen blocks for
sectioning.
4. Transfer the heads/brains, unfixed, into the wells containing
Tissue Freezing Medium®. Shock freeze the wells by placing
them in a cuvette containing 2-methylbutane maintained at
−20 °C. Do not add sucrose as this adversely affects the quality
of later dye coupling.
5. Orient the frozen block containing the head/brain in the
Cryostat according to the desired plane of sectioning. We gen-
erally sectioned horizontally with respect to the body axis and
at 16 μm thickness using a Leica low profile 819 microtome
blade. Replace this blade as required. This slice thickness
allowed individual cells surrounding the central body of the
brain to be easily targeted for intracellular dye injection under
DIC optics and without damaging primary glial processes.
6. Each brain slice (still frozen and attached to the blade) is trans-
ferred by direct application onto a separate frozen Superfrost®
Plus (Menzel-Gläser) microscope slide. For older embryos the
depth of the brain could mean that 20 or so slides are neces-
sary. This 1:1 procedure is important because it means that
only one brain slice is being investigated (at room tempera-
ture) at any given time. Any additional brain slices on the slide
would dry out during the experiment.
7. Draw a complete ring around the section with the Pap pen
for frozen slides. This ring will limit the spread of the saline
solution covering the tissue slice and allow maximum fluid
contact between the slide and the water-immersion objective.
106 George Boyan and Yu Liu

We also marked the location of the brain slice itself by placing a

dot on the underside of the glass slide with the indelible marker.
8. Transfer all the frozen slides from one brain to the same
prefrozen glass slide container and store containers in a freezer
at −18 °C. We stored sectioned tissue for up to 5 days without
noticeable changes in tissue preservation or dye coupling.

3.2 Intracellular 1. In preparation for dye injection, transfer the frozen slide with
Dye Injection its single brain slice to the fixed-stage microscope. Immediately
cover with physiological saline at room temperature to a depth
of at least 5 mm to allow good contact with the immersion
objective. Too little fluid and the surface tension forces gener-
ated by the objective might tear the slice from the slide. Allow
1–2 min for temperature acclimatization. Confirm using DIC
optics at ×10 magnification that the preparation is centered
within the optical field of the microscope. This is critical because
once the ×63 objective is in use, repositioning of the slice is
limited by the restricted visual field offered by the objective.
2. Take the Eppendorf containing FCS from the refrigerator and
allow to come to room temperature.
3. Remove an electrode carefully from the array within the petri
dish (too sudden a movement may cause the tip to snap off!).
Fill the electrode with the fluorochrome solution by placing its
blunt end vertically into the solution and allowing it to travel
via capillary action along the internal filament into the tip.
This will take about 1–2 min. The entire electrode should then
be filled with 200 mM KCl from its blunt end via the fine
syringe needle to enable a good electrical contact with the Ag/
AgCl electrode within the electrode holder. Dip the electrode
tip several times into the FCS to act as a lubricant facilitating
penetration and subsequent removal of the electrode from the
target cell.
4. Slide the electrode into the perspex microelectrode holder
(1.0 mm diameter, WPI Inc., FL, USA). Ensure there are no
salt residues on the electrode holder by wiping down with a
tissue wetted with aqua dest.
5. Position the electrode holder into an appropriate microma-
nipulator equipped with adjustment (manual or motorized) of
X, Y, Z axes. The shielded cable from the microelectrode
holder is connected to the positive pole on the head stage of a
DC amplifier (in our case a Getting 5) equipped with current
passing facility via a virtual ground circuit.
6. Place the tip of the chlorided silver wire acting as a reference
electrode into the physiological solution covering the prepa-
ration and connect to the negative pole of the amplifier head
stage.
 Dye Coupling Between Glia in Brain Slices 107

7. Using transmitted light, position the microelectrode tip so

that it is centered in the field of view of the ×10 objective on
the compound microscope. Lower the electrode via the micro-
manipulator so that the tip dips into the saline covering the
preparation, and continue lowering until the tip is located just
above the surface of the brain slice.
8. Now switch on your preamplifier and use the electrode
resistance-testing capability of the amplifier to check that the
electrode has a resistance of 30–40 MΩ when filled with the
fluorochrome. Resistances can be calculated by injecting a
1 nA current pulse (of 50 ms duration) across the bridge
circuit of the amplifier and cancelling the evoked voltage shift
being viewed on the oscilloscope screen with an internally cali-
brated current of opposite polarity. If the resistance exceeds
50–60 MΩ, reject the electrode as it will probably block on dye
injection. Also see Note 2 for checking your system for electri-
cal neutrality.
9. Switch to the high-power (×63) water-immersion objective.
Lower the objective into the saline and continue lowering until
the microelectrode tip is in view. Center the microelectrode tip
and lower to the tissue surface while constantly changing the
focus so that the target cell also comes into view. The angle of
deflection of the microelectrode holder will have to be adjusted
so that the microelectrode tip does not travel through too
much nontarget tissue (we use a 15° depression angle). The
steeper the angle, the closer the tip will have to be positioned
to the cell before penetration. Microelectrode tip and target
cell are both kept in focus and the electrode tip is propelled
into the cell via the micromanipulator (Fig. 1a).
10. Intracellular penetration of the target cell is monitored using
the DIC optics (transmitted light) of the Zeiss Axioskop 2
microscope. We performed penetrations exclusively of the
cell soma to avoid damage to, and inadvertent staining of, pro-
jections from neighboring cells. On penetration, switch from
transmitted light to epifluorescence (for Alexa Fluor® 568, use
rhodamine or TRITC filters; see ref. 30 for settings).
11. Commence dye injection by applying constant hyperpolarizing
current via the amplifier and observe the initial passing of dye
into the cell (Fig. 1b). The fluorochrome carries a charge so
use negative current to inject dye into the cell. Do not exceed
5 nA injected current as monitored on an oscilloscope screen.
Use the microscope shutter to block the fluorescence excita-
tion and continue injection in the dark. Unblock occasionally
to check injection progress on the computer screen. Make
photographs with the camera software for documentation
purposes (see Note 3). In our case dye injection proceeded for
108 George Boyan and Yu Liu

Fig. 1 Intracellular injection of Alexa Fluor® 568 reveals dye-coupled glia cells associated with the central
complex of the embryonic grasshopper brain. Photomicrographs of horizontal embryonic brain slices: (a–c)
normal, (d–f) after addition of bathing medium containing 1 × 10−5 M n-heptanol (a gap junctional blocker).
(a, d) DIC images show dye-filled (purple) electrode (white arrowheads) on penetration of a single target cell.
Central complex neuropils (CB central body, EB ellipsoid body, N nodulus) are outlined dashed white. Arrow
indicates anterior (a). Combined transmitted light/fluorescence images show the brain slices on commence-
ment (b, e: 0 min) and after 10 min. (c, f) Dye injection. Note the large collection of dye-coupled cells normally
(c) and their complete absence when n-heptanol is present (f). Scale bar represents 70 μm. Panels modified
from ref. 26

10–15 min beyond which there was no change in the quality of

staining or dye coupling (Fig. 1c). Constantly monitor elec-
trode resistance on the oscilloscope (see Note 4).
12. On completion of staining, switch off the amplifier, revert to
transmitted light, and use the micromanipulator to carefully
withdraw the microelectrode tip from the cell. Remove the
reference electrode from the solution. Clean the ×63 objective
with distilled water to remove salt residues.

3.3 Gap Junctional 1. Alexa Fluor® 568 has a small molecular weight [41] so that its
Blocker passage between glia cells might be mediated by gap junctions.
We tested this possibility by applying TES buffer containing
1 × 10−5 M n-heptanol—a proven gap junctional blocker
(see refs. 28, 29)—and then repeating the dye injection experi-
ment as described above (Fig. 1d).
2. In each experiment involving n-heptanol we found that only
the penetrated cell was stained after 10 min of injection with
Alexa Fluor® 568 (Fig. 1e). This shows that the injection of the
dye itself was not affected by the presence of n-heptanol.
Complete absence of dye coupling (Fig. 1f) supports the
presence of gap junctions between the labeled cells in our
brain slices.
 Dye Coupling Between Glia in Brain Slices 109

3.4 Immunolabeling 1. Immunolabeling is performed directly on the brain slice still

attached to the glass slide. Remove the slide from the micro-
scope and immediately apply fixative (aqueous Bouin or para-
formaldehyde as appropriate) to the saline covering the slice.
Drain the fixative/saline mixture from the slice after 1–2 min
and replace with fresh pure fixative. Keep slides overnight at
4 °C in the dark.
2. After fixation drain the fixative using a micropipette with an
exchangeable 50 μL tip and replace with 0.1 M PBS as a wash-
ing solution. Exchange this solution at least 6 times over the
course of 1 h. Bring a 5 mL aliquot of preincubation medium
to room temperature. Then exchange the washing solution
with preincubation medium for 2 h at room temperature to
block unspecific binding sites.
3. The following primary antibodies were then applied to the brain
slice, either singly or in combination depending on the data
required: (a) anti-glutamine synthetase, dilute 1:200 in the
preincubation medium; (b) anti-horseradish peroxidase, dilute
1:150 in preincubating medium; and (c) anti-repo, dilute 1:500
in preincubation medium. Drain the preincubation medium
currently covering the slice and replace with fresh medium con-
taining the primary antibody(s). In each case, slices remain
exposed to primary antibodies for 3 days at 4 °C in the dark.
4. After exposure to the primary antibody(s), drain medium
from the brain slice with a micropipette and wash sections by
repeated exchanges of 0.1 M PBS using a fresh pipette tip.
Then replace with the same preincubation medium as above
to which the relevant secondary antibody has been added as
follows: for anti-glutamine synthetase, DAM-Alexa Fluor®
488:1:300 dilution; for anti-HRP, GAR-Cy5 or GAR-Cy3:1:150
dilution; and for anti-Repo, GAR-Cy5 or GAR-Cy3:1:150
dilution. See Note 5 on avoiding spectral confusion. Exposure
of the brain slice to the secondary antibody(s) was for 24 h at
4 °C in the dark.
5. Incubation medium was then removed and replaced by several
changes of 0.1 M PBS. The brain slices remained in PBS over-
night at 4 °C in the dark and were subsequently covered in
Vectashield® (Vector laboratories) to minimize photobleach-
ing. Cover slips were applied and sealed to the glass slide with
fast-drying nail varnish to prevent movement under the oil
immersion objectives used in confocal microscopy.

3.5 Imaging 1. We acquired optical sections of preparations with a Leica TCS

SP5 confocal laser scanning microscope equipped with ×10
oculars and ×20 and ×63 oil immersion objectives (see Note 6).
2. Confocal images following double immunolabeling carried
out on brain slices prepared as above or via agarose embedding
110 George Boyan and Yu Liu

Fig. 2 Immunolabeling confirms glial identity of cells. (a) Confocal image of a brain
slice following agarose embedding and double immunolabeling shows cells
associated with the central body (CB) to be glutamine synthetase (GS) positive
(green)/HRP negative (red). Two cells (white stars) direct glia podia around the CB
(open white arrowhead) as well as (white arrowhead) into neighboring neuropil
of the protocerebrum (PC). Arrow indicates anterior (a). (b, c) All GS-positive cells
(green) associated with the ellipsoid body (EB) (b), and central body (CB) (c),
co-express the glial-specific nuclear antigen Repo (red, white stars) confirming
their identity as astrocyte-like glia. Note the extensive GS-positive gliopodia
(white arrowheads) surrounding these neuropils. (d–f) Confocal images confirm
glial identity of a pair of dye-coupled cells (black, white stars) associated with the
central body (CB) following Alexa Fluor® 568 injection in a brain slice. (d) Red
channel shows dye-injected (black star) and dye-coupled (white star) cells
following dye injection. (e) Green channel shows both cells to be glutamine syn-
thetase positive. (f) Combined channels confirm co-labeling (yellow) of anti-GS
and Alexa and therefore glial identity of both cells. Scale bar in (a) represents
12 μm, 25 μm in (b), 20 μm in (c–f). Panel (a) modified from ref. 26; panels (b, c)
modified from ref. 9

reveal that the cells surrounding central complex neuropils are

both glutamine synthetase positive/HRP negative (Fig. 2a)
and glutamine synthetase positive/repo positive (Fig. 2b, c),
confirming their identity as astrocyte-like glia. Confocal exami-
nation of a brain slice following injection of Alexa Fluor® 568
into a single cell associated with the central body reveals
 Dye Coupling Between Glia in Brain Slices 111

dye-coupled cells (Fig. 2d) which on immunolabeling prove to

be glutamine synthetase positive (Fig. 2e, f), consistent with
their being astrocyte-like glia (Fig. 2b, c).

4 Notes

1. The Alexa Fluor® 568 fluorochrome was preferred over Alexa

Fluor® 488 and Lucifer Yellow (which also revealed dye cou-
pling) because of its low molecular weight, staining intensity,
and because its emission wavelength is sufficiently distant so as
to prevent spectral overlap with those of the secondary antibod-
ies used in the subsequent immunohistochemistry. For a com-
parison of results using these various fluorochromes, see [30].
2. You will need to check the electrical neutrality of your record-
ing/injection system with an artificial electrode (a 10 MΩ
resistor in parallel with a 4.7 pF ceramic capacitor) connected
to your preamplifier and oscilloscope to ensure that no 60-cycle
interference is present. Once the real recording and reference
electrodes are in place, you will need to check that no voltage
displacement is present on the oscilloscope. If there is, it is a
sign that current is flowing across the electrode resistance and,
depending on its polarity, this could lead to uncontrolled dye
release into the tissue prior to penetration of a cell. Salt bridges
caused by saline residues on microscope or electrode holder
are a common source of polarizations. Wipe these down regu-
larly using ion-free water.
3. To allow better definition of stained cells and considerably
reduce background when the brain slice was simultaneously
illuminated with transmitted light on the Zeiss Axioskop 2
fluorescence microscope, the injected Alexa® Fluor 568 was
also excited at violet excitation wavelengths (430–440 nm)
and, using the camera-specific software (Visicapture™),
captured at yellow/orange emission wavelengths (535 nm).
4. If resistance rises sharply under current injection, it indicates
that the electrode is blocking and dye is no longer flowing
freely into the cell. If this happens try to relieve the block with
repeated pulses of positive current. If this does not work, break
off injection, turn on fluorescence, and examine the quality of
staining. If adequate, apply fixative to the preparation.
Otherwise start again with a fresh electrode and another target
cell elsewhere in the slice. Never reuse an electrode as tissue
invariably adheres to the tip causing blockage.
5. In experiments that involved combinations of antibodies with
Alexa Fluor® 568, the fluorochrome conjugated with each
secondary antibody needs to emit at a distinctly different wave-
length to avoid spectral confusion. The advantage of the Cy5
112 George Boyan and Yu Liu

fluorochrome in this regard is that it signals in infrared

(670 nm) and so can be allocated any color by imaging software
after confocal scanning.
6. Tip: Use the ×63 objective, rather than digitally zooming the
×20 objective, to capture high-resolution morphological
details of the stained cells.

Acknowledgements

This work was supported by DFG grant BO 1434/3–5 and the

Graduate School of Systemic Neuroscience, University of Munich.
We thank S. Götz for assistance with dye injection and immunola-
beling of glia (Fig. 2d–f).

References
1. Jones BW, Fetter RD, Tear G, Goodman CS barrier of an insect: recordings from identified
(1995) Glial cells missing: a genetic switch that neuroglia. J Exp Biol 109:307–318
controls glial versus neuronal fate. Cell 82: 11. Swales LS, Lane NJ (1985) Embyronic develop-
1013–1023 ment of glial cells and their junctions in the locust
2. Jones BW (2001) Glial cell development in the central nervous system. J Neurosci 5:117–127
Drosophila embryo. Bioessays 23:877–887 12. Schmidt J, Deitmar JW (1996) Photoin-
3. Hidalgo A (2003) Neuron-glia interactions activation of the giant neuropil glial cells in the
during axon guidance in Drosophila. Biochem leech Hirudo medicinalis: effects on neuronal
Soc Trans 31:50–55 activity and synaptic transmission. J Neuro-
4. Klämbt C (2009) Modes and regulation of physiol 76:2861–2871
glial migration in vertebrates and invertebrates. 13. Alexopoulos H, Böttger A, Fischer S et al
Nat Rev Neurosci 10:769–779 (2004) Evolution of gap junctions: the missing
5. Vanhems E (1995) Insect glial cells and their link? Curr Biol 14:879–880
relationships with neurons. In: Vernadakis A, 14. Phelan P, Goulding LA, Tam JLY et al (2008)
Roots B (eds) Neuron-Glia interrelations dur- Molecular mechanism of rectification at an
ing phylogeny: II. Plasticity and regeneration. identified electrical synapse in the Drosophila
Humana, Totowa, NJ, pp 49–77 giant fibre system. Curr Biol 18:1955–1960
6. Awasaki T, Lai S-L, Ito K, Lee T (2008) 15. Koussa MA, Tolbert LP, Oland LA (2011)
Organization and postembryonic development Development of a glial network in the olfactory
of glial cells in the adult central brain of nerve: role of calcium and neuronal activity.
Drosophila. J Neurosci 28:13742–13753 Neuron Glia Biol 6(4):245–261
7. Edwards TN, Meinertzhagen IA (2010) The 16. Zahs KR, Newman E (1997) Asymmetric gap
functional organisation of glia in the adult junctional coupling between glial cells in the
brain of Drosophila and other insects. Prog rat retina. Glia 20:10–22
Neurobiol 90:471–497 17. Newman EA (2001) Propagation of intercel-
8. Awasaki T, Lee T (2011) New tools for the lular calcium waves in retinal astrocytes and
analysis of glial cell biology in Drosophila. Glia Müller cells. J Neurosci 21:2215–2223
59:1377–1386 18. Theiss C, Meller K (2002) Aluminum impairs
9. Boyan G, Loser M, Williams L, Liu Y (2011) gap junctional intercellular communication
Astrocyte-like glia associated with the embry- between astroglial cells in vitro. Cell Tissue Res
onic development of the central complex in the 310:143–154
grasshopper Schistocerca gregaria. Dev Genes 19. Houades V, Koulakoff A, Ezan P et al (2008)
Evol 221:141–155 Gap junction-mediated astrocytic networks
10. Schofield PK, Swales LS, Trehern JE (1984) in the mouse barrel cortex. J Neurosci 28:
Potentials associated with the blood-brain 5207–5217
 Dye Coupling Between Glia in Brain Slices 113

20. Tanaka M, Yamaguchi K, Tatsukawa T et al grasshopper Schistocerca gregaria. Dev Genes

(2008) Connexin43 and bergmann glial gap Evol 220:297–313
junctions in cerebellar function. Front Neurosci 31. Chapman JA, Kirkness EF, Simakov O et al
2:225–233 (2010) The dynamic genome of Hydra. Nature
21. Parys B, Cote A, Gallo V et al (2010) 464:592–596
Intercellular calcium signaling between astro- 32. Wedler FC, Horn BR (1976) Catalytic mecha-
cytes and oligodendrocytes via gap junctions in nisms of glutamine synthetase enzymes. J Biol
culture. Neuroscience 167:1032–1043 Chem 251:7530–7538
22. Rela L, Bordey A, Greer CA (2010) Olfactory 33. Martinez-Hernandez A, Bell KP, Norenberg
ensheathing cell membrane properties are MD (1977) Glutamine synthetase: glial local-
shaped by connectivity. Glia 58:665–678 ization in brain. Science 195:1356–1358
23. Hossain MZ, Ernst LA, Nagy JI (1995) Utility 34. van der Hel WS, Notenboom RGE, Bos IWM
of intensely fluorescent cyanine dyes (Cy3) for et al (2005) Reduced glutamine synthetase in
assay of gap junctional communication by dye- hippocampal areas with neuron loss in tempo-
transfer. Neurosci Lett 184:71–74 ral lobe epilepsy. Neurology 64:326–333
24. Ball KK, Gandhi GK, Thrash J et al (2007) 35. Ward M, Jobling A, Puthussery T et al (2004)
Astrocytic connexin distributions and rapid, Localization and expression of the glutamate
extensive dye transfer via gap junctions in the transporter, excitatory amino acid transporter
inferior colliculus: implications for [14C] 4, within astrocytes of the rat retina. Cell Tissue
Glucose metabolite trafficking. J Neurosci Res Res 315:305–310
85:3267–3283 36. Snow PM, Patel NH, Harrelson AL et al
25. Lanosa XA, Reisin HD, Santacroce I et al (2008) (1987) Neural-specific carbohydrate moiety
Astroglial dye-coupling: an in vitro analysis of shared by many surface glycoproteins in
regional and interspecies differences in rodents Drosophila and grasshopper embryos. J Neurosci
and primates. Brain Res 1240:82–86 7:4137–4144
26. Boyan GS, Liu Y, Loser M (2012) A cellular 37. Jan LY, Jan YN (1982) Antibodies to
network of dye-coupled glia associated with horseradish-peroxidase as specific neuronal
the embryonic central complex in the grass- markers in Drosophila and grasshopper embryos.
hopper Schistocerca gregaria. Dev Genes Evol Proc Natl Acad Sci U S A 79:2700–2704
222:125–138 38. Haase A, Stern M, Wächtler K et al (2001)
27. Burt JM, Spray DC (1980) Single-channel A tissue-specific marker of Ecdysozoa. Dev
events and gating behavior of the cardiac gap Genes Evol 211:428–433
junction channel. Proc Natl Acad Sci U S A 39. Halter DA, Urban J, Rickert C et al (1995)
85:3431–3434 The homeobox gene repo is required for the
28. Weingart R, Bukauskas FF (1998) Long-chain differentiation and maintenance of glial func-
n-alkanols and arachidonic acid interfere with tion in the embryonic nervous system of
the Vm-sensitive gating mechanisms of gap Drosophila melanogaster. Development 121:
junction channels. Eur J Physiol 435:310–319 317–322
29. Juszczak GR, Swiergiel AH (2009) Properties of 40. Bentley D, Keshishian H, Shankland M et al
gap junction blockers and their behavioural, cog- (1979) Quantitative staging of embryonic
nitive and electrophysiological effects: animal and development of the grasshopper, Schistocerca
human studies. Prog Neuropsychopharmacol nitens. J Embryol Exp Morphol 54:47–74
Biol Psychiatry 33:181–198 41. Weber PA, Chang HC, Spaeth KE et al (2004)
30. Boyan GS, Niederleitner B (2011) Patterns of The permeability of gap junction channels to
dye coupling involving serotonergic neurons probes of different size is dependent on con-
provide insights into the cellular organization nexin composition and permeant-pore affinities.
of a central complex lineage of the embryonic Biophys J 87:958–973
 Part III

Molluscs
 Chapter 8

Methods in Brain Development of Molluscs

Andreas Wanninger and Tim Wollesen

Abstract
Representatives of the phylum Mollusca have long been important models in neurobiological research.
Recently, the routine application of immunocytochemistry in combination with confocal laser scanning
microscopy has allowed fast generation of highly detailed reconstructions of neural structures of even the
smallest multicellular animals, including early developmental stages. As a consequence, large-scale com-
parative analyses of neurogenesis—an important prerequisite for inferences concerning the evolution of
animal nervous systems—are now possible in a reasonable amount of time. Herein, we describe immuno-
cytochemical staining protocols for both whole-mount preparations of developmental stages—usually
70–300 μm in size—as well as for vibratome sections of complex brains. Although our procedures have
been optimized for marine molluscs, they may easily be adapted for other (marine) organisms by the
creative neurobiologist.

Key words Immunocytochemistry, Fluorescence, Antibody staining, Whole-mount, Neurogenesis,

Free-floating vibratome sections, Complexity, Cephalopods, Brains, Lophotrochozoa

1 Introduction

The Mollusca constitutes one of the most diverse animal phyla,

comprising, to name but a few, small worm-shaped groups in the
millimeter range, valve-bearing polyplacophorans, bivalves, and
snails, as well as the cephalopods with their highly complex behav-
ior [1]. The wide variety in molluscan gross morphology is also
represented by the neuroanatomical features of its representatives.
As such, while basal taxa as well as small-sized individuals often
exhibit a weakly ganglionated central nervous system (CNS) that
mostly consists of interconnected longitudinal nerve cords,
“higher” gastropods and all cephalopods have sophisticated brains
that result from fusion of various ganglia [2–4]. Despite the long
history of gastropods and cephalopods as model systems in neuro-
physiological and neuroanatomical research [5, 6], surprisingly
little is known concerning the ontogeny of the nervous system
on the one hand and precise neurotransmitter and neuropeptide
distribution within the adult molluscan CNS on the other.

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_8, © Springer Science+Business Media, LLC 2014

117
118 Andreas Wanninger and Tim Wollesen

Fig. 1 Maximum intensity projection of a whole-mount staining of a veliger

larva of the marine gastropod Aplysia californica (dorsal view, anterior faces
downwards). The serotonergic nervous system is labeled in red and the muscu-
lature (by F-actin staining using phalloidin) in green. Scale bar: 100 μm

Since immunocytochemical staining in combination with confocal

microscopy and 3D reconstruction techniques, which allow
detailed high-throughput analyses, have now become routine lab
procedures, this picture is likely to change rapidly. Indeed, a
number of comparative studies on invertebrate—and hence also
molluscan—neurogenesis have become available in the past few
years and have injected important novel data into the discussion on
functional as well as evolutionary aspects concerning molluscan
nervous systems [7–15]. Herein, we present an easy-to-do and
reliable protocol for whole-mount studies of the molluscan ner-
vous system of minute specimens, including embryonic and larval
stages (see Fig. 1), as well as a convenient procedure for nervous
tissue staining of free-floating vibratome sections (see Fig. 2) that
allows for precise determination of immunoreactive substances in
complex brains of gastropods and cephalopods.

2 Materials

Use deionized H2O for the preparation of all solutions. Buffer

solutions may be stored at room temperature. Sera and antibodies
are stored undiluted at −20 °C and may be kept in the fridge for a
few days after reconstitution. Avoid repeated freezing and thawing.
 Molluscan Brain Development 119

Fig. 2 Maximum intensity projection of a 100 μm thick free-floating vibratome

section of the CNS of the squid Loligo vulgaris (dorsal view, anterior faces
downwards). FMRF amide-like immunoreactivity is labeled in red and perikarya
are blue (DAPI). Scale bar: 300 μm

2.1 Whole-Mount 1. Make solution for relaxation of live embryos and larvae. 7.14 %
Staining of Neural stock solution: 38.1 g MgCl2 × 6H2O, 100 mL water. Use
Tissue parts of it to make a 3.5–3.6 % MgCl2 solution.
2. 4 % Paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB)
for fixation: 33.48 g of Na2HPO4 × 2H2O, 7.93 g
NaH2PO4 × H2O, and 100 mL H2O. Alternatively: 26.7 g
Na2HPO4, 6.9 g NaH2PO4, and 100 mL H2O. Adjust pH to
7.3. Dissolve 4 g of PFA powder in 100 mL PB or make a
1:10 dilution of custom-made 4 % PFA solution in PB (see
Notes 1 and 2).
3. Washing solution: 0.1 % NaN3 to PB is recommended to avoid
bacterial or fungal growth (not necessary for the fixative).
4. Many molluscan larvae bear shells, which are best decalcified
after fixation in 0.05 M EGTA (dissolve 1.9 g EGTA powder
in 100 mL H2O) (see Note 3).
5. PBT: 0.2–10 % Triton X-100 (concentration depending largely
on specimen size and tissue), 0.1 M PB + 0.1 % NaN3.
6. Block-PBT solution: 6 % normal goat serum (NGS) in PBT.
7. Antibody solution: dilute all primary and secondary anti-
bodies at the respective working concentration in Block-PBT
(see Note 4).
8. Fluorescent counterstains: e.g., DAPI (for nuclei), phalloidin
or phallacidin (for F-actin including musculature).
120 Andreas Wanninger and Tim Wollesen

9. Clearing medium: Murray’s Clear (“BBA”), a 2:1 mixture of

benzyl benzoate:benzyl alcohol; prepare also a 1:1 dilution of
100 % EtOH and BBA as intermediate.
10. 75 % EtOH.
11. Mounting medium (e.g., Vectashield, Fluoromount G).

2.2 Staining 1. Vibratome.

of Vibratome Sections 2. Dissection forceps.
3. Paper towels.
4. Rubber mold.
5. Razor blade.
6. Instant glue.
7. Coated objective slides.
8. Cover slip.
9. Vibratome embedding medium: 24.2 g ovalbumin in 66 mL
H2O. Adjust the stirrer to minimize agglutination of the pow-
der and skimming of proteins. Close beaker with lid to avoid
desiccation. Heat up 25 mL H2O to 50 °C in a water bath.
Add 5.5 g of gelatin from porcine skin in little portions to
avoid agglutination of the powder. Close beaker with lid to
avoid desiccation. Slowly combine both solutions as soon as
gelatin and albumin are dissolved. Add 1.2 g NaN3 to avoid
fungal and/or bacterial growth. Embedding medium may be
stored up to ½ year at 4 °C. The entire process takes approxi-
mately 5 h.
10. Elvanol mounting medium [16]: 5 g Mowiol, 20 mL PB, pH
7.3. Prepare in a covered Erlenmeyer flask and stir for 16 h at
room temperature (RT). Add 10 mL of glycerol and stir again
for 16 h at RT. Centrifuge solution at either 4,000 rpm for
15 min, 3,000 rpm for 20 min, or 2,000 rpm for ½ h at RT.
Aliquot supernatant into 1.5 mL portions in Eppendorf tubes
and store at −20 °C.
11. Coating medium for objective slides: Thoroughly rinse
objective slides in vial containing a mix of 50 % acetone and
50 % pure EtOH. Dry objective slides with paper towels and
incubate at 60 °C for 24 h. Heat up 100 mL H2O to 50 °C
in a covered Erlenmeyer flask in a water bath.
Add 1 g gelatin and 0.1 g chrome alum (potassium chro-
mium sulfate) when temperature reaches 50 °C and stir for
15 min. Filter solution and transfer into a vial. Submerse objec-
tive slides thrice in coating medium for 3 s per step. Dry
coated objective slides at 40 °C in oven overnight.
 Molluscan Brain Development 121

3 Methods

All procedures are carried out at RT unless stated otherwise.

3.1 Fixation Identical for whole-mount and vibratome-sectioned samples:

1. Animals are carefully anesthetized by adding 7.14 % MgCl2 to
the seawater or by cooling the samples on ice prior to fixation,
especially important for larvae that may retract into their shells.
For marine species, eventually replace all seawater by 3.5 % MgCl2
to avoid precipitation of the PB salts once you add the fixative.
2. Specimens are fixed in 4 % PFA in 0.1 M PB, for 1–3 h at RT
(or overnight at 4 °C for larger specimens).
3. Samples are rinsed in 0.1 M PB thrice for 10 min and twice for
1 h at RT. They are stored in 0.1 M PB + 0.1 % NaN3 at 4 °C.
Alternatively, samples may be stepped into 75 % EtOH and
stored at −20 °C.

3.2 Whole-Mount 1. Decalcify in 0.05 M EGTA if necessary (samples are best

Immunocytochemistry watched under polarized light to check progress of
decalcification).
2. Transfer specimens into Block-PBT. Incubate for 2–24 h at 4 °C.
3. Incubate specimens overnight (18–24 h) in primary antibody(ies)
at 4 °C (see Notes 5 and 6).
4. Wash specimens for 2–24 h in Block-PBT at 4 °C with a minimum
of four changes.
5. Incubate specimens overnight (18–24 h) in secondary
antibody(ies) at 4 °C in the dark (see Note 7).
6. If nucleic acid counterstain is desired, add drops of DAPI
(working concentration: 1 %) at least 1 h prior to the subse-
quent washes.
7. If phalloidin staining is desired (e.g., for neuropil staining),
add 2.5 % phalloidin or phallacidin at least 1 h prior to the
subsequent washes.
8. Wash specimens for 2–24 h in PB without NaN3 at 4 °C with
a minimum of four changes in the dark (see Note 8).
9. If no clearing is required, mount specimens on objective slides,
preferably in a medium with antifade reagent (e.g., Vectashield,
Fluoromount G).
10. Larger and nontransparent specimens (e.g., yolky embryos and
larvae) may require clearing. Step samples into H2O and an
ascending EtOH series (5–10 min per step with three changes
in 100 % EtOH) after step 6. Then transfer them into a 1:1
dilution of EtOH:BBA (10 min) and finally into 100 % BBA.
Mount specimens on objective slides in this solution. OBS:
EtOH will abolish phalloidin and phallacidin staining. If any of
122 Andreas Wanninger and Tim Wollesen

Fig. 3 Hatchlings of the squid Loligo vulgaris embedded in gelatin-albumin

medium in a rubber mold for vibratome sectioning (left ). Trimmed block in pyra-
midal shape ready for sectioning (right )

these stains are used, dehydration should be carried out in an

isopropanol series with very short steps (30 s each), followed
by 1:1 isopropanol:BBA and 100 % BBA application.

3.3 Immunocyto- 1. Step specimens into 0.1 M PB if stored in 75 % EtOH.

chemistry on 2. Remove all hard parts from sample (e.g., beak, radula, eye lens,
Vibratome Sections or shell) manually or by decalcification (see Subheading 3.2,
step 1) since they might interfere with sectioning.
3. Excise portion of embedding medium, transfer to small
beaker, and incubate at 60 °C in oven for 15 min.
4. Dry sample gently with paper towels (do not overdry).
5. Pour embedding medium into rubber mold and quickly sub-
merse specimen (see Note 9).
6. Center specimen in rubber mold with forceps and ensure its
proper orientation for the sectioning process (see Note 10).
7. Store rubber mold at 4 °C for 15 min to solidify embedding
medium.
8. Store rubber mold in a 10 mL 37 % formaldehyde and 90 mL
0.1 M PB solution at 4 °C overnight. This step will interlink
proteins of the embedding medium and the specimen.
9. Transfer specimen in rubber mold into 0.1 M PB and store at
4 °C until further processing (see Note 11).
10. Carefully remove embedded sample block from rubber mold
(see Fig. 3).
 Molluscan Brain Development 123

11. Trim block with razor blade to pyramidal shape and cut off a
single corner. This will facilitate orientation of the sections
(see Fig. 3).
12. Fill tray with chilled 0.1 M PB.
13. Glue trimmed sample with instant glue on stage and let dry
for 10 s.
14. Submerse sample in 0.1 M PB and adjust thickness of sections
(see Note 12).
15. Adjust begin and end, frequency, and speed of sectioning
process.
16. Section trimmed block with a vibratome and collect all sections
of interest with forceps. Ensure that problematic hard parts
of the specimen are located at the distal end of each section.
By doing so, only a little portion of the specimen will rupture
in case that these parts are impossible to section.
17. Apply immunocytochemistry protocol to the sections of interest
as described above (see Subheading 3.2, steps 2–6).
18. Mount vibratome sections on coated objective slides.
19. Orient and dry sections with forceps and paper towels (do not
overdry).
20. Apply approximately 100 μl of mounting medium alongside
the objective slide. Gently cover sections with a cover slip and
avoid air bubbles.
21. Let sample harden in the dark at 4 °C for a few hours to sev-
eral days.

4 Notes

1. Either 5× or 10× PB stock solutions may be made. Against

common belief, it is not necessary to store these at 4 °C. If so,
the PB salts may form crystals, which have to be dissolved by
heating.
2. As with PB, the PFA fixative may be stored at RT. If stored at
4 °C, PFA may form a cloudy precipitation at the bottom of
the container. On the other hand, formaldehyde molecules
may cross-link after a while (days to weeks) and may hamper
rapid tissue penetration which may result in poor tissue preser-
vation. It is always best to use freshly made fixative.
3. PFA and EGTA powders dissolve poorly. Heat the solution but
avoid boiling. Use fume hood. Let solution cool down and
adjust pH to 7.3.
4. Concentrations of antibodies vary depending on the species
investigated. As a rule of thumb, one may start with a 1:400
124 Andreas Wanninger and Tim Wollesen

dilution for commercially available primary antibodies (e.g.,

anti-serotonin) and 1:100–1:300 for commercial secondaries,
but often concentrations as low as 1:2,000 may be feasible.
5. For double or multi-labeling with antibodies, the primary
antibodies must be raised in different hosts. Antibodies may be
applied as “cocktail” of the respective primaries and secondaries
in their respective working concentrations.
6. Perform negative controls by omitting the primary and
secondary antibodies (only one component at a time), respec-
tively, in independent experiments. This should prevent
labeling of neuronal elements and ensures that fluorescence
signal due to autofluorescence or unspecific binding is not mis-
interpreted as positive signal. In addition, a primary antibody
can be incubated in its respective antigen with conjugated
bovine serum albumin. This preabsorption experiment should
not reveal any immunoreactivity, suggesting that the given
antibody specifically binds to the respective antigen.
7. Since the secondary antibodies bear the signal-producing fluo-
rochrome, all steps from secondary antibody incubation
onwards should be carried out in the dark. Subsequent washes
and mounting is best done by using a stereomicroscope with
illumination from below, whereby the white light is shielded by
a red filter (simple plastic sheet is sufficient).
8. NaN3 quenches fluorescence. Therefore, it is vital to omit NaN3
in the last washes prior to mounting.
9. Compared to protocols that use, e.g., resins, embedding
specimens with gelatin-albumin medium is quick and straight-
forward. However, since this medium does not penetrate the
tissue, delicate internal structures may be damaged during the
sectioning or subsequent mounting process. Hence, it is
important to ensure proper exposure of the entire specimen to
the embedding medium.
10. The orientation of the block and the embedded specimen
should always be in accordance with its stability during the
sectioning process, i.e., one large side will be cut horizontally
(see Fig. 3).
11. Blocks should not be stored for longer than 2 days since the
embedding medium will soften.
12. The section thickness should be determined experimentally
when employing different antibodies or direct stains (e.g., phal-
loidin) on different species. The working distance and other
parameters of the microscope objectives should also be taken
into account. As a rough estimate, physical sections with a thick-
ness of 50–100 μm have been proven well-suited for cephalopod
and gastropod brains.
 Molluscan Brain Development 125

Acknowledgments

We thank Rudi Loesel (RWTH Aachen University) for advice

concerning vibratome sectioning. This work was supported by the
FWF (Austrian Science Fund) grant P24276-B22 to A.W.

References

1. Haszprunar G, Wanninger A (2012) Molluscs. 10. Braubach OR, Dickinson AJG, Evans CCE,
Curr Biol 13:R510–R514 Croll RP (2006) Neural control of the velum
2. Bullock TH, Horridge GA (1965) Structure and in larvae of the gastropod, Ilyanassa obsoleta.
function in the nervous systems of invertebrates. J Exp Biol 209:4676–4689
Freeman and Company, San Francisco, p 1611 11. Voronezhskaya EE, Nezlin LP, Odintsova NA,
3. Lin MF, Leise EM (1996) Gangliogenesis in Plummer JT, Croll RP (2008) Neuronal devel-
the prosobranch gastropod Ilyanassa obsoleta. opment in larval mussel Mytilus trossulus
J Comp Neurol 374:180–193 (Mollusca: Bivalvia). Zoomorphology 127:
97–110
4. Shigeno S, Tsuchiya K, Segawa S (2001)
Embryonic and paralarval development of the 12. Todt C, Wanninger A (2010) Of tests, trochs,
central nervous system of the loliginid squid shells, and spicules: development of the basal
Sepioteuthis lessoniana. J Comp Neurol 437: mollusk Wirenia argentea (Solenogastres) and
449–475 its bearing on the evolution of trochozoan lar-
val key features. Front Zool 7:6
5. Chase R (2002) Behavior & its neural control
13. Wollesen T, Loesel R, Wanninger A (2009)
in gastropod molluscs. Oxford University
Pygmy squids and giant brains: mapping the
Press, Oxford, p 314
complex cephalopod CNS by phalloidin stain-
6. Nixon M, Young JZ (2003) The brains and ing of vibratome sections and whole-mount
lives of cephalopods. Oxford University Press, preparations. J Neurosci Methods 179:63–67
Oxford, p 392 14. Wollesen T, Cummins SF, Degnan BM,
7. Marois R, Carew TJ (1997) Projection pat- Wanninger A (2010) FMRFamide gene and
terns and target tissues of the serotonergic cells peptide expression during central nervous sys-
in larval Aplysia californica. J Comp Neurol tem development of the cephalopod mollusk,
386:491–506 Idiosepius notoides. Evol Dev 12:113–130
8. Friedrich S, Wanninger A, Brückner M, 15. Wollesen T, Nishiguchi MK, Seixas P, Degnan
Haszprunar G (2002) Neurogenesis in the BM, Wanninger A (2012) The VD1/RPD2
mossy chiton, Mopalia muscosa (Gould) α1-neuropeptide is highly expressed in the
(Polyplacophora): evidence against molluscan brain of cephalopod mollusks. Cell Tissue Res
metamerism. J Morphol 253:109–117 348:439–452
9. Wanninger A, Haszprunar G (2003) The devel- 16. Rodriguez J, Deinhardt F (1960) Preparation
opment of the serotonergic and FMRF-amidergic of a semipermanent mounting medium for
nervous system in Antalis entalis (Mollusca, fluorescent antibody studies. Virology 12:
Scaphopoda). Zoomorphology 122:77–85 316–317
 Part IV

Xenopus Protocols
 Chapter 9

In Situ Hybridization and Immunostaining

of Xenopus Brain
Kai-li Liu, Xiu-mei Wang, Zi-long Li, Rong-qiao He, and Ying Liu

Abstract
The dynamic expression pattern analysis provides the primary information of gene function. Differences of
the RNA and/or protein location will provide valuable information for gene expression regulation.
Generally, in situ hybridization (ISH) and immunohistochemistry (IHC) are two main techniques to
visualize the locations of gene transcripts and protein products in situ, respectively. Here we describe the
protocol for the whole brain dissection, the in situ hybridization and immunostaining of the developing
Xenopus brain sections. Additionally, we point out the modification of in situ hybridization for microRNA
expression detection.

Key words Expression pattern, Brain dissection, Section, Immunostaining, In situ hybridization,
Xenopus, microRNA

1 Introduction

In situ hybridization and immunostaining are widely used practical

techniques to detect the locations of gene transcript and protein
product in situ, respectively. Moreover, the co-application of in situ
hybridization and immunostaining in one section is increasingly
employed for simultaneously observing location of the gene tran-
scripts of interest as well as the spatial distribution of another gene
product at the biochemical level.
For the gene expression analysis in early developing Xenopus
brain, usually earlier than stage 35/36 (st.35/36), in situ hybrid-
ization and immunostaining are often conducted with fixed whole-
mount embryo followed by sectioning for following analysis.
Whole-mount in situ hybridization (WISH) was adapted accord-
ing to Harland [1], with modifications [2–5]. Whole-mount
immunohistochemistry (WIHC) was carried out according to the
described [3, 4]. Whole-mount in situ hybridization for detecting
microRNA (miRNA) expression was performed according to our
previous work [6–9].

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_9, © Springer Science+Business Media, LLC 2014

129
130 Kai-li Liu et al.

However, as embryonic development proceeds, it gradually

encounters difficulty for probe and antibody to penetrate into the
brain and other tissues completely if the embryos or tissues are too
large. This low permeability can cause poor signal-to-background
ratio in ISH and IHC. Therefore, it is necessary to dissect brain
and other tissues at later embryonic (st.37/38 and later) and adult
stages. Here, we describe the method for dissecting and preparing
the brain for in situ hybridization and immunostaining. Particularly,
we will present the updated approach for examining miRNA
expression.

2 Materials

2.1 Tools for 1. A stereotype microscope is required for observation during

Xenopus Embryo embryo or tissue dissection. If the embryos were injected with
Manipulation, Brain gfp mRNA, a stereotype fluorescent microscope is needed.
Dissection 2. Forceps with blunt or sharp tips (e.g., Sigma Tweezers style #5)
are used to hold and dissect embryos or tissues, which will be
placed on agarose plate for operation.
3. Small sharp surgical scissors for eye operation are used to cut
skin and nerve.
4. Stainless-steel bone clamp is adequate for opening adult head
skull to obtain the intact brain and separate nerve fiber.
5. Superfrost® Plus slides or other Poly-L-Lysine coated slides.
6. Hairloop is used for moving embryos gently.
7. Whatman No. 2 filter paper.

2.2 Chemical Representative chemical reagents for ISH and IHC are listed as
Reagents follows:
1. HEPES, tricaine methanesulfonate (Sigma).
2. Formaldehyde.
3. Paraformaldehyde (PFA).
4. Tween-20.
5. Methanol.
6. Ethanol.
7. Sucrose.
8. OCT.
9. Paraffin.
10. Xylene.
11. Proteinase K.
12. BSA (BSA V; Sigma).
 In Situ and Immunostaining in Xenopus 131

13. Polyvinylpyrrolidone (PVP-40).

14. Ficoll 400 (phamacia).
15. Formamide (redistilled).
16. Torula RNA (Type IX, Sigma).
17. Heparin.
18. CHAPS.
19. Blocking reagent (Roche).
20. Lamb serum.
21. Anti-digoxigenin–alkaline-phosphatase antibody (Roche
cat. No. 11093274910).
22. Tetramisole (Sigma cat. No. L9756-5G).
23. NBT/BCIP (Roche cat. No. 11681451001).
24. BM purple (Roche cat. No. 1144207400).
25. Fast Red Tablets (Roche cat. No. 11496549001).
26. DAB substrate kit (zsbio).
27. AEC substrate kit (zsbio).
28. Hoechst 33258 (Sigma).

2.3 Solutions Solutions for in situ hybridization should be prepared using ultrapure
for In Situ water (18.2 MΩ at 25 °C). For IHC, double distilled water
Hybridization and (ddH2O) is feasible. Chemical reagents are the analysis pure:
Immunostaining 1. MMR: 0.1 M NaCl, 2.0 mM KCl, 1 mM MgSO4, 2 mM
CaCl2, 5 mM HEPES (pH 7.8), 0.1 mM EDTA. For 1 L 10×
MMR, add NaCl 58.44 g, KCl 1.49 g, MgCl2·6H2O 2.03 g,
CaCl2(anhyd) 2.22 g, and HEPES 11.92 g into 800 mL water,
mix dissolve and adjust pH to 7.5 with NaOH, and bring to
1 L with water. Autoclave and store at room temperature (RT).
2. MS222 anesthetic solution: MS222 0.2 mg/mL in 0.1×
MMR, pH 7.5–7.8.
3. 1× PBS: For 1 L 10× PBS, dissolve NaH2PO4·2H2O 2.89 g,
Na2HPO4·12H2O 26.73 g, and NaCl 102.2 g in 800 mL
ultrapure water, and then adjust pH to 7.4 with NaOH. Bring
volume to 1 L and autoclave. Store this solution at RT. Dilute
the stock solution with water before use and adjust pH if
necessary.
4. MEMFA: 0.1 M MOPS (pH 7.4), 2 mM EGTA, 1 mM
MgSO4, 3.7 % formaldehyde (see Note 1).
5. 4 % PFA/1× PBS: For 50 mL, dilute 10 mL 20% PFA with
5 mL 10× PBS with water prior to use (see Note 2).
6. 20 % Sucrose/1× PBS: For 50 mL, dissolve 10 g RNase-free
sucrose with 1× PBS, sterilize the solution through 0.45 μM
filter, and store at 4 °C.
132 Kai-li Liu et al.

7. 50× Denhardt’s solution: 1 % BSA, 1 % PVP-40, 1 % Ficoll

400. For 50 mL, dissolve 0.5 g BSA, 0.5 g PVP-40, and 0.5 g
Ficoll 400 in water, sterilize the solution through 0.22 μM filter,
and store at −20 °C.
8. 20× SSC: Dissolve 175.3 g NaCl and 88.2 g sodium citrate in
800 mL of water, adjust to pH 7.0 with 1 N HCl. Make up to
1,000 mL of water. Autoclave and store the solution at RT.
9. Hybridization buffer (for ordinary RNA probe): 50 % for-
mamide, 5× SSC, 0.1 % Tween-20, 100 μg/mL Heparin,
1 mg/mL torula RNA, 1× Denhardt’s solution, 0.1 %
Tween-20, 10 mM EDTA, 0.1 % CHAPS. Sterilize the solu-
tion with 0.22 μM filter and store it at −20 °C.
10. Hybridization buffer (for LNA probe for miRNA): 50 %
Formamide, 5× SSC, 0.1 % Tween-20, 50 μg/mL Heparin,
500 μg/mL yeast tRNA or torula RNA, 10 mM Citric acid
(pH 6.0). Filter the solution with 0.22 μM filter and store it at
−20 °C.
11. Washing solution (post hybridization): 50 % formamide, 2×
SSC, 0.1 % Tween-20 for ordinary probe; 2× SSC for LNA probe.
12. PBT: 1× PBS, 0.1 % Tween-20.
13. MABT: 100 mM maleic acid, 150 mM NaCl, 0.1 % Tween-20,
pH 7.5 (see Note 3).
14. Blocking buffer: 2 % blocking reagent, 20 % heat inactivated
sheep serum in MABT for ordinary probe; 1 % blocking
reagent, 1 % heat inactivated sheep serum in MABT for LNA
probe (see Note 4).
15. Alkaline-phosphatase buffer: 100 mM NaCl, 100 mM Tris–
HCl pH 9.5, 50 mM MgCl2, 0.1 % Tween-20, 2 mM tetrami-
sole (see Note 5).
16. NBT/BCIP staining solution: For every mL of alkaline-
phosphatase buffer, add 1 μL of NBT and 3.5 μL of BCIP
(see Note 6).
17. Fast Red staining solution: Add one Roche tablet in 2 mL
0.1 M Tris–HCl, pH 8.2, vortex 2 min and centrifuge 30 s to
discard the red precipitates.

3 Methods

3.1 Brain Dissection Induction of ovulation in females, in vitro fertilization, embryo

and Fixation culture, and staging are carried out as described [10, 11]:
1. For tadpoles (st.37/38 and later), transfer the embryos in a
small dish (diameter 35 mm), and make them anesthetized
by replacing the culture solution with MS222 solution (see
Note 7).
 In Situ and Immunostaining in Xenopus 133

2. When the embryos stop moving, place them into agarose plate
with precooling 1× PBS. Use the forceps with blunt tips to
hold the embryo at the trunk; use another forceps with sharp
tips to make a cut through the spinal cord at the dorsal truck
just behind the hindbrain. Tip up the skin of the front cut with
forceps and carefully peel off the skin of the dorsal head to
expose the brain. Then insert the tips of the forceps from the
cut to the bottom of the brain, and lift up the brain by a careful
forward movement of the forceps and shear the nerve bundles
with forceps tips or scissors. Transfer the dissected brain imme-
diately into fixation solution with a pipette or forceps.
3. For the brain dissection of developing frogs to adult, make the
frog anesthetized in the MS222 for around 20 min and then
place it on the ice. Cut off the spine and tear the skin from
foramen magnum toward the head with a sharp scissors and
then use bone clamp to open the skull and expose the brain
(see Note 8). The procedure of dissecting brain follows as above
(step 2). If this step takes long time, to prevent the decay of
brain tissue, add several drops of fixation solution onto the brain
to fix it in situ for 1 h before taking it out (see Note 9).
4. Fix the dissected brain in MEMFA or 4 % PFA at 4 °C over-
night (see Note 10).
5. For whole-mount in situ hybridization or paraffin section prepa-
ration. Then the brain could be dehydrated gradually by replacing
the fixation solution with 25 % ethanol/PBS, 50 % etha-
nol/50% PBS, 75% ethanol, 100% ethanol sequentially, 5–10 min
each; and stored in 100% ethanol at −20 °C (see Note 11).
For whole-mount immunohistochemistry, ethanol should be
replaced with methanol. For cryosection section preparation,
go on Subheading 3.2 directly without dehydration.

3.2 Cryosection 1. Pipette off the fixation solution and add precooling 1× PBS to
Preparation and wash for 3 times.
Pretreatment Prior to 2. Pipette off the PBS, and add 20 % sucrose/1× PBS to cryoprotect
Hybridization the brains at 4 °C for 4 h to overnight until the brains sink to
the bottom.
3. Label molds the name and the direction of specimens. Transfer
the brain(s) into the mold and remove the sucrose solution as
much as possible. Add appropriate volume (enough to immerse
the whole specimen) of OCT in the molds. Orientate the brains
in right directions with a long syringe needle, and then carefully
move the mold on the iron plate upon dry ice or into the 70 %
ethanol precooling at −80 °C. After the OCT has been com-
pletely frozen, store the block at −80 °C before sectioning.
4. Slightly tickle or press the mold to take out the frozen block,
mount the block on the specimen stand of cryostat at right
direction, and cut the edges to modify it into right shape before
134 Kai-li Liu et al.

sectioning. Cut 10–15-μm sections of the embedded brain at

−24 °C.
5. Collect sections one by one on ready-to-use Superfrost® Plus
slides or other Poly-L-Lysine coated slides, air-dry the sections,
and store the slides at −80 °C.
6. The stored sections should be defrosted at RT for at least
30 min prior to use for hybridization or immunostaining.

3.3 Paraffin Section 1. Dissolve paraffin at 60 °C before the day of embedding, and
Preparation and keep it at 60 °C.
Pretreatment Prior to 2. Wash the stored dehydrated brains with 100 % ethanol, and
Hybridization then transfer specimens to xylene by gradually replacing the
solution with 75 % ethanol/25 % xylene, 50 % ethanol/50 %
xylene, 25 % ethanol/75 % xylene, and 100 % xylene, and shake
at RT for 5–10 min each time.
3. Wash the brains with new 100 % xylene, pour the specimens
with xylene into a small beaker (specimens should be immersed
in xylene), and incubate at 60 °C for 20 min (see Note 12).
4. Embedding: Add the same volume of prewarmed paraffin,
incubate at 60 °C for 45 min, and then replace with 100 %
paraffin, incubated at 60 °C for 20 min. Replace with 100 %
new paraffin, incubated at 60 °C for 3 h; wash specimens with
100 % new paraffin, incubate at 60 °C for 20 min; replace with
new paraffin, pour specimens with paraffin into prewarmed
mold or plastic disc, and set the specimens at the right orienta-
tion and positions with prewarmed needle. Turn off the oven,
and leave the paraffin block with specimens to be solidified
slowly in the oven overnight (see Note 13).
5. Modify the embedded block and cut 5–10 μm sections in a
microtome. Transfer sections on the surface of 0.2 % ethanol
on slides prewarmed at 37 °C (see Note 14). Discard the solu-
tion when sections have been completely extended, dry the
sections at 37 °C overnight, and then store the slides at 4 °C.
6. For hybridization or immunostaining. Dewax the section in
100 % xylene by three washes, 10 min each time, 100 % etha-
nol wash twice, and then rehydrate the section by washes with
graded alcohol sequentially reverse to the dehydration, 2 min
each wash.

3.4 In Situ 1. Digoxigenin (DIG)- or fluorescein-labeled antisense RNA

Hybridization probes should be generated prior to hybridization step. The
of Sections labeled LNA probe for miRNA is available from EXIQON
Company.
3.4.1 Pretreatment
of Sections and 2. Wash the sections with PBT twice at RT, 2–5 min each wash.
Hybridization 3. For ordinary RNA probe, the sections could be applied for
hybridization (step 4) directly. For the LNA probe for miRNA,
 In Situ and Immunostaining in Xenopus 135

rinse sections in PBT with 1 μg/mL proteinase K at RT for

10 min, followed with two washes in PBT with 2 mg/mL
glycine, 5 min each wash, and three washes in PBT, 3 min each
wash. Then refix the sections in the 4 % PFA solution for
15 min, followed with three washes in PBT, 3 min each time
(see Note 15).
4. Place slides horizontally in humid hybridization chamber bot-
tomed with filter paper immersed in 1× SSC/50 % formamide.
Prewarm hybridization buffer with 1 μg/mL probe at
60–65 °C for ordinary probe or at the temperature suggested
by the manufacturer for LNA probe for 5–10 min. Add 120–
150 μL probe mix per slide, and carefully cover slide with a
coverslip. Seal the box with plastic film and transfer the box
into oven prewarmed at the right temperature to hybridize
overnight (see Note 16).

