Micro Glia

Methods in
Molecular Biology 1041
Bertrand Joseph
José Luis Venero Editors
Microglia
Methods and Protocols
METHODS IN M O L E C U L A R B I O LO G Y ™
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
Microglia
Methods and Protocols
Edited by
Bertrand Joseph
Department of Oncology-Pathology, Karolinska Institutet, Cancer Centrum Karolinska, Stockholm, Sweden
José Luis Venero

Departamento de Bioquímica y Biología Molecular, Facultad de Farmacia, Universidad de Sevilla,
Sevilla, Spain; Instituto de Biomedicina de Sevilla, Hospital Universitario, Sevilla, Spain
Editors
Bertrand Joseph José Luis Venero
Department of Oncology-Pathology Departamento de Bioquímica
Karolinska Institutet y Biología Molecular
Stockholm, Sweden Facultad de Farmacia
Universidad de Sevilla
Sevilla, Spain
Instituto de Biomedicina de Sevilla
Hospital Universitario,Sevilla, Spain
ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-62703-519-4 ISBN 978-1-62703-520-0 (eBook)
DOI 10.1007/978-1-62703-520-0
Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2013940920
© Springer Science+Business Media New York 2013

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
Microglia are resident myeloid and immune effector cells of the central nervous system. Pío
del Río-Hortega, a disciple of Santiago Ramón y Cajal, named these cells “microglia” in
1919 [1]. The cells had been previously described in the 1880s by Franz Nissl and William
Ford Robertson, but Pío del Río-Hortega conducted the first systematic studies on this cell
type and remarkably many of his observations are still valid [2]. He is accordingly consid-
ered as the “Father of Microglia.”
Publications/Period of time
20000
18000
16000
14000
12000
10000
8000
6000
4000
2000
41 115 291 3203 8535 4698
0
1920-69 1970-79 1980-89 1990-99 2000-2009 2010-2012
Considering this historical note, it becomes evident that for a long period of time little
improvement was made in our knowledge of microglia. Indeed, this field of research
remained rather confidential for more than half a century. However, ever since the 1990s
there is a clear regain of interest for microglia as evidenced by the constant increase at an
almost exponential rate in the number of publications. In fact, there are none less than
8,000 publications including the term microglia during the past decade and if one consid-
ers the first quarter of the current decade one may expect over 18,000 publications on the
subject at the end of this decade.
The reasons for this recent and considerable interest for microglia are certainly multiple.
First and foremost, key discoveries concerning the different biological functions of microglia
in health and disease have attracted scientists from various fields. Indeed, microglia act as
sentinels of infection and injury, and participate in both innate and adaptive immune
v
vi Preface
responses in the central nervous system. Microglia can also be dysregulated in the context
of neurodegenerative disease and cancer, and thereby contribute to disease severity.
Microglia were as well as recently shown to play roles in normal brain development, adult
neuronal plasticity, and circuit function. A second factor which has undoubtedly contrib-
uted to the expansion of this field of research is the development of model systems and
methods to investigate microglia biological functions both in vitro and in vivo.
In light of the interest for microglia, the aim of Microglia: Methods and Protocols is to
provide a selection of the key cellular, molecular, and biochemical techniques that are used
in studying the many and varied functions of this fascinating cell. The chapters of Microglia:
Methods and Protocols are written by experts who have hands-on experience with the par-
ticular method. They provide a comprehensive step-by-step guide to many techniques for
microglia cell culture, for studying microglia activation and functions, as well as their inter-
action with other cell types, both in vitro and in vivo. These methods and protocols should
provide a useful resource for cell biologists, molecular biologists, immunologists, oncolo-
gists, and neuroscientists.
This book is dedicated to the memory of Dr. Laia Acarin (1970–2011), Professor at the
Medical School and member of the Institute of Neuroscience at the Autonomous University
of Barcelona, Spain. She was a brilliant scientist, teacher, and mentor and made prominent
contributions to the field of microglia.
We wish to thank all the authors for their excellent contributions and Prof. John M.
Walker for sound advice and assistance throughout the editorial process. It is our sincerest
hope that you will find Microglia: Methods and Protocols a useful lab companion for years
to come.
Stockholm, Sweden Bertrand Joseph

Sevilla, Spain José Luis Venero
1. Del Río-Hortega P (1919)El ‘tercer elemento’ de los centros nerviosos. I. La microglia en estado nor-
mal. II. Intervencíon de la microglia en los procesos patológicos. III. Naturaleza probable de la
microglia. Bol de la Soc esp de biol 9:69–129
2. Del Río-Hortega P (1932) Microglia. In: Cytology & cellular pathology of the nervous system. Paul
B. Hoeber, Inc., New York, p 483–534
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
PART I OVERVIEW OF MICROGLIA BIOLOGY
1 A Brief Overview of Multitalented Microglia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Bertrand Joseph and José Luis Venero
PART II ISOLATION AND CULTURE OF MICROGLIA
2 Cell Culturing of Human and Murine Microglia Cell Lines . . . . . . . . . . . . . . . . . 11

Johanna Rodhe
3 Microglia Isolation from Adult Mouse Brain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Jae-Kyung Lee and Malú G. Tansey
4 Preparation of Primary Microglia Cultures from Postnatal Mouse
and Rat Brains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Tomas Deierborg
5 Isolation of Murine Postnatal Brain Microglia for Phenotypic
Characterization Using Magnetic Cell Separation Technology . . . . . . . . . . . . . . . 33
Ashley S. Harms and Malú G. Tansey
6 Isolation and Culture of Adult Human Microglia Within Mixed
Glial Cultures for Functional Experimentation and High-Content Analysis . . . . . 41
Amy M. Smith, Hannah M. Gibbons, Claire Lill,
Richard L.M. Faull, and Mike Dragunow
PART III DEPLETION AND TRANSDUCTION OF MICROGLIA
7 Depletion of Microglia from Primary Cellular Cultures . . . . . . . . . . . . . . . . . . . . 55

Lorena Pont-Lezica, Sabrina Colasse, and Alain Bessis
8 Lentiviral Transduction of Cultured Microglia . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Takahiro Masuda, Makoto Tsuda, Hidetoshi Tozaki-Saitoh,
and Kazuhide Inoue
PART IV ANALYSIS OF MICROGLIAL CYTOKINE PRODUCTION
9 Microglial Activation: Measurement of Cytokines by Flow Cytometry. . . . . . . . . 71

Deepak Kumar Kaushik and Anirban Basu
10 In Situ Hybridization of Cytokine mRNA Using Alkaline
Phosphatase-Labelled Oligodeoxynucleotide Probes. . . . . . . . . . . . . . . . . . . . . . . 83
Bettina Clausen, Christina Fenger, and Bente Finsen
vii
viii Contents
11 Use of Meso-Scale Discovery™ to Examine Cytokine Content

in Microglia Cell Supernatant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Miguel A. Burguillos
PART V ANALYSIS OF MICROGLIA ACTIVATION
12 Analysis of Microglial Production of Reactive Oxygen and Nitrogen Species . . . . 103

Urte Neniskyte and Guy C. Brown
13 Quantification of Active Caspase-3 and Active Caspase-8 in Microglia Cells. . . . 113
Edel Kavanagh
14 Quantification of Microglial Phagocytosis by a Flow
Cytometer-Based Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Refik Pul, Kandiyil Prajeeth Chittappen, and Martin Stangel
15 Quantification of Microglial Proliferation and Apoptosis by Flow Cytometry . . . . 129
Alicia A. Babcock, Martin Wirenfeldt, and Bente Finsen
16 Fluorescence Imaging of Intracellular Ca2+, Na+, and H+ in Cultured Microglia. . . . 147
Tom Schilling and Claudia Eder
17 Patch Clamp Protocols to Study Ion Channel Activity in Microglia . . . . . . . . . . 163
PART VI ANALYSIS OF MICROGLIA POLARIZATION
18 Studying M1 and M2 States in Adult Microglia . . . . . . . . . . . . . . . . . . . . . . . . . 185

Sadanand M. Gaikwad and Michael T. Heneka
19 Isolating, Culturing, and Polarizing Primary Human
Adult and Fetal Microglia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Bryce A. Durafourt, Craig S. Moore, Manon Blain, and Jack P. Antel
PART VII CO-CULTURE SYSTEMS TO ANALYSIS MICROGLIA INTERACTIONS

WITH OTHER CELL TYPES
20 Understanding Microglia–Neuron Cross Talk: Relevance

of the Microglia–Neuron Cocultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Fernando G. Correa, Miriam Hernangómez, and Carmen Guaza
21 Preparation of Rodent Primary Cultures for Neuron–Glia, Mixed
Glia, Enriched Microglia, and Reconstituted Cultures with Microglia . . . . . . . . 231
Shih-Heng Chen, Esteban A. Oyarzabal, and Jau-Shyong Hong
PART VIII ANALYSIS OF MICROGLIA FUNCTIONS IN VIVO
22 Microglia Detection by Enzymatic Histochemistry . . . . . . . . . . . . . . . . . . . . . . 243

Beatriz Almolda, Berta González, and Bernardo Castellano
23 Tomato Lectin Histochemistry for Microglial Visualization . . . . . . . . . . . . . . . . 261
Nàdia Villacampa, Beatriz Almolda, Berta González,
and Bernardo Castellano
Contents ix
24 Immunohistochemical Detection of Microglia . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Rocío M. de Pablos, Ana M. Espinosa-Oliva, and Antonio J. Herrera
25 Intrathecal Infusion of Microglia Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
and Kazuhide Inoue
26 Intracranial Injection of LPS in Rat as Animal Model
of Neuroinflammation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Ana M. Espinosa-Oliva, Rocío M. de Pablos, and Antonio J. Herrera
27 Analyses of Microglia Effector Function Using CX3CR1-GFP
Knock-In Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Jenny A. Garcia, Sandra M. Cardona, and Astrid E. Cardona
28 In Vivo Two-Photon Microscopy of Microglia. . . . . . . . . . . . . . . . . . . . . . . . . . 319
Satoru Kondo and Shigeo Okabe
29 Use of Confocal Microscopy in the Study of Microglia
in a Brain Metastasis Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Manuel Sarmiento
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Contributors
BEATRIZ ALMOLDA • Universitat Autònoma de Barcelona, Barcelona, Spain

JACK P. ANTEL • Neuroimmunology Unit, Montréal Neurological Institute, McGill
University, Montreal, Canada
ALICIA A. BABCOCK • Institute of Molecular Medicine, University of Southern Denmark,
Odense, Denmark
ANIRBAN BASU • National Brain Research Centre, Manesar, India
ALAIN BESSIS • Ecole Normale Supérieure, Paris, France
MANON BLAIN • McGill University, Montreal, Canada
GUY C. BROWN • University of Cambridge, Cambridge, UK
MIGUEL A. BURGUILLOS • Karolinska Institutet, Stockholm, Sweden
ASTRID E. CARDONA • Department of Biology and South Texas Center for Emerging
Infectious Diseases, The University of Texas at San Antonio, San Antonio, TX, USA
SANDRA M. CARDONA • The University of Texas at San Antonio, San Antonio, TX, USA
BERNARDO CASTELLANO • Universitat Autònoma de Barcelona, Bellaterra, Spain
SHIH-HENG CHEN • National Institute of Environmental Health Sciences, Research
Triangle Park, NC, USA
BETTINA CLAUSEN • University of Southern Denmark, Odense, Denmark
SABRINA COLASSE • Ecole Normale Supérieure, Paris, France
FERNANDO G. CORREA • Instituto Cajal, Consejo Superior de Investigaciones Científicas,
Madrid, Spain; Universidad de Buenos Aires-Consejo Nacional de Investigaciones
Científicas y Técnicas, Buenos Aires, Argentina
ROCÍO M. DE PABLOS • Universidad de Sevilla, Sevilla, Spain; Instituto de Biomedicina
de Sevilla, Sevilla, Spain
TOMAS DEIERBORG • Lund University, Lund, Sweden
MIKE DRAGUNOW • The University of Auckland, Auckland, New Zealand
BRYCE A. DURAFOURT • McGill University, Montreal, Canada
CLAUDIA EDER • St. Georges’, University of London, London, UK
ANA M. ESPINOSA-OLIVA • Universidad de Sevilla, Sevilla, Spain; Instituto de Biomedicina
de Sevilla, Sevilla, Spain
RICHARD L.M. FAULL • The University of Auckland, Auckland, New Zealand
CHRISTINA FENGER • University of Southern Denmark, Odense, Denmark
BENTE FINSEN • Neurobiology Research, Institute of Molecular Medicine, University of
Southern Denmark, Odense, Denmark
SADANAND M. GAIKWAD • University of Bonn, Bonn, Germany
JENNY A. GARCIA • The University of Texas at San Antonio, San Antonio, TX, USA
HANNAH M. GIBBONS • The University of Auckland, Auckland, New Zealand
BERTA GONZÁLEZ • Universitat Autònoma de Barcelona, Barcelona, Spain
CARMEN GUAZA • Instituto Cajal, Consejo Superior de Investigaciones Científicas,
Madrid, Spain
ASHLEY S. HARMS • Emory University School of Medicine, Atlanta, GA, USA
xi
xii Contributors
MICHAEL T. HENEKA • Clinical Neuroscience Unit, Department of Neurology,

University of Bonn, Bonn, Germany
MIRIAM HERNANGÓMEZ • Instituto Cajal, Consejo Superior de Investigaciones Científicas,
Madrid, Spain
ANTONIO J. HERRERA • Facultad de Farmacia, Departamento de Bioquímica y Biología
Molecular, Universidad de Sevilla, Sevilla, Spain; Instituto de Biomedicina de Sevilla,
Sevilla, Spain
JAU-SHYONG HONG • National Institute of Environmental Health Sciences,
Research Triangle Park, NC, USA
KAZUHIDE INOUE • Kyushu University, Fukuoka, Japan
BERTRAND JOSEPH • Department of Oncology-Pathology, Karolinska Institutet,
Cancer Centrum Karolinska, Stockholm, Sweden
DEEPAK KUMAR KAUSHIK • Department of cellular and Molecular Neurosciences,
National Brain Research Centre, Manesar, India
EDEL KAVANAGH • Karolinska Institutet, Stockholm, Sweden
SATORU KONDO • Department of Molecular Physiology, Kyushu University, Fukuoka, Japan
JAE-KYUNG LEE • Emory University School of Medicine, Atlanta, GA, USA
CLAIRE LILL • Department of Anatomy with Radiology, The University of Auckland,
Auckland, New Zealand
TAKAHIRO MASUDA • Kyushu University, Fukuoda, Japan
CRAIG S. MOORE • McGill University, Montreal, QC, Canada
URTE NENISKYTE • University of Cambridge, Cambridge, UK
SHIGEO OKABE • University of Tokyo, Tokyo, Japan
ESTEBAN A. OYARZABAL • National Institute of Environmental Health Sciences,
Research Triangle Park, NC, USA
LORENA PONT-LEZICA • Ecole Normale Supérieure, Paris, France
KANDIYIL PRAJEETH CHITTAPPEN • Hannover Medical School, Hannover, Germany
REFIK PUL • Hannover Medical School, Hannover, Germany
JOHANNA RODHE • Karolinska Institutet, Stockholm, Sweden
MANUEL SARMIENTO • Department of Oncology, CR-UK/MRC Gray Institute for
Radiation Oncology and Biology, University of Oxford, Oxford, UK
TOM SCHILLING • St. George’s, University of London, London, UK
AMY M. SMITH • The University of Auckland, Auckland, New Zealand
MARTIN STANGEL • Hannover Medical School, Hannover, Germany
MALÚ G. TANSEY • Emory University School of Medicine, Atlanta, GA, USA
HIDETOSHI TOZAKI-SAITOH • Kyushu University, Fukuoda, Japan
MAKOTO TSUDA • Kyushu University, Fukuoda, Japan
JOSÉ LUIS VENERO • Departamento de Bioquímica y Biología Molecular, Facultad de
Farmacia, Universidad de Sevilla, Sevilla, Spain; Instituto de Biomedicina de Sevilla,
Hospital Universitario, Sevilla, Spain
NÀDIA VILLACAMPA • Universitat Autònoma de Barcelona, Barcelona, Spain
MARTIN WIRENFELDT • University of Southern Denmark, Odense, Denmark
Part I
Overview of Microglia Biology

Chapter 1
A Brief Overview of Multitalented Microglia

Bertrand Joseph and José Luis Venero
Abstract
Microglia are the resident immune cells of the central nervous system, and accumulating data demonstrates
a vast array of tasks in the healthy and injured brain. Microglia participate in both innate and adaptive
immune responses. These cells contribute to the brain homeostasis, including the regulation of cell death,
synapse elimination, neurogenesis, and neuronal surveillance. However, microglia can also become acti-
vated and/or deregulated in the context of neurodegenerative diseases, brain injuries, and cancer and thereby
contribute to disease severity. As a consequence of these developments, microglia have attracted substantial
attention on themselves.
Key words Immune cells, Central nervous system, Microglia, Brain homeostasis, Cell death
1 Overview of the Multitalented Microglia
Microglia are key glial cell elements of the central nervous system
(CNS) and are considered the major immunocompetent cells in
the brain. Consequently, microglia trigger signaling cascades well
established in the immune system involving chemokines and cyto-
kines and their receptor systems. In keeping with this view, microg-
lia share many features of monocytes, and they are constantly
scavenging for damaged neurons, plaques, and infectious agents
using their multiple surface receptors [1, 2].
Microglia were first described by Rio-Hortega, a disciple of
Ramón y Cajal, in 1919 in a visionary article [3] in which he antici-
pated key features of microglia cells. For instance, Rio-Hortega
deduced that (1) microglia enter the brain during early develop-
ment; (2) they acquire a branched, ramified morphology in the
mature brain; (3) they undergo a morphological transformation
into an amoeboid-shape morphology in response to different types
of injury; and (4) they have the capacity to migrate, proliferate, and
phagocytose. As Rio-Hortega predicted, microglia populate the
mammalian CNS in early embryonic development. By adulthood,
Bertrand Joseph and José Luis Venero (eds.), Microglia: Methods and Protocols, Methods in Molecular Biology, vol. 1041,
DOI 10.1007/978-1-62703-520-0_1, © Springer Science+Business Media New York 2013
3
4 Bertrand Joseph and José Luis Venero
microglia are found in all regions of the brain and spinal cord and
comprise 10–15 % of all CNS cells. In addition to being sensitive
to changes in their environment, each microglial cell also regularly
physically surveys its domain. Microglia are believed to derive from
monocytes that invade the developing CNS and persist over the
adult life as resident macrophages [4, 5]. An elegant study using
fate-mapping analysis confirmed that these glial cells derive from
primitive myeloid progenitors that arise before embryonic day 8
and that postnatal hematopoietic progenitors do not contribute to
microglia homeostasis in the adult brain [6].
Recent studies using in vivo two-photon imaging in undam-
aged normal brain reveal that microglial cells are highly active in
their presumed resting state, continually surveying their microen-
vironment with extremely motile processes and protrusions [7, 8].
In the healthy brain, microglia habitually interact with neuronal
and nonneuronal elements, both structurally and functionally,
including phagocytosis of synaptic structures during postnatal
development, phagocytosis of newborn neurons during adult neu-
rogenesis, and active remodeling of the perisynaptic environment
and release of soluble factors in the mature and aging brain [9].
Thus, microglial cells fulfil an astonishing variety of tasks within
the CNS; they are multitalented cells demanding multidisciplinary
approaches and methodologies to get further insights into their
multiple tasks.
When microglia are “activated,” they take on an amoeboid
shape with phagocytic activity (again anticipated by del Rio-
Hortega [3]), and they increase their gene expression leading to
the production of numerous potentially neurotoxic mediators, such
as proteases and proinflammatory cytokines. Another group of
potentially neurotoxic mediators are reactive oxygen species (ROS)
and NO. ROS, including superoxide, hydroxyl radicals, and hydro-
gen peroxide, are highly reactive molecules, involved in various signal
transduction cascades [10]. ROS form little threat in low concentra-
tions as cells possess various defense and repair mechanisms for minor
oxidative stress. However, during inflammation, they are released
in high concentrations by the oxidative burst of activated microg-
lia. High ROS concentrations overrule cellular defense mecha-
nisms and cause oxidative damage to proteins, lipids, and nucleic
acids with subsequent risk for neuronal populations.
Activated microglia also release various neurotrophic factors
and cytokines that can modify neuronal circuits [11, 12]. These
mediators are important in the normal functions of microglia, and
their production is usually decreased once their task is complete.
During recent years, researchers have tended to differentiate
between acute inflammation and chronic inflammation and their
effects on neurons. Neuroinflammation is a complex innate
response of neural tissue against the harmful effects of diverse stim-
uli within the central nervous system (CNS). Infections, trauma,
A Brief Overview of the Multitalented Microglia 5
stroke, toxins, and other stimuli are capable of producing an acute

neuroinflammatory response within the CNS with the activation of
resident immune cells (microglia, astrocytes) and the release of
inflammatory mediators such as cytokines and chemokines.
Neurodegenerative disorders, including Alzheimer’s disease (AD),
Parkinson’s disease (PD), Huntington’s disease (HD), multiple
sclerosis (MS), and amyotrophic lateral sclerosis (ALS), are associ-
ated with chronic neuroinflammation [10]. Chronic neuroinflam-
mation can be seen as a long-lasting and often self-perpetuating
neuroinflammatory response. Chronic neuroinflammation
involves the sustained activation of microglia, and consequent
release of proinflammatory mediators, but also an increased oxida-
tive and nitrosative stress which are both detrimental for the neu-
ronal cell population. Increasing evidences indicate that microglia
can become chronically activated in response to dying/damaged
neurons, fuelling a self-renewing cycle of microglial activation,
causing a self-propelling cycle of neuron death. This repeating
cycle of neurotoxic microglial activation in response to neuron
injury is commonly referred to as reactive microgliosis, which is a
proposed mechanism of chronic neuronal loss across diverse neu-
rodegenerative diseases [13]. However, not all immune responses
in the CNS are detrimental, and in many cases, they actually aid
repair and regeneration [14]. For example, microglia clear debris
after myelin damage and when this is impeded, delayed regenera-
tion occurs [15]. Immune activation is also crucial to limit neuro-
tropic viral infections, and immune cells remove necrotic cells
following ischemia [16]. Consequently, when studying the roles of
microglia in different scenarios, it is very important to define the
paradigm of microglia/macrophage activation [17]. Microglia and
macrophages can be classified into at least two subsets with distinct
molecular phenotypes and effector functions depending on the
activation pathway. The “classically activated” proinflammatory
M1 macrophages, activated by LPS and by the proinflammatory
cytokine IFN-γ, produce high amounts of oxidative metabolites,
proteases, and proinflammatory cytokines. They play a central role
in the host defense against pathogens and tumor cells, but they can
also damage healthy cells such as neurons and glial cells. In con-
trast, M2 macrophages or “alternatively activated,” induced by
IL-4 and IL-13, exhibit anti-inflammatory properties and promote
tissue remodeling/repair and angiogenesis [17, 18]. Microglial
cells are attracted toward glioma in large numbers—glioma tissue
consists of as much as 30 % microglial cells—and microglia density
in gliomas positively correlates with malignancy, invasiveness, and
grading of tumors. Soluble factors released from glioma stimulate
microglial Toll-like receptors, resulting in microglia activation and
infiltration of the tumor. Thereafter, tumor cells shut down their
immune properties and stimulate the microglia to release tumor
growth-promoting factors. In fact, microglia release many factors,
including extracellular matrix proteases and cytokines, which

directly or indirectly may influence tumor invasiveness and growth
[19–21]. As further evidence of their essential role in glioma biol-
ogy, their removal has been shown in vivo to inhibit glioma
invasiveness.
Another major flow of microglia research relies on knowledge
of the impressive array of multiple receptors located on these cells.
Among others, they include α-amino-3-hydroxy-5-methyl-4-
isoxazolepropionic acid (AMPA) receptors to respond to gluta-
mate [22], P2 purinoceptors to respond to ATP and other
nucleotides [23, 24], cytokine receptors to respond to tumor
necrosis factor (TNF)-α and interleukin (IL)-1β [24], and recep-
tors for the major histocompatibility complex (MHC) class anti-
gens and chemokines [25–27]. Activation of these receptors
induces signaling cascades in microglia that have been shown to be
involved in microglia chemotaxis, activation, and phagocytosis.
Furthermore, microglial P2XRs and P2Yrs receptors play a key role
in pain signaling in the spinal cord under pathological conditions,
such as following peripheral nerve injury (neuropathic pain) [28].
Further, microglia express pattern recognition receptors (PRRs),
which act as sensor for inflammatory triggers. The inflammatory
response can be triggered by the recognition of several microbial
signals, through conserved molecular pattern known as pathogen-
associated molecular patterns (PAMPs) [29] or intracellular mole-
cules, normally found in the cytosol or in the nucleus and are
released from the cell after tissue damage known as danger-
associated molecular patterns (DAMPs) [30]. PAMPs and DAMPs
are detected by members of Toll-like receptors (TLRs) family of
PRR [31]. The nucleotide oligomerization domain (NOD)-like
receptor (NLR) family is a large family of cytosolic PRRs, special-
ized in the recognition of intracellular signals [32]. Once activated
by the inflammatory stimuli, the PRRs (TLRs and NLRs) partici-
pate in the formation of multiprotein oligomers known as inflam-
masomes, which orchestrates the activation of caspase-1, which in
turn contributes to cytokine maturation and the inflammatory
responses [33]. We recently described a novel and unexpected role
for so-called apoptotic caspases in the activation of microglia and
associated neurotoxicity [34]. Activation of TLRs with proinflam-
mogens leads to the orderly activation of caspase-8 and caspase-3.
Caspase-3 activates the NF-κB pathways through processing and
activation of protein kinase C-δ. This finding may be relevant in
the onset of chronic neurodegenerative diseases, mainly PD and
AD [35]. In fact, several intracellular signaling pathways contrib-
ute to microglial activation. Activation of most TLRs results in the
recruitment of the adaptor protein myeloid differentiation factor-
88 (MyD88), serine/threonine kinase IL-1R-associated kinase
(IRAK), tumor necrosis factor (TNF) receptor-associated factor
(TRAF) adaptor protein TRAF6, NF-kB-inducing kinase (NIK),
A Brief Overview of the Multitalented Microglia 7
and IkB kinase (IKK), thereby leading to the nuclear translocation