3.4.2 Washing Steps 1. Set a glass trough with washing solution stand in water bath.
and Antibody Visualization Warm up the solution and the glass trough to the hybridization
temperature. Add 100 mL washing solution to a glass trough
for ten slides.
2. Transfer slides to the glass trough with prewarmed washing
solution (see Note 17), and take away the coverslips after they
detach off the slides (usually in 1–2 min). Then, for ordinary
probe, make two washes at 65 °C with stirring or shaking,
30 min each; for LNA probe, make two washes in 2× SSC at
RT, 10 min each, followed with another wash in 0.2× SSC at
RT for 10 min.
3. Wash the slides 2 times in MABT at RT, 30 min each for ordinary
probe, or five times in PBT, 3 min each, for LNA probe.
4. Take the slide out one by one with forceps, wipe away the
excessive solution with filter paper, and then lay the slides on
the holder of the humid chamber with filter paper soaked in 1×
PBS or water. Add around 1mL blocking buffer per slide to
cover all the sections (without coverslip), and block for a half
to 1 h at RT.
5. Discard the blocking buffer, add 150 μL antibody mix (anti-
body diluted 1:2,000 to 1:8,000 in blocking buffer) per slide,
cover with a piece of parafilm cut as the same size of coverslip,
and incubate the slide in a humid chamber at RT overnight.
6. Discard the antibody solution, transfer the slides in a glass
trough with MABT, and make five washes on a shaker, 30 min
each wash at RT (see Note 18), followed with another five
washes in PBT, 5 min each at RT.
7. Staining: Wash the slides three times in alkaline-phosphatase
buffer for 5 min each at RT. Take out the slides and set them
in a humid chamber, add 150 μL NBT/BCIP or Fast Red
staining buffer per slide, cover with parafilm, and then develop
136 Kai-li Liu et al.

in the dark for usually 1 h to overnight with NBT/BCIP or

20 min to 6 h with Fast Red, depending on the abundance
of the RNA.
8. Stop the staining reaction by two washes in PBT, 1× PBS
sequentially. Then the sections could be mounted for observa-
tion (go step 13) and go on with subsequent immunostaining
(see Subheading 3.5) or the following counterstaining (e.g.,
Hoechst nuclear staining) for histological examination.
9. Hoechst staining for counterstaining of cell nuclei: Add several
drops of freshly prepared 1 μg/mL Hoechst 33258/1× PBS to
the slides, cover with a piece of parafilm to spread the staining
solution to all section, incubate for around 10 min at RT to
stop the staining reaction, and wash sections with 1× PBS
twice, 3 min each.
10. Mounting: For fluorescent staining, mount in a water-soluble,
non-fluorescing mounting medium, e.g., Aqua-Poly/Mount
(Polysciences) for observation. For chemical staining, dehy-
drate and clear sections by graded alcohols two washes in 100 %
ethanol and two washes in xylene, 2–3 min each, and then
mount in Histomount or Canadian gel solution (see Fig. 1).

3.5 Immunohisto- 1. For paraffin sections, antigen should be retrieved in 0.01 M

chemistry of Sections sodium citrate pH 6.0 at 80 °C for 20 min. For cryosections,
this step can be omitted. When HRP-conjugated second anti-
body is applied, 0.3 % H2O2 in 1× PBS treatment for 10–30 min
would be needed to inactivate the endogenous peroxidase.
2. Rinse sections with PBS for three times, then PBT for three
times, 5 min for each time.
3. Add around 1 mL 10 % goat serum/2 % BSA/PBT per slide to
cover all the sections, and block for 30 min.
4. Discard the blocking buffer, and overlay with 150 μL primary
antibody solution (antibody diluted in blocking solution as
recommended by the manufacturer), covered with coverslips
and incubated for 6 h at RT or overnight at 4 °C in a humidi-
fied chamber.
5. Rinse the sections with PBT for three times, 5 min each.
6. Re-block sections with 10 % goat serum/2 % BSA/PBT for
30 min. Then incubate sections with second antibody (dilute
antibody in the blocking solution as recommended by the manu-
facturer) in the wet chamber for 2 h at RT or overnight at 4 °C.
7. Wash sections with PBT for 3 times, with PBS for 3 times,
5 min.
8. If the second antibody is conjugated with fluorescent dye like
FITC and TRITC, sections can be directly mounted and
observed under fluorescent microscope. If the second antibody
is AP or HRP conjugated, the substrate such as BM purple/
 In Situ and Immunostaining in Xenopus 137

Fig. 1 In situ hybridization (ISH) of st.46 Xenopus brain sections for detecting tran-
scripts of zic2 (a, b), ath3/neurod4 (c), microRNA-181 (d). ISH for Zic2 was carried
out in cryosection, stained by using AP substrate Fast Red (red), which shows
expressed in the dorsal region of the developing brain (b), the nuclei of which were
counterstained with Hoechst (a). ISH for ath3/neurod4 and microRNA-181 were
conducted in paraffin sections and stained by applying AP substrate BM purple
(c, d). Ath3/neurod4 was expressed higher in ventricular zone (proximal) while lower
in marginal zone (distal) of developing brain/neural tube (c). microRNA-181 is rarely
detectable in developing brain while relatively highly expressed in the retina inner
nuclear layer (d). Br brain, Re retina, Vz ventricular zone, Mz marginal zone

FastRed for AP and DAB or AEC for HRP should be added on

the slides for staining as described in Subheading 3.4 or the kit
instruction (see Fig. 2). The post staining washes and mount
are as described above.

4 Notes

Generally, the volume of solution for fixation and washing should

be no less than 10 volumes of the specimens unless indicated.
During the process of ISH or IHC, tissue sections should not
be dry after initiating PBS washing, especially when the antibody
138 Kai-li Liu et al.

Fig. 2 Immunostaining of Xenopus brain sections for detecting protein locations of N-tubulin (a, b) and
phosphohistone-H3 (pH3) (c–e). IHC for n-tubulin was carried out in cryosection in (a, b), stained by TRITC-
conjugated secondary antibody showing red fluorescence and expressed in the whole developing brain/neural
tube and otic vesicle (b), the nuclei of which were counterstained with Hoechst (a). IHC for pH3 was conducted
in paraffin sections and stained with TRITC-conjugated secondary antibody (c, d) and HRP substrate AEC (e).
pH3 is specifically expressed in the proliferating cells in the midline of the developing brain as shown by the
arrows. Br brain, Ov otic vesicle

is incubated. Otherwise, the background will appear to be too

high. Sense probe control and blank control should be set in in situ
hybridization. For immunohistochemistry, the blank control with-
out adding the primary antibody is also required. The signal-to-
background ratio of in situ hybridization or immunohistochemistry
depends on the abundance of transcript/antigen of interest and
also the quality of the antibody. Therefore, the final quality of the
transcript/protein detection may be optimized in modulating the
corresponding step when necessary:
1. The solution with formaldehyde is all right when stored at
4 °C and used within a week at most of the cases. But it’s bet-
ter to make the 10-times solution without formaldehyde (10×
MEM) as stock solution and add fresh formaldehyde prior to
use. For 100 mL 10× MEM, add MOPS 20.93 g, 0.2 M EGTA
10 mL, and 1 M MgSO4 1 mL, adjust pH to 7.4 with NaOH,
and filter sterilized and aliquoted with 50 mL plastic tube,
stored at −20 °C.
 In Situ and Immunostaining in Xenopus 139

2. 20 % PFA is commercially available. For self-preparation of

100 mL solution, heat 80 mL distilled water at 60–65 °C.
Add 20 g paraformaldehyde and slowly add drops (<100 μL)
of 10 N NaOH, with stirring on hot plate in fume hood until
solution becomes clear; this should take no longer than
30–60 min! Add distilled water to 100 mL, and then filter with
Whatman No. 2 filter paper. Aliquot with 50 mL plastic tube,
stored at −20 °C.
3. 5× MAB stock solution could be prepared without Tween-20,
and NaOH pellets could be applied for the pH adjustment.
Autoclave and store the solution at RT. Dilute the stock and
add Tween-20 prior to use. It’ll be easier to get the right
volume of the dense Tween-20 by cutting the sharp pipette tip
a little.
4. The blocking solution could be prepared prior to use by dilut-
ing 5× MAB, 10 % blocking reagent, 100 % serum (inactivated
by incubated at 55 °C for 30 min), and Tween-20 with water.
The 10 % blocking reagent and inactivated serum should be
aliquoted and stored at −20 °C.
5. The solution should be freshly prepared prior to use by diluting
the 5 M NaCl, 1 M Tris–HCl pH 9.5, and 1 M MgCl2 with
water. Add Tween-20 and tetramisole and mix well prior to
use. The solution works well without tetramisole for in situ
hybridization for ordinary probe in our experiences.
6. The ready-to-use mix is also available, as Sigma cat. No. B1911;
BM purple could be apply for ordinary RNA probe staining.
7. The brain should be dissected within 10 min after the tadpoles
fall asleep, or the neurons/embryos would die.
8. Pay attention to avoid damaging the brain and bleeding. Be
patient and operate gently and carefully.
9. The in situ fixation should be conduit in hood. Add more
fixation solution from time to time to keep the brain immersed
in the fix.
10. For immunohistochemistry, Dent’s fixative (80 % metha-
nol/20 %DMSO) is another optional for some antibodies.
11. For small embryonic brain, the dehydration step can be simpli-
fied by directly replacing the fix with 100 % alcohol.
12. Extension of incubation time in xylene will lead to damage of
the tissue.
13. The whole embedding process should be conduit in oven.
The solid paraffin appeared should be remelted before going to
the next step.
14. Avoid the attachment of sections with the slide before the
sections have been completely extended.
140 Kai-li Liu et al.

15. The proteinase K and glycine solution should be freshly

prepared with 10 or 20 mg/mL stock; gentle mixing is
required to avoid denaturing the proteinase. Avoid shaking the
slides during the proteinase and postfixation treatment.
16. For LNA probe, hybridization temperature of Tm-22 °C is
suggested as described [12], but lower temperature could be
applied in our experience, especially for non-abundant miRNAs.
In addition, a 30 min prehybridization step is optional by add-
ing 200 μL hybridization buffer without probe for each 60 mm
slides; a piece of parafilm cut as the size of coverslip could
be applied to spread the hybridization buffer and cover the
section. But the parafilm “coverslip” could not be applied at
hybridization temperature higher than 38 °C. In addition, the
parafilm and hybridization buffer should be removed carefully
to avoid the damage of sections.
17. This step needs to be fast. Never let the slides cool down.
18. This step is optional for probes give low background or faint
signal.

Acknowledgments

We’re grateful to Federico Cremisi for sharing the protocol of

in situ hybridization with cryosections. This work was supported
by the NSFC No. 31271387.

References
1. Harland RM (1991) In situ hybridization: an 6. Decembrini S, Bressan D, Vignali R, Pitto L,
improved whole-mount method for Xenopus Mariotti S, Rainaldi G, Wang X, Evangelista M,
embryos. Methods Cell Biol 36:685–695 Barsacchi G, Cremisi F (2009) MicroRNAs cou-
2. Lupo G, Liu Y, Qiu R, Chandraratna RA, ple cell fate and developmental timing in retina.
Barsacchi G, He RQ, Harris WA (2005) Proc Natl Acad Sci USA 106(50):21179–21184
Dorsoventral patterning of the Xenopus eye: 7. Liu K, Liu Y, Mo W, Qiu R, Wang X, Wu JY,
a collaboration of Retinoid, Hedgehog and He R (2011) MiR-124 regulates early neuro-
FGF receptor signaling. Development 132: genesis in the optic vesicle and forebrain,
1737–1748 targeting NeuroD1. Nucleic Acids Res 39:
3. Liu Y, Lupo G, Marchitiello A, Gestri G, He 2869–2879
RQ, Banfi S, Barsacchi G (2001) Expression of 8. Qiu R, Liu Y, Wu JY, Liu K, Mo W, He R
the Xvax2 gene demarcates presumptive ventral (2009) Misexpression of miR-196a induces eye
telencephalon and specific visual structures in anomaly in Xenopus laevis. Brain Res Bull
Xenopus laevis. Mech Dev 100:115–118 79:26–31
4. Lan L, Liu M, Liu Y, He R (2007) Expression 9. Qiu R, Liu K, Liu Y, Mo W, Flynt AS, Patton
and antibody preparation of POU transcription JG, Kar A, Wu JY, He R (2009) The role of
factor qBrn-1. Protein Pept Lett 14:7–11 miR-124a in early development of the Xenopus
5. Lan L, Liu M, Liu Y, Zhang W, Xue J, Xue Z, eye. Mech Dev 126:804–816
He R (2006) Expression of qBrn-1, a new 10. Sive HL, Grainger RM, Richard RM (2002)
member of the POU gene family, in the early Early development of Xenopus laevis: a labora-
developing nervous system and embryonic tory manual. Cold Spring Harbor Laboratory
kidney. Dev Dyn 235:1107–1114 Press, Cold Spring Harbor, NY
 In Situ and Immunostaining in Xenopus 141

11. Nieuwkoop PD, Faber J (1994) Normal table 12. Kloosterman WP, Wienholds E, de Bruijn E,
of Xenopus laevis (Daudin): a systematical and Kauppinen S, Plasterk RH (2006) In situ detec-
chronological survey of the development from tion of miRNAs in animal embryos using LNA-
the fertilized egg till the end of metamorphosis. modified oligonucleotide probes. Nat Methods
Garland, New York 3:27–29
 Chapter 10

Microinjection Manipulations in the Elucidation

of Xenopus Brain Development
Cristine Smoczer, Lara Hooker, Saqib S. Sachani,
and Michael J. Crawford

Abstract
Microinjection has a long and distinguished history in Xenopus and has been used to introduce a surprisingly
diverse array of agents into embryos by both intra- and intercellular means. In addition to nuclei, investiga-
tors have variously injected peptides, antibodies, biologically active chemicals, lineage markers, mRNA,
DNA, morpholinos, and enzymes. While enumerating many of the different microinjection approaches
that can be taken, we will focus upon the mechanical operations and options available to introduce mRNA,
DNA, and morpholinos intracellularly into early stage embryos for the study of neurogenesis.

Key words Microinjection, Morpholino, mRNA, Lineage marker, Xenopus, Over-expression,

Knockdown, Transgenic, Mutant

1 Introduction

Microinjection approaches have facilitated numerous studies on

neurogenesis; however, fine injection needles drawn from heated
glass capillaries have been used for many other purposes, especially
in studies employing amphibian eggs and embryos. For example,
the first nuclear transfer and cloning exercises were accomplished
in frog [1, 2]. Injection rigs can be used to introduce materials
both inter- and intracellularly, and the gamut of injectables runs
from nuclei, to peptides, antibodies, biologically active chemicals,
lineage markers, mRNA, DNA constructs, morpholinos, and
enzymes. In Xenopus laevis, ovulation and insemination are easily
induced, and the large size and accessibility of eggs and embryos
makes the species ideal for many studies on early neural patterning.
Moreover, cheap and effective transgenic approaches have been
developed; however, the lack of suitable stem cell lines still precludes
knockouts by homologous recombination. The pseudotetraploid
nature of the X. laevis genome also imposes limits upon the scope
experiments.

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_10, © Springer Science+Business Media, LLC 2014

143
144 Cristine Smoczer et al.

1.1 Intercellular The object of intercellular injection has been to introduce agents
Injection either between germinal layers or into the blastocoel.

1.1.1 Localized Agents In studies of neural activation and posterior transformation,

retinoic acid was co-injected with the lipophilic marker DiI
(1,1′-Dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine iodide)
between deep and surface ectoderm in early gastrula [3]. The
approach permits the introduction of a teratogen that remains rela-
tively localized over the course of neurogenesis. Moreover, the intro-
duction of many varieties of marker (such as Invitrogen’s CellTracker
and MitoTracker or BrdU; see ref. [4] for review) offers a plethora of
opportunities for consideration in lineage marker studies. Finally,
while there are not many examples of the virally mediated intro-
duction of genes to amphibians, vaccinia was successfully injected
and used to ectopically introduce Sonic hedgehog into axolotl tissues,
and this approach could be considered in Xenopus [5].

1.1.2 Blastocoel Peptides have been injected into the blastocoel either to activate
[6–8] or to competitively block cell surface or matrix-mediated
pathways [8].
A similar approach that used antibodies as a blocker has also
been successfully used [9]. Other, more complex mixtures contain-
ing biological agents have been tested in this manner, for example,
the contents of germinal vesicles [10].

1.2 Intracellular Intracellular injection, particularly if accompanied by inclusion of a

Injection fluorescent lineage marker, provides a useful entrée to study gene
perturbation effects and cell lineages. Considerable versatility is
introduced by the availability of different developmental stages
for injection. For example, nuclear and DNA injections typically
take place at the one-cell stage; however, individual blastomeres
can be injected when effects upon restricted lineages are desired
[11]. Injection at the 2-cell stage permits gene perturbation on one
side of the future embryo—the other side serves as contralateral
control [12].
Intracellular approaches can facilitate subsequent analysis
through the grafting of heterologously treated tissues or through
the production of embryos that serve as donors for animal cap
explants [13].

1.2.1 Exogenous DNA Several decades ago, linearized DNA was microinjected into fertil-
ized Xenopus eggs to study tissue-specific expression of the intro-
duced gene [14]. Although a majority of the transgenic embryos
were highly mosaic, a very small percentage survived to adulthood
and transmitted the transgene to the next generations [15, 16].

1.2.2 REMI (Restriction Many years passed following pioneering nuclear transfer studies
Enzyme-Mediated [1, 17] before nuclear transplantation coincident with the introduction
Integration) of plasmid evolved to facilitate the generation of transgenic frogs [18].
 Microinjection Manipulations in the Elucidation… 145

The latter technique was developed to address the high mosaicism

that accompanies microinjection solely of linearized DNA. In
order to control mosaicism, an in vitro fertilization procedure facil-
itates transgene incorporation into decondensing pronuclei. Sperm
nuclei, restriction endonuclease, and linearized DNA were
co-injected into eggs by means of an infusion pump (sperm nuclei
are too big to pass through the narrow needle used in most micro-
injections) [18, 19]. Despite the reduced incidence of mosaicism,
there are major drawbacks to this approach: the large bore needle
that is required to introduce nuclei damages and impairs the viabil-
ity of injected embryos; the concomitant restriction enzyme-medi-
ated integration (REMI) does not always occur immediately—mosaic
embryos still result; concentrations of restriction endonuclease
have to be carefully titered to avoid genome fragmentation and
aneuploidy; and finally, integration copy number and variable pen-
etrance/expressivity can complicate phenotype interpretation
[18, 20, 21]. The long generation time of X. laevis renders its use
as a genetic model problematic, and typically, numerous transgen-
ics must be assessed in an experiment to control for integration
site-specific variations. With a shorter generation time and diploid
genome, Xenopus (Silurana) tropicalis is better suited for experi-
ments requiring founder lines and genetic crosses.

1.2.3 I-Sce I An approach utilizing a rare-cutting meganuclease is based on the

Meganuclease presence of 18 bp restriction sites for I-Sce I within the Xenopus
genome. The transgene construct is flanked by I-Sce I restriction
sites. Following digestion with the meganuclease, the mixture
is microinjected close to the forming nucleus of fertilized eggs.
This procedure is simple and effective and yields high survival
rates among transgenic embryos that can later transmit the trans-
gene to their offspring [22–24]. The disadvantages of the method
reflects the tendency for random and multiple insertions of the
transgene both singly and as concatemers, as well as a degree of
mosaicism [24].

1.2.4 Phi-C31 Integrase Integrase-mediated transgene insertion relies on Phi-C31-mediated

recombination between attB sites flanking the transgene construct
and pseudo-attP sites present in the amphibian genome. Exogenous
DNA and the mRNA encoding Phi-C31 integrase are co-injected
into fertilized eggs. This procedure results in relatively efficient
single-copy transgene incorporation at specific sites [25, 26].
Efficacy can be improved by deploying a vector that brackets
the transgene with chicken β-globin 5′-HS4 insulators [25, 26].
The same insulators have been used to flank I-Sce I sites and thereby
improve transgene activity in that system as well [27]. The disad-
vantage of the approach is that the vector becomes somewhat
large, and this, in conjunction with the repetitive nature of the
insulators, renders the construct prone to recombination during
amplification.
146 Cristine Smoczer et al.

1.2.5 Lineage Markers Several strategies are available to deploy fluoresceinated lineage
markers. Intracellular injection of fluorescently labelled hydrophilic
dextrans has a well-tried history [28, 29]. An advantage that dex-
trans confer is that they tend to diffuse relatively well throughout
the injected cell, and yet they remain bounded by their large
molecular mass to that cell and its progeny. Tissues derived from
injected embryos can be combined with wild type or differently
injected embryos via heterotypic grafting [12, 30]. Alternatively,
the mRNA for GFP [31] or fluorescent fusion markers [32] can be
deployed either by co-injection or as bi-cistronic transcripts derived
from IRES vectors.

1.2.6 Biologically Transcripts are routinely injected into Xenopus embryos to perturb
Active In Vitro neurological development. Ectopic gain-of-function experiments
Synthesized mRNAs are straightforward [33]. Alternatively, chimeric constructs that
deploy fused activator (VP16) or repressor (engrailed) domains are
used to deliver phenotypes that can reveal whether a candidate
gene exerts an activating or repressive effect [34]. Mutations can
be introduced to alter the behavior or to define the role of encoded
sub-domains. For example, a point mutation introduced to Mix.1
impaired the DNA binding functions of its encoded homeodomain
[34]. Finally, mRNA “rescue” experiments can be mounted to
assess hierarchical relationships between genes. If morpholino-
mediated knockdowns or chimeric activator/repressor transcripts are
introduced, then the capacity of potential downstream signalling
partners can be assessed by co-injecting them in complementation
studies. All of these approaches suffer some disadvantages: transcripts
will be nonspecifically distributed throughout all the progeny of an
injected blastomere; ectopic mRNAs have a limited lifespan—their
sensitivity to degradation limits their utility to early developmental
stages; and different species of transcript might vary with regard to
longevity and distribution pattern in a context-specific manner.

1.2.7 Morpholino Morpholinos can be injected to abrogate transcript translation

Antisense Oligonucleotides [35] or, at least in zebrafish, to specifically inhibit the generation of
a particular alternatively spliced isoform [36]. Their use is described
elsewhere in this volume.

2 Materials

A good source for general recipes and handling instructions for

Xenopus and related reagents is available and recommended [37].

2.1 Frogs 1. Xenopus laevis males and females (Xenopus I, Michigan, USA).
and Embryos Wild caught specimens preferred.
2. Human chorionic gonadotropin (HCG) (AkzoNobel or Sigma).
Prepare a 1,000 U/ml solution in the manufacturer provided buffer.
Unused remainder can be stored at −20 °C for several weeks.
 Microinjection Manipulations in the Elucidation… 147

3. Single-use tuberculin syringe, 1 cc, 26-gauge needle.

4. Tricaine methanesulfonate MS-222 (Sigma). Prepare 1 l of a
0.2 % solution buffered to pH 7.2 with sodium bicarbonate
fresh the morning of use in water and store at 4 °C.
5. Scissors, forceps, small finger bowl or 80 ml beaker, and Petri
dishes.
6. A means to incubate embryos at 12–13 °C.
7. Petri dishes, capillary tubes (not heparin coated or chemically
treated), and water-resistant tape (masking tape or the colored
paper tape normally used to label lab ware is fine).
8. Pasteur pipettes with small rubber bulb for the transfer of
embryos. These can be cut and flame polished to leave an
opening of 1.5–2 mm.

2.2 Solutions Stock Part A (20×): 1.16 M NaCl, 13.4 mM KCl, 16.6 mM
MgSO4, 6.7 mM Ca(NO3)2, pH to 7.4.
2.2.1 Steinberg’s
Solution Stock Part B (20×): 92.5 mM Tris base, 0.08 N HCl.
Prepare both parts using distilled water, autoclave, and store at
room temperature.
Dilutions of stock solution for typically used Steinberg’s solutions:

200 % 100 % 80 % 20 %
Part A per liter 100 ml 50 ml 40 ml 10 ml
Part B per liter 100 ml 50 ml 40 ml 10 ml

2.2.2 MBS (Modified Part A. 10× MBS: 880 mM NaCl, 25 mM NaHCO3, 10 mM KCl,
Barth’s Saline, Stored in 50 mM HEPES of pH 7.8, 10 mM MgSO4. Adjust pH to 7.8 with
Two Parts) NaOH, top up with distilled water, and then autoclave. Store
at room temperature and use within the month.
Part B. 0.1 M CaCl2—into 7 ml aliquots. Store frozen.
On the day of experiments, prepare 1× MBS by adding 100 ml of
10× MBS and 7 ml 0.1 M CaCl2 to distilled water—volume
adjusted to 1 l.

2.2.3 L-Cysteine 2 % L-cysteine hydrochloride (Sigma or ICN) is dissolved in H2O and

adjusted to pH 8.0 with NaOH. Prepare fresh on the morning of use.

2.2.4 Ficoll Ficoll 400 is dissolved in 0.3× MBS in a ratio 2–5 % weight to
volume, depending upon the needle size. Typically, RNA and
morpholino injections deploy very fine needles and embryos only
require 2–3 % Ficoll for structural support during injection and
recovery. Larger needles required for nuclear transfer will require
5 % Ficoll. Prepare the morning of use and refrigerate to 12–13 °C.
148 Cristine Smoczer et al.

2.3 Glass Needles 1. Microinjection needles. Drawn from borosilicate glass capillaries
(1 mm outside diameter and 0.58 mm inside diameter—
Harvard Apparatus Ltd., Edenbridge, UK, or Drummond
3-000-203-G/X) using a microneedle puller (such as model
PC-10, Narishige, Tokyo, Japan) (see Note 1).
2. 1 cc inoculation (tuberculin) syringe with a 26-gauge needle.
3. Mineral oil (Eppendorf).
4. Micromanipulator.
5. Positive displacement injector (e.g., Drummond Nanoject
II—see Note 2).
6. Stereomicroscope.
7. Paraplast film or glass slide.

3 Methods

3.1 Preparation of 1. Select two females exhibiting prominent red cloacae. Inject
Frogs and Fertilization each with 0.75 cc HCG and incubate overnight (if frogs are to
of Eggs [38] be kept at 18 °C, inject at 6:00 pm in the afternoon the day
before needed or at 11:00 if females are to be held at room
temperature 22–23 °C).
2. Next morning, check to see if the females are shedding eggs.
If they are, then anesthetize the male in tricaine methanesulfo-
nate (MS-222) and sacrifice by cervical section or decapitation.
Remove the testes to 200 % Steinberg’s solution and keep cool
either on ice or in a refrigerator.
3. Macerate one quarter testes in 2 ml of 80 % Steinberg’s solution
and draw up and down in 1 cc syringe (no needle), to break the
tissue up further and to evenly distribute the sperm.
4. Strip eggs from the female into a Petri dish and add 0.5 cc
sperm suspension. Leave at room temperature for 10 min.
5. After 10 min has elapsed, flood the fertilized embryos with
0.1× MBS.
6. In the meantime, prepare fresh 2 % L-cysteine solution.
7. Successfully fertilized embryos rotate so that their dark animal
pole faces up, and the cortex visibly contracts.
8. Once rotation has occurred, remove the embryos to a vessel
for de-jellying (finger bowl or 80 ml beaker). Remove as much
of the 0.1× MBS as easily possible, and flood with L-cysteine
solution. Swirl them very gently periodically. After 2–4 min, the
jelly should have dissolved, and the eggs will pack close together
in the bottom of the beaker when it is tilted.
9. Immediately decant the bulk of the solution being careful
never to fully expose the embryos to air, and gently replace
 Microinjection Manipulations in the Elucidation… 149

with 0.1× MBS to wash off the L-cysteine. It is important to

remove all jelly, but to minimize the duration of contact with
L-cysteine.

10. Repeat this 9 more times taking great care not to violently
agitate the embryos.
11. Remove to a Petri dish of 0.1× MBS, and cool the embryos
to 14 °C.
12. For subsequent manipulation/transfers of embryos, we use a
Pasteur pipette that has been cut and flame polished at the end
leaving an opening approximately 1.5–2 mm wide (wider that
an embryo by a generous margin).
Good quality and undamaged embryos will look round and
firm and will not adhere to the plastic (see Note 3).

3.2 Preparation 1. We recommend prepared needles be placed in a large Petri dish

of the Injection and baked to degrade any RNases. Only experience will help
Apparatus you to determine what parameters of forge or flame will work
best to produce an optimal needle. You need needles that can
easily discharge a fixed volume of injectable fluid, but that will
not leave a wound large enough to cause cytoplasmic leaking
from punctured embryos.
2. Under a stereomicroscope, break off the tip of the needle to
open it. Prepared needles will be sharp enough that the ends
will be discernable only under a stereomicroscope.
3. Using a 1 cc syringe equipped with a 26-gauge needle, backfill
the glass microinjection needle with mineral oil. A tiny droplet
of oil should become barely discernable at the tip. If air bub-
bles are introduced, begin again with a fresh needle.
4. Fix the microinjector into an orientation that is angled
approximately 45° downward. It is good practice to clean the
plunger of the microinjector with 70 % ethanol prior to use.
5. Taking great care to avoid the introduction of air bubbles, load
the needle onto the microinjector.
6. Empty the microinjector of oil to the extent that the positive
displacement plunger has gone as far as it can without reaching
the constricted end of the glass needle.
7. Under the stereomicroscope, fill the needle from a droplet of
injectable substance that is placed center field upon paraplast
or a glass slide. Exercise caution to avoid sucking up air
bubbles.
8. For mRNA injections, mRNA is diluted in RNase-free water
which is drawn up into the needle from glass slides which have
been washed with 70 % ethanol, baked (65–95 °C for 2 h),
cooled to room temperature, and wiped with RNase AWAY®
solution (Life Technologies).
150 Cristine Smoczer et al.

3.3 Preparation Morpholinos can be obtained through Gene Tools LLC, and their
of Injectables administration and qualities are described elsewhere in this
Morpholinos volume.

3.4 Preparation 1. Prepare a workspace that is essentially RNase-free. This can be

of Injectables done cheaply by wiping surfaces and pipettors with a solution
Morpholinos mRNA of 0.1 N NaOH and 0.1 % SDS.
2. mRNA is prepared by in vitro transcription from linearized
template. Template must be of high quality and RNase-free.
We have found Ambion’s mMESSAGE mMACHINE (Life
Technologies–Invitrogen) to provide good quality transcript.
Capped transcript is diluted to the desired concentration in
freshly distilled or nano-pure RNase-free water (older aliquots
of water absorb CO2 and acidify with time. See Note 4).
3. The quality and size of transcript should be checked on a fresh
gel (1 % agarose in 1× TAE) that is prepared using RNase-free
gel casting trays, gel boxes, loading tips, running and loading
buffers, etc.
4. We recommend aliquoting transcript in known quantities
suitable for one-time use—repeated freeze–thaw cycles
damage transcript integrity, and repeated handling increases
the risk of RNase contamination. Store the transcripts in ali-
quots at −80 °C.
5. Fluoresceinated dextran is a poor choice of marker to monitor
the efficacy of mRNA injection since it is hard to acquire the
fluor free of RNase activity. We typically co-inject with tran-
script for GFP and use this component both as a vital marker
(under UV illumination) and to balance total mRNA concen-
trations to match control injected quantities.
6. Normally, a dilution series is prepared in order to develop a
dose–response curve.

3.5 Injection One of the challenges in later assessment is to separate un-injected

of Embryos or poorly injected embryos from viable experimental specimens.
Vital markers facilitate assessment of injection efficacy at relatively
early stages (e.g., during early neurulation), when the label remains
intense. It is advantageous to ensure that a fluorescent marker (flu-
oresceinated dextran), tag (e.g., fluorescein- or lissamine-tagged
morpholino), or mRNA for GFP (green fluorescent protein) is
part of the injection strategy. A directly tagged morpholino or
bi-cistronic marker transcript permits more direct estimations of
intensity and distribution of product within an embryo. Co-injected
markers like dextran or mRNA for GFP help by inference only.
Experiments usually start with a characterization of the effects of
an injected substance following introduction at the one-cell stage.
Once a dose–response curve has been established, injection into
 Microinjection Manipulations in the Elucidation… 151

specific blastomeres can ensue. Generally speaking, when injected

embryos are going to be used to produce explants for culture or
grafting, they are injected at the one-cell stage:
1. Prepare a few Petri dishes as platforms for the injection proce-
dure by placing a capillary tube (not heparin coated or chemi-
cally treated) so that it aligns down the center of the plate. Fix
it in place at both ends using small strips of tape.
2. Place a prepared Petri injection platform under the stereomi-
croscope and extrude a few milliliters of Ficoll solution down
one side of the centered capillary tube.
3. Gently drop a line of de-jellied embryos into the Ficoll. Let them
sit for a minute or 2 (see Note 5).
4. Gently remove as much of the Ficoll solution as possible
without disrupting the aligned embryos. Use a tissue on the
opposite side of the capillary to wick away much of the remain-
ing Ficoll leaving the embryos damp, but not submerged.
Surface tension will keep them in place for injection.
5. Lower the injection needle and pierce each embryo at a rela-
tively steep angle near, but not into, the area likely to house the
nucleus. The sperm entry site is usually visible as a slight differ-
ence in pigmentation: an ideal injection latitude is between the
animal pole and this discoloration point. Keep the injection
movement fluid and smooth. Piercing can be quick, but with-
drawal should be a little slower, and the whole process for each
injection will lower to just over a second with practice.
6. When all embryos in the line have been injected, raise the glass
injection needle, and reserve the injection rig somewhere safe
and convenient while the Petri dish is prepared for removal.
7. Gently flood the injected embryos with Ficoll solution by gently
adding the Ficoll from the capillary side of the line or from
either end. The object is to submerge them without floating
too many since this would introduce stress via surface tension.
Floaters can usually be sunk by gently adding Ficoll solution
dropwise overtop of them.
8. Remove the embryos to a cool incubator (12–13 °C) for at
least 1 h.
9. If you plan to inject different concentrations of substance in a
single sitting, start with the lowest concentration and then
reuse the needle. This can be done between specimen lots by
expelling all aqueous fluid and refilling with a higher concen-
tration of the same reagent.
10. Preferably after an hour or two, but certainly before gastrulation
(long-term storage in Ficoll inhibits gastrulation movements),
the Ficoll solution should be replaced with 0.1× MBS. Replace
the solutions by serially and gently pouring 0.1× MBS onto the
152 Cristine Smoczer et al.

plate and then decanting the solution until the Ficoll has been
largely replaced. Embryos may float, but can be immersed
again by dripping fluid gently overtop.
11. At this point survey the embryos under a stereomicroscope
and remove all dead or nondividing embryos. Remove the
surviving embryos to a fresh dish of 0.1× MBS. Continue to
incubate at 12–13 °C.
12. Each day, be sure to change the 0.1× MBS and to remove dead
embryos before they render the media toxic.
13. If a fluorescent marker has been employed, use a fluorescence
stereomicroscope to remove unsuccessfully injected embryos
from the plates. Incubate the remainder to the developmental
stage desired.

4 Notes

1. Needles can also be pulled manually over a small flame, but this
takes practice and skill. Affix a modified large gauge needle to
a stand, and use it as a miniature gas burner to heat a small
region of a glass capillary. Be careful to keep flammables away
from the metal that can become very hot. Roll the capillary
between your fingers above the flame until the glass become
soft and ductile, and then pull the two halves apart while lifting
the glass away from the flame. This takes caution, care, and
practice.
2. We have previously employed a homemade air pressure
displacement rig to inject embryos. This is accomplished by
attaching a “y” of tubing to a source of pressurized air. One
arm of the “y” is partially clamped, and the other goes to the
glass needle held in a micromanipulator. A glass needle is back-
filled with injectable fluid and then plugged into one end of
the air delivery tube. The clamp on the other arm is adjusted
to allow the bulk of air to escape while permitting the fluid to
remain in the needle. When delivery of injection fluid is desired,
the partially clamped tube is pinched by folding to prevent any
air from bleeding out, and this in turn raises pressure sufficient
to expel injection fluid. Injection volumes are calibrated by
time and diameter, and recalibration has to be repeated peri-
odically. The process is slow and cumbersome, but surprisingly
accurate and reproducible.
3. Flattened or floppy eggs/embryos or sticking to the bottom of
the Petri dish reflect damaged materials. Possible causes include
poor fertilization (check sperm motility and Steinberg’s solu-
tion pH), poor egg quality from the female, too much time in
L-cysteine, incomplete washing in 0.1× MBS after de-jellying,
 Microinjection Manipulations in the Elucidation… 153

or rough handling via too much agitation or sharp transfer

pipettes—this damages the fertilization membrane and permits
direct contact of the egg/embryo surface with the Petri dish.
4. The use of DEPC (diethylpyrocarbonate) to inactivate RNase
is common in many RNA-based experiments. Unfortunately, it
may be toxic to embryos, and we have found that it is unneces-
sary to use DEPC-treated water as a carrier for injections.
5. If embryos leak substantially after injection, there may be
several causes: the eggs were not exposed to Ficoll long enough
to lend support; the needle is too thick; or finally, the needle
injection path was not stable and tore the membrane too much
either during penetration or, more likely, during withdrawal.

Acknowledgment

M.J.C. is supported by the Natural Sciences and Engineering

Research Council (NSERC) of Canada Grant #203549.

References
1. Briggs R, King TJ (1952) Transplantation of embryos. Proc Natl Acad Sci USA 91(8):
living nuclei from blastula cells into enucleated 3049–3053
frogs’ eggs. Proc Natl Acad Sci USA 38(5): 9. Purcell L, Gruia-Gray J, Scanga S, Ringuette
455–463 M (1993) Developmental anomalies of
2. Gurdon JB (2006) Nuclear transplantation in Xenopus embryos following microinjection
Xenopus. Methods Mol Biol 325:1–9 of SPARC antibodies. J Exp Zool 265(2):
3. Drysdale TA, Crawford MJ (1994) Effects of 153–164
localized application of retinoic acid on 10. Kao KR, Elinson RP (1985) Alteration of the
Xenopus laevis development. Dev Biol 162(2): anterior-posterior embryonic axis: the pattern
394–401 of gastrulation in macrocephalic frog embryos.
4. Dolbeare F (1995) Bromodeoxyuridine: a Dev Biol 107(1):239–251
diagnostic tool in biology and medicine, part II: 11. Shi J, Severson C, Yang J, Wedlich D,
oncology, chemotherapy and carcinogenesis. Klymkowsky MW (2011) Snail2 controls
Histochem J 27(12):923–964 mesodermal BMP/Wnt induction of neural
5. Roy S, Gardiner DM, Bryant SV (2000) crest. Development 138(15):3135–3145
Vaccinia as a tool for functional analysis in 12. Khosrowshahian F, Wolanski M, Chang WY,
regenerating limbs: ectopic expression of Shh. Fujiki K, Jacobs L, Crawford MJ (2005) Lens
Dev Biol 218(2):199–205 and retina formation require expression of
6. Otto C, Schutz G, Niehrs C, Glinka A (2000) Pitx3 in Xenopus pre-lens ectoderm. Dev Dyn
Dissecting GHRH- and pituitary adenylate 234(3):577–589
cyclase activating polypeptide-mediated sig- 13. Zuber ME, Gestri G, Viczian AS, Barsacchi G,
nalling in Xenopus. Mech Dev 94(1–2): Harris WA (2003) Specification of the verte-
111–116 brate eye by a network of eye field transcription
7. Cooke J, Smith JC (1989) Gastrulation and factors. Development 130(21):5155–5167
larval pattern in Xenopus after blastocoelic 14. Rusconi S, Schaffner W (1981) Transformation
injection of a Xenopus-derived inducing factor: of frog embryos with a rabbit beta-globin gene.
experiments testing models for the normal Proc Natl Acad Sci USA 78(8):5051–5055
organization of mesoderm. Dev Biol 131(2): 15. Etkin LD, Pearman B, Ansah-Yiadom R (1987)
383–400 Replication of injected DNA templates in
8. Saint-Jeannet JP, Dawid IB (1994) Vertical Xenopus embryos. Exp Cell Res 169(2):
versus planar neural induction in Rana pipiens 468–477
154 Cristine Smoczer et al.

16. Etkin LD, Pearman B (1987) Distribution, 28. Wetts R, Fraser SE (1991) Microinjection of
expression and germ line transmission of exog- fluorescent tracers to study neural cell lineages.
enous DNA sequences following microinjec- Development Suppl 2:1–8
tion into Xenopus laevis eggs. Development 29. Gimlich RL, Braun J (1985) Improved fluo-
99(1):15–23 rescent compounds for tracing cell lineage.
17. Gurdon JB (1962) The transplantation of Dev Biol 109(2):509–514
nuclei between two species of Xenopus. Dev 30. Keller R, Tibbetts P (1989) Mediolateral cell
Biol 5:68–83 intercalation in the dorsal, axial mesoderm of
18. Kroll KL, Amaya E (1996) Transgenic Xenopus laevis. Dev Biol 131(2):539–549
Xenopus embryos from sperm nuclear trans- 31. Zernicka-Goetz M, Pines J, Ryan K, Siemering
plantations reveal FGF signaling requirements KR, Haseloff J, Evans MJ, Gurdon JB (1996)
during gastrulation. Development 122(10): An indelible lineage marker for Xenopus using
3173–3183 a mutated green fluorescent protein.
19. Amaya E, Kroll KL (1999) A method for gen- Development 122(12):3719–3724
erating transgenic frog embryos. Methods Mol 32. Itoh K, Sokol SY (2011) Polarized transloca-
Biol 97:393–414 tion of fluorescent proteins in Xenopus ecto-
20. Bronchain OJ, Hartley KO, Amaya E (1999) A derm in response to Wnt signaling. J Vis Exp
gene trap approach in Xenopus. Curr Biol (51). doi:2700 [pii] 10.3791/2700
9(20):1195–1198 33. Wolanski M, Khosrowshahian F, Kelly LE,
21. Sparrow DB, Latinkic B, Mohun TJ (2000) El-Hodiri HM, Crawford MJ (2009) xArx2: an
A simplified method of generating trans- aristaless homolog that regulates brain regional-
genic Xenopus. Nucleic Acids Res 28(4): ization during development in Xenopus laevis.
E12 Genesis 47(1):19–31
22. Loeber J, Pan FC, Pieler T (2009) Generation 34. Lemaire P, Darras S, Caillol D, Kodjabachian L
of transgenic frogs. Methods Mol Biol 561: (1998) A role for the vegetally expressed
65–72 Xenopus gene Mix.1 in endoderm formation
23. Ogino H, McConnell WB, Grainger RM (2006) and in the restriction of mesoderm to the mar-
High-throughput transgenesis in Xenopus ginal zone. Development 125(13):2371–2380
using I-SceI meganuclease. Nat Protoc 1(4): 35. Bayramov AV, Martynova NY, Eroshkin FM,
1703–1710 Ermakova GV, Zaraisky AG (2004) The home-
24. Pan FC, Chen Y, Loeber J, Henningfeld K, odomain-containing transcription factor X-nkx-
Pieler T (2006) I-SceI meganuclease-mediated 5.1 inhibits expression of the homeobox gene
transgenesis in Xenopus. Dev Dyn 235(1): Xanf-1 during the Xenopus laevis forebrain
247–252 development. Mech Dev 121(12):1425–1441
25. Allen BG, Weeks DL (2005) Transgenic 36. Pei W, Noushmehr H, Costa J, Ouspenskaia
Xenopus laevis embryos can be generated using MV, Elkahloun AG, Feldman B (2007) An
phiC31 integrase. Nat Methods 2(12): early requirement for maternal FoxH1 during
975–979 zebrafish gastrulation. Dev Biol 310(1):10–22
26. Allen BG, Weeks DL (2006) Using phiC31 37. Sive H, Grainger RM, Harland RM (2000)
integrase to make transgenic Xenopus laevis Early development of Xenopus laevis. A labora-
embryos. Nat Protoc 1(3):1248–1257 tory outline, 1st edn. Cold Spring Harbor
27. Sekkali B, Tran HT, Crabbe E, De Beule C, Laboratory Press, Cold Spring Harbor
Van Roy F, Vleminckx K (2008) Chicken beta- 38. Drysdale TA, Elinson RP (1991) Development
globin insulator overcomes variegation of trans- of the Xenopus laevis hatching gland and its
genes in Xenopus embryos. FASEB J 22(7): relationship to surface ectoderm patterning.
2534–2540 Development 111:469–478
 Chapter 11

Morpholino Studies in Xenopus Brain Development

Jennifer E. Bestman and Hollis T. Cline

Abstract
Antisense morpholino oligonucleotides (MOs) have become a valuable method to knock down protein
levels, to block mRNA splicing, and to interfere with miRNA function. MOs are widely used to alter gene
expression during development of Xenopus and zebra fish, where they are typically injected into the fertil-
ized egg or blastomeres. Here, we present methods to use electroporation to target delivery of MOs to the
central nervous system of Xenopus laevis or Xenopus tropicalis tadpoles. Briefly, MO electroporation is
accomplished by injecting MO solution into the brain ventricle and driving the MOs into cells in the brain
with current passing between two platinum plate electrodes, positioned on either side of the target brain
area. The method is straightforward and uses standard equipment found in many neuroscience labs. A
major advantage of electroporation is that it allows spatial and temporal control of MO delivery and there-
fore knockdown. Co-electroporation of MOs with cell-type specific fluorescent protein expression plas-
mids allows morphological analysis of cellular phenotypes. Furthermore, co-electroporation of MOs
with rescuing plasmids allows assessment of specificity of the knockdown and phenotypic outcome. By
combining MO-mediated manipulations with sophisticated assays of neuronal function, such as electro-
physiological recording, behavioral assays, or in vivo time-lapse imaging of neuronal development,
the functions of specific proteins and miRNAs within the developing nervous system can be elucidated.
These methods can be adapted to apply antisense morpholinos to study protein and RNA function in a
variety of complex tissues.

Key words Xenopus, Electroporation, Morpholino oligonucleotide, Antisense, Knockdown, In vivo

imaging, Neuron, Brain, CNS phenotype

1 Introduction

Antisense morpholino oligonucleotides (MOs) are DNA analogs

that are widely used interfere with and knock down gene expres-
sion. Through the substitution of the morpholine ring for DNA’s
riboside moiety, MOs were conceived and designed to overcome
the instability of other DNA antisense oligonucleotides. By virtue
of their chemistry, MOs and their targets are not recognized and
degraded by the RNAse H and therefore offer a stable and specific
steric-blocking method to interfere with RNA processing. When
injected into embryos, antisense MO technology has been shown

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_11, © Springer Science+Business Media, LLC 2014

155
156 Jennifer E. Bestman and Hollis T. Cline

to suffer from off-target effects [1], as do other antisense technolo-

gies (i.e., RNA interference (RNAi; [2])), and so a series of con-
trols have been recommended when MOs are used [3]. The
length of MOs (25 bases) affords MO technology a higher degree
of target sequence specificity than other knockdown technologies
such as short interfering RNAs (siRNAs), for example, where in
addition to the target gene, siRNAs have been shown to repress
the translation of other genes with partial complementarity [2, 4].
It is possible to make siRNAs work in Xenopus by supplementing
RISC (RNA-induced silencing complex) components with exoge-
nous expression [5, 6], thereby permitting genetically regulated
cell-type specific or spatial and temporal control of knockdown.
Nevertheless, the ease of use, faster onset of knockdown, and fewer
problems with target specificity are among the advantages of MO
technology [3].
The major application of MOs is to interfere with translation
of specific gene targets. When MOs are designed to bind to the
sequence immediately surrounding the start codon, they inhibit
the 40S ribosomal subunit from scanning the 5′UTR and finding
the start codon (see Note 1). Therefore, MOs prevent translation
in a highly effective and gene-specific manner. MOs have also been
used to interfere with RNA splicing [7, 8], with microRNAs and
their targets [9, 10], and other RNA-protein interactions [11].
The Xenopus tadpole model system has many advantages for
CNS research. The tadpoles develop externally and are transpar-
ent when they are young. The CNS is positioned directly under
the skin making it accessible for experimental manipulation and
direct observation. Furthermore, barriers to MO use in other
model systems, namely, MO delivery and cell uptake, are relatively
insignificant.
In Xenopus, MOs have been administered by a variety of meth-
ods. A common method is to inject MOs into one cell of Xenopus
embryos during the two-cell stage. The developing frog will then
have one affected side and a non-affected control side [12, 13].
Other methods that allow more tissue and temporally specific MO
administration include lipofection [14, 15] and electroporation
[16–21]. MOs that are tagged with fluorophores (lissamine or
carboxyfluorescein) are charged molecules, which allow them to be
efficiently delivered to cells through electroporation, which we will
detail below.
Here, we describe the methods we use to make observations
of neurodevelopmental events in the developing optic tectum,
the area of the Xenopus tadpole brain that receives direct input
from the retinal ganglion cell axons coming from the eye and that
controls visually guided behaviors. We have also used electropora-
tion to target other areas of the CNS and spinal cord, and methods we
describe here could easily be adapted for those tissues [20, 22, 23].
We co-electroporate fluorophore-tagged MOs and fluorescent
 Morpholino Studies in Xenopus Brain Development 157

protein reporter plasmids (e.g., GFP, CFP, YFP, tdTomato) and

use confocal or 2-photon microscopy to take 3D stacks of the
transfected cells to measure whether interfering with genes of
interest with MO knockdown effects the development of the CNS.

2 Materials

1. Borosilicate glass capillary pipettes. We suggest 1.0 mm

outer × 0.75 mm inner diameters and single barrel with filament
(e.g., World Precision Instruments).
2. Micropipette puller (P-97 Flaming/Brown, Sutter
Instruments). Tapered micropipettes for injecting reagents
through the tadpole skin and into the ventricle of the brain
must be crafted. Ideal micropipettes are shaped so that they are
wispy and sealed at the tip. Before use, the tip is broken with
forceps to create a ~20 μm tip. The pipette tip must be rigid
enough to penetrate the tadpole’s skin without producing an
excessively large hole, but still pass MO solution into the brain
ventricle without clogging. Instructions for injection micropi-
pettes, which are specific to the puller’s filament type (box or
trough or differing diameters) and gauge of glass, can be found
in Chapter 2 of the micropipette fabrication manual, Pipette
Cookbook [24].
3. Microinjector (e.g., Picospritzer II, General Valve Corporation).
4. Dissection microscope (e.g., Wild M8, Leica).
5. Two coarse micromanipulators to hold the micropipette
injection needle and the platinum electrode (e.g., MM33,
Märzhäuser Wetzlar).
6. A square pulse generator (e.g., Grass SD9 Stimulator). Settings:
1.6 ms pulse duration, 30–70 V, and 1–5 pulses. Equipment
that provides precise interpulse intervals is not necessary.
7. 3 μF capacitor (e.g., Part Number, 97F5437, Genteq
Capacitors). The capacitor is wired between the stimulator
output and the platinum electrode to add an exponential decay
waveform to the square pulse.
8. Insulated wires with alligator and/or banana clips to connect
the output of the Grass Stimulator to the capacitor and from
the capacitor to the platinum electrode leads.
9. Micropipette holder. Must have an opening to fit the outer
diameter of the micropipette glass (recommended 1.0 mm),
a port for the input tubing of the picospritzer, and a mount
onto a rod that fits the micromanipulator. As shown in the
photo in Fig. 1a and in the diagram in Fig. 1b. For example,
MP series holders from Warner Instruments, Hamden, CT.
158 Jennifer E. Bestman and Hollis T. Cline

Grass Stimulator

Picospritzer

micropipette electrode
holder

Grass SD9 Stimulator

Duration = 1.6 ms
Volts = 30-70 V
Twin Pulses = regular
MODE = single
Polarity = (see methods)
+
OUTPUT = mono - 3 µF Capacitor
Picospritzer
Duration = 50-500 ms
Pressure = 10-50 psi
Pulses = 5-10

Fig. 1 Configuration of electroporation equipment. (a) A typical setup of the apparatus needed to electroporate
includes a Grass Stimulator square pulse generator, a Picospritzer pressure injector, dissecting microscope,
and micromanipulators to hold the electrode and micropipette for injection. (b) Simple wiring diagram and
settings for the Grass Stimulator and Picospritzer and diagram to indicate the position of the injection micro-
pipette and the plates of the electrode
 Morpholino Studies in Xenopus Brain Development 159

Fig. 2 Custom platinum electrode. (a) The plates of the electrode are constructed out of platinum foil that is
soldered to insulated wire leads. The plates are isolated from one another with shrink tubing. A larger piece of
shrink tubing holds the aligned plates in place, and a still larger piece of shrink tubing holds the electrode onto
the rod that fits in the micromanipulator. Scale bar = 5 mm. (b) A magnified view of the tip of the electrode
showing the folded platinum foil that is folded and cut into ~1 mm wide plates. The bracket indicates the
shrink tubing that insulates electrodes from one another. Scale bar = 2 mm

10. Custom platinum electrode. Description of the fabrication

of the electrode (Figs. 1b and 2) is given in Subheading 3.2,
step 2.
11. Apparatus for housing/moving/positioning tadpoles.
Tadpoles can be reared in large (200–500 ml) bowls (lower
left, Fig. 1a). After imaging experiments begin in which indi-
vidual animals are followed over time, we house tadpoles in
plastic 6-well tissue culture plates with lids. We move individ-
ual tadpoles around by sucking them into disposable plastic
transfer pipettes with openings cut large enough (with a razor
blade) to encompass the tadpoles without injury. For electro-
poration under the microscope, we transfer the tadpoles to an
inverted petri dish with wet tissue on top of it. The petri dish
then can be slid around and the tadpole positioned under
the microscope. A fine paintbrush can be used to adjust the
position of the tadpoles.
We make custom imaging chambers. To do this, we make
thin (5–10 mm) sheets of Sylgard elastomer (Dow Corning),
cut a square piece of the Sylgard that fits on top of a microscope
slide, and then carve tadpole-shaped recesses in the Sylgard so
that the top of the tadpoles’ heads are at the same level as the
surface of the Sylgard sheet (Fig. 3). Tadpoles are placed in the
recesses with a bit of MS-222 solution, and glass coverslips are
placed directly on their heads/on the Sylgard surface. If the
Sylgard is clean and the surface is smooth and flat, the glass
coverslip will make a tight seal. Check to make sure the head
and brain are not squished. Brisk blood flow through the brain
must be maintained throughout the imaging protocol.
160 Jennifer E. Bestman and Hollis T. Cline

Fig. 3 Tadpole imaging chamber. We carve tadpole-shaped recesses in blocks of

Sylgard elastomer. The chambers are placed on a microscope slide so that they
can be locked onto the stage of the microscope. The recesses must be large
enough to hold the tadpole so that the top of their heads is just at the surface of
the Sylgard and the tadpole is not compressed and does not shift during image
acquisition. To hold the tadpoles in place under a water immersion lens, we place
a slide coverslip over the tadpoles on the surface of the Sylgard

12. 100× Steinberg’s stock: 0.5 g KCl, 0.8 g Ca(NO3)2·4H2O,

2.1 g MgSO4·7H2O, 34 g NaCl, and 119 g HEPES to 1 l
dH2O, pH to 7.4, and keep at 4 °C.
13. 1× Steinberg’s working solution: 100 ml 100× Steinberg’s
stock to 10 l dH2O. Add 500 μl penicillin/streptomycin
(5,000 U/ml/5,000 mg/ml, Invitrogen), pH to7.4. Keep
and use at room temperature.
14. Anesthesia: 0.02 % MS-222/Tricaine (3-Aminobenzoic acid
ethyl ester, Sigma-Aldrich) dissolved in 1× Steinberg’s working
solution, pH 7.0.
15. Endotoxin-free plasmid DNA encoding gene of interest/fluo-
rescent protein. Plasmids should be prepared with Qiagen
EndoFree Plasmid Maxi Kit, for example. Vectors with CMV
promoters (e.g., Clontech pEGFP and similar) are used at
1–5 μg/μl. Vectors with Gal4/UAS repeats (e.g., [25]) are
used at 0.1–1 μg/μl.
16. Lissamine- or carboxyfluorescein-tagged morpholino
(GeneTools), diluted in endotoxin-free dH2O; working con-
centrations for electroporation are 0.1–0.5 mM.
17. ~1 % Fast Green stock solution, prepared with endotoxin-free
molecular grade water and filtered through a syringe filter. Fast
Green is added to the DNA/morpholino solution to equal
~0.01–0.1 % so that the injected solution can be seen in the
ventricle of the brain.
 Morpholino Studies in Xenopus Brain Development 161

18. 1 ml disposable plastic syringe for loading micropipettes with

plasmid/morpholino reagents. By melting the center of the
syringe with the plunger pulled out over a flame and pulling
the tip (or carefully letting the end fall to the floor or counter-
top), it will form a wispy capillary that is >1 mm in diameter.
Cut the tip of the pulled syringe until you find it is not sealed
and is sufficiently small to fit inside the back of the micropipettes
(see Fig. 3 in ref. 18). This syringe can be kept and reused to
load micropipettes with plasmid and MO reagents.
19. Xenopus laevis tadpoles. Fertilized eggs are acquired from
hormone-induced matings of albino Xenopus laevis frogs in
our colony. Tadpoles raised at 23 °C with a 12 h light/12 h
dark diurnal cycle until used for experiments. Alternatively,
tadpoles can also be purchased as fertilized eggs/young
embryos from commercial sources (e.g., Nasco, Xenopus
Express, or Xenopus One).
20. Aquarium air pump. Not necessary, but recommended to add
to the bowl to speed recovery from anesthesia. For example,
Tetra 77851“Whisper” Air Pumps for 10 gallon tank.
21. Compound microscope equipped with epifluorescence and
appropriate filter cubes and a low-power (20×) air objective,
optional (e.g., Nikon Optiphot-2). This microscope is used to
screen quickly tadpoles in order to determine whether suffi-
cient cells are transfected and MOs are taken up into cells.
22. A multiphoton or confocal microscope system equipped with
filters and/or lasers appropriate for the fluorescent molecules
in the experiment (e.g., PerkinElmer UltraVIEW spinning disc
confocal system using a Nikon Eclipse FN1 with a 25× 1.1NA
water immersion objective).
23. Incubator for tadpole housing. Set to 23 °C with a 12 h
light/12 h dark diurnal cycle.