of NF-kB [29]. However, other alternative adaptor molecules
including TIR-associated protein (TIRAP)/MyD88-adaptor-like
(MAL), TIR-domain-containing adaptor protein-inducing IFN-β
(TRIF)/TIR-domain-containing molecule 1 (TICAM1), and
TRIF-related adaptor molecule (TRAM) exist, which transduce
signals from TLRs via an MyD88-independent pathway [29]. In
canonical interferon (IFN)-γ-Janus kinase (Jak)-signal transducer
and activator of transcription 1 (STAT1) signaling (reviewed in ref.
[36]), the binding of IFN-γ to its receptor leads to activation of
receptor-associated Jak1 and Jak2 and phosphorylation of a recep-
tor tyrosine residue working as a docking site for STAT1. STAT1 is
then activated by phosphorylation and translocates to the nucleus
where it stimulates transcription of STAT1 target genes. Another
way for IFN-γ to activate macrophages is by enhancing macro-
phage responsiveness to other inflammatory stimuli, such as TLR
ligands and TNF. This mechanism is known by the name of “prim-
ing.” By this “priming” of IFN-γ, the TLR-dependent expression
of several proinflammatory cytokines and chemokines is greatly
augmented [37].
The clarification of the microglia biological functions in the
healthy brain, the discovery of their significant contribution to
various CNS diseases, ranging from neurodegenerative disorders,
stroke, brain injuries, to cancer, as well as the advances in the
understanding of the signaling pathways controlling their activa-
tion and functions have collectively contributed to the literal
“explosion of interest” for these cells which used to be rather
ignored. As a result, methods, protocols, and tools have been
developed to study the multitalented microglia.
References
1. Perry VH, Gordon S (1988) Macrophages and 6. Ginhoux F, Greter M, Leboeuf M, Nandi S,
microglia in the nervous system. Trends See P, Gokhan S, Mehler MF, Conway SJ, Ng
Neurosci 11(6):273–277 LG, Stanley ER, Samokhvalov IM, Merad M
2. Kreutzberg GW (1996) Microglia: a sensor for (2010) Fate mapping analysis reveals that adult
pathological events in the CNS. Trends microglia derive from primitive macrophages.
Neurosci 19(8):312–318 Science 330(6005):841–845
3. Rio-Hortega Pd (1919) El "tercer elemento" 7. Davalos D, Grutzendler J, Yang G, Kim JV,
de los centros nerviosos. Poder fagocitario y Zuo Y, Jung S, Littman DR, Dustin ML, Gan
movilidad de la microglia. Bol de la Soc espa- WB (2005) ATP mediates rapid microglial
nola de Biol. Ano ix:154–166 response to local brain injury in vivo. Nat
4. Alliot F, Godin I, Pessac B (1999) Microglia Neurosci 8(6):752–758
derive from progenitors, originating from 8. Nimmerjahn A, Kirchhoff F, Helmchen F
the yolk sac, and which proliferate in the (2005) Resting microglial cells are highly
brain. Brain Res Dev Brain Res 117(2): dynamic surveillants of brain parenchyma in
145–152 vivo. Science 308(5726):1314–1318
5. Gomes-Leal W (2012) Microglial physiopa- 9. Tremblay ME, Stevens B, Sierra A, Wake H,
thology: how to explain the dual role of Bessis A, Nimmerjahn A (2011) The role of
microglia after acute neural disorders? Brain microglia in the healthy brain. J Neurosci
Behav 2(3):345–356 31(45):16064–16069
10. Venero JL, Burguillos MA, Brundin P, Joseph Jacobson KA, Kohsaka S, Inoue K (2007)
B (2011) The executioners sing a new song: UDP acting at P2Y6 receptors is a mediator of
killer caspases activate microglia. Cell Death microglial phagocytosis. Nature
Differ 18(11):1679–1691 446(7139):1091–1095
11. Merrill JE (1992) Tumor necrosis factor alpha, 25. Finsen BR, Jorgensen MB, Diemer NH,
interleukin 1 and related cytokines in brain Zimmer J (1993) Microglial MHC antigen
development: normal and pathological. Dev expression after ischemic and kainic acid
Neurosci 14(1):1–10 lesions of the adult rat hippocampus. Glia
12. Biber K, Neumann H, Inoue K, Boddeke HW 7(1):41–49
(2007) Neuronal ‘On’ and ‘Off’ signals 26. Bessis A, Bechade C, Bernard D, Roumier A
control microglia. Trends Neurosci 30(11): (2007) Microglial control of neuronal death
596–602 and synaptic properties. Glia 55(3):233–238
13. Block ML, Zecca L, Hong JS (2007) 27. Cullheim S, Thams S (2007) The microglial
Microglia-mediated neurotoxicity: uncovering networks of the brain and their role in neuro-
the molecular mechanisms. Nat Rev Neurosci nal network plasticity after lesion. Brain Res
8(1):57–69 Rev 55(1):89–96
14. Aguzzi A, Barres BA, Bennett ML (2013) 28. Tsuda M, Tozaki-Saitoh H, Inoue K (2012)
Microglia: scapegoat, saboteur, or something Purinergic system, microglia and neuropathic
else? Science 339(6116):156–161 pain. Curr Opin Pharmacol 12(1):74–79
15. Neumann H, Kotter MR, Franklin RJ (2009) 29. Akira S, Uematsu S, Takeuchi O (2006)
Debris clearance by microglia: an essential link Pathogen recognition and innate immunity.
between degeneration and regeneration. Brain Cell 124(4):783–801
132(Pt 2):288–295 30. Beg AA (2002) Endogenous ligands of Toll-
16. Amor S, Puentes F, Baker D, van der Valk P like receptors: implications for regulating
(2010) Inflammation in neurodegenerative inflammatory and immune responses. Trends
diseases. Immunology 129(2):154–169 Immunol 23(11):509–512
17. Gordon S, Martinez FO (2010) Alternative 31. Akira S, Takeda K (2004) Functions of toll-like
activation of macrophages: mechanism and receptors: lessons from KO mice. C R Biol
functions. Immunity 32(5):593–604 327(6):581–589
18. Czeh M, Gressens P, Kaindl AM (2011) The 32. Strober W, Murray PJ, Kitani A, Watanabe T
yin and yang of microglia. Dev Neurosci (2006) Signalling pathways and molecular
33(3–4):199–209 interactions of NOD1 and NOD2. Nat Rev
19. Badie B, Schartner J (2001) Role of microglia Immunol 6(1):9–20
in glioma biology. Microsc Res Tech 54(2): 33. Yazdi AS, Guarda G, D’Ombrain MC, Drexler
106–113 SK (2010) Inflammatory caspases in innate
20. Markovic DS, Vinnakota K, Chirasani S, immunity and inflammation. J Innate Immun
Synowitz M, Raguet H, Stock K, Sliwa M, 2(3):228–237
Lehmann S, Kalin R, van Rooijen N, Holmbeck 34. Venero JL, Burguillos MA, Joseph B (2013)
K, Heppner FL, Kiwit J, Matyash V, Lehnardt Caspases playing in the field of neuroinflam-
S, Kaminska B et al (2009) Gliomas induce mation: old and new players. Dev Neurosci.
and exploit microglial MT1-MMP expression doi:10.1159/000346155
for tumor expansion. Proc Natl Acad Sci USA 35. Burguillos MA, Deierborg T, Kavanagh E,
106(30):12530–12535 Persson A, Hajji N, Garcia-Quintanilla A,
21. Saijo K, Glass CK (2011) Microglial cell origin Cano J, Brundin P, Englund E, Venero JL,
and phenotypes in health and disease. Nat Rev Joseph B (2011) Caspase signalling controls
Immunol 11(11):775–787 microglia activation and neurotoxicity. Nature
22. Noda M, Nakanishi H, Nabekura J, Akaike N 472(7343):319–324
(2000) AMPA-kainate subtypes of glutamate 36. Stark GR (2007) How cells respond to inter-
receptor in rat cerebral microglia. J Neurosci ferons revisited: from early history to current
20(1):251–258 complexity. Cytokine Growth Factor Rev
23. Tsuda M, Shigemoto-Mogami Y, Koizumi S, 18(5–6):419–423
Mizokoshi A, Kohsaka S, Salter MW, Inoue K 37. Hu X, Paik PK, Chen J, Yarilina A, Kockeritz
(2003) P2X4 receptors induced in spinal L, Lu TT, Woodgett JR, Ivashkiv LB (2006)
microglia gate tactile allodynia after nerve IFN-gamma suppresses IL-10 production and
injury. Nature 424(6950):778–783 synergizes with TLR2 by regulating GSK3 and
24. Koizumi S, Shigemoto-Mogami Y, Nasu-Tada CREB/AP-1 proteins. Immunity 24(5):
K, Shinozaki Y, Ohsawa K, Tsuda M, Joshi BV, 563–574
Part II
Isolation and Culture of Microglia

Chapter 2
Cell Culturing of Human and Murine Microglia Cell Lines

Johanna Rodhe
Abstract
Despite the fact that microglia cells were first described almost a century ago, microglia-derived immortalized
cell lines have only been established in the last two decades. One should be aware of their limitations but
also of their advantages. Cell lines offer a potentially powerful tool to investigate some functional aspects
of microglia. Cell culturing of human and murine microglia cell lines will be described in this chapter.
It includes a presentation of equipment needed, cell culture medium and supplements, cell culture
monitoring, and a protocol describing the steps for subculturing of microglia cell lines.
Key words Microglia cell line, BV2, CHME5, Cell culture, Medium, Serum
1 Introduction
The concept of microglia cells was first described by Rio-Hortega

in 1919 in studies of tissue sections of the brain [1]. Even though
isolation of primary microglia cells for in vitro cultures was already
described in 1930, the first immortalized cell lines from rodents
were not established until the late 1980s when microglia cells from
primary cultures of mouse brain were infected with myc oncogenes
[2, 3]. Several rodent cell lines were established in the following
years, before microglia cells could be isolated from human embry-
onic spinal cord and cortex and immortalized to form the first
human microglia cell line in 1995 [4].
Microglia cell lines are derived from primary cultures of
microglia cells from the brain or spinal cord, which are then
immortalized by, e.g., infection by a retroviral vector carrying a
v-raf, v-myc, or c-myc oncogene, the simian virus 40 (SV40)
large T antigen, or spontaneously immortalized in culture (see
Table 1) [2–12].
Immortalized microglia cells can express several typical microglia/
macrophage markers, like CD11b/Mac1, CD68, FcR, MHC class
I, MHC class II, B4, and Mac2 [13]. They should keep some of
11
12 Johanna Rodhe
Table 1
Table showing some of the existing murine and human microglia cell lines
Cell line Species Immortalization method Year Authors

N9, N13 Mouse v-myc oncogene 1989 Righi et al. [2]
BV2 (BV-2) Mouse v-raf/v-myc oncogenes 1990 Blasi et al. [3]
EOC Mouse Spontaneous 1995 Walker et al. [5]
C8-B4 Mouse Spontaneous 1996 Alliot et al. [6]
MG5 Mouse p53-deficient mice 1997 Ohsawa et al. [7]
MG6 (MG6-1) Mouse c-myc oncogene 2005 Takenouchi et al. [8]
MLS-9 Rat 1998 Zhou et al. [9]
Ra2 Rat Spontaneous 1998 Sawada et al. [10]
HAPI Rat Spontaneous 2001 Cheepsunthorn et al. [11]
CHME Human SV40 large T antigen 1995 Janabi et al. [4]
HMO6 Human v-myc oncogene 2001 Nagai et al. [12]
the important characteristics of microglial cells, like the ability of

phagocytosis and secretion of cytokines upon activation, e.g. IL-6
and TNF-α [2, 11, 12].
A comparison with primary microglia culture shows that there
are both advantages as well as limitations when using continuous
cell lines. A primary cell culture is time-consuming to prepare and
gives a limited number of cells per animal, which can only be used
for a small number of passages. The established cell lines on the
other hand can be thawed and frozen when needed and give a vir-
tually unlimited number of cells, which can be used for numerous
of experiments over many passages. Cell lines also have the advan-
tage of being characterized and are generally at the same stage of
maturation. They are a more homogenous population of cells
derived from one clone, which gives less variation in experiments
and makes it easier to compare results between different labs. This
can also be seen as a disadvantage; since the cell lines are selected
from the expansion of one clone, there is a possibility that the cell
line has lost important characteristics or markers from the popula-
tion in primary culture. Continuous cell lines are also rapidly divid-
ing and have therefore the ability to differentiate and change
characteristics over time, which can affect the outcome of long-
term experiments. Since there are differences between the cell lines
in expression of antigens and in response to different stimuli, it is
important to select the cell line which has the characteristics needed
to study the question in focus.
Culturing of Microglia Cell Lines 13
The access to human samples for isolating microglia from the

brain and spinal cord is very limited in both availability and yield,
which makes it a great advantage of having continuous cell lines.
Microglia cell lines are adherent cells that grow in monolayer in
cell culture. The cells are grown in a H2O-saturated atmosphere
with 5 % CO2 at 37 °C. The following section describes a routine
protocol for subculturing of common microglia cell lines, such as
BV2 and CHME5.
2 Material
1. Dulbecco’s Modified Eagle Medium (DMEM: high glucose

with glutamax/glutamine), supplemented with 10 % heat-
inactivated Fetal Bovine Serum (FBS), penicillin (100 U/ml),
and streptomycin (100 μg/ml) (see Notes 1 and 2).
2. 0,05 % trypsin–EDTA.
3. Phosphate-buffered saline (PBS) pH 7.4.
4. Cell culture flasks.
5. 5 and 10 ml pipettes.
6. Falcon tubes.
3 Methods
3.1 Preparation 1. Warm cell culture medium and trypsin bottle to 37 °C.
2. Prepare working area and bring out all equipment needed. All
cell culture work should be performed in laminar flow hood
using sterile techniques, and all equipment and solutions used
for cell culture must be sterile.
3. Check the cells under an inverted microscope to ensure healthy
cells (see Note 3), mainly attached to bottom of flask and
expanding at normal growth rate (see Note 4). Observe cell
culture to ensure absence of contamination (see Note 5) and
look out for color changes in the culture medium.
3.2 Subculturing 1. Remove and discard the medium from the cell flask.
2. Wash the cells with 5 ml room temperature PBS, tip the flask
gently to rinse off remaining FBS from the culture media, and
then remove the PBS.
3. Add 2–3 ml of trypsin–EDTA (ca. 1 ml per 25 cm2 of surface
area), rock the flask gently for complete coverage of the cell
layer, and leave in an incubator at 37 °C for 1–5 min for the
cells to detach. Avoid prolonged periods in trypsin by checking
the cells oft during the trypsinization. If cells are slow to
detach, tap gently on the bottle for the cells to lift off.
14 Johanna Rodhe
4. As soon as the cells have detached, add 5–10 ml of complete

medium (2–3 times the amount of trypsin added) to the cells
and transfer cell suspension to a Falcon tube.
5. Spin down the cells in a centrifuge at 200 × g for 5 min.
6. Remove and discard the supernatant and resuspend the cell
pellet in new medium.
7. Add desired amount of cell suspension into a new culture flask
and add complete culture medium (see Notes 6 and 7).
8. Return cell flask to incubator.
4 Notes
1. RPMI 1640 medium can also be used for culturing of microglia

cell lines [2, 3] and can be used to replace DMEM if preferred.
Another option for some microglia cell lines is to use MEM as
culture medium [9, 10]. Although most cell lines are cultured
in regular culture medium, some cell lines do require special
supplements to proliferate, like GM–CSF or insulin [10].
Consult original publications of the cell lines for information
about eventual need of specific supplements.
2. For experiments with microglia cells, the medium used for the
treatment of cells can be supplemented with a lower amount of
FBS than in continuous cell culture, e.g., when activating cells
using the proinflammatory stimulant lipopolysaccharide (LPS).
Medium containing 5 % FBS can be used for experiments [11,
14] or even serum-free medium during the treatments [15].
The amount of serum that is used, the source, batch, and
storage of serum (e.g., repeated freeze and thaw cycles) can be
a source of diversity between experiments both within and
between labs. Keeping the same standards within experiments
is important for consistency in results. The choice of what
amount of serum to use depends on duration of experiments
and possible effect of components in the serum. In cell culture,
serum is the source of nutrients for the cells, but in vivo the
cells only come in contact with complete serum in pathological
conditions when breakdown of blood–brain barrier exposes
microglia to several serum components that can cause microg-
lia activation [16, 17]. This is something to keep in mind for
experiments using compounds to stimulate activation of
microglia cells, since the serum exposure can give a higher
baseline of microglia activation. An example of this is the LPS-
binding protein (LBP), which can be found in serum. The
binding of LBP to LPS increases its affinity for binding to
microglia surface receptors, which results in an increased sensi-
tivity to LPS and a higher degree of microglia activation [18].
Culturing of Microglia Cell Lines 15
Another source of diversity here can also be the choice of LPS.

The source, batch, storage conditions, and methods of
application of compound affect the sensitivity of microglia
cells to LPS.
3. A good habit is to look at the cells under the microscope at a
regular basis. Get to know how the cell line looks under normal
conditions to be able to observe any changes in the culture and
recognize changes in morphology and signs of activated cells
when evaluating effect of experiments. Also note that healthy
cells should have light refracting in a ring around the cells in
the inverted light microscope.
4. Many microglia cell lines are sensitive to being over-confluent
and should be subcultured at ca. 80–90 % confluency.
Subculturing should be done every 2–3 days depending on
growth speed of cell line.
5. Always look out for traces of contamination in the cell culture.
Even very low levels of bacterial contamination can activate
microglia cells.
6. Cells are often split in a ratio of 1:4–1:10, depending on
growth rate of the cell line. Add ca. 5–10 ml of complete
DMEM per 25 cm2 of surface area of the flask.
7. It is good to keep track of the number of passages for the cell
line, e.g., write passage number on the culture bottle before
returning it to the incubator. Keep the number of passages
relatively low to avoid selection pressure and diversion from
the parental cell line. The maximum number of passages can
differ between cell lines. Never let the amount of passages
exceed the number where you have to question if the charac-
teristics that are studied are compromised.
References
1. Rio Hortega PD (1919) El tercer elemento de with the SV40 large T antigen. Neurosci Lett
los centros nerviosos. Bol Soc Esp Biol 195(2):105–108
9:69–129 5. Walker WS, Gatewood J, Olivas E, Askew D,
2. Righi M, Mori L, De Libero G, Sironi M, Havenith CE (1995) Mouse microglial cell
Biondi A, Mantovani A, Donini SD, Ricciardi- lines differing in constitutive and interferon-
Castagnoli P (1989) Monokine production by gamma-inducible antigen-presenting activities
microglial cell clones. Eur J Immunol for naive and memory CD4+ and CD8+ T
19(8):1443–1448 cells. J Neuroimmunol 63(2):163–174
3. Blasi E, Barluzzi R, Bocchini V, Mazzolla R, 6. Alliot F, Marty MC, Cambier D, Pessac B
Bistoni F (1990) Immortalization of murine (1996) A spontaneously immortalized mouse
microglial cells by a v-raf/v-myc carrying ret- microglial cell line expressing CD4. Brain Res
rovirus. J Neuroimmunol 27(2–3):229–237 Dev Brain Res 95(1):140–143
4. Janabi N, Peudenier S, Héron B, Ng KH, 7. Ohsawa K, Imai Y, Nakajima K, Kohsaka S
Tardieu M (1995) Establishment of human (1997) Generation and characterization of a
microglial cell lines after transfection of pri- microglial cell line, MG5, derived from a
mary cultures of embryonic microglial cells p53-deficient mouse. Glia 21(3):285–298
16 Johanna Rodhe
8. Takenouchi T, Ogihara K, Sato M, Kitani H 13. De Vries GH, Boullerne AI (2010) Glial cell
(2005) Inhibitory effects of U73122 and lines: an overview. Neurochem Res 35(12):
U73343 on Ca2+ influx and pore formation 1978–2000
induced by the activation of P2X7 nucleotide 14. Burguillos MA, Deierborg T, Kavanagh E,
receptors in mouse microglial cell line. Biochim Persson A, Hajji N, Garcia-Quintanilla A,
Biophys Acta 1726(2):177–186 Cano J, Brundin P, Englund E, Venero JL,
9. Zhou W, Cayabyab FS, Pennefather PS, Joseph B (2011) Caspase signalling controls
Schlichter LC, DeCoursey TE (1998) HERG- microglia activation and neurotoxicity. Nature
like K+ channels in microglia. J Gen Physiol 472(7343):319–324
111(6):781–794 15. Pocivavsek A, Burns MP, Rebeck GW (2009)
10. Sawada M, Imai F, Suzuki H, Hayakawa M, Low-density lipoprotein receptors regulate
Kanno T, Nagatsu T (1998) Brain-specific microglial inflammation through c-Jun
gene expression by immortalized microglial N-terminal kinase. Glia 57(4):444–453
cell-mediated gene transfer in the mammalian 16. Möller T, Weinstein JR, Hanisch UK (2006)
brain. FEBS Lett 433(1–2):37–40 Activation of microglial cells by thrombin:
11. Cheepsunthorn P, Radov L, Menzies S, Reid J, past, present, and future. Semin Thromb
Connor JR (2001) Characterization of a novel Hemost 32(Suppl 1):69–76
brain-derived microglial cell line isolated from 17. Ransohoff RM, Perry VH (2009) Microglial
neonatal rat brain. Glia 35(1):53–62 physiology: unique stimuli, specialized
12. Nagai A, Nakagawa E, Hatori K, Choi HB, responses. Annu Rev Immunol 27:119–145
McLarnon JG, Lee MA, Kim SU (2001) 18. Schumann RR, Leong SR, Flaggs GW, Gray
Generation and characterization of immortal- PW, Wright SD, Mathison JC, Tobias PS,
ized human microglial cell lines: expression of Ulevitch RJ (1990) Structure and function of
cytokines and chemokines. Neurobiol Dis lipopolysaccharide binding protein. Science
8(6):1057–1068 249(4975):1429–1431
Chapter 3
Microglia Isolation from Adult Mouse Brain

Jae-Kyung Lee and Malú G. Tansey
Abstract
Although microglia isolation from embryonic or postnatal mouse brain is possible using a number of
different protocols, microglia isolation from adult brain is more challenging and often results in low yields.
Here, we describe a protocol to isolate intact microglia from adult mouse brain for functional assays,
immunocytochemistry, and/or flow cytometry analysis. This protocol involves enzymatic dissociation in
medium supplemented with dispase II, papain, and DNase I followed by mechanical dissociation. Cell
separation is achieved via percoll gradients of various densities. Microglia isolated using this protocol is
suitable for flow cytometry analysis, RNA isolation for gene expression by real-time PCR or microarrays,
and for functional assays including cytokine production, chemotaxis, and phagocytosis.
Key words Microglia, CD11b, Inflammatory gene expression, Flow cytometry, Immunocytochemistry
1 Introduction
Microglia, the myeloid-derived resident macrophages of the brain,

play the primary role of immune surveillance and respond to
environmental stress and immunological challenges [1, 2]. Initial
physical or pathogenic events in the CNS can trigger microglial
expansion through recruitment of peripheral macrophages to the
CNS, as a result of increased permeability of the BBB, differentia-
tion from progenitor cells, or proliferation of residual microglia
[3]. Acute activation of microglia often results in secretion of neu-
rotrophic factors such as glial-derived neurotrophic factor family
ligands (GFLs) that limit tissue injury by protecting vulnerable
populations of neurons and aiding in repair processes [4, 5].
However, activated microglia can also overproduce prostaglandins,
chemokines, cytokines, and reactive oxygen and nitrogen species
(ROS and RNS) including nitric oxide (NO), which can have a
deleterious effect on neuronal survival by enhancing oxidative
stress and activating cell death pathways [6]. If glial activation
persists for extended periods, a cycle of chronic neuroinflammation
ensues and contributes to tissue damage. This protocol describes a
17
18 Jae-Kyung Lee and Malú G. Tansey
stepwise method (modified from one previously described [7]) to

isolate intact microglia from adult mouse brain. Microglia isolated
using this method are suitable for in vitro studies aimed at under-
standing the mechanisms that regulate the balance of neuroprotective
and neurotoxic microglial activities.
2 Materials
2.1 Reagents 1. PBS-perfused brains from adult mice (age >8 weeks).
2. Hank’s balanced salt solution (HBSS) without calcium and
magnesium.
3. 10× HBSS.
4. DNase I (final concentration; 20 U/mL, Invitrogen).
5. Dispase II (final concentration; 1.2 U/mL, Roche).
6. Papain (1 mg/mL, Sigma-Aldrich).
7. DMEM/F12 medium.
8. 100× penicillin/streptomycin.
9. Percoll.
3 Media and Solutions
1. Dissection medium (50 mL).

49.5 mL of HBSS plus 0.5 mL of penicillin/streptomycin
(100×).
Sterile filter all solutions after mixing/before use.
2. Glucose.
Prepare 0.45 g/mL in PBS (100×).
Add 500 μL in 50 mL reagents to make final concentration of
4,500 mg/L.
3. Serum-free medium (50 mL).
49 mL of DMEM/F12 plus 0.5 mL of penicillin/streptomy-
cin (100×) and 0.5 mL of glucose (100×).
4. Neutralization medium (50 mL).
44 mL of DMEM/F12 plus 0.5 mL of penicillin/streptomy-
cin (100×), 0.5 mL of glucose (100×), and 5 mL of heat-
inactivated FBS (or FCS).
Adult Mouse Microglia 19
5. Culture medium (50 mL).

44.5 mL of DMEM/F12 plus 0.5 mL of penicillin/streptomy-
cin (100×) and 5 mL of heat-inactivated FBS (or FCS).
6. Dispase II
Dissolve the non-sterile lyophilized enzyme in HEPES-
buffered saline (50 mM HEPES/KOH pH 7.4, 150 mM
NaCl) (10 mg/mL). Dilute further with the culture medium
to be used for the isolated cells at the final concentration of
2.4 U/mL. Concentrations higher than 2.4 U/mL are not
recommended. Sterilize through a 0.22 μm filter membrane.
7. Dissociation medium.
Prepare dispase, DNase, and papain (DDP) solution.
● 0.028 g papain (final concentration 1 mg/mL).
● 14 mL of DMEM/F12.
● 14 mL premade dispase II (final concentration 1.2 U/mL).
Aliquot in −20 °C.
Add DNase I (final concentration 20 U/mL) right before use.
8. Percoll preparation.
Prepare stock isotonic percoll (SIP); mix nine parts of percoll
with one part 10× HBSS.
70 % percoll: 7.0 mL of SIP plus 2.0 mL of 1× HBSS.
4 Methods
1. Place the freshly perfused adult brain in 2 mL serum-free

media into the 35 mm dish and mince the freshly perfused
adult brain as finely as possible using the 15 T scalpel blade.
Make sure the amount of serum-free medium is equivalent to
the amount of dissociation medium.
2. Transfer minced brain to 15 mL tube containing 3 mL of dis-
sociation medium.
3. Gently rock (or invert the tube every 5 min) cell suspension in
tissue culture incubator for 20 min.
4. Neutralize the enzymes in dissociation media by adding 5 mL
of neutralization medium.
5. Centrifuge 5 min at 250 × g in room temperature.
20 Jae-Kyung Lee and Malú G. Tansey
6. Aspirate the medium slowly (be extremely cautious not to

disturb the pellet as it can easily be aspirated by the aspirating
pipette).
7. Resuspend the pellet in 5 mL of serum-free medium. Repeat
steps 5 and 6.
8. Add 3 mL DMEM/F12 and pipette up and down with
polished large size-hole Pasteur pipette against bottom of tube
until large clumps of tissue are broken up. Let it sit for 1 min until
large clumps settle. Transfer upper part (containing dissociated
cells) into new 15 mL conical tube and keep it on ice.
9. Add 3 mL of DMEM/F12 and pipette up and down with
polished Pasteur pipette (medium-size hole) against bottom of
tube until large clumps of tissue are broken up. Let it sit for
1–2 min until large clumps settle. Transfer upper part into col-
lecting tube with previous cell suspension and keep it on ice.
10. Add 2 mL DMEM/F12 and pipette up and down with pol-
ished Pasteur pipette (small-size hole) to break clumps. Let it
sit for 1 min until large clumps settle.
11. Combine upper part (containing dissociated cells) into previous
cell suspension.
12. Wet 40 μm cell strainer with 2 mL of DMEM/F12 and then
filter cell suspension through cell strainer.
13. Spin at 250 × g for 4 min.
14. Resuspend cells in 5 mL DMEM/F12, spin 250 × g for 4 min,
and remove supernatant.
15. Resuspend cell pellet in 4 mL per brain of 37 % SIP.
16. Transfer 4 mL of the 37 % SIP (from step 15) to 15 mL coni-
cal tubes and slowly underlay 4 mL of 70 % percoll (see Note
1). Then on top of the 37 % layer, slowly pipette 4 mL of 30 %
percoll, followed by 2 mL of HBSS (Fig. 1).
17. Centrifuge gradient 40 min at 300 × g (18 °C) with no brake.
***Important!! Make sure centrifuge will stop with no brake
so that the interphase is not disturbed.
18. Using a transfer pipette, gently remove layer of debris and col-
lect 2.0–2.5 mL of the 70–37 % interphase into a clean 15 mL
conical tube.
19. Add 6 mL of HBSS for each 2 mL of interphase volume
collected to ensure the percoll containing the interphase is
diluted about three times.
20. Centrifuge 7 min at 500 × g at 4 °C.
Fig. 1 Schematic of the percoll gradient setup for isolation of adult mouse
microglia
21. Resuspend pellet in 500 μL of HBSS and transfer to small 0.6

mL or 1.5 mL tubes and wash three times in a volume of
500 μL, using a micro-centrifuge at 800 × g at 4 °C.
22. Count cells using a hemocytometer and seed cells on the cul-
ture plate for immunocytochemistry or functional assays (see
Notes 2–4).
5 Notes
1. Percoll should be kept at room temperature.