3 Methods

3.1 Animal Tadpoles are reared as groups of ~100 in large bowls at 23 °C with
Husbandry a 12 h light/12 h dark diurnal cycle. For our experiments, we typi-
cally use animals that are between Nieuwkoop and Faber stage 46
(5 days postfertilization, dpf) and stage 49 (~12 dpf) [26], though
electroporation is effective on younger and old animals. Once
experiments begin, single tadpoles are kept in 6-well tissue culture
plates so that time-lapse images from identified individual tadpoles
can be acquired. All procedures are conducted with Xenopus tad-
poles anesthetized with 0.02 % MS-222 solution. Before electro-
poration or imaging protocols, the tadpoles are transferred into
the MS-222 solution, and within minutes, they are immobile,
162 Jennifer E. Bestman and Hollis T. Cline

unresponsive to touch, and ready for the procedure. Afterward,

they are revived within minutes of placing them back to their rear-
ing containers with Steinberg’s solution.

3.2 Preparing the 1. Configuring the electroporation equipment. We use an equip-

Electroporation ment configuration shown in Fig. 1a, so that the Grass
Equipment Stimulator and Picospritzer are within arm’s reach while look-
ing through the microscope. The wiring diagram is shown in
Fig. 1b. What is not visible in the photo in Fig. 1a is the 3 μF
capacitor shown in the wiring diagram in Fig. 1b. Before you
begin the electroporation procedure, practice adjusting the
micropipette and electrode so that they both can reach where
the stage/tadpole in the field of view is. Minimizing the time
that the tadpoles are anesthetized and on the stage of the micro-
scope by preparing the equipment will help their recovery from
the procedure.
2. Fabricating the electrode. Two platinum foil plates are fabri-
cated from new or discarded filaments from a Sutter puller that
are folded and cut to a shape that fits the contour of the tadpole
head/brain (Fig. 2). Typical platinum electrodes are approxi-
mately 0.5–1 mm wide and 5–10 mm long and are soldered to
wire leads. These plates must remain electrically isolated but
secured so that the tips of the plates are ~0.5 mm apart and can
straddle the area of the brain to be electroporated. We use heat
shrink tubing around the soldered connection of one electrode
(bracket, Fig. 2b) and then shave down the tubing so the second
platinum plate can be positioned very close to it. Next, position
the plates so they are parallel and use heat shrink tubing to
secure them together. Lastly, secure the plates with shrink-wrap
along a rod that fits in the micromanipulator. Connect the wire
leads as shown in Fig. 1b.

3.3 In Vivo Tectal 1. Preparation of transfer materials for electroporation. Using

Cell Transfection of molecular biology grade, endotoxin-free water, prepare
Plasmid Constructs 10–20 μl working solutions of plasmid, MO, and Fast Green in
and MOs with microcentrifuge tube. Recommended solution concentrations
Electroporation are 1–5 μg/μl for CMV-promoter plasmids, 0.1–1 μg/μl for
Gal4-UAS plasmids, 100–500 μM MO (see Note 2), and
0.01–0.1 % Fast Green. Though there are reports that some
MOs become unstable over time [27], we have had success
keeping these working solutions at 4 °C for weeks and repeat-
edly drawn from them.
2. Loading micropipettes. Using the disposable 1 ml syringe that
has been melted and pulled to a <0.5 mm capillary, carefully
draw 1–2 μl of plasmid into the capillary part of the syringe. Be
careful not to let the vacuum pressure pull the solution into the
body of the syringe because it will be difficult to expel it. Insert
the syringe tip into the micropipette and expel the solution.
 Morpholino Studies in Xenopus Brain Development 163

3. Insert the pipette into the micropipette holder. It must be

secured well or the pressure of the Picospritzer will eject the
micropipette from the holder.
4. Break back the tip of the glass micropipette by brushing it
with a tissue or breaking it with forceps until it has a tip size of
~20 μm.
5. Check that the micropipette is open by using low pressure
(10 psi) and short durations (10 ms) to eject the plasmid/MO.
It should form a small droplet at the tip of the micropipette.
6. Check that the electrode is approximately in the correct posi-
tion and can reach the stage under the microscope where the
tadpole will be placed.
7. Anesthetize tadpoles. Initially anesthetize only the tadpole that
will be used so that there is no chance of overexposing the
tadpole to the anesthetics. As you become more familiar and
faster with the procedure, five to ten tadpoles can be anesthe-
tized simultaneously. To do this, suck up the tadpole into the
transfer pipette and squirt it into the MS-222 solution.
Depending upon how much Steinberg’s is moved with the
tadpole, the MS-222 will become dilute and will need to be
replaced after multiple tadpoles are transferred.
8. Place an anesthetized tadpole along with a little MS-222 solu-
tion on the tissue-covered petri dish using the disposable plas-
tic transfer pipette. Gently arrange the tadpole with the
paintbrush and by moving the petri dish platform so that the
brain is aligned with the glass micropipette as in Fig. 1b.
Younger animals (stage 42 and younger) that still have enlarged
yolky guts tend to tip onto their sides, and using the paint-
brush to curl their tails up along the side of the tadpoles will
help orient them so their dorsal side is up. Alternately, the tad-
poles can be propped up against ridges formed in the tissue to
orient them. Keep the tadpole moist with the MS-222
solution.
9. Insert tip of the micropipette through the skin, into the ventricle
of the brain. The angle of the micropipette must not be too shal-
low or it will slide/glance off the skin. The angle of the pipette
is limited by working distance of the microscope. The tadpole’s
head has a slight downward angle between the midbrain and
its mouth, which can help the pipette penetrate through the
skin when the tadpole is oriented as in Fig. 1b.
10. Eject the plasmid/MO/Fast Green solution from the pipette.
The approximate amount of plasmid/MO solution is ~10 nl,
but because the pipette may break as it penetrates the skin, the
blue color of the Fast Green filling the ventricle is a better indi-
cator of the volume. The Xenopus tectum has two lobes that
are separated by a large ventricle. When we wish to electropor-
ate a large number of tectal cells, we fill the ventricle until the
164 Jennifer E. Bestman and Hollis T. Cline

caudal/lateral corner of the midbrain ventricle is blue from the

Fast Green Dye and the triangular shape of the ventricle is
clear. Overfilling the ventricle will put pressure on the tissue
and may cause damage. Ejecting less plasmid/MO will result
in fewer electroporated cells. Depending on the bore of the
micropipette and the pressure of the Picospritzer, filling the
ventricle may take 10 or more pulses, but 5–10 pulses of 30 ms
duration at 20 psi are a good starting point. Ejecting more,
shorter duration and lower pressure pulses is preferred because
it causes less damage to the brain. As the micropipette is reused
on multiple tadpoles, its tip may break and the pulse duration
and/or pressure should be reduced so that single Picospritzer
ejections do not harm the tadpole.
11. You may choose to leave the micropipette in place or back it up
and out of the way.
12. Positioning the electrode. Lower down the platinum electrode
so that it just spans the target area of the brain. The edges of
the electrode tips should be depressed firmly down on the skin
of the tadpole. The shape of the electrode and the orientation
of how it is placed on the tissue can both be used to target
areas of the brain for transfection. Placement of the electrodes
is important to consider because positively charged molecules
(e.g., lissamine-conjugated MOs) will move toward the nega-
tive pole, while negatively charged molecules (e.g., DNA and
carboxyfluorescein-conjugated MOs) will move toward the
positive pole of the electrode during electroporation.
Electroporation of untagged MOs, however, has also been
shown in some tissues [28]. By orienting the electrode and
electroporating with a single polarity, it is possible to drive
reagents in a single direction in order to target areas of the tis-
sue for transfection. The Grass Stimulator is capable of revers-
ing the polarity of the voltage pulses, which can be used to
electroporate reagents that have opposite charges. To target
both lobes of the tectum, we place the electrode so that the
tips of the electrode plates press down just beyond the lateral
edge of each side of the tectum.
13. Electroporation. Apply voltage pulses by tapping the toggle
switch on the Grass Stimulator. The interpulse interval has not
been found to affect electroporation efficiency. Keep in mind
when co-electroporating plasmid and MOs, that MOs are
more easily/efficiently electroporated than plasmid vectors
(see Note 3). As a rule, increasing the voltage and the number
of pulses will increase the number of cells that are transfected
and the expression levels of individual cells. Efficiency of plas-
mid transfection is also affected by the expression vector used
(as discussed in Subheading 3.2). Regardless of the voltage
level used, delivering more than 5 pulses (with one polarity)
 Morpholino Studies in Xenopus Brain Development 165

is not advised because higher number of pulses may damage

the tissue. For an application where we want to transfect
~25 cells in each lobe of the tectum with a Gal4-UAS plasmid
(i.e., a high efficiency expression vector) and a lissamine-tagged
MO, we typically apply 2–3, 1.6 ms duration, and 30 V pulses
of each polarity. With a CMV-promoter-driven fluorescent
reporter combined with a lissamine-tagged MO, the setting
would change slightly to 4–5, 1.6 ms duration, and 40 V pulses
of each polarity, because the expression vector drives weaker
expression. Signs to look for when electroporating are dis-
cussed in Note 4.
14. Retract electrode and micropipette. Often, the electrode plates
may need to be cleaned (dabbing them with a folded tissue).
15. Place tadpole into recovery chamber. This can be done by
moving the tadpole while it is on the petri dish cover over a
larger bowl and then gently squirting it off the tissue with a
plume of Steinberg’s from the transfer pipette. We return tad-
poles to a bowl of Steinberg’s solution and add the output of
an aquarium air pump to the bowl until the tadpoles are revived
and swimming.
16. Tadpoles are returned to 23 °C incubator and 12 h light/12 h
dark diurnal cycle.

3.4 In Vivo Imaging 1. We wait about 24 h after electroporation before the first image is
of Xenopus Tadpoles acquired to allow sufficient fluorescent protein to be expressed.
to Screen for CNS Another factor to consider in deciding when to collect images
Phenotypes is the rate of protein turnover of the gene targeted by the MO.
Images should be acquired over an interval when the MO is
capable of knocking down protein levels.
2. Individual tadpoles should be transferred into MS-222 and
anesthetized.
3. Anesthetized tadpoles in the MS-222 solution are transferred
to the Sylgard imaging chamber, which is mounted on a micro-
scope slide (Fig. 3). Under a dissecting microscope, tadpoles
are positioned within the carved indentation of the Sylgard so
that the dorsal side of the tadpole is up. Place a coverslip over
the tadpole/Sylgard and press it down so that it attaches to the
wet surface of the Sylgard and presses gently on the surface of
the tadpole’s head. A bit of liquid from the tadpole chamber
on the Sylgard will ensure that the coverslip sticks to the
Sylgard. If tadpoles are selected for confocal/multiphoton
image acquisition, care must be taken so that the tadpole does
not shift midway through the acquisition of the z-stack. See
Note 5 on positioning tadpoles.
4. (Optional) Prescreen tadpoles for imaging experiments.
Tadpoles may be viewed quickly under the screening microscope
166 Jennifer E. Bestman and Hollis T. Cline

with an air objective in order to prioritize them based on the

number of reporter-expressing cells and the quality of plasmid
and MO transfection. The tadpoles and CNS can be inspected
for damage at this point (see Note 6). Simply secure the micro-
scope slide with the Sylgard chamber on the stage of the micro-
scope. Similar to Subheading 3.3, step 15, after prescreening,
individual tadpoles are returned to a larger bowl with an aquar-
ium pump until they are revived. Once swimming, each identi-
fied tadpole is transferred to a 6-well plastic tissue culture dish
so that individuals can be identified and imaged.
5. Acquire confocal/multiphoton z-stack image. Sylgard cham-
ber holding the tadpole is placed under the microscope and
positioned so that the field of view is centered on the region of
interest. Top and bottom levels of the stack are set, filters and
laser(s) are selected, settings are modified to maximize signal
and minimize saturation, and images are acquired.
6. Tadpoles are given time to revive in a bowl with the aquarium
pump input before transferred back to their well of the 6-well
cell culture dish.
7. Return tadpoles to 23 °C incubator and 12 h light/12 h dark
diurnal cycle.

3.5 MO Uptake MOs are more efficiently electroporated into tissue than are
and Fluorescent DNA plasmids. MOs are taken up by more cells over a physically
Protein Expression larger area of the brain than seen by GFP expression from
in the Tectum co- electroporated plasmids. Figure 4 shows two typical examples
of right tectal lobes ~24 h after electroporation of a GFP plasmid
vector and lissamine-tagged control MO. The animals were
injected with 0.1 mM lissamine-tagged control MO and 0.5 μg/μl
turboGFP plasmid under the control of the Gal4/UAS expression
system described in Bestman et al. [29]; the electrode spanned the
tectum as shown in Fig. 1b and electroporated with 3, 1.6 ms, and
40 V pulses with each polarity.
The confocal stacks in Fig. 4 show that the majority of the
tectal cells have taken up the MO (magenta signal), and relatively
few cells express the GFP (green). The right side of the CNS of the
tadpole is lightly outlined in Fig. 4a. The right tectal lobe is shown
in the bright-field image (Fig. 4b) and in the flattened confocal
image of a 50 μm stack taken ~20 μm below the surface of the skin
(Fig. 4c). The image in Fig. 4d is a flattened 180 μm stack through
the right tectal lobe, with magnified views of cell bodies (Fig. 4d
(i), (ii)) and distal processes (Fig. 4d (iii–vi)). These images show
that the MO signal in the tectum is extensive and that the GFP-
expressing cells also contain MO. The differences in the number of
cells expressing GFP and the MO may be due to the different size
and charge of MOs and plasmid DNA. Adjustment of the DNA to
MO ratio may increase the correlation between distribution of
 Morpholino Studies in Xenopus Brain Development 167

Fig. 4 Examples of MO uptake and fluorescent protein expression in two electroporated tectal lobes. The tadpoles
were electroporated 24 h prior to image acquisition with 0.1 mM lissamine-tagged MO (magenta) and 0.5 μg/
μl turboGFP (green). (a) A tadpole head. The right side of the brain is outlined and the bracket indicates the
tectum, which is magnified in (b). (b) A bright-field image of the right tectal lobe with the plane of focus about
60 μm below the surface of the skin. (c) An image of a flattened 50 μm confocal stack taken about 20 μm
below the surface of the skin showing the distribution of the lissamine-tagged control MO and turboGFP-
expressing cells in the right tectal lobe. Scale bar = 50 μm. (d) An image of a flattened 180 μm confocal stack
taken through the right tectal lobe of a stage 46 tadpole. The boxes in (d) are magnified to show the distribution
of MO in the cells. Scale bar = 50 μm. (i, ii) MO fills the soma of the turboGFP-expressing cell, as well as many
of its neighbors. The arrows indicate the position of the turboGFP-expressing cell. (iii–vi) We see that many
cells also show evidence of the MO signal extending >100 μm into the distal processes of the cells. Arrows
indicate the distal tips of tectal cells expressing turboGFP and show the lissamine MO signal. Panels (i)–(vi) are
50 μm square

MOs and expression of reporter from DNA plasmids [30]. If it is

critical for the experimental design to have precise co-delivery of
MOs and plasmid, single-cell electroporation, where transfer material
is delivered to a single cell by delivery of voltage pulse through a
modified patch pipette, may be a better approach for combined
delivery of MO and plasmid reporter [18].

3.6 Verification Control and knockdown verification experiments for MO use have
of Target Knockdown been outlined by Eisen and Smith [3] and Bedell et al. [27] and
and Analysis of also can be found on the GeneTools website (http://www.gene-
Morpholino Efficacy tools.com). We note here that the methods suggested in these
and Specificity resources have been largely developed for experiments where MOs
168 Jennifer E. Bestman and Hollis T. Cline

are administered by the injection into fertilized eggs or blastomeres.

With embryo injection methods, it is possible to calculate precisely
the concentration of MO that the cell receives. A recommended
strategy to investigate specificity of knockdown is to test whether
the knockdown phenotype decreases severity with decreased con-
centrations of MO. With electroporation, the MO concentration
injected into the ventricle is known, but the diluted MO concen-
tration within the ventricle is hard to estimate because it is not a
closed system. More importantly, the efficiency of MO transfer
from the ventricle into cells by electroporation is difficult to
estimate and whether transfer is linearly related to ventricular MO
concentration is unknown. Therefore, electroporation may effec-
tively concentrate MOs in target cells, so injecting a range of MO
concentrations into the ventricle may not result in a corresponding
range of MO concentrations in target cells in a complex tissue like
the brain. Figure 4 gives an indication of the heterogeneity of MO
distribution in electroporated optic tecta and suggests that MO
distribution is uneven across the tissue.
The specificity of an MO on the knockdown of a target gene
can be analyzed through a variety of experiments. First, the results
of experimental MO electroporation should be compared to the
results of electroporation of control MOs. MO knockdown can also
be compared to other results of knockdown methods, such as the
electroporation of dominant-negative plasmid DNA constructs or
other interfering reagents [31–33]. Electroporation also works well
to coadminister a “rescue” construct to replace the gene targeted
by MO knockdown along with the MO [34].
Several authorities recommend using multiple MOs against
the same target transcript to validate specificity of knockdown.
For translation-blocking MOs, the MO is most effective when it
overlaps with the start site, which limits the region against which
MOs can be designed. This is a particular issue for neural genes,
which have many different splice isoforms in the 5′UTR, which
govern promoter specificity. Because the Xenopus genome is not
yet completely annotated, shifting the target region of the MO
around the start site may unwittingly end up targeting alternate
transcripts and generate confusing or erroneous results with respect
to specificity of knockdown. Similarly, recommendations to vali-
date knockdown by comparing effects of translation-blocking MOs
and MOs that interfere with splicing are valid in systems in which
genomic information is complete, but cannot be applied where this
is not the case. The central nervous system has the highest density
of splice variants of an organ. Even as this feature makes studying
the molecular genetic control of brain development particularly
exciting, it also means that the challenges unique to studying nervous
system development and plasticity must be recognized.
When antibodies to the target protein are available, it is quite
common to test whether MOs decrease levels of protein expression
 Morpholino Studies in Xenopus Brain Development 169

with immunocytochemistry or Western blots. However, Western

blot analysis may not detect protein knockdown in the brain
when MOs are administered with electroporation. Unlike MO
injection into a fertilized egg or blastomere, where protein knock-
down can be widespread, targeted electroporation of the tectum,
retina or other areas of the CNS, may result in fewer cells that take
up the MO. The sensitivity of the Western blot to reveal knock-
down will be hindered by the relatively few MO-containing cells in
the tissue that will be homogenized with the surrounding cells that
have not taken up the MO. Despite these caveats, this method has
been shown to verify MO knockdown [31, 33, 35].
Lastly, alternative methods to evaluate knockdown of genes of
interest in the development of the CNS include functional analysis,
such as electrophysiological recordings from neurons electroporated
with MOs against neurotransmitter receptor subunits [33, 36].

4 Notes

1. Design of experimental MOs. The MO vendor, GeneTools,

provides a free design service. Discussion of MO design can be
found here [37].
2. MO concentration. The MO vendor, GeneTools, provides
detailed methods for storing and reconstituting the MO.
Briefly, MOs are stable when reconstituted in pure water and
stored at room temperature. Fluorescently tagged MOs should
be stored in the dark. As stated on the GeneTools website, the
lissamine tag decreases the solubility of the MO and lissamine-
tagged MOs in particular should be heated to 65 °C for 10 min
and vortexed for 30 min until the MO solution is fully resus-
pended. MO concentration can be measured using the proto-
col on the GeneTools website. Considerations for long-term
storage and dilution of MOs are also discussed in [27].
3. Control MOs, choices, and their design. Though the knock-
down efficacy will be attenuated, MOs are capable of comple-
menting a target sequence that shares 21/25 bases [3, 38].
“Specific” control MOs that are incapable of interacting with
the target sequence consist of a 5/25 base mispair and can be
designed for each experimental MO.
4. There are two signs to look for as you electroporate as a way to
verify that the settings are correct and the equipment is wired
correctly. The first is the formation of small electrolysis-induced
bubbles (>10 bubbles). These will appear where the electrode
makes contact with the moist skin. Large violently rolling
bubbles are an indication of a problem with the settings/
equipment. The second indicator is a slight twitch of the ocular
muscles and movement of the eyes.
170 Jennifer E. Bestman and Hollis T. Cline

5. Positioning tadpoles in Sylgard chamber. For prescreening,

this can be done quickly with little regard to the position of
the coverslip and tadpole. When z-stacks are to be acquired, take
care that no bubbles are transferred with the tadpoles inside
of the Sylgard chamber because they might move and shift the
tadpole during the image acquisition. Bubbles can usually be
brushed out with the fine paintbrush. It is also important that
the coverslip gently press on the head of the tadpole. The best
image is acquired when there is no space between the tadpole
and the coverslip. The coverslip must be secure, as it will have the
water droplet for the immersion objective on it. If the coverslip
becomes loose during the acquisition, it will shift the tadpole
and disrupt the image mid-acquisition.
6. Inspecting tadpoles for damage as a result of the electroporation
procedure. We occasionally find tadpoles that do not recover,
or do not recover properly, from electroporation. This may
result from the tadpole drying out during the procedure, and
with experience, this can be minimized. Tadpoles that sustain
damage can be spotted because they fail to swim properly, and
inspection of the brain may show herniation of cells into the
ventricle or signs of bleeding. The evidence of bleeding usually
clears up within 48 h, but these are signs that indicate that the
electrode pressed too hard on the tadpole, the volume injected
into the ventricle was too great, and/or voltage level and
number of electroporation pulses should be scaled back.

References

1. Robu ME, Larson JD, Nasevicius A, Beiraghi 7. Draper BW, Morcos PA, Kimmel CB (2001)
S, Brenner C et al (2007) p53 activation by Inhibition of zebrafish fgf8 pre-mRNA splicing
knockdown technologies. PLoS Genet 3:e78 with morpholino oligos: a quantifiable method
2. Hardy S, Legagneux V, Audic Y, Paillard L for gene knockdown. Genesis 30:154–156
(2010) Reverse genetics in eukaryotes. Biol 8. Morcos PA (2007) Achieving targeted and
Cell 102:561–580 quantifiable alteration of mRNA splicing with
3. Eisen JS, Smith JC (2008) Controlling mor- Morpholino oligos. Biochem Biophys Res
pholino experiments: don’t stop making anti- Commun 358:521–527
sense. Development 135:1735–1743 9. Kloosterman WP, Lagendijk AK, Ketting RF,
4. Summerton JE (2007) Morpholino, siRNA, Moulton JD, Plasterk RH (2007) Targeted
and S-DNA compared: impact of structure and inhibition of miRNA maturation with morpho-
mechanism of action on off-target effects adn linos reveals a role for miR-375 in pancreatic
sequence specificity. Curr Top Med Chem islet development. PLoS Biol 5:e203
7:651–660 10. Choi WY, Giraldez AJ, Schier AF (2007)
5. Chen CM, Chiu SL, Shen W, Cline HT (2009) Target protectors reveal dampening and bal-
Co-expression of Argonaute2 enhances short ancing of Nodal agonist and antagonist by
hairpin RNA-induced RNA interference in miR-430. Science 318:271–274
Xenopus CNS neurons in vivo. Front Neurosci 11. Bruno IG, Jin W, Cote GJ (2004) Correction
3:63 of aberrant FGFR1 alternative RNA splicing
6. Lund E, Sheets MD, Imboden SB, Dahlberg JE through targeting of intronic regulatory ele-
(2011) Limiting ago protein restricts RNAi and ments. Hum Mol Genet 13:2409–2420
microRNA biogenesis during early development 12. Heasman J, Kofron M, Wylie C (2000) Beta-
in Xenopus laevis. Genes Dev 25:1121–1131 catenin signaling activity dissected in the early
 Morpholino Studies in Xenopus Brain Development 171

Xenopus embryo: a novel antisense approach. chronological survey of the development from
Dev Biol 222:124–134 fertilized egg till the end of metamorphosis.
13. Tandon P, Showell C, Christine K, Conlon FL North-Holland Publishing Co, Amsterdam,
(2012) Morpholino injection in Xenopus. p 243
Methods Mol Biol 843:29–46 27. Bedell VM, Westcot SE, Ekker SC (2011)
14. S-i O, Mann F, Boy S, Perron M, Harris WA Lessons from morpholino-based screening in
(2002) Lipofection strategy for the study of zebrafish. Brief Funct Genomics 10:181–188
Xenopus retinal development. Methods 28: 28. Kos R, Tucker RP, Hall R, Duong TD,
411–419 Erickson CA (2003) Methods for introducing
15. Ando H, Okamoto H (2006) Efficient trans- morpholinos into the chicken embryo. Dev
fection strategy for the spatiotemporal control Dyn 226:470–477
of gene expression in zebrafish. Mar Biotechnol 29. Bestman JE, Lee-Osbourne J, Cline HT
(NY) 8:295–303 (2012) In vivo time-lapse imaging of cell pro-
16. Sasagawa S, Takabatake T, Takabatake Y, liferation and differentiation in the optic tec-
Muramatsu T, Takeshima K (2002) Improved tum of Xenopus laevis tadpoles. J Comp Neurol
mRNA electroporation method for Xenopus 520:401–433
neurula embryos. Genesis 33:81–85 30. Sauka-Spengler T, Barembaum M (2008)
17. Eide FF, Eisenberg SR, Sanders TA (2000) Gain- and loss-of-function approaches in the
Electroporation-mediated gene transfer in free- chick embryo. Methods Cell Biol 87:237–256
swimming embryonic Xenopus laevis. FEBS 31. Bestman JE, Cline HT (2008) The RNA bind-
Lett 486:29–32 ing protein CPEB regulates dendrite morpho-
18. Bestman JE, Ewald RC, Chiu SL, Cline HT genesis and neuronal circuit assembly in vivo.
(2006) In vivo single-cell electroporation for Proc Natl Acad Sci USA 105:20494–20499
transfer of DNA and macromolecules. Nat 32. Chiu S-L, Chen C-M, Cline HT (2008) Insulin
Protoc 1:1267–1272 receptor signaling regulates synapse number,
19. Falk J, Drinjakovic J, Leung K, Dwivedy A, dendritic plasticity, and circuit function in vivo.
Regan A et al (2007) Electroporation of Neuron 58:708–719
cDNA/Morpholinos to targeted areas of 33. Shen W, McKeown CR, Demas JA, Cline HT
embryonic CNS in Xenopus. BMC Dev Biol (2011) Inhibition to excitation ratio regulates
7:107 visual system responses and behavior in vivo.
20. Haas K, Jensen K, Sin WC, Foa L, Cline HT J Neurophysiol 106(5):2285–2302
(2002) Targeted electroporation in Xenopus 34. Sharma P, Cline HT (2010) Visual activity reg-
tadpoles in vivo–from single cells to the entire ulates neural progenitor cells in developing
brain. Differentiation 70:148–154 Xenopus CNS through musashi1. Neuron
21. Haas K, Sin W-C, Javaherian A, Li Z, Cline HT 68:442–455
(2001) Single-cell electroporation for gene 35. Schwartz N, Schohl A, Ruthazer ES (2011)
transfer in vivo. Neuron 29:583–591 Activity-dependent transcription of BDNF
22. Javaherian A, Cline HT (2005) Coordinated enhances visual acuity during development.
motor neuron axon growth and neuromuscu- Neuron 70:455–467
lar synaptogenesis are promoted by CPG15 in 36. Ewald RC, Van Keuren-Jensen KR, Aizenman
vivo. Neuron 45:505–512 CD, Cline HT (2008) Roles of NR2A and
23. Ruthazer ES, Li J, Cline HT (2006) NR2B in the development of dendritic arbor
Stabilization of axon branch dynamics by syn- morphology in vivo. J Neurosci 28:850–861
aptic maturation. J Neurosci 26:3594–3603 37. Zhao Y, Ishibashi S, Amaya E (2012) Reverse
24. Osterele A (2011) P-1000 & P-97 pipette genetic studies using antisense morpholino
cookbook, 2011. In:Rev G (ed) Sutter oligonucleotides. Methods Mol Biol 917:
Instrument Company 143–154
25. Koster RW, Fraser SE (2001) Tracing trans- 38. Rana AA, Collart C, Gilchrist MJ, Smith JC
gene expression in living zebrafish embryos. (2006) Defining synphenotype groups in
Dev Biol 233:329–346 Xenopus tropicalis by use of antisense morpho-
26. Nieuwkoop PD, Faber J (1956) Normal table lino oligonucleotides. PLoS Genet 2:
of Xenopus laevis (Daudin); a systemical and 1751–1772
 Part V

Zebrafish Protocols
 Chapter 12

Sensitive Whole-Mount Fluorescent In Situ Hybridization

in Zebrafish Using Enhanced Tyramide Signal Amplification
Gilbert Lauter, Iris Söll, and Giselbert Hauptmann

Abstract
Whole-mount in situ hybridization is the preferred method for detecting transcript distributions in whole
embryos, tissues, and organs. We present here a sensitive fluorescent in situ hybridization method for
colocalization analysis of different transcripts in whole embryonic zebrafish brains. The method is based
on simultaneous hybridization of differently hapten-labeled RNA probes followed by sequential rounds
of horseradish peroxidase (POD)-based transcript detection. Sequential detection involves enhancement of
fluorescent signals by tyramide signal amplification (TSA) and effective inactivation of the antibody–POD
conjugate prior to the following detection round. We provide a detailed description of embryo preparation,
hybridization, antibody detection, POD–TSA reaction, and mounting of embryos for imaging. To achieve
high signal intensities, we optimized key steps of the method. This includes improvement of embryo
permeability by hydrogen peroxide treatment and efficacy of hybridization and TSA–POD reaction by
addition of the viscosity-increasing polymer dextran sulfate. The TSA–POD reaction conditions are further
optimized by application of substituted phenol compounds as POD accelerators and use of highly efficient
bench-made tyramide substrates. The obtained high signal intensities and cellular resolution of our method
allows for co-expression analysis and generation of three-dimensional models. Our protocol is tailored to
optimally work in zebrafish embryos, but can surely be modified for application in other species as well.

Key words Fluorescent in situ hybridization, FISH, Tyramide signal amplification, TSA, Peroxidase,
Zebrafish

1 Introduction

With the rapid progress of high-throughput sequencing and

improved cloning technologies, available sequence data and clone
collections of genomic DNA and cDNA have been steadily increas-
ing. The use of these resources allows producing gene expression
data to construct genomic atlases of different tissues and organs.
Vertebrate forebrain molecular maps of the telencephalon [1, 2],
hypothalamus [3, 4], thalamus [3, 5], and pretectum [6, 7] have
been generated to potentially reveal novel aspects of the underly-
ing basic brain organization. Such gene expression data are rou-
tinely produced by in situ hybridization technologies to allow

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_12, © Springer Science+Business Media, LLC 2014

175
176 Gilbert Lauter et al.

Fig. 1 Two-color whole-mount FISH using enhanced tyramide signal amplification. Lateral views of dlx2a
and lhx1a expression in a 28-hpf zebrafish forebrain with anterior oriented to the left. Transcript signals are
visualized in cyan and magenta as indicated in the panels. Single-channel detection and the overlay of both
channels in the same confocal plane are shown. (a) Expression of dlx2a was visualized by using a DNP-labeled
RNA probe together with the DyLight 633-tyramide. The TSA reaction was allowed to proceed for 20 min in the
presence of 0.15 mg/ml 4-iodophenol. (b) Expression of lhx1a was visualized by using a DIG-labeled RNA
probe together with the FAM-tyramide. The TSA reaction was allowed to proceed for 30 min in the presence of
0.15 mg/ml 4-iodophenol. (c) Overlay of both single channels shows the expression of dlx2a and lhx1a in the
developing forebrain with areas of overlap indicated in yellow. Scale bars = 50 μm

global and detailed views on gene expression patterns [8].

Because of the complexity of the brain, it is however often essen-
tial to relate a distinct expression domain of a gene of interest to
well-known tissue- or cell-type-specific molecular markers. For
this purpose, in situ hybridization methods have been developed
to detect two or more transcript patterns in different colors simul-
taneously in the same tissue or embryo [9–12].
In the original whole-mount in situ hybridization protocol,
a digoxigenin-labeled nucleic acid probe specific for a unique
transcript is detected by anti-digoxigenin antibodies conjugated
to alkaline phosphatase. The transcript distribution is visualized by
formation of an insoluble cellular purple precipitate during the chro-
mogenic substrate reaction [8]. Protocols for two-color detection
add a second nucleic acid probe with another hapten label such as
biotin or fluorescein to the hybridization mix and use two different
chromogenic substrates to obtain contrasting color precipitates.
These protocols were first developed in zebrafish and flies and
included alkaline phosphatase, horseradish peroxidase, and ß-galac-
tosidase as reporter enzymes [9, 10, 12–15].
One drawback of chromogenic detection, however, can be
compromised cellular resolution especially in case of overlapping
expression domains. To overcome this caveat, fluorescent detec-
tion methods with increased sensitivity have been developed, so
that two or more different transcripts could be detected with high
signal strength and signal-to-noise ratio. The problem of weak
fluorescence signal was largely solved by implementation of tyra-
mide signal amplification (TSA) [16] into fluorescent in situ
hybridization (FISH) protocols [17]. TSA dramatically enhances
 FISH in Zebrafish 177

the FISH signals and makes multiplex mRNA detection feasible in

Drosophila [18], Platynereis [19], zebrafish [20, 21], frog [22],
and chicken [23]. One problem with TSA however can be the
amplification of background staining in parallel to specific signal
enhancement, which may significantly reduce achievable signal-to-
noise ratios. Strong autofluorescence of zebrafish and other
embryos may additionally lower signal-to-noise ratios to a nonac-
ceptable level for multiplexing. Therefore, further optimization of
embryo preparation, hybridization, and detection steps have been
necessary to achieve optimal signal strengths and results in zebraf-
ish embryos (Fig. 1) [24].
We describe here an optimized protocol for colocalization
analysis of different mRNA transcripts in whole embryonic zebraf-
ish brains. For best results we optimized key steps of the procedure
[21, 24]. After fixation, permeability of embryos is improved by
treatment with hydrogen peroxide aside standard proteinase K
digestion. Hybridization efficacy is enhanced by addition of the
viscosity-increasing polymer dextran sulfate. Differently hapten-
labeled RNA probes (e.g. dinitrophenol and digoxigenin) are
simultaneously hybridized and sequentially detected by their
respective anti-hapten antibodies, each coupled to horseradish per-
oxidase (POD). Inclusion of dextran sulfate in the TSA reaction
and addition of substituted phenol compounds as accelerators
strongly enhance achievable fluorescent signal intensity. Between
detection rounds consisting of antibody–POD conjugate incubation
and TSA reaction, the POD activity of the prior round has to be
efficiently removed by acidic pH. Finally, we provide a short proto-
col for synthesis of tyramides, as we find that bench-made substrates
are more efficient than commercially available tyramides in produc-
ing high signal sensitivity [24]. Our FISH protocol is tailored to
work optimally in whole zebrafish embryos and embryonic brains
(Fig. 1), but may be easily adapted for application to other model
organisms.

2 Materials

2.1 General Buffers 1. Paraformaldehyde (PFA): 4 % (w/v) paraformaldehyde in 1×

PBS, pH 7.3. Dissolve 4 g PFA in 100 ml PBS and stir in a
fume hood at about 65 °C until everything has dissolved
(1–2 h). Let solution cool to room temperature (RT) and
adjust to pH 7.3 with NaOH. Store the fixative in 5–10 ml
aliquots at −20 °C.
2. Phosphate-buffered saline (1× PBS): 8 % (w/v) NaCl, 0.2 %
(w/v) KCl, 16 mM Na2HPO4, 4 mM NaH2PO4, pH 7.3.
3. Phosphate-buffered saline plus Tween (PBST): 1×PBS, 0.1 %
(v/v) Tween-20.
4. Methanol (MeOH).
178 Gilbert Lauter et al.

5. 30 % hydrogen peroxide (H2O2), stabilized (Sigma 31642).

6. Rehydration series: 75 %, 50 %, and 25 % (v/v) of methanol in
PBST.
7. Proteinase K: 20 mg/ml stock in TE (10 mM Tris–HCl
pH 8.0, 1 mM EDTA), store in aliquots at −20 °C. Prepare
working solution of 10 μg/ml proteinase K in PBST just prior
to use.
8. Glycine: 2 mg/ml glycine in PBST. Store glycine as a 100 mg/
ml stock in ddH2O at –20 °C in aliquots.

2.2 Tyramide 1. Succinimidyl esters: 5-(and-6)-carboxyfluorescein succin-

Synthesis imidyl ester (FAM-SE; Molecular Probes), DyLight
633 N-hydroxysuccinimide ester (DyLight 633-SE; Pierce).
Prepare a 10 mg/ml stock solution of each succinimidyl ester
in dimethylformamide (Sigma) just before use.
2. Tyramine hydrochloride: Prepare a 10× stock solution at a
concentration of 100 mg/ml in dimethylformamide just before
use. Dilute with dimethylformamide to yield the 1× working
solution and add 10 μl triethylamine (Sigma) per 1 ml
solution.
3. Absolute ethanol (EtOH).

2.3 Hybridization 1. Prehybridization buffer (HB): 50 % (v/v) deionized for-

mamide, 5× SSC, 5 mg/ml torula RNA (Sigma), 50 μg/ml
heparin sodium salt, 0.1 % (v/v) Tween-20. Store in aliquots at
−20 °C. Store heparin as a 50 mg/ml stock in ddH2O at
−20 °C.
2. Dextran sulfate (Sigma): Prepare a 50 % (w/v) stock solution
in water. When used for hybridization, autoclave stock solu-
tion for 30 min at 110 °C, otherwise store in aliquots at
−20 °C.
3. Hybridization buffer (HBD5): HB including 5 % (w/v) dex-
tran sulfate.
4. Hybridization wash: 50 % (v/v) deionized formamide
(AppliChem), 2× SSC, 0.1 % (v/v) Tween-20. Store in ali-
quots at −20 °C.
5. 20× SSC: 3 M NaCl, 300 mM trisodium citrate, pH 7.0.
6. 2× SSCT: 2× SSC, 0.1 % (v/v) Tween-20.
7. 0.2× SSCT: 0.2× SSC, 0.1 % (v/v) Tween-20.

2.4 Antibody 1. Blocking solution: 8 % (v/v) normal sheep serum in 1×

Detection PBST. Heat inactivate sheep serum stock (Sigma) at 56 °C for
30 min and store in aliquots at −20 °C.
2. Sheep-anti-digoxigenin-POD Fab fragments (Roche): Prepare
1:500 working dilution in blocking solution prior to use.
 FISH in Zebrafish 179

3. Anti-dinitrophenyl-POD (PerkinElmer TSA Plus DNP

System): Prepare 1:100 working dilution in blocking solution
prior to use.

2.5 Fluorogenic 1. 4-Iodophenol (Fluka 58020): Prepare a 150 mg/ml stock in

Reaction DMSO. Store tightly sealed at 4 °C (see Note 1).
2. Vanillin (Sigma V110-4): Prepare a 150 mg/ml stock in
DMSO. Store tightly sealed at 4 °C.
3. TSA reaction buffer: 100 mM borate buffer pH 8.5, 2 % (w/v)
dextran sulfate, 0.1 % (v/v) Tween-20, 0.003 % H2O2.
4. Borate buffer: 100 mM borate pH 8.5 plus 0.1 % (v/v)
Tween-20.
5. POD inactivation solution: 100 mM glycine-HCl pH 2.0 plus
0.1 % (v/v) Tween-20.

2.6 Mounting 1. Glycerol series: 25 %, 50 %, and 75 % (v/v) of glycerol in PBST,

40 mM NaHCO3.
2. Mounting gel: 1 % (w/v) low melting agarose in 75 % (v/v)
glycerol in PBST, 40 mM NaHCO3. Keep the solution stirring
at 50 °C, as it will turn yellow after repeated heating.
3. Mounting slides: Two stacks of coverslips are glued to a micro-
scope slide with Eukitt (Fluka) leaving a gap, which can be
easily bridged by a 24 × 32 mm coverslip. To mount samples of
varying thickness, prepare a series of mounting slides using one
to several coverslips for each stack.

3 Methods

3.1 Tyramide Despite the advantage of being cost-efficient, bench-made tyrami-

Synthesis des are also highly concentrated and offer the possibility to specifi-
cally match the requirements of the microscope in use.
1. Mix the freshly prepared tyramine working solution and the
respective succinimidyl esters at a 1:1.1 equimolar ratio.
2. Allow the reaction to proceed for 2 h in the dark without
agitation.
3. The resulting tyramide product is diluted with absolute etha-
nol to a concentration of 1 mg/ml and stored protected from
light at −20 °C (see Note 2).

3.2 Embryo 1. Transfer dechorionated embryos of the desired developmental

Pretreatment stage to a 2 ml microcentrifuge tube. Aspirate supernatant while
taking care that embryos remain immersed in liquid. Fixate
embryos in 1 ml of 4 % PFA for 24 h at 4 °C (see Note 3).
2. Rinse embryos four times for 5 min with 1× PBST. Transfer
embryos into 100 % MeOH. Exchange 100 % MeOH after
180 Gilbert Lauter et al.

5 min and incubate embryos at −20 °C for at least 30 min.

Alternatively, embryos can be kept in 100 % MeOH at −20 °C
for long-term storage.
3. Incubate embryos in a 2 % hydrogen peroxide solution in 100 %
MeOH for 20 min. Gradually rehydrate embryos by going
through a series of 5-min washing steps of 75 %, 50 %, and 25 %
MeOH in PBST followed by two PBST washes (see Note 4).
4. Embryos developed to tail bud stage or further have to be
digested with proteinase K in order to increase permeability.
The optimal digestion times for different stages have to be
determined experimentally. After treatment with proteinase
K solution (10 μg/ml in PBST) at RT and under gentle
agitation, stop the reaction by rinsing twice with 2 mg/ml
glycine in PBST. Postfix embryos for 20 min in 4 % PFA.
Afterwards, wash four times for 5 min in PBST. Transfer
embryos into 0.5 ml prehybridization buffer (HB) and
exchange with 1 ml HB after 5 min. Embryos are now ready
to use or can be stored in HB at −20 °C.
5. Transfer up to 25 embryos into 200 μl of HB in a 2 ml
microfuge tube with round bottom. Incubate in a water
bath at 60 °C for 1 h for prehybridization.

3.3 Hybridization 1. Prepare probe mix by adding dinitrophenol (DNP)- and

digoxigenin (DIG)-labeled RNA probes together in 150 μl
HBD5 (see Note 5). Denature the probe mix for 5 min at
80 °C and equilibrate to 60 °C.
2. Carefully aspirate prehybridization solution from embryos and
add the pre-warmed probe mix instead. Incubate in a water
bath at 60 °C overnight (min. 15 h).
3. For all post-hybridization washes, solutions have to be pre-
warmed to 60 °C.
4. Incubate embryos twice in 1 ml SSCT hybridization wash
solution for 30 min.
5. Wash once with 1.5 ml 2× SSCT for 15 min.
6. Wash twice with 1.5 ml 0.2× SSCT for 30 min.
7. Add fresh 0.2× SSCT and let cool down to RT.
8. Rinse twice with PBST.

3.4 Antibody The differently hapten-labeled RNA probes are sequentially

Detection detected using peroxidase (POD)-conjugated antibodies directed
against the respective hapten. Typically DNP-labeled probes are
detected first (for reasons see Note 6) using an anti-DNP-POD
antibody (1:100) followed by detection of DIG-labeled probes
with sheep-anti-DIG-POD (1:500) (see Note 7).
1. Incubate embryos in 100 μl blocking solution for 30 min on a
gently rocking table at RT.
 FISH in Zebrafish 181

2. Carefully remove the blocking solution and add antibody solu-

tion. Incubate overnight at 4 °C without agitation.
3. To remove excess antibody, wash six times with PBST for
20 min at RT with gentle agitation.

3.5 Fluorogenic The TSA reaction can be significantly enhanced by the use of a
Reaction POD accelerator. Use 4-iodophenol or vanillin at a concentration
of 0.15 mg/ml and 0.45 mg/ml, respectively (see Note 8).
1. For 1 ml reaction buffer, combine 500 μl borate buffer pH 8.5
(200 mM), 40 μl 50 % dextran sulfate, 10 μl of 10 % Tween-
20, 6 μl of 0.5 % H2O2, and an appropriate accelerator and
adjust the volume with water to 1 ml (see Note 9).
2. Addition of 4-iodophenol will result in the appearance of a
cloudy smear. Mix the reaction buffer well by pipetting up and
down until the dextran sulfate has dispersed and the solution
appears opaque.
3. Dilute the desired tyramide with the reaction buffer and mix
well by pipetting. Use 6 μl of DyLight 633-tyramide or 4 μl
of FAM-tyramide stock solution per 1 ml reaction buffer
(see Note 10).
4. Rinse embryos twice with 100 mM borate buffer pH 8.5 con-
taining 0.1 % Tween-20. Remove supernatant borate buffer as
closely as possible and add 90 μl of the tyramide solutions to
the embryos. Gently mix by pipetting using a cut tip. Incubate
for the desired length of time in the dark at RT without agitation
(see Note 11).
5. To stop the TSA reaction, wash for four times with PBST. For
each washing step, fill the entire tube with PBST. After invert-
ing the tubes for several times, wait until the embryos are at
the bottom and then remove excess buffer carefully. From this
step onward, embryos should be protected from light.
6. To inactivate POD activity of the first antibody, incubate
samples in 100 mM glycine-HCl pH 2.0 for 10 min followed by
four 5-min washes in PBST under agitation (see Note 12).
7. For the second antibody detection round, repeat the steps
described under Subheading 3.4 using sheep-anti-DIG-POD
Fab fragments (1:500).
8. Repeat steps 1–5 for fluorogenic detection (see Note 13).

3.6 Mounting 1. To avoid shrinkage, gradually transfer the embryos through

5-min washing steps in 25 %, 50 %, and 75 % glycerol in PBST,
40 mM NaHCO3 (see Note 14).
2. After embyo preparation under a dissecting microscope,
immerse the sample in mounting gel solution.
182 Gilbert Lauter et al.

3. Transfer the sample onto a microscope slide between spacers

of respective heights and apply coverslip.
4. By gently moving the coverslip, the sample can be rotated into
the desired orientation.
5. Put the mounted sample into the fridge until the agarose has
solidified.

4 Notes

1. Be careful when working with 4-iodophenol, as it is a highly

aggressive substance and should only be handled under a fume
hood wearing appropriate protections at all times (even when
highly diluted).
2. Tyramide reagents diluted in absolute ethanol and stored at
−20 °C are at least stable for 3 years.
3. The optimal fixing conditions for different stages have to be
determined experimentally by adjusting fixation time and
temperature. Embryos that are older than 1 day are usually
fixed for 24 h at 4 °C.
4. Hydrogen peroxide treatment promotes embryo permeabili-
zation properties resulting in improved signal detection.
5. As a rule of thumb, use comparable concentration of RNA
probes as with chromogenic BCIP/NBT detection.
Exaggerated probe concentrations will result in a decreased
signal-to-noise ratio. In case three different mRNA transcript
patterns are to be compared in the same embryo, in addition
to DIG- and DNP-, a fluorescein-labeled mRNA probe is
included in the probe mix. Labeling probes with biotin is not
recommended because the biotin label may produce high
background signals in zebrafish embryos [13]. However,
biotin-labeled probes work well in Drosophila [10]. For detailed
descriptions of RNA probe preparation, see [11].
6. The anti-DNP-POD antibody shows cross-reactivity with the
periderm, which surrounds the zebrafish embryo as a thin cell
layer during early stages of development. This usually results in
visualization of the outline of the embryo. As the first round of
detection usually requires only short staining times resulting in
low background, this effect is minimized when detecting the
DNP RNA first.
7. In case a fluorescein-labeled probe is included in the hybridiza-
tion mix, it is detected by rabbit-anti-fluorescein/Oregon Green
488-POD diluted 1:500 in blocking solution prior to use.
8. In general 4-iodophenol is a more potent enhancer than
vanillin. Higher concentrations of accelerator maybe used to
 FISH in Zebrafish 183

further increase signal intensity, although adverse effects on

the signal-to-noise ratio should be kept in mind.
9. Always make fresh TSA reaction buffer just before use. Preparing
1 ml reaction buffer in a 2 ml tube helps to minimize eventual
spill of aggressive 4-iodophenol-rich solution during mixing.
10. Since acidic POD inactivation greatly diminishes the signal
intensity of FAM-tyramide staining [25], DyLight633-
tyramide is usually used in the first and FAM-tyramide in the
second round of detection. In case of three-color detection,
the third transcript is visualized with carboxytetramethylrho-
damine (TAMRA) tyramide [24].
11. Since the 50 % dextran sulfate stock solution is very viscose, use
a cut 200 μl tip for pipetting. Take special care not to transfer
embryos by accident from one tube to another. In order to
avoid a decline in the signal-to-noise ratio, the reaction time
should not exceed 30 min.
12. The acidic inactivation step is pivotal to avoid that the first
detected probe is visible in two fluorescence detection channels.
Incubation in hydrogen peroxide solutions bears the risk of
partial POD inactivation [24, 26]. This may cause false-positive
colocalization signals.
13. In case a third fluorescein-labeled probe is included in the
hybridization mix, the second antibody–POD conjugate is also
inactivated (repeat step 6) and a third antibody detection (repeat
the steps described under Subheading 3.4) and fluorogenic
reaction (repeat steps 1–5 of Subheading 3.5) round is added.
The fluorescein-labeled probe is detected with rabbit-anti-
fluorescein/Oregon Green 488-POD and TAMRA-tyramide is
used as third substrate for the fluorogenic reaction. We found
that best results in a three-color experiment are obtained when
the fluorescein label is detected first with DyLight 633 followed
by DNP with TAMRA and DIG with FAM.
14. Using 40 mM NaHCO3 ensures that the pH is above 8
(see also Note 10). Stained embryos stored in 75 % glycerol in
PBST, 40 mM NaHCO3 are stable for several months and the
use of special anti-fading agents is usually not necessary.

Acknowledgments

The Stockholm Greater Council and the Karolinska Institutet

supported this work. Imaging was in part performed at the Live
Cell Imaging Unit, Department of Biosciences and Nutrition,
Karolinska Institutet, Huddinge, Sweden, supported by grants
from the Knut and Alice Wallenberg Foundation, the Swedish
Research Council, and the Centre for Biosciences.
184 Gilbert Lauter et al.

References
1. Abellan A, Medina L (2009) Subdivisions and 8. Tautz D, Pfeifle C (1989) A non-radioactive in
derivatives of the chicken subpallium based on situ hybridization method for the localization
expression of LIM and other regulatory genes of specific RNAs in Drosophila embryos reveals
and markers of neuron subpopulations during translational control of the segmentation gene
development. J Comp Neurol 515(4):465–501 hunchback. Chromosoma 98(2):81–85
2. Puelles L, Kuwana E, Puelles E, Bulfone A, 9. Hauptmann G, Gerster T (1994) Two-color
Shimamura K, Keleher J, Smiga S, Rubenstein whole-mount in situ hybridization to verte-
JL (2000) Pallial and subpallial derivatives in brate and Drosophila embryos. Trends Genet
the embryonic chick and mouse telencephalon, 10(8):266
traced by the expression of the genes Dlx-2, 10. Hauptmann G, Gerster T (1996) Multicolour
Emx-1, Nkx-2.1, Pax-6, and Tbr-1. J Comp whole-mount in situ hybridization to
Neurol 424(3):409–438, doi:10.1002/1096- Drosophila embryos. Dev Genes Evol 206(4):
9861(20000828)424:3<409::AID- 292–295
CNE3>3.0.CO;2-7 [pii] 11. Hauptmann G, Gerster T (2000) Multicolor
3. Diez-Roux G, Banfi S, Sultan M, Geffers L, whole-mount in situ hybridization. Methods
Anand S, Rozado D, Magen A, Canidio E, Mol Biol 137:139–148, doi:1-59259-066-7-
Pagani M, Peluso I, Lin-Marq N, Koch M, 139 [pii] 10.1385/1-59259-066-7:139
Bilio M, Cantiello I, Verde R, De Masi C, 12. Jowett T, Lettice L (1994) Whole-mount in
Bianchi SA, Cicchini J, Perroud E, Mehmeti S, situ hybridizations on zebrafish embryos
Dagand E, Schrinner S, Nurnberger A, Schmidt using a mixture of digoxigenin- and
K, Metz K, Zwingmann C, Brieske N, Springer fluorescein-labelled probes. Trends Genet
C, Hernandez AM, Herzog S, Grabbe F, 10(3):73–74
Sieverding C, Fischer B, Schrader K,
Brockmeyer M, Dettmer S, Helbig C, Alunni 13. Hauptmann G (1999) Two-color detection of
V, Battaini MA, Mura C, Henrichsen CN, mRNA transcript localizations in fish and fly
Garcia-Lopez R, Echevarria D, Puelles E, embryos using alkaline phosphatase and beta-
Garcia-Calero E, Kruse S, Uhr M, Kauck C, galactosidase conjugated antibodies. Dev
Feng G, Milyaev N, Ong CK, Kumar L, Lam Genes Evol 209(5):317–321
M, Semple CA, Gyenesei A, Mundlos S, 14. Hauptmann G (2001) One-, two-, and three-
Radelof U, Lehrach H, Sarmientos P, Reymond color whole-mount in situ hybridization to
A, Davidson DR, Dolle P, Antonarakis SE, Drosophila embryos. Methods 23(4):359–372,
Yaspo ML, Martinez S, Baldock RA, Eichele G, doi:10.1006/meth.2000.1148
Ballabio A (2011) A high-resolution anatomi- S1046-2023(00)91148-4 [pii]
cal atlas of the transcriptome in the mouse 15. O’Neill JW, Bier E (1994) Double-label in situ
embryo. PLoS Biol 9(1):e1000582 hybridization using biotin and digoxigenin-
4. Shimogori T, Lee DA, Miranda-Angulo A, tagged RNA probes. Biotechniques 17(5):870,
Yang Y, Wang H, Jiang L, Yoshida AC, Kataoka 874, 875
A, Mashiko H, Avetisyan M, Qi L, Qian J, 16. Bobrow MN, Harris TD, Shaughnessy KJ, Litt
Blackshaw S (2010) A genomic atlas of mouse GJ (1989) Catalyzed reporter deposition, a
hypothalamic development. Nat Neurosci novel method of signal amplification.
13(6):767–775 Application to immunoassays. J Immunol
5. Suzuki-Hirano A, Ogawa M, Kataoka A, Methods 125(1–2):279–285, doi:0022-1759
Yoshida AC, Itoh D, Ueno M, Blackshaw S, (89)90104-X [pii]
Shimogori T (2011) Dynamic spatiotemporal 17. van de Corput MP, Dirks RW, van Gijlswijk RP,
gene expression in embryonic mouse thalamus. van Binnendijk E, Hattinger CM, de Paus RA,
J Comp Neurol 519(3):528–543 Landegent JE, Raap AK (1998) Sensitive
6. Ferran JL, Sanchez-Arrones L, Sandoval JE, mRNA detection by fluorescence in situ
Puelles L (2007) A model of early molecular hybridization using horseradish peroxidase-
regionalization in the chicken embryonic pre- labeled oligodeoxynucleotides and tyramide
tectum. J Comp Neurol 505(4):379–403. signal amplification. J Histochem Cytochem
doi:10.1002/cne.21493 46(11):1249–1259
7. Lauter G, Söll I, Hauptmann G (2013) 18. Kosman D, Mizutani CM, Lemons D, Cox WG,
Molecular characterization of prosomeric and McGinnis W, Bier E (2004) Multiplex detection
intraprosomeric subdivisions of the embryonic of RNA expression in Drosophila embryos.
zebrafish diencephalon. J Comp Neurol 521(5): Science 305(5685):846, doi:10.1126/sci-
1093–1118 ence.1099247 305/5685/846 [pii]
 FISH in Zebrafish 185

19. Tessmar-Raible K, Steinmetz PR, Snyman H, 23. Denkers N, Garcia-Villalba P, Rodesch CK,
Hassel M, Arendt D (2005) Fluorescent Nielson KR, Mauch TJ (2004) FISHing for
two-color whole mount in situ hybridization in chick genes: triple-label whole-mount fluores-
Platynereis dumerilii (Polychaeta, Annelida), cence in situ hybridization detects simultane-
an emerging marine molecular model for evo- ous and overlapping gene expression in avian
lution and development. Biotechniques embryos. Dev Dyn 229(3):651–657.
39(4):460, 462, 464. doi:000112023 [pii] doi:10.1002/dvdy.20005
20. Clay H, Ramakrishnan L (2005) Multiplex 24. Lauter G, Söll I, Hauptmann G (2011)
fluorescent in situ hybridization in zebrafish Multicolor fluorescent in situ hybridization to
embryos using tyramide signal amplification. define abutting and overlapping gene expres-
Zebrafish 2(2):105–111. doi:10.1089/ sion in the embryonic zebrafish brain. Neural
zeb.2005.2.105 Dev 6(1):10, doi:1749-8104-6-10
21. Lauter G, Söll I, Hauptmann G (2011) Two- [pii]10.1186/1749-8104-6-10
color fluorescent in situ hybridization in the 25. Fischer AH, Jacobson KA, Rose J, Zeller R
embryonic zebrafish brain using differential (2008) Media for mounting fixed cells on
detection systems. BMC Dev Biol 11:43 microscope slides. CSH Protoc 2008:pdb ip52
22. Vize PD, McCoy KE, Zhou X (2009) 26. Liu G, Amin S, Okuhama NN, Liao G, Mingle
Multichannel wholemount fluorescent and LA (2006) A quantitative evaluation of peroxi-
fluorescent/chromogenic in situ hybridiza- dase inhibitors for tyramide signal amplification
tion in Xenopus embryos. Nat Protoc mediated cytochemistry and histochemistry.
4(6):975–983, doi:nprot.2009.69 [pii] Histochem Cell Biol 126(2):283–291.
10.1038/nprot.2009.69 doi:10.1007/s00418-006-0161-x
 Chapter 13

Dynamic Neuroanatomy at Subcellular

Resolution in the Zebrafish
Adèle Faucherre and Hernán López-Schier

Abstract
Genetic means to visualize and manipulate neuronal circuits in the intact animal have revolutionized
neurobiology. “Dynamic neuroanatomy” defines a range of approaches aimed at quantifying the architecture
or subcellular organization of neurons over time during their development, regeneration, or degeneration.
A general feature of these approaches is their reliance on the optical isolation of defined neurons in toto
by genetically expressing markers in one or few cells. Here we use the afferent neurons of the lateral line
as an example to describe a simple method for the dynamic neuroanatomical study of axon terminals in the
zebrafish by laser-scanning confocal microscopy.

Key words Neuroanatomy, Zebrafish, Live imaging, Microscopy, Fluorescent protein

1 Introduction

Neural circuits involved in behaviors evoked by environmental

signals convey sensory information from peripheral receptors to the
central nervous system and return to the musculature via premotor
neural pathways [1]. Because sensory-motor circuits perform their
main physiological function only in their natural context, explanted
tissues rarely provide an ideal model system for their study [2, 3].
Therefore, enormous effort has been invested in developing tech-
niques for imaging and manipulating neurons over extended periods
in the intact animal [4–7]. The zebrafish (Danio rerio) has emerged
as a powerful model system for the implementation of systematic
dynamic neuroanatomical studies [8–10]. It compares favorably
with other animal models, offering a number of advantages such as
its rapid sexual development and fecundity, which facilitates trans-
genesis and permits the generation of hundreds of animals for
quantitative/statistical analyses. The zebrafish embryo develops
rapidly, externally, and is optically transparent, which facilitates the
visualization of neurons for detailed cellular, molecular, and physi-
ological analyses [11]. Finally, because all vertebrates share many

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_13, © Springer Science+Business Media, LLC 2014

187
188 Adèle Faucherre and Hernán López-Schier

of the cellular and physiological mechanisms that underlie neural

circuit development and performance, the zebrafish provides a highly
relevant model to human neurobiology and neuropathology [12].
Small fluorescent proteins can be expressed in genetically
defined neuronal types and also targeted to specific subcellular
compartments, serving as excellent markers for the entire neuron
and its axonal projections or dendritic trees. However, for densely
packed axons or overlapping dendritic trees, expressing a fluores-
cent protein in a large population of neurons may prevent detailed
visualization and quantification [13, 14]. Mosaic gene expression
to mark a small fraction of neurons within a large population can
overcome this limitation [10, 15, 16]. There are several strategies
for the systematic expression of fluorescent markers in single or
few genetically defined neurons. Some of these methods are based
on the generation of a silent transgene in all the neurons, which
can later be expressed in cell clones by recombinase-based excision
of the silencing/repressing genetic elements in the transgene [17].
Other approaches are based on site-specific chromosomal recombi-
nation, such as the Mosaic Analysis with Double Markers (MADM)
or Mosaic Analysis with Repressible Cell Marker (MARCM) [18].
Finally, the Brainbow and derived approaches allow the individual-
ization of every neuron within a complex population by the ran-
dom expression of different combinations of spectral variant of
fluorescent proteins [19, 20]. However, all these methods require
the generation and maintenance of stable transgenic lines, with a
consequent increment of the time and costs of every study.
Here we detail a simple, fast, and inexpensive strategy for
dynamic neuroanatomy in the zebrafish [10]. We use the afferent
neuronal pathway of the mechanosensory lateral line as an example
of its implementation [10, 15, 21, 22]. This protocol is based on
transient expression of engineered transgenes coding for fluores-
cent proteins in single neurons by DNA injection in the fertilized
egg. Our procedure only requires a small zebrafish colony, a basic
knowledge of molecular biology, an access to a laser-scanning
confocal microscope for image acquisition, and finally, a desktop
computer equipped with free or inexpensive commercially available
software packages for image analysis and rendering. A single change
in the mounting of the samples would allow imaging by Selective
Plane Illumination Microscopy (SPIM) [23–25].

2 Materials

2.1 Embryos 1. E3 embryo medium stock solution (60×): Dissolve 172 g

Obtainment and NaCl, 7.6 g KCl, 29 g CaCl2·2H2O, and 49 g MgSO4·7H2O
Injections Components in 10 L of distilled H2O (dH2O).
2. Necessary equipment for raising fish and collecting eggs.
 Dynamic Neuroanatomy in Zebrafish 189

3. Necessary equipment for fish eggs injections: Glass micro-

capillary, micropipette puller, microinjector using air pressure,
and micropipette holder.
4. Necessary equipment to clone DNA and DNA purification kit.

2.2 Mounting 1. Tricaine (MS222 or MESAB: 3-aminobenzoic acid ethyl ester)

Embryos Components stock solution (25×): 0.4 g tricaine, 97.9 mL dH2O. Adjust
the pH to ~7 by adding 9.1 mL of 1 M Tris–HCl (pH 9).
Store at 4 °C.
2. 1 % LMP (low-melting-point) agarose in E3 to mount embryos
for observation.
3. Cover-glass-bottomed culture dishes.

2.3 Observation 1. Fluorescence stereomicroscope equipped with a filter set for

and Analysis of GFP and RFP.
Neuron Behaviors 2. Confocal microscope equipped with lasers for excitation at
488 nm and 532 nm.
3. Imaris software (Bitplane).

3 Methods

3.1 Expressing 1. Prepare DNA for injection (see Fig. 1). To achieve mosaic
Membrane Red expression in single neurons, clone the cDNA encoding the
Fluorescent Protein in red fluorescent protein (e.g., mCherry or tdTomato) under
the Zebrafish Lateral the control of a neuronal promoter (e.g., HuC). For a better
Line Afferent Neurons visualization of neurites, the red fluorescent protein has to be
fused to a membrane-targeting sequence (e.g., CAAX domain
or first 20 codons of GAP43).

HuC:mem-TdTomato DNA

Brn3c:memEGFP
embryo

Fig. 1 Scheme of the injection and selection of the samples to image. Overview
of the injection strategy to label single lateralis neurons. Injection of cDNA coding
with HuC:mem-TdTomato into 1–2-cell Tg[Brn3c:mem-eGFP] host embryos
190 Adèle Faucherre and Hernán López-Schier

2. Purify DNA using a plasmid purification kit following the

manufacturer’s instructions.
3. One day before carrying out injections, set the male and female
fish as a pair. Use transgenic fish expression the GFP in the hair
cells of the lateral line (e.g., sqET4 or Brn3c:mem-eGFP
transgenic lines).
4. For transient expression in single afferent neurons, inject 20 ng
of circular DNA (see Note 1). Dilution of the purified plasmid
is carried out in milliQ water.
5. Raise the injected embryos at 28.5 °C in an incubator to obtain
a standard developmental speed until the desired stage.
6. Screen and select embryos with a stereomicroscope under ultra-
violet light and select those expressing red fluorescence in afferent
neurons and GFP in the hair cells for live imaging. The exact time
for the start of expression after injection of DNA depends on
the promoter/enhancer used in the expression vector.

3.2 Imaging Afferent 1. Transfer the embryo directly to a cover-glass-bottomed culture

Arborization dish in a drop of 1 % low-melting-point agarose in E3 medium
with tricaine and carefully rotate the embryo with a hair loop
so that the desired side of the embryo faces the bottom.
2. Wait for a few minutes. Carefully fill the dish with E3 medium
and tricaine.
3. Use excitation light from a mercury lamp passed through
the filter set to observe red fluorescence. Select a neuromast
innervated by a single red afferent neuron (see Notes 2–4).
4. Image embryos with a confocal microscope with a 40× oil
immersion objective or with a 20× dry objective. Z-stacks are
acquired at 0.8 or 0.5 mm intervals, imaging GFP (488 nm
excitation, 500–550 nm emission), and TdTomato (532 nm
excitation, 570–630 nm emission).
5. Generate three-dimensional images through the entire extent
of the neuromast and the neuronal arbor over time for a period
of 1 h and every 10 min (6 time points).

3.3 Neuronal-Arbor 1. Image stacks are rendered three-dimensional using the Imaris
Tracing (See Fig. 2) software. Under the file menu, chose open and select the
desired file. In order to accelerate the analysis process, resample
the dataset before loading it (see Note 5).

Fig. 2 (continued) A and B, respectively, after 10 min. The dendrite C is no more present. (e, f) Example of
tracing of the whole neuronal arbor at T1 (e) and T2 (f). (g) Example of data for four-dimensional quantification.
The dendrite length is given by the software in micrometers. To calculate the growth or retraction of a neurite,
values obtained at T2 are subtracted from the values obtained at T1 for each neurite. In this case, the dendrite
A has extended 4.1 μm whereas the dendrite B has retracted 2.45 μm. In the case of the dendrite C, which
was present at T1 but absent at T2, the distance of retraction is calculated from the branching point, here Y
 Dynamic Neuroanatomy in Zebrafish 191

T1 T2
a b

c d

X X

Y
C

B B’
A
A’

e f
Begining point

Branching point

End point

g dendrite length A-X B-X C-X C-Y

T1 21.79 28.50 11.23
T2 25.89 26.05 7.99
T2-T1 4.10 -2.45 -3.24
event growth retraction retraction
stability stable stable transient

Fig. 2 Neuronal-arbor tracing. (a, b) Neuronal arbor at time points T1 and T2. (c, d) Example of dendrite tracing.
X represents the beginning point. A, A′; B, B′; and C represent 3 end points and Y represents a branching point.
In this case, at T1 (c), 3 dendrites have been traced. At T2 (d), A′ and B′ represent the position of the dendrites
192 Adèle Faucherre and Hernán López-Schier

2. Trace the neuronal arbor using the FilamentTracer Imaris

module. Open the Filament wizard and select the Autopath
(no loops) algorithm. Select the starting point manually at the
first bifurcation of the afferent neuron and assign it as Dendrite
Beginning Point in the Process Selection box of the Edit tab.
End points are also manually selected and filaments automati-
cally traced. Check the final tracing manually, and if needed,
correct the path using the object menu. Repeat the process for
every time point (see Note 6).
3. In order to visualize the different positions of a specific tip
over time (four-dimensional rendering), chose Mouse selects
point in the Edit tab and select a terminal point. In the
Statistics tab, press Duplicate Selection to New Filament and
attribute one different color in the Color tab to each terminal
point. Repeat for each time point, keeping the same color for
the same neurite.

3.4 Four- 1. Select a terminal point. In the Statistics menu, display the
Dimensional Selection values and chose Pt distance (i.e., dendrite length)
Quantification (see Note 7). Copy the value of Pt distance and paste it in a
of Axonal Arbors new Excel file.
(See Fig. 3) 2. For the same neurite, follow the terminal point from one time
point to the following one and repeat step 1.
3. In the case of transient neurites, select the branching point and
repeat step 15.
4. To express the Neurites stability, calculate the percentage of
stable neurites (present throughout all the movie) versus the
percentage of transient neurites (neurites visible during less
than 6 time points).
5. Calculate the Neurites persistence by counting the percentage
of neurites present for only one, two, three, four, five, or six
time frames.
6. To calculate growth and retraction, subtract the position value
of a neurite at a specific time point of the value of the previous
time point. Positive values correspond to growth and negative
values to retraction. Decide of an arbitrarily absolute value that
reflects significant growth or retraction movements (in our
study, the arbitrarily absolute value reflecting significant move-
ment is 1.5 μm) (see Note 8).
7. Calculate the number of neurites that have achieved a defined
number of events. Neurites consistency is defined as the number
of events (growth or retraction) observed for each neurite.
“Consistent” corresponds to a neurite that has neither grown
nor retracted over the 6 time frames (i.e., no events). “One
event” corresponds to a neurite that has only grown or only
retracted over the 6 time frames. “Two events” correspond to a
 Dynamic Neuroanatomy in Zebrafish 193

Fig. 3 Four-dimensional representation. (a, b) One color is attributed to each neurite and end points are duplicated
in order to visualize the positions of a specific tip over time. At T1 (a), the end points of dendrites A, B, and C have
been plotted. At T2 (b), A′ and B′ indicate the positions of the tip of the neurites A and B, respectively, after 10 min.
(c) Four-dimensional representation of the whole arbor after 1 h (six times 10 min). (d) Same as in (c) with the
maximal projection of the mem-tdTomato-expressing neuron and GFP-expressing hair cells

neurite that completed a sequence of two countermovements,

namely, growing followed by retraction or the opposite.
Similarly, “three events” represent a sequence of three coun-
termovements (e.g., growth followed by retraction followed
by growth or vice versa).
8. To calculate the Neurites presence, go to the Statistics menu,
display the Overall values and select No. Dendrite Terminal
Pts. Repeat for each time

4 Notes

1. The amount of injected circular DNA may depend on the

construct itself. It is recommended to perform a range of injec-
tions with different amount of DNA prior to the experiment in
order to determine the right amount of DNA for expression of
the fluorescent protein in single neurons.
194 Adèle Faucherre and Hernán López-Schier

2. Select for analysis only those specimens with single red fluores-
cent neurons whose soma is localized within the lateralis affer-
ent ganglion, which can be marked using the HGn39D
transgenic line.
3. If more than one afferent neuron are expressing the red fluores-
cent protein, you can try to inject less DNA. Nevertheless, if
the neuromast is innervated by only one neuron and if the soma
of this neuron can be localized within the lateralis afferent
ganglion, the experiment can be carried away.
4. If none of the afferent neurons is expressing the red fluorescent
protein, you can try to inject more DNA. Check as well that
you are using a promoter/enhancer allowing expression in the
afferent neurons.
5. If the sample has moved during the time of the acquisition,
align the three-dimensional reconstructions of each time point
using the Spot/Correct drift function of Imaris.
6. It is very important to assign the starting point as Dendrite
Beginning Point in the Edit menu since every following calcu-
lation is based on the position of this point.
7. If, in the statistic menu of Filament tracer, there is no value for
the dendrite length, check that the starting point has been
assigned as the Dendrite Beginning Point. Check as well that
there is no discontinuity in the tracing of the neuronal arbor
reconstruction.
8. In the case of transient neurites, the value before appearance or
after disappearance is the distance between the beginning point
and the branching point.

Acknowledgements

We thank the generosity of C.B. Chien for the Tol2kit and advice
and dedicate this article to his memory. We also thank T. Zimmermann
and J. Swoger for advice on, respectively, confocal and SPIM micros-
copy. The original research that encouraged the development of
this methodology was supported by a grant from the European
Research Council (ERC-2007-StG SENSORINEURAL) and by
the Ministerio de Ciencia e Innovación of Spain to H.L. S.A.F. was
supported by an Intra-European Marie Curie Fellowship from the
European Union.

References

1. Arber S (2012) Motor circuits in action: speci- 2. Seelig JD, Jayaraman V (2011) Studying senso-
fication, connectivity, and function. Neuron rimotor processing with physiology in behaving
74:975–989 Drosophila. Int Rev Neurobiol 99:169–189
 Dynamic Neuroanatomy in Zebrafish 195

3. Koch M (1999) The neurobiology of startle. 15. Pujol-Martí J, Baudoin JP, Faucherre A et al
Prog Neurobiol 59:107–128 (2010) Progressive neurogenesis defines late-
4. Piggott BJ, Liu J et al (2011) The neural circuits ralis somatotopy. Dev Dyn 239:1919–1930
and synaptic mechanisms underlying motor ini- 16. Collins RT, Linker C, Lewis J (2010) MAZe: a
tiation in C. elegans. Cell 147:922–933 tool for mosaic analysis of gene function in
5. Anikeeva P, Andalman AS et al (2011) zebrafish. Nat Methods 7:219–223
Optetrode: a multichannel readout for optoge- 17. Hans S, Kaslin J, Freudenreich D et al (2009)
netic control in freely moving mice. Nat Temporally-controlled site-specific recombina-
Neurosci 15:163–170 tion in zebrafish. PLoS One 4:e4640
6. Ahrens MB, Li JM et al (2012) Brain-wide 18. Luo L (2007) Single-neuron labeling using the
neuronal dynamics during motor adaptation in genetic MARCM method. CSH Protoc 2007:
zebrafish. Nature 485:471–477 pdb.prot4789
7. Moser EI, Kropff E, Moser MB (2008) Place 19. Livet J, Weissman TA, Kang H et al (2007)
cells, grid cells, and the brain’s spatial represen- Transgenic strategies for combinatorial expres-
tation system. Annu Rev Neurosci 31:69–89 sion of fluorescent proteins in the nervous sys-
8. Gahtan E, Baier H (2004) Of lasers, mutants, tem. Nature 450:56–62
and see-through brains: functional neuroanat- 20. Pan YA, Livet J, Sanes JR et al (2011)
omy in zebrafish. J Neurobiol 59:147–161 Multicolor brainbow imaging in zebrafish.
9. Friedrich RW, Jacobson GA, Zhu P (2010) CSH Protoc 2011(1):pdb.prot5546
Circuit neuroscience in zebrafish. Curr Biol 21. Raible DW, Kruse GJ (2000) Organization of
20:R371–R381 the lateral line system in embryonic zebrafish.
10. Faucherre A, Baudoin JP, Pujol-Martí J et al J Comp Neurol 421:189–198
(2010) Multispectral four-dimensional imag- 22. Metcalfe WK, Kimmel CB, Schabtach E (1985)
ing reveals that evoked activity modulates Anatomy of the posterior lateral line system in
peripheral arborization and the selection of young larvae of the zebrafish. J Comp Neurol
plane-polarized targets by sensory neurons. 233:377–389
Development 137:1635–1643 23. Swoger J, Muzzopappa M, López-Schier H
11. Simmich J, Staykov E, Scott E (2012) Zebrafish et al (2011) 4D retrospective lineage tracing
as an appealing model for optogenetic studies. using SPIM for zebrafish organogenesis studies.
Prog Brain Res 196:145–162 J Biophotonics 4:122–134
12. Santoriello C, Zon LI (2012) Hooked! 24. Kaufmann A, Mickoleit M, Weber M et al
Modeling human disease in zebrafish. J Clin (2012) Multilayer mounting enables long-
Invest 122:2337–2343 term imaging of zebrafish development in a
13. Shu X, Lev-Ram V, Olson ES et al (2011) Spiers light sheet microscope. Development 139:
Memorial lecture. Breeding and building molec- 3242–3247
ular spies. Faraday Discuss 149:63–77 25. Huisken J, Stainier DY (2009) Selective plane
14. Detrich HW 3rd (2008) Fluorescent proteins illumination microscopy techniques in devel-
in zebrafish cell and developmental biology. opmental biology. Development 136:
Methods Cell Biol 85:219–241 1963–1975
 Chapter 14

Anatomical Dissection of Zebrafish Brain Development

Katherine J. Turner, Thomas G. Bracewell, and Thomas A. Hawkins

Abstract
Zebrafishbrain.org is an online neuroanatomical atlas of the embryonic zebrafish. The atlas uses high-
resolution confocal images and movies of transgenic lines to describe different brain structures. This chapter
covers detail of materials and protocols that we employ to generate data for the atlas.

Key words Zebrafish, Neuroanatomy, Brain atlas, Brain dissection, Labeling

1 Introduction

Increasingly zebrafish are becoming a model system for the study

of behavior [1–5]. To truly understand zebrafish behavior, we
must first understand the connectivity of the neuronal circuits driv-
ing behavior. This requires detailed neuroanatomical characteriza-
tion of the zebrafish brain. Zebrafish are a relatively new model
system; as such, the zebrafish does not benefit from the extensive
neuroanatomical descriptions that have been undertaken in other
model species [6, 7]. Most neuroanatomical studies in the zebrafish
and other teleosts focus on adult neuroanatomy [8, 9]. These
studies mainly use serial sections of adult brains which can be chal-
lenging to interpret by non-experts and can be difficult to apply to
the embryonic brain [10]. Specific description of the neuroanat-
omy of the embryonic and larval zebrafish brain is not comprehen-
sive. To address this issue, in collaboration with other laboratories,
we are in the process of building an online neuroanatomical
resource (an atlas) called zebrafishbrain.org.
Zebrafishbrain.org is being built to communicate current knowl-
edge about the neuroanatomy of the developing zebrafish brain, and
this is achieved in two ways. The first is to provide a hub for commu-
nity experts to provide data and write descriptions of neuroanatomi-
cal structures, and the second is to provide data and descriptions
ourselves. To do this we are mining currently available collections of
transgenic zebrafish (and making a few of our own) to generate a

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_14, © Springer Science+Business Media, LLC 2014

197
198 Katherine J. Turner et al.

collection of transgenic lines with thoroughly described embryonic/

larval (and sometimes adult) brain-specific expression patterns. We
employ the descriptions of the expression patterns of these lines in the
construction of tutorials describing brain structures.
This chapter details the core methods that are currently
employed by us when studying transgenic zebrafish embryos and
larvae for zebrafishbrain.org. The protocols describe in detail how
we undertake brain dissection, immunohistochemistry, and the
several available methods we employ to mount embryos for confo-
cal microscopy. We also list software we have found useful for pro-
cessing the raw data.
The protocols are based upon methods that are widely used in
the zebrafish field but with a few adjustments to adapt to charac-
terization of zebrafish developmental neuroanatomy. We also spec-
ify some reagents that we have found to be reliable for use in the
zebrafish. Many of these protocols have flexible aspects to them,
and we have tried to include detail of this flexibility where possible
as an indication of how optimization for particular conditions can
be undertaken.

2 Materials

2.1 Equipment 1. Petri dishes.

for Embryonic Culture 2. Watchmakers forceps for dechorionation: Dumont #5 (Fine
and Fixation Science Tools).
3. 7 ml bijou tubes: Appleton Woods.

2.2 Embryonic 1. Fish/embryo water (filter-sterilized aquarium system water) or

Culture and Fixation embryo medium (zebrafish book).
Reagents 2. 1-Phenyl 2-thiourea (PTU) 3 mg/ml: Keep frozen 40 ml ali-
quots of 25× stock solution (75 mg/ml). One aliquot makes
1 l of PTU when mixed with fish water. Fresh PTU should be
made every 2 days.
3. Sweet fix: 4 % PFA with 4 % sucrose. Make from 20 % PFA stock:
40 ml H2O + 10 g paraformaldehyde. Heat to 60 °C with stir-
ring. Add 10 drops of 10 M NaOH. Cool and filter through
funnel with filter paper, add 10× phosphate-buffered saline
(PBS) (pH 7.3), sucrose, and H2O and adjust pH with HCl to
give a final concentration of 4 % paraformaldehyde, 4 % sucrose,
and 1× PBS at pH 7.3. Aliquot into 5 ml aliquots and freeze.
Defrost fresh fix at room temperature (RT) just prior to fixation.
Do not use fix that has been defrosted for longer than 48 h.
4. 2 % TCA in PBS. 10 % TCA (in H2O) stock can be stored at
−20 °C. Defrost and mix with 10× PBS and H2O to make final
concentration of 2 % in 1× PBS.
5. PBS: phosphate-buffered saline pH 7.3.
 Anatomical Dissection of Zebrafish Brain Development 199

2.3 Brain Dissection 1. Glass petri dish.

Equipment 2. Bunsen burner.
3. Oven (60 °C).
4. Minute pins A1.
5. Superfine Dumont forceps (Fine Science Tools).
6. Needle holders (Fine Science Tools or VWR)
7. Tungsten wire 0.125 mm.
8. 12v DC power supply.
9. Two wires with crocodile clips.
10. Glass jar with lid.
11. Electrode.
12. Microcentrifuge tubes.

2.4 Brain Dissection 1. Sylgard: 1 kit 184 from Dow Corning WPI Sylgard http://
Reagents and www.dowcorning.com/applications/search/products/
Solutions Details.aspx?prod=01064291&type=&country=GBR.
2. Saturated NaOH.
3. PBS.

2.5 Brain Dissection 1. Sylgard is an elastomer curing agent. The Sylgard kit contains
Sylgard Dishes two solutions that set to a hard rubberlike texture when mixed.
A layer of Sylgard in a glass petri dish makes a great dissecting
dish as you can pin the embryos using insect/minute pins to
the Sylgard, immobilizing them while you dissect. You also
minimize damage to forceps and dissecting needles.
2. Mix Part A liquid from the Sylgard 184 kit with Part B liquid
in a ration of 10:1 in a beaker. You will need 40–50 ml per
petri dish.
3. Thoroughly mix the two solutions together using a disposable
stirrer such as a tongue depressor. Be careful not to stir too vig-
orously so as not to create unwanted bubbles in the mixture.
4. After mixing, pour the Sylgard slowly into the glass petri dishes
laid on a flat, non-vibrating surface where they can remain undis-
turbed for several days. Fill each petri dish to 1/3–1/2 full.
5. After the Sylgard has settled for a few minutes, you will see that
bubbles have risen to the surface of the liquid. Burst these
using a Bunsen burner by passing the flame lightly close to the
surface of the Sylgard.
6. Put the petri dish lids on top to prevent dust settling on the
plates while they set.
7. Sylgard dishes need at least 24 h to set and improve if left for
several days. It is possible to speed up this process by heat
curing them at 60 °C for several hours in a mini oven or
incubator.
200 Katherine J. Turner et al.

Fig. 1 Sharpening dissection needles using an electric current and saturated

sodium hydroxide(NaOH). (a) Connect an electrode to the negative terminal of a
12v DC power supply. Connect your dissection needle to the positive terminal. (b)
Submerge the negative electrode into the saturated NaOH. (c) Dip the tip of the
tungsten wire into the NaOH and slowly draw back up out of the solution. Repeat
this motion several times until you have a very sharp point. When making a new
dissecting needle, you will have to repeat the dipping action several times. When
you are sharpening a dissecting needle that has already been sharpened before,
you only need to dip the needle two or three times

2.6 Brain Dissection To remove the skin and eyes from zebrafish embryos, we use
Needles dissecting needles fashioned from fine tungsten wire. The
tungsten wire is clamped into a needle holder and cut using
strong scissors so it protrudes from the tip of the holder by about
1 cm. The wire at this point is blunt but can be sharpened to a
very fine sharp point using an electric current and saturated
sodium hydroxide (Fig. 1).

2.7 Immunohisto- 1. 1.5 ml microcentrifuge tubes.

chemistry Equipment 2. Plastic fine tip Pasteur pipettes or aspirator.
3. 50 ml Falcon tubes.
 Anatomical Dissection of Zebrafish Brain Development 201

2.8 Immunohisto- 1. PBS pH 7.3 + 0.5–0.8 % Triton-X100 (PBTr).

chemistry Reagents 2. 50 % MeOH in PBS.
and Solutions
3. 100 % MeOH.
4. Proteinase K(PK) 10 mg/ml (this is a 1,000× stock) store in
aliquots at −20 °C.
5. Trypsin (0.25 % in PBS). Trypsin stock is 2.5 % (10×).
6. Immunohistochemistry blocking solution (IB): For 1 ml of IB,
100μl normal goat serum (NGS), 10μl of DMSO, 0.89 ml
PBTr.
7. Anti-GFP: polyclonal rabbit α-GFP from AMS Biotechnology
gives great results. Use 1:1,000.
8. Anti-acetylated tubulin antibody: mouse monoclonal (IgG2b)
from Sigma. Use 1:500 [11].
9. Anti-SV2: mouse monoclonal (IgG1) DSHB. Use 1:500
[12, 13].
10. Anti-GFP(Rat): rat monoclonal (IgG2a) from Nacalai Tesque.
Use 1:1,000.
11. Chk pAb to GFP: chicken polyclonal from Abcam. Use
1:1,000 [15].
12. Anti-RFP: rabbit polyclonal from MBL. Use 1:2,500 [16].
13. Anti-DsRed (Living Colors): rabbit polyclonal from Clontech.
Use 1:300 [14].
14. Anti-Kaede (M-125-3): mouse monoclonal from MBL. Use
1:200 (does not distinguish between red and green forms of
protein).
15. Anti-Kaede (PM 102): rabbit polyclonal from MBL. Use
1:1,000 (does not distinguish between red and green forms of
protein).
16. Secondary antibody depends upon primary antibody and
detection method: For fluorescence, Molecular Probes (www.
probes.com) Alexa Fluor 488 goat α-rabbit (highly cross-
adsorbed) IgG. Use at 1:200. To detect anti-acetylated tubulin
and anti-SV2 in the same sample, use isotype-specific second-
aries: Alexa Fluor 568(IgG2b) and Alexa Fluor 633(IgG1).
Use at 1:200. Molecular Probes also make fluorescent second-
aries against rat and chicken primary antibodies conjugated
with various fluorescent dyes.

2.9 Confocal 1. Silicone grease: Fisher Scientific.

Mounting and 2. Glass rings for mounting.
Imaging Equipment
3. Disposable Pasteur pipettes, glass, short tip.
4. Araldite Rapid: Fisher Scientific.
5. Rubber bulbs for pipettes 1 ml (VWR).
202 Katherine J. Turner et al.

6. Small coverslips 22 × 22 (VWR).

7. Microscope slides (VWR).
8. Heat block set to 40 °C.
9. Microwave.
10. Beaker with microcentrifuge tube plastic stand to melt
agarose.
11. Dissecting microscope.
12. Confocal microscope with appropriate laser lines.
13. Long working distance objective lens. Highest possible numer-
ical aperture (NA). Approximately 500 μm (±250) working
distance (WD) is required. For example, Leica HCX IRAPO L
25×/0.95 W is a water immersion lens with working distance
of 2.4–2.5 mm, and coverglass and non-coverglass corrected
versions are available.

2.10 Confocal 1. Fish water without methylene blue.

Mounting and 2. PBS.
Imaging Reagents
3. Low gelling temperature agarose (Sigma): 1 ml aliquots of 1 %
and Solutions
agarose in filter-sterilized fish water (with no methylene blue)
or embryo medium (E3) for live time-lapse imaging. For fixed
embryos use 1 % agarose in PBS or 1 % regular (not low melt)
agarose in PBS/80 % glycerol. The latter solution can be chal-
lenging to make. Use a water bath and stirrer to make a 5 %
agarose solution, and then add 4× volume of glycerol.
4. Tricaine:(3-amino benzoic acid ethyl ester) (Sigma) 400 mg
tricaine powder, 97.9 ml DD H2O, ~2.1 ml 1 M Tris (pH 9).
Adjust pH to ~7. Freeze this solution into aliquots and use
4.2 ml per 100 ml of fish water or E3.
5. CyGEL Sustain (500 μl) (BioStatus Limited).
6. 60× E3 embryo medium: NaCl 3 M, KCl 0.1 M, CaCl 0.2 M,
MgSO4 0.2 M dissolved in deionized water.
7. Ice pack or ice bucket.

2.11 Image 1. Volocity (Perkin Elmer).

Processing Software 2. XuvTools.
3. ImageJ/Fiji.

3 Methods

3.1 Embryonic 1. Embryos should be collected and raised at 28.5 °C in fish

Culture and Fixation water or embryo medium without methylene blue. This
medium minimises the auto-fluorescence of the skin caused by
 Anatomical Dissection of Zebrafish Brain Development 203

methylene blue that is not desirable with fluorescent imaging

methods.
2. At 24 hpf transfer embryos to fish water containing PTU again
without methylene blue. PTU inhibits pigmentation of the
embryos.
3. Defrost sweet fix just prior to fixation at room temperature.
Embryos should be dechorionated using forceps prior to fixa-
tion. Fix embryos in large volumes of sweet fix. Transfer
embryos to 7 ml bijou tube; remove as much fish water as pos-
sible before adding sweet fix. Up to 200 embryos can be fixed
per 5 ml of sweet fix (see Note 1).
4. Fix embryos at 25 °C (employ an incubator if necessary). 1 h
of fixation per day of development, e.g., for 24 hpf embryos fix
for 1 h at 25 °C.
5. After fixation, remove most but not all sweet fix and top up
tube with PBS (do not use any detergent (e.g., triton) at this
stage).
6. Leave embryos overnight at 4 °C before starting dissection.
Embryos can be stored for several days at 4 °C before dissec-
tion but will deteriorate in that time so prompt dissection is
advised. If the embryos are not going to be dissected prior to
performing the immunohistochemistry, then dehydrate the
embryos using these steps.
7. Transfer to 1.5 ml tubes, rinse 2×, and wash 3 × 10 min in
PBTr (phosphate-buffered saline + 0.5–0.8 % Triton-X100) on
side on shaker.
8. Transfer to MeOH—1 × 5 min wash in 50 % MeOH/50 %
PBTr and 3 × 5 min in 100 % MeOH. NB Mixing PBTr and
MeOH is exothermic—this solution must be made up at least
15 min before use.
9. Store at −20 °C for at least 6 h (or up to 6 months+) to
delipidate.
10. For most antibodies formaldehyde fixation with sweet fix gives
the best results; however, some antibodies work better with
TCA fixation. This is often true of older embryos (5dpf+). For
TCA fixation, simply replace 2 % TCA for sweet fix in the pro-
tocol above. TCA-fixed embryos should be permeabilized
using trypsin (see below); TCA-fixed embryos are also more
opaque and will benefit from clearing in glycerol at the end of
the protocol (see below).

3.2 Brain Dissection Dissecting the skin and eyes off of zebrafish embryos prior to
antibody staining vastly improves the penetration of antibodies and
reduces auto-fluorescence of the skin common to many transgenic
lines. Although this technique requires some dexterity, with practice
it becomes routine and the results justify the effort (Fig. 2).
204 Katherine J. Turner et al.

Fig. 2 Dissection. (a) Pin embryo laterally to a Sylgard dish. (b) Remove eye using a sharpened tungsten nee-
dle. (c) Remove pins and flip the embryo so the other side now faces upwards. (d) Remove second eye using
sharpened tungsten needle. (e) Cut away the yolk and jaw from the ventral surface of the embryo. (f) At the
level of the otic vesicle using either a sharpened tungsten needle or fine forceps, grip a piece of skin and tease
away from the brain. (g) Pull this flap of skin rostrally, removing all of the skin from the surface of the brain. (h)
Close up showing completely dissected brain ready to be transferred into methanol for immunohistochemistry
or in situ hybridization

1. Under a dissecting microscope, transfer embryo to Sylgard

dish.
2. Using forceps (a low-grade type), pin the embryo with minute
pins. Embryo should be pinned on its side (laterally). Pin the
embryo twice through the tail at the level of the notochord.
Push down on the pins so they stick into the Sylgard.
3. Using a dissecting needle, gently cut the skin surrounding the
eye and tease the eye off the head. The eye should pop off easily
at stages post-48 hpf but is trickier to remove at earlier stages.
4. Unpin embryo, taking care not to break the tail. Turn the
embryo to lie on the opposite side and re-pin to the Sylgard.
5. Remove second eye using the same procedure as above.
6. Cut skin between anterior yolk and heart/head using dissect-
ing needle. Then cut away the skin attaching the yolk sack to
the body. The yolk should come off as an intact structure.
7. Remove the jaw and other connective tissue from the ventral
surface of the brain.
8. Make a shallow cut at the level of the otic vesicle. Using fine
forceps tease a flap of skin off and pull this flap rostrally remov-
ing all of the skin from the surface of the brain.
9. Using fine forceps (see Note 2) or a dissecting needle, pull off
any remaining skin, leaving just the brain attached to the tail.
10. Unpin the embryo and transfer to a 1.5 ml tube filled with
PBTr using normal forceps.
 Anatomical Dissection of Zebrafish Brain Development 205

11. After all embryos are dissected and transferred to 1.5 ml tube.
Wash several times with PBTr, then transfer to MeOH (see
Subheading 3.1), and store at −20 °C until you want to start
the immunohistochemistry (see Note 3).

3.3 Fluorescent 1. Rehydrate embryos: Wash embryos 5 min in 50 % MeOH/50 %

Immunohisto- PBTr, and wash 3 × 5 min in PBTr.
chemistry Protocol 2. Permeabilize embryos (PFA-fixed embryos): Proteinase K (pK)
digestion times vary with embryo age, and also pK batches vary.
The times below are a guide, and tests should be carried out with
the particular batch employed (store pK at −20 °C at 10 mg/
ml—this is 1,000× stock). Dilute pK stock 1/1,000 (1×) in PBTr.
● Up to tailbud, no PK
● 2–10ss, in and out of 1× PK
● 10–15ss, 1 min 1× PK
● 16–26ss, 2 min 1× PK
● 24 h, 10–15 min 1× PK
● 30 h, 20 min 1× PK
● 36–48 h, 30–40 min 1× PK
● 2.5d, 30–40 min 1.5× PK
● 3d, 30–40 min 2× PK
● 4d, 30–40 min 3× PK
● 5d, 30–40 min 4× PK.
Digestions are at room temperature (18–22 °C) with tube
lying on its side. Dissected embryos should be treated as
somite-stage embryos (minimal permeabilization). Rinse 3× in
PBTr. Postfix in 4 % PFA for 20 min at room temperature (also
denatures pK). Wash 3 × 5min in PBTr.
3. For TCA-fixed embryos: Rinse embryos 3 × 5 min in PBS. Prechill
trypsin solution (0.25 % in PBS) and 5 ml per sample of PBTr on
ice until cold. Incubate embryos in trypsin, on ice for 5–10 min
(according to age 36 h to 5 days), longer for older embryos,
depending on trypsin batch; titrate upon first use. Rinse 2× in
cold PBTr then 3 × 10 min in cold PBTr, bring to RT.
4. Block endogenous binding sites: Incubate in IB for at least 1 h
at room temperature on shaker.
5. Primary antibody incubation: Incubate in IB + primary antibody
overnight at 4 °C on shaker. Some antibodies work better with
longer incubations or room temperature incubations, and for
room temperature incubation, consider adding 2 mM sodium
azide to the IB to inhibit mold growth. For longer incubations
increase the number and length of post-incubation washes.
206 Katherine J. Turner et al.

6. Postprimary incubation washes: Remove primary antibody

(can be kept at 4 °C for reuse within a week). Rinse 3× in
PBTr. Wash at least 4 × 30 min in PBTr on shaker, can be lon-
ger/more washes.
7. Secondary antibody incubation: Incubate in IB + secondary
antibody overnight on shaker at 4 °C or room temperature.
8. Postsecondary incubation washes: Rinse 3× in PBTr. Wash at
least 4 × 30 min on shaker. Fluorescent-stained embryos are
now ready to be imaged. Transfer either to 80 % glycerol
(through 25 % and 50 % glycerol/PBS solutions) and mount.
Alternatively, keep in PBTr and mount in agarose for imaging.
Keep at 4 °C in the dark and image as soon as possible.

3.4 Antibodies Zebrafishbrain.org focuses mainly on the characterization of

for Neuroanatomy transgenic zebrafish lines. Many hundreds of transgenic and
enhancer trap lines have been created by the zebrafish community
and can be generated easily using protocols provided elsewhere. To
generate data for zebrafishbrain.org, high-resolution confocal

Fig. 3 Transgenic zebrafish embryo labelled with anti-GFP (green ), anti-SV2

(blue ), and anti-acetylated tubulin (red ). This is a dorsal view of a 4dpf Tg(dlx
4/6:GFP) embryo. In this image we can see GFP-positive cells in the telencepha-
lon, optic tectum, and the cerebellum. The SV2 staining labels the synaptic neu-
ropil and axonal tracts are visualized by labeling with anti-acetylated tubulin
antibody. This is an example of the type of images we use to populate the zebrafishbrain.
org database (Image generated by Monica Folgueira)
 Anatomical Dissection of Zebrafish Brain Development 207

imaging is performed on suitable transgenics generated in-house

and externally that have fluorescent expression in specific brain
regions or nuclei (Fig. 3). With the majority of transgenic lines
expressing GFP, the primary antibody used most frequently is a
polyclonal rabbit anti-GFP from AMS Biotechnology (TP401).
This antibody has excellent penetration and works equally well in
whole-mount embryos at all stages and on sections.
To aid anatomical orientation transgenic specimens can also be
labelled with anti-acetylated tubulin (IgG2b, Sigma) and/or anti-SV2
(IgG1, DSHB). These antibodies label beautifully the axonal
connections and neuropil, respectively. In addition to these anti-
bodies being informative from a neuroanatomical perspective, they
are also invaluable as a tool for anatomical localization. These anti-
bodies can be used as a framework to easily compare the expression
patterns of different transgenic lines and locate GFP-positive struc-
ture in the context of the brain. Both antibodies are mouse mono-
clonals; fortunately, they are different subtypes and can be detected
in the same specimen using subtype-specific secondary antibodies.
We have trialed many neurotransmitter antibodies in
whole-mount zebrafish embryonic/larval preps with little success.
Colocalization to check which neurotransmitters a particular cell is
expressing in a transgenic normally needs to be done using immu-
nohistochemistry on cryosections. There is a protocol for this on
zebrafishbrain.org. Some exceptions of antibodies that have
worked well in whole-mount can also be seen there. Different fixa-
tion methods can improve the efficacy of some of these antibodies
for immunohistochemistry. Many antibodies that do not work
after PFA fixation will work better after fixation with TCA or other
fixatives such as glutaraldehyde.

3.5 Cell Dyes Using a nuclear label in conjunction with a fluorescent immuno-
for Neuroanatomy staining can be very useful to delineate brain nuclei, neuropil, and
ventricles through the tissue (Fig. 4). Nuclear staining has also
been employed by the Driever lab to produce a 3D reference brain
for their ViBE-Z software. They have also used acetylated tubulin
immunohistochemistry with their reference brain [17, 18].
For nuclear staining, use SYTOX Orange or Green at a con-
centration of 1:10,000 or Topro3 at a concentration of 1:5,000
depending upon the wavelengths required. These dyes can be
added with the secondary antibody incubation. Staining with
nuclear dyes works best following room temperature incubation in
the dye so it is advisable to add 2 mM azide with the secondary
antibody/nuclear dye IB mixture to inhibit mold formation. The
dyes bleach very easily so minimize exposure to light and keep the
laser intensity on the confocal as low as possible when imaging, and
also, timely imaging following the staining process produces the
best results.
208 Katherine J. Turner et al.

Fig. 4 Nuclear dyes. Nuclear dyes show brain nuclei and also expose morpho-
logical boundaries between brain areas. Neuropil is not marked and is revealed
as dark spaces. In this confocal slice through the forebrain of this enhancer trap
line, we can see expression of GFP in the olfactory epithelia (Image generated by
Tim Geach and TAH)

3.6 Confocal 1. Mounting media: The choice of mounting medium depends

Mounting with upon the experimental procedure that is being undertaken.
Glass Ring For live imaging (e.g., timelapse), embryos can be mounted in
agarose (made with fish water) or CyGEL Sustain (see
Subheading 3.9 below). For imaging of fixed preparations,
embryos can be mounted in agarose (made with PBS) or glyc-
erol agarose. Agarose in simple PBS leaves embryos slightly
opaque; depth penetration can be improved if the embryo is
mounted in glycerol agarose (as described in the protocol asso-
ciated with the recent Driever lab paper: [17, 18]). The meth-
ods of ring mounting described below can be achieved using
either aqueous or glycerol agarose. Alternatively, embryos can
be mounted between stacked coverslips. Aqueous agarose is
slightly preferable as a mountant when using water immersion
lenses, and glycerol immersion lenses should be employed
where available when mounting in glycerol agarose. Water
immersion lenses can produce acceptable results with glycerol
agarose as a mountant despite being an optically suboptimal
system.
2. Mounting in glass rings (Fig. 5): The advantage of this prepa-
ration is that the embryos are securely mounted in a large
volume of agarose that permits the orientation of the embryos
as is required for imaging. It can be achieved using either aque-
 Anatomical Dissection of Zebrafish Brain Development 209

Fig. 5 Mounting in glass rings. (a) On a dissecting microscope, place a 22 × 22 coverslip onto a slide. Smear
silicone grease on both sides of a glass ring, and press it onto the coverslip forming a watertight seal. Pipette
the embryo into the agarose and then onto the coverslip in the center of the glass ring. Orientate the embryo
so the surface of the embryo you wish to image is against the coverslip. (b) Fill the ring 2/3 full with agarose
and let it set. Once the agarose has set firm, pipette enough PBS onto the surface of the agarose to form a
convex meniscus over the top of the ring. Overfilling the ring with PBS means that no bubbles will be trapped
when you place the slide on top. (c) Take another slide and press down onto the top of the glass ring expelling
the excess PBS. Make sure the slide and the glass ring have formed a watertight seal with the silicone grease.
(d) The prep can now be inverted with the coverslip on top and slide on the bottom. Your embryo will now be
at the top of the prep just under the coverslip ready for imaging

ous or glycerol agarose as the mounting medium although

glycerol agarose should only be used for fixed preparations.
Before beginning this protocol, several tubes (as many as
required) of agarose should be melted in the microwave and
transferred to a heat block set at 45 °C.
3. Place a slide on the dissecting microscope.
4. Place a 22 × 22 coverslip onto the slide.
210 Katherine J. Turner et al.

5. Smear silicone grease on to the top and bottom of a glass ring,

and press it onto the coverslip to form a watertight seal.
6. Using a glass pipette, suck up the embryo to be mounted in a
small volume of medium, and pipette it into the molten
agarose, which should be around 40 °C (slightly cooled from
the hot block temperature). Refill the pipette with 0.5–1 ml of
agarose with the embryo. Expel the pipette contents (agarose
and embryo) onto the coverslip in the center of the glass ring.
There should be enough agarose to fill the glass ring around
two-thirds full.
7. Moving quickly, using forceps or other suitable tool, manipu-
late the embryo to place in the correct orientation for imaging.
Bear in mind that after step 7, the preparation will be inverted;
thus the side of the embryo to be imaged should be closest to
the coverslip; for example, if the embryo is to be imaged from
a dorsal aspect, the dorsal side of the embryo will be touching
the coverslip with the ventral side facing upwards at this point.
8. Once the agarose has set firm, pipette some PBS (see Note 5)
onto the surface of the agarose to form a convex meniscus over
the top of the ring (it is important to slightly overfill the ring
as this prevents bubbles getting trapped inside the ring once
the top slide is secured).
9. Take another slide and press down onto the top of the glass
ring starting from a slight angle, expelling the excess PBS.
Make sure the slide and the glass ring have formed a watertight
seal with the silicone grease. The prep can now be inverted
with the coverslip on top and slide on the bottom. Use a tissue
to remove the excess liquid. Your embryo will now be at the
top of the prep just under the coverslip, ready for imaging.

3.7 Confocal If glass rings are not available, a similar mounting method can be
Mounting Without used where a well of silicone grease is constructed directly onto a
Glass Ring slide (Fig. 6). A syringe filled with silicone grease should be used
to squeeze the grease onto the slide. The embryo is mounted on a
coverslip as above in a large drop of agarose (the diameter and
depth of this drop should not exceed the width and depth of the
silicone grease well). The silicone grease well is filled with PBS to
form a convex meniscus. The coverslip with the embryo attached
is inverted and pressed gently down into the well expelling the
excess PBS. Make sure the coverslip forms a watertight seal with
the top of the silicone grease well. The embryo is now ready for
imaging. This method also has the advantage of allowing for a
small adjustment in the orientation of the sample.

3.8 Mounting For imaging of live embryos (particularly for timelapse), embryos
of Live Embryos can be mounted in a large epoxy resin chamber or well. This allows
for Confocal Imaging for constant gaseous exchange and thus improves the health of the
embryo. These chambers are easily made using Araldite or similar
 Anatomical Dissection of Zebrafish Brain Development 211

Fig. 6 Mounting fixed embryos in silicone grease wells. (a) Using a syringe filled with silicone grease, pipe a
well onto a microscope slide. (b) On a coverslip orientate your embryo in a drop of 1 % agarose. The surface
of the embryo you wish to image should be touching the coverslip. Half fill the silicone grease well with PBS.
(c) Once the agarose is set, invert your coverslip so that the embryo is submerged in PBS inside the silicone
grease well. Lightly push down on the coverslip to create a sealed chamber

epoxy compounds. Mix the two components and make a rectangular

well, a few millimeters deep on a clean slide. Make sure the well is
continuous so the liquid won’t leak out. Allow the resin to
polymerize until hard before using. These large wells mean that it
is possible to mount several embryos on the same slide. This is
particularly useful for multiple timelapses on a confocal with a
motorized X/Y stage. Once made, these slides can also be reused
many times. Mounting protocol:
1. Anesthetize embryo(s) in a petri dish by adding tricaine (final
concentration 1.6 mg/l) to the fish water. Wait until the
embryo has completely stopped twitching before trying to
mount. Check that heart beat of the embryo is still strong
under the dissecting scope.
2. Place resin chamber slide onto the dissecting scope.
3. Remove tube of agarose from the heat block and allow to cool
down to below 37 °C to avoid heat-shocking the embryo.
4. Pipette the anesthetized embryo into the warm molten agarose
then refill the pipette with agarose and the embryo. Pipette
212 Katherine J. Turner et al.

embryo and agarose onto the slide with enough agarose to

form a small bubble around the embryo.
5. Orientate the embryo very gently using forceps or other appro-
priate implement. This prep will not be inverted so the side of
the embryo to be imaged should face upmost; it should also
not be too deep in the agarose.
6. Repeat this procedure for two or three other embryos per
slide. Do this as quickly as possible so the embryos mounted
earlier do not dry out.
7. When the agarose is set, flood the chamber with fish water
containing tricaine.
8. The embryos are now ready for imaging. Embryos mounted in
this way should be imaged using a non-coverslip corrected
water immersion lens. These lenses can be dipped straight into
the fish water as long as it does not contain methylene blue.
For the modifications required to adapt this method to live
embryos (see Note 4).