2. Typical yield is from 300,000 to 500,000 microglial cells per
mouse brain (see Fig. 2).
3. All the procedures need to be performed under the hood and
reagents needs to be filtered to prevent contamination.
4. Cells can be maintained in the culture media for functional
analysis (see Fig. 3).
Fig. 2 Purity of isolated adult mouse microglia population. (a) Primary microglia cells from C57BL/6 mice were
plated into a 4-well chamber plate at the density of 8,000 cells/well in growth medium. Twenty-four hours
after seeding, cells were fixed with 4 % paraformaldehyde for 15 min. Cells were then stained with CD68 (as
microglia marker, green) and GFAP (as astrocyte marker, red ) antibodies followed by incubating with Alexa
fluorophore-conjugated secondary antibody (Invitrogen). (b) Primary microglia cells were labeled with fluores-
cently conjugated CD68-FITC antibodies (green ), and fixed and analyzed by flow cytometry. Unstained microg-
lia (red) were used as negative controls. Histogram plot indicates the presence of a single-labeled population.
Scale bar represents 100 μm
Fig. 3 Isolated adult mouse microglia secrete cytokines upon stimulation with LPS. Primary microglia cells
were plated at a density of 3,000 cells per well in 96-well plate. Cells were treated with 1 μg/mL of LPS for
18 h. Conditioned media were collected and analyzed for inflammatory factor production by multiplexed
immunoassay (Meso Scale Discovery). Student t-test was used for statistical analysis. The asterisks denote
significant differences between PBS and LPS treatment at p < 0.001
Acknowledgements
This work was supported by grant 1R01 NS072467 (MGT) from
NIH/NINDS.
References
1. Puntambekar SS, Doose JM, Carson MJ (2008) 4. Wyss-Coray T, Mucke L (2002) Inflammation
Microglia: a CNS-specific tissue macrophage. in neurodegenerative disease—a double-edged
In: Lane TE, Carson M, Bergmann C, Wyss- sword. Neuron 35(3):419–432
Coray T (eds) Central nervous system diseases 5. Marchetti B, Abbracchio MP (2005) To be or not
and inflammation. Springer, New York, pp 1–12 to be (inflamed) – is that the question in anti-
2. Tansey MG, Wyss-Coray T (2008) Cytokines in inflammatory drug therapy of neurodegenerative
CNS inflammation and disease. In: Lane TEC, disorders? Trends Pharmacol Sci 26(10):517–525
Carson M, Bergmann C, Wyss-Coray T (eds) 6. McGeer PL, McGeer EG (2004) Inflammation
Central nervous system diseases and inflamma- and neurodegeneration in Parkinson’s disease.
tion. Springer, New York, pp 59–106 Parkinsonism Relat Disord 10(Suppl 1):S3–S7
3. Bjorkqvist M, Wild EJ, Thiele J et al (2008) A 7. Cardona AE, Huang D, Sasse ME et al (2006)
novel pathogenic pathway of immune activation Isolation of murine microglial cells for RNA
detectable before clinical onset in Huntington’s analysis or flow cytometry. Nat Protoc 1(4):
disease. J Exp Med 205(8):1869–1877 1947–1951
Chapter 4
Preparation of Primary Microglia Cultures

from Postnatal Mouse and Rat Brains
Tomas Deierborg
Abstract
Microglia are the inflammatory cells of the brain and are activated in neuropathological conditions. To
study the biology of microglia, these cells can be isolated from the brain and analyzed in terms of pro- and
anti-inflammatory cytokine production, involvement of intracellular signaling pathways upon inflamma-
tory stimuli, phagocytosis, and several more biological aspects to understand their role in the brain. In this
book chapter, I will discuss microglial cells and describe how these cells can be cultured from a postnatal
mouse and rat brain.
Key words Microglia, Inflammation, Cytokine, Postnatal, Cell culture
1 Introduction
Microglial cells are the immune cells of CNS and play an important
role not only in inflammation processes in neuropathological
conditions but also to maintain normal homeostasis under physi-
ological conditions [1].
In the situation of neurodegeneration and acute brain injuries,
microglia are proliferating, adapting a hypertrophic morphology,
and migrate towards an injury.
Microglia is a phagocytizing cell type and can typically be
found at the site of injury where the cells take care of cell debris.
Microglia in the brain have two origins: one is the parenchymal
origin meaning that the cells come from precursors in the brain
parenchyma and the second origin is from blood-borne mono-
cytes. These two different origins cannot be distinguished at an
immunohistological level, but the different populations can be dis-
tinguished in terms of expression levels of CD11b, where infiltrat-
ing cells from the blood have a high expression of CD11b [2].
Microglia respond quickly to a multitude of stimulus, which
leads to a production of cytokines, predominantly proinflammatory
25
26 Tomas Deierborg
Fig. 1 Mixed culture of glial cells from postnatal mouse brains showing clearly
the microglial cells as small round cells on top of the astrocytic monolayer. Scale
bar, 50 μm
cytokines (e.g., IL-1β, TNF-α) that are initiating and maintaining

the inflammatory response but also anti-inflammatory cytokines
(e.g., IL-10 and IL-4) that are important in the regulation of the
inflammatory cascade. Hence, determining the role of microglia in
neuroinflammatory conditions and their role in inflammation prop-
agation and attenuation is important for understanding and treat-
ing pathological conditions in the brain. Studying primary cultures
of microglia from rats and mice provides important and reproduc-
ible methods to unravel the role of microglia in brain pathologies.
Isolation and culturing of microglia from the postnatal rodent
brain is a well-characterized and reliable method to study microglia
[3]. Though, it should be emphasized that microglia in culture are
cells that have been altered by the isolation and culturing condition
and might therefore not resemble the normal environment in the
brain. The preparation and isolation protocol is straightforward
where a mixed population of glial cells is isolated and grown
together before the microglia is isolated after a couple of weeks.
The small amoeboid microglial cells are growing on top of a mono-
layer of astrocytes where the microglia typically appears as clusters
or colonies suggesting a clonal expansion of precursor cells or indi-
vidual cells (see Fig. 1). Earlier studies have characterized this dis-
tinct phenotype of amoeboid microglial cells. Microglial cells are
similar to macrophages and show ability to secrete cytokines, pro-
duce reactive oxygen species, phagocytize, and proliferate. The
ramified microglia population on the other hand has a reduced
proliferation rate and ability of phagocytosis. When cultures are
maintained for longer time periods, the ramified microglial cells
Primary Microglia Cultures 27
will integrate into the astrocytic layer [4, 5]. Interestingly, increasing
intracellular calcium or cAMP will make the cells convert back to a
round amoeboid morphology [4] showing, at least in vitro, that
the two distinct phenotypes can convert from one to another.
Microglia has an amoeboid morphology during embryogenesis
that is reduced postnatally and increases following brain injuries.
It is suggested that brain-derived microglia is different to bone
marrow-derived brain macrophages [6] in terms of inflammatory
profile and morphology and that microglial cells respond to
granulocyte/macrophage colony-stimulating factor (GM-CSF) unlike
bone marrow-derived monocytes or tissue macrophages [7].
In mixed glial cultures, which will be described later in terms of
how to harvest microglia, the astrocytes are important for the pro-
liferation of microglia, and astrocytes also have an effect on the
ramification of the microglia [7, 8]. Astrocytes produce trophic fac-
tors, e.g., macrophage colony-stimulating factor (M-CSF) and
GM-CSF, that will regulate microglial cells [9, 10]. Since these fac-
tors will have the ability to change the phenotypes of the microglia,
i.e., inducing ramification [11], augments phagocytic ability, chang-
ing cytokine production, and antigen-presenting ability [9, 12, 13],
there are reasons to avoid these mitogens in the culturing proce-
dures. It has been suggested that microglia from adult mice that has
been treated with GM-CSF can acquire a dendritic cell phenotype,
whereas the presence of the growth factor M-CSF can convert the
microglia into a proliferative phenotype [14]. Also, the rat microg-
lia after treatment with GM-CSF have a reduced period of prolifera-
tion which corresponds to shortening of telomeres [15].
In this book chapter I will describe a simple and reproducible
method to isolate brain cells, culturing mixed glial cultures with-
out additional mitogens, only using the supportive function of
astrocytes, and harvest pure microglial cells that could be seeded
and cultured for detailed cellular experiments.
2 Materials
The material needed to dissect out and culture microglial cells

from mouse pups postnatal day 1–4.
1. Dissection microscope.
2. Ice-cold Hank’s Balanced Salt Solution without bivalent ions
(HBSS).
3. Freezing block.
4. Petri dishes.
5. Two pairs of fine forceps.
6. Fine scissors.
7. Trypsin (0.1 %).
28 Tomas Deierborg
8. DNase (0.05 %).

9. DMEM + GlutaMAX medium containing 4.5 g/l glucose
without sodium pyruvate.
10. Fetal Bovine Serum (FBS).
11. Penicillin/streptomycin (100 U/ml and 0.1 mg/ml,
respectively).
12. Eppendorf tubes.
13. Falcon tubes (15 or 50 ml).
14. T75 culture flasks.
15. Incubator.
16. Centrifuge.
17. Sterile dissections hood/bench.
18. Pipette 1 ml and pipette tips.
3 Methods
The isolation of tissue from mouse and rats is similar. I will describe
the typical isolation of tissue from the mouse brain, but the same
protocol can be used for the rat brain. Microglia is derived from
mouse pups postnatal days 1–4.
1. Mouse pups are decapitated and heads are put in ice-cold HBSS.
For the whole procedure, brains and dissected out tissue
are kept in ice-cold HBSS on a cold freezing block or ice in the
dissection hood.
2. The skull is cut open with a scissors along the sagittal fissure,
and the bone and cartilage of the left and right hemisphere are
removed or dragged to the side.
3. After the bone and cartilage have been removed, the cortical
tissue can be removed in one operation by going down with a
forceps sagittally around one hemisphere, squeezing the for-
ceps while pulling upwards. By this maneuver, one hemisphere
including cortex, striatum, and hippocampus can be isolated.
4. The hippocampus and striatum are removed from the cortex as
well as the choroid plexus using a dissection microscope. The
dura mater surrounding the cortex can be removed as much as
reasonable possible. Even though the removal of dura mater is
not essential for the microglia culture, this membrane can
damage/remove the astrocytic monolayer when microglia is
removed by smacking the culture flask.
5. The cortex tissue is cut up in pieces with scissors, thereafter
trypsin (0.1 %) and DNase (0.05 %) are added to the HBSS
(without bivalent ions) and incubated in an Eppendorf tube
for 15–20 min at 37 °C.
6. The tissue is mechanically dissociated by using a 1 ml pipette

tip by aspirating the tissue up and down 5–10 times until a cell
suspension is obtained. Avoid air bubbles to minimize the
damage to the cells.
7. The cell suspension is then spun down at 400 × g for 4 min in
a Falcon tube.
8. After centrifugation the supernatant is removed and the pellet,
containing the cells, is taken up from the tube and put into a
T75 flask with 10 ml of DMEM + GlutaMAX medium contain-
ing 4.5 g/l glucose without sodium pyruvate (Gibco by Life
Technologies) supplemented with 10 % of Fetal Bovine Serum
(FBS) and penicillin/streptomycin (100 U/ml and 0,1 mg/ml,
respectively). Cultures are maintained at 37 °C and 5 % CO2.
9. Normally, the cortices from three mouse pups should be cul-
tured in one T75 flask, and it is not beneficial for the cultures
to have too low concentration of cells, which will compromise
the yield of microglia in the end. For rat, the typical cell con-
centration per T75 flask should be 10 million cells.
10. Optionally, a cell strainer, 70 μm mesh, can be used to get rid
of bigger tissue/cell aggregates. Even though some protocols
suggest changing of medium after 1 day, it is often better to
wait around 2–4 days before changing medium to let the cells
attached to the bottom of the culture flask.
11. The medium can then be changed two times a week, by
exchanging about 60 % of the culture medium, or once a week
and then changing about 90 % of the culture medium.
12. After around 2–3 weeks of culture, the microglia can be har-
vested. The microglial cells will appear as small round cells on
top of the monolayer of astrocytes, typically appearing in clus-
ters (Fig. 1).
13. The microglial cells can easily be removed from the T75 flasks
by smacking the cultures flask 10–20 times, which will let the
round microglial cells let go from the astrocytic monolayer.
Too vigorous smacking can result in that the astrocytic mono-
layer detaches from the bottom of the culture flask. Smacking
the flasks and continuous looking in the microscope that satis-
factory number of microglial cells have detached are recom-
mended. Even though several protocols describe the use of a
rotating shaker that could be used for up to 1 h to isolate the
microglia from the flask, smacking the flask works equally
good. All culture medium that now contains microglia should
be removed from the T75 flask.
14. The cell-containing medium should be centrifuged at 400 × g
for 4 min and the supernatant removed to get the pellet with
the microglial cells. Around 95 % of microglial purity can be
30 Tomas Deierborg
expected using this approach. The experiments can be pursued

directly with the harvested microglial cells (see Notes 1 and 2).
The cells can also be cultured for a few days (to get a ramified
morphology) and then the experiments can be performed. For
culture in a 96-well plate, a concentration of 20,000 microglial
cells/well can be used. During the experiment the serum
should be reduced (e.g., 2 %) to maximize the inflammatory
response by the microglia.
4 Notes
1. It is common to only use these mixed cultures to harvest one

yield of microglia. However, it has been shown by Floden and
Combs that it is possible to get out more microglia from sub-
sequent yields [16]. They show that two more yields can be
obtained after another 7 + 7 days of culture with fresh culture
medium with a yield that is about 20 % of the first yield in
terms of cell numbers. The two subsequent yields isolated con-
tain microglia that show similar morphology and immunoreac-
tivity of the microglial markers CD11b, CD68, MHCII, and
CD45, and cell survival from yield 1 and 3 was not changed.
However, the secretion of two cytokines examined, IL-6 and
TNF-α, showed significantly decreased secretion of these cyto-
kines 24 h after an LPS challenge, suggesting that cytokine
production or secretory ability is reduced after the first yield.
The phagocytic activity, measured by the ability to phagocyte
FITC-conjugated bioparticles, was not altered.
2. In this book chapter, I describe a protocol that uses the intrin-
sic ability of glial cells to proliferate in culture without adding
any mitogens (e.g., GM-CSF, M-CSF). With this protocol it is
easy to retrieve good yield of microglia that will give robust
and reproducible results that can uncover new knowledge
about the biology of the microglia.
Acknowledgement
This work has been supported by grants provided by the Swedish

Research Council grant no. 2012–2229 by the Gyllenstierna
Krapperup, the Royal Physiographic Society, Swedish stroke asso-
ciation, Crafoord, A.E. Berger, Wiberg, Bergvall, G & J Kock
foundations.
References
1. Ransohoff RM, Brown MA (2012) Innate 10. Gehrmann J (1996) Microglia: a sensor to
immunity in the central nervous system. J Clin threats in the nervous system? Res Virol
Invest 122(4):1164–1171 147(2–3):79–88
2. Lambertsen KL et al (2011) Differences in ori- 11. Fujita H et al (1996) Effects of GM-CSF and
gin of reactive microglia in bone marrow chime- ordinary supplements on the ramification of
ric mouse and rat after transient global ischemia. microglia in culture: a morphometrical study.
J Neuropathol Exp Neurol 70(6):481–494 Glia 18(4):269–281
3. Giulian D, Baker TJ (1985) Peptides released 12. Aloisi F et al (2000) Functional maturation of
by ameboid microglia regulate astroglial pro- adult mouse resting microglia into an APC is
liferation. J Cell Biol 101(6):2411–2415 promoted by granulocyte-macrophage colony-
4. Kalla R et al (2003) Loss of microglial ramifica- stimulating factor and interaction with Th1
tion in microglia-astrocyte cocultures: involve- cells. J Immunol 164(4):1705–1712
ment of adenylate cyclase, calcium, phosphatase, 13. Lee SC et al (1993) Macrophage colony-
and Gi-protein systems. Glia 41(1):50–63 stimulating factor in human fetal astrocytes
5. Tanaka J et al (1999) Morphological differen- and microglia. Differential regulation by cyto-
tiation of microglial cells in culture: involve- kines and lipopolysaccharide, and modulation
ment of insoluble factors derived from of class II MHC on microglia. J Immunol
astrocytes. Neurosci Res 34(4):207–215 150(2):594–604
6. Lambertsen KL et al (2009) Microglia protect 14. Ponomarev ED et al (2005) Development of
neurons against ischemia by synthesis of tumor a culture system that supports adult microg-
necrosis factor. J Neurosci 29(5):1319–1330 lial cell proliferation and maintenance in the
7. Giulian D et al (1995) Cell surface morphol- resting state. J Immunol Methods 300(1–2):
ogy identifies microglia as a distinct class of 32–46
mononuclear phagocyte. J Neurosci 15(11): 15. Flanary BE, Streit WJ (2004) Progressive
7712–7726 telomere shortening occurs in cultured rat
8. Tanaka J, Maeda N (1996) Microglial ramifi- microglia, but not astrocytes. Glia 45(1):
cation requires nondiffusible factors derived 75–88
from astrocytes. Exp Neurol 137(2):367–375 16. Floden AM, Combs CK (2007) Microglia
9. Giulian D, Ingeman JE (1988) Colony- repetitively isolated from in vitro mixed glial
stimulating factors as promoters of ameboid cultures retain their initial phenotype.
microglia. J Neurosci 8(12):4707–4717 J Neurosci Methods 164(2):218–224
Chapter 5
Isolation of Murine Postnatal Brain Microglia

for Phenotypic Characterization Using Magnetic
Cell Separation Technology
Ashley S. Harms and Malú G. Tansey
Abstract
To shorten the time between brain harvesting and microglia isolation, and characterization, we utilized the
MACS® neural dissociation kit followed by OctoMACS® CD11b magnetic bead isolation technique to
positively select for brain microglia expressing the pan-microglial marker CD11b, a key subunit of the
membrane attack complex (MAC). This protocol yields a viable and highly pure (>95 %) microglial
population of approximately 500,000 cells per pup that is amenable for in vitro characterization within
hours or days after being harvested from brain tissue. Primary microglia from C57Bl/6 mice were plated
for next-day analyses of morphology and cellular markers by immunocytochemistry or for analysis of gene
expression under resting or LPS-stimulated conditions. The ease of isolation enables investigators to
perform molecular and cellular analyses without having to wait 1–2 weeks to isolate microglia by conven-
tional methods involving mechanical agitation to dislodge these from astrocyte beds.
Key words Microglia, CD11b, Inflammatory gene expression, Flow cytometry, Immunocytochemistry
1 Introduction
Microglia are the monocyte-derived resident macrophages of the

brain in charge of immune surveillance [1]. Activated microglia
secrete neurotrophic factors to limit tissue injury, protect vulnerable
neuronal populations, and aid in brain repair processes. However,
activated microglia can also overproduce prostaglandins, chemokines,
cytokines, and reactive oxygen and nitrogen species which can have
a deleterious effect on neuronal survival by enhancing oxidative
stress and activating cell death pathways (reviewed in ref. 2), rais-
ing the interesting possibility that chronic microglia activation may
contribute to the etiology and/or progression of neurodegenera-
tive disease (reviewed in ref. 3). In vitro analyses of microglia
phenotype and their effector functions are expected to significantly
improve the ability to generate testable hypotheses about the in
vivo role of microglia in normal and pathophysiological states in
33
34 Ashley S. Harms and Malú G. Tansey
the brain. Yet conventional methods used to obtain homogeneous

populations of microglia require maintenance of primary neuron-
glia cultures for extended periods of time (from 1 to 2 weeks),
increasing the likelihood for loss of morphological and functional
microglial phenotype.
2 Materials
2.1 Tissue Harvest 1. Mice at postnatal day 3–5 (P3–P5).

and Primary Cell 2. DMEM/F12 with 10 % Heat-Inactivated FBS.
Culture Components
3. Lipopolysaccharide (LPS E. coli strain O111:B4).
2.2 Magnetic Cell 1. MACS® neural dissociation kit (Miltenyi Biotec).

Isolation Components 2. OctoMACS® CD11b magnetic bead separation (Miltenyi
Biotec).
2.3 Immuno- 1. 4 % paraformaldehyde in 0.01M PBS pH 7.4.

cytochemistry 2. Antibodies for CD45 (Serotec) and CD68 (Serotec).
Components
3. Alexa dye-conjugated secondary antibodies for immunocyto-
chemistry (Invitrogen).
2.4 Components 1. F4/80-Texas Red (Caltag) and CD11b-FITC (Miltenyi Biotec).

for Flow Cytometry 2. Hoechst 33258 (Invitrogen) for nuclear counterstain.
2.5 Components 1. RNA STAT-60 (Tel-Test, Friendswood, TX).

for Quantitative PCR 2. DNase I (Invitrogen).
3. SuperScript II RNase H- Reverse Transcriptase (Invitrogen).
4. SYBR Green PCR Master Mix (Applied Biosystems Inc.).
5. Oligonucleotide forward and reverse PCR primers for TNF,
IL-1β, iNOS, MIP-1α, and CD45 (Integrated DNA
Technologies).
6. ABI PRISM 7000 Detection System (Applied Biosystems Inc.).
3 Methods
3.1 Positive 1. Obtain eight postnatal day 3–5 (P3–P5) C57Bl/6 wild-type
Selection of Primary pups.
Microglia Using CD11b 2. Isolate primary microglia using the MACS® neural dissociation
Magnetic Beads for kit (Miltenyi Biotec).
Next-Day 3. Determine the total cell number by trypan-blue exclusion
Immunocytochemical (~40 million total).
Analyses
Postnatal Mouse Microglia 35
4. To enrich for primary microglia, OctoMACS® CD11b magnetic

bead separation (Miltenyi Biotech) can be employed according
to the protocol provided by the manufacturer. We recommend
shortening the centrifugation time from 10 min to 5 in order
to increase cell viability.
5. The number of cells obtained post-magnetic CD11b bead
isolation will be ~ 4 million or 10 % of the total number of cells
in the single-cell suspension obtained with the MACS neural
dissociation kit, which is in the expected range.
6. After [1] a day in vitro in DMEM/F12 with 10 % HI-FBS, the
cells can be treated with lipopolysaccharide (LPS E. coli strain
O111:B4, 10 ng/mL or 1 μg/mL for 24 h) and then fixed in
4 % paraformaldehyde in 0.01M PBS and processed for
fluorescence immunocytochemistry as previously described
[4] (see Notes 1, 2, and 3).
7. Acquire images of stained cells with a fluorescence microscope
equipped with a digital camera.
3.2 Positive 1. Obtain 10 postnatal day 3–5 (P3–P5) C57Bl/6 wild-type

Selection of Postnatal pups.
Microglia Using CD11b 2. Isolate primary microglia using the MACS® neural dissociation
Magnetic Beads for kit followed by OctoMACS® CD11b magnetic bead positive
Gene Expression selection.
Analyses Induced by 3. Plate cells in 6-well plates at a density of 0.5 million/well.
Inflammatory Stimuli
4. After 1 day in culture in DMEM/F-12 supplemented with
10 % FBS, stimulate cells with lipopolysaccharide (LPS,
1 μg/mL) for 4 h at 37° in a humidified CO2 incubator.
5. Harvest cells in RNA STAT-60 (Tel-Test, Friendswood, TX)
for isolation of total RNA and briefly treat with DNase I
(Invitrogen) and then reverse-transcribe into cDNA using
SuperScript II RNase H- Reverse Transcriptase (Invitrogen)
(see Note 4).
6. Perform real-time quantitative polymerase chain reaction
(qPCR) as previously described [5, 6] using an ABI PRISM
7000 Detection System (Applied Biosystems Inc.) (see Note 5).
3.3 Positive 1. Obtain eight postnatal day 3–5 pups from wild type or TNF
Selection of Postnatal null (or another knockout mouse).
Microglia Using CD11b 2. Isolate primary microglia using MACS® neural dissociation kit
Magnetic Beads for followed by OctoMACS® CD11b magnetic bead separation
Quantitative Flow- (Miltenyi Biotech).
Cytometric Analyses 3. Label cells with a fluorescently conjugated antibodies specific
of Cell-Surface for the pan-marker CD11b and the activation marker F4/80.
Markers
104 104
1.5 10.4 1.03 16.7
103 WT 103 TNF KO

FL3-H: F480 TxRed
FL3-H: F480 TxRed

102 102
101 101
53.1 35 40.5 41.7

100 100
0 1 2 3 4 0 1 2 3
10 10 10 10 10 10 10 10 10 104
FL1-H: Cd11b Fitc FL1-H: Cd11b Fitc
100 100
WT CD11b WT F4/80
80 80
TNF KO CD11b TNF KO F4/80
% of Max
% of Max
60 60
40 40
20 20
0 0
100 101 102 103 104 100 101 102 103 104
FL1-H: Cd1 1b Fitc FL3-H: F480 TxRed
Fig. 1 Primary microglia positively selected with CD11b magnetic bead separation display the expected
microglial cell-surface markers as measured by flow cytometry. Primary microglia were isolated from postna-
tal day 3 (P3) TNF-deficient (TNF KO) and wild-type (WT) pups by MACS® neural dissociation and CD11b
magnetic bead separation. Cells were double-labeled with an antibody against the activation marker F4/80
conjugated to the fluorophore Texas Red (Caltag) and an antibody against the microglial marker CD11b conju-
gated to the fluorophore FITC (Miltenyi Biotec), fixed in 1 % paraformaldehyde and subjected to flow cytometry.
FACS analyses of the live cell population revealed the activation status of the microglia isolated from the two
different genotypes
4. Fix briefly in 1 % PFA in PBS and sort using flow cytometry as

previously described [6].
5. Figure 1 demonstrates a representative scatter plot from
fluorescence-activated cell sorting (FACS) analysis from the
cell population. The data reveals the expected CD11b expres-
sion in the population and relatively low F4/80 expression
indicative of minimal activation; no difference between geno-
types was noted.
4 Notes
1. The antibody dilutions should be as follows: CD45 (Serotec)

1:250, CD68 (Serotec) 1:250, and Hoechst 33258 (Invitrogen)
1:20,000.
2. The appropriate Alexa-conjugated secondary antibodies
(Invitrogen) should be used at a dilution of 1:1,000 in the
dark.
3. Figure 2 illustrates images from a representative experiment
demonstrating that the number of cells found to be immuno-
reactive for CD45 was approximately 99 % and for CD68 was
98 %, indicating that the purity of the isolated microglia popu-
lation is high.
Fig. 2 Primary microglia positively selected with CD11b magnetic beads reveal
normal morphological changes upon LPS stimulation as measured by immuno-
cytochemistry. Primary microglial cells were isolated from postnatal day 3–5
(P3–P5) C57Bl/6 pups using the MACS® neural dissociation kit followed by
CD11b magnetic bead separation. Cells were plated and treated for 24 h with
LPS or TNF, fixed, and stained with anti-CD45 (Serotec) and anti-CD68
(Serotec). Immunocytochemical analysis revealed morphological changes and
increased expression of microglial markers CD45 and CD68 after stimulation
with LPS or TNF
NFκB Target Genes

Relative mRNA Abundance 6000 *** 200 *** Vehicle
180
160 1 ug/mL LPS
140
4000 120
100
*** 10
*** 8
2000
6
4
2
0 0
TNF IL-1β iNOS MIP-1α CD45
Fig. 3 Primary microglia positively selected with CD11b magnetic bead separation are responsive to inflam-
matory stimuli as measured by real-time PCR. Primary microglia were isolated from ten postnatal day 3–5
(P3–P5) C57Bl/6 pups by MACS® neural dissociation followed by CD11b magnetic separation and treated for
4 h with 1 μg/mL LPS. Total RNA was harvested using phenol/chloroform extraction and reverse-transcribed
into cDNA. Quantitative PCR analysis revealed expression of mRNA for iNOS, MIP1α, IL-1β, TNF, and CD45 at
rest and after LPS stimulation
4. All reactions can be done in 384-well format with 50 ng cDNA,

10 μL SYBR Green PCR Master Mix, and 150 nM each
forward and reverse primer. Oligonucleotide primers for TNF,
IL-1β, iNOS, MIP-1α, and CD45 can be designed using
primer design software freely available on the World Wide
Web. Figure 3 illustrates changes in mRNA levels for these
markers in response to LPS stimulation.
5. All reactions should be performed in triplicate and levels of
various mRNAs should be normalized to the geometric mean
of several housekeeping genes such as TATA binding protein,
cyclophilin, and GAPDH.
Acknowledgement
The content of this article was adapted from one that appeared in
the [Date] issue of the Miltenyi Biotec newsletter. This work was
supported by NIH/NINDS grant 5R011NS049433 and The
Michael J. Fox Foundation for Parkinson’s Research.
References
1. Ransohoff RM, Perry VH (2009) Microglial 3. Tansey MG, McCoy MK, Frank-Cannon TC
physiology: unique stimuli, specialized (2007) Neuroinflammatory mechanisms in
responses. Annu Rev Immunol 27:119–145 Parkinson’s disease: potential environmental
2. McGeer PL, McGeer EG (2004) Inflammation triggers, pathways, and targets for early thera-
and neurodegeneration in Parkinson’s disease. peutic intervention. Exp Neurol 208:
Parkinsonism Relat Disord 10(Suppl 1):S3–S7 1–25
4. McCoy MK, Martinez TN, Ruhn KA et al tumor necrosis factor signaling. Glia 60(2):
(2006) Blocking soluble tumor necrosis factor 189–202
signaling with dominant-negative tumor necro- 6. Kurrasch DM, Huang J, Wilkie TM et al (2004)
sis factor inhibitor attenuates loss of dopami- Quantitative real-time polymerase chain reac-
nergic neurons in models of Parkinson’s disease. tion measurement of regulators of G-protein
J Neurosci 26:9365–9375 signaling mRNA levels in mouse tissues.
5. Harms AS, Lee JK, Nguyen TA et al (2012) Methods Enzymol 389:3–15
Regulation of microglia effector functions by
Chapter 6
Isolation and Culture of Adult Human Microglia Within