3.9 Mounting of Live An alternative to agarose for mounting live embryos is a compound
Embryos Using CyGEL called CyGEL Sustain. CyGEL is liquid at low temperatures and
Sustain changes from a sol to a gel at 23–24 °C. The sol–gel conversion
can be reversed, by simply placing the sample on ice for a few
seconds. This is particularly useful for short procedures that require
a fast and easy way of recovering the embryo undamaged after
manipulation, such as electroporation or Kaede photoconversion.
CyGEL Sustain is a compound commercialized by BioStatus
Limited (www.biostatus.com). They provide protocols on their
web page, but these are optimized for the growth of cells. Here we
provide a protocol adapted for zebrafish mounting:
1. Place vial of CyGEL Sustain on ice and make sure it is a sol.
2. Add E3 so that the final concentration will be 1× (8.4 μl of
60XE3 for 500 μl of CyGEL). Keep it on ice.
3. Prepare a chamber on a slide with silicone grease, or use a glass
ring as described in Subheading 3.6.
4. Warm the vial slightly using hands so that it is not too cold
(but not too much because it gels).
5. Pipette the anesthetized embryo into the CyGEL then refill
the pipette with the embryo and enough CyGEL to fill the sili-
cone chamber or glass ring.
6. Orientate embryo as required the consistency of CyGEL
quickly becomes dense, becoming a tight gel within minutes.
At that point, the embryo is ready for microscopy.
No additional liquid should be placed on the gel as it immedi-
ately becomes a sol again. If the embryo needs to be imaged using
 Anatomical Dissection of Zebrafish Brain Development 213

a water immersion lens for a long period, it should be mounted

upside down in a glass ring on a coverslip filled completely with
CyGEL and tightly sealed with a slide. This ensures that the
preparation stays moist for the length of the procedure and that no
extra liquid disrupts the gel. Tricaine should be added to the E3
medium (4.2 ml tricaine per 100 ml E3) used to prepare the
CyGEL, to ensure that the embryo remains anesthetized for the
length of the procedure.

3.10 Confocal Setup of the imaging system for optimal imaging depends very
Microscopy Setup much on the individual sample employed. High-NA, long working
distance objectives (water and/or glycerol immersion) are available
from most microscope manufacturers. These lenses are ideally
suited to imaging the zebrafish brain as they permit imaging
through the whole depth of the brain at high resolution. There are
many tradeoffs in the setup of the imaging parameters. For high-
quality three-dimensional reconstructions when imaging fixed,
antibody-stained preparations, it is best to aim for approximately
isometric voxels at resolutions approaching the limit of diffraction;
however, this should be traded off against the time taken and the
file size. For live imaging of fluorophores, it is best to minimize
bleaching of the (usually dim) fluorescent proteins. Thus
minimizing the dwell time at any particular voxel is advisable.

3.11 Image To process/visualize images, there are a number of free and

Processing commercial packages available. To produce zebrafishbrain.org, we
mainly use Volocity (PerkinElmer), but other software or a
combination of packages can be used for processing/visualization.
Examples of other packages are Imaris (Bitplane), Amira,
FluoRender, OsiriX, Drishti, and ImageJ/Fiji. There are also some
web-based processing servers that are available to do various tasks
(e.g. XuvTools and ViBE-Z).

4 Notes

1. Depending on the antigen certain antibodies may work better

after fixation with TCA: 2 % TCA in PBS for exactly 3 h at RT
in 5 ml bijous. Transfer to 1.5 ml tubes, rinse 2×, and wash
3 × 5 min PBS (on side). Store at 4 °C for a week, but add
20 mM azide if storing for longer (to prevent mold). In azide
PBS embryos should keep for a month.
2. Take great care of superfine grade forceps. Do not use them to
touch anything except the embryo, and try not the stick them
into the Sylgard as even this can bend and blunt them. To pin
and manipulate the embryo, use normal dechorionation-grade
forceps. Superfine forceps are expensive and not strictly neces-
214 Katherine J. Turner et al.

sary for dissecting. A combination of dissecting needle and

normal forceps can be sufficient. Forceps can be rehoned using
fine pliers and a whetstone.
3. It is possible to dissect an embryo without turning it half-
way through. See http://zebrafishbrain.org/movies/brain_
dissection.
4. To mount live embryos in glass rings for timelapse. Follow the
procedure described in Subheading 3.4.2 with the following
changes. Embryos must first be anesthetized by adding tricaine
to the fish water prior to mounting. Use 0.5–08 % low melt
agarose in filtered fish water or E3. Use fish water with tricaine
instead of PBS to flood the ring.
5. Glycerol agarose sets very slowly. When using glycerol agarose,
replace the PBS used to top up the ring with 80 % glycerol/
PBS.

References

1. Norton W, Bally-Cuif L (2010) Adult zebraf- 11. Wilson SW, Ross LS, Parrett T et al (1990)
ish as a model organism for behavioural genet- The development of a simple scaffold of axon
ics. BMC Neurosci 11:90 tracts in the brain of the embryonic zebrafish,
2. Ahrens MB, Li JM, Orger MB et al (2012) Brachydanio rerio. Development
Brain-wide neuronal dynamics during motor 108(1):121–145
adaptation in zebrafish. Nature 485:471–477 12. Buckley K, Kelly RB (1985) Identification of a
3. Stewart A, Gaikwad S, Kyzar E et al (2012) transmembrane glycoprotein specific for secre-
Modeling anxiety using adult zebrafish: a concep- tory vesicles of neural and endocrine cells.
tual review. Neuropharmacology 62:135–143 J Cell Biol 100:1284–1294
4. Mathur P, Guo S (2010) Use of zebrafish as a 13. Hendricks M, Jesuthasan S (2007) Asymmetric
model to understand mechanisms of addiction innervation of the habenula in zebrafish.
and complex neurobehavioral phenotypes. J Comp Neurol 502:611–619
Neurobiol Dis 40:66–72 14. Miyasaka N, Morimoto K, Tsubokawa T et al
5. Prober DA, Rihel J, Onah AA et al (2006) (2009) From the olfactory bulb to higher
Hypocretin/orexin overexpression induces an brain centers: genetic visualization of second-
insomnia-like phenotype in zebrafish. ary olfactory pathways in zebrafish. J Neurosci
J Neurosci 26:13400–13410 29:4756–4767
6. Lein ES, Hawrylycz MJ, Ao N et al (2007) 15. Cerveny KL, Cavodeassi F, Turner KJ et al
Genome-wide atlas of gene expression in the (2010) The zebrafish flotte lotte mutant
adult mouse brain. Nature 445:168–176 reveals that the local retinal environment pro-
7. Hawrylycz MJ, Lein ES, Guillozet-Bongaarts motes the differentiation of proliferating pre-
AL et al (2012) An anatomically comprehen- cursors emerging from their stem cell niche.
sive atlas of the adult human brain transcrip- Development 137:2107–2115
tome. Nature 489:391–399
16. Collins RT, Linker C, Lewis J (2010) MAZe:
8. Rink E, Wullimann MF (2002) Connections
a tool for mosaic analysis of gene function in
of the ventral telencephalon and tyrosine
zebrafish. Nat Methods 7:219–223
hydroxylase distribution in the zebrafish brain
17. Ronneberger O, Liu K, Rath M et al (2012)
(Danio rerio) lead to identification of an
ViBE-Z: a framework for 3D virtual colocal-
ascending dopaminergic system in a teleost.
ization analysis in zebrafish larval brains. Nat
Brain Res Bull 57:385–387
Methods 9:735–742
9. Mueller T, Dong Z, Berberoglu MA et al (2011)
18. Meta R, Roland N, Alida F, Olaf R et al (2012)
The dorsal pallium in zebrafish, Danio rerio
Generation of high quality multi-view confocal
(Cyprinidae, Teleostei). Brain Res 1381:95–105
3D datasets of zebrafish larval brains suitable for
10. Wullimann MF, Rupp B, Reichert H (eds)
analysis using Virtual Brain Explorer (ViBE-Z)
(1996) Neuroanatomy of the zebrafish brain: a
software. Nat Protoc Exchange. doi:10.1038/
topological atlas. Basel, Switzerland: Birkhäuser
protex.2012.031
Verlag, p. 160
 Part VI

Chicken Protocols
 Chapter 15

Immunohistochemistry and In Situ Hybridization

in the Developing Chicken Brain
Richard P. Tucker and Qizhi Gong

Abstract
One of the first steps in studies of gene function is the spatiotemporal analysis of patterns of gene expres-
sion. Indirect immunohistochemistry is a method that allows the detection of a protein of interest by
incubating a histological section with an antibody or antiserum raised against the protein and then local-
izing this primary antibody with a tagged secondary antibody. To determine the cellular source of a protein
of interest, or if a specific antibody is not available, specific transcripts can be localized using in situ hybrid-
ization. A histological section is incubated with a labeled RNA probe that is complementary to the target
transcript; after hybridization with the target transcript, the labeled RNA probe can be identified with an
antibody. Here we describe materials and methods used to perform basic indirect immunohistochemistry
and in situ hybridization on frozen sections through the developing chicken brain, emphasizing controls
and potential problems that may be encountered.

Key words Immunohistochemistry, In situ hybridization, Riboprobe, Antibody, Fluorescence,

Protocol, Technique, Cryosection

1 Introduction

Immunohistochemistry was first used in the 1940s to identify bac-

terial antigens in mouse tissues by applying fluorescein-labeled
antibodies to frozen histological sections [1]. The technique was
modified in the following decades to amplify the signal by “indi-
rectly” identifying the so-called primary antibodies with labeled
secondary antibodies [2]. The basic method of indirect immuno-
histochemistry is unchanged to this day: a histological section is
incubated with a primary antibody (e.g., a mouse monoclonal anti-
body specific to an antigen of interest), then the section is incu-
bated with a tagged secondary antibody (e.g., fluorescently tagged
rabbit polyclonal antibodies against mouse antibodies), and the
protein of interest is then observed using a microscope fitted with
special illumination and optics. The technique is widely used in the

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_15, © Springer Science+Business Media, LLC 2014

217
218 Richard P. Tucker and Qizhi Gong

neurosciences for determining patterns of expression of novel gene

products as well as for identifying specific types of neurons and
supporting cells in brain sections.
In situ hybridization is used to detect transcripts in their native
tissue environment. The technique was first used in the early 1980s
to study gene expression in Drosophila embryos [3, 4], but it was
quickly adapted to studies of vertebrate brain [5]. In short, a tissue
section is treated to make it amenable to hybridization, and then it
is incubated with a labeled nucleic acid probe with sequences com-
plementary to the target transcript. Excess probe is removed and
the label is detected. There are several established methods for tis-
sue in situ hybridization [6]. One of the major differences between
the published methods is the choice of hybridization probes; in
general, RNA probes provide more sensitive detection and cleaner
background [7].
Here we present detailed protocols for fixing and cryosection-
ing embryonic chicken brains as well as methods for performing
indirect immunohistochemistry and in situ hybridization with
RNA probes on these frozen sections. The methods are easily
adapted to other tissues and to tissues from other species. For illus-
trative purposes we have localized the extracellular matrix glyco-
protein tenascin-C, which plays critical roles in brain development
[8], and its transcript.

2 Materials

The required materials are listed below in Subheadings 2.1, 2.2,

and 2.3.

2.1 Cryosectioning 1. Phosphate-buffered saline, pH 7.4 (PBS): 10 g NaCl, 2 g KCl,

Components 11.5 g Na2HPO4⋅7H2O, and 2 g KH2PO4 in 800 mL of
ddH2O. Bring volume to 1 L and adjust pH.
2. Fixative: 4 % paraformaldehyde in PBS, pH 7.4, made fresh
(same day). Add 2 g of reagent grade paraformaldehyde pow-
der to 40 mL of PBS in a 50 mL conical tube. Add two NaOH
pellets, cap, and allow the pellets to dissolve with gentle shak-
ing. When the paraformaldehyde is in solution, adjust the pH
(using pH-sensitive paper) with 2 N HCl, and then top off
with PBS to a final volume of 50 mL. Store on ice (see Note 1).
3. Cryoprotection: sucrose.
4. Embedding medium: Tissue-Tek O.C.T. Compound (Sakura
Finetek).
5. Embedding molds: disposable base molds (various sizes, e.g.,
15 × 15 × 5 mm).
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 219

Fig. 1 Schematic illustration of an immunohistochemistry incubation chamber, viewed from above (top ) and
the side (bottom ). The chamber should have a flat bottom and either a securely fitting lid or a lid fashioned
from plastic wrap. As the slides rest directly on the moist paper towel, care should be taken not to let the slides
touch each other, and the “puddle” of antibody should not extend to the very edge of the slide

6. Freezing solution: 2-methylbutane. Just before use, break up a

10 cm × 10 cm × 3 cm block of dry ice into powder and chunks
approximately 1 cm3. Put the dry ice into a glass dish, and add
the 2-methylbutane to make a slurry 1–2 cm deep.
7. Slides: Superfrost Plus precleaned microscope slides
(25 × 75 × 1 mm).
8. Cryostat microtome: for example, Leica CM3050.

2.2 Immunohisto- 1. Buffer: PBS (see above).

chemistry Components 2. Blocking agent: 0.5 % bovine serum albumin (BSA) in PBS.
3. Coplin jars: polypropylene Coplin staining jar.
4. Staining tray: any flat-bottomed glass or plastic tray, approxi-
mately 5 cm deep. Place a layer of paper towels on the bottom
and dampen with water, pushing out excess water and air bub-
bles to keep the surface flat. Cover with sticky plastic wrap
(Fig. 1).
5. Antibodies: The primary antibody used as an example in this
protocol is M1-B4 (mouse anti-chicken tenascin-C [9];
Developmental Studies Hybridoma Bank). Secondary anti-
bodies should be against the animal source of the primary anti-
body. The secondary antibody used here is Alexa 594 rabbit
anti-mouse IgG (Invitrogen).
6. Coverslips: various sizes (e.g., 22 × 30 mm).
7. Nuclear stain: Hoechst (Sigma-Aldrich). Make a 2 mg/mL
100× stock and store in a foil-wrapped tube at 4 °C.
8. Wet coverslip mounting medium: 5 mL PBS/5 mL glycerol.
220 Richard P. Tucker and Qizhi Gong

2.3 In Situ 1. cDNA clones in an RNA polymerase binding site containing

Hybridization plasmid: Example given in this protocol is a tenascin-C cDNA
Riboprobes clone, which is in pCRII-TOPO plasmid (Invitrogen).
2. Desired restriction enzymes to linearize the plasmid: We used
Spe I and EcoR V in this experiment.
3. Phenol/chloroform/isoamyl alcohol (25:24:1) and chloroform.
4. 100 % ethanol and 70 % ethanol in RNase-free H2O.
5. T7 and Sp6 RNA polymerase 20 U/μL (Promega, Madison,
WI, USA). 5× transcription buffer and 100 mM DTT are pro-
vided with the enzyme.
6. DIG RNA labeling mix (Roche Applied Science, Indianapolis,
IN, USA).
7. RNase-free DNase 10 U/μL (Promega).
8. RNase inhibitor 20 U/μL (New England Biolabs, Ipswich,
MA, USA).
9. 4 M LiCl: Dissolve 17 g of LiCl in 100 mL of ddH2O. Add
100 μL of diethyl pyrocarbonate (DEPC) and mix well. Let
the solution stand overnight and then autoclave for at least
15 min.

2.4 In Situ 1. Hair dryer with cool air setting.

Hybridization 2. Slides.
Components
3. Plastic storage box with a tight-fitting lid.
4. Microscope cover glass (e.g., 40 × 22 or 50 × 22 mm; Fisher
Scientific or Corning).
5. DEPC-treated H2O: 1 mL of DEPC, 1 L of ddH2O. Mix well.
Let stand at room temperature overnight or 37 °C for 2 h.
Autoclave at least 15 min to inactivate the DEPC.
6. DEPC-treated PBS: 100 mL 10× PBS (see above), 900 mL of
ddH2O, 1 mL of DEPC. Mix well by shaking, and let stand in
room temperature overnight. Autoclave to inactivate the
DEPC.
7. 1 M Tris–Cl, pH 8.0: 60.5 g Tris base in 400 mL DEPC-
treated H2O. Adjust pH to 8.0 with concentrated HCl, and
bring volume up to 500 mL with DEPC-treated H2O. Filter
sterilize (see Note 2).
8. 1 M Tris–Cl, pH 7.5 and pH 9.5: Same as above; adjust pH to
7.5 and 9.5, respectively.
9. 0.5 M ethylenediaminetetraacetic acid (EDTA), pH 8.0: 19 g
of EDTA, 80 mL DEPC-treated H2O, 10 N NaOH (about
4 mL). Mix to dissolve, adjust pH to 8.0 with NaOH, and
bring volume to 100 mL with DEPC-treated H2O. Filter
sterilize.
Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 221

10. TE buffer, pH 8.0: 5 mL of 1 M Tris–Cl, pH 8.0, and 1 mL of

0.5 M EDTA. Bring to 500 mL with DEPC-treated H2O.
11. Proteinase K: 12.8 μL of 15.6 mg/mL proteinase K, 20 mL of
TE buffer, pH 8.0.
12. 2 N HCl.
13. 1 M triethanolamine–HCl stock: 66.25 mL triethanolamine,
11.25 mL of HCl, 500 mL of DEPC H2O. Dilute with DEPC-
treated H2O to make 0.1 M triethanolamine–HCl.
14. Acetic anhydride.
15. Ethanol series: Dilute ethanol with DEPC-treated H2O to
make 50 %, 75 %, and 90 % ethanol.
16. Hybridization solution:

Stock For 10 mL solution Final concentration

Formamide 5 mL 50 %
Yeast tRNA, 10 mg/mL 200 μL 0.2 mg/mL
Dextran sulfate, 50 % 2 mL 10 %
1 M Tris–Cl, pH 8.0 100 μL 10 mM
0.5 M EDTA, pH 8.0 20 μL 1 mM
Denhardt’s, 50× 100 μL 0.5×
5 M NaCl 1.2 mL 600 mM
10 % SDS 250 μL 0.25 %
DEPC-treated H2O To 10 mL

17. 20× SSC: 175.3 g NaCl, 88.2 g sodium citrate

(Na3C6H5O7·2H2O), 800 mL DEPC-treated H2O. Adjust pH
to 7.0 with 1 M HCl, and add DEPC-treated H2O to 1 L.
18. RNase A stock solution 10 mg/mL.
19. 5 M NaCl: 146.1 g NaCl, 450 ddH2O. Mix by stirring. Add
ddH2O to 500 mL.
20. TNE buffer: 5 mL 1 M Tris–Cl, pH 7.5, 50 mL 5 M NaCl,
1 mL 0.5 M EDTA pH 8.0, and 444 mL ddH2O.
21. Buffer 1: 100 mM Tris–Cl pH7.5, 150 mM NaCl
22. Blocking reagent (Roche Applied Science).
23. Alkaline phosphatase (AP)-conjugated anti-DIG antibody
(Roche Applied Science).
24. Buffer 2: 4 mL 1 M Tris–Cl, pH 9.5, 2 mL 1 M MgCl2, 800 μL
5 M NaCl, ddH2O to make 40 mL. Make fresh for each use.
25. NBT/BCIP stock solution (Roche Applied Science).
26. Flouromount-G (SouthernBiotech).
222 Richard P. Tucker and Qizhi Gong

3 Methods

If studying a range of developmental stages, it can be convenient to

put the fertilized chicken eggs into the incubator at different inter-
vals and then collect and process all of the embryos at once. Be sure
to follow appropriate institutional and regional ethical guidelines
for the treatment of research animals. The fixation and sectioning
protocol described below is used to make frozen sections suitable
for both immunohistochemistry and in situ hybridization.

3.1 Cryosectioning 1. Remove the embryo from the egg by cracking the egg into a
small bowl containing a small volume of PBS (room tempera-
ture). After trimming put the whole embryo (embryonic day
[E]4–E7) or the head (E8–E16) into a 50 mL conical centri-
fuge tube with ice-cold 4 % paraformaldehyde in PBS (pH 7.6).
Cap the tube and place it on ice on a rotary shaker for 1–6 h,
and then store the tube at 4 °C overnight (see Note 3).
2. Pour off and dispose of the fixative appropriately and add ice-
cold PBS to the tube. Cap the tube and place on a rotary shaker
in an ice bath for 30 min. Pour off the PBS and repeat twice
with fresh PBS.
3. Add sucrose to the third PBS rinse to a final concentration of
20–25 %. Cap the tube and place on rotating shaker until the
sucrose has gone into solution. The tissues should “float” on
the sucrose. Store the tube with floating tissues at 4 °C over-
night (see Note 4).
4. Put the sucrose infiltrated (cryoprotected) tissues into a petri
dish or onto a piece of Parafilm©, and trim them with a fresh
razor blade such that the trimmed surface will match the ori-
entation of the sections to be cut in the cryostat. Transfer the
trimmed tissues to a “puddle” of embedding compound to
minimize the presence of any sucrose solution, and then trans-
fer the tissue to an embedding mold partially filled with embed-
ding medium. With forceps orient the tissue so that the cut
surface is resting along the bottom of the mold (see Note 5).
Float the mold in a shallow slurry of 2-methylbutane and
crushed dry ice, taking care not to let the 2-methylbutane
come into direct contact with the embedding medium. When
the embedding medium is solid white, transfer the mold with
the frozen tissue to dry ice (see Note 6).
5. Cut sections with a cryostat using appropriate protocols, and
collect the sections on pre-subbed slides (see Note 7). Allow the
sections to air-dry for 4–6 h, and then either proceed with the
protocol or store the slides at −20 °C to −70 °C (see Note 8).
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 223

3.2 Immunohisto- All of the immunohistochemistry procedures described below are

chemistry carried out at room temperature.
1. Allow slides to come to room temperature (if stored in the
freezer), and then place them into a Coplin jar or slide staining
rack (depending on the number of slides being processed) with
PBS for 10–30 min. Pour off the PBS and add PBS with BSA
blocking solution (see Note 9). Incubate the slides in blocking
solution for 15–30 min (longer blocking can reduce back-
ground, but typically is not necessary).
2. Dilute the primary antibody in blocking solution in a 1.5 mL
snap cap tube. It may be necessary to use a range of dilutions
(e.g., 1:10, 1:100, and 1:1,000) with antibodies that have not
been characterized before. Spin the diluted antibodies in a
tabletop microfuge for 5 min. Reserve a few mL of blocking
solution for the secondary antibody (store at 4 °C until used).
3. As the primary antibody is spinning, remove the slides from
the blocking solution. Wipe away excess solution using a
Kimwipe© taking care not to disturb the section and taking
care not to let the section dry out (see Note 10). Label the
slide and place it on the moist paper towel in the incubation
chamber; add an appropriate volume of diluted and centri-
fuged primary antibody until the section is completely covered
(about 200 μL). The antibody should stay on the region that
is wet and not spread onto the parts of the slide that were dried
with the Kimwipe©.
4. Cover the chamber (e.g., with a clingy plastic wrap) and leave
undisturbed on the bench top overnight (Fig. 1).
5. Gently tap the primary antibody from the slides onto a
Kimwipe©, and rinse the slides three times for 10 min in PBS.
6. Dilute the secondary antibody (following the manufacturer’s
instructions) in the reserve blocking solution in a 1.5 mL snap
cap tube, and centrifuge in a tabletop microfuge for 5 min.
7. As the secondary antibody is spinning, remove the slides from
the PBS rinse and carefully dry them (as above), leaving the
area where the section is found wet. Place the slides back into
the incubation chamber, and add the secondary antibody until
it covers the section. Take the same precautions noted above
for adding the primary antibody.
8. Cover the chamber and incubate the slides with the secondary
antibody for 1–3 h (longer incubations are possible, but usu-
ally not necessary).
9. Gently tap the secondary antibody off each slide onto a
Kimwipe©, and rinse the slides three times for 10 min in PBS.
Optional: Sections can be counterstained with a fluorescent
nuclear dye at this stage. Between the second and third PBS
224 Richard P. Tucker and Qizhi Gong

rinse, put the slides into a Coplin jar with nuclear stain (Hoechst)
diluted to the 1× working concentration in PBS for 1 min.
10. Coverslip carefully, without introducing air bubbles, using
either the PBS/glycerol wet mount or a permanent mounting
medium suited for fluorescence microscopy (e.g.,
Fluoromount-G) and appropriately sized glass coverslips (see
Note 11).
11. One or more control incubations should also be run. If the anti-
body is well characterized, it may be sufficient to incubate a sec-
tion in the blocking solution overnight and then treat it with the
secondary antibody the following day. Such “secondary anti-
body only” controls will reveal background fluorescence or non-
specific binding of the secondary antibody. When illustrating
this type of control (Fig. 2), be sure to use the same camera
settings that were used when imaging the experimental sections.
If a polyclonal antiserum is used, one should run a control that
includes similarly diluted pre-immune serum.

3.3 In Situ In situ hybridization is a multistep procedure that can take 3–4
Hybridization days. For better planning, make all solutions ahead of time. Frozen
sections can be prepared days to weeks ahead and stored at −20 °C.
DIG-labeled RNA probes need to be made and the quality of the
probe evaluated before starting in situ hybridization.

3.4 Preparation For this protocol we used riboprobes generated against tenascin-C
of DIG-Labeled to match the immunohistochemistry described above.
Riboprobes
1. cDNAs of tenascin-C (GENE ID: 396440 TNC, 2,761–
3,179 bp) were obtained by RT-PCR using primers, 5′- acag-
gactaccattgacctctctg -3′ and 5′- atcagtgccagcattaatggtagc -3′,
and cloned into pCRII-TOPO. Orientation and the sequence
were confirmed by sequencing (see Note 12).
2. 10 μg of cDNA-containing plasmids were linearized either at
the 5′ end for an antisense probe or at the 3′ end of the cDNA
for a control sense probe, with appropriate restriction endo-
nucleases (Fig. 3). The restriction digest should be done over-
night to ensure completeness (see Notes 13 and 14).
3. Take out 1/20 of the reaction to run on an agarose gel to con-
firm that the restriction digest is complete.
4. Purify the restriction digest reactions with phenol extraction:
Bring the volume to 100 μL with ddH2O in a 1.5 mL tube. Add
100 μL of phenol/chloroform/isoamyl alcohol (25:24:1);
vortex for 1 min and centrifuge at 16,000 × g for 5 min; carefully
remove the top aqueous layer and add it to a new tube; add an
equal volume of chloroform to the aqueous layer and vortex for
1 min and centrifuge at 16,000 × g for 5 min. Again, collect the
top aqueous layer and transfer to a new tube.
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 225

Fig. 2 Examples of indirect immunohistochemistry and in situ hybridization on adjacent sections through the
E12 chicken brain. The primary antibody recognizes tenascin-C, and the antisense riboprobe recognizes
tenascin-C transcripts. (a) Anti-tenascin-C labels the optic fiber layer (ofl) and inner plexiform layer (ipl) of the
E12 retina. gcl, ganglion cell layer; inl, inner nuclear layer; rpe, retina pigment epithelium. (b) An adjacent sec-
tion through the E12 retina subjected to in situ hybridization with a tenascin-C antisense probe. The dark
reaction product is found in the gcl and inl. (c). An adjacent section incubated with a control sense probe,
illustrating the specificity of the hybridization signal in b. The rpe is dark due to the presence of melanosomes
and not the target transcript. Scale bar: 100 μm (a–c). (d) Anti-tenascin-C immunoreactivity is found around
the nucleus of Edinger-Westphal (EW) and oculomotor nucleus (ON). AQ, cerebral aqueduct. (e) The tenascin-C
antisense probe hybridizes in large cells within the EW and ON as well as in the ependymal layer (el). (f). The
control section incubated with a sense probe shows that the reaction products seen in E are specific.
(g). The tenascin-C antibody labels the region underlying the ependymal layer (el) of the optic tectum as well
as the stratum griseum periventriculare (SGP). VT ventricle of the optic tectum, SGC stratum griseum centrale.
(h) A secondary antibody only control shows the absence of background fluorescence in the same region in an
adjacent section. (i) An adjacent section incubated with a tenascin-C antisense probe. The reaction product is
found in the el and SGC. Scale bar: 500 μm (d–i)
226 Richard P. Tucker and Qizhi Gong

Fig. 3 Steps for making tenascin-C riboprobes. A cDNA fragment of tenascin-C is cloned into the pCRII-TOPO
plasmid with an Sp6 promoter at its 5′ end. Restriction enzyme Spe I and EcoRV sites are located at different
ends of the cloned cDNA. To make antisense probes, plasmids are digested with Spe I and transcribed with T7
RNA polymerase. To make control sense probes, plasmids are digested with EcoR V and transcribed with Sp6
RNA polymerase. DIG-labeled UTPs are used to incorporate DIG (indicated by black dots) into the riboprobes

5. Precipitate the linearized plasmids: Add 1/10 volume of 3 M

sodium acetate and 3 volumes of ethanol; mix by briefly vor-
texing and leave the reaction at −20 °C for 2 h or −80 °C for
1 h; centrifuge at 16,000 × g in a refrigerated centrifuge for
20 min.
6. Take out the liquid and wash the pellet with 1 mL of 70 %
ethanol (made with DEPC-treated H2O); spin again in the
refrigerated centrifuge for 10 min, empty the tube, and let the
pellet dry at room temperature for 10 min.
7. Dissolve the pellet with 20 μL of TE, pH 8.0.
8. To make DIG-labeled RNA probes, assemble the reaction as
follows:

Template DNA 1 μg
5× transcription buffer 4 μL
DIG RNA labeling mix 2 μL
0.1 M DTT 1 μL
RNA polymerase 2 μL
RNase inhibitor 0.5 μL
RNase-free water to 20 μL
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 227

9. Incubate the reaction at 37 °C for 2 h.

10. Add the following and incubate at 37 °C for 15 min:

RNase inhibitor 1 μL
DNase 1 μL

11. Precipitate RNA probes: Add 1/10 volume of 4 M LiCl and

2.5 volumes of ethanol; incubate at −20 °C for 2 h or −80 °C
for 1 h.
12. Pellet RNA probes for 20 min at high speed in a 4 °C centri-
fuge. Wash with 70 % ethanol (RNase-free) and air dry.
Resuspend in 20 μL RNase-free water (see Note 15).

3.5 In Situ 1. Dry sections with cold air using a hair dryer for up to 1 min
Hybridization: Day 1. (see Note 16).
Pre-hybridization 2. All pre-hybridization and subsequent washing steps are done
and Hybridization: in a vertical slide mailer that holds five slides. Complete immer-
RNase-Free sion of tissue sections requires 15 mL of solution for each
mailer.
3. Place slides into a slide mailer and postfix with 4 % paraformal-
dehyde in PBS for 15 min.
4. Rinse in DEPC-treated PBS three times, 1 min each.
5. Digest sections with proteinase K at room temperature for
6–12 min (see Note 17).
6. Inactivate the proteinase K by incubating the sections in 4 %
paraformaldehyde in PBS for 10 min at room temperature.
7. Rinse in DEPC-treated PBS three times, 1 min each.
8. Treat the sections with 0.2 N HCl for 10 min at room tem-
perature (see Note 18).
9. Rinse in DEPC-treated PBS three times, 1 min each.
10. Measure 15 mL of 0.1 M triethanolamine–HCl into a 15 mL
tube. Add 37.5 μL of acetic anhydride drop by drop. Mix by
inverting and pour into the mailer immediately.
11. Transfer slides into the mailer and incubate sections for 10 min
at room temperature.
12. Wash in DEPC-treated PBS three times, 1 min each.
13. Dehydrate sections in increasing concentrations of ethanol
(50 %, 75 %, 95 %, 100 %, 100 %) for 2 min each. Sections are
subsequently air-dried for 2 min.
14. Add 400 μL of hybridization solution to two 1.5 mL tubes.
Heat the hybridization solution at 85 °C for 10 min.
228 Richard P. Tucker and Qizhi Gong

15. Add DIG-labeled sense and antisense probes (~ 1 μg/mL) to

the preheated hybridization solution, and incubate at 85 °C
for 3 min (see Note 19).
16. Add 400 μL of probe solution onto each slide. Cover the slide
with a glass coverslip. Be sure there are no bubbles and that the
hybridization solution covers the entire coverslip area.
17. Place the slides into a sealed, humid chamber. We line the bot-
tom of the plastic box with a piece of paper towel soaked with
50 % formamide in 5× SSC. Instead of placing the slides directly
on the paper towel, place a 1.5 mL tube rack upside down to
build a platform for the slides. Cover the box with a tight lid.
18. Hybridize at 60 °C overnight (see Notes 20 and 21).

3.6 In Situ 1. Carefully remove the coverslips by dipping the slides in and
Hybridization: Day 2. out of 2× SSC (see Note 22).
Post-hybridization 2. Incubate in 2× SSC and 50 % formamide at 60 °C for 30 min.
Washes and
3. Transfer sections to TNE buffer at 37 °C for 10 min.
Immunological
Detection: 4. Replace the TNE buffer with TNE plus RNaseA (20 μg/mL)
and incubate at 37 °C for 30 min.
Non-RNase-Free
5. Wash with TNE for 10 min at 37 °C.
6. Further wash the sections in 2× SSC at 60 °C for 30 min.
7. Proceed with high-stringency washes in 0.2× SSC twice for
30 min at 60 °C (see Note 23).
8. Let the sections stay cool at room temperature for 5 min.
9. Wash with Buffer 1 for 5 min at room temperature.
10. Incubate with 1 % blocking reagent for 1 h at room
temperature.
11. Briefly rinse the sections with Buffer 1 plus 0.1 % Tween 20
(B1-T). Put 300 μL of alkaline phosphatase-conjugated anti-
DIG antibody (1:1,000 with B1-T) onto each slide. Cover the
sections with coverslips and place the slides into a humidified
chamber (e.g., a flat-bottomed box with a moist paper towel as
in Fig. 1, but with the sections coverslipped to reduce the vol-
ume of antibody solution).
12. Incubate overnight at 4 °C.

3.7 In Situ 1. Remove the coverslips by dipping the slides in B1-T. Wash in
Hybridization: Day 3. B1-T three times 10 min each at room temperature.
Signal Detection: 2. Precondition with Buffer 2 (B2) for 5 min.
Non-RNase-Free
3. Replace with 20 mL of B2 with 400 μL of NBT/BCIP. The
NBT/BCIP solution is light sensitive. Keep the reaction in the
dark.
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 229

4. The color reaction may take 30 min to 2 h or longer. Check

the staining at 30 min intervals. If the reaction is weak at the
end of 2 h, leave the staining overnight in a dark place at room
temperature (see Note 24). The reaction product is blue-pur-
ple in color and located in the cytoplasm.
5. Stop reaction by washing with TE, pH 7.5 for 10 min.
6. Rinse quickly with ddH2O and mount with Fluoromount-G
(see Note 25).
7. Determine appropriate conditions for imaging the sections
hybridized with antisense, and then photograph the control
sense slides with the same camera settings (Fig. 2).

4 Notes

1. Paraformaldehyde causes skin, eye, and respiratory tract irrita-

tion and is a suspect carcinogen. All handling of paraformalde-
hyde should be done in the fume hood. Read and follow
precautions provided by the manufacturer.
2. DEPC inactivates RNase at 0.1 % concentration and is there-
fore used to treat solutions used for in situ hybridization. It is
important to remember that DEPC should not be used to
treat Tris buffer. When making RNase-free Tris buffer, dissolve
Tris base into DEPC-treated water and filter sterilize.
3. It is important to use freshly prepared formaldehyde as fixative
made more than a day or two before use can increase back-
ground fluorescence. Do not over fix. Tissues left in fixative for
more than 24 h can become so cross-linked that in situ hybrid-
ization will not work, and over fixation can also adversely affect
epitopes. Some antibodies will not work on formaldehyde-
fixed tissues. It may be necessary to experiment with different
fixatives, like methanol-, ethanol-, or acetone-based fixatives.
4. The sucrose will prevent tissue disruption from the expansion
of water in tissues as it freezes. Tissues that have taken up suf-
ficient sucrose to withstand freezing will sink into the solution
over time. Smaller tissues will sink sooner and can be processed
the same day. Larger tissues will take more time, but most will
sink after 24 h. Air bubbles trapped in the tissues may prevent
them from sinking even when they have taken up the sucrose,
so avoid introducing air bubbles into the specimens.
5. It is important to orient the material properly before it is
frozen. Chose traditional orientations that correspond to
orientations found in atlases to help interpret your results.
6. Tissues frozen in embedding medium can typically be stored in
ziplock storage bags at −20 °C for several weeks or at −70 °C
230 Richard P. Tucker and Qizhi Gong

for several months. Tissues can usually be stored for longer

periods if the sections will be used for immunohistochemistry
than if they are to be used for in situ hybridization.
7. Wear gloves when handling slides that will be used for in situ
hybridization. It is easier to cut and collect thicker sections,
but try to collect sections in the 12–16 μm range. The ideal
temperature for cutting sections through chicken brain is usu-
ally colder than the ideal temperature for cutting murine tis-
sues; precise block, knife, and chamber tissues will vary from
cryostat to cryostat and require practice before working with
precious specimens.
8. Store the slides in a clean box and wrap the edges of the box
with Parafilm©, and place it in a ziplock bag before refreezing.
It is always best to work with fresh slides, but not always practi-
cal. Slides kept in the freezer for several years are often suitable
for immunohistochemistry, but slides used for in situ hybrid-
ization should be used within a few weeks.
9. The BSA-based blocking solution is easy to make and use and
works well for most antibodies, but with some antibodies it
may be necessary to use a fetal calf serum or condensed milk-
based blocking solution.
10. The moisture left behind after drying around the section will
help keep the primary antibody from spreading away from the
section during the antibody incubation step. There is an
increased chance of auto-fluorescence if the section dries out
during this step. To avoid this, work with small batches of
slides.
11. The wet mounting medium (PBS/glycerol) is easy to use, but
care must be taken not to use too much or the medium may
spread onto the microscope stage and/or objectives. Dab at the
edge of the coverslip after mounting to remove excess medium,
and let the coverslipped slides air-dry for several hours before
use. Another advantage of this medium is that it is easy to “float
off” the coverslip after photography, allowing the section to be
counterstained or stained with another primary and secondary
antibody (double label immunohistochemistry).
12. The most important factor for probe selection is its sequence
specificity to the target gene. When choosing cDNA fragments
as template, perform a blast search to find out whether or not
they have homology to other transcripts. The optimal probe
size is 100–1000 bp. If the probe is too long, it can inhibit
tissue penetration.
13. In pCRII-TOPO, T7 and Sp6 polymerase binding sites are
located on different sides of the cloned cDNA. In this experi-
ment, sequencing indicated that the tenascin-C cDNA clone
was oriented so that the T7 RNA polymerase transcribes the
Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 231

antisense riboprobe and the Sp6 transcribes the sense (control)

probe. When selecting restriction enzymes, it is preferable to
use enzymes that result in either a 5′ overhang or a blunt end
DNA template. We used Spe I to linearize the template for the
antisense probe and EcoR V for the sense probe. Selected
restriction enzymes will cut on either ends of the cDNA,
respectively, to allow the RNA polymerase to transcribe
through the cDNA, but relatively little of the plasmid
sequences.
14. When performing in situ hybridization, it is important to
incorporate controls. The sense probe often serves as a nega-
tive control for nonspecific hybridization signals. Under high-
stringency condition, the sense probe should give no in situ
signal while the antisense probe hopefully hybridizes specifi-
cally to its target. Others suggest to evaluate tissue transcript
quality by using labeled poly T probes [10].
15. The yield of the labeled RNA probe is usually 10–15 μg. To
evaluate the quantity of the probe, measure the concentration
using a nanodrop or conventional spectrophotometer. To
determine the quality and the size of the RNA probe, run
0.5–1 μg of the probe on a 2 % agarose formaldehyde gel. To
assemble a 2 % agarose formaldehyde gel, combine 2 g of aga-
rose and 73.3 mL of DEPC-treated ddH2O, and heat to boil;
in the fume hood add 16.7 mL formalin (37 % formaldehyde)
and 10 mL of 10× MOPS (200 mM MOPS, pH 7.0, 20 mM
sodium acetate, 10 mM EDTA, pH 8.0); mix and pour into
the tray to set. The probes are diluted with 1× MOPS, 20 %
formaldehyde, and 50 % formamide; heat denature at 70 °C
for 10 min before loading onto the gel. The gel is run in 1×
MOPS.
16. RNases are long lived and difficult to inactivate. Small amount
of contamination can destroy the transcripts in the tissue sec-
tion, so it is important to prevent RNase contamination. On
day 1 of in situ hybridization, all solutions should be RNase-
free. Use DEPC-treated water to make all solutions. Wear
gloves during day 1. It is important to make sure that the slide
mailers for pre-hybridization steps are brand new and free of
RNase.
17. Proteinase K treatment is used to allow better access of the
probe to its target transcripts in tissue sections. Excessive diges-
tion will result in damage to tissue morphology. The concen-
tration, duration, and the temperature of the treatment should
be determined empirically. Proteinase K is commonly used at
10 μg/mL. Dependent upon the tissue type, digestion can be
done between 25 and 37 °C for 5–15 min. A good starting
point is 10 min at room temperature. When treating embryonic
232 Richard P. Tucker and Qizhi Gong

tissue, it is suggested to start with lower concentration of pro-

teinase K (2 μg/mL) or a shorter digestion time.
18. The function of HCl is not entirely known. It is believed that
HCl extracts protein and hydrolyzes the target sequence. The
treatment may allow better permeability and also appears to
reduce background.
19. Dependent upon the abundance of the target transcript, the
optimal concentration of the labeled probe should be deter-
mined empirically. A good starting range is 0.5–1 μg/mL.
20. The theoretical melting temperature (Tm), which is the tem-
perature at which 50 % of the probe is dissociated from the
target, is determined by a number of factors including mon-
ovalent cation concentration, the presence of formamide,
probe length, and GC content. Tm can be calculated as follows:
Tm = 79.8+18.8 × log [Na]+0.58 × GC %+11.8 × [GC %]2 −
0.35× %formamide-820/length [11]. Hybridization is often
carried out at 20 °C below the Tm.
21. The Tm on tissue (tissue Tm) is different from the theoretical
Tm. While theoretical Tm is determined by DNA or RNA
behavior in solution, tissue Tm is determined by chromogenic
detection of a specific hybridization signal. In general, tissue
Tm is around 10° higher than the theoretical Tm. It is important
to keep this mind when designing hybridization and washing
conditions [12–15].
22. During post-hybridization washes and all subsequent steps, it
is not necessary to maintain an RNase-free environment. In
general, RNases cleave single-stranded but not double-
stranded RNAs. After hybridization, probes are hybridized
with their transcript targets and are resistant to degradation by
RNase contamination.
23. The stringency of the washes needs to be determined empiri-
cally for each probe. Washing stringency is determined by both
the salt concentration and the washing temperature. For RNA
probes larger than 300 bp, washing temperatures are generally
between 55 and 65 °C, and the salt concentrations vary
between 1× SSC and 0.1× SSC. As discussed in (see Note 21),
tissue Tm is generally 10–15° higher than theoretical Tm. In
this experiment, washing with 0.2× SSC at 60 °C yields the
most specific in situ signal for TNC.
24. NBT/BCIP are AP substrate-chromogens. The reaction
results in a blue-purple precipitate where hybridization has
occurred. If the tissue has endogenous AP activity, it will yield
a false positive signal. Adding 0.1 mM levamisole to the NBT/
BCIP solution will help inhibit endogenous AP activity.
 Immunohistochemistry and In Situ Hybridization in the Developing Chicken Brain 233

25. Mounting should be done with a quick wash with water and
using an aqueous mount. Do not dehydrate with xylene as the
treatment will result in the formation of crystals.

Acknowledgments

We would like to thank Huaiyang Chen for her help in the design
of the riboprobe and the cloning of the cDNA, and Chenghao
Qian for his technical help with the in situ hybridization.

References
1. Coons AH, Creech HJJ, Jones RN, Berliner E 9. Chiquet M, Fambrough DM (1984) Chick
(1942) The demonstration of pneumococcal myotendinous antigen. I. A monoclonal anti-
antigen in tissues by the use of fluorescent body as a marker for tendon and muscle mor-
antibody. J Immunol 45:159–170 phogenesis. J Cell Biol 98:1926–1936
2. Beutner EH, Witebsky E (1963) Studies on 10. Iezzoni JC, Kang JH, Bucana CD, Reed JA,
organ specificity. Xv. Immunohistologic evalu- Brigati DJ (1993) Rapid colorimetric detec-
ation of reactions produced by thyroid autoan- tion of epidermal growth factor receptor
tibodies. J Immunol 91:204–209 mRNA by in situ hybridization. J Clin Lab
3. Akam ME (1983) The location of Ultrabithorax Anal 7:247–251
transcripts in Drosophila tissue sections. 11. Bodkin DK, Knudson DL (1985) Assessment
EMBO J 2:2075–2084 of sequence relatedness of double-stranded
4. Hafen E, Levine M, Garber RL, Gehring WJ RNA genes by RNA-RNA blot hybridization.
(1983) An improved in situ hybridization J Virol Methods 10:45–52
method for the detection of cellular RNAs in 12. Herrington CS, McGee JO (1994)
Drosophila tissue sections and its application Discrimination of closely homologous human
for localizing transcripts of the homeotic genomic and viral sequences in cells and tis-
Antennapedia gene complex. EMBO J sues: further characterization of Tmt.
2:617–623 Histochem J 26:545–552
5. Garner CC, Tucker RP, Matus A (1988) 13. Herrington CS, Anderson SM, Graham AK,
Selective localization of messenger RNA for McGee JO (1993) The discrimination of high-
cytoskeletal protein MAP2 in dendrites. risk HPV types by in situ hybridization and the
Nature 336:674–677. doi:10.1038/336674a0 polymerase chain reaction. Histochem J
6. Wilkinson DG (1998) In situ hybridization: a 25:191–198
practical approach, 1999/01/30th edn. 14. Herrington CS, Graham AK, Flannery DM,
Oxford University Press, Oxford, p 224 Burns J, McGee JO (1990) Discrimination of
7. Zeller R, Rogers M, Haramis AG, Carrasceo closely homologous HPV types by nonisotopic
AS (2001) In situ hybridization to cellular in situ hybridization: definition and derivation
RNA. Current protocols in molecular biol- of tissue melting temperatures. Histochem J
ogy/edited by Frederick M. Ausubel … [et al.] 22:545–554
Chapter 14, Unit 14 13, doi:10.1002/ 15. Evans MF, Aliesky HA, Cooper K (2003)
0471142727.mb1403s55 Optimization of biotinyl-tyramide-based in
8. Chiquet-Ehrismann R, Tucker RP (2011) situ hybridization for sensitive background-
Tenascins and the importance of adhesion free applications on formalin-fixed, paraffin-
modulation. Cold Spring Harb Perspect Biol embedded tissue specimens. BMC Clin Pathol
3, doi:10.1101/cshperspect.a004960 3:2. doi:10.1186/1472-6890-3-2
 Chapter 16

Transplantation of Neural Tissue: Quail-Chick Chimeras

Andrea Streit and Claudio D. Stern

Abstract
Tissue transplantation is an important approach in developmental neurobiology to determine cell fate, to
uncover inductive interactions required for tissue specification and patterning as well as to establish tissue
competence and commitment. Avian species are among the favorite model systems for these approaches
because of their accessibility and relatively large size. Here we describe two culture techniques used to gener-
ate quail-chick chimeras at different embryonic stages and methods to distinguish graft and donor tissue.

Key words Chick, Brain, Neural plate, Neural tube, Quail, Transplantation

1 Introduction

During development the entire central nervous system arises from

the neural plate, which is induced in the ectoderm by signals from
the organizer [1–3]. Shortly thereafter, precursors for the fore-
brain, midbrain, hindbrain, and spinal cord occupy different, albeit
overlapping, territories [4–7]. As the neural plate folds to form the
neural tube, anterior–posterior and dorsoventral patterning is
established through signals from surrounding tissues but also
through the action of local organizers like the midbrain–hindbrain
boundary, the floor plate, and roof plate [8]. Many of the molecular
mechanisms that control these processes have now been identified,
but originally these paradigms were established by transplantation
experiments using various ways to distinguish host and donor
tissue. In particular, grafting experiments established fate maps of
the neural plate and brain [4, 5, 9–12], the location and action of
organizers [8, 13], and also the time of competence during which
tissue can respond to organizer signals and the time when cells
become committed to a particular fate [14–16].
Avian model systems have been very popular for such studies,
because the development of their nervous system parallels that of
mammalian embryos in many aspects and has been described in
great detail. Unlike mammals, however, avian embryos are easily

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_16, © Springer Science+Business Media, LLC 2014

235
236 Andrea Streit and Claudio D. Stern

accessible and relatively cheap to obtain, and little specialized

equipment is needed for operations and for growing the embryos.
Most, if not all, transplantation experiments require a reliable
system to distinguish host and donor tissue to locate the graft, to
follow its progeny or examine cell behavior like axonal projections,
but also to establish whether, e.g., changes in gene expression or
neuronal morphology occur cell autonomously (e.g., within the
graft) or are induced in surrounding cells (e.g., in neighboring
tissues). Many studies have used transplantation of tissues labelled
with fluorescent dyes (e.g., [17], infected with retroviral vectors
(transplanted into resistant hosts) [18] and more recently tissues
from GFP transgenic chickens [19, 20]). However, one of the
most extensively used techniques is cross-species transplantations
generating chick-quail chimeras to provide permanent cell tracing
[9–12]. Quail and chick are closely related species; their early
development is fairly similar, but they differ very slightly in timing.
The chimeras generated by transplantation of neural tissue and
neural crest cells develop normally and are even able to hatch. Early
experiments to distinguish quail and chick tissue made use of the
fact that quail nucleoli are associated with a fair amount of hetero-
chromatin, which is absent in most other species including the
chick. Therefore, histological staining for DNA can differentiate
quail and chick tissue [21]. More recently, however, quail-specific
antibodies have become available, which recognize either all quail
cells or quail neurites [22, 23]. These are now frequently used and
their detection can be combined with other techniques like in situ
hybridization [24].
This chapter focuses on quail-chick transplantation of neural
tissue at early neural plate and at later neural tube stages. After
embryonic days 3–4, the brain becomes less accessible due to the
formation of blood vessels and extraembryonic membranes, while
the spinal cord remains accessible. The procedures described are used
to replace neural tissue from chick with the identical tissue from
quail (orthotopic) at the same stage of development (isochronic).
However, similar strategies can be used for heterotopic or heteroch-
ronic grafts as well as to any other tissue.

2 Materials

All procedures described below require two pairs of watchmakers’

forceps (number 5); one pair of coarse forceps, about 15 cm long;
one pair of small, fine scissors, with straight blades about 2 cm
long; Pasteur pipettes (short form), end lightly flamed to remove
sharp edges and rubber teats; container for egg waste; and small
beakers (50–100 ml). Instruments should be cleaned with lightly
soapy water, rinsed in distilled water, and washed in 70 % ethanol
before drying on a tissue. You need a good stereomicroscope with
 Quail-Chick Chimeras 237

transmitted light base and for in ovo work a cold light source
(fiber optics) for illumination from the top. Fertile hens’ or quails’
eggs are incubated in a humidified incubator at 38 °C until they
have reached the stage desired; staging of host and donor embryos
is performed according to Hamburger and Hamilton [25]. All
solutions are diluted from autoclaved stock solutions in distilled
water immediately before use; beakers for salines are autoclaved
before use.

2.1 Preparing Chick Operations on primitive streak to early somite stage (HH3+-8) chick
Hosts for New Culture host embryos are performed in modified New culture [26, 27]; at
this stage embryos are fragile and difficult to manipulate in ovo, and
survival rate in ovo is poor. On the other hand, in New culture
embryos can only be grown for 24–36 h even in an expert’s hands.
In addition to the above materials, this method requires:
1. Pyrex baking dish about 5 cm deep with 2 l capacity.
2. Watch glasses about 5–7 cm diameter.
3. Rings cut from glass tubing (approx. 27 mm outer diameter,
24 mm inner diameter and 3–4 mm deep; obtained from a
local glass blower).
4. 35 mm plastic dishes with lids (bacteriological grade).
5. Plastic box with lid for incubating culture dishes.
6. Pannett-Compton saline is prepared from two stock solutions,
which can be kept at 4 °C if autoclaved. Solution A: 121 g
NaCl, 15.5 g KCl, 10.42 g CaCl2 ⋅ 2H2O, 12.7 g MgCl2 ⋅ 6H2O,
H2O to 1 l. Solution B: 2.365 g Na2HPO4 · 2H2O, 0.188 g
NaH2PO4 · 2H2O, H2O to 1 l. To prepare working solution
just before use, mix (in order) 120 ml A, 2,700 ml H2O, and
180 ml B. Do not mix concentrated stocks of A and B.

2.2 Preparing Chick After HH8-9, operations are performed in ovo; under perfect
Hosts for In Ovo conditions embryos can be grown for a long time, even until hatch-
Operation ing. Collect the following materials in addition to those listed at
the beginning of the methods section.
1. Scalpel with No. 3 handle and No. 11 blades.
2. Plasticine or foam (from packaging) to make a ring for resting
eggs on their side.
3. PVC tape to seal the eggs.
4. 5 ml syringe with 21G needle (for removing albumen).
5. 1 ml syringe with 27G (or finer) needle (for ink injection).
6. 1 ml syringe with 21G needle (for antibiotics).
7. Paper tissues.
8. Indian ink (Pelikan Fount India or Winsor and Newton;
diluted 1:10 in saline).
238 Andrea Streit and Claudio D. Stern

9. Silicone grease in a 10 ml syringe (no needle).

10. 10× stock Tyrode’s saline: 80 g NaCl, 2 g KCl, 0.5 g
NaH2PO4⋅2H2O, and 10 g glucose in 1 l H2O. Prepared in
advance, autoclaved, and kept at 4 °C after opening. Dilute
this to 1× working solution with autoclaved water (about
100 ml are needed).
11. Antibiotics: 100× penicillin/streptomycin solution (Sigma
A9909).
12. 70 % ethanol is also required.

2.3 Preparing The following materials are required to harvest quail embryos:
Quail Donors
1. Glass Petri dish (10 or 15 cm diameter, depending on the
number of embryos to be collected).
2. Spoon spatula for collecting embryos.
3. Glass Pasteur pipette cut at the shoulder and fire polished
(for embryo transfer).
4. Rubber teats.
5. 500 ml Tyrode’s saline.
6. Dissecting microscope with transmitted light base.

2.4 Grafting To dissect quail and chick tissue for grafting, the following materi-
als are required:
1. 35 mm Sylgard-coated dish for dissecting (this should never
come into contact with fixative).
2. Entomological insect pins (A1; steel) for pinning out embryos
on the Sylgard dish.
3. Insect pins mounted on Pasteur pipettes or tungsten needles
(Goodfellow; 100 μm diameter, mounted on aluminum
holders or glass rods using sealing wax; sharpen by repeated
exposure of the tip to a very hot Bunsen flame).
4. 30G needles mounted on 1 ml syringes.
5. P20 Gilson pipette and yellow tips.
In addition, for in ovo transplantation in older embryos, the
following materials are needed:
6. Micro-knife (e.g., micro-feather microsurgery knives for eye
surgery 15E blade angle).
7. Aspirator tube (Sigma A5177).
8. 50 μl borosilicate glass capillaries (for trypsin injection) pulled
to fine injection needles using an electrode puller, tips broken
off (puller settings need to be determined; needles should be
fine enough to avoid fluid uptake by capillary forces but large
enough to deliver small amounts of trypsin by air pressure).
 Quail-Chick Chimeras 239

9. 50 ml 0.12 % trypsin (Difco) in Tyrode’s saline.

10. 5 % serum (any species) in Tyrode’s saline for stopping the
trypsin.