Mixed Glial Cultures for Functional Experimentation
and High-Content Analysis
Amy M. Smith, Hannah M. Gibbons, Claire Lill,
Richard L.M. Faull, and Mike Dragunow
Abstract
Microglia are thought to be involved in diseases of the adult human brain as well as normal aging processes.
While neonatal and rodent microglia are often used in studies investigating microglial function, there are
important differences between rodent microglia and their adult human counterparts. Human brain tissue
provides a unique and valuable tool for microglial cell and molecular biology. Routine protocols can now
enable use of this culture method in many laboratories. Detailed protocols and advice for culture of human
brain microglia are provided here. We demonstrate the protocol for culturing human adult microglia
within a mixed glial culture and use a phagocytosis assay as an example of the functional studies possible
with these cells as well as a high-content analysis method of quantification.
Key words Human brain tissue, Primary human culture, Mixed glial cultures, Phagocytosis, High-
content analysis
1 Introduction
Microglia are specialized immune cells that are prevalent through-

out the brain. They maintain homeostasis and respond sensitively to
changes in their microenvironment [1]. Microglia influence many
brain processes, from neurogenesis to learning. There are numer-
ous reports of microglial activity in human brain disorders, yet we
still lack knowledge about the basic molecular mechanisms that
govern different activation states and phenotypes of microglia.
Although most microglial research has been carried out using
rodent cells and models, it is becoming increasingly clear that there
are important differences between rodent microglia and their
human counterpart [2, 3]. While neonatal and rodent microglia
are often used in studies investigating microglial function, more
41
42 Amy M. Smith et al.
research is required to discover which microglial processes and

hallmarks are conserved in the adult human brain and where they
are divergent from rodent microglia. There is also evidence that
immune responsiveness changes with age [4, 5]. Furthermore,
microglia may respond differently to peripheral immune cells due
to their brain-specific environment.
Thus, isolation and culture of human adult microglia is a
unique and valuable tool to complement other approaches to
microglial research. Primary adult human microglia are a relevant
tool for studying human adult brain disorders due to their tissue
and species origin. Cultured primary human microglia can be used
with relative ease for a range of studies including functional assays,
drug screening and development, and gene profiling. The use of
human adult brain cells as a tool for research can play a major role
in conjunction with in vivo mechanistic proof-of-principle studies
and play an essential role in translation of microglial-based thera-
pies to the clinic.
We culture microglia within a mixed glial culture also contain-
ing astrocytes and fibroblast-like cells (the benefits of which are
discussed in Subheading 4) [6]. The high yield of microglia
obtained from this protocol is ideal for high-throughput analysis
when used in conjunction with 96-well microplates and high-
content imaging methods [7]. A variety of characterization and
functional studies are possible with these primary human cells. We
provide a protocol for examining the quintessential microglial
function of phagocytosis. The ability of microglia to phagocytose
extracellular debris is important throughout life from development
to neurodegeneration. This reliable assay can also be used in con-
junction with other treatments. We have developed a high-content
image analysis protocol to rapidly and reliably quantify microglial
phagocytosis. High-content analysis is a useful tool for objective,
quick, and accurate examination of human microglial biology [8].
For a review of the different culturing protocols for human
microglia in the literature, see ref. 9. The protocol outlined here is
the one routinely used in our laboratory.
2 Materials
Human Adult Brain Tissue

For our experiments, autopsy adult human brain tissue is obtained
from the Neurological Foundation of New Zealand Human
Brain Bank. Biopsy adult human brain tissue is obtained from
patients undergoing surgery for intractable temporal lobe epi-
lepsy. This research is approved by the Northern Regional
Ethics Committee and the University of Auckland Human
Participants Ethics Committee, and informed consent was
obtained from all tissue donors.
Adult Human Microglia for High Content Analysis 43
Ethics approval must be obtained from the relevant committees

prior to tissue collection and informed consent must be
obtained from all donors.
2.1 Isolation and Reagents

Culture of Human
1. Hank’s balanced salt solution (HBSS; Ca2+ and Mg2+ free,
Adult Mixed Glia
Gibco BRL).
Including Microglia
2. Earle’s balanced salt solution (EBSS; Ca2+ and Mg2+ free, Gibco
BRL).
3. Hibernate A (Gibco BRL) (see Note 1).
4. DMEM/F12 media (Gibco BRL) with 10 % Fetal Bovine
Serum (FBS) (Gibco BRL), 1 % Penicillin-Streptomycin-
Glutamine (Gibco BRL) (final concentrations: penicillin
(100 U/ml), streptomycin (100 μg/ml), and l-glutamine
(0.29 mg/ml)).
5. DNase (Invitrogen), 1,000 U/ml stock made up in HBSS
(can be stored at −20 °C).
6. Papain (Worthington Biochemical Corporation), 20 U/ml
stock made fresh in EBSS.
Equipment
1. Ice bucket.
2. 50 ml tube rack.
3. 50 ml tubes (BD Biosciences).
4. 2 × 10 cm dishes (Nunc, Roskilde, Denmark).
5. 2× autoclave-sterilized scalpel handles (Swann-Morton,
Sheffield, England).
6. 2× autoclave-sterilized forceps (ProSciTech, Queensland,
Australia).
7. 2× scalpel blades (Swann-Morton, Sheffield, England).
8. Waste beaker containing TriGene (Medichem International, UK).
9. 3× T75 flasks (Nunc).
10. 2 × 100 μm filters (BD Biosciences, New Jersey, USA).
11. 10 ml strip pipettes.
12. Pipette gun.
Plating cells for experiments
1. Phosphate-buffered saline (PBS).
2. 0.25 % trypsin/1 mM EDTA (Gibco BRL).
3. Cell scrapers (1 per flask) (BD Biosciences).
4. Hemocytometer.
5. 96-well tissue culture plate (Nunc).
2.2 Phagocytosis Reagents

Assay for Primary
1. Amyloid-β1–42 amino acid peptide (Aβ1–42; Bachem, Bubendorf,
Human Adult Microglia
Switzerland).
in a Mixed Glial
Culture 2. Thioflavin S (Sigma).
3. Hoechst 33258 (Sigma).
4. Analytical grade ethanol.
6. 4 % Paraformaldehyde solution (PFA).
7. PBS-Triton 0.2 % (v/v) (PBS-T).
8. TNE buffer (10 mM Tris, 200 mM NaCl, 1 mM EDTA;
pH 7.4).
Equipment
1. Aspirator/vacuum pump (a multichannel pipette can be used
to remove media/PBS from the wells if an aspirator/vacuum
pump is unavailable).
2. Plate rocker.
3. Multichannel pipette.
2.3 High-Content Equipment

Analysis of Primary
1. Discovery-1 Automated Fluorescence Microscope (Molecular
Adult Human
Devices, Sunnyvale, CA, USA).
Microglial
Phagocytosis 2. MetaMorph Image Analysis System (6.2.6 software, Molecular
Devices).
3 Methods
3.1 Isolation and 1. Prepare suitable biological safety hood (for safe and sterile han-
Culture of Human dling of samples, see Note 2), equipment, and reagents prior to
Adult Mixed Glia collecting tissue. In a sterile tissue culture hood, place 1 × 50 ml
Including Microglia tube rack, 2 × 10 cm dishes, 2× autoclave-sterilized scalpel han-
dles, 2× autoclave-sterilized forceps, 2× scalpel blades, waste
container with TriGene, 3× T75 flasks, and 2 × 100 μm filters.
2. In 37 °C water bath, pre-warm Hibernate A and DMEM/F12
media with 10 % FBS and 1 % PSG (complete media).
3. For tissue collection, add 20 ml Hibernate A to a 50 ml tube
and weigh contents; store on ice.
4. Thaw DNase and allow papain to come to room temperature.
5. Prepare papain by adding 5 ml EBSS to vial and invert to mix
(20 units of papain per ml).
6. Collect ~2 g white and gray matter from the middle temporal
gyrus (see Note 3) in ice-cold Hibernate A.
7. Weigh tube containing tissue and calculate weight of tissue by

subtracting previously weighed mass of tube and Hibernate A.
8. Wash tissue twice in Hibernate A to remove contaminating
blood. Carefully tip waste into waste container with TriGene
(see Note 4).
9. Place tissue in a sterile Petri dish and remove meninges and
visible blood vessels aseptically using autoclave-sterilized
forceps.
10. Place tissue into a new 50 ml falcon tube.
11. Wash tissue twice to remove contaminating blood and care-
fully tip waste into waste container with TriGene.
12. Place tissue in a sterile Petri dish and dice tissue into ~1 mm3
pieces using sterile scalpel blades. Dispose of scalpel blades
using a safety blade remover device.
13. Place tissue in a tube containing 50 ml Hibernate A and allow
tissue to sink to the bottom of the tube.
14. Prepare enzyme dissociation mix (10 ml/g tissue) with pre-
warmed 37 °C Hibernate A containing 2.5 U/ml papain and
10 U/ml DNase.
15. Use a 10 ml strip pipette to remove Hibernate A from tube
containing tissue and replace with 10 ml warm enzyme disso-
ciation mix per g tissue.
16. Incubate tissue with enzymes for 15 min at 37 °C with agita-
tion (e.g., MACSmix Tube Rotator (Miltenyi Biotec, Bergisch
Gladbach, Germany) on continuous full speed in incubator, or
place tubes in water bath and regularly invert).
17. Remove from the incubator and triturate tissue ~5× through a
10 ml strip pipette to aid digestion.
18. Return to incubator for a further 15 min (as above).
19. Triturate well through a 10 ml strip pipette (~5× until resis-
tance decreases).
20. Add an equal volume of complete media (DMEM/F12 with
10 % FBS, 1 % PSG) and gently triturate the tissue ~5× more
using a 10 ml strip pipette.
21. Using a 10 ml strip pipette, pass the cell suspension through a
100 μm nylon cell strainer into a new 50 ml tube.
22. Centrifuge for 10 min at 160 × g.
23. Carefully pipette off the supernatant and resuspend the pellet
in 15 ml media.
24. Plate the cell suspension into 75 cm2 tissue culture flasks. 2 g
should be split between 3× T75 flasks for biopsy or short
(up to 3 h) postmortem delay tissue (see Note 5).
25. Add 5 ml cell suspension to each of 3× 75 cm2 tissue culture

flasks. Add 20 ml warm, complete media to each flask for a
final volume of 25 ml per flask.
26. Incubate cells at 37 °C in a humidified atmosphere with 95 %
air/5 % CO2.
27. First thing the next morning (~16 h later), refresh the media.
28. For each flask, remove culture media containing debris and
unattached cells and transfer to a 50 ml tube. Wash the adhered
cells carefully with 10 ml fresh media to remove debris and
transfer to the same 50 ml tube.
29. Add 10 ml complete media to the flask of attached cells and
incubate at 37 °C. Centrifuge 50 ml tube containing debris
and unattached cells for 10 min at 160 × g.
30. Carefully pipette off the supernatant and resuspend the pellet
in 15 ml medium. Replate onto the adhered cells (for a total of
25 ml per flask) and incubate for a further 24 h.
31. The following day, remove culture media containing debris
and unattached cells and discard. Wash the adherent cells care-
fully 2× with 5 ml complete media to remove debris and add
20 ml fresh complete media to each flask.
32. Culture cells for ~1–2 weeks, with weekly media changes,
depending on density and use of cells (see Notes 6 and 7).
To detach and plate cells for experiments
1. Remove media from flask.
2. Rinse cells with warm PBS.
3. Detach cells with 0.25 % trypsin/1 mM EDTA (~2 ml per
flask). Incubate at 37 °C for 5 min.
4. Tap the side of flask to dislodge cells and observe cells under a
microscope. If cells are still adherent, return to incubator for
5 min.
5. Tap the side of flask to dislodge cells. Add 5 ml complete media
to flask and transfer cells and media to a 50 ml tube. Repeat.
6. Observe flask surface under a microscope. If some cells are still
adherent, a cell scraper can be used to detach remaining cells
(see Note 8). Collect scraped cells by rinsing with 5 ml media
and transfer to the same 50 ml tube.
7. Centrifuge for 10 min at 160 × g.
8. Carefully pipette off the supernatant, resuspend the pellet in
medium, and count cells using a hemocytometer.
9. Plate cells by resuspending in medium at 50,000 cells/ml and
plate into 96-well plates (100 μl/well) for experiments.
10. Allow cells to settle and recover from trypsinization for at least
48 h before beginning experiments (see Notes 9 and 10).
3.2 Phagocytosis 1. Dissolve Aβ1–42 (1 mg) in distilled H2O (444 μl) to a concen-
Assay for Primary tration of 500 μM (Aβ1–42 solution can be stored at −20 °C).
Human Adult Microglia 2. Vortex Aβ1–42 solution to break up large crystals.
in a Mixed Glial
3. Prepare working stock of Aβ1–42 by adding to an equal volume
Culture
of PBS, i.e., dilute 1:1 in PBS immediately prior to addition to
cells.
4. In a suitable biological safety cabinet, add 5 μM Aβ1–42 to the
cells (see Note 11), i.e., from a stock concentration of 500 μM,
diluted 1:1 with PBS, add 2 μl per well (containing 100 μl
media).
5. Incubate for 24 h in normal incubating conditions of 37 °C in
5 % CO2 (see Note 12).
6. Wash cells twice with warm PBS to remove extracellular Aβ1–42
using an aspirator and multichannel pipette.
7. Fix cells with 4 % PFA for 15 min at room temperature.
8. Wash with PBS-T for 10 min.
Thioflavin S is used to visualize phagocytosed Aβ1–42
1. Prepare a 0.01 % solution of Thioflavin S in 50 % ethanol
(e.g., 1 mg Thioflavin S in 10 ml 50 % ethanol solution)
(see Note 13).
2. Add 50 μl per well of 96-well plate.
3. Incubate with gentle rocking for 10 min at room temperature
in the dark.
4. Wash cells in 50 % ethanol, followed by distilled H2O, for
10 min each.
5. Add 60 μl PBS-T per well.
6. Visualize by fluorescence microscopy using a fluorescein iso-
thiocyanate (FITC) filter (see Note 14).
Hoechst 33258 is used to counterstain cell nuclei (see Note 15)
1. Wash cells for 5 min with TNE buffer containing 10 mM Tris,
200 mM NaCl, 1 mM EDTA; pH 7.4.
2. Prepare a 20 μM Hoechst 33258 solution in TNE buffer.
3. Add 50 μl per well of 96-well plate.
4. Incubate with gentle rocking for 30 min at room temperature
in the dark.
5. Wash cells 2 × 5 min with TNE buffer.
6. Add 60 μl PBS-T per well.
7. Visualize under UV light (Fig. 1b) (see Note 16).
Fig. 1 Adult human microglia in mixed glial cultures can phagocytose Aβ1–42 and be quantified by high-content
analysis. (a) Primary adult human microglia express the transcription factor PU.1 (pink ) and cell-surface
marker CD45 (green). All glial nuclei are labelled with Hoechst (blue). (b) Adult human microglia phagocytose
Aβ1–42 (green, stained with Thioflavin S) in vitro. (c) Microglial phagocytosis can be automatically quantified
using the “Count Nuclei” application in MetaMorph image analysis software. The corresponding FITC image of
Aβ1–42 from (b) has been thresholded to count the number of phagocytic microglia. Images were acquired using
a Discovery-1 automated fluorescence microscope. Scale bar = 100 μm
3.3 High-Content Images of labelled cells in a 96-well plate format can be acquired by a
Analysis of Primary high-throughput system such as a Discovery-1 microscope
Adult Human
1. Using a 10× objective, take images of 9 sites per well (3 × 3) for
Microglial high-content analysis.
Phagocytosis
2. Two (or three) wavelengths can be imaged together for the
same site. For Thioflavin S-stained Aβ1–42, use a FITC filter set
(470Ex/535Em) and an exposure time of approximately
200 ms (see Note 17).
3. For corresponding fluorescent images of Hoechst-stained cell
nuclei, use a UV filter set (403Ex/465Em) and an exposure
time of approximately 500 ms.
The images can then be processed with image analysis software such as
MetaMorph
4. “Count Nuclei” is a simple application that identifies and iso-
lates cell nuclei, and similarly shaped objects, through image
segmentation (see Notes 18–20).
5. To measure total cell number from Hoechst-stained nuclei:
6. Open the “Count Nuclei” application.
7. Adjust the intensity threshold to segment the nuclei from
background.
8. Adjust the cell size thresholds (approximate minimum width
and approximate maximum width) to segment individual
whole nuclei.
9. Apply “Count Nuclei” analysis to all images to count the num-
ber of nuclei per image.
10. Results are logged automatically to Microsoft Excel spread-

sheets for easy analysis.
To count the number of microglia that have phagocytosed Thioflavin
S-stained Aβ1–42:
11. Open the “Count Nuclei” application.
12. Adjust the intensity threshold to segment the Thioflavin
S-positive cells from background.
13. Adjust the cell size thresholds (approximate minimum width
and approximate maximum width) to segment individual
whole cells and exclude extracellular Thioflavin S-stained
Aβ1–42 blobs.
14. Apply “Count Nuclei” analysis to all images to count the num-
ber of fluorescent cells per image (Fig. 1c).
15. Results are logged automatically to Microsoft Excel spread-
sheets for easy analysis.
4 Notes
1. Hank’s balanced salt solution (HBSS; Ca2+ and Mg2+ free,

Gibco BRL) can be used in place of Hibernate A for tissue col-
lection, washes, and enzymatic dissociation.
2. The highest health and safety practices need to be adhered to
when using human cells and tissue. Assume that all tissue is
potentially contaminated and work accordingly. All manipula-
tion of cells/tissues should be undertaken in a suitable biologi-
cal safety cabinet/hood with surgical gloving at all times. All
staff should also be immunized against Hepatitis B as a precau-
tionary measure.
3. In our laboratory, we routinely culture from middle temporal
gyrus tissue, but other brain regions can also be used for isola-
tion of human adult microglia. Both white and gray matter
give rise to viable microglia.
4. Carefully discard all waste that has been in contact with human
tissue by treating with TriGene.
5. If there is reason to believe that fewer cells may be obtained
from a particular culture (e.g., long postmortem delay), cell
suspension can be plated into fewer flasks, i.e., 1 or 2 × 75 cm2
flasks for 2 g tissue.
6. Changes in the proportions of cell types in mixed glial cultures
occur over time due to proliferation of fibroblast-like cells and
very limited proliferation of astrocytes and microglia (see ref. 6).
This should be taken into account for timing of microglial
experiments.
7. Some other protocols for microglial cultures use the mitogens

macrophage colony-stimulating factor (M-CSF) and
granulocyte-macrophage colony-stimulating factor (GM-CSF).
We do not routinely use these mitogens in our cultures as they
may prime/activate the microglia towards a particular pheno-
type (unpublished observations).
8. Adult human microglia are very adherent cells (more so than
astrocytes or fibroblast-like cells) and do not always detach
from the flask with trypsinization. Using a cell scraper does not
appear to lead to increased cell death and can increase the
microglia yield from the culture.
9. A heterogeneous cell population in mixed glial cultures mimics
the in vivo environment better than a single cell population.
While being able to characterize and quantify microglia-specific
functions (Fig. 1; see phagocytosis assay below), it is also pos-
sible to assess how microglia interact with other glial cell types.
10. As each primary mixed glial culture comes from a different
donor, each culture of cells will be different (e.g., number of
cells isolated, cell composition of culture). Variability in the
isolation protocol should be controlled as much as possible,
but the heterogeneity of primary human cultures reflects the
heterogeneity of individual humans.
11. The final concentration of 5 μM Aβ1–42 added to cells is in
excess of what they require for phagocytosis and can be
adjusted if necessary.
12. The Aβ1–42 incubation time may need to be adjusted to suit the
cells’ phagocytic ability and the test conditions.
13. Thioflavin S is best made fresh before each staining.
14. Thioflavin S may also be visualized through other filters, but a
FITC filter gives the strongest signal.
15. Under these culture conditions, phagocytosis of Aβ1–42 is rarely
seen by cells other than microglia. However, immunocyto-
chemistry can be used to identify microglia using antibodies to
microglia-specific antigens, e.g., PU.1 and CD45 [10].
16. Internalization of Aβ1–42 can be confirmed by confocal micros-
copy [11].
17. Exposure times need to be adjusted to acquire the best images
for each experiment but must be kept constant between
conditions.
18. Count Nuclei application can also be used to quantify other
aspects of microglial biology such as immuno-positive markers.
19. A range of assays are available to analyze other end points such
as morphology, co-localization, and intensity of staining.
20. For a review of high-content analysis applications to neuroscience
and microglial biology, refer to [8].
Acknowledgments
We are very grateful to the tissue donors and their families for their
generous and precious gift of brain tissue for our research into
brain disorders. We also thank specialist epilepsy nurse Lynair
Roberts, neurologist Dr. Peter Bergin, neurosurgeon Dr. Edward
Mee, and pathologist Dr. Robyn Oldfield at Auckland City Hospital
for providing biopsy tissue. We thank the staff of the Neurological
Foundation of New Zealand Human Brain Bank and Centre for
Brain Research Biobank for technical assistance. This protocol has
been optimized with the help of funding from the National
Research Centre for Growth and Development, Coker Charitable
Trust, Hugh Green Foundation, and the Health Research Council
of New Zealand (Program Grant).
References
1. Hanisch U-K, Kettenmann H (2007) 7. Narayan P, Gibbons H, Mee E et al (2007)
Microglia: active sensor and versatile effector High throughput quantification of cells with
cells in the normal and pathologic brain. Nat complex morphology in mixed cultures.
Neurosci 10(11):1387–1394 J Neurosci Methods 164:339–349
2. Davis MM (2012) Immunology Taught by 8. Dragunow M (2008) High-content analysis in
Humans. Sci Transl Med 4(117) neuroscience. Nat Rev Neurosci 9(10):
3. Dragunow M (2008) The adult human brain 779–788
in preclinical drug development. Nat Rev 7: 9. Gibbons HM, Dragunow M (2010) Adult
659–666 human brain cell culture for neuroscience
4. Lynch AM, Murphy KJ, Deighan BF et al research. Int J Biochem Cell Biol 42(6):
(2010) The impact of glial activation in the 844–856
aging brain. Aging Dis 1(3):262–278 10. Gibbons HM, Smith AM, Teoh HH et al
5. Streit WJ (2006) Microglial senescence: does (2011) Valproic acid induces microglial dys-
the brain’s immune system have an expiration function, not apoptosis, in human glial cul-
date? Trends Neurosci 29(9):506–510 tures. Neurobiol Dis 41(1):96–103
6. Gibbons HM, Hughes SM, Van Roon-Mom 11. Smith AM, Gibbons HM, Dragunow M
W et al (2007) Cellular composition of human (2010) Valproic acid enhances microglial
glial cultures from adult biopsy brain tissue. phagocytosis of amyloid-b1-42. Neuroscience
J Neurosci Methods 166(1):89–98 169(1):505–515
Part III
Depletion and Transduction of Microglia

Chapter 7
Depletion of Microglia from Primary Cellular Cultures

Lorena Pont-Lezica, Sabrina Colasse, and Alain Bessis
Abstract
Primary cultures are an important in vitro tool to study cellular processes and interactions. These cultures
are complex systems, composed of many cell types, including neurons, astrocytes, oligodendrocytes,
microglia, NG2 cells, and endothelial cells. For some studies it is necessary to be able to study a pure
culture of one cell type, or eliminate a particular cell type, to better understand its function. There exist
cell culture protocols for making pure astrocyte or microglia cultures. Here we present two protocols to
produce cultures depleted for microglia: in the first case, from a mixed astrocyte–microglia culture and, in
the second, for eliminating microglia from neuronal cultures.
Key words Microglia depletion, LME, Conjugated Saporin, Pure primary astrocyte culture, Primary
neuronal culture
1 Introduction
Brain function is more than just signaling between neurons.

Neuroglia constitute more than half of the cells of the vertebrate
adult brain [1] and contribute to many functions of the central
nervous system (CNS) including neurotransmission and synapto-
genesis, homeostasis of the blood–brain barrier, and immune
surveillance (refs within [2]). To get a complete picture of how the
brain functions, it is thus important to understand the contribu-
tion of the different cell types. Simplified primary cultures of
neurons, astrocytes, and microglia from neonate animals have
allowed efficient dissection of several complex cell–cell interactions
[3–6]. One of the major caveats of primary cultures, however, is
that the main cell population can be contaminated by a small num-
ber of cells of other cell types.
Astrocyte-enriched cultures are usually prepared from brains
dissected from neonate mice. Microglia invade the brain at early
stages of development, and so, because the brains of neonates con-
tain microglia, astrocyte cultures are consistently contaminated by
these myeloid cells. In initial protocols, microglia were routinely
55
56 Lorena Pont-Lezica et al.
removed by shaking [7]. Treatment with cytosine β-D-

arabinofuranoside hydrochloride (AraC) was used to further
deplete microglia, leading to >95 % pure astrocyte cultures [8, 9].
Such astrocyte-enriched primary cultures have been extensively
used for several years. However, it recently appeared that despite
the fact that microglia make up only 2–5 % of the cell population
[2, 10], their presence strongly biased the interpretation of results
obtained with these cultures. For instance, NOS2 and TLR4
proteins were detected in astrocyte cultures and their expression was
initially wrongly attributed to astrocytes ([11, 12] see also ref. 13)
until it was shown that they were expressed solely by contaminat-
ing microglia [2, 4, 14]. Similarly, work using conditioned medium
from astrocyte-enriched cultures showed that TNF-α was impor-
tant for synaptic scaling [3, 15], but the cellular source of TNF-α
is still controversial [16]. These examples show that it is difficult to
tease apart the contributions of each cell type when the cultures are
not pure, and thus one must use caution when interpreting results
in studies using non-purified cell cultures.
Various techniques have been used to eliminate microglia,
including repeated mechanical depletion (shaking [8]), genetic
depletion [8, 17], pharmacological depletion with L-leucine methyl
ester (LME) [9] or clodronate [18], and targeted immunotoxin
depletion [19, 20].
Combining mechanical and pharmacological depletion has
proven to be an efficient method to get rid of all microglia [4, 9].
LME is a lysomotropic compound that traverses cell and organ-
elle membranes to accumulate in the lysosome. Once within this
compartment, LME is hydrolyzed by serine or cysteine proteases
to produce free L-leucine [21], and the accumulation of this
L -amino acid brings about osmotic swelling and rupture of
lysosomes [21, 22]. The LME protocol appears to be highly
specific for microglia [9, 10].
The key advantage of using LME to deplete primary astrocyte
cultures of microglia is the low cost of the reagent and its specific-
ity for microglia. A possible disadvantage is that, although no obvi-
ous deleterious effects have been reported, it is unclear what effect
LME might have on the function of other cell types.
The impact of contaminating microglia on neuronal cultures
has not been evaluated. However, some of our unpublished data
show that, similarly to astrocyte cultures, caution should also be
exercised. Since the effect of LME on neurons has not been stud-
ied, we adapted another, more specific, strategy to deplete microg-
lia from primary cultures of neurons. Saporin is an unusually stable,
ribosome-inactivating protein (RIP) found in the seeds of
Saponaria officinalis. RIPs enzymatically inactivate ribosomes
using their N-glycosidase activity, which removes a single adenine
base from the rRNA of the large ribosome subunit, thus effectively
shutting down protein synthesis and eventually leading to cell
Microglia Depletion 57
death. Specific targeting and internalization of Saporin have been

obtained by conjugating Saporin to antibodies or other molecules
that recognize proteins on the surface of the targeted cell [23]. For
the elimination of microglia, Saporin can be commercially found
conjugated to isolectin B4 or CD11b antibodies. The key advan-
tage of conjugated Saporin is its efficacy (ED50 in the low picomo-
lar range), the possibility of very specifically targeting a particular
cell type, and its speed of action. Despite the fact that the targeted
cytotoxin can be used at such low concentrations, the main disad-
vantage of this system is its high cost.
2 Materials
All work must be done in a hood, under sterile cell culture

conditions.
1. Stock solution of conjugated Saporin/conj Saporin (Advanced
Targeting Systems): dilute to 0.1 μg/μl and store aliquots of
stock solution at −80 °C for long-term storage, or at −20 °C
for no more than 2 months. If done under sterile conditions,
no need to sterilize solutions by filtration.
2. Stock solution cytosine β-D-arabinofuranoside hydrochloride/
AraC: make a concentrated stock solution at 100 mM with sterile
water and store as 250 μl aliquots at −20 °C. The concentrated
stock solution is then diluted in sterile water to make a working
solution at 10 mM and stored as 250 μl aliquots at −20 °C.
3. Stock solution of L-leucine methyl ester/LME: stock solution
at 500 mM in DMEMc, stored as 2 ml aliquots at −20 °C for
no more than a month.
4. Astrocyte culture medium/DMEMc: DMEM supplemented
with 10 % Fetal Bovine Serum, 1 % glutamine, 1 % sodium
pyruvate, and 0.1 % antibiotics (Invitrogen). Sterilize solutions
by filtration.
5. Neuron culture medium/NBc: Neurobasal medium supple-
mented with B27, 1 % L-GLUTAMINE (stock conc 200 mM), and
0.05 % antibiotics. Sterilize solutions by filtration.
3 Methods
3.1 Protocol 1: Confluent primary astrocytes cultured in 10 cm Petri dishes are