2.5 Fixing Embryos Embryos are fixed several hours or days after transplantation; they
and Analyzing Results can be analyzed by in situ hybridization or immunostaining to label
specific tissues or cell types, followed by labelling with quail-specific
antibodies to detect the graft by in situ hybridization using chick-
and quail-specific probes or by histological sectioning followed by
Feulgen and Rossenbeck staining to reveal nucleoli [21]. Depending
on the analysis, different fixatives are used. Fixing requires:
1. Petri dish for collecting embryos (for in ovo).
2. 3.5 or 10 mm Sylgard dish for pinning out embryos.
3. Insect pins (see above).
4. 7 ml glass vials and phosphate-buffered saline (PBS).
5. For whole-mount in situ hybridization, embryos are fixed in
4 % PFA, 1 mM EGTA in PBS, and 0.1 % Tween-20. Embryos
are stored in methanol.
6. For Feulgen staining and in situ hybridization on sections
[28], embryos are fixed in Zenker’s fixative: 50 g HgCl2, 25 g
K2Cr2O7, 10 g of Na2SO4 × 10 H2O in 1 l distilled water; before
use, add 5 ml glacial acetic acid to 100 ml of the solution: or
fixed in Carnoy’s fixative: 50 % ethanol, 11.1 % formaldehyde,
10 % glacial acetic acid [24, 28].
The most important point for further analysis is the detection
of quail tissue. While traditionally Feulgen staining has been used to
reveal the difference between chick and quail nucleoli [22, 29, 30],
now quail-specific antibodies are the favorite method. The mono-
clonal mouse antibody QCPN (Developmental Studies Hybridoma
Bank) labels all quail cells, while QN is specific for neurites [23].
We generally perform QCPN staining after whole-mount in situ
hybridization; this procedure requires:
1. Glass vials.
2. Pasteur pipettes.
3. Rubber teats.
4. Rocking platform.
5. PBS, blocking buffer: 1 % goat serum, 0.5 % Triton X-100 in
PBS.
6. Anti mouse IgG-HRP coupled (Jackson).
7. 100 mM Tris–HCl pH 7.4.
8. 50 mg/ml 3,3′-diaminobenzidine (DAB) in 100 mM Tris–
HCl pH 7.4.
9. 0.3 % H2O2 in 100 mM Tris–HCl pH 7.4.
240 Andrea Streit and Claudio D. Stern

It is also possible to perform in situ hybridization with a probe

directed against sequences that differ between chick and quail
(most likely the 3′ UTR of a specific gene of interest) to distinguish
transcripts produced by cells of the graft and donor [31].

3 Methods

3.1 Preparing Quail To dissect the donor tissue, quail embryos are first removed from
Donor Embryos for the egg and cleaned using the following steps.
Grafting
1. Remove quail eggs from incubator. Using fine scissors, gently
tap the shell near the blunt end of the egg to penetrate the
shell. Use the tip of the scissors to cut off a small cap of shell;
avoid damaging the yolk.
2. Pour thin egg white into waste; use the scissors to help and cut
through the rather thick albumen if required (see Note 1).
3. Once most albumen is removed, turn the yolk by stroking it
very gently with the sides of the scissors to make the embryo
become visible on top of the yolk.
4. Using the scissors make four cuts into the vitelline membrane
around the embryo; make sure that all the cuts meet (see Note 2).
5. Using the spoon, pick up the square of embryo and membrane
including a little yolk; try to collect as little yolk as possible.
6. Transfer the embryo including yolk and membrane into the
large Petri dish containing Tyrode’s saline under a dissecting
microscope by sliding it carefully off the spoon. Using fine for-
ceps, turn the square of yolk/membrane/embryo so you can
see the embryo.
7. Once all donor embryos have been collected in the Petri dish,
separate the embryos from adhering yolk. Work at low magni-
fication; use two pairs of forceps to pick up a corner of the
vitelline membrane with one and slowly but steadily fold it
back, steadying the yolk with the other. Make sure that the
membrane and embryo remain submerged in saline. The
embryo will remain attached to the membrane. If not, peel off
the membrane completely, and then use the forceps gently to
remove the embryo from the underlying yolk.
8. Using the wide-mouth Pasteur pipette, pick up the embryo,
with (better) or without membrane, and transfer to a 10 cm
dish with clean saline. To clean the embryo use a fire-polished
Pasteur pipette and gently blow saline over it; this will
remove yolk particles. The embryos are now ready for dissec-
tion and grafting and can be kept for 1–2 h before proceed-
ing further.
 Quail-Chick Chimeras 241

3.2 Preparing Chick At primitive streak and neural plate stages, operations in avian
Hosts for New Culture embryos are most easily performed in New culture. The method
described below is based on New’s original technique [27] modi-
fied by Stern and Ireland [26]. This modified culture method uses
rings cut from glass tubes, instead of rings bent from glass rods and
35 mm plastic dishes instead of glass watch glasses resting inside a
large glass Petri dish. The rings cut from tubing generate a slightly
rough surface that grips the vitelline membrane and therefore
allows easy transfer of the culture into the plastic dish. To set up
the cultures, proceed as follows (Fig. 1):
1. Remove eggs from the incubator.
2. Fill the large Pyrex dish with about 1.5 l of Pannett-Compton
saline; the volume should be large enough that eggs yolks are
submerged completely.
3. To open an egg, tap its blunt egg with the coarse forceps and
carefully remove pieces of the shell. Discard the thick albumen
into the waste bucket, assisted with the coarse forceps. Collect
the thin albumen in a small beaker (see Note 3).
4. Carefully tip the yolk into the Pyrex dish containing saline, tak-
ing care not to damage the membrane on the edges of the
shell. Carefully turn the yolk with the side of the coarse forceps
so the embryo is facing upwards. Now place a watch glass and
a glass ring into the dish.
5. Cut the vitelline membrane enveloping the yolk just below the
equator using small scissors; you can use one pair of forceps to
push the yolk around gently while continuing to cut all the way
around its circumference (see Note 4).
6. Using both pairs of fine forceps, peel the vitelline membrane
with the embryo attached slowly but very steadily off the yolk.
Use one pair of forceps to pull the edge of the membrane
slightly upwards (about 25–30° angle from the yolk surface)
and the other to hold the yolk down. Do not pull tangentially
along the yolk: this may detach the embryo from the mem-
brane. Do not stop during this process. The embryo should
come off with the membrane (see Note 5).
7. Turn the vitelline membrane with the inner face containing the
embryo pointing upwards and slide it, preserving its orientation,
onto the watch glass. Place the ring over it so that the mem-
brane protrudes around the ring and the embryo sits in its
center. Remove the watch glass, ring, and embryo from the
dish; tilt the assembly gently to pour off some saline while
steadying the ring with one finger (see Note 6).
8. Dry the bottom of the watch glass on some tissue. Using fine
forceps, carefully wrap the lose edges of the vitelline membrane
242 Andrea Streit and Claudio D. Stern

Fig. 1 Setting up New culture. (a) Set up. (b) Instruments. (c) Opening eggs. (d) Removing albumen. (e) Yolks
in Pyrex dish. (f) Cutting the vitelline membrane at the equator. (g) Removing membrane and embryo from the
yolk. (h) Membrane with embryo facing upwards on watch glass. (i) Placing glass ring on the membrane.
(j) Culture assembled on ring and watch glass. (k) Removing assembly form Pyrex dish. (l) Cleaning the culture.
(m) Primitive streak stage embryo on watch glass. (n) Embryo after removal of a piece of neural plate.
(o) Setting up the culture in a Petri dish with albumen. (p) Finished New culture

over the edge of the ring, all the way around its circumference.
Pull the membrane slightly so its bottom is smooth and free
from wrinkles, but be careful not to pull so tight that it breaks
(see Note 7).
9. Using the fire-polished Pasteur pipette, rinse the outside of the
ring to remove yolk particles. If there is a lot of egg albumen
remaining under the membrane, lift the ring gently and use the
 Quail-Chick Chimeras 243

Pasteur pipette to remove it. Clean the yolk over and around
the embryo using clean saline; be careful not to dislodge the
embryo form the membrane. Damaged embryos do not grow
well or normal. If there is a lot of vitelline membrane inside the
ring, trim off the excess with the fine scissors while lifting the
edges with fine forceps. At this stage embryos are ready for
transplantation and can be kept on the bench for some time;
make sure they remain well submerged under saline and there
is sufficient saline on the watch glass. If keeping them for a few
hours, place them on a wet tissue and cover with a large glass
or plastic plate/dish.
10. To finalize the cultures after transplantations, work under the
microscope; carefully remove any remaining saline, both inside
and outside the ring. Drying helps the graft and host tissue to
heal faster. During culture the embryo and the inside of the
ring must remain dry.
11. Pour some thin albumen (about 2–3 mm thick layer) on the
bottom of a 35 mm Petri dish. Using fine forceps, slide the
ring with vitelline membrane off the watch glass, and transfer
it to the dish; lower it onto the albumen making sure that no
air is trapped underneath. Press the ring lightly onto the bot-
tom of the dish using two forceps to allow it to adhere.
12. Remove the excess albumen if its level comes close to the edge
of the ring. At this point, remove any remaining liquid from
inside the ring using a fire-polished Pasteur pipette. The vitel-
line membrane should be slightly dome shaped; this will help to
drain off any fluid that accumulates during culture (see Note 8).
13. To seal the Petri dish, wet the inside of the lid with a thin film of
albumen all around the edge, discard the excess, and place onto
the bottom part. Press lid down slightly to seal (see Note 9).
14. Place the dish in a plastic box containing a piece of wet tissue,
seal the box, and place it into an incubator at 38 °C.

3.3 Preparing Chick Later-stage embryos are generally grown in ovo, which allows
Hosts for In Ovo embryos to grow for long periods, even until hatching. Eggs must
Culture be incubated lying on their side, so the yolk turns with the embryo
facing upwards to make it accessible for manipulation. To prepare
hosts for in ovo operations, use the following procedure (Fig. 2):
1. Remove the eggs from the incubator, place one egg onto the
egg rest, and clean with 70 % ethanol; be careful not to rotate
the egg.
2. Hold the 5 ml syringe with 21G needle nearly vertical, and
insert the needle into the blunt end of egg until you feel the
shell at the bottom. Remove about 1 ml egg albumen and
discard. This lowers the embryo away from the top of the shell.
244 Andrea Streit and Claudio D. Stern

Fig. 2 In ovo culture. (a) Eggs are incubated on their side and placed on egg rest. Circle labels the blunt end of
the egg. (b) Blunt end is used to remove albumen with 5 ml syringe. (c) Inserting syringe. (d) Scoring the win-
dow. (e) Windowed egg. (f) Windowed egg with Vaseline border surrounding the window. (g) Embryo before ink
injection. (h) Ink injection using a 1 ml syringe. (i) Ink injection: Embryo is clearly visible. (j) Embryo after ink
injection. (k) Removing albumen to lower the embryo. (l) Eggs sealed with tape

3. Using the scalpel score a 1 cm × 1 cm square on the top of the

shell and lift it up using the blade or a pair of forceps.
4. Moisten the white membrane under the shell with a little
Tyrode’s saline and remove it with fine forceps, to the edge of
the window. Be careful to avoid damage to the embryo
underneath.
5. Add saline to the egg so that the embryo floats up to the level
of the window.
6. Take up diluted ink into a 1 ml syringe with a 27G needle;
make sure there are no air bubbles in the syringe. Insert the
needle under the vitelline membrane almost parallel to the yolk
surface; choose a position away from the embryo proper, and
point the needle towards and underneath the embryo. Inject
about 50–100 μl ink; the amount should be kept as little as
possible. Avoid moving the needle after initial penetration;
otherwise the hole will become too big and yolk and/or ink
 Quail-Chick Chimeras 245

will leak out. The embryo should now be clearly visible on a

dark background (see Note 10).
7. Line the shell window with a shallow edge of Vaseline by eject-
ing it from the syringe. This will allow you to cover the embryo
with a drop of saline during the operation for moisture and
good optics. Fill the chamber with saline until there is a good
dome of fluid. Adjust the fiber optic light so the light shines
tangentially onto the embryo, which is now ready for manipu-
lation (see Note 11).
8. Once the embryo has received the graft and it is in the correct
position, remove the saline very carefully from above the
embryo using a Pasteur pipette. Watch under low magnifica-
tion to ensure the graft does not move. If required, reposition
it using a pin mounted on a Pasteur pipette.
9. Carefully insert the 5 ml syringe with 21G needle into the
original hole in the blunt end of the shell. Carefully remove
about 3 ml egg albumen to lower the embryo to its original
position.
10. Add 1–2 drops (50–100 μl) of antibiotic solution (see Note 12).
11. Use a tissue with 70 % ethanol to wipe the Vaseline off the
shell; dry the shell thoroughly with another tissue.
12. Cut a piece of PVC tape about 3–4 cm long, stretch it slightly,
and allow it to relax again. To seal the egg, place the tape over
the window and carefully smoothen any wrinkles without put-
ting too much pressure on the shell. Make sure the edges of
the tape are firmly attached to the shell; if not they will roll up
and expose the embryo.
13. Incubate the egg in a well-humidified incubator at 38 °C; next
day you can turn the egg window side down which helps to
keep the embryo moist and improves their development.
Incubate for the desired period; generally the 2–4 day survival
rate should be 80–90 % (see Note 13).

3.4 Grafting The procedure below describes orthotopic, isochronic neural plate
Procedure: Quail grafts from quail donors into chick hosts at HH3+/4. The same
Neural Plate into Chick method can be applied for heterotopic and heterochronic grafts or
Hosts in New Culture transplantation of other tissues.
1. To prepare the host follow the procedure described under
Subheading 3.1 until step 9.
2. Replace one of the eyepieces of the microscope with an eyepiece
containing a graticule; a protractor, to measure angles, is par-
ticularly useful.
3. Place a host embryo from Subheading 3.1, step 9 (kept on a
watch glass) under the microscope and center the graticule on
the node.
246 Andrea Streit and Claudio D. Stern

4. Define the area to be replaced using the graticule coordinates.

New culture embryos face ventral side up; to reach the ecto-
derm lower layers need to be removed. Fold back the endo-
derm and mesoderm above the area to be replaced by quail
tissue using 30G needles on a 1 ml syringe serving as a holder.
Use the sharp side of the needle gently to score both layers on
three sides (e.g. anterior, posterior and lateral leaving them
attached medially); then using the back of the needle carefully
peel away the endoderm and mesoderm overlying the area to
be grafted.
5. Use tungsten needles, mounted insect pins or 30G needles to
cut out the region of the neural plate to be replaced by quail
tissue (see Note 14; Fig. 1n). Set aside the host embryo and
turn to the quail donor.
6. Pin out a quail embryo of the same stage, ventral side up, on a
Sylgard dish containing Tyrode’s saline; use insect pins through
the extraembryonic region to stretch the embryo slightly.
7. Locate the area to be grafted using the graticule, remove the
endoderm and mesoderm as described in step 4 and excise the
underlying neural plate as described in step 5.
8. Working under low magnification, use a Gilson pipette to pick
up the graft in 3–5 μl saline.
9. Move the host embryo back on the stage and, working under
low magnification, place the graft close to the target site.
10. Use mounted insect pins to move the graft into the hole cut
previously. It is crucial to maintain the apical-basal orientation
of the graft; after excision the ectoderm generally contracts
basally and the tissue curves slightly.
11. Carefully remove all the liquid outside the ring and most of the
fluid inside the ring.
12. Flip back the mesoderm and endoderm to secure the graft in
its position and carefully remove all remaining liquid inside the
ring. Excess liquid around the grafted area should be removed
using pulled capillaries on aspirator tubes (see Note 15).
13. Now finish setting up the cultures by following steps 10–13 in
Subheading 3.2. Make sure that the dome of albumen is rather
flat; a high dome causes too much tension and the grafts do
not integrate properly. Leave embryos on the bench for
30–60 min to let the grafts heal and then proceed to step 14
in Subheading 3.2.

3.5 Grafting This section describes orthotopic and isochronic neural tube grafts
Procedures: Quail in embryos older than HH9; as with transplantations in younger
Neuroectoderm into embryos described above, the same techniques can be used for
Chick Host In Ovo heterotopic and heterochronic experiments. First, prepare the
donor embryo, and then turn to the host.
 Quail-Chick Chimeras 247

1. To prepare the quail donor, pin out the embryo dorsal side up
on a Sylgard dish in Tyrode’s saline and place under a dissect-
ing microscope with transmitted light base.
2. Using tungsten needles or a micro-knife, make a longitudinal
incision into the ectoderm dorsal to the neural tube on both of
its sides.
3. Replace the Tyrode’s with trypsin solution; working at high mag-
nification peel the ectoderm away from the neural tube using the
back of a 30G needle. In the same way, gently scrape any loosely
attached cells (neural crest depending on stage) off the neural
tube and free it from the adjacent tissues (see Note 16).
4. Progressively separate the neural tube from the underlying
notochord using a micro-knife to push it from side to side,
allowing the Trypsin to penetrate; when completely detached,
cut the neural tube transversely at its anterior and posterior
ends to free it.
5. Remove the excised neural tube using a Gilson pipette set to
3–5 μl and place into a 35 mm Petri dish containing saline with
5 % serum (see Note 17). Graft can be kept on ice until use.
6. Prepare the chick host by following steps 1–7 in Subheading 3.2.
Use a mounted insect pin make a small hole into the vitelline
membrane just over the area to be operated. The hole should
be as small as possible.
7. Replace the drop of Tyrode’s saline with trypsin solution and
follow steps 2–4 above to excise the same section of the neural
tube as in the quail donor.
8. Remove the excised neural tube using a Gilson pipette set to
3–5 μl and replace the trypsin solution with fresh Tyrode’s
saline twice.
9. Pick up the graft using a Gilson pipette; rinse in Tyrode’s saline
without serum before transferring it to the host.
10. Using a Gilson pipette and working under low magnification,
transfer the graft into the saline bubble over the host embryo.
11. Use a mounted insect pin or 30G needle to place the graft into
the hole made by removal of the host neural tube. Preserve
anterior-posterior and dorsoventral orientation (see Note 18).
12. Once the transplant is in position, carefully remove the saline
using a Pasteur pipette while observing under low magnifica-
tion. If needed, reposition the graft using a mounted pin.
13. Finish the egg by following steps 8–13 in Subheading 3.2.

3.6 Detecting As outlined above, grafted embryos can be analyzed in various ways
Quail Tissue depending on the question; these include whole-mount or section in
situ hybridization and tissue- or cell-specific immunohistochemistry.
248 Andrea Streit and Claudio D. Stern

All of these techniques can be combined with the antibody staining

using the quail-specific antibody QCPN. We generally perform
whole-mount in situ hybridization for embryos up to embryonic day
3 and section in situ for older embryos followed by QCPN staining.
The protocol below describes the whole-mount procedure; for other
applications, see [19, 20, 28]. For embryos incubated to HH13 or
older, it is a good idea to treat them with 6 % H2O2 in PBS, 0.1 %
Tween after fixation or rehydration after storing in methanol
(see [24]); this reduces background for the in situ hybridization and
immunostaining signal.
1. After developing the in situ hybridization color reaction, wash
and fix embryos as normal. Remove fixative by washing in PBS
three times for 10–30 min depending on the age of the embryo.
2. Block embryos for 1–3 h in blocking buffer at room tempera-
ture on a rocking platform.
3. Replace blocking solution with QCPN antibody solution
(dilute antibody in blocking buffer; determine concentration
for each batch of antibody) and incubate at 4 °C on a rocking
platform for two nights.
4. Remove antibody and wash embryos in PBS for five to seven
times 1 h each; for older embryos leave the final wash over-
night at 4 °C.
5. Incubate embryos in secondary antibody (generally 1:1,000 in
blocking buffer, but may need titration) over one or two nights
at 4 °C.
6. Wash as in step 4.
7. Wash twice for 15 min in 100 mM Tris–HCl pH 7.4 (see Note
19); in the second wash measure the volume (generally 1 ml is
sufficient).
8. Add DAB from the stock to vial with embryos to a final
concentration of 0.5 mg/ml; incubate for 10–15 min rocking
in the dark (see Note 20).
9. Add the appropriate amount of H2O2 from the 0.3 % stock to
the embryos to make a final concentration of 0.003 %.
10. Incubate in the dark until brown color develops, normally
within 5–10 min; check occasionally using illumination from
the top on a white background.
11. Stop reaction by rinsing several times in distilled H2O to
remove residual substrate and postfix embryos in 4 %
formaldehyde.
12. Embryos can now be cleared, photographed, and embedded
for paraffin or vibratome sectioning as required.
 Quail-Chick Chimeras 249

4 Notes

1. Remove as much albumen as possible; the yolks move less in

the next step.
2. Make sure scissors are cleaned after each egg; crusts of egg yolk
make the vitelline membrane stick to the scissors and the
embryos tend to sink into the yolk. Do not hesitate when mak-
ing the cuts; work rapidly so the embryos do not move.
3. Try to remove as much albumen as possible; albumen adhering
to the vitelline membrane makes the following steps more
difficult.
4. Make sure to cut at or slightly below the equator; otherwise it
will be difficult to fit the membrane around the ring in the next
step.
5. Occasionally embryos remain attached to the yolk particularly
at early primitive streak stages. Keep the membrane because it
can be used for other embryos, in case you accidentally punc-
tured a membrane. The embryo can also be retrieved from the
yolk but requires thorough cleaning: use a pair of fine forceps,
close them, and gently push the edges of the extraembryonic
region away from the yolk. Work all the way around the edge
of the embryo. Transfer the embryo using a wide-mouthed
pipette into a dish with fresh saline, ventral side up gently blow
the attached yolk plug off the embryo using a Pasteur pipette.
The embryo can now be returned to the membrane; make sure
to keep it orientation: embryos do not grow with the ventral
side on the membrane.
6. Make sure you do not turn the membrane inside out; embryos
do not grow on the outer surface.
7. Be careful not to make any holes into the membrane; this will
allow albumen to accumulate inside the ring and prevent the
embryo from growing.
8. Be careful not to use too much albumen for grafted embryos;
this will increase the tension and prevent healing.
9. Sealing is important to prevent condensation on the lid during
incubation.
10. Too much or too high concentration of ink is toxic for the
embryos. Recently Pelikan Indian ink has become difficult to
obtain; if you are using other brands they need to be tested for
toxicity.
11. A dome of liquid considerably improves the optics and also
prevents the embryo from drying. The latter is critical as dry-
ing out reduces the survival rate.
250 Andrea Streit and Claudio D. Stern

12. Antibiotics are generally only required for long culture peri-
ods. Normally embryos survive well for 2–3 days without this.
13. Low survival rates can be due to a number of factors. The most
common problems are dehydration, damage to critical blood
vessels, and infection. Working in a drop of saline helps to alle-
viate dehydration. To avoid infection ensure you use clean
solutions; instruments can be cleaned periodically while work-
ing in distilled H2O and ethanol (make sure it evaporates
before using to operate), while tungsten needles are flamed
periodically to keep them sharp.
14. Be careful not to make any holes in the vitelline membrane.
15. Drying the area surrounding the graft greatly improves heal-
ing; be careful not to suck up the graft into the capillary.
16. Be patient; the trypsin works almost by itself, and it generally
sufficient to push the adjacent tissues away using your instru-
ments. Avoid going too deep and cut a hole into the endo-
derm; this will make ink and yolk leak out.
17. This inactivates trypsin; this is important because the tissue
should be exposed to proteolytic enzymes for as short as pos-
sible to avoid disintegration.
18. If you encounter difficulties to preserve orientation, mark one
end of the neural tube with a small crystal of carmine powder.
19. Make sure the pH is properly adjusted to 7.4; the reaction is
pH sensitive.
20. DAB is carcinogenic; make sure to wear appropriate protective
clothing (lab coat, gloves) and consult the local health and
safety regulations for inactivation and disposal.

Acknowledgment

This work is supported by the BBSRC, NIH, and ERC. We thank

Anneliese Norris for assistance with New culture photography.

References
1. Stern CD (2005) Neural induction: old prob- 3. Spemann H, Mangold H (1924) Über
lem, new findings, yet more questions. Induktion von Embryonalanlagen durch
Development 132:2007–2021 Implantation artfremder Organisatoren.
2. Waddington CH (1934) Experiments on Wilhelm Roux’ Arch Entwicklungsmech Org
embryonic induction. Part I: the competence 100:599–638
of the extra-embryonic ectoderm. Part II: 4. Cobos I, Shimamura K, Rubenstein JL,
experiments on coagulated organisers in the Martinez S, Puelles L (2001) Fate map of the
chick. Part III: A note on inductions by chick avian anterior forebrain at the four-somite
primitive streak transplanted to the rabbit stage, based on the analysis of quail-chick
embryo. J Exp Biol 11:211–227 chimeras. Dev Biol 239:46–67
 Quail-Chick Chimeras 251

5. Fernandez-Garre P, Rodriguez-Gallardo L, orthotopic and heterotopic grafts. Development

Gallego-Diaz V, Alvarez IS, Puelles L (2002) 118:527–538
Fate map of the chicken neural plate at stage 4. 18. Fekete DM, Cepko CL (1993) Retroviral
Development 129:2807–2822 infection coupled with tissue transplantation
6. Garcia-Martinez V, Alvarez IS, Schoenwolf GC limits gene transfer in the chicken embryo.
(1993) Locations of the ectodermal and non- Proc Natl Acad Sci USA 90:2350–2354
ectodermal subdivisions of the epiblast at stages 19. Barraud P, Seferiadis AA, Tyson LD, Zwart
3 and 4 of avian gastrulation and neurulation. MF, Szabo-Rogers HL, Ruhrberg C, Liu KJ,
J Exp Zool 267:431–446 Baker CV (2010) Neural crest origin of olfac-
7. Eagleson GW, Harris WA (1990) Mapping of tory ensheathing glia. Proc Natl Acad Sci USA
the presumptive brain regions in the neural 107:21040–21045
plate of Xenopus laevis. J Neurobiol 21: 20. Sabado V, Barraud P, Baker CV, Streit A (2012)
427–440 Specification of GnRH-1 neurons by antago-
8. Vieira C, Pombero A, Garcia-Lopez R, Gimeno nistic FGF and retinoic acid signaling. Dev Biol
L, Echevarria D, Martinez S (2010) Molecular 362:254–262
mechanisms controlling brain development: an 21. Feulgen R, Rossenbeck H (1924) Mikroskopisch-
overview of neuroepithelial secondary organiz- chemischer Nachweis einer Nukleinsaeure vom
ers. Int J Dev Biol 54:7–20 Typus der Thymonukleinsaeure und die daruf
9. Couly G, Le Douarin NM (1988) The fate beruhende elecktive Faerbung von Zellkernen in
map of the cephalic neural primordium at the miroskopischen Praeparaten. Hoppe-Seyler’s Z
presomitic to the 3-somite stage in the avian Physiol Chem 135:203–248
embryo. Development 103:101–113 22. Teillet MA, Ziller C, Le Douarin NM (2008)
10. Couly G, Le Douarin NM (1990) Head mor- Quail-chick chimeras. Methods Mol Biol
phogenesis in embryonic avian chimeras: evi- 461:337–350
dence for a segmental pattern in the ectoderm 23. Tanaka H (1990) Selective motoneuron out-
corresponding to the neuromeres. Development growth from the cord in the avian embryo.
108:543–558 Neurosci Res Suppl 13:S147–S151
11. Couly GF, Le Douarin NM (1985) Mapping of 24. Streit A, Stern CD (2001) Combined whole-
the early neural primordium in quail-chick chi- mount in situ hybridization and immunohisto-
meras. I. Developmental relationships between chemistry in avian embryos. Methods
placodes, facial ectoderm, and prosencephalon. 23:339–344
Dev Biol 110:422–439 25. Hamburger V, Hamilton HL (1951) A series of
12. Couly GF, Le Douarin NM (1987) Mapping of normal stages in the development of the chick
the early neural primordium in quail-chick chi- embryo. J Morph 88:49–92
meras. II. The prosencephalic neural plate and 26. Stern CD, Ireland GW (1981) An integrated
neural folds: implications for the genesis of experimental study of endoderm formation in
cephalic human congenital abnormalities. Dev avian embryos. Anat Embryol (Berl) 163:
Biol 120:198–214 245–263
13. Martinez S, Wassef M, Alvarado-Mallart RM 27. New DAT (1955) A new technique for the cul-
(1991) Induction of a mesencephalic pheno- tivation of the chick embryo in vitro. J Embryol
type in the 2-day-old chick prosencephalon is Exp Morph 3:326–331
preceded by the early expression of the homeo- 28. Lassiter RN, Dude CM, Reynolds SB, Winters
box gene en. Neuron 6:971–981 NI, Baker CV, Stark MR (2007) Canonical
14. Baker CV, Stark MR, Marcelle C, Bronner- Wnt signaling is required for ophthalmic tri-
Fraser M (1999) Competence, specification geminal placode cell fate determination and
and induction of Pax-3 in the trigeminal plac- maintenance. Dev Biol 308:392–406
ode. Development 126:147–156 29. Storey KG, Selleck MA, Stern CD (1995)
15. Bhattacharyya S, Bronner-Fraser M (2008) Neural induction and regionalisation by differ-
Competence, specification and commitment to ent subpopulations of cells in Hensen’s node.
an olfactory placode fate. Development Development 121:417–428
135:4165–4177 30. Le Douarin N, Dieterlen-Lievre F, Creuzet S,
16. Groves AK, Bronner-Fraser M (2000) Teillet MA (2008) Quail-chick transplanta-
Competence, specification and commitment in tions. Methods Cell Biol 87:19–58
otic placode induction. Development 127: 31. Izpisua-Belmonte JC, De Robertis EM, Storey
3489–3499 KG, Stern CD (1993) The homeobox gene
17. Guthrie S, Prince V, Lumsden A (1993) goosecoid and the origin of organizer cells in
Selective dispersal of avian rhombomere cells in the early chick blastoderm. Cell 74:645–659
 Chapter 17

RNAi-Based Gene Silencing in Chicken Brain Development

Irwin Andermatt and Esther T. Stoeckli

Abstract
The mouse is the most commonly used vertebrate model for the analysis of gene function because of the
well-established genetic tools that are available for loss-of-function studies. However, studies of gene func-
tion during development can be problematic in mammals. Many genes are active during different stages of
development. Absence of gene function during early development may cause embryonic lethality and thus
prevent analysis of later stages of development. To avoid these problems, precise temporal control of gene
silencing is required.
In contrast to mammals, oviparous animals are accessible for experimental manipulations during
embryonic development. The combination of accessibility and RNAi-based gene silencing makes the
chicken embryo a powerful model for developmental studies. Depending on the time window during
which gene silencing is attempted, chicken embryos can be used for RNAi in ovo or cultured in a domed
dish for easier access during ex ovo RNAi. Both techniques allow for precise temporal control of gene
silencing during embryonic development.

Key words Neural development, Chicken embryo, Artificial miRNA, RNA interference,
Electroporation, Gene silencing

1 Introduction

The chicken embryo has been used to study developmental pro-

cesses for a long time. The advantage of the chicken as a model is
its easy accessibility for experimental manipulation during embry-
onic development. Compared to the mouse, however, the chicken
cannot be manipulated with the same powerful genetic tools.
Although discovered a couple of years ago, chicken embryonic
stem cells cannot be maintained and manipulated in vitro.
Furthermore, size of the adult animals and long generation time
would also be unfavorable features of the chicken with respect to
genetic approaches.
However, the development of RNAi-based loss-of-function
approaches has overcome these problems and made the chicken
embryo a great model organism for developmental studies [1–4].

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_17, © Springer Science+Business Media, LLC 2014

253
254 Irwin Andermatt and Esther T. Stoeckli

In fact, the precise temporal control and the cell-type specificity of

gene silencing provide unique opportunities for the analysis of
gene function during development.
Since many genes involved in neural development are also
involved in the development of other tissues, such as the heart,
classical genetic knockout strategies can lead to early lethality. One
approach to overcome this restriction in the mouse model is a con-
ditional, tissue-specific knockout of a gene of interest provided that
suitable Cre lines are available to excise the floxed target gene.
Conversely, in chicken, organ- or even cell-type specificity is easy to
achieve and can be combined with temporal control of gene
silencing.
We have developed approaches for temporally and spatially
controlled gene silencing based on RNAi in chicken embryos [4–6].
Temporal control can be achieved with both in ovo and ex ovo
RNAi by using long dsRNA or plasmid-based miRNA/shRNA for
gene silencing [1, 4]. The latter, miRNA-based RNAi comes with
the additional advantage of cell-type specificity of gene silencing.
Both dsRNA- and miRNA-based RNAi can be used to silence sev-
eral genes at the same time.
For gene analysis during early stages, when embryos can easily
be accessed through a window in the eggshell, in ovo RNAi is the
method of choice, as it has higher survival rates compared to ex
ovo RNAi. To analyze brain function during late developmental
stages, embryos can be cultured ex ovo (shell-less) in a dish which
enables direct access to the desired brain areas for injection and
electroporation [3, 7].
Timing is important! Keep in mind that RNAi does not remove
the preexisting protein. Therefore, electroporation has to be done
before the protein of interest has accumulated. It is thus necessary
to carefully analyze the temporal expression of the target gene. We
routinely use in situ hybridization to analyze gene expression dur-
ing embryonic development [8]. Timing of injection and electro-
poration of dsRNA or miRNA is determined by the temporal
expression of the target gene, not by the time point of analysis or
the developmental milestones of the part of the nervous system
that is analyzed. In general, injection and electroporation are easier
at younger stages. However, cell proliferation will dilute the active
RNA-induced silencing complex (RISC) loaded with the specific
siRNA produced from the injected dsRNA or the miRNA. Thus,
gene silencing will be very effective over long periods of time in
neurons but will be less efficient in cells that keep proliferating after
electroporation.
Once manual skills for handling embryos and efficient injec-
tion and electroporation have been acquired, RNAi will provide
results on gene function very rapidly. For a first approach, it is
 RNAi in Brain Development 255

often easiest to just use long dsRNA prepared by in vitro

transcription from cDNA fragments. For thorough functional
analysis of a target gene, miRNA-based plasmids can be used in a
cell-type-specific manner. Since these miRNAs are coupled to a
fluorescent protein, efficiency of electroporation is directly visual-
ized [4, 9]. These vectors can be used in combinatorial knock-
downs with different tissue-specific promoters or in combination
with misexpression constructs resistant to the miRNA to perform
rescue experiments.
An excellent example demonstrating the power of precise
temporal control of gene silencing is provided by our discovery
of a role of the morphogen Shh in axon guidance [10–12].
Morphogens are required for cell differentiation and patterning
of the nervous system during early stages of development. Thus,
precise temporal control of gene silencing was key to our finding
of a direct and an indirect effect of Shh on post-crossing commis-
sural axon guidance [10–12]. Loss of Shh function during the
morphogenesis phase of neural development would have pre-
vented these findings as cell types in the neural tube would not
have been induced properly.
Here, we describe how to culture chicken embryos in ovo or
ex ovo for subsequent electroporation of miRNA constructs
(described in [4]) or long dsRNA [1, 3] to analyze neural develop-
ment. In ovo RNAi is well-suited for manipulation of embryos at
young stages and limited to about the fourth day of embryonic
development [10, 13]. After these stages the brain of the embryo
is no longer easily accessible. For manipulations at older stages,
chicken embryos can be transferred from the egg to a plastic dish
enabling injection and electroporation of diverse brain regions at
late stages [3, 7].
To study neural crest derivatives, injection and electroporation
have to be carried out in the first 2.5 days of development.
Injections into the eye are easier in the first 3–4 days, as the poor
elasticity of the sclera makes good injections without leakiness
more difficult at later stages. The cerebellum, for instance, starts to
emerge very late, only after about 1 week of embryonic develop-
ment. At this stage, the head needs to be fixed for injection in a
more upright position, and the blood vessels at the back of the
head can be used as landmarks to guide the injection needle [3].
We include two protocols in this chapter, one for injections
and electroporation of the neural tube within the first 3–3.5 days
of development and the second one for electroporation of the cer-
ebellum at late stages (HH34–36; Hamburger and Hamilton
stages 34–36 [14]).
256 Irwin Andermatt and Esther T. Stoeckli

2 Materials

2.1 Windowing 1. Fertilized eggs from a local hatchery.

the Eggs 2. Incubator set at 38.5 °C and 45 % humidity. We use two incu-
bators, one to incubate eggs before they are windowed and a
second one for incubation during the experiment (e.g.,
Heraeus/Kendro Model B12, Kendro Laboratory Products,
Germany, or Juppiter 576 Setter + Hatcher; FIEM, Italy).
3. Facial tissues.
4. 70 % ethanol.
5. Paraffin wax (Paraplast tissue embedding medium).
6. Heating plate set at 80 °C to melt paraffin.
7. Paint brush.
8. Scalpel.
9. Scotch tape or coverslips (24 × 24 mm).
10. Soldering iron (when eggs are closed with coverslips).
11. Sterile syringe with 18G needle.
12. Fine scissors.

2.2 Ex Ovo Culture 1. Fertilized eggs from a local hatchery.

2. Domed dish with a diameter of 80 mm and a depth of 40 mm
(these dishes are produced for the food industry from oriented
polystyrene (OPS; Bellaplast, Altstaetten, Switzerland)).
3. Lid for domed dish (we use the lid of a 10 cm Ø petri dish).
4. Incubator set at 38.5 °C and 45 % humidity. We use two incu-
bators, one to incubate eggs and a second one for incubation
of embryos in the dishes (e.g., Heraeus/Kendro Model B12,
Kendro Laboratory Products, Germany, for dishes; Juppiter
576 Setter + Hatcher; FIEM, Italy, for eggs).
5. Facial tissues.
6. 70 % ethanol.

2.3 In Ovo and Ex 1. Borosilicate glass capillaries (outer Ø/inner Ø:

Ovo Electroporation 1.2 mm/0.68 mm; World Precision Instruments; 1B120F-4).
2. Glass needle puller (Narishige, Japan; PC-10).
3. Square wave electroporator (BTX ECM 830).
4. Spring scissors (Fine Science Tools; 15003-08).
5. Dumont #5 forceps (Fine Science Tools; 11252-20).
6. For in ovo electroporation: Platinum electrodes (4 mm length,
4 mm distance between cathode and anode).
 RNAi in Brain Development 257

7. For ex ovo electroporation: Platelet electrode of 7 mm diameter

(Tweezertrodes Model #520, BTX Instrument Division,
Harvard Apparatus, Holliston, MA, USA, also see http://
www.btxonline.com/tweezertrodes/).
8. Spritz bottle filled with ddH2O.
9. Polyethylene tubing (Ø 1.24 mm).
10. 0.2-μm filter (Sarstedt, Switzerland).
11. Trypan blue solution, 0.4 % (Invitrogen).
12. 20× phosphate-buffered saline (PBS).
13. Sterile PBS.
14. Spatula bent to a hook (ex ovo electroporation).

3 Methods

3.1 Windowing Eggs To obtain access to the developing embryo, a window is cut into
the eggshell (Fig. 1). Independent of the time of electroporation,
eggs are windowed at the second or third day of incubation (E2
and E3, respectively; see Note 1).
1. Incubate eggs at 38.5 °C and 45 % humidity (see Notes 2–4).
2. Let embryos develop until they have reached the desired stage
for the experimental manipulation. Staging of the embryo is
done according to Hamburger and Hamilton [14]; see Note 5.
3. About 20 min before windowing, place the egg on its side to
allow the embryo to reposition on top of the egg yolk.
4. Wipe the eggshell using facial tissue and 70 % ethanol to avoid
contamination (see Note 6).
5. Place a strip of Scotch tape to cover the area of the planned
window. This will prevent pieces of eggshell from falling onto
the embryo when cutting the window.
6. Use a scalpel to drill a hole into the corner of the intended
window and into the blunt end of the egg (see Fig. 1 and
Note 7).
7. Remove ~3 ml of albumen by pushing the needle of the syringe
at an angle greater than 45° into the hole at the blunt end of
the egg. This will avoid damage to the egg yolk (see Note 8).
8. Use a paint brush to seal the hole at the blunt end of the egg
with melted paraffin.
9. Cut a window into the eggshell. To avoid damaging the
embryo, carefully hold scissors horizontally.
10. Seal the window with a coverslip and paraffin. Use a brush to
apply melted paraffin to the edges of the window and carefully
press a coverslip onto the hot paraffin (alternatively, use Scotch
258 Irwin Andermatt and Esther T. Stoeckli

Fig. 1 Egg windowing. (a) After the egg is cleaned with 70 % ethanol, a strip of sticky tape is put on the top of
the egg, and two holes are drilled into the edge of the presumptive window and the blunt end of the egg
(arrowheads). (b) The needle is inserted at an angle greater than 45° in order not to damage the egg yolk.
Removal of about 3 ml of albumen will detach the embryo and the blood vessels from the eggshell. (c) The hole
at the blunt end is sealed (arrowhead ), and a window is cut into the shell starting at the previously drilled hole.
(d) With a brush melted paraffin is applied to the edges of the window and immediately covered with a cover-
slip (open arrows )

tape to seal the window; see Note 9). As paraffin cools down
quickly, carefully pressing a soldering iron on the coverslip will
remelt the paraffin below the glass and lead to proper sealing.
11. Put the windowed egg in the incubator at 38.5 °C and 45 %
humidity until further use (see Notes 3 and 4).

3.2 In Ovo 1. Clean the working space with 70 % ethanol and autoclave your
Electroporation working tools (see Note 6).
2. Prepare capillaries to make injection needles using the glass
needle puller (see Note 10).
3. Remove the coverslip or the tape covering the window. To
remove the coverslip briefly press the hot soldering iron onto
the glass.
4. To get direct access to the embryo, carefully remove the extra-
embryonic membranes with forceps and spring scissors (Fig. 2).
5. Break off the tip of the previously pulled needles to obtain a
diameter of 5–7 μm. Plug the needle into the polyethylene
tubing and fill the tip of the needle with your injection mix
(see Note 10).
 RNAi in Brain Development 259

Fig. 2 (a, b) Schematic drawing of in ovo electroporation of the caudal neural tube. (a) Injection of dsRNA or
miRNA into the central canal of the spinal cord. (b) Electroporation with anode and cathode positioned parallel
to the body axis. (c) Depending on the positioning of the electrodes, electroporation can be spatially controlled.
(d) Successful unilateral transfection of EGFP in the right hemisegment is visualized on a cryosection of a HH25
embryo. (e) Unilateral in ovo injection and electroporation of an E3 chicken embryo. Using forceps, the extra-
embryonic membranes are removed to obtain access to the spinal cord. The injection mix is injected into the
central canal of the lumbar region of the spinal cord. The injection volume is controlled by mouth. The maximal
injection volume is achieved when the blue solution reaches both the ventricle and the tail (arrowheads ). For
unilateral electroporation the electrodes are placed laterally in parallel to the body axis of the embryo. Make
sure not to touch major blood vessels to avoid fatal bleeding. (f) For dorsal or ventral electroporation HH20
embryos are easy to handle because their body is more detached from the egg yolk and slightly tilted to the
side compared to the younger embryo shown in (e). Note that the lumbar region is tilted to the side and there-
fore you see more lateral parts of the spinal cord. Use the forceps to pull the extraembryonic membranes
sidewards to position the embryo and stabilize it in the desired position for injection and electroporation with
anode and cathode positioned dorsally and ventrally, respectively. For dorsal electroporation the polarity would
be reversed. Scale bar: 2 mm

6. Inject the mix containing miRNA plasmids (see Notes 11 and

12) or long dsRNA (see Notes 11 and 13) into the central canal
of the neural tube and control injection volume by mouth
(Fig. 2e, f). The maximum volume is reached when the blue
solution extends from the brain ventricle to the tail of the embryo.
260 Irwin Andermatt and Esther T. Stoeckli

Fig. 3 Ex ovo culture. (a) The embryo is transferred from the egg into a domed dish at E2.5 and kept at 38.5 °C.
(b and c) Ex ovo culture of a HH34 embryo (E8) ready for cerebellar injection and electroporation

7. Before applying an electric field, add a few drops of PBS to the

embryo to prevent overheating and high electrical resistance
during subsequent electroporation.
8. Place the electrodes parallel to the anterior-posterior axis of the
spinal cord. Do not touch major blood vessels while applying
current to avoid bleeding. Use 5 pulses of 50 ms duration at
18 V for E2 or at 25 V for E3 embryos for electroporation
(Fig. 2, see Notes 14–16).
9. Put the resealed egg back into the incubator until the
embryo reaches the desired developmental stage for analysis
(see Note 4).

3.3 Ex Ovo Culture 1. Incubate fertilized eggs for 2.5 days at 38.5 °C and at least
of Chicken Embryos 45 % humidity (see Notes 4 and 17).
2. Position the eggs on the side for 20 min to allow for the
embryo to position on top of the egg yolk.
3. Wipe the egg with 70 % ethanol and crack it on a sharp edge.
Transfer the whole egg content into the domed dish without
destroying the egg yolk (Fig. 3).
4. Cover the culture with a lid from a petri dish to minimize evap-
oration and keep the ex ovo cultures in the incubator until
further use.

3.4 Ex Ovo Injection Here we describe injection and electroporation of the developing
and Electroporation cerebellum of embryos at E8 as an example and as previously
described [3]. Other areas of the brain can be targeted with the
same parameters as a starting point.
1. Use autoclaved tools and clean the work space with 70 %
ethanol.
2. Stage the embryos according to Hamburger and Hamilton [14].
 RNAi in Brain Development 261

Fig. 4 Ex ovo electroporation. (a and b) Using spring scissors and forceps, a small hole is cut into the extraem-
bryonic membranes in order to get access to the desired injection site. In the example shown here, the injec-
tion would be planned into the eye. (c) For injections into the developing cerebellum, the blood vessels on the
back of the head can be used as landmarks. (d) The tweezer electrodes are placed parallel to the head for
electroporation. Importantly, electrodes should not touch the embryo. Scale bar in (d): 2 mm. (e) and (f) Two
examples of successful electroporation of the cerebellum. Coronal sections are shown with EGFP-positive cells
(green ) indicating successful transfection. Axonin-1/contactin-2 is a marker for parallel fibers, axons of
granule cells (red ). V ventricle, PS pial surface

3. Remove the lid and cut a small hole of 3–4 mm diameter into
the extraembryonic membranes above the head, where the
injection is planned (Fig. 4).
4. Depending on the age of the embryo and the injection site, it
may be required to fix the head of the embryo with a spatula
bent to a hook by placing it underneath the neck.
5. Use a glass capillary connected to a piece of tubing to inject
the mix (see Notes 10–13) into the cerebellum (Fig. 4c). To
target all the cerebellar layers, insert the glass capillary first into
the ventricular zone and apply constant pressure while pulling
the needle out (see Note 18).
6. Put a few drops of sterile PBS on the injection site and place
the platelet electrodes on either side of the head of the embryo.
Avoid touching embryonic tissue with the electrodes to pre-
vent damage. Use six pulses of 40 V and 99 ms duration with
1 s interpulse intervals (Fig. 4d; see Note 16).
7. Cover the dish and put it back to the incubator until the desired
stage for analysis is reached.
262 Irwin Andermatt and Esther T. Stoeckli

3.5 Verification First, electroporation efficiency has to be high to get strong

of Gene Silencing downregulation of the gene of interest. The best way to monitor
successful targeting of the tissue is to check for the expression of
fluorescent proteins (Figs. 2d and 4e, f; see Note 19). A first
impression can be obtained by simply looking at the embryo in ovo
or ex ovo under a fluorescent stereomicroscope. The embryos
expressing the reporter protein can then be further analyzed for
verification of gene knockdown. The efficacy of electroporation at
late stages in ex ovo cultures has to be verified after dissection [3].
There are several possibilities to analyze gene silencing.
Immunohistochemistry as described previously is certainly the best
way to verify the downregulation of protein expression [1, 4, 10,
15]. Lysates from spinal cord or brain tissue can be used for analy-
sis by Western blot [15]. Alternatively, if there are no antibodies
available, in situ hybridization on cryosections [4, 10] or maybe on
whole-mount embryos at very young stages [16, 17] can monitor
expression levels of the targeted mRNA. Also RT-PCR can serve to
demonstrate efficient downregulation, although spatial informa-
tion is lost [18]. The nontargeted half of the spinal cord or the
cerebellum may serve as internal control when compared to the
electroporated hemisegment (see Note 20).

3.6 Phenotype Depending on the target gene and its function, the methods of
Analysis phenotype analysis will vary. It is not within the scope of this
chapter to list methods for immunohistochemistry, axonal tracing
[1, 19], or analysis of cell migration. No matter what kind of
analysis you will use, keep in mind that phenotypes can only be
assessed by comparison of your experimental group with two
control groups (see Note 20).

4 Notes

1. Eggs have to be windowed no later than the third day of incu-

bation even when the injection and electroporation are made
at later stages. After 3 days of development, the blood vessels
no longer detach from the eggshell and will therefore be
damaged during windowing.
2. Eggs can be stored at 15 °C for up to 1 week before incuba-
tion. Longer storage will impair development and viability of
the embryos.
3. After windowing, the eggs need to be sealed properly. Loss of
humidity from the egg will strongly decrease the viability of
the embryo. Furthermore, high humidity in the incubator is
crucial (~45 %). Place a tray of water containing 0.1 g/l of cop-
per sulfate into the incubator. Copper sulfate prevents con-
tamination of the water.
 RNAi in Brain Development 263

4. We routinely use two incubators, one to incubate eggs before

they are windowed or used for ex ovo cultures and another one
for windowed eggs or dishes. Avoid opening incubators too
many times during an experiment, as temperature fluctuations,
especially going up to ~40 °C and more, will have negative
effects on embryo development and strongly decrease survival
rates.
5. Staging of the embryos at the beginning and at the time of
analysis is very important. Careful comparison of the develop-
mental progress between experimental and control embryos
that have not been handled can give important information
about potential interference of embryo handling and/or injec-
tion and electroporation procedure with normal
development.
6. To reduce the risk of contamination, always clean your work
space with 70 % ethanol and use autoclaved tools and sterile
solutions. Additionally, keep the time during which the egg is
unsealed to a minimum. Working in a laminar flow hood is not
necessary, however.
7. Drilling a hole into the corner of the intended window is nec-
essary for allowing airflow during removal of albumen which
will result in the detachment of the embryo from the
eggshell.
8. Damaging the yolk will result in death of the embryo and
therefore has to be avoided for both in ovo and ex ovo
development.
9. If the eggs need to be reopened several times, sealing with a
coverslip is advantageous since the paraffin can be quickly
melted by placing a soldering iron on the coverslip. No matter
whether you seal the egg with Scotch tape or a coverslip, make
sure the window is properly closed, because dehydration will
dramatically lower survival rate.
10. The diameter of the injection needle should be kept at 5–7 μm
to minimize tissue damage and to prevent leakage of the injec-
tion mix upon retraction of the needle. It is important to have
no leakage in order to get efficient and reproducible gene
silencing.
11. Make sure that salt concentrations and pH of the injection mix
are in the physiological range. Plasmids used in the mix should
be purified carefully by making sure that there is no alcohol
remaining from previous precipitation steps. Tris buffers
should be avoided as they tend to cause unspecific, toxic effects.
12. DNA injection mix: RNAi plasmid at a concentration that
must be determined by the user and trypan blue diluted in PBS
(for detailed designing of RNAi plasmids and concentration of
264 Irwin Andermatt and Esther T. Stoeckli

ingredients of the injection mix, see [4]). We design artificial

miRNAs using GenScript’s siRNA Target Finder http://www.
genscript.com/siRNA_target_finder.html.
13. dsRNA injection mix: sterile PBS containing dsRNA derived
from the gene of interest (200–400 ng/μL), EGFP reporter
plasmid (20 ng/μL), and 0.04 % (vol/vol) trypan blue (for
detailed description of dsRNA synthesis and troubleshooting
advice, see [1, 5]). Long dsRNA is easily produced by in vitro
transcription from a cDNA template. Expressed sequence tags
(ESTs) are available from ARK Genomics or Source BioScience
Life Sciences.
14. For targeting dorsal or ventral cell types, electrodes can be
positioned accordingly along the body of the embryo (see
Fig. 2c). Make sure not to touch any major blood vessels dur-
ing electroporation because this will lead to fatal bleeding.
Keep the electrodes away from the heart.
15. For in ovo spinal cord electroporation, we use custom made
wire electrodes. There are also commercially available ones
from BTX, Harvard Apparatus (http://www.btxonline.com/
genetrodes/).
16. Make sure to clean the electrodes from denatured proteins
after each electroporation to maintain a proper electric field
between the electrodes for the following embryos.
17. For best survival rate the eggs should be cracked at E2.5. E2
works also but older stages will not give good survival rates.
18. Since injection depth and volume cannot be seen easily at older
stages of development, we recommend using injection of a mix
of a dye (DiI, CFSE (carboxyfluorescein succinimidyl ester)) or
fluorescent dextran beads together with an EGFP reporter
plasmid to establish the landmarks that can be used to guide
injection. For instance, for the cerebellar anlage, the blood ves-
sels on the back of the head can be used as landmarks [3].
19. If you have low transfection rates, consider the following
points:
● Make sure the electrodes are positioned close enough to
the embryo without actually touching it.
● The injection volume should be maximized but kept small
enough to avoid tissue damage or leakage. It is absolutely
important to prevent leakage of the injection mix before
electroporation in order to get reproducible and effective
gene silencing.
● Take into account that in highly proliferating cells, the
electroporated plasmid will be diluted strongly over time.
In this case, reconsider the time point of injection and
electroporation.
 RNAi in Brain Development 265

● Finally, injection and electroporation require extensive

practice to obtain adequate manual skills for an efficient
downregulation of the targeted mRNA.
20. Note depending on time of electroporation and position of the
electrodes, it is not possible to compare the two halves of the
embryo, as both sides would be affected by the manipulation.
Therefore it is important to have adequate controls. At least
two control groups are always required. One group consists of
untreated embryos taken out of unwindowed eggs at the time
of analysis. The second control group consists of mock-treated
embryos. These are handled exactly the same way as the exper-
imental group(s) but without dsRNA derived from the gene of
interest. In these embryos either only a reporter plasmid or
dsRNA from a gene that is not expressed in the nervous system
is injected. The best control would be dsRNA- or miRNA-
based constructs targeting a family member of the gene of
interest. This is of course not always available. The treated con-
trol group and the embryos from the untreated group have to
be indistinguishable to make sure that the handling of the
embryos did not cause any artifacts that may be mistaken as a
phenotype caused by silencing the target gene. The compari-
son between the control-treated and the experimental group
will provide the desired result indicating the function off the
gene of interest.

Acknowledgments

We thank Evelyn Avilés and Nicole Wilson for their help in Figs. 2
and 4. Work in the laboratory of E.S. is supported by a grant from
the Swiss National Science Foundation.

References
1. Pekarik V, Bourikas D, Miglino N, Joset P, 5. Baeriswyl T, Mauti O, Stoeckli ET (2008)
Preiswerk S, Stoeckli ET (2003) Screening for Temporal control of gene silencing by in ovo
gene function in chicken embryo using RNAi electroporation. Methods Mol Biol
and electroporation. Nat Biotechnol 442:231–244
21:93–96 6. Mauti O, Baeriswyl T, Stoeckli ET (2008)
2. Chesnutt C, Niswander L (2004) Plasmid- Gene silencing by injection and electropora-
based short-hairpin RNA interference in the tion of dsRNA in Avian embryos. CSH Protoc
chicken embryo. Genesis 39:73–78 2008:pdb.prot5094
3. Baeriswyl T, Stoeckli ET (2008) Axonin-1/ 7. Luo J, Redies C (2005) Ex ovo electropora-
TAG-1 is required for pathfinding of granule tion for gene transfer into older chicken
cell axons in the developing cerebellum. embryos. Dev Dyn 233:1470–1477
Neural Dev 3:7 8. Mauti O, Sadhu R, Gemayel J, Gesemann M,
4. Wilson NH, Stoeckli ET (2011) Cell type spe- Stoeckli ET (2006) Expression patterns of
cific, traceable gene silencing for functional plexins and neuropilins are consistent with
gene analysis during vertebrate neural develop- cooperative and separate functions during neu-
ment. Nucleic Acids Res 39:e133 ral development. BMC Dev Biol 6:32
266 Irwin Andermatt and Esther T. Stoeckli

9. Das RM, van Hateren NJ, Howell GR, Farrell of the chick embryo. 1951. Dev Dyn
ER, Bangs FK, Porteous VC, Manning EM, 195:231–272
McGrew MJ, Ohyama K, Sacco MA et al 15. Rao M, Baraban JH, Rajaii F, Sockanathan
(2006) A robust system for RNA interference S (2004) In vivo comparative study of
in the chicken using a modified microRNA RNAi methodologies by in ovo electropor-
operon. Dev Biol 294:554–563 ation in the chick embryo. Dev Dyn
10. Bourikas D, Pekarik V, Baeriswyl T, Grunditz 231:592–600
A, Sadhu R, Nardó M, Stoeckli ET (2005) 15. Stepanek L, Stoker AW, Stoeckli E, Bixby JL
Sonic hedgehog guides commissural axons (2005) Receptor tyrosine phosphatases guide
along the longitudinal axis of the spinal cord. vertebrate motor axons during development.
Nat Neurosci 8:297–304 J Neurosci 25:3813–3823
11. Domanitskaya E, Wacker A, Mauti O, 16. Katahira T, Nakamura H (2003) Gene silenc-
Baeriswyl T, Esteve P, Bovolenta P, Stoeckli ing in chick embryos with a vector-based small
ET (2010) Sonic hedgehog guides post- interfering RNA system. Dev Growth Differ
crossing commissural axons both directly and 45:361–367
indirectly by regulating Wnt activity. J Neurosci 17. Dai F, Yusuf F, Farjah GH, Brand-Saberi B
30:11167–11176 (2005) RNAi-induced targeted silencing of
12. Wilson NH, Stoeckli ET (2013) Sonic hedge- developmental control genes during chicken
hog regulates its own receptor on post-crossing embryogenesis. Dev Biol 285:80–90
commissural axons in a Glypican-dependent 18. Sato F, Nakagawa T, Ito M, Kitagawa Y,
manner. Neuron in press Hattori M-A (2004) Application of RNA
13. Niederkofler V, Baeriswyl T, Ott R, Stoeckli interference to chicken embryos using small
ET (2010) Nectin-like molecules/SynCAMs interfering RNA. J Exp Zool A Comp Exp
are required for post-crossing commissural Biol 301:820–827
axon guidance. Development 137:427–435 19. Perrin FE, Stoeckli ET (2000) Use of lipo-
14. Hamburger V, Hamilton HL (1992) A philic dyes in studies of axonal pathfinding in
series of normal stages in the development vivo. Microsc Res Tech 48:25–31
 Part VII

Mouse Protocols
 Chapter 18

Immunohistochemistry and RNA In Situ Hybridization

in Mouse Brain Development
Jinling Liu and Aimin Liu

Abstract
During development, the mouse brain is progressively divided into functionally distinct compartments.
Numerous neuronal and glial cell types are subsequently generated in response to various inductive signals.
Each cell expresses a unique combination of genes encoding proteins from transcription factors to
neurotransmitters that define its role in brain function. To understand these important and highly sophis-
ticated processes, it is critical to accurately locate the various proteins and cells that produce them. In this
chapter, we introduce the techniques of immunohistochemistry, which detects the localization of specific
proteins, and RNA in situ hybridization, which enables the visualization of specific mRNAs.

Key words Immunohistochemistry, RNA in situ hybridization, Cryosection, Antibody, Digoxigenin,

Fluorescent

1 Introduction

The mouse brain consists of multiple divisions (cerebrum,

epithalamus, thalamus, hypothalamus, cerebellum, and brain
stem) and close to 100 million neurons and glia [1]. Extrinsic
inductive signals and intrinsic cellular programs both play key
roles in the compartmentalization of the brain as well as cellular
behaviors such as proliferation, differentiation, migration, and cell
death. To better understand the developmental processes involved
in mouse brain development, it is important to obtain information
regarding the spatial and temporal patterns of gene expression. In
this chapter, we will describe methods for the detection of protein
(immunohistochemistry, IHC) and mRNA (RNA in situ hybrid-
ization, RISH) in brain sections.
IHC detects particular proteins present in tissues. The princi-
ple underlying this technique is the specific bindings between
antibodies and antigens. To visualize the antibody-antigen interac-
tion, an antibody is tagged with a fluorophore, which can be
conveniently detected with a fluorescent microscope. Alternatively,

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_18, © Springer Science+Business Media, LLC 2014

269
270 Jinling Liu and Aimin Liu

the antibody can be conjugated to an enzyme that catalyzes a

color-producing reaction, which can be visualized under a regular
microscope. In this chapter we will describe the method of IHC
with the fluorophore-labeled antibody. The procedure comprises
tissue preparation, blocking, primary and secondary antibody
incubation, mounting, and visualization.
The success of IHC heavily depends on the availability of high-
quality antibodies. In addition, secreted signaling proteins as well
as proteins of the extracellular matrix are not restricted to the cells
producing them, preventing a high-resolution identification of the
signaling centers. On the other hand, RISH allows the detection of
the expression of virtually all genes in the cells, even genes that do
not encode proteins, providing more flexibility compared with the
antibody-based IHC method. Traditionally, RISH depends on the
hybridization of the specific RNA sequence in situ to radiolabeled
probes [2]. Currently, Digoxigenin-labeled probes are more com-
monly used in RISH, which can be recognized with antibodies
coupled with fluorophore or enzymes such as alkaline phosphatase
or peroxidase [3].
RISH can be performed on both frozen sections and paraffin
sections, with frozen sections allowing more sensitive detection of
weak signals [4]. The RISH method we introduce in this chapter
uses Digoxigenin-labeled riboprobes (complementary RNA
probes) to detect specific mRNA on frozen brain sections. The
procedure includes synthesis of riboprobes, hybridization of sec-
tions with Digoxigenin-labeled riboprobes, post-hybridization
washes, incubation with alkaline phosphatase (AP)-conjugated
anti-Digoxigenin antibody, and a color reaction using the phos-
phatase substrate BM purple solution.

2 Materials

The materials should be stored at room temperature unless other-

wise specified.

2.1 Cryosection 1. Dissection tools: student quality iris scissors (Fine Science
Preparation Tools), Dumont forceps (Fine Science Tools), and spoon (Fine
Science Tools).
2. 6 cm petri dishes.
3. 24-well tissue culture plates (see Note 1).
4. Stereomicroscope, such as Nikon SMZ645.
5. 4 % paraformaldehyde (PFA): Prepare 16 % stock by adding
32 g PFA powder and 100 μl 5 N NaOH in 150 ml pre-
warmed distilled, deionized water (ddH2O, at ~65 °C) (see
Note 2). Once PFA is dissolved, add 20 ml 10× PBS and adjust
the volume to 200 ml with ddH2O. Filter through Whatman
 In Situ and Immunostaining in Mouse 271

paper and make 10 ml aliquot in 50 ml centrifuge tubes. Store

at −20 °C. Thaw one tube at ~65 °C and dilute to 4 % PFA
with PBS before use. 4 % PFA can be stored at 4 °C for up to
1 week (see Note 3).
6. Phosphate-buffered saline (PBS): Prepare 10× stock by dis-
solving 80 g NaCl, 2 g KCl, 14.4 g Na2HPO4 and 2.4 g
KH2PO4 in 1 l of ddH2O. Adjust pH to 7.4 and autoclave at
121 °C for 25 min. Dilute to 1× solution with ddH2O for use.
7. Nutator mixer (VWR).
8. 30 % sucrose: Dissolve 12 g sucrose powder in 40 ml PBS.
Store at 4 °C.
9. O.C.T. compound (VWR): Store at 4 °C.
10. Disposable embedding molds, such as Polysciences.
11. Dry ice or liquid nitrogen in appropriate containers.
12. Cryostat, such as Leica CM1900.
13. Superfrost Plus microscopic slides, such as VWR (see Note 4).
14. Slide boxes, such as VWR.
15. Micro slide trays, such as VWR.

2.2 Immunohisto- 1. Coplin jars, such as Wheaton.

chemistry 2. Blocking buffer: 1 % normal goat serum, 0.1 % Triton X-100,
in PBS. Keep at 4 °C (see Note 5).
3. A humidified slide incubation chamber (Fig. 1): Cut two 5 ml
serological pipettes and tape them to the bottom of a flat-
bottom plastic box with lid, such that the two pipettes are par-
allel and 5 cm apart. Place paper towels soaked with ddH2O on
the bottom to keep a moist environment.
4. Paper towels.
5. Fluorophore-conjugated secondary antibodies: Store at
−20 °C in 50 % glycerol.
6. Micro slide trays, such as VWR.
7. Coverslips, such as VWR, 25 × 50 mm.
8. Forceps: Dumont forceps (Fine Science Tools).
9. Dabco: Aldrich. Store at 4 °C (see Note 6).
10. Nail polish.

2.3 Synthesis 1. A plasmid with the promoters for viral RNA polymerases (T3,
of RNA Probes T7, and Sp6) flanking the multiple cloning sites. Store at
−20 °C.
2. Restriction endonucleases: Store at −20 °C.
3. Horizontal DNA electrophoresis apparatus and power supply.
4. Agarose.
272 Jinling Liu and Aimin Liu

Fig. 1 A homemade humidified slide incubation chamber. This chamber is made

by taping two 5 ml serological pipettes, broken to appropriate length, to the bot-
tom of a plastic container with lid. Paper towels soaked in ddH2O provide humid-
ity during the incubation. For RISH, a microwavable box or a container that can
withstand 55 ºC temperature is needed

5. 10 mg/ml ethidium bromide solution: Keep away from light

(see Note 7).
6. TBE buffer: Dissolve 54 g Tris base, 27.5 g boric acid in
ddH2O, add 20 ml 0.5 M EDTA (pH 8.0), and bring the
volume to 5 l with ddH2O.
7. 1-kb DNA ladder.
8. 3 M sodium acetate (NaOAc), pH 5.2: Dissolve 408.3 g of
NaOAc⋅3H2O in 800 ml ddH2O. Adjust pH to 5.2 with acetic
acid and adjust the volume to 1 l with ddH2O. Aliquot and
autoclave.
9. Phenol saturated with Tris–HCl, pH 8.0: Store at 4 °C.
10. Chloroform.
11. 70 % and 100 % ethanol.
12. Tabletop microcentrifuge, such as Eppendorf 5415D.
13. Magnetic stir plate.
14. Diethylpyrocarbonate (DEPC): Store at 4 °C (see Note 8).
15. DEPC-H2O: Add 0.1 % v/v DEPC to ddH2O and mix o/n on
a magnetic stir plate at room temp; autoclave at 121 °C for
25 min (see Note 9).
16. 10× transcription buffer (Roche): Store at −20 °C.
 In Situ and Immunostaining in Mouse 273

17. 10× DIG-labeling mix (Roche): Store at −20 °C (see Note 10).
18. RNase inhibitor (Promega, 40 U/μl): Store at −20 °C.
19. RNA polymerase (Roche T3, T7, or Sp6, 20 U/μl): Store at
−20 °C.
20. DNaseI (RNase-free): Store at −20 °C.

2.4 RNA In Situ The amount of reagents in this section is enough for processing
Hybridization five slides in a 50 ml Coplin jar and should be adjusted according
to the number of slides and size of the container used.

2.5 RNA In Situ All containers and reagents need to be RNase-free on day 1.
Hybridization Day 1:
1. Coplin jars: 50 ml.
Hybridization of
Cryosections on the 2. Diethylpyrocarbonate (DEPC): Store at 4 °C (see Note 8).
Slides 3. Magnetic stir plate.
4. DEPC-H2O (400 ml): Add 400 μl DEPC to 400 ml ddH2O.
Mix o/n on a magnetic stir plate at room temperature and
autoclave at 121 °C for 25 min (see Note 9).
5. DEPC-PBS (400 ml): Add 400 μl DEPC to 400 ml PBS. Mix
o/n on a magnetic stir plate at room temperature and auto-
clave at 121 °C for 25 min (see Note 9).
6. 4 % PFA in DEPC-PBS (80 ml): Thaw two 10 ml aliquots of
16 % PFA (see Subheading 2.1 for 16 % PFA preparation) and
dilute each with 30 ml DEPC-PBS. Store at 4 °C.
7. 0.25 % acetic anhydride in 0.1 M TEA-HCl (40 ml): Add
0.742 g triethanolamine-HCl and 360 μl 5 N NaOH to
39.8 ml DEPC-H2O to make TEA-HCl. Add 100 μl acetic
anhydride right before use (see Note 11).
8. 20 μg/ml proteinase K (proK; Roche #03115879001):
Dissolve one vial of proK (100 mg) in 5 ml DEPC-H2O to
make 20 mg/ml stock. Aliquot and store at −20 °C. Thaw an
aliquot before use and add 40 μl into 40 ml DEPC-PBS (see
Note 12).
9. 70 % ethanol in DEPC-H2O (40 ml): 28 ml 100 % ethanol,
12 ml DEPC-H2O.
10. 95 % ethanol in DEPC-H2O (40 ml): 38 ml 100 % ethanol,
2 ml DEPC-H2O.
11. Hybridization solution (40 ml): Mix the following in a clean
50 ml tube, aliquot, and store at −20 °C (see Note 13). 20 ml
formamide (deionized, aliquot in 50 ml tubes, and store at
−20 °C); 8 ml 50 % dextran sulfate; 400 μl 100× Denhardt’s
(VWR, aliquot in 1.5 ml tubes and store at −20 °C); 1 ml yeast
tRNA (10 mg/ml, aliquot in 1.5 ml tubes, store at −20 °C);
2.4 ml 5 M NaCl; 800 μl 1 M Tris–HCl (pH 8.0); 400 μl
0.5 M EDTA; 400 μl 1 M NaPO4; 4 ml 10 % sarkosyl; 2.6 ml
DEPC-H2O.
274 Jinling Liu and Aimin Liu

12. Parafilm (cut into 20 mm × 50 mm pieces) or RNase-free plas-

tic coverslips.
13. A hybridization oven.
14. A humidified slide incubation chamber: see Subheading 2.2.

2.6 RNA In Situ Starting from day 2, reagents do not need to be RNase-free.
Hybridization Day 2:
1. 20× saline-sodium citrate (SSC) buffer: Mix the following in
Post-hybridization
an appropriate container: 800 ml ddH2O; 175.3 g NaCl;
Washes and Antibody 88.2 g sodium citrate. Adjust pH to 7.0 with a few drops of
Incubation 12 N HCl. Adjust the volume to 1 l with ddH2O. Aliquot and
autoclave at 121 °C for 25 min.
2. 5× SSC (40 ml): 10 ml 20× SSC, 30 ml ddH2O.
3. 2× SSC (40 ml): 4 ml 20× SSC, 36 ml ddH2O.
4. 0.1× SSC (40 ml): 0.2 ml 20× SSC, 39.8 ml ddH2O.
5. High-stringency wash buffer (120 ml) (make fresh): 60 ml for-
mamide, 12 ml 20× SSC, 48 ml ddH2O.
6. PBT (350 ml, enough for both days 2 and 3): Add 350 μl
Tween 20 to 350 ml PBS in a 500 ml bottle. Mix well by shak-
ing vigorously. Alternatively, 10 % Tween 20 stock can be made
in ddH2O in advance and stored at room temperature.
7. RNase buffer (400 ml) (see Note 14): 40 ml 5 M NaCl; 4 ml
1 M Tris–HCl, pH 7.5; 4 ml 0.5 M EDTA, pH 8.0; 352 ml
ddH2O.
8. 10 mg/ml RNase A stock (DNase-free): Add 10 ml 0.01 M
NaOAc (pH 5.2) to a vial of 100 mg RNase A. Heat at 100 °C
for 15 min. Cool to room temperature. Add 1 ml 1 M Tris–
HCl (pH 7.5) and aliquot. Store at −20 °C (see Note 15). On
the day of experiment, thaw one aliquot and make 20 μg/ml
RNase A in RNase buffer by adding 80 μl 10 mg/ml RNase A
stock to 40 ml RNase buffer.
9. Alkaline phosphatase-conjugated anti-Digoxigenin antibody
(Fab fragments from sheep; Roche). Store at 4 °C.
10. A humidified slide incubation chamber: See Subheading 2.2.

2.7 RNA In Situ 1. NTMT (80 ml) (make fresh; see Note 16): 8 ml 1 M Tris–HCl
Hybridization Day 3: 9.5; 4 ml 1 M MgCl2; 1.6 ml 5 M NaCl; 800 μl 10 % Tween
Color Reaction 20; 40 mg levamisole; adjust to 80 ml with ddH2O.
2. Levamisole: make 50 mg/ml stock in ddH2O, aliquot, and
store in 1.5 ml tubes at −20 °C.
3. BM purple (Roche). Store at 4 °C.
4. Water-based mounting medium, such as Mount Quick aq.
(SPI). Store at 4 °C.
5. Coverslips: VWR micro cover glass, 25 × 50 mm.
 In Situ and Immunostaining in Mouse 275

6. Nuclear fast red solution (100 ml): Dissolve 5 g aluminum

sulfate in 100 ml ddH2O, then add 0.1 g nuclear fast red. Boil
and stir on a heated magnetic stir plate to dissolve nuclear fast
red. Filter the solution right before use.

3 Methods

Conduct all procedures at room temperature unless otherwise

specified.

3.1 Mouse Brain 1. Dissect embryos in ice-cold PBS in a 6 cm dish (see Note 17).
Cryosection 2. Cut the embryos at the shoulder level and transfer the heads
Preparation into a 24-well plate with a spoon.
3. Fix the embryos in 4 % PFA on a nutator at 4 °C for 1 h for
IHC or o/n for RISH.
4. Rinse the embryos with PBS and then wash the embryos in
PBS on a nutator o/n at 4 °C.
5. Immerse the embryos in 30 % sucrose o/n at 4 °C, on a nuta-
tor (see Note 18).
6. Change the 30 % sucrose and incubate for another 2–3 h for
further infiltration at 4 °C, on a nutator.
7. Transfer the embryos into a disposable embedding mold and
immerse the embryos in O.C.T. compound for 1 h at 4 °C
(see Note 19).
8. Position the samples at desired orientation and freeze them on
dry ice or, alternatively, in a paperboard box floating on the
surface of liquid nitrogen. Wait for 5 min (see Note 20).
9. Transfer the frozen O.C.T. block containing the brain samples to
the cryostat and wait at least 1 h so that the temperature of the
block can reach the optimal cutting temperature (see Note 21).
10. Cut 10 μm sections using a cryostat (see Note 22). Collect sec-
tions on Superfrost Plus slides.
11. Dry the slides in a micro slide holder for 1 h at room tempera-
ture. Store them in a slide box at −80 °C (see Note 23).

3.2 Immunostaining 1. Remove sections from −80 °C freezer and dry slides in a micro
on Mouse Brain slide holder at room temperature for about 45 min (see Note 24).
Sections 2. Place the slides in a Coplin jar and incubate the sections with
blocking buffer for 1 h.
3. Take slides out of blocking buffer; wipe the backside (the one
without sections) and edges with paper towel. Place them with
the front side (the one with sections) up on the pipettes of the
humidified slide incubation chamber (Fig. 1; see Note 25).
276 Jinling Liu and Aimin Liu

Fig. 2 Immunostaining of brain sections. Shown are two coronal sections of an E12.5 brain incubated with a
monoclonal antibody against Pax6 and a Cy3-conjugated secondary antibody. (a) Pax6 is weakly expressed in
the dorsal region and strongly expressed in the intermediate region in the hindbrain. (b) Pax6 is expressed in
the diencephalon (Di), but not the mesencephalon (mes). Dashed line demarcates the boundary between the
diencephalon and mesencephalon

4. Apply primary antibodies to the sections (diluted in blocking

buffer, 300 μl per slide) and incubate o/n at 4 °C (see Note 26).
5. Pour primary antibody onto a paper towel and place the slides
in a Coplin jar. Wash the slides with blocking buffer, 3 × 10 min.
6. Similar to step 3, place slides in the humidified chamber and
incubate in the dark with appropriate fluorophore-conjugated
secondary antibody (diluted in blocking buffer, 300 μl per
slide) for 2 h (see Note 27).
7. Pour secondary antibody onto a paper towel and place the
slides in a Coplin jar. Wash with blocking buffer, 3 × 10 min.
8. Take slides out of blocking buffer and wipe the backside and
edges with paper towel. Place them into a micro slide tray.
9. Apply 45 μl Dabco evenly onto each slide and mount with a
coverslip (see Notes 28 and 29); seal the slides with nail polish.
10. Observe the sections under fluorescent microscope and take
photos with a cooled CCD camera. Some examples are shown
in Fig. 2.
11. Slides can be stored at 4 °C protected from light (see Note 30).

3.3 Synthesis of RNA 1. Clone the cDNA (or part of the cDNA if the full-length cDNA
Probes for RISH is longer than 1,500 bps; see Note 31) of the gene of interest
into a plasmid containing the promoters for the viral RNA
polymerases (T3, T7, and Sp6; such as pBluescript).
 In Situ and Immunostaining in Mouse 277

2. Linearize the template:

Choose a unique restriction site in the multiple cloning sites
on the 5′ end of the cDNA. Cut ~10 μg DNA in 20–50 μl
reaction with the corresponding restriction endonuclease o/n
using at least 20 IU enzyme. Run small amount (0.1–0.5 μg)
in a 0.8 % agarose gel to check the efficiency of restriction
enzyme digestion.
3. Purify the linearized DNA:
Bring the volume of above reaction to 200 μl with ddH2O,
add 20 μl 3 M NaOAc (pH 5.2), and mix well. Perform
standard phenol/chloroform and chloroform extraction. Add
600 μl 100 % ethanol to the purified DNA, mix well, and leave
at −80 °C for 30 min. Centrifuge at top speed (>14,000 × g) in
a tabletop centrifuge for 10 min at 4 °C. Rinse the pellet with
70 % ethanol without disturbing it. Dry the pellet briefly and
dissolve it in 10 μl DEPC-H2O. The linearized template can
be stored at −20 °C.
4. In vitro transcription:

2 μl 10× transcription buffer

3 μl DIG-labeling mix
2 μl Linearized template DNA
0.5 μl RNase inhibitor
2 μl RNA polymerase (T3, T7, or Sp6; use the one
whose promoter is at the 3′ end of the cDNA)
10.5 μl DEPC-H2O

Incubate at 37 °C for 2 h.
5. Run 1 μl in 1 % agarose gel for 20 min to 1 h to check the yield
of the probe.
6. DNase I treatment (optional):
Add 1 μl DNaseI and 1 μl RNase inhibitor to the RNA probe
and incubate at 37 °C for 15 min to remove the template.
7. Add 180 μl DEPC-H2O, 20 μl 3 M NaOAc, and mix well.
Then add 600 μl 100 % ethanol, mix well, and leave at −80 °C
for 30 min.
8. Centrifuge at top speed (>14,000 × g) in a tabletop centrifuge
for 15 min at 4 °C (see Note 32).
9. Discard the supernatant and rinse the pellet with 70 % ethanol
once without disturbing the pellet.
10. Discard 70 % ethanol and dry the pellet for 5 min. Dissolve the
RNA probe in 40–50 μl DEPC-H2O and store at −80°C.
278 Jinling Liu and Aimin Liu

3.4 RNA In Situ Day 1: hybridization of cryosections (all steps are carried out in Coplin
Hybridization of jars unless otherwise specified; avoid RNase contamination)
Mouse Brain Sections
1. Dry slides at room temperature for about 45 min
(see Note 24).
2. Postfix slides in 4 % PFA in DEPC-PBS for 10 min
(see Note 33).
3. Wash with DEPC-PBS, 2 × 5 min.
4. Drain excess DEPC-PBS and incubate for 6 min in 20 μg/ml
proteinase K in DEPC-PBS (see Note 34).
5. Drain and wash with DEPC-PBS for 5 min.
6. Refix in 4 % PFA for 5 min, then wash 5 min in DEPC-PBS
(see Note 35).
7. Acetylate sections with acetic anhydride in 0.1 M TEA-HCl for
10 min (see Note 36).
8. Wash in DEPC-PBS for 5 min; dehydrate in 70 % ethanol for
5 min and 95 % ethanol for 2 min. Air dry for 30 min to 2 h.
9. Add 2 μl appropriate probe (approx. 1 μg) to 1 ml hybridiza-
tion solution and heat at 80 °C for 2 min (see Note 37).
10. Place slides horizontally in a humidified slide incubation cham-
ber. Cover sections with 200 μl of hybridization solution with
the probe and lower parafilm coverslips over sections avoiding
bubbles (see Note 38).
11. Seal the slide incubation chamber carefully and hybridize at
55 °C in a hybridization oven overnight (16–18 h).
Day 2: post-hybridization washes and antibody incubation (all
steps are carried out in Coplin jars; RNase-free environment is not
required)
1. Dip slides gently in a Coplin jar filled with 5× SSC to let the
coverslips float off the slides (see Note 39).
2. Incubate the sections in high-stringency wash at 65 °C in a
Coplin jar for 30 min (see Note 40).
3. Wash in RNase buffer at 37 °C, 3 × 10 min.
4. Wash in RNase buffer with 20 μg/ml RNase A at 37 °C for
30 min (see Note 41).
5. Wash in RNase buffer at 37 °C for 15 min.
6. Repeat high-stringency wash (as in step 2) at 65 °C, 2 × 20 min.
7. Wash in 2× SSC, then 0.1× SSC for 15 min each at 37 °C.
8. Wash with PBT for 15 min.
9. Take the slide out of PBT; wipe the backside and edges of the
slide with paper towel. Place slides horizontally in a humidified
slide incubation chamber and block for 1 h with 10 % goat
serum in PBT (300 μl per slide).
 In Situ and Immunostaining in Mouse 279

10. Pour the blocking buffer onto paper towels. Wipe the backside
and edges of the slide with paper towel. Incubate with
AP-conjugated anti-Digoxigenin antibody (diluted 1/5,000
in PBT with 1 % goat serum, 320 μl per slide) at 4 °C over-
night in the same humidified chamber.
Day 3: color reaction
1. Pour the antibody onto paper towels. Place slides in a Coplin
jar and wash in PBT for five times, 1 h each.
2. Wash 2 × 10 min in freshly prepared NTMT buffer.
3. Wipe the backside and edges of the slide with paper towel.
Place the slides horizontally in a humidified slide incubation
chamber and incubate o/n to several days in BM purple solu-
tion (300 μl per slide) supplemented with 0.5 mg/ml levami-
sole in the dark (see Note 42).
4. Observe periodically the progress of the color reaction under a
microscope. If the staining is not ready, reapply BM purple
solution and incubate for longer time (see Note 43).
5. When the staining is ready, i.e., when the signal is strong and
the background staining just begins to show, place slides back
into a Coplin jar. Wash slides in PBS for 2–5 min and dip briefly
in ddH2O (see Note 18).
6. (Optional) Counterstain the sections with nuclear fast red until
the sections turn slightly pink. Usually it takes 2–3 min.
7. Wash the excess nuclear fast red in slow-running tap water.
8. Wipe the backside and edges of the slide with paper towel.
Apply mounting medium to the slides and put coverslips on
(see Note 44).
9. Observe the staining under a microscope and take photos with
a color camera. Some examples are shown in Fig. 3.
10. The mounted sections can be stored at 4 °C.

4 Notes

1. The 24-well plates are for the convenient storage of individual

small sample, such as the brains of E12.5 embryos or younger.
Vials of appropriate size should be used for older/bigger brain
samples.
2. Distilled, deionized water (ddH2O) used in this protocol is
ultra-purified water with a resistance of 18.2 MΩ.
3. PFA is highly volatile and irritant. Prepare the 16 % PFA in the
fume hood.
280 Jinling Liu and Aimin Liu

Fig. 3 RNA in situ hybridization of brain sections. Shown are two sections of E12.5 brains hybridized with a
Digoxigenin-labeled RNA probe against Otx2. (a) A coronal section through the forebrain region shows Otx2
expression in ventral telencephalon (tel) and the retinal pigment epithelium of the eye, but not in ventral dien-
cephalon (di). (b) A sagittal section shows that Otx2 is expressed in the mesencephalon (mes) and diencephalon
(di), but not in the metencephalon (met) and pons. The arrowhead points to the boundary between diencephalon
and mesencephalon. Dashed lines demarcate the boundary between the midbrain and hindbrain, known as the
isthmus (is)

4. It is critical to use Superfrost Plus slides as they contain a special

coating to prevent the sections from falling off the slides during
the incubation and washing steps.
5. Make fresh blocking buffer for each experiment. Do not store
for more than a week.
6. Dabco is irritant. Avoid contact with skin and eyes. Avoid
inhalation of vapor or mist.
7. Ethidium is highly carcinogenic. Avoid direct contact and dis-
pose ethidium-containing waste properly.
8. DEPC is toxic. Avoid direct contact.
9. All DEPC-treated solutions need to be autoclaved to degrade
DEPC, which can react with RNA.
10. When thawing the DIG-labeling mix, avoid prolonged incuba-
tion at 37 °C to reduce the chance of NTP degradation.
11. Keep acetic anhydride from moisture (keep the container
tightly closed all the time) and add acetic anhydride immedi-
ately before use.
12. Do not refreeze proK for RNA in situ hybridization. The activ-
ity of proK should be tested every time a new batch is
introduced.
 In Situ and Immunostaining in Mouse 281

13. To make 50 % dextran sulfate, mix 10 ml dextran sulfate and

10 ml ddH2O in a 50 ml tube by inverting, shaking, vortexing,
and heating at 60 °C. When dextran sulfate starts to get into
water, the total volume will decrease. Add more ddH2O to
keep the final volume at 20 ml.
14. Sterile RNase buffer can be stored at room temperature.
15. Take caution to avoid contaminating the bench and other lab
materials with RNase A.
16. NTMT should be freshly prepared.
17. At this step, we usually remove the extraembryonic membranes
or the tips of the tails to genotype the embryos.
18. The tissue should float on the surface of the 30 % sucrose ini-
tially and sink to the bottom of the well after sucrose has fully
infiltrated the tissue. Make sure the tissue is completely
immersed in the sucrose, as the morphology of tissues staying
at the air/liquid interface can be distorted by surface tension.
19. The time the embryos immersed in O.C.T. depends on the size
and density of the tissue. For large and/or dense tissue, longer
time is needed.
20. Do not freeze samples by directly placing the embedding mold
in liquid nitrogen.
21. If not cutting sections right away, wrap the embedding mold
with parafilm and keep it at −80 °C for up to 1 week.
22. See manufacturer’s manual for how to cut cryosections.
23. Cryosections can be stored for months at −80 ºC without
noticeable degradation of proteins and mRNAs.
24. Dried sections do not fall off the slides in subsequent experi-
ments. In addition, drying creates holes in the subcellular
structure, permeabilizing the cells for further experiments.
25. Take caution not to let the sections dry completely. If neces-
sary, process slides one at a time. This applies to all steps that
involve taking slides out of the solution.
26. The antibody solution stays on the top of slides only if the
edges and the bottom of the slides are dry. Therefore, add the
antibody solution to the center of the slide and avoid moving
the chamber once the antibody is added. Place a sign on the
chamber to warn others not to move it during incubation.
27. Cover the humidified slide incubation chamber with foil or
perform the incubation in a cabinet.
28. Cut the pipette tip to make a large orifice because Dabco is
sticky.
29. To avoid bubbles, lower the coverslip slowly.
30. The fluorescent signal decreases over time, so try to observe
the fluorescence as early as you can.
282 Jinling Liu and Aimin Liu

31. In general, shorter probes (<500 bps) tend to yield weaker

signals, whereas very long probes (>1,500 bps) have difficulty
penetrating the cell membranes.
32. Take caution when removing the supernatant because the pel-
let may not be visible.
33. Postfixation ensures the tissue is fixed equally with cross-linked
RNA molecules. It also improves the retention of the tissue on
the slide.
34. To drain excess DEPC-PBS, hold the slides and tilt the Coplin
jar onto the paper towels. Proteinase K treatment improves the
signal intensity by allowing greater access of the target mRNA
for the probes.
35. Refixation improves the section stability after proteolytic
digestion.
36. Acetylation chemically modifies proteins and reduces their
nonspecific bindings.
37. Preheat the hybridization solution at 80 °C before use.
38. To avoid bubbles, lower the parafilm coverslips slowly from
one side to the other. Take caution when transporting the slide
chamber to the oven such that the coverslips do not fall off the
slides.
39. Do not force the coverslips off the slides with forceps, or the
sections may tear.
40. Pre-warm solution for this step and steps 3–7. During the
high-stringency wash, low salt concentration and high tem-
perature inhibit nonspecific bindings.
41. RNase A digests single-stranded RNA to reduce the back-
ground signal.
42. Wrap the slide chamber in foil or place the chamber in a dark
cabinet.
43. Before checking the staining status of the sections, prepare BM
purple solution with 0.5 mg/ml levamisole in case the staining
is not ready and more incubation with BM purple is needed.
Otherwise, the sections may become dry before the BM purple
solution is ready.
44. The mounting medium is very sticky and solidifies quickly. It is
better to apply the mounting medium before the slides dry and
put coverslip on immediately.

Acknowledgments

We would like to thank Dr. Simeone for providing the RNA in situ
probe for Otx2. The monoclonal antibody against Pax6 developed
by Dr. Jessell was obtained from the Developmental Studies
 In Situ and Immunostaining in Mouse 283

Hybridoma Bank developed under the auspices of the NICHD

and maintained by The University of Iowa, Department of
Biological Sciences, Iowa City, IA 52242. A. L. is supported by an
NSF grant (IOS-0949877) and a Penn State new lab start-up fund.

References
1. Kandel E, Schwartz J, Jessell T (2000) Principles 3. Komminoth P, Merk FB, Wolfe HJ, Roth J
of neural science, 4th edn. McGraw-Hill, Health (1992) Comparison of 35S-and digoxigenin-
Professions Division, New York, NY. ISBN labeled RNA and oligonucleotide probes for
0-8385-7701-6 in situ hybridization. Histochemistry 98:
2. Joseph G. Gall, Mary Lou Pardue (1969) 217–228
Formation and detection of RNA-DNA hybrid 4. Wilcox JN (1993) Fundamental principles of in
molecules in cytological preparations*. Proc situ hybridization. J Histochem Cytochem
Natl Acad Sci USA 63:378–383 41:1725–1733. doi:10.1177/41.12.8245419
 Chapter 19

In Utero Electroporation to Study Mouse Brain

Development
Emilie Pacary and François Guillemot

Abstract
In utero electroporation is a rapid and powerful technique to study the development of many brain regions.
This approach presents several advantages over other methods to study specific steps of brain development
in vivo, from proliferation to synaptic integration. Here, we describe in detail the individual steps necessary
to carry out the technique. We also highlight the variations that can be implemented to target different
cerebral structures and to study specific steps of development.

Key words Electroporation, In utero, Brain, Embryo, Neuronal development, Gain and loss of func-
tion, Transfection, Mouse

1 Introduction

In utero electroporation is a simple and rapid approach to overex-

press or silence genes in the developing mouse brain, representing
a major alternative to time-consuming knockout, knock-in, and
transgenic strategies to alter the mouse genome. Indeed, injection
of plasmid DNA solution into the embryonic brain and application
of an electric current are sufficient to allow area- and time-
dependent transfection of brain cells. Since DNA is negatively
charged, the injected plasmid moves towards the positive pole of
the electrode and is thus selectively introduced into a specific brain
region. Consequently, distinct areas can be targeted (Table 1),
depending on the position and/or angle of the positive electrode.
This is an important advantage over viral infections where the
infected area is usually large and dependent on the titer of the viral
suspension used. The region of interest can thus be individually
targeted without affecting the rest of the embryo. The size of the
transfected area can also be adjusted by using paddle-type elec-
trodes of different diameter or by using wire-type electrodes
(microelectrodes). In addition to the orientation and type of

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_19, © Springer Science+Business Media, LLC 2014

285
286 Emilie Pacary and François Guillemot

Table 1
The different brain structures that can be
targeted by in utero electroporation

Brain structures References

Somatosensory cortex [5–7]
Prefrontal cortex [8]
Piriform cortex [9]
Amygdala [9–11]
Hippocampus [3, 12, 13]
Retina [14–16]
Ganglionic eminences [17–19]
Preoptic area [20]
Thalamus [21–23]
Cerebellum [24–26]
Midbrain [4, 27, 28]

electrodes, the location of DNA injection is another parameter that

can be altered to target specific cell populations. For instance,
while injection into the ventricles allows the targeting of a large
region because of the diffusion of DNA, DNA can be introduced
directly into a specific structure to better control the size of the
transfected area. Finally, the ability to choose a particular develop-
mental time point for the electroporation extends the range of
brain areas that can be targeted. Moreover, depending on this time
point, different developmental events can be studied including cell
proliferation, cell migration, axon growth, dendrite development,
or synaptogenesis (Table 2). This technique can also be used to
electroporate plasmids that provide inducible gene expression or
inducible RNAi-mediated silencing, thus refining the temporal
control of gene expression. By using this system, the timing of
gene targeting can be finely controlled, which allows for a detailed
analysis of the cellular functions of the target gene.
Besides the spatial and temporal restrictions that can be
achieved in the developing brain, in utero electroporation presents
also the advantage to allow introduction of multiple plasmids in
the same cells, which is difficult to achieve with viral constructs.
Gain- and loss-of-function studies using in utero electropora-
tion have greatly contributed to the characterization of molecular
mechanisms controlling specific steps of brain development.
The improvement of this technique, with for instance transposon
systems to promote stable gene expression [1, 2], should help in
the future to further understand mouse brain development.
 In Utero Electroporation in Mouse 287

Table 2
Developmental events that can be studied using in utero
electroporation and corresponding references

Developmental events Examples of references

Regional patterning [23, 29]
Proliferation [18, 30, 31]
Neuronal differentiation [19, 32, 33]
Neuronal migration [5, 26, 34]
Dendrite and spine development [3, 35]
Axon formation and guidance [21, 36, 37]
Synaptogenesis and synapse maturation [38, 39]
Astrogliogenesis [40, 41]

2 Materials

2.1 DNA Injection 1. DNA solution: Purify plasmid DNA with an endotoxin-free
Maxi-prep kit. Prepare plasmid DNA solution to desired con-
centrations in water and add Fast Green (final concentration
of 0.05 %; Sigma) (see Note 1).
2. Needles: Pull borosilicate glass capillaries (1.0 mm O.D. × 0.58 mm
I.D.) (Harvard Apparatus) using a micropipette puller.
3. Microloader tips (Eppendorf).

2.2 Surgery 1. Phosphate-buffered saline (PBS), pH 7.4: 10 g NaCl, 2 g KCl,

11.5 g Na2HPO4 7H2O, and 2 g KH2PO4 in 800 ml of
ddH2O. Bring volume to 1 l and adjust pH.
2. Autoclave: Autoclave surgical instruments and PBS.
3. Buprenorphine: Prepare analgesic solution in PBS to a final
concentration of 30 μg/ml.
4. Isoflurane.
5. Anesthetic induction chamber.
6. Heating pads.
7. Recovery chamber.
8. Water bath.
9. Surgical instruments: Extra thin Iris scissors (Fine Science
Tools), Curved Forceps (Fine Science Tools), Ring forceps
(Fine Science Tools), Needle holder (Fine Science Tools),
Graefe Forceps (Fine Science Tools).
10. Beaker.
11. Electric razor.
288 Emilie Pacary and François Guillemot

12. Eye gel.

13. Chlorhexidine.
14. Mask and sterile gloves.
15. Scalpel.
16. Sterile drapes.
17. Sterile swabs.
18. Cotton buds.
19. Vicryl absorbable suture (Ethicon Inc.).

2.3 Electroporation 1. Electroporator.

2. Electrodes: Paddle-type or wire-type electrodes.
3. Microinjector (Femtojet Eppendorf).
4. Foot control for Femtojet microinjector (Eppendorf).
5. Capillary holder (Eppendorf).

2.4 Tissue 1. Sodium pentobarbitone.

Processing 2. Pump for perfusion.
3. Fluorescent binocular.
4. 4 % Paraformaldehyde (PFA) in PBS.
5. 20 % sucrose in PBS.
6. OCT Compound (VWR International).
7. Molds for embedding (HistoMold, Leica).
8. Vibratome or cryostat.
9. Microscope slides.
10. Coverslips.
11. Mounting medium: Aqua Poly/mount (Polysciences Inc).

3 Methods

3.1 Preparation of 1. Weigh a pregnant mouse at appropriate stage of pregnancy

the Surgery and inject subcutaneously 0.1 mg/kg of the analgesic
buprenorphine at least 30 min before the surgery.
2. Turn on the heating pads and the recovery chamber. Place
sterile PBS in warm water bath and all sterile surgical instru-
ments and materials (scalpel, drape, swabs, cotton buds, and
sutures) on a sterile drape. Place the electrodes into a PBS-
filled beaker and connect to the electroporator.
3. Fill the needle with the DNA solution using a microloader tip
and connect the needle to the capillary holder.
 In Utero Electroporation in Mouse 289

3.2 Surgery Before 1. Anesthetize the pregnant mouse with isoflurane in oxygen
DNA Injection and carrier (oxygen 2 l/min) using an anesthetic induction cham-
Electroporation ber. Wait until the animal loses righting reflex.
2. Transfer the animal to a “presurgery” mask. Place a drop of
eye gel on each eye to prevent corneal ulceration of the eyes
while it is under general anesthesia. Use an electric razor to
shave the hair of the abdomen. Clean the shaved area once
with chlorhexidine to collect flying hair.
3. Transfer the animal to a second mask in the surgical area. Place
the mouse with its back on the heating pad. Start the surgery
when the pedal reflex has been lost.
4. Put on mask and sterile gloves. Cover the animal with a sterile
drape (with a small hole over the abdomen) to prevent the tis-
sues and instruments from being contaminated by the areas of
skin that have not been shaved and disinfected. Clean the
shaved area at least three times with chlorhexidine. Use a dif-
ferent sterile cotton bud each time. Use a scalpel to make a
vertical incision along the midline (~1 in. long) through the
skin. Using scissors, make a similar incision of the muscle of
the abdomen along the linea alba (white line composed mostly
of collagen connective tissue).
5. Choose the most accessible embryos, place the ring forceps
between two embryos, and carefully pull the embryonic chain
out of the abdominal cavity (see Note 2). From this point on,
keep the embryos hydrated with sterile pre-warmed PBS.

3.3 Injection of DNA 1. Pinch off the tip of the needle with a forceps.
and Electroporation 2. Start with one of the most lateral embryos, making it easier to
keep track of which embryos were electroporated. Manipulate
the position of the embryo inside the amniotic sac using the ring
forceps and stabilize the head of the embryos between the rings.
Squeeze gently to push up the embryo closer to the uterine wall.
3. With the other hand, take the capillary holder and insert the
needle carefully into the middle of the hemisphere to target
the lateral ventricle or into the desired structure (see Note 3).
Press the pedal to inject the DNA solution (see Note 4). The
location of the DNA solution is monitored with Fast Green
(see Note 5).
4. Place the electrodes on the sides of the embryo head to target
the desired area (see Note 6).
5. Apply electrical pulses (see Notes 7 and 8).
All the embryos in a pregnant mouse can be electroporated,
usually with the same DNA construct to avoid confusion when
harvesting electroporated embryos or pups (see Note 9). However,
290 Emilie Pacary and François Guillemot

a long surgery decreases the survival rate of embryos. The abdomi-

nal cavity should not remain open longer than 30 min.

3.4 Surgery Post 1. After electroporating the embryos, add PBS into the abdomi-
Electroporation nal cavity and use the ring forceps to replace the uterine horn
in its original location.
2. Suture the abdomen wall and skin with Vicryl absorbable sutures.
3. Place the animal in a recovery chamber until it wakes up (usually
5–10 min) and then transfer in a cage placed on a heating pad.

3.5 Post-surgery Check the behavior of the mice to assess pain, suffering, or distress
and weigh the animals 24 and 48 h after the surgery. If needed,
analgesics can be administered to minimize pain and discomfort.

3.6 Tissue Collect the electroporated embryos or pups at the embryonic or

Processing the postnatal stages required for the experiment.
1. For analysis at embryonic stages (for example to study cell
proliferation or migration):
Euthanize mother via cervical dislocation and collect the
embryos. After decapitation, select the brains that have been
properly electroporated, as indicated by the amount and loca-
tion of the fluorescent signal, visualized across the skull using
a fluorescent binocular microscope. Dissect the brain out of
the skull, fix overnight in 4 % PFA, and then place in 20 %
sucrose/PBS overnight. Embed in OCT compound, freeze at
−80 °C, and section using a cryostat.
2. For analysis at postnatal stages (for example to study
synaptogenesis):
Anesthetize pups or adult mice with intraperitoneal injec-
tion of pentobarbitone (40–60 mg/kg) and perform transcar-
dial perfusion with PBS, followed by 4 % PFA in PBS. Dissect
the brains out of the skull and postfix in 4 % PFA overnight.
After washings in PBS, section the brains using a vibratome.
In both types of analysis, sections can be mounted in Aqua
Poly/mount directly or after immunostaining.

4 Notes

1. The efficiency of electroporation is highly dependent on the

DNA concentration. A concentration of 1 μg/μl is generally
used. This concentration is usually sufficient to visualize elec-
troporated neurons (when a GFP-expressing plasmid is elec-
troporated) without affecting their development. However,
depending on the promoter used (low expression level with
cytomegalovirus promoter/enhancer, stronger expression
 In Utero Electroporation in Mouse 291

level with the cytomegalovirus immediate early enhancer and

chicken β-actin promoter fusion (CAG) promoter) as well as
the size and the stability of the protein expressed, plasmid con-
centration can be adjusted (0.25–5 μg/μl).
2. Do not pull too much on the embryos as this will increase the
risk of hemorrhage.
3. The sharpness of the needle is critical to pierce properly the
uterine wall. It is important to minimize the movement of the
needle at the surface of the uterine wall and after insertion into
the brain because an enlargement of the hole will result in the
leakage of amniotic fluid and embryo death. Avoid piercing
blood vessels in the uterine wall, as this will result in bleeding
and embryo death.
4. It is important not to inject a too large volume of DNA into
the lateral ventricle because it will induce hydrocephalus (do
not exceed 2 µl at E14.5). The volume of DNA is adjusted for
the purpose of the experiment. 1 µl is generally used for most
experiments but a smaller volume of DNA may be required
(for example to electroporate fewer cells and visualize the den-
dritic arbor of isolated electroporated neurons).
5. In case of injection into the ventricle, it is worth noting that
the size of the electroporated area increases with the time
between the injection and the application of currents.
6. Selection of the correct size of electrodes and adequate arrange-
ment of electrodes are of key importance for successful electro-
poration. For instance, to target the cerebral cortex, 5 mm
platinum electrodes are often used and the positive paddle is
placed on the dorsolateral side of the injected ventricle [3].
7. Avoid applying current across the placenta as this will result in
embryo death.
8. The electric pulses induce the formation of transient pores in cell
membranes, thus allowing the uptake of DNA. These electro-
pores last longer when higher voltage and longer pulse duration
are applied. However, this also increases the risk of inducing non-
reversible cell membrane damage, which will cause cell death.
For instance, to electroporate the cerebral cortex at E14.5,
five 30 V electrical pulses (50 ms duration) at 1-s intervals are
often applied.
See Saito and Nakatsuji [4]: Table 1 “Effect of Voltage and
Pulse Numbers on Survival and EYFP-Positive Rate” and Table 2
“Optimal Conditions for Electroporation at Different Stages.”
9. Conditions (size electrodes, voltage …) should be optimized
for each electroporation experiment depending on the stage of
the embryo and brain region electroporated. See details in ref-
erences of Tables 1 and 2.
292 Emilie Pacary and François Guillemot

References
1. Chen F, LoTurco J (2012) A method for S (2009) Development of the mouse amygdala
stable transgenesis of radial glia lineage in rat as revealed by enhanced green fluorescent pro-
neocortex by piggyBac mediated transposi- tein gene transfer by means of in utero electro-
tion. J Neurosci Methods 207(2):172–180 poration. J Comp Neurol 513(1):113–128
2. Iguchi T, Yagi H, Wang CC, Sato M (2012) A 12. Nakahira E, Yuasa S (2005) Neuronal genera-
tightly controlled conditional knockdown sys- tion, migration, and differentiation in the
tem using the Tol2 transposon-mediated tech- mouse hippocampal primoridium as revealed
nique. PLoS One 7(3):e33380 by enhanced green fluorescent protein gene
3. Pacary E, Haas MA, Wildner H, Azzarelli R, transfer by means of in utero electroporation.
Bell DM, Abrous DN, Guillemot F (2012) J Comp Neurol 483(3):329–340
Visualization and genetic manipulation of 13. Navarro-Quiroga I, Chittajallu R, Gallo V,
dendrites and spines in the mouse cerebral Haydar TF (2007) Long-term, selective gene
cortex and hippocampus using in utero expression in developing and adult hippo-
electroporation. J Vis Exp (65):e4163. doi: campal pyramidal neurons using focal in
10.3791/4163 utero electroporation. J Neurosci 27(19):
4. Saito T, Nakatsuji N (2001) Efficient gene trans- 5007–5011
fer into the embryonic mouse brain using in vivo 14. Garcia-Frigola C, Carreres MI, Vegar C,
electroporation. Dev Biol 240(1):237–246 Herrera E (2007) Gene delivery into mouse
5. Pacary E, Heng J, Azzarelli R, Riou P, Castro retinal ganglion cells by in utero electropora-
D, Lebel-Potter M, Parras C, Bell DM, Ridley tion. BMC Dev Biol 7:103
AJ, Parsons M, Guillemot F (2011) Proneural 15. Petros TJ, Rebsam A, Mason CA (2009) In
transcription factors regulate different steps of utero and ex vivo electroporation for gene
cortical neuron migration through Rnd- expression in mouse retinal ganglion cells. J
mediated inhibition of RhoA signaling. Vis Exp (31):e1333. doi:10.3791/1333
Neuron 69(6):1069–1084 16. Punzo C, Cepko CL (2008) Ultrasound-
6. Shimogori T, Ogawa M (2008) Gene applica- guided in utero injections allow studies of the
tion with in utero electroporation in mouse development and function of the eye. Dev Dyn
embryonic brain. Dev Growth Differ 237(4):1034–1042
50(6):499–506 17. Borrell V, Yoshimura Y, Callaway EM (2005)
7. Tabata H, Nakajima K (2001) Efficient in Targeted gene delivery to telencephalic inhibi-
utero gene transfer system to the developing tory neurons by directional in utero electro-
mouse brain using electroporation: visualiza- poration. J Neurosci Methods 143(2):
tion of neuronal migration in the developing 151–158
cortex. Neuroscience 103(4):865–872 18. Castro DS, Martynoga B, Parras C, Ramesh V,
8. Niwa M, Kamiya A, Murai R, Kubo K, Gruber Pacary E, Johnston C, Drechsel D, Lebel-
AJ, Tomita K, Lu L, Tomisato S, Jaaro-Peled Potter M, Garcia LG, Hunt C, Dolle D, Bithell
H, Seshadri S, Hiyama H, Huang B, Kohda K, A, Ettwiller L, Buckley N, Guillemot F (2011)
Noda Y, O’Donnell P, Nakajima K, Sawa A, A novel function of the proneural factor Ascl1
Nabeshima T (2010) Knockdown of DISC1 in progenitor proliferation identified by
by in utero gene transfer disturbs postnatal genome-wide characterization of its targets.
dopaminergic maturation in the frontal cortex Genes Dev 25(9):930–945
and leads to adult behavioral deficits. Neuron 19. Rouaux C, Arlotta P (2010) Fezf2 directs the
65(4):480–489 differentiation of corticofugal neurons from
9. Bai J, Ramos RL, Paramasivam M, Siddiqi F, striatal progenitors in vivo. Nat Neurosci
Ackman JB, LoTurco JJ (2008) The role of 13(11):1345–1347
DCX and LIS1 in migration through the lat- 20. Gelman DM, Martini FJ, Nobrega-Pereira S,
eral cortical stream of developing forebrain. Pierani A, Kessaris N, Marin O (2009) The
Dev Neurosci 30(1–3):144–156 embryonic preoptic area is a novel source of
10. Remedios R, Huilgol D, Saha B, Hari P, cortical GABAergic interneurons. J Neurosci
Bhatnagar L, Kowalczyk T, Hevner RF, Suda 29(29):9380–9389
Y, Aizawa S, Ohshima T, Stoykova A, Tole S 21. Bonnin A, Torii M, Wang L, Rakic P, Levitt P
(2007) A stream of cells migrating from the (2007) Serotonin modulates the response of
caudal telencephalon reveals a link between the embryonic thalamocortical axons to netrin-1.
amygdala and neocortex. Nat Neurosci Nat Neurosci 10(5):588–597
10(9):1141–1150 22. Kataoka A, Shimogori T (2008) Fgf8 controls
11. Soma M, Aizawa H, Ito Y, Maekawa M, Osumi regional identity in the developing thalamus.
N, Nakahira E, Okamoto H, Tanaka K, Yuasa Development 135(17):2873–2881
 In Utero Electroporation in Mouse 293

23. Vue TY, Bluske K, Alishahi A, Yang LL, King C, Sklar P, Sive H, Tsai LH (2011)
Koyano-Nakagawa N, Novitch B, Nakagawa Y Common DISC1 polymorphisms disrupt
(2009) Sonic hedgehog signaling controls tha- Wnt/GSK3beta signaling and brain develop-
lamic progenitor identity and nuclei specifica- ment. Neuron 72(4):545–558
tion in mice. J Neurosci 29(14):4484–4497 34. Elias LA, Wang DD, Kriegstein AR (2007)
24. Kawauchi D, Taniguchi H, Watanabe H, Saito Gap junction adhesion is necessary for radial
T, Murakami F (2006) Direct visualization of migration in the neocortex. Nature
nucleogenesis by precerebellar neurons: 448(7156):901–907
involvement of ventricle-directed, radial fibre- 35. Cubelos B, Sebastian-Serrano A, Beccari L,
associated migration. Development 133(6): Calcagnotto ME, Cisneros E, Kim S, Dopazo
1113–1123 A, Alvarez-Dolado M, Redondo JM, Bovolenta
25. Nishiyama J, Hayashi Y, Nomura T, Miura E, P, Walsh CA, Nieto M (2010) Cux1 and Cux2
Kakegawa W, Yuzaki M (2012) Selective and regulate dendritic branching, spine morphol-
regulated gene expression in murine Purkinje ogy, and synapses of the upper layer neurons of
cells by in utero electroporation. Eur J the cortex. Neuron 66(4):523–535
Neurosci 36(7):2867–2876 36. Fang WQ, Ip JP, Li R, Ng YP, Lin SC, Chen Y,
26. Okada T, Keino-Masu K, Masu M (2007) Fu AK, Ip NY (2011) Cdk5-mediated phos-
Migration and nucleogenesis of mouse precer- phorylation of Axin directs axon formation
ebellar neurons visualized by in utero electro- during cerebral cortex development.
poration of a green fluorescent protein gene. J Neurosci 31(38):13613–13624
Neurosci Res 57(1):40–49 37. Yi JJ, Barnes AP, Hand R, Polleux F, Ehlers
27. Alvarez-Maya I, Navarro-Quiroga I, Meraz- MD (2010) TGF-beta signaling specifies
Rios MA, Aceves J, Martinez-Fong D (2001) axons during brain development. Cell
In vivo gene transfer to dopamine neurons of 142(1):144–157
rat substantia nigra via the high-affinity neuro- 38. Elias GM, Elias LA, Apostolides PF, Kriegstein
tensin receptor. Mol Med 7(3):186–192 AR, Nicoll RA (2008) Differential traffick-
28. Willett RT, Greene LA (2011) Gata2 is ing of AMPA and NMDA receptors by
required for migration and differentiation of SAP102 and PSD-95 underlies synapse
retinorecipient neurons in the superior collicu- development. Proc Natl Acad Sci USA
lus. J Neurosci 31(12):4444–4455 105(52):20953–20958
29. Fukuchi-Shimogori T, Grove EA (2001) 39. Trimbuch T, Beed P, Vogt J, Schuchmann S,
Neocortex patterning by the secreted signaling Maier N, Kintscher M, Breustedt J, Schuelke
molecule FGF8. Science 294(5544): M, Streu N, Kieselmann O, Brunk I, Laube G,
1071–1074 Strauss U, Battefeld A, Wende H, Birchmeier
30. Lange C, Huttner WB, Calegari F (2009) C, Wiese S, Sendtner M, Kawabe H,
Cdk4/cyclinD1 overexpression in neural stem Kishimoto-Suga M, Brose N, Baumgart J,
cells shortens G1, delays neurogenesis, and Geist B, Aoki J, Savaskan NE, Brauer AU,
promotes the generation and expansion Chun J, Ninnemann O, Schmitz D, Nitsch R
of basal progenitors. Cell Stem Cell 5(3): (2009) Synaptic PRG-1 modulates excitatory
320–331 transmission via lipid phosphate-mediated
31. Weimer JM, Yokota Y, Stanco A, Stumpo DJ, signaling. Cell 138(6):1222–1235
Blackshear PJ, Anton ES (2009) MARCKS 40. Garcia-Marques J, Lopez-Mascaraque L
modulates radial progenitor placement, prolif- (2013) Clonal Identity Determines Astrocyte
eration and organization in the developing cere- Cortical Heterogeneity. Cereb Cortex 23(6):
bral cortex. Development 136(17):2965–2975 1463–1472
32. Nguyen L, Besson A, Heng JI, Schuurmans C, 41. Subramanian L, Sarkar A, Shetty AS,
Teboul L, Parras C, Philpott A, Roberts JM, Muralidharan B, Padmanabhan H, Piper M,
Guillemot F (2006) p27kip1 independently Monuki ES, Bach I, Gronostajski RM,
promotes neuronal differentiation and migra- Richards LJ, Tole S (2011) Transcription fac-
tion in the cerebral cortex. Genes Dev tor Lhx2 is necessary and sufficient to suppress
20(11):1511–1524 astrogliogenesis and promote neurogenesis in
33. Singh KK, De Rienzo G, Drane L, Mao Y, the developing hippocampus. Proc Natl Acad
Flood Z, Madison J, Ferreira M, Bergen S, Sci USA 108(27):E265–E274
 Chapter 20

The Cre/Lox System to Assess the Development

of the Mouse Brain
Claudius F. Kratochwil and Filippo M. Rijli

Abstract
Cre-mediated recombination has become a powerful tool to confine gene deletions (conditional knock-
outs) or overexpression of genes (conditional knockin/overexpression). By spatiotemporal restriction of
genetic manipulations, major problems of classical knockouts such as embryonic lethality can be circum-
vented. Furthermore Cre-mediated recombination has broad applicability in the analysis of the cellular
behavior of subpopulations and cell types as well as for genetic fate mapping. This chapter will give an
overview about applications for the Cre/LoxP system and their execution.