Treatment with AraC first treated with the antimitotic agent AraC for 5 days to eliminate
and LME replicating microglia. At the end of the AraC treatment, cultures
may still contain microglia embedded in the astrocyte layer. To
remove these, we carry out an additional treatment with the lyso-
somotropic agent LME (see Note 1). Primary astrocyte cultures
become confluent about 7–10 days after culturing.
Fig. 1 Effect of AraC and LME treatment on a confluent culture of astrocytes. (a) After 13 days in vitro (DIV)
including 5 days of AraC treatment, the astrocytes are confluent and microglia are visible as round, refractive
cells (arrows). (b) After 90 min treatment with LME, microglia are not visible anymore; astrocytes are more
refractive and their morphology has changed. (c) After 24 h, the monolayer of astrocytes has recovered its
initial paved-like appearance and is free from microglia
1. AraC treatment.
(a) Prepare fresh DMEMc culture medium and add AraC
from 10 mM working stock to a final concentration of
5 μM. Sterilize solutions by filtration.
(b) Warm to 37 °C, then replace old medium with AraC-
containing DMEMc medium.
(c) Change medium every day for 5 days with fresh AraC-
containing DMEMc medium.
2. LME treatment.
(a) Dilute LME to 75 mM in DMEMc culture medium.
Sterilize solutions by filtration.
(b) Adjust pH to 7.5 by adding 1N NaOH.
(c) Warm to 37 °C, then replace astrocyte medium with
LME-containing DMEMc medium.
(d) Put cultures back into the incubator and allow LME to act
for 1h30.
(e) Replace LME-containing DMEMc medium with standard
DMEMc culture medium and wait 24 h before using the
now pure astrocyte culture (see Note 2 and Fig. 1).
3.2 Protocol 2: Conjugated Saporin can be used to deplete microglia from primary
Treatment with neuronal cultures at any stage of the culture (see Note 3). Although
Conjugated Saporin the ED50 of Saporin is in the hundreds of nanomolar range, the
conjugated versions of the cytotoxin are four orders of magnitude
more sensitive. The exact ED50 is tested by Advanced Targeting
Systems for each lot (see Notes 4 and 5).
1. Conjugated Saporin treatment.
Dilute stock solution of conjugated Saporin in NBc medium to
a final concentration of 0.35 μg/ml. We use 7 μl of a
0.1 μg/μl stock solution for 2 ml of NBc medium in a 35 mm
Petri dish containing five 12 mm coverslips (see Notes 6 and 7).
We allow the conj Saporin to act for 72 h before using the

neuronal cultures. No additional medium changes are necessary
unless prescribed for the neurons.
4 Notes
1. A similar protocol was recently published [24]. The key differ-

ence is that in our protocol, AraC-containing medium is
changed daily over 5 days and that LME is added for no less
and no more than 90 min. This seems to result in a much more
effective depletion of microglia (compare RT-QPCR results
[4] and Western blots in [24]).
2. The morphology of astrocytes will change during LME treat-
ment but will return to normal afterwards (Fig. 1).
3. Isolectin B4 from Griffonia simplicifolia conjugated to Saporin
targets cells expressing the α-D-galactopyranoside group on
their cell surface. This includes activated macrophages [25],
microglia [26], as well as endothelial cells [27] and dorsal root
ganglion neurons [28]. Thus, if working with spinal cord cul-
tures, it would be more appropriate to use another form of
conjugated Saporin, such as CD11b-Saporin.
4. Do not use reducing agents (DTT, β-mercaptoethanol, ascor-
bic acid) in conjunction with Saporin as these will inactivate
the toxin.
5. Conjugated Saporin is an extremely potent cytotoxin and
should be handled with appropriate caution (gloves and safety
glasses). Saporin can be inactivated by autoclaving or exposure
to 0.2M NaOH. All labware that has come into contact with
the cytotoxin should be cleaned in this way. For further details,
see information from the Advanced Targeting Systems infor-
mation sheet.
6. If adding conj Saporin to neurons after seeding, dilute into
fresh NBc medium. If adding to an existing neuronal culture,
add conj Saporin directly into culture. Careful! Neurons are
very sensitive to changes in their culture medium: never change
all of the medium at once; rather change half or a third of the
medium and then add the conj Saporin to the culture.
7. We tested various concentrations (0.1, 0.2, and 0.35 μg/ml)
of IB4-Saporin and found that 0.35 μg/ml is sufficient to
deplete all microglia from a mixed neuronal culture within
24–48 h, for up to 6 days. We have not examined the presence
of microglia at later time points. Furthermore, although this
concentration has worked effectively in our hands for isolectin
B4-Saporin and CD11b-Saporin, effective concentrations for
other Saporin conjugates need to be determined.
Acknowledgments
We thank Sarrah Ben Achour for preliminary experiments and

thank Yasmine Belarif-Cantaut for the helpful discussion with the
Saporin protocol. The work was supported by a grant from the
seventh Framework Program Moodinflame (222963).
References
1. Pelvig DP, Pakkenberg H, Stark AK et al 12. Alfonso-Loeches S, Pascual-Lucas M, Blanco
(2008) Neocortical glial cell numbers in AM et al (2010) Pivotal role of TLR4 recep-
human brains. Neurobiol Aging 29(11): tors in alcohol-induced neuroinflammation
1754–1762 and brain damage. J Neurosci 30(24):
2. Saura J (2007) Microglial cells in astroglial cul- 8285–8295
tures: a cautionary note. J Neuroinflammation 13. Hewett JA, Hewett SJ, Winkler S et al (1999)
4:26 Inducible nitric oxide synthase expression in
3. Stellwagen D, Malenka RC (2006) Synaptic cultures enriched for mature oligodendrocytes
scaling mediated by glial TNF-alpha. Nature is due to microglia. J Neurosci Res 56(2):
440(7087):1054–1059 189–198
4. Pascual O, Ben Achour S, Rostaing P et al 14. Sola C, Casal C, Tusell JM et al (2002)
(2012) Microglia activation triggers astrocyte- Astrocytes enhance lipopolysaccharide-
mediated modulation of excitatory neuro- induced nitric oxide production by microglial
transmission. Proc Natl Acad Sci USA 109(4): cells. Eur J Neurosci 16(7):1275–1283
E197–205 15. Beattie EC, Stellwagen D, Morishita W et al
5. Saijo K, Winner B, Carson CT et al (2009) A (2002) Control of synaptic strength by glial
Nurr1/CoREST pathway in microglia and TNFalpha. Science 295(5563):2282–2285
astrocytes protects dopaminergic neurons 16. Barres BA (2008) The mystery and magic of
from inflammation-induced death. Cell glia: a perspective on their roles in health and
137(1):47–59 disease. Neuron 60(3):430–440
6. Bezzi P, Domercq M, Brambilla L et al (2001) 17. Burnett SH, Kershen EJ, Zhang J et al (2004)
CXCR4-activated astrocyte glutamate release Conditional macrophage ablation in trans-
via TNFalpha: amplification by microglia trig- genic mice expressing a Fas-based suicide
gers neurotoxicity. Nat Neurosci gene. J Leukoc Biol 75(4):612–623
4(7):702–710 18. Marin-Teva JL, Dusart I, Colin C, Gervais A
7. McCarthy KD, de Vellis J (1980) Preparation et al (2004) Microglia promote the death of
of separate astroglial and oligodendroglial cell developing Purkinje cells. Neuron
cultures from rat cerebral tissue. J Cell Biol 41(4):535–547
85(3):890–902 19. Montero Dominguez M, Gonzalez B, Zimmer
8. Holm TH, Draeby D, Owens T (2012) J (2009) Neuroprotective effects of the anti-
Microglia are required for astroglial Toll-like inflammatory compound triflusal on ischemia-
receptor 4 response and for optimal TLR2 and like neurodegeneration in mouse hippocampal
TLR3 response. Glia 60(4):630–638 slice cultures occur independent of microglia.
9. Hamby ME, Uliasz TF, Hewett SJ et al (2006) Exp Neurol 218(1):11–23
Characterization of an improved procedure for 20. Montero M, Gonzalez B, Zimmer J (2009)
the removal of microglia from confluent Immunotoxic depletion of microglia in mouse
monolayers of primary astrocytes. J Neurosci hippocampal slice cultures enhances ischemia-
Methods 150(1):128–137 like neurodegeneration. Brain Res 1291:
10. Giulian D, Baker TJ (1986) Characterization 140–152
of ameboid microglia isolated from developing 21. Reeves JP (1979) Accumulation of amino
mammalian brain. J Neurosci 6(8): acids by lysosomes incubated with amino acid
2163–2178 methyl esters. J Biol Chem 254(18):
11. Brahmachari S, Fung YK, Pahan K (2006) 8914–8921
Induction of glial fibrillary acidic protein 22. Thiele DL, Kurosaka M, Lipsky PE (1983)
expression in astrocytes by nitric oxide. Phenotype of the accessory cell necessary for
J Neurosci 26(18):4930–4939 mitogen-stimulated T and B cell responses in
human peripheral blood: delineation by its simplicifolia I-B4 isolectin. Proc Natl Acad Sci
sensitivity to the lysosomotropic agent, L- USA 79(1):166–170
LEUCINE methyl ester. J Immunol 131(5): 26. Streit WJ, Kreutzberg GW (1987) Lectin
2282–2290 binding by resting and reactive microglia.
23. Wiley RG, Oeltmann TN, Lappi DA (1991) J Neurocytol 16(2):249–260
Immunolesioning: selective destruction of 27. Laitinen L (1987) Griffonia simplicifolia lec-
neurons using immunotoxin to rat NGF recep- tins bind specifically to endothelial cells and
tor. Brain Res 562(1):149–153 some epithelial cells in mouse tissues.
24. Uliasz TF, Hamby ME, Jackman NA et al Histochem J 19(4):225–234
(2012) Generation of primary astrocyte cul- 28. Fang X, Djouhri L, McMullan S et al (2006)
tures devoid of contaminating microglia. Intense isolectin-B4 binding in rat dorsal root
Methods Mol Biol (Clifton, NJ) 814:61–79 ganglion neurons distinguishes C-fiber noci-
25. Maddox DE, Shibata S, Goldstein IJ (1982) ceptors with broad action potentials and high
Stimulated macrophages express a new glyco- Nav1.9 expression. J Neurosci 26(27):
protein receptor reactive with Griffonia 7281–7292
Chapter 8
Lentiviral Transduction of Cultured Microglia

and Kazuhide Inoue
Abstract
Microglial cells are the resident immune-related glial cells of the central nervous system (CNS) that are
crucial for maintaining homeostasis and sensing pathological alterations in the nervous system. To improve
our understanding of the biological function of microglia, gene-transfer techniques have been improved
and become widely used over the past several years. Here, we describe lentiviral-mediated transduction as
a valuable tool for transduction of cultured microglial cells.
Key words Microglia, Lentiviral vector, Gene transduction
1 Introduction
Microglia are the CNS immune cells that survey around environ-
ment by elongating and contracting their processes. During CNS
pathologies such as injury, they show rapid responses including
change in morphology and cell-surface antigen expression and are
crucial for CNS pathology [1–4]. During recent years, gene-
transfer techniques have undergone rapid progress; they are now
widely used as a useful procedure for the understanding of inher-
ent biological cellular function [5]. However, monomyeloic lin-
eage cells, including microglia, share similar characteristics in that
they are difficult to transduce [6]. That is why a limited number of
studies utilizing transgene approaches with primary cultured microg-
lia have achieved a satisfactory response. Meanwhile, several types of
gene-transfer tools based on viral vectors became more technically
advanced and have been spotlighted as a new effective means of
transducing such cells [6]. Among them, Lentivirus, a genus of ret-
roviruses that can transfer genetic information into the DNA of the
host cell, has the unique ability among retroviruses of being able to
effectively replicate in both dividing and nondividing cells [6, 7].
63
64 Takahiro Masuda et al.
Here, we demonstrate the lentiviral-mediated transduction of

cultured microglia with high transduction efficiency using some
unique methods.
2 Materials
Prepare and store the following at room temperature (unless indi-

cated otherwise). We do not add sodium azide to the reagents.
2.1 For Preparation 1. Poly-L-lysine hydrobromide MW 30,000–73,000 (Wako):

of Lentiviral Particles 0.1 μg/mL solution in PBS.
2. 100-mm cell culture dish: for coating, incubate poly-L-lysine
solution in a culture dish for at least 3 h. Then, remove the
solution and wash twice with PBS.
3. Culture medium for human HEK293T cells: 5 % fetal bovine
serum (FBS), penicillin (100 U/mL), streptomycin
(100 μg/mL), L-glutamine (2 mM) in Dulbecco’s modified
Eagle medium (DMEM). Store at 4 °C.
4. Plasmid vectors including targeted vector plasmid, packaging
plasmid (pCAG-HIVgp, RIKEN) and envelop plasmid
(pCMV-VSV-G-RSV-Rev, RIKEN).
5. 2 mg/mL polyethylenimine (PEI) solution: Store at 4 °C.
6. Forskolin.
7. PEG solution (4×): Make 80 mL of 50 % polyethylene glycol
6,000 (PEG). Add 8 mL of 5 M NaCl and then 4 mL of 1 M
HEPES (see Note 1). Adjust pH with NaOH and dilute to
100 mL with water. Sterilize using an autoclave. Store at 4 °C.
8. Filter: cellulose acetate filter (0.45 μm) (Advantec).
2.2 For Treatment 1. Primary cultured microglia: prepare according to our previous
of Lentiviral Particles paper [8].
2. 24-well plate.
3. Culture medium for mixed glial culture: 10 % FBS (Gibco), peni-
cillin (100 U/mL), streptomycin (100 μg/mL), L-glutamine
(2 mM) in DMEM (Gibco, Cat. 11965). Store at 4 °C.
4. 10 mg/ml polybrene (hexadimethrine bromide) solution:
Store at 4 °C.
3 Methods
Carry out all procedures at room temperature unless otherwise

specified.
Lentiviral-Mediated Transduction in Microglia 65
3.1 Lentiviral Vector Generate a targeted vector plasmid by cloning target cDNA into
Construction the lentiviral vector construct (see Note 2). Put the fluorescent
protein construct [e.g., green fluorescent protein (GFP)] into the
vector to determine the transduction efficiency.
3.2 Lentiviral 1. Seed HEK293T cells in poly-L-lysine-coated 100-mm cell cul-

Particles Preparation ture dishes at a density of 3.0 × 106 cells per dish in culture
medium (see Note 3).
2. Incubate the cells for 24 h at 37 °C in a humidified incubator
with an atmosphere of 5 % CO2 (about 70 % confluent).
3. Mix three plasmids, targeted vector plasmid (2.76 μg), packag-
ing plasmid (1.62 μg, pCAG-HIVgp), and envelope plasmid
(1.62 μg, pCMV-VSV-G-RSV-Rev), in a total volume of
600 μL of serum-free DMEM. Add 14 μL of polyethylenimine
(PEI, 2 mg/mL) and mix well (vortex). Incubate the mixture
for 15 min at room temperature.
4. Remove the culture medium from HEK293T cells, and add
10 ml of prewarmed fresh medium, and then add the mixture
to the dish.
5. Incubate for 16 h at 37 °C in a humidified incubator with an
atmosphere of 5 % CO2.
6. Remove the mixture-containing medium and add 10 ml of
prewarmed fresh medium containing 10 μM forskolin.
7. Incubate further for 48 h at 37 °C.
8. Filter the medium containing lentiviral particles through a fil-
ter (0.45 μm) and collect in a 50-mL conical tube.
9. Add a one-third volume of PEG solution, mix well, and incu-
bate at 4 °C overnight.
10. Centrifuge the mixed solution at 2,600 × g for 30 min at 4 °C
and discard the supernatant.
11. Resuspend the pellet in an appropriate solution by pipetting
(see Note 4). Store at –80 °C.
3.3 Preparing Collect the supernatant of the mixed glial culture and filter through
Conditioned Medium a filter (0.45 μm). Store at –20 °C (see Note 5).
of the Mixed Glial
Culture
3.4 Treatment of 1. Seed primary microglial cells in 24-well plates at a density of

Lentiviral Particles 1.2 × 105 cells per well (see Note 6).
with Primary Cultured 2. After cells have adhered to the bottom, add an appropriate
Microglia volume of viral particles and 8 μg/ml polybrene onto cultured
microglia (see Note 7).
Fig. 1 GFP-positive microglial cells after lentiviral transduction. GFP or bright field (BF) images of cultured
microglia transduced with a lentiviral vector encoding GFP. White or red arrowheads indicate GFP-positive or
negative microglial cells, respectively. To determine the transduction efficiency, count GFP+ cells among all
microglia cells 72 h after transduction
3. After a 12-h treatment with the lentiviral particles, change

culture medium to 1 mL of prewarmed conditioned medium
prepared from the mixed glial culture (see Subheading 3.3).
4. After a further 60 h, validate the transduction efficiency by
counting the microglial cells expressing fluorescent protein,
such as GFP (Fig. 1).
4 Notes
1. Mix PEG, NaCl, and HEPES in this order with stirring con-
tinually, or it may dissolve poorly (for Subheading 2.1, item 7).
2. We usually use the lentiviral CS2 vector (RIKEN) (for
Subheading 3.1).
3. For coating 100-mm culture dishes with poly-L-lysine, add
poly-L-lysine bromide solution to the dish and incubate over-
night (or for at least 3 h) (for Subheading 3.2, step 1).
4. We usually use approximately 200 μl of PBS, HBSS, and fresh
culture medium for resuspension. Try to avoid introducing air
bubbles to keep transduction efficiency high (for
Subheading 3.2, step 11).
5. We find that it is better to prepare the conditioned medium
from the mixed glial culture as freshly as possible (for
Subheading 3.3).
Lentiviral-Mediated Transduction in Microglia 67
6. For using the microglial cell line BV2, seed in 24-well plates at
a density of 1 × 104 cells per well and use an appropriate culture
medium that is the same in composition as that used before
and after transduction (see ref. 8) (for Subheading 3.4).
7. During treatment with viral particles, to achieve a high trans-
duction efficiency of the lentiviral vector, reduce the culture
medium of microglial cells at least to 500 μl per well (for
Acknowledgment
Dr. Hiroyuki Miyoshi (RIKEN BioResource Center) kindly pro-

vided the lentiviral vector and its packaging plasmids.
References
1. Hanisch UK, Kettenmann H (2007) Microglia: 5. Kamimura K, Suda T, Zhang G et al (2011)
active sensor and versatile effector cells in the Advances in gene delivery systems. Pharm Med
normal and pathologic brain. Nat Neurosci 25:293–306
10:1387–1394 6. Burke B, Sumner S, Maitland N et al (2002)
2. Perry VH, Nicoll JA, Holmes C (2010) Macrophages in gene therapy: cellular delivery
Microglia in neurodegenerative disease. Nat Rev vehicles and in vivo targets. J Leukoc Biol
Neurol 6:193–201 72:417–428
3. Glass CK, Saijo K, Winner B et al (2010) 7. Follenzi A, Naldini L (2002) Generation of
Mechanisms underlying inflammation in neuro- HIV-1 derived lentiviral vectors. Methods
degeneration. Cell 140:918–934 Enzymol 346:454–465
4. Tsuda M, Inoue K, Salter MW (2005) 8. Masuda T, Tsuda M, Yoshinaga R et al (2012)
Neuropathic pain and spinal microglia: a big IRF8 is a critical transcription factor for trans-
problem from molecules in “small” glia. Trends forming microglia into a reactive phenotype.
Neurosci 28:101–107 Cell Rep 1:334–340
Part IV
Analysis of Microglial Cytokine Production

Chapter 9
Microglial Activation: Measurement of Cytokines

by Flow Cytometry
Deepak Kumar Kaushik and Anirban Basu
Abstract
Cytokine measurement is a prerequisite to understand the inflammatory state of the body. Quantitative
analysis of cytokines by Western blotting and ELISA is a daunting task as these are time-consuming and
error-prone protocols. With the advent of flow cytometry, the estimation of cytokines using the classical
antigen–antibody reaction has become a popular choice with researchers/clinicians. Here, we describe a
protocol for multiple cytokine analysis using flow cytometry.
Key words Microglia, Cytokines, Cytokine bead array, Flow cytometry, Inflammation
1 Introduction
Microglia are the resident macrophages of the central nervous system

(CNS) which represent the innate immune arm of the CNS [1].
Activation of microglia is often detected by the levels of cytokines
and chemokines secreted in response to any pathogenic stimuli
such as lipopolysaccharide (LPS) [2]. Cytokines and chemokines
are low-molecular weight proteins which are involved in cell-to-
cell communication and execute key regulatory functions in differ-
ent target cell types [3]. The quantification of cytokines can be
achieved using enzyme-linked immunosorbent assay (ELISA)
and/or immunoblot techniques. However, both these techniques
are time consuming and labor intensive [4]. Flow cytometry is
routinely used for quantitative analysis of cell populations for either
cell cycle analysis, staining of surface, or intracellular proteins as
well as sorting of different populations (achieved by a cell sorter).
Cytokine bead array (CBA), a flow cytometry-based application,
offers efficient, reliable, and quick quantification of cytokines and
chemokines [4, 5]. The CBA principle is based on the antigen–
antibody (Ag–Ab) reactions similar to ELISA where the polymer
beads are coated with specific antibodies to individual cytokines.
71
72 Deepak Kumar Kaushik and Anirban Basu
The detection reagent is also bound to the antibodies which bind

to the already-bound cytokines on the beads at different sites. This
enables a highly specific binding of the cytokines with the beads.
The beads are then run on the cytometer and resolved on different
fluorescent channels in order to estimate the levels of different
cytokines. CBA offers a flexibility to carry out analysis of several
cytokines in a single experiment without much hassle [6, 7].
Various companies offer ready-to-use kits and softwares for the
quantification and analysis of these cytokines.
Here, we discuss the methodology of analyzing cytokine analy-
sis using mouse inflammation kit from BD (Becton Dickinson)
Biosciences using a BD™ FACSCalibur instrument. The applica-
tions and principles are necessarily the same with slight variations
observed between different suppliers.
2 Materials
2.1 Brain In vivo analysis of cytokines from mice brain can be detected from
Homogenate/Cell whole brain homogenates. However, the source of cytokines can
Culture Supernatant be redundant as astrocytes also secrete cytokines [8, 9]; it is the
microglial population that predominates in the CNS in secreting
2.1.1 Brain Homogenate
the cytokines and carrying out the innate immune responses dur-
ing CNS pathologies [10]. The analysis of cytokines from brain
homogenates provides the information regarding the inflamma-
tory state of this organ. For the estimation of cytokines, the amount
of protein homogenate needs to be standardized; however, a suc-
cessful sensitive assay can detect as low as 10 ng/L [5]. In practice,
we usually take 10 μg protein in 50 μl sample volume for the analy-
sis of cytokines (see Note 1).
Protocol for isolation of brain homogenate/cell lysate:
1. The brain tissue is homogenized in 1× lysis buffer (1 %
Triton-X-100, 10 mM Tris–HCl (pH 8.0), 150 mM NaCl,
0.5 % Nonidet P (NP-40), 1 mM EDTA, 0.2 % EGTA, 0.2 %
sodium orthovanadate, and protease inhibitor cocktail).
2. The homogenate is kept on ice with intermittent vortexing for
an hour and then centrifuged at 12,000 × g at 4 °C for 15 min.
Similarly, the protein from cells is also harvested by incubating
the cells in 1× lysis buffer and then pelleting them.
3. The supernatant from cell lysate or tissue homogenate is then
collected and measured for protein concentration (see Note 2).
2.1.2 Cell Culture The cytokines are small proteins that act as messengers between
Supernatant different cell types. They are secreted out into the cell culture
medium in which the cells are grown. This supernatant can be used
for analysis of cytokine secreted by cultured microglia. It is desired
that the volume of cell culture medium be as low as possible such
that the cytokine concentration is increased per μl (see Note 3).
Flow Cytometry Based Cytokine Analysis 73
Fig. 1 Gating of bead population on FSC/SSC (forward scatter/side scatter) plot.

All the beads having the same shape and size cluster together and can be seen
on the dot plot as a single cluster. It is then tightly gated which is represented as
R1 (Region 1) in the figure
The volume of supernatant to be taken for each sample may range

from 50 to 200 μl depending upon the sensitivity of the assay. The
volumes may be standardized before setting up an actual experi-
ment (see Note 4).
2.2 Cytokine Beads The cytokine beads are precoated with antibodies for detecting dif-
and Detection Reagent ferent cytokines. Each bead may correspond to a single cytokine.
Depending on the provider, these beads may be similar to each
2.2.1 Cytokine Beads
other except for the presence of different intensities of fluoro-
phores (enabling them to be detected as separate beads on a differ-
ent channel) [5]. Some providers have different sizes of beads for
different cytokines and can be detected based on their sizes. In the
former case, a single bead population is gated on FSC (forward
scatter) and SSC (side scatter) plot (Fig. 1), while later may have
different bead populations to be gated and analyzed separately.
The CBA kit from BD Biosciences have six different beads corre-
sponding to different cytokines and have different intensities of
fluorophore that enables the user to identify different bead popula-
tions (Fig. 2). However, the number of cytokines that can be
detected depends upon the assay system as some systems offer
analysis of multiple cytokines and can detect a large number of
cytokines simultaneously. For example, the following beads corre-
sponding to the following cytokines are distributed on FL3 chan-
nel from brightest to dimmest on a FACSCalibur machine using
mouse inflammation CBA kit from BD.
Fig. 2 Resolution of different beads corresponding to different individual cyto-

kines on FL3 channel. There is no change in the levels of these cytokines on FL1
channel as they are not labeled with FITC
Bead population Cytokines

A1 (Brightest) IL-6
A2 IL-10
A3 MCP-1
A4 IFN-γ
A5 TNF-α
A6 (Dimmest) IL-12p70
2.2.2 PE (Phycoerythrin) The CBA kit supplied by BD has phycoerythrin-conjugated anti-

Detection Reagent bodies that bind to the cytokine beads. The detection is then made
on FL2 channel on a BD™ FACSCalibur machine (Fig. 3). Other
suppliers may provide other detection reagents.
2.2.3 Recombinant The CBA kits from various sources have recombinant cytokines
Cytokines for Standard whose concentrations are known. Using serial dilutions of these
Curve Generation cytokines, a standard curve is generated which is used as a refer-
(Inflammation Standards) ence for the estimation of cytokines in a given sample.
Fig. 3 Levels of cytokines measured by distinct bead population on FL2 channel.

Different bead levels corresponding to changes in cytokine levels can be noticed
on FL2 channel which detects PE
2.3 Instrumentation A FACS instrument equipped with argon 488 nm laser is required.
and Buffers This enables to detect and distinguish fluorescence emissions at
670 nm on a FACSCalibur machine. Additional lasers may be
2.3.1 FACS Instrument
required if other dyes emitting at higher wavelength are used. For
with Cell Analysis Software
example, phycoerythrin (PE) emits at 575 nm; therefore, using a
machine which is equipped with either 3-color or 4-color laser is a
prerequisite for running this application (this application may be
required if the beads are clustered as different populations and
coated with different dyes).
2.3.2 Instrument These beads are provided for setting up the instruments. Using
Setup Beads FACSComp™ software from BD Biosciences, this can easily be
performed in a stepwise fashion according to the manufacturer’s
protocol. Instrument setup generates a file which is stored in the
“instrument setup” folder, and these settings are used while CBA
is carried out.
2.3.3 Sheath Fluid Sheath fluid is a proprietary product provided by the manufacturers.
for Sample Running Many workers also use filtered 1× PBS as sheath fluid which may
also contain antifungal/antibacterial agents such as sodium azide.
Some labs also use 0.9 % saline solution with no antimicrobial
agents for this application. Always refer to the recommended
buffers from the manufacturer in order to prolong the life of the
machine (see Note 5).
2.3.4 Wash Buffer Wash buffer is usually supplied with the kit. Wash buffer can also
be made in the laboratory: To make 500 ml of wash buffer, take
480 ml of 1× PBS, pH 7.4, and add 20 ml of fetal bovine serum
(FBS) to it. To this add 0.09–0.1 % sodium azide, mix well, and
store at 4°C till further use (see Note 5).
2.3.5 Buffer for FACS Clean™ Solution provided by BD Biosciences is a routinely

Cleaning the Fluidics used for cleaning the instrument after the sample has been ana-
lyzed. Ten percent bleach solution (sodium hypochlorite) may also
be used after the analysis for cleaning of the fluidics of flow cytom-
eter. This step is crucial for the proper cleaning of the fluidics in
order to avoid accumulation of debris from previous runs. This
also ensures uniform flow rates during a machine’s run. The solu-
tion is run for 5 min at high flow rates.
3 Method
3.1 Instrument Setup For CBA application, instrument setup beads are usually supplied
Using Instrument by the provider along with the CBA kit. Using interactive software
Setting Beads provided by the manufacturer, these setup beads are run on the
machine during which the various parameters are set automatically.
During the instrument setup, the various parameters such as
threshold and compensations are adjusted. This is then saved sepa-
rately as an “instrument setup” file in “applications” folder. This
path may be different for different manufacturers and machines.
3.2 Instrument Instrument calibration is routinely carried out using the beads con-
Calibration jugated to different dyes in order to set the threshold and compensa-
tions as well as check the instrument sensitivity for proper instrument
functioning. BD Biosciences offer BD™ Calibrite beads for a regular
instrument calibration. These beads are usually polymers. The
Calibrite beads provided with BD CBA kit are made of polymethyl-
methacrylate microspheres of approximately 6 μm diameter. The kit
provides the following beads for calibration of the instrument:
1. A two-color calibrite kit: contains unlabeled beads, fluorescein
isothiocyanate (FITC)-labeled beads, and phycoerythrin (PE)-
labeled beads.
2. A three-color calibrite kit: contains unlabeled, FITC-labeled,
PE-labeled, and peridinin chlorophyll (PerCP)-labeled beads.
3. A four-color calibrite kit: contains unlabeled, FITC-labeled,
PE-labeled, PerCP-labeled, and allophycocyanin (APC)-
labeled beads.
The following PMTs carry out following signal detection on
FACSCalibur (see Note 6):
FL1 (fluorescence 1): FITC (yellow-green)

FL2 (fluorescence 2): PE (red-orange)
FL3 (fluorescence 3): PerCP (red)
FL4 (fluorescence 4): APC (red)
3.3 Standard Curve Recombinant cytokines are provided in the kit or can be procured
separately from various companies as recombinant proteins. In order
3.3.1 Preparation
to carry out the standards, these recombinant cytokines are diluted
of Standards
in assay diluent in recommended volumes. These cytokines are
then serially diluted. Five thousand pg of each cytokine is provided
in the kit. Using CBA kit from BD Biosciences, the standards are
double diluted in a stepwise fashion to make the second standard
of 2,500 pg and so on such that the last dilution has 20 pg of cyto-
kines. Assay diluent (a solution provided by the manufacturer) is
also run on the cytometer which has 0 pg of standards. These are
then run on the instrument and standard curve for the following
concentrations are generated:
Concentration (pg) Dilution

0 Negative control (assay diluent)
20 (1:250)
40 (1:128)
80 (1:64)
156 (1:32)
312.5 (1:16)
625 (1:8)
1,250 (1:4)
2,500 (1:2)
5,000 Top standard
3.3.2 Acquiring the Data 1. Gating the bead population: On a FSC and SSC dot plot, bead
population may be represented by a cluster of cells (beads in this
case) which is then gated using the gating tool. Care must be
taken to avoid scattered population on the plot as it may represent
degraded beads or nonspecific bead population (see Note 7).
2. Setting the parameters and events: Go to the cytometer appli-
cation, select the latest instrument setup file, and apply it to the
current settings. Set minimum event rate to 3,000 of the total
bead population to be analyzed. A further increase in the event
rate may offer better results. Using the “setup” mode, the volt-
ages are adjusted to put the values of the highest dilution of
Fig. 4 A four-parametric standard curve generated for TNF-α, one of the cyto-
kines in mouse inflammation kit. X-axis labels the concentration in pg/ml while
Y-axis represents the mean fluorescent intensities (MFI) on FL2 channel
standard such that it occupies a left quadrant on the dot plot

between FL3 (separates different beads) and FL2 (detects the
levels of individual cytokines) channel. Once the voltages are
adjusted, the standards are run without the “setup” mode and
the data is acquired.
3. Generating the standard curve: Standard curves may either be
created manually from the readings obtained for individual
cytokines and then generating a line equation on a simple excel
sheet. This equation can be then used to analyze the levels of
various cytokines individually. Automatic analysis files or soft-
wares are also available with the instrument. A template file for
CBA analysis such as FCAP™ software may offer analysis of
individual cytokines with reference to the standard files. An
example of a typical 4-parametric logistic graph generated by
the software can be seen in Fig. 4. (The graph represents the
standard curve for TNF-α cytokine.)
4. Running the samples: The samples are run similar to the standard
samples. The samples are first gated and voltages are set such that
the values from untreated (the sample which is expected to have
a low cytokine level) samples are on the left quadrant of the
FL3-FL2 plot (on a FACSCalibur machine; remember phycoer-
ythrin is detected on FL2 channel) (see Note 8).
4 Detailed Protocol for Running of Samples
1. Preparation of samples: The beads corresponding to different

cytokines are mixed in 1:1 ratio. The bead vials must be vor-
texed before use (see Note 9). Depending upon the detection
capability, the volume of the bead mixture is defined. BD CBA
kit recommends 50 μl of bead mix for each sample. For a six-
cytokine bead array, the volume of each bead required is
50 μl ÷ 6 = 8.333 μl.
This number is then multiplied by the number of samples
to be analyzed (see Note 10). For example, in order to analyze
four samples, the volume of each bead required is
8.333 μl × 4 = 33.333 μl.
Once the bead mixture is prepared, 50 μl of sample is
added to each vial (see Note 11) along with 50 μl of phycoery-
thrin (PE) detection reagent (see Note 12) to a 50 μl bead
mixture. This step may be carried out in 1.5 ml plastic centri-
fuge tubes.
2. Incubation of samples: The mixture is then incubated for rec-
ommended time durations. Often, these time ranges from 2 to
3 h depending upon the manufacturer or the kit being used.
For analysis of mouse cytokines using the BD CBA bead array,
the samples must be incubated for 2 h at room temperature
under dark conditions (see Note 13).
3. Washing the samples: Upon incubation, the samples are then
washed using the wash buffer. Approximately, 1 ml of wash
buffer is added to each sample mixture which is then centri-
fuged at 200 × g for 5 min (see Note 14).
4. The wash buffer is now carefully aspirated without disturbing
the bead pellets and then additional 300 μl of wash buffer is
added to each vial to resuspend the pelleted bead population.
5. The samples are ready to be analyzed on a flow cytometer.
4.1 Instrument 1. Instrument is started and flow check is performed. Calibration of

Start-up and Data the machine (explained in the Subheading 3.2) may also be car-
Acquisition ried out at this juncture if not carried out recently (see Note 15).
2. CellQuest Pro™ software is launched and BD CBA instrument
setup template is opened (generated as explained in
Subheading 3.1). The software may vary from manufacturer

to manufacturer but the basic principles remain the same.
3. Samples are then run at low to medium flow rates in the “setup”
mode first to adjust the settings such that the bead population
is observed as a single cluster on FSC/SSC dot plot. Once the
gating is carried out of this population, the data is then acquired
without the “setup” mode as described in earlier sections. These
settings are kept constant during the acquisition of the rest of
the samples.
Parameter description: Using this application of the CellQuest
Pro, the directory (folder) is assigned to a run along with
file names for each sample.
Acquisition and storage: Using this application, events pertain-
ing to either total count or the gated population can be set.
Counters: The counters provide information regarding the total
cells being analyzed and flow rates as events per second.
4. Analysis of results: The results are analyzed with reference to
the standard curves generated (Fig. 4). Using the line equation
from the standard curve (shown in Fig. 4), the amount of cyto-
kines can be measured in pg/ml. Analysis of result is also carried
out automatically by automatic softwares such as FCAP CBA
flex from BD Biosciences.
5 Notes
1. The amount of protein depends upon the levels of cytokines

secreted. Usually 10 μg of total protein is sufficient to analyze
several cytokines that are measured in pg/ml.
2. In case the samples are obtained from infected animals/cells,
they must be handled with care and any direct contact must be
avoided. Discarding such samples must be done according to
the institutional chemical discard guidelines. A 10 % bleach
solution may come handy and should be used to treat such
samples before discarding them.
3. The samples must not be too concentrated as this may result
in high background. In a case where sample is too concen-
trated, number of washes must be increased.
4. The sample volume may depend on the amount of cytokine
secreted in the supernatant and also depends upon the cell
type in culture conditions. For example, while BV-2 mouse
microglia cell line often exhibits increased levels of MCP-1,
IL-6, and TNF-α, there is a relatively decreased expression of
IL-10, IL-12p50, and IFN-γ. Therefore, the amount of super-
natant depends upon the cell phenotype and the cytokines
secreted that are to be analyzed.
5. The beads, the wash buffer, and assay diluents contain sodium
azide. It is therefore advised not to come directly in contact
with these reagents and wear gloves and lab coats during per-
formance of these assays.
6. The fluorescence channels and their detection wavelength may
vary among different models of instruments. It is therefore
imperative that information about different PMTs of an instru-
ment is known to the user.
7. Beads representing a single population must be tightly gated.
This avoids false analysis of nonspecific bead population which
is often found scattered around a single cluster of bead
population.
8. The detection limit of cytokine analysis is 5,000 pg/ml and
therefore samples having high concentration of individual
cytokines may need to be diluted such that the cytokines are
within the range of linearity.
9. The bead vials must be thoroughly vortexed before aspirating
out the beads because the beads are often adhered to the plas-
tic surfaces and get aggregated.
10. It is advised to take additional beads. For example, if the sam-
ple number is n, then beads must be calculated for n + 1 sam-
ples such that there is no shortage of bead mix due to pipetting
errors.
11. The samples to be analyzed must be fresh and kept on ice such
as to avoid cytokine degradation. Avoid freeze–thaw cycles of
the samples.
12. PE must be added in dark conditions and the incubation with
PE must be carried out under dark conditions at room tem-
perature. The light conditions and incubation time limits must
be carefully read from the manufacturer’s guidelines. In case
this information is not accessible to the user, the user may
standardize these time points.
13. Incubation of samples with bead mix and detection reagent
usually results in increased debris or nonspecific binding of
beads if kept for longer durations. It is therefore necessary to
keep in mind that the incubation must not exceed the recom-
mended time.
14. Beads must be not be centrifuged beyond recommended grav-
itational forces (measured in relative centrifugal field) as this
may result in rupture of the beads. Once the beads are settled
after the centrifugation, the wash buffer is gently discarded
without disturbing the beads.
15. It is recommended that calibration of the machine is carried
out regularly just before the samples are analyzed.
References
1. del Rio-Hortega P (1932) Cytology and cel- 6. McHugh TM (1994) Flow microsphere
lular pathology of the nervous system. In: immunoassay for the quantitative and
Penfield W (ed) Microglia. P. B. Hoebaer, simultaneous detection of multiple soluble
New York, NY, pp 483–534 analytes. Methods Cell Biol 42(Pt B):
2. Kaushik DK, Gupta M, Das S et al (2010) 575–595
Kruppel-like factor 4, a novel transcription factor 7. Carson RT, Vignali DA (1999) Simultaneous
regulates microglial activation and subsequent quantitation of 15 cytokines using a multi-
neuroinflammation. J Neuroinflammation 7:68 plexed flow cytometric assay. J Immunol
3. Borish LC, Steinke JW (2003) 2. Cytokines Methods 227:41–52
and chemokines. J Allergy Clin Immunol 8. Carpentier PA, Begolka WS, Olson JK et al
111:S460–S475 (2005) Differential activation of astrocytes by
4. Chen R, Lowe L, Wilson JD et al (1999) innate and adaptive immune stimuli. Glia
Simultaneous Quantification of Six Human 49:360–374
Cytokines in a Single Sample Using 9. Ransohoff RM, Brown MA (2012) Innate
Microparticle-based Flow Cytometric immunity in the central nervous system. J Clin
Technology. Clin Chem 45:1693–1694 Invest 122:1164–1171
5. Oliver KG, Kettman JR, Fulton RJ (1998) 10. Kaushik DK, Gupta M, Basu A (2011)
Multiplexed analysis of human cytokines by Microglial response to viral challenges: every
use of the FlowMetrix system. Clin Chem silver lining comes with a cloud. Front Biosci
44:2057–2060 17:2187–2205
Chapter 10
In Situ Hybridization of Cytokine mRNA Using Alkaline

Phosphatase-Labelled Oligodeoxynucleotide Probes
Bettina Clausen, Christina Fenger, and Bente Finsen
Abstract
In situ hybridization is a powerful tool for visualizing cellular gene expression in morphologically
preserved brain tissue giving precise information on the regional expression of specific mRNA sequences
in cells of diverse phenotype. Here, we describe a sensitive, simple, and robust method using alkaline
phosphatase (AP)-labelled oligodeoxynucleotide probes to detect cytokine mRNA in the acutely injured
or inflamed mouse CNS.
Key words In situ hybridization, mRNA, Enzyme-conjugated oligodeoxynucleotide probes, Tumor

necrosis factor, Interleukin-1, Cytokines, CNS
1 Introduction
Hybridization of specific cytokine mRNA in situ is an important

supplement to immunohistochemistry for unequivocal identification
of the site of cellular transcription/translation, as the majority of
cytokines are cleaved or released into the extracellular space. We
have for the detection of cytokine transcripts in the CNS opti-
mized our in situ hybridization procedure for the use of alkaline
phosphatase (AP)-labelled, short 26–30 mer oligodeoxynucleotide,
probes [1–3]. The major advantages of this non-isotopic technique
are that it provides similar high resolution [1–4] as digoxigenin-
labelled probes [5–7] and reaches the same high sensitivity as
radiolabelled probes [8, 9]. Thus, the threshold levels of detection
are in the region of 10–20 mRNA copies per cell [10–12].
Hybridization occurs when the AP-labelled oligodeoxynucleotide
probe recognizes and binds by the means of hydrogen bonding to
the target mRNA sequence in the tissue section [13], followed by
probe-signal development using an AP substrate solution. Critical
for successful in situ hybridization is to avoid premature degradation
of target mRNA by nucleases. The stability of a given probe-mRNA
hybrid depends on its melting temperature (Tm), which for a given
83
84 Bettina Clausen et al.
Table 1
In situ hybridization probes for TNF and IL-1β mRNAs
No. of
Probe ID Probe sequence bases Target region % GC
TNF-I 5′CT TCT CAT CCC TTT GGG GAC CGA TCA CC 28 305–332 [17] 57
TNF-II 5′CG TAG TCG BBB CAG CCT TGT CCC TTG AA 28 570–597 [17] 60
TNF-III 5′CT TGA CGG CAG AGA GGA GGT TGA CTT TC 28 644–671 [17] 53
IL-1β 5′CT GTC CAT TGA GGT GGA GAG CTT TCA GCT C 30 501–530 [18] 57
IL-1β 5′GC TTG TGA GGT GCT GAT GTA CCA GTT GGG G 30 779–808 [18] 53
GADPH 5′CC TGC TTC ACC ACC TTC TTG ATG TCA 26 846–871 [19] 50
hybrid is a function of its cytosine (C) and guanine (G) content.

High GC content equals high Tm of the probe-mRNA hybrid,
since these nucleobases bind more strongly by the means of three
hydrogen bonds, as compared to adenine (A) and thymidine (T).
The strength of the hydrogen bonds is influenced by the strin-
gency factors [13, 14], which are extrinsic factors affecting the
melting temperature (Tm) of the newly formed hybrid. These fac-
tors are pH, salt concentration, temperature, and organic solvents
such as formamide. Increased salt concentration stabilizes the
hybrid while increased pH, temperature, and high formamide con-
tent destabilize the hybrid [13]. Here, we provide a detailed pro-
tocol for specific and sensitive detection of microglial- and
macrophage-produced TNF and IL-1β mRNA in the injured or
diseased mouse CNS using short AP-labelled oligodeoxynucleo-
tide probes (Table 1) [2, 3, 14].
2 Materials
2.1 Probe Design 1. Identify the specific mRNA sequence in the species of interest
(here Mus musculus) using the nucleotide search tool at the
NCBI website.
2. Identify the exon–exon junctions in the selected cytokine mRNA
(see Note 1) and use the FASTA format of a cDNA sequence
spanning one or more exon–exon junction(s) as input sequence
for the probe design software (for instance, Oligo-design soft-
ware v. 6.0, Molecular Biology Insight, CO, USA).
3. Design antisense oligodeoxynucleotide (~26–30 bases)
probe(s) that span an exon–exon junction site.
4. The CG content must be approx. 50 % and the DNA Tm must
be approx. 37 °C in the used hybridization conditions (1×
SSC, 50 % formamide, 165 mM NaCl) (see Note 2).
In Situ Detection of Cytokine mRNA 85
Fig. 1 In situ hybridization specificity and sensitivity illustrated on stroke-lesioned mouse brain showing TNF
mRNA expression in microglial-like cells. (Upper row) In situ hybridizations performed with Probe I, Probe II, or
Probe III show identical cellular and regional localization of the in situ signal (shown for Probe I, only).
Furthermore, hybridization with Probe I+II or Probe I+II+III yields stronger in situ signal than hybridization with
a single probe only, here Probe I. Photomicrographs were obtained from the border zone of the infarct 12 h
after induction of the stroke lesion, at which time single-cell expression is very high. (Lower row) Hybridization
with Probe I+II+III on RNAse-treated sections yields no signal. Similarly, hybridization with 100-fold excess of
each of the unlabelled probes, and incubation with hybridization buffer, yields no signal. Photomicrographs
were obtained of the cingulate cortex 4 h after a stroke lesion in a mouse (for details, see ref. 3). Arrows point
to TNF mRNA+ cells. Scale bar: 20 μm
5. Reject suggested probes with tendency to make stable dimers

or hairpins (∆G ≤ −5 kcal/mol).
6. Confirm that the probe(s) selectively recognize(s) the target
mRNA by using standard nucleotide BLAST (blastn) tool at
the NCBI website.
2.2 Solutions and 1. AP-labelled antisense oligodeoxynucleotide probes written in

Reagents Used for In the direction from 5′ to 3′ (Table 1) are purchased from DNA
Situ Hybridization Technology Aps, Denmark. The probe is synthesized with a
modified 5′ amino terminal which is coupled to AP thereby
generating the AP-probe conjugate [14]. Using two or more
probes targeting different, non-overlapping sequences of the
target mRNA increases signal strength in an additive way
(Fig. 1, upper row).
2. Diethylpyrocarbonate (DEPC)-treated water: 1 % DEPC solu-
tion in dH2O (see Note 3). Stir overnight and autoclave the
DEPC-water to remove excess DEPC.
3. Hybridization buffer: 500 ml 50 % formamide, 200 ml 20×

SSC, 25 ml 40× Denhardt, 200 ml 50 % dextran sulfate, 100 μl
single-stranded salmon sperm DNA (see Note 4).
(a) Formamide: 100 ml formamide and 5 g Amberlite for 1 h
in the dark at room temperature (RT), filter and store at
−20 °C.
(b) 20× Saline sodium citrate (SSC): 150 mM NaCl and
15 mM sodium citrate dissolved in 500 ml DEPC-water.
Store at 4 °C.
(c) 40× Denhardt: 0.04 g Ficoll, 0.04 g polyvinylpyrrolidone,
and 0.04 g bovine serum albumin dissolved in 5 ml DEPC-
water (see Note 5).
(d) 50 % dextran sulfate: 5 g dextran sulfate dissolved in 10 ml
DEPC-water. Stir at RT.
(e) Single-stranded salmon sperm DNA: 50 μg single-stranded
salmon sperm DNA dissolved in 50 ml Tris–EDTA (TE)
buffer.
4. TE buffer: 0.5 ml 2 M Tris buffer (pH 7.4) adding up to
100 ml DEPC-water, 20 μl 0.5 M EDTA. Adjust pH to 8.
5. Tris buffer 2 M: 24.2 g Tris(hydroxymethyl)aminomethane
TRIS (Sigma) dissolved in 100 ml DEPC-water. Add concen-
trated HCl to pH 7.4.
6. EDTA: 4.65 g EDTA dissolved in 100 ml DEPC-water. Adjust
pH to 8.
7. 1× SSC: 50 ml of 20× SSC and 950 ml dH2O.
8. Tris–HCl buffer (rinsing buffer): 0.01 M Trizma hydrochloride
and 0.15 M NaCl dissolved in 1 L dH2O. Adjust pH to 9.5.
9. Tris–HCl–MgCl2 (developer buffer): 0.01 M Trizma hydro-
chloride, 0.01 M NaCl, and 0.05 magnesium chloride
(MgCl2⋅6H2O) dissolved in 1 L dH2O. Adjust pH to 9.5.
10. AP substrates: Nitroblue tetrazolium (NBT) and 5-bromo-4-
chloro-3-indolyl phosphate (BCIP).
(a) NBT stock solution: 70 mg NBT dissolved in 1 ml 70 %
N,N-dimethylformamide.
(b) BCIP stock solution: 50 mg BCIP dissolved in 1 ml 100 %
N,N-dimethylformamide (see Note 6).
11. AP developer: 100 ml Tris–HCl–MgCl2 (developer buffer)
containing 450 μL NBT stock solution and 350 μL BCIP
stock solution.
12. RNAse A solution (control reaction): 200 μg/ml RNAse A
diluted 1:3 in RNAse buffer to give a final concentration of
50 μg/ml (see Note 7).
13. RNAse buffer: 0.1 M Tris–HCl, 0.5 M NaCl, 1 mM EDTA
diluted in 800 ml DEPC-water.
3 Methods
3.1 Tissue The PBS perfused or unperfused brain is quickly removed and
Processing frozen by the use of CO2 snow (see Note 8). The brain is cut into
series of 30 μm (see Note 9) thick cryostat sections and mounted
on RNAse-free SuperfrostPlus glass slides (Hounisen) (see Note
10) and stored at −80 °C until further used (see Note 11).
3.2 Hybridization, Day 1

Posthybridization
1. Place glass slides containing the fresh frozen tissue sections on
Rinsing, and
a paper tray (RNAse-free) and dry the sections at 55 °C for
Development
10 min.
2. Immediately after, place the sections in 96 % ethanol for a
minimum of 30 min (see Note 12).
3. Air dry the section at RT for approx. 20 min.
4. Add probe(s) to the hybridization buffer and mix gently, but
well (see Note 13). Keep probes on ice while pipetting. The
optimal probe concentration must be determined every time
the procedure is set up for a new target mRNA (see Note 14).
The volume of hybridization buffer per slide depends on the
number of sections per slide (see Note 15). Carefully add the
hybridization buffer containing the probe onto the slide, but
next to the sections (see Note 16). Cover the section with a
60 mm long mounting glass, while gently pushing the drop of
liquid over the sections, thereby avoiding air bubbles
(see Note 17).
5. Place the cover-slipped sections in a humid plastic hybridization
chamber (Life Technology) containing 10 ml distilled H2O at
the bottom, which reduces evaporation of hybridization buffer.
6. Seal the plastic chamber using aluminum foil/parafilm and
place it in a heating chamber at 37 °C overnight.
7. Prepare 1× SSC and place in a 55 °C heating chamber over-
night. Cover to prevent evaporation.
Day 2
1. Remove the mounting glass and rinse the hybridized section.

2. Rinse sections 3 × 30 min in 1× SSC, pH 8, at 55 °C.
3. Rinse for 2 × 10 min in Tris–HCl, pH 9.5 (see Note 18).
4. Mix Tris–HCl–MgCl2 buffer, pH 9.5, with the AP substrates
(AP developer). Place sections in upright rack and add the AP
developer. Wrap the rack in aluminum foil and allow signal
development to take place in darkness at RT.
Days 4–5
1. Arrest signal development after 48–72 h and rinse well in run-

ning tap water (see Note 19).
2. Cover slip sections using Aqua-Mount™. When successful
hybridization has occurred, the target mRNA will appear as
blue formazane deposits within the mRNA-expressing cells.
3.3 Control Specificity controls are performed on sections parallel to those of

Reactions interest. A control panel for in situ hybridizations for TNF mRNA
in stroke-lesioned mouse brain is shown in Fig. 1.
1. Probe specificity. Confirmed by the use of two or more probes,
alone and in combination. This is illustrated for detection of
TNF in Fig. 1 (upper row) and Table 1. The probability that
Probe I, II, or III yields identical regional and cellular localiza-
tion of the in situ hybridization signal is negligible, if one of
the probes should happen to detect a different mRNA than
TNF mRNA. Besides adding to the sensitivity, yielding a stron-
ger signal in sections hybridized with the probe mixtures
(Probe I+II or Probe I+II+III) compared to a single probe
also provides evidence of probe specificity (Fig. 1, upper row).
2. RNAse pretreatment. The binding specificity of the probe for
mRNA is tested on RNAse pretreated sections (see Note 7).
Absence of signal suggests that the probe binds to RNA within
the tissue section (Fig. 1, bottom row). For RNAse pretreat-
ment, 1–2 control sections are incubated with RNAse A solu-
tion for 1–2 h(s) at 37 °C. Rinse 2 × 10 min in 2× SSC and 10
min in DEPC-water at RT and dehydrate sections in 96 %
ethanol before reentering the main protocol again, at step 3
(Day 1).
3. Competition with unlabelled probe. Hybridization with
AP-labelled probes and a 100-fold excess of unlabelled probes
results in absence of hybridization signal (Fig. 1, bottom row)
(see Note 20).
4. In situ hybridization for GAPDH mRNA. Finally, as a test for
tissue quality, hybridization with a probe specific for GAPDH
mRNA is routinely included, since GAPDH mRNA is highly
expressed throughout the brain (see Note 21).
5. Detection of the cytokine encoded by the mRNA of interest.
Detection of the cytokine itself by immunohistochemistry in
the cells with the same morphology and location reinforces the
specificity of the in situ hybridization reaction as illustrated in
Lambertsen et al. [3]. However, absence of protein does not
mean that the in situ hybridization signal for the mRNA is
unspecific (see Note 22).
4 Notes
1. To make sure that the oligodeoxynucleotide probes selectively

recognize a specific mRNA member of protein family, design
probes against regions of high variability within the protein
family. Furthermore, make sure that the selected exon–exon
junction(s) is/are present in all the mRNA isoforms or the
single mRNA isoform of interest, depending on the purpose.
2. The Tm is the temperature at which half the DNA is present in
a single-stranded form. However, the RNA:DNA stability is
higher than the DNA:DNA stability. Therefore, we use a
Thyb ≈ Tm (DNA:DNA).
3. Addition of DEPC inactivates RNAses present in the water. In
situ hybridization should be carried out under RNAse-free
conditions until the hybridization step is completed. To secure
RNAse-free condition, glass articles should be heated at 260 °C
for 6–12 h, and solutions for the hybridization buffer need to
be made based on DEPC-water.
4. Formamide reduces the thermal stability of the hydrogen
bonds allowing hybridization to be carried out at a lower tem-
perature. The monovalent cations of SSC decrease the electro-
static repulsion between the two strands of the DNA–RNA
duplex. EDTA removes free divalent cations from the hybrid-
ization solution that strongly stabilize the DNA–RNA duplex.
Dextran sulfate accelerates the hybridization by generating a
local increase of the probe concentration. Denhardt’s solution
and salmon sperm DNA decrease the chance of nonspecific
binding of the oligodeoxynucleotide probe. Hybridization
buffer can be stored at −20 °C up to 12 months.
5. Add reagents in the indicated order.
6. AP developer should be mixed just before use and be light yel-
low. NBT and BCIP stock solutions should be shielded from
light.
7. Perform the RNAse control reaction away from the rest of
the experiment using separate glass articles, pipettes, and
equipment.
8. Liquid nitrogen or dry ice can be used as alternatives to CO2
snow. The usage of paraformaldehyde-perfused, sucrose-
immersed tissue introduces a requirement for enzymatic treat-
ment of the tissue sections prior to hybridization.
9. Sections can be cut thicker or thinner; however, they must fit
into the small hybridization chamber, which is formed when
covering slipping the RNAse-free SuperfrostPlus glass slide.
This small hybridization chamber facilitates a uniform
hybridization.
10. It is possible to perform in situ hybridization on standard

gelatin-coated glass slides.
11. It is possible to perform in situ hybridization on tissue stored
for >5 years, at −80 °C and −20 °C.
12. Tissue can be immersed longer for convenience. However,
immersion in ethanol for longer than 24 h leads to precipita-
tion of the mRNA and thereby to reduced hybridization
signal.
13. The hybridization buffer is stored at −20 °C and is quite vis-
cous. Therefore, leave at RT for 30 min before adding the
probe. Good mixing (do not vortex) facilitates even
hybridization.
14. Start with testing the probe at a concentration of 2, 4, 6, 8,
and 10 pmol/ml hybridization buffer. Then continue with the
concentration giving the best signal as compared to
background.
15. For 2 mouse brain sections per glass slide, dilute probes into
100 μl hybridization buffer and apply approx. 95 μl per slide.
For 4 or more brains per glass slide, apply up to 120 μl per
slide.
16. Important that the drop does not touch the tissue to avoid
hybridization artifacts caused by quick evaporation of the
hybridization buffer due to its content of formamide.
17. Avoid toughing the cover slip on the tissue slide. If air bubbles,
carefully press them out.
18. If necessary, the sections can rest here.
19. The optimal time of signal development depends on the
amount of mRNA in the tissue. Low mRNA expression equals
longer time of signal development. Sections can (quickly) be
inspected macroscopically (GAPDH mRNA) or microscopi-
cally (GAPDH and cytokine mRNA) during development.
20. Competition control ensures that the in situ hybridization sig-
nal is not due to nonspecific binding of the AP-probe conju-
gate to the tissue section [14]. Excess unlabelled probe
prevents (by competition for binding sites and excess amount)
binding of the AP probe to the target mRNA and thereby
leaves the section devoid of signal.
21. For illustration of GADPH mRNA expression in murine CNS,
see Clausen et al. [15].
22. Cytokines are subject to posttranscriptional processing and are
not necessarily translated into immunohistochemically detect-
able levels of protein. This is illustrated in Fenger et al. [16].
Acknowledgment
Lene Jørgensen and Sussanne Petersen are acknowledged for their

excellent technical assistance. The experimental material shown in
Fig. 1 was kindly provided by Dr. Kate Lambertsen. The work was
supported by grants from the Lundbeck Foundation, the Novo
Nordisk Foundation, and the Danish Multiple Sclerosis Society.
References
1. Gregersen R, Lambertsen KL, Finsen B (2000) sections using antisense RNA probes.
Microglia and macrophages are major sources Histochem J 18:597–604
of tumor necrosis factor in permanent middle 11. VandenBroecke C, Tovey MG (1991)
cerebral occlusion in mice. J Cereb Blood Expression of the genes of class I interferons
Flow Metab 20:53–65 and interleukin-6 in individual cells. J
2. Clausen BH, Lambertsen KL, Meldgaard M Interferon Res 11:91–103
et al (2005) A quantitative in situ hybridiza- 12. Lu J, Tsourkas A (2009) Imaging individual
tion and polymerase chain reaction study of microRNAs in single mammalian cells in situ.
microglial-macrophage expression of Nucleic Acids Res 37:e100
interleukin-1beta mRNA following permanent 13. Tecott LH, Eberwine JH, Barchas JD et al
middle cerebral artery occlusion in mice. (1987) Methodological considerations in the
Neuroscience 132:879–892 utilization of in situ hybridization. In:
3. Lambertsen KL, Clausen BH, Babcock AA Eberwine JH, Barchas JD, Valentino KL (eds)
et al (2009) Microglia protect neurons against In situ hybridization: amplification to neurobi-
ischemia by synthesis of tumor necrosis factor. ology. Oxford Press, New York, NY, pp 3–24
J Neurosci 29:1319–1330 14. Finsen B, Gregersen R, Lehrmann E et al
4. Christensen JE, Simonsen S, Fenger C et al (2004) In situ hybridization. In: Evans SM,
(2009) Fulminant lymphocytic choriomeningitis Janson AM, Nyengaard JR (eds) Quantitative
virus-induced inflammation of the CNS involves methods in neuroscience—a neuroanatomical
a cytokine-chemokine-cytokine-chemokine cas- approach. Oxford University Press, New York,
cade. J Immunol 182:1079–1087 NY, pp 115–145
5. Buttini M, Appen K, Sauter A et al (1996) 15. Clausen BH, Lambertsen KL, Finsen B (2006)
Expression of tumor necrosis factor alpha after Glyceraldehyde-3-phosphate dehydrogenase
focal cerebral ischemia in the rat. Neuroscience versus toluidine blue as a marker for infarct
71:1–16 volume estimation following permanent mid-
6. Meltzer JC, Sanders V, Grimm PC et al (1998) dle cerebral artery occlusion in mice. Exp
Production of digoxigenin-labelled RNA Brain Res 175:60–67
probes and the detection of cytokine mRNA in 16. Fenger C, Drøjdahl N, Wirenfeldt M et al
rat spleen and brain by in situ hybridization. (2006) Tumor necrosis factor or its TNFp55
Brain Res Prot 2:339–351 and -p75 receptors are not required for axonal
7. Woodroofe MN, Cuzner ML (1993) Cytokine lesion-induced microgliosis in mouse fascia
mRNA expression in inflammatory multiple dentata. Glia 54:591–605
sclerosis lesion: detection by non-radioactive 17. Pennica D, Hayflick JS, Bringman TS et al
in situ hybridization. Cytokine 5:583–588 (1985) Cloning and expression in Escherichia
8. Kiefer R, Funa K, Schweitzer T et al (1996) coli of the cDNA for murine tumor necrosis fac-
Transforming growth factor-beta 1 in experimen- tor. Proc Natl Acad Sci USA 82:6060–6064
tal autoimmune neuritis. Cellular localization and 18. Gray BW, Glaister D, Chen E et al (1986) Two
time course. Am J Pathol 148:211–223 interleukin 1 genes in the mouse: cloning and
9. Turrin NP, Rivest S (2006) Tumor necrosis expression of the cDNA for murine interleu-
factor a but not interleukin 1β mediates neuro- kin-1b. J Immunol 137:3644–3648
protection in response to acute nitric oxide 19. Sabath DE, Broome HE, Prytowsky MB
excitotoxicity. J Neurosci 26:143–151 (1990) Glyceraldehyde-3-phosphate dehydro-
10. Höfler H, Childers H, Montminy MR et al genase mRNA is a major interleukin-2-induced
(1986) In situ hybridization methods for the transcript in a cloned T-helper lymphocyte.
detection of somatostatin mRNA in tissue Gene 91:185–191
Chapter 11
Use of Meso-Scale Discovery™ to Examine Cytokine