Key words Cre recombinase, Transgenesis, Conditional knockout, Conditional knockin, CreERT2,
Flpe recombinase, MADM, Split-Cre, Brainbow

1 Introduction

After the first gene knockout (KO) in the mouse was obtained by
Thomas and Capecchi using site-directed mutagenesis of the
HPRT gene in 1987 [1], the in vivo function of many genes has
been analyzed using this technique. Nevertheless the KO approach
has two main restrictions. First of all, genes whose inactivation is
embryonically lethal cannot be analyzed for their function in late
development and adulthood. Secondly, KOs of genes which have a
function in multiple tissues and/or cell types are difficult to ana-
lyze, since phenotypes might be combinations of multiple distinct
defects and therefore quite complex to dissect.
In 1994 Gu et al. described for the first time the use of the
Cre/Lox recombination system to induce a gene knockout in mice
[2]. This technology made it possible to conditionally knockout
genes solely in subsets of cells (i.e., in a cell-type- or tissue-specific
manner), where Cre recombinase is expressed. Two years before,
the technique had been already used to conditionally overexpress
the SV40 large tumor antigen in mice [3]. The key principle of the

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9_20, © Springer Science+Business Media, LLC 2014

295
296 Claudius F. Kratochwil and Filippo M. Rijli

Cre-mediated recombination is that the recombinase enzyme can

catalyze the deletion or inversion of a genomic fragment depend-
ing on the orientation of small recognition sequences, called Lox or
LoxP sites, flanking such fragment.
Since then, Cre-mediated recombination has been used suc-
cessfully for many applications, tissues [4–9] and model organisms
including, e.g., Drosophila [10], Xenopus [11], zebrafish [12], and
plants [13]. Also the recombination mechanism has been eluci-
dated by the analysis of the crystal structure of Cre and the
Cre/LoxP interface [14].
In mammalian cells, Cre is the predominantly used recombinase
for site-specific recombination and has been shown to be more effec-
tive than another (in Drosophila widely used) recombination system,
the Flp/Frt system from Saccharomyces cerevisiae [15]. The Flp
recombinase has a similar ability as Cre to delete or invert (and there-
fore was named Flp (pronounced “Flip”) recombinase) DNA frag-
ments, though recognizing distinct target sequences. Later on a more
efficient version, Flpe has been created which works reliably in mice
[16, 17] and offers a suitable alternative for some applications.
To further restrict the knockout in time, inducible variants of
Cre were designed. The most widely used version is CreERT2 [18].
Hereby Cre is fused to a mutated ligand-binding domain of the
estrogen receptor (ER) [19]. The fusion protein is normally con-
fined to the cytoplasm, while in the presence of the synthetic ligand
tamoxifen or 4-hydroxytamoxifen, it translocates to the nucleus,
where it can trigger recombination events (see Subheading 3.5).
A light-activatable Cre recombinase to control activity in time and
space has also been generated [20].
Another idea to restrict Cre or CreERT2 further in space came
from Hirrlinger et al. [21–23]. By splitting Cre into two inactive
fragments, which regain Cre activity when co-expressed, recombi-
nation could be restricted to the intersection of two expression
domains (by using different enhancers/promoters for the two
fragments). A similar system was established by Farago et al. [24],
using a combination of the two recombination systems Cre/Lox
and Flpe/Frt (see Subheading 3.6). Both systems have been also
tested in vivo in transgenic animals.
A powerful tool to construct a fate map of cells as well as to
analyze gene function on a single-cell resolution was introduced
with the MADM (mosaic analysis with double markers) system
[25, 26] an adaptation of the MARCM (mosaic analysis with a
repressible cell marker) system from Drosophila [27]. MADM uses
Cre-mediated interchromosomal recombination. By employing
two markers (e.g., GFP and RFP), not only single-cell progenies
can be traced, but the combination with mutations makes it
possible to distinguish mutant, heterozygous, and wild-type cells
by their distinct fluorescent markers [28] (see in detail in
Subheading 3.7).
 Cre/Lox in Mouse Development 297

Another attempt to achieve single-cell resolution was given by

the brainbow system from Livet et al. [29]. Here, recombination
leads to a stochastic choice of expression of a fluorescent protein.
Multiple integrations result in a combinatorial color code, creating
approximately 100 distinguishable colors (see Subheading 3.8),
which is especially helpful for the analysis of the nervous system.
In this chapter, we will focus on the basic principles and appli-
cations of the Cre/LoxP and CreERT2/LoxP systems and their
variations (Split-Cre, MADM, brainbow) and will provide strate-
gies and protocols for their use.

2 Materials

2.1 Reporter Mice 1. ROSA26 LacZ reporter line [30]; available at Jackson
Laboratories:
# 003504 B6.129S4-Gt(ROSA)26Sor tm1Sor/J.
2. ROSA26 tdTomato reporter line [31]; available at Jackson
Laboratories:
# 007914 B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J.
3. ROSA26 ZsGreen reporter line [31]; available at Jackson
Laboratories:
# 007906 B6.Cg-Gt(ROSA)26Sor tm6(CAG-ZsGreen1)Hze/J.
4. Brainbow reporter lines [29]; six transgenic lines available at
Jackson Laboratories:
# 007901 B6.Cg-Tg(Thy1-Brainbow1.0)HLich/J.
# 007910 B6;CBA-Tg(Thy1-Brainbow1.0)LLich/J.
# 007911 B6.Cg-Tg(Thy1-Brainbow1.1)MLich/J.
# 007921 B6.Cg-Tg(Thy1-Brainbow2.1)RLich/J.
# 013731 STOCK Gt(ROSA)26Sor tm1(CAG-Brainbow2.1)Cle/J.
# 017492 B6.129P2-Gt(ROSA)26Sortm1(CAG-Brainbow2.1) Cle/J.
5. MADM mice [25, 26, 32, 33]; several transgenic lines are
available at Jackson Laboratories, the following are the most
widely used:
# 013749 STOCK Tg(ACTB-EGFP,-tdTomato)11Luo/J.
# 013751 STOCK Tg(ACTB-tdTomato,-EGFP)11Luo/J.
# 017932 STOCK Tg(ACTB-EGFP*)10Luo/J.
# 017923 STOCK Tg(ACTB-EGFP*,-tdTomato)10Luo/J.
# 017912 STOCK Gt(ROSA)26Sor tm6(ACTB-EGFP*,-tdTomato)Luo/J.
# 017921 STOCK Gt(ROSA)26Sortm7(ACTB-EGFP*)Luo/J.
298 Claudius F. Kratochwil and Filippo M. Rijli

2.2 Generation of 1. Genomic DNA (for amplifying enhancers).

Genetically Modified 2. Cloning plasmid(s) (for Cre, CreERT2, Lox sites, minimal
Mice promoter, fluorescent proteins, resistance cassettes (e.g., Addgene)).
3. Standard reagents for molecular biology (restriction enzymes
(NEB), ligase (NEB), competent E. coli (e.g., DH5alpha,
Top10 (Invitrogen)), primers, antibiotics).
4. Modified recombinogenic bacterial strains for recombination
[34] (see Note 1).

2.3 Tamoxifen 1. Gavage/feeding needle to administer tamoxifen (Fine Science

Treatment for CreERT2- Tools).
Mediated 2. Tamoxifen (SIGMA).
Recombination 3. Corn oil (SIGMA) or sunflower oil.
4. Syringe (graded in 100 μl intervals).

2.4 Genotyping 1. Standard reagents for DNA extraction (1 M Tris–HCl pH 8.5;

5 M NaCl; 0.5 M EDTA; 20 % SDS; proteinase K; isopropa-
nol; ethanol).
2. Standard reagents and primers for polymerase chain reaction
(PCR) or GoTaq® Green Master Mix (Promega).
3. Standard reagents for gel electrophoresis.

3 Methods

3.1 Applications of To create a conditional knockout, the gene of interest is flanked with
the Cre/LoxP System recognition sites (LoxP sites; Locus of crossing [x-ing]-over of
bacteriophage P1) for the bacteriophage P1 Cre recombinase [35]
(Fig. 1). If Cre recombinase is present, the sequence flanked by two
LoxP sites (the “floxed” sequence) is excised (if Lox sites have the
same orientation) or inverted (if Lox sites have opposing orientation)
(Fig. 1). It is also possible to recombine between two different plas-
mids or chromosomes (interchromosomal recombination), which is,
for example, used in the MADM system (see Subheading 3.7). The
excision of genomic fragments gives the possibility to perturb gene
function either (a) completely, by removing the whole gene or the
start codon; (b) partly, by removing certain parts/exons of the gene
or by truncating it; or by (c) changing its expression pattern and
levels by excising/replacing/modifying promoter or enhancer ele-
ments. In a similar way, Cre-mediated recombination can be used to
overexpress genes, by excision of an intervening transcriptional ter-
mination sequence, flanked by LoxP sites, that prevents the tran-
scription of the target gene [3] (Fig. 2b). Since these constructs
might result in low levels of leaking readthrough transcription, espe-
cially if many copies are present (e.g., when using gene transfer by
 Cre/Lox in Mouse Development 299

Fig. 1 Distinct genomic outcomes of Cre/LoxP-mediated recombination. (a) If two Lox sites have opposing direction-
ality, the sequence in between is inversed upon Cre-mediated recombination. The recombination is reversible. (b) If
Lox sites have the same orientation, the sequence in between is recombined out, resulting in a circular fragment and
the sequence lacking this fragment. (c) If LoxP sites are present on two chromosomes, recombination can also occur
interchromosomally, albeit at a lower frequency than if the LoxP sites are on the same chromosome. (d) The LoxP
site is 34 base pair (bp) long. Two palindromic 13 bp sequences, containing the Cre binding site (boxed), flank an
asymmetric 8 bp sequence. Arrows indicate the sites of cleavage during recombination. The underlined base pairs
are mutated in the most commonly used variants LoxN and Lox2722 (modified from [36])

viruses or electroporation), a different approach (also called flip-

excision (FLEX) switch) was created, using LoxP sites, which are put
in inverted orientation (Fig. 2c), causing an inversion of the inter-
vening DNA (Fig. 1). Thereby the transcription of a gene can be
initiated by inverting the open-reading frame. Because here none of
the LoxP sites is removed, the inversion would continue forth and
back. To interfere with that, a second LoxP-incompatible site pair is
introduced (e.g., Lox2272) which results in the termination of the
ongoing recombination by cutting out the specific recognition part-
ners of both Lox sites (Fig. 2) [37].
It should be mentioned that the efficiency of Cre/LoxP-
mediated recombination decreases in general with increasing
genetic distance, but in principle any desired rearrangement can be
made with the Cre/LoxP system [38].

3.2 Designing Cre/LoxP is a binary system. On the one hand, the gene of interest
Constructs has to be flanked by LoxP sites (“floxed”) (Fig. 1). On the other hand
recombinase expression in the cell or tissue of interest is provided
300 Claudius F. Kratochwil and Filippo M. Rijli

Fig. 2 Three examples for applications using the Cre/LoxP system. (a) The classic approach to generate a
conditional knockout is to place LoxP sites in introns either flanking the exon, which contains the ATG, or exons
that contain functionally important domains. Using this strategy, the probability to interfere with gene expres-
sion before Cre recombination is reasonably smaller than if LoxP sites are placed in promoter regions. (b) To
overexpress genes usually an intervening sequence is placed between promoter and target gene, which blocks
transcription. Upon Cre-mediated recombination this sequence is removed, and the gene starts to be expressed
under the control of the upstream promoter. (c) To reduce the risk of leaking, recently the use of double-
inverted Lox sites became an alternative approach (flip-excision switch, or FLEX). Here, the gene of interest is
placed in inverted or antisense orientation. Cre mediates the inversion of the sequence in the presence of a
pair of Lox sites in inverted orientation. The use of a second pair of Lox sites also in reverse orientation, though
incompatible with the first pair, eventually results in an irreversible sense configuration, since both sites lack
at the end a partner site, allowing stable transcription

either by (a) a knockin (KI) of Cre, (b) a transgene in which Cre

expression is driven by a specific promoter/enhancer, (c) a bacterial
artificial chromosome (BAC) in which Cre is inserted in-frame at a
specific locus, or (d) gene transfer using, e.g., viruses or in utero
electroporation (Fig. 3). Many tissue- or cell-type-specific enhanc-
ers have been described in the literature. Others can be found in
enhancer databases such as the VISTA Enhancer Browser [39] or
can be identified by selecting evolutionarily conserved sequences in
proximity of genes with expression patterns of interest [40] and by
cloning a fragment of a few kilobases of core and proximal pro-
moter in front of the Cre recombinase gene. Finally, relevant
 Cre/Lox in Mouse Development 301

Fig. 3 Two approaches to obtain tissue-/cell-type-specific Cre expression. (a) A tissue-specific enhancer is
combined with a minimal promoter to drive tissue-specific Cre expression, where the enhancer is active. (b)
Another approach is site-directed recombination into a bacterial artificial chromosome (BAC) or into the
genome of embryonic stem cells

enhancers can be also identified by genome-wide search, e.g., by

ChIP-Seq for the p300 protein [41]. If enhancers are used which
lack the core promoter element, they are usually combined with the
minimal/core promoter (minP) of the human ß-globin promoter
[42], which is unable to drive efficient transcription if no additional
enhancer sequences are present proximally.

3.3 Mouse Mating Once Cre KI (i.e., inserted at specific loci) or Cre-expressing trans-
Schemes genic mice have been created, they need to be crossed to mice
carrying LoxP site-bearing conditional alleles, in order to generate
double heterozygotes for Cre and the floxed locus. Since most
laboratories dispose of multiple Cre drivers, it is more space effi-
cient to generate a few double heterozygous males for the Cre-
expressing line and the conditional allele and mate them to
homozygous conditional mutant females, which can be readily
maintained as a pool. However, it should be noted that in this type
of crossings, double heterozygous Cre/conditional allele and/or
homozygous conditional mutant specimen in the absence of Cre-
mediated excision usually displays a wild-type-like phenotype, thus
serving as controls. It may be useful to additionally cross a floxed
conditional reporter line into the background of Cre/conditional
allele double heterozygotes. In many projects this will ease the
analysis, as it will allow to (1) have a direct readout of the cells in
which the gene has been knocked out, (2) directly analyze cellular
behavior between heterozygotes and homozygotes, and (3) select
tissue or cells using methods such as fluorescent-activated cell sort-
ing (FACS). Conditional Cre-inducible reporter mice in which the
reporter genes are inserted in housekeeping gene loci are available
from the Jackson Laboratories. These lines provide stable and con-
stitutive reporter expression once activated by Cre-mediated exci-
sion. In particular, three KIs in the ROSA26 locus are the most
frequently used reporter lines since they are highly sensitive to Cre-
mediated activation, which express either ß-galactosidase (LacZ)
[30], ZsGreen or tdTomato [31], but additional reporter lines
expressing other fluorescent proteins are also available (e.g., [43]).
The choice depends on the application. ß-Galactosidase catalyzes
the transformation of X-gal into an insoluble blue enzymatic product
302 Claudius F. Kratochwil and Filippo M. Rijli

Fig. 4 Mating scheme to analyze gene function by conditional knockout. The use of pools for the conditional
knockout (with or without reporter) eases the analysis if multiple Cre drivers are used to test gene function in
different tissues. The offspring contains animals displaying mutant (Cre tg/+, Gene fl/fl ) and control (Cre tg/+,
Gene fl/+, and Gene fl/fl ) phenotypes

and is preferentially used for nonfluorescent histochemistry or

whole mount stainings of embryos or organs. The strongly fluores-
cent proteins ZsGreen (see Note 2) or tdTomato are a better choice
for fluorescence histochemistry, live imaging, or cell sorting. The
crossing scheme (Fig. 4) will give litters containing all needed con-
trols: 25 % homozygotes (+Cre), 25 % homozygotes (−Cre), and
25 % heterozygotes (−Cre). The controls can be used to exclude an
effect of the Cre (heterozygotes +Cre) or of the floxed locus
(homozygotes −Cre) when analyzing the phentoype of the homo-
zygous mutants.

3.4 Genotyping To genotype mice for maintenance and experiments, polymerase

chain reaction (PCR) is the method of choice. Primers have to be
designed to simultaneously detect (1) the Cre, (2) the floxed locus,
(3) the locus after Cre-mediated deletion, and (4) the wild-type
locus of the respective gene. For (1), specific primers for Cre can
be used or, alternatively, primers which span the Cre extremities
and its flanking regions. For (2–4), usually three primers are needed
(Fig. 5). One primer pair is chosen on each side of one or the other
of the LoxP sites to detect the small size difference, as compared to
the wild-type fragment, due to the presence of the LoxP site
(Fig. 5b). Another pair of primers is designed 5′ and 3′ to the two
LoxP sites, respectively, to be able to detect the Cre-mediated dele-
tion of the locus and confirm the efficiency of the recombination
(Fig. 5c). PCRs for genotyping can be done using standard PCR
protocols for genotyping on clipped toes, ears, or tails or any other
tissue, where recombination has to be tested. We recommend the
 Cre/Lox in Mouse Development 303

Fig. 5 Genotyping by PCR to detect gene deletion. (a) The targeted allele can be detected by primer pair 1 (F1
and R1). (b) The floxed allele can be detected by the same primers, yielding a longer amplicon. (c) A distinct
primer pair (F2 and R1) detects the Cre-mediated recombination. The non-recombined locus usually cannot be
detected, as the amplicon will be too long. (d) Theoretical result of a PCR with different genotypes in Cre-
positive (Cre+) and Cre-negative (Cre−) tissues

use of the GoTaq® Green Master Mix (Promega), because it is

faster, reduces pipetting mistakes and contaminations, and increases
reproducibility.
DNA extraction:
1. Prepare tail buffer (500 ml stock):

100 mM Tris–HCl pH 8.5 50 ml of 1 M

200 mM NaCl 20 ml of 5 M
5 mM EDTA 5 ml of 0.5 M
0.2 % SDS 5 ml of 20 % SDS
ddH2O 420 ml

2. Clip the tails or ears and collect them in 1.5 ml Eppendorf

tubes.
3. Add 500 μl tail buffer and 10 μl proteinase K (20 mg/ml).
4. Digest overnight in a water incubator at 55 °C or 2 h on a
shaking dry incubator at 55 °C.
5. Vortex tubes and centrifuge for 5 min at 12,000–16,000 × g.
6. Collect supernatant in new tube (to remove undigested tissue)
with 500 μl isopropanol (2-propanol).
304 Claudius F. Kratochwil and Filippo M. Rijli

7. Shake vigorously and centrifuge for 10–15 min at

12,000–16,000 × g.
8. Remove supernatant, add 500 μl 70 % ethanol to wash pellet,
and remove supernatant again.
9. Add 500 μl ddH2O and vortex vigorously.
Genotyping using GoTaq® Green Master Mix:
1. Aliquot 2× Master Mix into aliquots between 200 μl and 1 ml.
2. Prepare Primer Mix for each genotype (200 μl stock):

Upstream primer, 100 μM 20 μl

Downstream primer, 100 μM 20 μl
ddH2O 160 μl

If two up- or downstream primers are needed, use 10 μl of each.

3. Prepare Master Mix for each primer set:

1 reaction 20 reactions
®
GoTaq Green Master Mix, 2× 4 μl 80 μl
Primer Mix, 10 μM for each primer 0.4 μl 8 μl

4. Pipet 4 μl genomic DNA template and 4 μl Master Mix (for

many samples a dispenser can be used) per reaction tube. Also
half reactions can be done.
5. Annealing conditions should be optimized. It is recommended
to design all primers for approximately the same annealing
temperature (e.g. 58 °C) because PCRs can be pooled in the
same block and different primers can be combined. The fol-
lowing program works for many primer sets:

Initial denaturation 30–35 cycles: 3 min 94 °C

Denaturation 40 s 94 °C
Annealing 40 s 52–62 °C
Extension 1 min 72 °C
Final extension 5 min 72 °C
Soak/refrigeration cycle – 4 °C

6. Load DNA samples on a 1.5 % agarose gel (one band) 2 %

agarose gel (multiple bands with <100 bp difference).
 Cre/Lox in Mouse Development 305

3.5 Inducible If the gene knockout needs to be induced at a specific developmen-

Conditional Knockouts tal stage, a tamoxifen-inducible form of Cre, CreERT2 [18], can be
used. In this case, Cre activity is only induced if the synthetic hor-
mone tamoxifen is administered. For the activation of CreERT2 in
embryos, pregnant mice are injected with tamoxifen (dissolved in
corn or sunflower oil) using a gavage needle.
Preparation of Tamoxifen (20 mg/ml):
1. Heat 5 ml of corn oil in a tube at 37 °C for 30 min.
2. Put tamoxifen at room temperature (RT).
3. Weigh 100 mg of tamoxifen and let the tamoxifen dissolve on
a shaker at 37 °C; vortex regularly. It will take a few hours till
the tamoxifen is completely dissolved.
4. Store at 4 °C (1 month) or make aliquots (0.5 ml) and store at
−20 °C.
Gavaging:
1. Hold the pregnant female mouse from the back, in a similar
fashion as for intraperitoneal injections, and secure such that
the head cannot move and the esophagus is straight.
2. Slowly enter the mouth of the animal and proceed along the
top of the mouth till the gavage needle enters first the esopha-
gus and then the stomach.
3. Slowly inject the oil with 1–10 mg tamoxifen (see Note 3) and
remove the needle.
4. Minimize the stress of the animal following tamoxifen admin-
istration (see Note 4).

3.6 Intersectional Specificity of Cre recombination in distinct cell subpopulations or

Recombination cell types may be limited by the lack of specific promoters to drive
Systems Cre expression. Two strategies have been designed to achieve
higher cellular specificity of gene recombination, which take advan-
tage of the intersectional activities of two promoters in the same
cell. The first approach has been elaborated in the Dymecki labora-
tory [6, 24] to analyze hindbrain development. Hereby the over-
lapping expression of two different recombinases, Cre recombinase
and Flpe recombinase, driven by two different promoters activates
the expression of a reporter gene (GFP) solely in cells where both
recombinases are present (Fig. 6b). Additionally the construct
contains a reporter (b-galactosidase), which is activated only if Cre
though not Flpe is expressed (Fig. 6b).
A second approach was created by Hirrlinger et al. [21, 22],
where Cre recombinase is split into two inactive parts, N-Cre (the
N-terminus) and C-Cre (C-terminus), which are fused to the
constitutive GCN4 protein–protein interaction domain, which has
been shown to create stable dimers [44]. When co-expressed in the
306 Claudius F. Kratochwil and Filippo M. Rijli

a Promoter 1 N-Cre Promoter 2 C-Cre

N-Cre N-Cre + C-Cre C-Cre

(not functional) (functional) (not functional)

P STOP Reporter P Reporter P STOP Reporter

Reporter expression
Activity promoter 1 (coexpressing area) Activity promoter 2

b
Promoter 1 Cre Promoter 2 Flpe

Promoter 1: off - Promoter 2: off No Cre / No Flpe P STOP Reporter Reporter

LoxP LoxP FRT FRT

Promoter 1: off - Promoter 2: on No Cre / + Flpe P STOP Reporter

Promoter 1: on - Promoter 2: off + Cre / No Flpe P Reporter Reporter

Promoter 1: on - Promoter 2: on + Cre / + Flpe P Reporter

Activity promoter 1 Reporter Expression Activity promoter 2

Fig. 6 Tools to generate intersectional recombination. (a) In the Split-Cre system, Cre is divided into two inac-
tive fragments expressed under the control of two distinct promoters. In the tissue, or cell type, where both
promoters are active, the two fragments are co-expressed resulting in a functional Cre. This triggers a recom-
bination event, e.g., the Cre-mediated activation of a reporter gene. (b) The Cre/Flpe system uses two different
recombinases recognizing distinct target sites. When both recombinases are active, a specific reporter gene is
expressed (e.g., GFP) which is not activated by either one of the recombinases alone
 Cre/Lox in Mouse Development 307

same cell, N-Cre-GCN4 and C-Cre-GCN4 were able to reconstitute

a functional Cre recombinase activity, which can induce recombina-
tion (Fig. 6a). The system was successfully tested with virus-mediated
gene transfer [21, 22] as well as by the use of transgenic mice [23].

3.7 MADM System The MADM (mosaic analysis with double markers) is an adapta-
tion in mouse of the conceptually similar MARCM (mosaic analy-
sis with a repressible cell marker) used in Drosophila [27]. Hereby,
rare interchromosomal recombination (Fig. 1c) during mitosis
results in sparse labeling of cells with one of two fluorescent
reporter genes (e.g., GFP or RFP). In differentiated cells inter-
chromosomal recombination can occur as well but can only result
in the expression of both (GFP and RFP) or no reporter (Fig. 7).
Because of the low frequency of recombinations during mitosis,
this allows (1) to follow the behavior of clonally related (i.e., gen-
erated from the same progenitor) cells [45] as well as (2) to sparsely
generate mutant cells in the background of a heterozygous mouse
labeled by a fluorescent marker (e.g., RFP).
Before recombination no fluorescent marker is expressed in
MADM mice, because the N-terminus sequence of either one of
the two fluorescent markers (separated by a LoxP site) is followed
by the C-terminus of the other marker (e.g., N-RFP-LoxP-C-GFP
and N-GFP-LoxP-C-RFP, respectively), resulting in nonfunctional
fluorescent proteins. After Cre-induced recombination, N-terminus
and C-terminus can be reconstituted to functional fluorescent pro-
teins (e.g., RFP and GFP in the old versions of MADM and GFP
and tdTomato in the new versions (Fig. 8)). Since recombination
can be tightly controlled by the use of tissue- or cell-specific Cre
drivers as well as by tamoxifen-inducible CreERT2 drivers for tempo-
ral induction, the lineage of single cells can be traced during devel-
opment and adulthood. Furthermore MADM can be used to
selectively knockout genes in labeled cells by introducing (by natu-
ral recombination) a mutant allele between the centromere and the
MADM transgene, while the second chromosome solely carries the
wild-type allele (besides the complementary MADM transgene/
knockin). In case of a X-segregation (sister chromatids segregating
into different daughter cells) in G2 phase of mitosis, two different
cells with different colors (red and green) and genotypes (homozy-
gote mutant and homozygote wild-type allele) are generated, while
a Z-segregation would result in two heterozygous sister cells (one
unlabeled and one labeled with both fluorescent markers) (Fig. 7).
In contrast to classical conditional knockouts, MADM
conditional knockouts have several advantages (Fig. 9). First of all,
non-cell-autonomous defects can be almost excluded, because just
a small proportion of cells carries the homozygous knockout
(which can be titered up or down by the use of a CreERT2 driver).
Secondly it has clear advantages, when the cellular character and
behavior (cell shape, cell migration, axon guidance) have to be
308 Claudius F. Kratochwil and Filippo M. Rijli

Fig. 7 The MADM system. The MADM system consists of two knockins, which are chimeric genes constituted
by the N-terminus and C-terminus of two distinct fluorescent proteins (e.g., RFP and GFP). Before Cre-mediated
recombination, the chimeric N-RFP/C-GFP or N-GFP/C-RFP alleles result in two nonfunctional proteins. During
the S phase of the mitosis, sister chromatids duplicate and pair. If an interchromosomal recombination hap-
pens, the correspondingly recombined sister chromatids segregate during the G2 phase into different daughter
cells (X-segregation). Two cells of different colors (i.e., expressing either red or green fluorescence) are then
generated. In case each sister chromatids segregate into the same cell, a double-labeled and colorless cell is
produced. A recombination during G1 or G0 (differentiated cell) phase results as well in a colorless and in a
double-labeled cell. If additionally a mutant allele is recombined onto a MADM-bearing chromosome (while the
other MADM containing chromosome has the wild-type allele), X-segregation results in the generation of sib-
ling mutant and wild-type cell, labeled with different fluorescent markers, while Z-segregation or recombina-
tion in G1 or G0 phase does not change the genotype of the cell

analyzed. Hereby it allows even the direct comparison of cells car-

rying homozygous mutations and no mutation.
Disadvantages of this technique are that MADM lines are not yet
available for all chromosomes, which restricts the use of it at the
moment to analyze knockouts of genes located on chr.6, chr.10, and
 Cre/Lox in Mouse Development 309

Fig. 8 MADM cassettes. To increase variability in the choice of the second reporter/gene, the old cassettes (a)
containing GFP and RFP have been replaced by one cassette containing an ATG and the C-terminus of GFP and
a second cassette containing N-GFP and an ATG-less tdTomato (b). If another gene except tdTomato should be
overexpressed (GeneX ), only one transgenic line has to be created making the system more universal

Fig. 9 Advantages of conditional knockouts and MADM conditional knockouts. While classical knockouts give
little information about the role of the knockedout gene in different tissues, the use of conditional knockouts
allows the analysis of gene function in selected cell subpopulations or tissues. The MADM system allows fur-
thermore to trace the lineage of single cells with different genotypes (homozygous knockout and wild type,
respectively) in an environment of heterozygous cells

chr.11 and only between the MADM insertion locus and the
centromere.
Recently the system has been further optimized by using con-
structs with tdTomato, a red fluorescent protein brighter than RFP,
as well as by making the choice of the second gene to be recom-
bined more flexible (e.g., by substituting a second fluorescent
reporter gene with a GeneX to be overexpressed) (Fig. 8) [26].

3.8 Brainbow An elegant approach to analyze the behavior of single cells in a tis-
System sue, especially for the analysis of connectivity patterns in neural
circuits, came from the Lichtman group [29]. By the use of
different, incompatible Lox sites (LoxN, Lox2722, and LoxP), these
authors generated mice, which, in a stochastic manner, are induced
to express only one fluorescent protein (YFP, CFP, or RFP) as a
final recombination event (Figs. 10 and 11).
Four different brainbow versions that stochastically express
different kinds of fluorescent proteins exist (Fig. 11a, b), and the
corresponding transgenic mouse lines are available from Jackson
310 Claudius F. Kratochwil and Filippo M. Rijli

Fig. 10 The brainbow system. Recombination of the Brainbow 1.1 constructs can result in three possible
outcomes. A recombination of the LoxN site results in the transcriptional activation of RFP. Because the other
two Lox sites (Lox2722 and LoxP ) lack a second site because they have been placed 3′ of the LoxN site, no
further recombination can occur. If the first recombination happens between the Lox2272 sites, the result is
expression of YFP; if it happens between the LoxP sites, it results instead in expression of CFP

Laboratories. Due to the effect that multiple integrations occur as

well as they independently result in the stochastic choice of one
color, combinations of color lead to almost 100 distinguishable
colors (Fig. 11c).

4 Notes

1. Detailed information and troubleshooting about the use of

recombination-mediated genetic engineering can be found here.
http://web.ncifcrf.gov/research/brb/recombineeringInfor
mation.aspx (31/07/13)
2. ZsGreen in contrast to tdTomato is not localizing strongly in
cellular extensions like axons or dendrites. For these kinds of
analyses, the tdTomato line (# 007914) would be the recom-
mended choice. For localization of cells ZsGreen (# 007906)
is better.
 Cre/Lox in Mouse Development 311

Fig. 11 Overview of brainbow constructs and principle of “multicolor” brainbow. (a, b) Four brainbow con-
structs have been generated (a) with different possible outcomes (b). While Brainbows 1.0 and 1.1 recombine
only once, Brainbows 2.0 and 2.1 continue to recombine changing the cassette expression pattern after every
recombination event. (c) Due to multiple insertions more than three different colors are possible. Three inserts
would give ten distinguishable colors. Here YFP channel would be set to green and CFP channel to blue, giving
a similar principle as the R(ed)G(reen)B(lue) color model of screen displays

3. For reporter lines and overexpression lines, usually 1–2 mg of

tamoxifen per pregnant female is sufficient. For the recombi-
nation of two alleles in mutants, 5–10 mg (varies between con-
ditional knockout lines) is necessary. The efficiency of the
knockout has to be controlled carefully by in situ hybridization
and/or immunohistochemistry.
312 Claudius F. Kratochwil and Filippo M. Rijli

4. In general, stress of the tamoxifen-treated pregnant female

should be kept to a minimum, e.g., by no/reduced change of
bedding and limited handling. If the pups are to be analyzed
postnatally, there may be problems with delivery, since tamoxi-
fen interferes with the female hormone system in a dosage-
dependent manner. To circumvent this problem, the pups can
be recovered by C-section and put in adoption with a foster
mother.

Acknowledgements

Work in F.M.R. laboratory in Friedrich Miescher Institute (Basel,

Switzerland) is supported by the Swiss National Science Foundation
(Sinergia CRSI33_127440), ARSEP, and the Novartis Research
Foundation.

References
1. Thomas KR, Capecchi MR (1987) Site- 10. Siegal ML, Hartl DL (1996) Transgene
directed mutagenesis by gene targeting in Coplacement and high efficiency site-specific
mouse embryo-derived stem cells. Cell recombination with the Cre/loxP system in
51(3):503–512 Drosophila. Genetics 144(2):715–726
2. Gu H, Marth JD, Orban PC, Mossmann H, 11. Werdien D, Peiler G, Ryffel GU (2001) FLP
Rajewsky K (1994) Deletion of a DNA poly- and Cre recombinase function in Xenopus
merase beta gene segment in T cells using cell embryos. Nucleic Acids Res 29(11):E53–3
type-specific gene targeting. Science 12. Dong J, Stuart GW (2004) Transgene manip-
265(5168):103–106 ulation in zebrafish by using recombinases.
3. Lakso M, Sauer B, Mosinger B, Lee EJ, Methods Cell Biol 77:363–379
Manning RW, Yu SH, Mulder KL, Westphal 13. Gilbertson L (2003) Cre-lox recombination:
H (1992) Targeted oncogene activation by Creative tools for plant biotechnology. Trends
site-specific recombination in transgenic Biotechnol 21(12):550–555
mice. Proc Natl Acad Sci USA 14. Guo F, Gopaul DN, van Duyne GD (1997)
89(14):6232–6236 Structure of Cre recombinase complexed with
4. Sauer B (1998) Inducible gene targeting in DNA in a site-specific recombination synapse.
mice using the Cre/lox system. Methods Nature 389(6646):40–46
14(4):381–392 15. Nakano M, Odaka K, Ishimura M, Kondo S,
5. O’Neal KR, Agah R (2007) Conditional tar- Tachikawa N, Chiba J, Kanegae Y, Saito I
geting: inducible deletion by cre recombinase. (2001) Efficient gene activation in cultured
Methods Mol Biol 366:309–320 mammalian cells mediated by FLP
6. Kim JC, Dymecki SM (2009) Genetic fate- recombinase-expressing recombinant adenovi-
mapping approaches: new means to explore rus. Nucleic Acids Res 29(7):E40
the embryonic origins of the cochlear nucleus. 16. Buchholz F, Angrand P-O, Stewart AF (1998)
Methods Mol Biol 493:65–85 Improved properties of FLP recombinase
7. Wilson TJ, Kola I (2001) The LoxP/CRE sys- evolved by cycling mutagenesis. Nat
tem and genome modification. Methods Mol Biotechnol 16(7):657–662
Biol 158:83–94 17. Rodríguez CI, Buchholz F, Galloway J,
8. Brault V, Besson V, Magnol L, Duchon A, Sequerra R, Kasper J, Ayala R, Stewart AF,
Hérault Y (2007) Cre/loxP-mediated chro- Dymecki SM (2000) High-efficiency deleter
mosome engineering of the mouse genome. mice show that FLPe is an alternative to Cre-
Handb Exp Pharmacol 178:29–48 loxP. Nat Genet 25(2):139–140
9. Feil S, Valtcheva N, Feil R (2009) Inducible 18. Feil R, Wagner J, Metzger D, Chambon P
Cre mice. Methods Mol Biol 530:343–363 (1997) Regulation of Cre recombinase activity
 Cre/Lox in Mouse Development 313

by mutated estrogen receptor ligand-binding 32. Luo L (2007) Fly MARCM and mouse
domains. Biochem Biophys Res Commun MADM: genetic methods of labeling and
237(3):752–757 manipulating single neurons. Brain Res Rev
19. Feil R, Brocard J, Mascrez B, LeMeur M, 55(2):220–227
Metzger D, Chambon P (1996) Ligand- 33. Potter CJ, Tasic B, Russler EV, Liang L, Luo L
activated site-specific recombination in mice. (2010) The Q system: a repressible binary sys-
Proc Natl Acad Sci USA 93(20):10887–10890 tem for transgene expression, lineage tracing,
20. Edwards WF, Young DD, Deiters A (2009) and mosaic analysis. Cell 141(3):536–548
Light-activated Cre recombinase as a tool for the 34. Zhang Y, Buchholz F, Muyrers JP, Stewart AF
spatial and temporal control of gene function in (1998) A new logic for DNA engineering
mammalian cells. ACS Chem Biol 4(6):441–445 using recombination in Escherichia coli. Nat
21. Hirrlinger J, Scheller A, Hirrlinger PG et al Genet 20(2):123–128
(2009) Split-cre complementation indicates 35. Sauer B, Henderson N (1988) Site-specific
coincident activity of different genes in vivo. DNA recombination in mammalian cells by
PLoS One 4(1):e4286 the Cre recombinase of bacteriophage P1.
22. Hirrlinger J, Requardt RP, Winkler U, Wilhelm Proc Natl Acad Sci USA 85(14):5166–5170
F, Schulze C, Hirrlinger PG (2009) Split- 36. Santoro SW, Schultz PG (2002) Directed evo-
CreERT2: temporal control of DNA recombi- lution of the site specificity of Cre recombinase.
nation mediated by split-Cre protein fragment Proc Natl Acad Sci USA 99(7):4185–4190
complementation. PLoS One 4(12):e8354 37. Schnütgen F, Doerflinger N, Calléja C,
23. Wang P, Chen T, Sakurai K, Han B-X, He Z, Wendling O, Chambon P, Ghyselinck NB
Feng G, Wang F (2012) Intersectional cre (2003) A directional strategy for monitoring
driver lines generated using split-intein medi- Cre-mediated recombination at the cellular
ated split-cre reconstitution. Sci Rep 2:497 level in the mouse. Nat Rev Neurosci
24. Farago AF, Awatramani RB, Dymecki SM 21(5):562–565
(2006) Assembly of the brainstem cochlear 38. Zheng B, Sage M, Sheppeard EA, Jurecic V,
nuclear complex is revealed by intersectional Bradley A (2000) Engineering mouse chromo-
and subtractive genetic fate maps. Neuron somes with Cre-loxP: range, efficiency, and
50(2):205–218 somatic applications. Mol Cell Biol
25. Zong H, Espinosa JS, Su HH, Muzumdar 20(2):648–655
MD, Luo L (2005) Mosaic analysis with dou- 39. Visel A, Minovitsky S, Dubchak I, Pennacchio
ble markers in mice. Cell 121(3):479–492 LA (2007) VISTA Enhancer Browser–a data-
26. Tasic B, Miyamichi K, Hippenmeyer S, Dani base of tissue-specific human enhancers.
VS, Zeng H, Joo W, Zong H, Chen-Tsai Y, Nucleic Acids Res 35:D88–D92
Luo L (2012) Extensions of MADM (mosaic 40. Engström PG, Fredman D, Lenhard B (2008)
analysis with double markers) in mice. PLoS Ancora: a web resource for exploring highly
One 7(3):e33332 conserved noncoding elements and their asso-
27. Lee T, Luo L (1999) Mosaic analysis with a ciation with developmental regulatory genes.
repressible cell marker for studies of gene func- Genome Biol 9(2):R34
tion in neuronal morphogenesis. Neuron 41. Visel A, Blow MJ, Li Z et al (2009) ChIP-seq
22(3):451–461 accurately predicts tissue-specific activity of
28. Hippenmeyer S, Youn YH, Moon HM, enhancers. Nature 457(7231):854–858
Miyamichi K, Zong H, Wynshaw-Boris A, Luo 42. Yee SP, Rigby PW (1993) The regulation of
L (2010) Genetic mosaic dissection of Lis1 myogenin gene expression during the embry-
and Ndel1 in neuronal migration. Neuron onic development of the mouse. Genes Dev
68(4):695–709 7(7A):1277–1289
29. Livet J, Weissman TA, Kang H, Draft RW, Lu 43. Srinivas S, Watanabe T, Lin CS, William CM,
J, Bennis RA, Sanes JR, Lichtman JW (2007) Tanabe Y, Jessell TM, Costantini F (2001) Cre
Transgenic strategies for combinatorial expres- reporter strains produced by targeted insertion
sion of fluorescent proteins in the nervous sys- of EYFP and ECFP into the ROSA26 locus.
tem. Nature 450(7166):56–62 BMC Dev Biol 1:4
30. Soriano P (1999) Generalized lacZ expression 44. Hope IA, Struhl K (1987) GCN4, a eukary-
with the ROSA26 Cre reporter strain. Nat otic transcriptional activator protein, binds as a
Genet 21(1):70–71 dimer to target DNA. EMBO J
31. Madisen L, Zwingman TA, Sunkin SM et al 6(9):2781–2784
(2010) A robust and high-throughput Cre 45. Espinosa JS, Luo L (2008) Timing neurogen-
reporting and characterization system for the esis and differentiation: insights from quantita-
whole mouse brain. Nat Neurosci tive clonal analyses of cerebellar granule cells.
13(1):133–140 J Neurosci 28(10):2301–2312
 INDEX

A Cryosection .....105, 133–134, 136–138, 207, 218–219,

222, 259, 262, 270–271, 273–275, 278, 281
Agarose .......................... 100, 109, 110, 130, 133, 150, Cy-fluorophore .................... 11, 13, 102, 104, 109, 113
179, 182, 189, 190, 202, 206–212, 214, 224,
231, 271, 277, 304 D
Alexa-fluorophore ........... 62, 65, 66, 68, 101, 102, 104,
Danio rerio .............................................................. 187
107–111, 201
Dechorionation ............................................... 198, 213
Alkaline phosphatase ................. 5, 20, 23, 50, 131, 132,
DiI.................................................... 37–55, 144, 264
135, 176, 221, 228, 270, 274
Drosophila melanogaster .................... 17, 35, 56, 69, 78,
Anatomy .................................................................. 195
93, 94, 96, 113
Anesthetize ......................... 84, 87, 121, 132, 133, 148,
161–163, 165, 211–214, 289, 290 E
Animal keeping .......................................................... 82
Antibody .............................. 5, 19, 39, 58, 74, 83, 100, Egg................ 9, 23, 24, 31, 32, 43, 53, 63, 66, 67, 73,
119, 130, 177, 201, 217, 239, 269 76, 82–85, 88, 89, 105, 143–145, 148–149, 152,
Anti-HRP .................................................... 4, 102, 109 153, 161, 168, 169, 188, 189, 222, 236, 237,
Anti-sense probe .............................. 224–226, 228, 231 240–245, 247, 249, 255–260, 262, 263, 265
Arthropods ................................................................ 97 Electroporation............... 156, 158–170, 212, 254–265,
Atlas.... ................................................................175 285–291, 299, 300
Attb/attP .......................................................... 60, 145 Embryo .......................... 3, 19, 37, 60, 72, 79, 99, 118,
129, 143, 155, 176, 187, 197, 218, 235, 253,
B 275, 285, 295
Ex ovo. ............................................ 254–257, 260–263
Blocking agent ......................................................... 219
Brainbow ............................. 57, 86, 188, 297, 309–311 F
BrdU... .............................................................. 144
Fetal calf serum .......................................... 40, 101, 230
FIJI............................................. 62, 66, 83, 202, 213
C
Filet preparation .................................................. 27–31
Central complex ................................ 99, 100, 108, 110 FISH. See Fluorescent in situ hybridization (FISH)
Cephalopod ............................................. 117, 118, 124 Fixative ....................... 6, 22, 24, 40, 49, 61, 72, 75, 76,
Cerulean ............................................ 58, 59, 62, 65, 66 101, 109, 111, 119, 121, 123, 139, 177, 207,
CFP................................................ 66, 157, 309–311 218, 222, 229, 238, 248
Chicken ............................ 88, 145, 177, 201, 217–233, FLP recombinase ....................... 57, 58, 63, 79, 80, 296
236, 253–265, 291 Fluorescent in situ hybridization (FISH) .......... 175–183
Chimera .......................................................... 235–250 Fluorescent labeling ................. 134, 146, 182, 183, 217
Circuit. ........................3, 5, 38, 57–68, 103, 104, 106, Fluorescine ................................. 21, 40, 104, 134, 150,
107, 187, 188, 197, 309 176, 182, 183, 217
Clones.................. 5, 38, 47–52, 54, 55, 58, 60, 62–64, Flybow .......................................................... 57–68, 86
79–83, 85, 86, 89, 92, 175, 188, 189, 220, 224, Forceps ....................... 8, 12, 14, 23, 28, 30, 40, 47, 49,
226, 230, 276 62–64, 68, 72–75, 82, 84, 87, 90–92, 100, 103,
Cover slips .................................... 8, 10, 11, 13, 15, 16, 105, 120, 122, 123, 130, 133, 135, 147, 157,
23, 27–30, 34, 40, 42–50, 52, 53, 62, 64, 71–76, 163, 198, 199, 203, 204, 210, 211, 213, 222,
100, 109, 120, 123 236, 240, 241, 243, 244, 249, 256, 258, 259,
Cre/Lox.......................................................... 295–312 261, 270, 271, 282, 287, 289, 290

Simon G. Sprecher (ed.), Brain Development: Methods and Protocols, Methods in Molecular Biology, vol. 1082,
DOI 10.1007/978-1-62703-655-9, © Springer Science+Business Media, LLC 2014

315
 BRAIN DEVELOPMENT: METHODS AND PROTOCOLS
316 Index

Formaldehyde ................... 6, 7, 22, 25, 40, 52, 72, 101, mCherry ................................................ 58, 59, 65, 189
122, 123, 130, 131, 138, 203, 229, 231, mCitrine ........................................................ 58, 59, 65
239, 248 Microinjection ................................................. 143–153
Fruit fly ..................................................................... 12 Microinjector ................................... 149, 157, 189, 288
micro RNA (mRNA) ......................... 29, 130, 143, 145,
G 146, 149, 150, 177, 182, 262, 265, 269, 270,
Gal80.. .............................................. 38, 80–83, 85–87 281, 282
Gal4/UAS ..................................... 57, 80, 87, 160, 166 Microscope ................................... 5, 20, 40, 62, 73, 82,
Gastropod ............................................... 117, 118, 124 100, 124, 130, 148, 157, 179, 188, 202, 217,
Gene silencing ................................................. 253–265 236, 262, 269, 288
Genotyping .................... 15, 28, 60, 63, 72, 82, 84, 88, Microscope slides .................... 8, 11, 13, 15, 16, 23, 26,
89, 281, 298, 302–304, 307–308 28, 30, 62, 64, 83, 87, 100, 105, 159, 160, 165,
GFP..... ............... 13, 41, 47, 48, 54, 76, 77, 80, 85, 86, 166, 179, 182, 202, 211, 219, 288
88, 92, 146, 150, 157, 166, 189, 190, 193, 201, Minutien pins .................................................. 8, 23, 28
206–208, 236, 290, 296, 305–309 Mitotic recombination ............................................... 79
Glass capillary .......................................... 152, 157, 261 Molluscs .......................................................... 117–124
Glass needles............................ 148, 149, 152, 256, 258 Monoclonal antibody ......................... 62, 217, 276, 282
Glia..... .......................... 3, 57, 67, 81, 86, 99–112, 269 Morpholino ..................... 143, 146, 147, 150, 155–170
Glutaraldehyde .................................................... 6, 207 Mosaic(s) .................... 5, 67, 79, 80, 86, 144, 145, 188,
Graft........................ 236, 239, 240, 243, 245–247, 250 189, 296, 307
Grasshopper....................................................... 99–112 Mosaic analysis with a repressible cell marker
(MARCM) ........ 5, 38, 67, 79–92, 188, 296, 307
I Mosaic analysis with double marker (MADM) ......... 188,
295–298, 307–309
ImageJ......................................... 66, 83, 104, 202, 213
Mounting medium ....................... 8, 10, 11, 16, 54, 83,
Immunostaining ....................3–16, 62, 84–85, 99–112,
120, 123, 136, 207, 208, 219, 224, 230, 274,
129–140, 207, 239, 248, 275–276, 290
279, 282, 288
Incubator ............................. 63, 67, 82, 151, 165, 166,
Mouse ................................. 11, 13, 62, 72, 76, 88, 102,
190, 199, 203, 222, 237, 240, 241, 243, 245,
192, 201, 207, 217, 219, 239, 253, 254,
256, 258, 260–263, 303
269–282, 285–291, 295–312
In ovo. ............. 237–239, 243–247, 254–260, 262–264
Insects. .......... 20, 37, 99, 102, 199, 238, 239, 246, 247 N
In situ hybridization ........................... 19–34, 129–140,
175–183, 204, 217–233, 236, 239, 240, 247, Neural plate ............................................ 235, 236, 241,
248, 254, 262, 269–282, 311 242, 245–246
In utero ................................................... 285–291, 300 Neural stem cell ....................................... 20, 37, 71–77
In vivo imaging................................................ 165–166 Neuroblast .................. 3, 20, 21, 29, 31, 34, 37–55, 71,
76, 77, 80, 81, 86, 89
K Neuron............... 3, 38, 57, 81, 99, 124, 139, 169, 187,
197, 218, 236, 254, 269, 287
Knock-in ......................................... 285, 300, 307, 308
Normal goat serum .............. 7, 8, 61, 83, 119, 201, 271
Knock-out ...... 143, 254, 285, 295, 296, 298, 300, 302,
305, 307–309, 311 P
L Paraffin section ........................ 133, 134, 136–138, 270
Paraformaldehyde ........................... 6, 8, 61, 66, 83, 84,
LacZ.............................................. 29, 31, 81, 297, 301 101, 109, 119, 130, 139, 177, 198, 218, 222,
Larva... ...................................... 49, 73, 88, 90, 92, 118 227, 229, 270, 288
LexA...................................................................81, 86 PCR. See Polymerase chain reaction (PCR)
Lophotrochozoa ...................................................... 117 Phalloidin ........................................ 118, 119, 121, 124
LoxP.... .................................... 296–302, 307, 309, 310 Phenotype .................... 5, 81, 145, 146, 165–166, 168,
262, 265, 295, 301, 302
M
Phi-C31................................................................... 145
MADM. See Mosaic analysis with double marker Pipette.......................8, 12, 13, 15, 16, 28, 33, 52, 62,
(MADM) 63, 67, 72, 74–76, 83, 84, 87, 91, 92, 100–102,
MARCM. See Mosaic analysis with a repressible cell 105, 109, 133, 139, 147, 149, 153, 157,
marker (MARCM) 159, 163, 165, 167, 200, 201, 209–212, 236,
 BRAIN DEVELOPMENT: METHODS AND PROTOCOLS
Index
317
238–240, 242, 243, 245–247, 249, 271, 272, SPIM. See Selective Plane Illumination Microscopy
275, 281 (SPIM)
Polyclonal antibody ................................................. 217 Split-Cre .......................................................... 297, 306
Polymerase chain reaction (PCR) .............. 87, 224, 262, Stereo microscope.............................. 23, 25, 26, 28, 34
298, 302–304
Proteinase K (PK) ................... 130, 135, 140, 177, 178, T
180, 201, 205, 221, 227, 231, 232, 273, 278, Tadpole .......................................... 132, 139, 156, 157,
282, 298, 303 159–167, 170
Tamoxifen ...................................... 296, 298, 305, 307,
Q
311, 312
Quail ................................................................ 235–250 Tissue dissection ...................................................... 130
Tissue fixation............................................................ 38
R Transgenesis ............................... 58, 59, 63, 65, 66, 68,
Restriction enzyme ................. 144–145, 220, 226, 231, 85–87, 187, 188
277, 298 Transplantation .................................. 79, 144, 235–250
RNA... ................ 19–34, 131, 132, 134, 136, 139, 147, Tyramide signal amplification (TSA) .............. 21, 23, 26,
153, 155, 156, 176–178, 180, 182, 218, 220, 33, 34, 50, 175–183
224, 226, 227, 230–232, 254, 269–282
V
interference ........................................................ 156
probe .................................. 20, 132, 134, 139, 176, Viratome ................................. 118–123, 248, 288, 290
177, 180, 182, 218, 224, 226, 227, 231, 232,
270–273, 276–277, 280 X
RNAse .................... 131, 149, 150, 153, 155, 220, 221, Xenopus ...........................129–140, 143–153, 155–170
226–229, 231, 232, 273, 274, 277, 278,
281, 282 Y
S YFP.......................................................... 157, 309–311

Schistocerca gregaria ........................................... 99–112 Z

Sectioning ...................... 105, 122–124, 129, 133, 134,
222, 239, 248 Zebrafish ..........146, 175–183, 187–194, 197–213, 296