Content in Microglia Cell Supernatant
Miguel A. Burguillos
Abstract
Cytokine production by activated microglia is one of the hallmarks of inflammatory response in the CNS.
The cytokines released by microglia cells can be very different depending on the proinflammatory
stimulus.
Traditionally, to quantify these different cytokines, the “Sandwich”-enzyme-linked immunosorbent
assay (Sandwich-ELISA) has been used. In this chapter we will discuss and describe an improved protocol
of the Sandwich-ELISA developed by Meso-Scale Discovery based on an electrochemiluminescence detec-
tion system, which allows the ultralow detection of multiple cytokines in microglia cell supernatant.
Key words Cytokine, ELISA, Antibody, SULFO-TAG, Inflammation, Electrochemiluminescence
1 Introduction
Inflammatory signals produced within the central nervous system

(CNS) and peripheral tissues regulate diverse biological processes.
Surveying microglia generate these signals upon several stimuli, for
instance, tissue damage, cellular dysfunction, and infection. These
inflammatory signals trigger intracellular signaling cascades that
eventually lead to immune cell activation, proliferation, cell recruit-
ment, or cellular demise. Of particular importance in the genesis of
inflammatory events are the immunomodulators referred to as
cytokines. These immunomodulators have been shown to play
very important roles in several processes such as acute neurodegen-
eration [1], immunodeficiency syndromes [2], and even during
the development of CNS [3].
Cytokines are heterogeneous, nonstructural proteins with
molecular weights ranging from 8 to 40,000 Da. There is no amino
acid sequence motif or three-dimensional structure shared among
the different cytokines. They are grouped depending on their bio-
logical activity [4].
93
94 Miguel A. Burguillos
Fig. 1 Principle of the Sandwich-ELISA. (a) The surface of the well is applied with a known amount of the
capture antibody. (b) The antigen-containing sample to the surface. (c) A specific antibody against our antigen
is added forming a “sandwich”—antibody-sample-antibody. (d) We add an enzyme-linked secondary antibody
that binds specifically to the nonspecific region antibody (Fc) of the previous antibody used in step c.
(e) A chemical solution is applied and converted by the enzyme into a color, fluorescent, or electrochemical
signal that can be quantified
Cytokines can be divided as:

– Th1-type cytokines: include interleukin 1-β, IFN-γ, and TNF-α.
They generate the proinflammatory phenotype responsible for
killing intracellular parasites and for perpetuating autoimmune
responses.
– Th2-type cytokines: include interleukins 4, 5, and 13, which are
associated with the promotion of IgE and eosinophilic
responses in atopy, and also interleukin 10, which contributes
to an anti-inflammatory response.
– Th17-type cytokines: include interleukin 17, which plays a role
in host defense against extracellular pathogens by mediating
the recruitment of neutrophils and macrophages to infected
tissues.
The ratio between Th1/Th2/Th17 cytokines is what defines
the type of activation of our microglia. The quantification of the
cytokine content is a very useful tool to characterize our microglia
population. Generally, these measurements have been performed
using the “Sandwich”-enzyme-linked immunosorbent assay (or
“Sandwich”-ELISA). This technique is based on the detection of
an analyte (in this case a cytokine) dissolved in a certain solution
(Fig. 1). Sandwich-ELISA is able to separate some components of
the analytical mixture by adsorbing these components onto a solid
Multiplex Cytokines Measurement 95
phase, which is physically immobilized. The principle of the

technique is the following:
1. On top of a stationary solid phase (for instance, a 96-well
plate), a molecule with special binding properties (the capture
antibody) is added.
2. Then, the blocking solution is added to decrease unspecific
binding. The plate is washed several times to take away the rest
of the blocking solution (Fig. 1a).
3. The sample is poured into the well and binds to the capture
antibody (Fig. 1b). Additional washes take away the rest of
proteins that have not bound to the antibody.
4. There is an incubation step with another antibody, the detect-
ing antibody that will bind to our protein, forming a
“sandwich”—antibody/sample/antibody (Fig. 1c).
5. A secondary antibody with an enzyme linked in its Fc region
recognizes and binds to the detecting antibody (Fig. 1d).
6. Finally, a developing solution is supplied, that is, a substrate for
the enzyme conjugated to the secondary antibody. The detec-
tion’s sensitivity depends on the amplification of the signal
during the analytic reactions. In order to obtain accurate read-
ings, fixed proportions of enzymes must be linked to the detec-
tion reagents.
Meso-Scale Discovery (MSD) MULTI-ARRAY® using an
improved electrochemiluminescence detection system is able to
detect ultralow levels of cytokines (up to five logs of linear dynamic
range). In addition, the assay can be performed with lower amounts
of sample as compared with the traditional ELISA. Finally, it pres-
ents the advantage of measuring several cytokines from a single
sample with a very simple protocol.
The Meso-Scale Discovery MULTI-ARRAY technology is
based on an electrochemiluminescence method to quantify the
detection antibody. This detection antibody is linked to SULFO-
TAG™ labels that emit light upon electrochemical stimulation initi-
ated at the electrode surface of MULTI-ARRAY®–MULTI-SPOT®
microplates. One advantage of this array is that the emission light
wavelength is at 620 nm, which avoid the color quenching prob-
lems. Only labeled antibodies located near to the electrode are
excited and detected.
MSD provides a plate coated with anti-species antibody (cap-
ture antibody) that allows the immobilization of the protein/s of
interest in the sample. Later on, the detecting antibody is added
together with the SULFO-TAG™ label. The user adds an MSD
Read Buffer (with the appropriate chemical environment for the
electrochemiluminescence reaction plus co-reactants that enhance
the signal) and quantifies the emitted light into an MSD SECTOR™
instrument.
You can choose between several predetermined mouse/

human/rat cytokine assays (is able to measure from 1 to 10 differ-
ent cytokines in a 96-well or 384-well format MULTI-ARRAY®
and MULTI-SPOT® cytokine assays).
2 Materials
2.1 Materials 1. Cytokine calibrators (stock concentration: 1 μg/ml). Store

Included in 1 Kit (and at −80 °C.
Storage Conditions) 2. Detection antibody mix (Anti-cytokine antibodies already pre-
labeled with SULFO-TAG reagent. 50× stock solution). Store
it at 2–8 °C.
3. MULTI-ARRAY or MULTY-SPOT plate. Store it at 2–8 °C.
4. Two different blocking solutions or “blockers” with stabilizing
agents (depending on the cytokine content).
(a) Blocker B (if the kits measure IL-12p40; room
temperature).
(b) Blocker D-B (only for kits that include TNF-α; −20 °C).
5. Diluent 1 (RPMI-based medium to dilute the calibrators;
2–8 °C).
6. Diluent 100 (blocking solution with stabilizing agents;
2–8 °C).
7. Read Buffer T (4× read buffer T with surfactant; room
temperature).
2.2 Materials and 1. Deionized water for diluting concentrated buffers.

Equipment Required 2. 15 and 50 ml tubes for reagent preparation.
3. Microcentrifuge tubes for preparing serial dilutions.
4. Phosphate-buffered saline (pH 7.4) + 0.05 % Tween-20 (wash
buffer).
5. Accurate pipette to dispense volume from 10 to 150 μl into a
96-well microtiter plate.
6. Automated plate washer of multichannel pipette.
7. Adhesive plate seals.
8. Rotating plate shaker.
9. SECTOR® Imager apparatus from Meso-Scale company
(Check the different models at Meso-Scale website: www.
mesoscale.com).
2.3 Buffers Needed 1. Blocker B solution (Only for assays that detect IL-12p40)
Prepare 20 ml of the solution per plate of 20 mg Blocker B in
20 ml of PBS (0.1 % w/v).
2. Calibrators and control solutions

Dilute the calibrators using Diluent 1 solution in the following
way to generate a standard curve with the following range
(2.4 pg/ml up to 10,000 pg/ml):
10,000 pg/ml: Mix 10 μl of the 1 μg/ml stock + 990 μl of
Diluent 1.
2,500 pg/ml: Mix 50 μl of the 10,000 pg/ml calibra-
tor + 150 µl of Diluent 1.
625 pg/ml: Mix 50 µl of the 2,500 pg/ml calibrator + 150 µl
of Diluent 1.
156 pg/ml: Mix 50 µl of the 625 pg/ml calibrator + 150 µl of
Diluent 1.
39 pg/ml: Mix 50 µl of the 156 pg/ml calibrator + 150 µl of
Diluent 1.
9.8 pg/ml: Mix 50 µl of the 39 pg/ml calibrator + 150 µl of
Diluent 1.
2.4 pg/ml: Mix 50 µl of the 9.8 pg/ml calibrator + 150 µl of
Diluent 1.
0 pg/ml: 150 μl of Diluent 1.
Diluent 1 is an RPMI medium with 10 % serum. Other
mediums can be used, instead of RPMI 10 % serum, if the
experiments have been performed with it (i.e., DMEM 2 %
serum). In general, the presence of some protein in the diluent
solution is desirable to prevent loss of the analyte through
adsorption with other surfaces like tubes and pipettes.
3. Detection antibody solutions
(CAUTION: Some of the antibodies are light sensitive. Keep
this solution in darkness.)
The antibodies are premixed at a 50× concentration
(50 mg/ml). The final concentration to work with should be
1× (1 mg/ml).
(a) For one plate you need to mix in a 15 ml tube: 60 μl of the
detection antibody mix solution + 2.94 ml of Diluent 100.
(b) If the assay measures TNF-α, then add 210 μl of Blocker
D-B (10 %) + 2.73 ml Diluent 100 + 60 μl of the detection
antibody mix solution in a 15 ml tube.
4. Read Buffer T solution
The Read Buffer T must be diluted in deionized water to
a final concentration of 2× Read Buffer T. For one plate, mix
10 ml of stock Read Buffer T (4×) + 10 ml of deionized water
in a 50 ml tube.
5. MSD plate
The plate can be used as delivered (the plate has been already
pre-coated with capture antibodies specific for the cytokines).
It has a stabilizing treatment to guarantee the stability of the
antibodies, but it has not been pre-blocked. Some cytokines

need a blocking step to improve the measurement (i.e.,
IL-12 p40).
6. Wash buffer
PBS (pH 7.4)—0.05 % Tween-20
2.4 Sample This assay can be used to measure the cytokine content in tissue/
Preparation cell culture supernatant and tissue lysates (see Note 1). The follow-
ing notes must be considered:
1. For most of the samples, there is no need of a dilution step
unless the experimental conditions promote an extremely high
production of cytokines. If a dilution step is necessary, it should
be done by a factor of 2–10.
2. Normally avoid multiple freeze/thaw cycles since the sensitiv-
ity for the detection of cytokines decrease greatly after the first
round of thawing due to the labile nature of some of them.
Also keep the samples at −80 °C, if they are not measured
immediately (see Note 2).
3. In tissue lysates, the lysis buffer should contain low levels of
denaturing detergents (i.e., less than 0.1 % SDS) and reducing
agents (i.e., less than 1 mM DTT). Also a carrier protein should
be added (i.e., 1 % BSA) to avoid loss of the analyte with tubes
or pipette tips.
3 Methods
(Carry out all procedures at room temperature unless otherwise

specified.) A schematic protocol of this method is presented in
Fig. 2.
3.1 Addition of the Dispense 150 μl of Blocker B solution per well. Seal the plate with
Blocker Solution (Only an adhesive plate seal and incubate with vigorous rotational shak-
for Assays Containing ing (300–1,000 rpm) for 1 h at room temperature. Wash three
IL-12p40 Cytokine) times with the wash buffer.
3.2 Addition of Add 25 μl of either your sample or calibrator into a separate well of
Sample and Calibrator the MSD plate (it is recommended to perform the analysis in dupli-
cates) (see Notes 3 and 4). Seal the plate with an adhesive plate seal
and incubate with vigorous rotational shaking (300–1,000 rpm)
for 1–2 h at room temperature. This step can be done without
shaking but in this case the incubation time is longer (4 h or lon-
ger) in order to get the same sensitivity.
3.3 Addition of Add 25 μl of the 1× detection antibody solution into each well of
Detection Antibody the MSD plate. Seal the plate with an adhesive plate seal and incu-
Solution bate with vigorous rotational shaking (300–1,000 rpm) for 1–2 h
Fig. 2 Schematic protocol for measuring cytokines with the MULTI-ARRAY and MULTI-SPOT cytokine assay
at room temperature. This step can be done without shaking but

in this case the incubation time is longer (4 h or longer) in order
to get the same sensitivity.
3.4 Wash and Wash three times with the wash buffer and add 150 μl of the 2×
Measurement of Read Buffer T to each well (Avoid bubble formation!). Plates may
the Plate be read immediately after adding the reading buffer. Quantify the
plate with SECTOR® Imager (see Notes 5–7).
4 Notes
1. In our experience, there’s a much higher production of cyto-

kines in primary microglial cultures (for instance, primary
microglia extract from the cortex) than in microglial cell line
(e.g., the murine microglial cell line BV2).
2. If you are measuring cytokine content from supernatant

obtained from cell cultures, it is good to fast freeze them once
(i.e., by putting the eppendorfs with the supernatant immedi-
ately in dry ice and keep the samples in −80 °C until they are
used).
3. Do not touch the pre-coated electrode surface with the pipette.
4. Use the reverse pipetting technique to deliver accurately the
different solutions and obtain reproducible measurements.
5. In each wash, remove the buffer thoroughly and tap the empty
plate on top of paper towels in order to get rid of as much buf-
fer as possible.
6. The stability of the signal after adding the reading buffer
decreases with time (around 20 % the first hour). Read within
the first 2 h after adding the reading buffer to the plate.
7. In order to increase the sensitivity of the assay, some modifica-
tions in the protocol can be done:
(a) Increasing incubation time (up to 4 h with shaking and at
least 12 h without shaking).
(b) Increase the sample volume up to 50 μl. In this case it is
very important to get a more efficient shaking or longer
incubation time. Also the concentration of the detection
antibody must be adjusted to the larger volume of
sample.
(c) Wash sample prior the addition of the detection antibody.
This step is useful if you are using larger volumes of sam-
ple. In this way you avoid a dilution of the detection anti-
body. CAUTION: Not all the cytokines will improve
with this step, but also some could even lose some
sensitivity.
Acknowledgement
This work has been supported by a grant provided by the Swedish

Research Council.
References
1. Allan SM, Rothwell NJ (2001) Cytokines and 3. Deverman BE, Patterson PH (2009) Cytokines
acute neurodegeneration. Nat Rev Neurosci and CNS Development. Neuron 64:61–78
2:734–744 4. Peck A, Mellins ED (2009) Plasticity of T-cell
2. Leonard WJ (2001) Cytokines and immunode- phenotype and function: the T helper type 17.
ficiency diseases. Nat Rev Immunol 1:200–208 Immunology 129:147–153
Part V
Analysis of Microglia Activation

Chapter 12
Analysis of Microglial Production of Reactive

Oxygen and Nitrogen Species
Urte Neniskyte and Guy C. Brown
Abstract
Reactive oxygen and nitrogen species are both regulators and effectors of microglial activation, and assays of
these oxidants can be used as a measure of acute and chronic activation of microglial cells. Here we describe
quick methods to assess the production of superoxide, hydrogen peroxide, nitric oxide, and peroxynitrite
by microglia.
Key words Reactive oxygen species, Reactive nitrogen species, Superoxide, Hydrogen peroxide, Nitric
oxide, Peroxynitrite, Phagocytic NADPH oxidase, Inducible nitric oxide synthase, 3-Nitrotyrosine
Abbreviations
iNOS Inducible NOS
LPS Lipopolysaccharide
LTA Lipoteichoic acid
nNOS Neuronal NOS
NO Nitric oxide
NOS Nitric oxide synthase
PHOX Phagocytic NADPH oxidase
RNS Reactive nitrogen species
RONS Reactive oxygen and nitrogen species
ROS Reactive oxygen species
1 Introduction
Reactive oxygen species (ROS) and reactive nitrogen species (RNS)

are molecules derived from oxygen and nitric oxide, respectively,
that are reactive, i.e., they react directly with other molecules with-
out requiring catalysis by enzymes. Reactive oxygen and nitrogen
species (RONS) include superoxide (O2−), hydrogen peroxide
103
104 Urte Neniskyte and Guy C. Brown
(H2O2), nitric oxide (NO), and peroxynitrite (OONO−). RONS are

both regulators and effectors of inflammation. Inflammatory
agents (such as LPS, β-amyloid, ATP, and cytokines) acutely activate
the phagocyte NADPH oxidase (PHOX, found mainly in microglia
but also in astrocytes and some neurons) to produce superoxide,
which then dismutates to hydrogen peroxide. This hydrogen per-
oxide contributes to the inflammatory activation of the microglia
through NF-κB, resulting in the expression of inflammatory
proteins such as inducible nitric oxide synthase (iNOS). iNOS
when expressed produces high levels of NO, which may react with
oxygen to produce NO2 and thence nitrite and nitrate or react with
superoxide to produce peroxynitrite. Thus, measurements of ROS
can be used as a measure of acute activation of microglia (seconds
to minutes), while measurements of iNOS, NO, or its relatively
stable product nitrite can be used as measurements of chronic acti-
vation of microglia (hours to days).
Superoxide production within cells can be assessed by nitro-blue
tetrazolium reduction, leading to the formation of a dark formazan
precipitate [1, 2]. Meanwhile extracellular superoxide may be
detected by the reduction of cytochrome c added to the medium
[3]. Extracellular superoxide readily dismutates into hydrogen
peroxide, whose formation rate can be measured in a continuous
Amplex Red fluorometric assay [4].
Nitric oxide production can be assessed either directly by NO
electrode [5] or indirectly by measurement of nitrite (a stable
breakdown product of nitric oxide) by the Griess reaction [6] or
the more sensitive fluorometric 2,3-diaminonaphthalene assay [7].
In addition, nitric oxide synthase activity can be visualized by
NADPH diaphorase staining [6].
Rates of peroxynitrite production can be determined by
peroxynitrite-mediated oxidation of nonfluorescent dihydrorhoda-
mine-1,2,3 to fluorescent rhodamine-1,2,3 [5]. Sites of peroxyni-
trite production and reaction can be visualized by 3-nitrotyrosine
immunostaining, detecting tyrosine residues in proteins that have
reacted with peroxynitrite [6].
2 Materials
2.1 Superoxide 1. Nitro-blue tetrazolium (Sigma).

Production Measured 2. Phosphate buffered saline (PBS): 10 mM phosphate, 150 mM
with NBT sodium chloride, pH 7.4 (available in tablets and solutions or
made from scratch).
3. (Optional) 2 M potassium hydroxide (KOH).
4. (Optional) Dimethyl sulfoxide (DMSO).
Microglial NO, ONOO-, O2- & H2O2 105
2.2 Superoxide 1. Buffered salt solutions:

Production Measured (a) Krebs-HEPES: 1 mM CaCl2, 11 mM glucose, 25 mM
with Cytochrome c HEPES, 4.8 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4,
118 mM NaCl, pH 7.4.
(b) Hanks buffered salt solution (HBSS): 1.2 mM CaCl2,
0.5 mM MgCl2, 0.4 mM MgSO4, 0.5 mM KCl, 0.4 mM
KH2PO4, 138 mM NaCl, 0.3 mM Na2HPO4, 5 mM glu-
cose, pH 7.4 supplemented with 5 mM glucose (see Note 1).
2. Cytochrome c (Sigma).
3. Catalase from human erythrocytes (Sigma C3556).
4. Superoxide dismutase (Sigma).
2.3 Cellular H2O2 1. Amplex Red (Invitrogen).

Production Measured 2. Dimethyl sulfoxide (DMSO).
by Amplex Red Assay
3. Horseradish peroxidase (Sigma).
4. HBSS: 1.2 mM CaCl2, 0.5 mM MgCl2, 0.4 mM MgSO4, 0.5 mM
KCl, 0.4 mM KH2PO4, 138 mM NaCl, 0.3 mM Na2HPO4,
5 mM glucose, pH 7.4 supplemented with 5 mM glucose.
2.4 Direct 1. Krebs buffer: 126 mM NaCl, 2.5 mM KCl, 25 mM NaHCO3,

Measurement 1.2 mM NaH2PO4, 1.2 mM MgCl2, 2.5 mM CaCl2, 10 mM
of NO Generation glucose, pH 7.2 (see Note 2).
Using a NO Electrode 2. L-arginine (Sigma).
3. Clark-type NO electrode (World Precision Instruments, WPI).
4. (Optional) Oxygen electrode (Rank Brothers) (see Note 3).
2.5 Assessment 1. (Optional) Nitrate reductase (NAD[P]H) from Aspergillus

of NO Generation niger (Sigma).
by Griess Reaction 2. (Optional) NADPH (Sigma).
3. Sulfanilamide: 2 mM in 1.2 M hydrochloric acid (Sigma).
4. N-1-(1-naphthyl)ethylenediamine: 3 mM in double-distilled
water (Sigma).
5. Nitrite standard solution (Sigma).
2.6 Assessment 1. 2,3-Diaminonaphthalene: 0.05 mg/ml in 0.62 M hydrochlo-

of NO Generation ric acid (Sigma).
by Fluorometric 2. NaOH: 0.28 M.
2,3-Diaminonaph-
3. Nitrite standard solution (Sigma).
thalene Assay
2.7 Diaphorase 1. 4 % Paraformaldehyde (PFA) in PBS: 10 mM phosphate,

Staining as a Measure 150 mM sodium chloride, pH 7.4 (available in tablets and
of iNOS Expression solutions or made from scratch) (see Note 4).
2. PBS.
3. Triton X-100 solution: 0.3 % in PBS (see Note 5).
4. NADPH (1 mg/ml) and nitro-blue tetrazolium (0.2 mg/ml)
solution in PBS with 0.3 % Triton X-100.
2.8 Peroxynitrite 1. Dihydrorhodamine-1,2,3 (Sigma).

Production Measured 2. Dimethyl sulfoxide (DMSO).
by Dihydrorhodamine
3. Krebs-HEPES buffer: 1 mM CaCl2, 11 mM glucose, 25 mM
Oxidation
HEPES, 4.8 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4,
118 mM NaCl, pH 7.4 (Sigma).
2.9 Peroxynitrite 1. PBS: 10 mM phosphate, 150 mM sodium chloride, pH 7.4

Production Measured (available in tablets and solutions or made from scratch).
by 3-Nitrotyrosine 2. 4 % Paraformaldehyde (PFA) in PBS (see Note 4).
Immunocytochemistry
3. Anti-nitrotyrosine monoclonal antibody (Upstate, Millipore).
4. Secondary antibody conjugated with horseradish peroxidase
or fluorescent tag (e.g., secondary antibodies from Jackson
ImmunoResearch Laboratories).
3 Methods
3.1 Superoxide 1. Dissolve nitro-blue tetrazolium (NBT) in double-distilled

Production Measured water to obtain 10 mg/ml concentration.
with NBT 2. At specific time points after stimulation of cells, add 30 μl of
NBT solution to 270 μl of medium taken from individual wells
of cultures in 24-well plate to obtain final concentration of
1 mg/ml. This will help to avoid local NBT concentration
differences.
3. Remove the remaining medium from the wells.
4. Add NBT solution to the respective wells and incubate for
30 min at 37 °C, 5 % CO2. Superoxide within cells reduces
nitro-blue tetrazolium, leading to the formation of a dark
formazan precipitate.
5. Wash cells with PBS.
6. The number of formazan-positive cells is counted under a
transmission light microscope.
7. (Optional) Alternatively, the reduction of nitro-blue tetrazolium
may be measured in a plate reader. Allow cells to dry in the air.
Dissolve intracellular formazan deposits with 240 μl of potas-
sium hydroxide and 280 μl of dimethyl sulfoxide with gentle
shaking for 10 min at room temperature. Transfer the dissolved
formazan solution to a 96-well plate and read the absorbance
at 630 nm.
3.2 Superoxide 1. Suspend cells at a density of about 106 cells/ml in appropriate

Production Measured medium (see Notes 1 and 6).
with Cytochrome c 2. Dissolve 60 mg of cytochrome c in 1 ml of double-distilled
water to obtain stock solution of 5 mM. Add 10 μl of cyto-
chrome c solution to 1 ml of cell suspension to obtain final
cytochrome c concentration of 50 μM.
3. (Optional) Add 100 U/ml catalase to prevent reoxidation of
reduced cytochrome c by H2O2 (but remember that catalase is
also a hemoprotein).
4. Stir the cells before and during measurement in a spectropho-
tometer or plate reader, which ideally should be thermostated
at 37 °C (see Note 7).
5. Superoxide directly reduces oxidized (Fe3+) cytochrome c to
reduced (Fe2+) cytochrome c, which absorbs light at 550 nm,
while oxidized cytochrome c has little absorbance at this wave-
length. Follow the reduction continuously, measuring the absor-
bance at 550 nm. If you have a dual-wavelength or multi-wavelength
spectrophotometer or plate reader, measure absorbance at both
550 and 540 nm (where there is no change in absorbance on
reduction of cytochrome c), and subtract the absorbance at
540 nm from the absorbance at 550 nm to correct for any absor-
bance changes not due to cytochrome c reduction.
6. Repeat measurements with and without 50 U/ml superoxide
dismutase to ensure that the signal is due to extracellular super-
oxide only.
3.3 Cellular H2O2 1. Suspend the cells in HBSS with glucose at a density of
Production Measured 3.5 × 105 cells/ml.
by Amplex Red Assay 2. Dissolve 5 mg of Amplex Red reagent in 1.94 ml of DMSO
to obtain stock solution of 10 mM. Prepare 1,000 U/ml of
horseradish peroxidase solution in double-distilled water.
3. Add 10 μl of Amplex Red and 10 μl of horseradish peroxidase
to 1 ml of cell suspension to obtain final concentration of
100 μM and 10 U/ml, respectively.
4. Horseradish peroxidase catalyzes H2O2 oxidation of Amplex
Red to fluorescent resorufin. Measure the rate of hydrogen
peroxide production in a stirred cuvette in a spectrofluoropho-
tometer or in a plate reader (use black plates to avoid reading
fluorescence from adjacent wells), which ideally should be
thermostated at 37 °C. Use excitation wavelength of 560 nm
and emission wavelength of 587 nm. H2O2 passes rapidly
through membranes, so extracellular H2O2 reflects both extra-
cellular and intracellular production and breakdown.
5. Repeat the measurements with and without cells as well as with
and without catalase to ensure that fluorescence is due to
microglial H2O2 production only.
3.4 Direct 1. Prepare activated cell suspension in Krebs buffer at a density

Measurement of NO between 105 and 106 cells/ml (see Notes 8–10).
Generation Using a NO 2. Add L-arginine (200 μM) to see NO generation. Glucose
Electrode (10 mM, a component of Krebs buffer) is required for the cells
to be able to produce NADPH.
3. Put cells in a stirred, thermostated, and sealed incubation cham-
ber (supplied by World Precision Instruments, WPI) with the
NO electrode. NO readily passes through cell membranes and
therefore NO from cellular iNOS is measurable by this method.
Before adding the cells, it is necessary to have a steady baseline
reading from the NO electrode. For this, all components of the
system (including particularly the electrode) need to come to
thermal equilibrium. Note that warming up medium or the elec-
trode can cause small bubbles to form, which can cause electrode
noise. After the measurement add a small amount of hemoglobin
to remove all NO from the chamber so that the location of the
electrode baseline can be reestablished.
4. Calibrate the NO electrode using either NO-saturated water
(NO gas bubbled through distilled water in gas-tight vial inside
a gas hood) or acidified nitrite (see WPI manual).
5. (Optional) It may be advisable to combine the measurement
with an oxygen electrode (see Note 3).
3.5 Assessment 1. (Optional) To avoid the underestimation of NO production, it

of NO Generation may be advisable to convert any nitrate formed to nitrite by
by Griess Reaction incubating cell-culture supernatants with nitrate reductase
(0.5 U/ml) and NADPH (80 μM) for 30 min at room tem-
perature. However, nitrate is not specific to NO production and
may arise from other sources. Therefore, the necessity of this
step has to be determined empirically.
2. In a 96-well plate, mix 50 μl of cell-culture supernatant with
50 μl of ice-cold sulfanilamide in hydrochloric acid and incubate
for 10 min at room temperature, protected from light.
3. Add 50 μl of N-1-(1-naphthyl)ethylenediamine and incubate the
mixture for another 10 min at room temperature, protected
from light.
4. Measure the absorbance of the resulting red azo dye in a plate
reader at 550 nm.
5. Calculate nitrite concentrations using standards prepared from
nitrite standard solution.
3.6 Assessment 1. Mix 100 μl of cell-culture supernatants with 20 μl of

of NO Generation 2,3-diaminonaphthalene solution in hydrochloric acid and
by Fluorometric incubate for 10 min at room temperature, protected from light.
2,3-Diaminonaph- 2. Add 100 μl of NaOH to stop the reaction.
thalene Assay
3. Measure fluorescence intensity in a spectrofluorophotometer

or a plate reader at an excitation wavelength of 363 nm and an
emission wavelength of 426 nm.
4. Calculate nitrite concentrations using standards prepared from
nitrite standard solution.
3.7 Diaphorase 1. After stimulation fix cells with 4 % PFA in PBS for 30 min at 4 °C,
Staining as Measure followed by three washes with PBS, 15 min each.
of iNOS Expression 2. Permeabilize fixed cells with 0.3 % Triton X-100 for 5 min at
room temperature.
3. Incubate cells in NADPH and nitro-blue tetrazolium solution
with Triton X-100 for 2 h at 37 °C. Nitric oxide synthase acts
as a NADPH diaphorase that reduces the chromogen nitro-
blue tetrazolium to form a dark formazan precipitate.
4. After incubation wash cells once with PBS and detect cells with
NOS activity under a transmission light microscope. nNOS-
expressing neurons have a light staining, whereas iNOS-express-
ing glia can be very dark.
3.8 Peroxynitrite 1. Suspend cells in pre-warmed (37 °C) Krebs-HEPES buffer at

Production Measured a density of 106 cells/ml.
by Dihydrorhodamine 2. Prepare 5 mM dihydrorhodamine stock solution in DMSO.
Oxidation
3. Add 10 μM of dihydrorhodamine stock solution to 1 ml of cell
suspension to obtain final 5 μM concentration.
4. Measure the rate of continuous fluorescence increase in a stirred
cuvette in a spectrofluorophotometer or a plate reader at an
excitation wavelength of 500 nm and an emission wavelength of
536 nm. Peroxynitrite mediates the oxidation of nonfluorescent
dihydrorhodamine-1,2,3 to fluorescent rhodamine-1,2,3, and
the rate of fluorescence increase is equal to the rate of per-
oxynitrite production (see Note 11).
3.9 Peroxynitrite 1. Fix stimulated cells with 4 % PFA in PBS for 20 min at room
Production Measured temperature, followed by three washes with PBS, 15 min each.
by 3-Nitrotyrosine 2. Incubate cells in PBS with 5 % of normal serum of the host
Immunocytochemistry species of the secondary antibody to block unspecific epitopes.
3. Incubate fixed cells with 10 μg/ml of anti-nitrotyrosine mono-
clonal antibody (1:100 dilution in PBS) for 1 h at room tempera-
ture. Wash three times with PBS, 15 min each.
4. Detect primary antibody with secondary antibody conjugated
with horseradish peroxidase or fluorescent tag (1:200 dilution
in PBS, 30 min at room temperature).
5. Visualize 3-nitrotyrosine-positive cells under a transmission
light microscope (for horseradish peroxidase immunocyto-
chemistry) or fluorescence microscope (for fluorescent secondary
antibodies) (see Note 12).
4 Notes
1. Any other medium that is not colored (i.e., without phenol

red), does not scatter light, does not scavenge superoxide, and
does not reduce cytochrome c (i.e., without ascorbate) can be
used as well.
2. It is best to use a relatively simple incubation medium such as
Krebs buffer rather than a cell culture medium such as DMEM,
as components of the latter including riboflavin react with NO
in the presence of light.
3. Rank Brothers supply oxygen electrode and chamber. Drill a
2 mm hole through the chamber top to allow the NO elec-
trode in and seal it carefully, because the chamber needs to be
gas tight and bubble free.
4. To dissolve 4 g of PFA in 100 ml of PBS, add 1 ml of 1 M
NaOH to increase pH and stir the solution in the water bath at
70 °C until it is clear. Filter the solution through the filter
paper and adjust the pH to 7.4. Aliquot 4 % PFA to avoid
repeated freezing–thawing cycles and store it at −20 °C.
5. Since Triton X-100 is very viscous, it is easier to measure it by
weight rather than volume.
6. In a plate reader, extracellular superoxide production by adher-
ent cells may be measured as well.
7. At this density the cells may run out of oxygen after a few
minutes—it may be an idea to saturate the medium with
95/100 % oxygen, but bear in mind that superoxide produc-
tion may depend on the oxygen level.
8. Direct NO measurement by NO electrode is difficult or impos-
sible for cells expressing nNOS or eNOS, but relatively easy for
any cells expressing iNOS (such as microglia), because iNOS
produces a lot more NO continuously.
9. For example, cells may be activated with 100 ng/ml LPS and
10 ng/ml IFN-γ. Addition of IFN-γ ensures strong iNOS
expression, which peaks between 8 and 24 h after activation.
10. It may be advisable to use arginine-free culture medium for the
activation or add arginase to the medium, because the NO
released by iNOS may damage the cells and limit iNOS
expression.
11. Other intra- and extracellular oxidants can oxidize dihydrorho-
damine; therefore, it is important to include controls with spe-
cific peroxynitrite scavengers, such as urate (100 μM) or
FeTPPS (5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrinato
iron (III) chloride, 25 μM).
12. To determine the site of peroxynitrite production and action,

3-nitrotyrosine immunostaining can be combined with staining
for markers of specific cell types, such as isolectin B4 (Invitrogen)
or antibodies against OX-42 to label microglia.
Acknowledgement
Relevant research in our laboratory has been funded by the
Wellcome Trust (Grant RG50995).
References
1. Wyss-Coray T (2006) Inflammation in 5. Bal-Price A, Matthias A, Brown GC (2002)
Alzheimer disease: driving force, bystander or Stimulation of the NADPH oxidase in activated
beneficial response? Nat Med 12:1005–1015 rat microglia removes nitric oxide but induces
2. Ransohoff RM, Perry VH (2009) Microglial peroxynitrite production. J Neurochem 80:
physiology: unique stimuli, specialized 73–80
responses. Annu Rev Immunol 27:119–145 6. Mander P, Brown GC (2005) Activation of
3. Colton CA, Gilbert DL (1987) Production of microglial NADPH oxidase is synergistic with
superoxide anions by a CNS macrophage, the glial iNOS expression in inducing neuronal
microglia. FEBS Lett 223:284–288 death: a dual-key mechanism of inflammatory
4. Jekabsone A, Mander P, Tickler A et al (2006) neurodegeneration. J Neuroinflammation 2:20
Fibrillar beta-amyloid peptide Abeta1-40 acti- 7. Jekabsone A, Neher J, Borutaite V et al (2007)
vates microglial proliferation via stimulating Nitric oxide from neuronal nitric oxide synthase
TNF-alpha release and H2O2 derived from sensitises neurons to hypoxia-induced death via
NADPH oxidase: a cell culture study. J competitive inhibition of cytochrome oxidase.
Neuroinflammation 3:24 J Neurochem 103:346–356
Chapter 13
Quantification of Active Caspase-3 and Active

Caspase-8 in Microglia Cells
Edel Kavanagh
Abstract
During microglia activation the levels of active caspase-3, caspase-7 and caspase-8 are increased, which
leads to the transcription of proinflammatory cytokines and factors. As such, the induction of caspase activity
in microglia can be used as a marker for activation. The use of sensitive and quantitative techniques has
made it possible to reproducibly detect these low levels of active caspases. This chapter outlines the materials
and methodology for three different ways to detect caspase activation in microglia.
Key words Flow cytometry, Luminometry, Fluorometry, Caspase-3, Caspase-8, Antibodies,

Luminogenic substrate, Fluorogenic substrate
1 Introduction
Cysteine-dependent aspartate proteases (caspases) contain a cysteine

in their active site, which is required for their activity, and cleave
their substrates after an aspartate residue preceded by a recognition
sequence [1]. Currently, there are 14 caspases described in the
mammalian genome, with a variety of signalling functions including
apoptosis, differentiation, and inflammatory response [2]. During
the activation of microglia, the levels of active caspase-1, caspase-3,
caspase-7, and caspase-8 increase [3]. The activity of each specific
caspase, or specific groups of caspases, can be experimentally calcu-
lated based on their substrate recognition sequences [4]. Active
caspase-1 cleaves its substrates after the peptide sequence YVAD;
caspase-3 and caspase-7 both share the same substrate recognition
sequence of DEVD, while active caspase-8 cleaves after IETD [4].
This characteristic of caspases is harnessed to experimentally measure
activity in the first two methods described in this chapter.
All caspases have a similar structure, consisting of an N-terminal
prodomain, a large subunit, and a small subunit. During activation,
cleavage occurs between the large and small subunits, followed by
cleavage between the prodomain and the large subunit [5]. A tetramer
113
114 Edel Kavanagh
consisting of two large and two small subunits comprises the active
caspase [5]. During activation, the caspase subunits which undergo
cleavage and conformational changes expose new epitopes. The
final method to measure caspase activity described in this chapter is
based on antibodies specifically designed to bind to these newly
exposed epitopes after caspase activation.
1.1 Chemilumine- In this assay, active caspases cleave a luminogenic substrate con-
scence Detection taining their specific substrate recognition sequence. Following
of Caspase Activity cleavage, a substrate for luciferase (aminoluciferin) is released and
reacts with luciferase to produce light. This assay can be used to
detect either DEVDase activity (for caspase-3 and caspase-7) or
IETDase activity (for caspase-8 and caspase-10) (see Note 1). The
reagents described in this assay are supplied as a kit called Caspase-
Glo from Promega.
1.2 Caspase Assay The activity of caspases can also be determined fluorometrically.
Using a Fluorescent A peptide substrate is synthesized with an attached fluorescent tag,
Caspase Substrate acetyl-L-aspartyl-L-glutamyl-L-valyl-L-aspartic acid α-(4-methyl-
coumaryl-7-amide) (DEVD-MCA). When the peptide substrate is
cleaved, the fluorescent 7-amino-4-methylcoumarin (AMC) is
released and can be measured over time to calculate the enzymatic
activity of the caspase.
1.3 Quantification Antibodies, specific for active or cleaved caspases, have been devel-
of Cleaved Caspases oped against the amino acid residue which is only exposed after
by Flow Cytometry cleavage between the large and small caspase subunits. Coupled
with the sensitivity, quantitative power, and speed of flow cytome-
try, antibodies against cleaved caspases can be used to accurately
measure small differences in levels of cleaved caspases.
2 Materials
2.1 Chemilumine- 1. Caspase-Glo 3/7 reagent or Caspase-Glo 8 reagent (Promega).

scence Detection of 2. White-walled 96-well plate.
Caspase Activity
3. Plate reader with chemiluminescence detection capability, e.g.,
Wallace plate reader.
4. Hemocytometer.
2.2 Caspase Assay 1. Ac-Asp-Glu-Val-Asp-α-(4-methyl-coumaryl-7-amide) (DEVD-

Using a Fluorescent AMC) and acetyl-Ile-Glu-Thr-Asp-4-methylcoumaryl-7-amide
Caspase Substrate (IETD-MCA) available from companies such as Peptide Institute
or BD Pharmingen.
2. Black 96-well microtiter plates.
Measuring Caspase Activity in Microglia Cells 115
3. Plate reader with the correct excitation and emission filters,

e.g., Wallace.
4. Substrate buffer (100 mM N-2-hydroxyethyl-piperazine-N′-
2-ethanesulphonic acid (HEPES) pH 7.0, 10 % sucrose, 0.1 %
3-[(3-cholamidopropyl)-dimethylammonio]-1-
propanesulfonate (CHAPS)).
5. 5 mM D dithiothreitol (DTT).
6. 0.0001 % Nonidet P-40 (make a working stock of 0.1 %).
7. Phosphate buffered saline (PBS 10×: 1.37 M NaCl, 27 mM
KCl, 100 mM Na2HPO4, 20 mM KH2PO4).
8. Liquid nitrogen.
2.3 Quantification 1. Primary antibodies against cleaved caspase-3 (Cell Signalling cat
of Cleaved Caspases no. 9664 against mouse, human, and rat) and cleaved caspase-8
by Flow Cytometry (Cell Signalling cat no. 8592 mouse specific or cat no. 9496
human specific).
2. Fluorescent secondary antibodies (must be compatible with
flow cytometer lasers, such as Alexa488).
3. Ice-cold methanol.
4. Phosphate buffered saline (PBS 10×: 1.37 M NaCl, 27 mM
KCl, 100 mM Na2HPO4, 20 mM KH2PO4).
5. Blocking buffer (0.5 % BSA dissolved in PBS).
6. 4 % Paraformaldehyde (dissolved in PBS).
7. Water bath set to 37 °C.
8. Centrifuge.
9. Vortex.
10. Orbital shaker.
11. Flow cytometer.
12. Flow cytometer tubes.
3 Method
3.1 Chemilumine- 1. Equilibrate the Caspase-Glo buffer and lyophilized Caspase-Glo

scence Detection of substrate to room temperature before use.
Caspase Assay 2. Transfer the contents of the Caspase-Glo buffer bottle into the
bottle containing Caspase-Glo substrate. Mix until the substrate
is thoroughly dissolved to form the Caspase-Glo reagent.
3. Store the reconstituted Caspase-Glo reagent at 4 °C and use
within 4 weeks.
4. Allow the Caspase-Glo reagent to equilibrate to room tem-
perature. Mix well.
116 Edel Kavanagh
Table 1
Generation of an AMC standard curve
nmol AMC AMC solution (μl) Substrate buffer (μl)

0 0 100
0.1 5 95
0.2 10 90
0.5 25 75
1.0 50 50
1.5 75 25
2.0 100 0
5. Grow and perform treatments on 100 μl cells in white-walled

96-well plates. Remove 96-well plates containing cells from
the incubator. Alternatively, grow and treat in different-sized
plates but detach cells from those plates and transfer to the
white-walled 96-well plates.
6. Add 100 μl of Caspase-Glo reagent to each well of a white-
walled 96-well plate containing 100 μl treated cells in culture
medium. Because of the sensitivity of this assay, be careful not
to touch pipette tips to the wells containing samples to avoid
cross-contamination. Cover the plate with a plate sealer or lid
(see Note 2).
7. Gently mix contents of wells using a plate shaker at 300–
500 rpm for 30 s. Incubate at room temperature for 1 h.
8. Measure the luminescence of each sample in a plate-reading
luminometer as directed by the luminometer manufacturer (see
Notes 3 and 4).
3.2 Caspase Assay 1. Set up a program on the plate reader following the manufac-
Using a Fluorescent turer’s instructions as follows: Assay temperature at 37 °C,
Caspase Substrate the number of plate reads at 25, delay between reads is 1 min,
excitation filter is 355–380 nm, and emission filter is
430–460 nm.
2. To generate an AMC standard curve, dissolve 7-amino-4-
methylcoumarin (AMC) in substrate buffer to a final concentra-
tion of 20 μM. Add increasing concentrations of AMC solution
to substrate buffer in wells of a black 96-well plate (Table 1).
Measure the fluorescence generated as relative fluorescence
units (RFU). Plot RFU (y-axis) against substrate concentration
(x-axis) and draw a linear standard curve. Record the RFU
value generated with 1 nmol AMC.
3. 2 × 105 cells are needed for this assay per sample.
4. After the appropriate treatment, scrape cells into culture medium

and centrifuge the resulting cell suspension at 1,000 × g for 5 min
to pellet the cells. Discard the supernatant.
5. Wash cell pellet once in 1 ml PBS and centrifuge at 1,000 × g
for 5 min. Discard the supernatant.
6. Resuspend the cell pellet in 50 μl PBS. Lyse cells in eppendorfs
by snap freezing in a bath of liquid nitrogen. The suspension
can be stored at −80 °C at this stage until further analysis.
7. Thaw the cell suspensions on ice and divide equally between
two wells of a black microtiter plate.
8. Just before use, add 5 mM DTT, 0.0001 % Nonidet P-40, and
50 μM of DEVD-MCA/IETD-MCA/YVAD-MCA to the
substrate buffer. Store in the dark and on ice. Use within
30 min.
9. When the plate reader temperature reaches 37 °C, add the
substrate buffer to the samples and start the program imme-
diately. Assay will take approximately 30 min to complete
(see Note 5).
10. Using the assay results, the fluorescent units (RFU, y-axis) vs.
time (min, x-axis) for each sample are graphed as an XY scatter
plot. Calculate the rate at which fluorescence (free AMC)
increases over time using the linear portion of the curve. This
value is the rate of AMC released/min.
11. RFU are converted to nmoles of AMC released per minute
using the standard curve generated with free AMC. Divide
value from the previous step by the RFU value which corre-
sponds to 1 nmol AMC. The new value obtained is nmol AMC
released/min.
12. Protein concentration in each sample is calculated using a
standard protein assay such as Bradford assay. Divide the previ-
ous value by the amount of protein (in mg) present in each
sample. The new value obtained is nmoles AMC released/
min/mg, which corresponds to the caspase activity in that
sample (see Note 6).
3.3 Quantification Carry out all procedures at room temperature unless otherwise
of Cleaved Caspases specified.
by Flow Cytometry
1. Grow and treat microglia cells (2 × 105).
2. Detach cells from plates/flasks by pipetting and transfer cells
with medium to 15 ml tubes. Collect cells by centrifugation at
1,000 × g for 5 min at 4 °C and discard supernatant.
3. Resuspend cell pellet in 0.5 ml of 4 % paraformaldehyde and
incubate tubes at 37 °C for 10 min.
4. Place tubes on ice for 1–2 min.
118 Edel Kavanagh
5. Using the vortex at a low speed, slowly add 4.5 ml ice-cold

methanol to the fixed cells to achieve a final methanol concentra-
tion of 90 % (see Note 7). Incubate the cells on ice for 30 min.
Alternatively, the procedure can be stopped here by storing the
cells at −20 °C (see Note 8).
6. Centrifuge cells at 1,500 × g for 5 min and discard
supernatant.
7. Wash cell pellet in 1 ml blocking solution (freshly prepared) and
centrifuge cells at 1,500 × g for 5 min. Discard supernatant.
Repeat this wash step once (see Note 9).
8. Resuspend pellet in 100 μl blocking solution (see Note 10)
and incubate at room temperature for 10 min.
9. Dilute the primary antibody in blocking solution to a concen-
tration of 1:1,600 for anti-cleaved caspase-3, 1:800 for cleaved
caspase-8 (mouse specific), or 1:100 for cleaved caspase-8
(human specific).
10. Add 100 μl primary antibody solution to cells and incubate
with gentle agitation for 1 h.
11. Wash cell pellet in 1 ml blocking solution and centrifuge cells
at 1,500 × g for 5 min. Discard supernatant. Repeat this wash
step once.
12. Dilute the appropriate secondary antibody in blocking buffer
at 1:200 concentration (see Notes 11–14).
13. Add 100 μl antibody solution to each sample and incubate in
the dark with gentle agitation for 30 min.
14. Wash cell pellet in 1 ml blocking solution and centrifuge cells
at 1,500 × g for 5 min. Discard supernatant. Repeat this wash
step once.
15. Resuspend pellet in 0.5 ml PBS and transfer to flow cytometry
tubes. Analyze immediately by flow cytometry or store in the
dark at 4 °C for up to 1 week.
16. Display the acquired data in the format of a histogram overlay.
A cell population which is undergoing apoptosis will display
two peaks, one corresponding to the healthy population
(no expression of cleaved caspase-3) and one corresponding to
the dying population (high expression of cleaved caspase-3).
On the other hand, activated microglia display only one peak,
but the whole population shows a low level of cleaved caspase-3
expression and appears shifted to the right when compared
with unstimulated microglia (Fig. 1) (see Note 15).
17. K/S statistics on the histogram peaks can be performed to
measure the difference between two experimental samples of
microglia.
Fig. 1 Histogram showing the distribution of cleaved caspase-3 in microglia cells

upon lipopolysaccharide (LPS) or staurosporine (STS) treatment. BV2 mouse
microglia cells were treated with 1 μM LPS for 6 h or 0.2 μM STS for 3 h. Cells
were stained with cleaved caspase-3 (Cell Signalling cat. no. 9664) and analyzed
by flow cytometry (FACS Calibur, Becton Dickinson)
4 Notes
1. In human cells IETDase activity assays also measure the activ-

ity of caspase-10, which is not present in mouse cells.
2. The assay works equally well using as little as 50 μl Caspase-
Glo reagent mixed with 50 μl cells in culture medium.
3. The luminescence values obtained must be normalized to the
number of cells in each sample. This can be done in two ways.
One way is to count the number of cells in each sample using
a hemocytometer or cell counter. Alternatively, lyse the remain-
der of the cells from the sample wells, extract the protein, and
perform a protein assay. Then divide the luminescence values
by the protein concentration.
4. As a control, measure the background luminescence from the
cell culture medium (fresh unused medium) and subtract that
value from the experimental samples.
5. Since there are no protease inhibitors in the lysis buffer, it is
vital that before the assay begins, the samples are kept on ice
and placed in the plate reader only when it has reached 37 °C.
6. When calculating the rate of free AMC over time, ensure that
only the linear portion of the curve is used. At the start of the
assay, there will be a lag phase with little or no cleavage of
substrate while the reagents reach the optimal temperature.
120 Edel Kavanagh
At the end of the assay, the substrate or enzyme may be

completely used up causing a plateau in activity.
7. For convenience, store a stock of 100 % methanol at −20 °C.
8. Alternative permeabilization step. After fixation, the cells can
be centrifuged at 1,500 × g and paraformaldehyde discarded.
Then the pellet can be resuspended in 1 ml methanol. The
advantage of this is the rest of the protocol can be carried out
in 1.5 ml tubes instead of 15 ml tubes.
9. There is only one wash step between antibody incubations to
avoid loss of cells.
10. Blocking solution can be prepared in advance and stored at
4 °C for up to 1 week. Discard if the solution is cloudy.
11. Ensure the fluorescent secondary antibodies are compatible
with the lasers of the flow cytometer. Information can be found
on the BD Biosciences website (http://www.bdbiosciences.
com/research/multicolor/spectrum_viewer/index.jsp).
12. Incubate one sample with a primary antibody isotype control
instead of the cleaved caspase primary antibody to show nonspe-
cific binding of the primary antibody.
13. Incubate one sample with just secondary antibody to show
nonspecific binding of the secondary antibody.
14. Leave one sample unstained to show autofluorescence of the cells.
15. An advantage of this method is the ability to detect and quantify
the cleavage of each specific caspase, compared with DEVDase
or IETDase activity assays, which measures both caspase-3 and
caspase-7 or caspase-8 and caspase-10, respectively. Another
advantage is the ability to quantify the relative quantities of
cleaved caspase in each individual cell. As a result, this technique
easily distinguishes apoptotic microglia cells from activated
microglia cells.
References
1. Alnemri ES, Livingston DJ, Nicholson DW, J, Brundin P, Englund E, Venero JL, Joseph B
Salvesen G, Thornberry NA, Wong WW, Yuan (2011) Caspase signalling controls microglia
J (1996) Human ICE/CED-3 protease activation and neurotoxicity. Nature
nomenclature. Cell 87(2):171. doi:S0092- 472(7343):319–324. doi:nature09788
8674(00)81334-3 [pii] [pii]10.1038/nature09788
2. Lamkanfi M, Declercq W, Kalai M, Saelens X, 4. Timmer JC, Salvesen GS (2007) Caspase sub-
Vandenabeele P (2002) Alice in caspase land. strates. Cell Death Differ 14(1):66–72.
A phylogenetic analysis of caspases from worm doi:4402059 [pii] 10.1038/sj.cdd.4402059
to man. Cell Death Differ 9(4):358–361. 5. Pop C, Salvesen GS (2009) Human caspases:
doi:10.1038/sj/cdd/4400989 activation, specificity, and regulation. J Biol
3. Burguillos MA, Deierborg T, Kavanagh E, Chem 284(33):21777–21781. doi:R800084200
Persson A, Hajji N, Garcia-Quintanilla A, Cano [pii] 10.1074/jbc.R800084200
Chapter 14
Quantification of Microglial Phagocytosis

by a Flow Cytometer-Based Assay
Refik Pul, Kandiyil Prajeeth Chittappen, and Martin Stangel
Abstract
Microglia represent the largest population of phagocytes in the CNS and have a principal role in immune
defense and inflammatory responses in the CNS. Their phagocytic activity can be studied by a variety of
techniques, including a flow cytometry-based approach utilizing polystyrene latex beads. The flow
cytometry-based microglial phagocytosis assay, which is presented here, offers the advantage of rapid and
reliable analysis of thousands of cells in a quantitative fashion.
Key words Phagocytosis, Microglia, Polystyrene latex beads, Flow cytometry, Assay
1 Introduction
The first extensive experiments on quantification of phagocytosis

by using solid particles were published in 1912 by the Dutch phys-
iologist Hartog Jacob Hamburger [1]. His usual procedure for
measuring phagocytosis was to incubate suspensions of neutrophils
and carbon mixed in a test tube for a given length of time and to
quantify with a microscope the percentage of neutrophils that
ingested carbon. In 1921 Fenn measured phagocytosis by count-
ing the number of quartz particles phagocytosed per leucocyte [2].
According to Fenn phagocytosis was the result of surface forces
that exist in a system containing phagocytes and particles. Whether
phagocytosis will occur in a given system was held to be governed
by the balance of these forces [2]. Even though the theory of
phagocytosis at that time was purely mechanistic, such simple
quantification methods are still widely in use, for example, by not-
ing the ingestion of cells, such as bacteria, red blood cells, or other
damaged (e.g., apoptotic) cells, as well as noncellular foreign par-
ticulate matter, including crystals, immiscible vacuolated liquids,
and polystyrene latex beads ([3], Fig. 1).
121
122 Refik Pul et al.
Fig. 1 Immunocytochemistry illustrates phagocytosis of fluorescent latex beads

(yellow–green) by microglia (red, labelled with anti-OX42). Nearly all microglia
cells contain beads after 1 h of phagocytosis
Polystyrene latex beads, initially used to produce experimen-

tally regional ischemia by intravascular injection, were used for the
first time by Sbarra and Karnovsky in 1959 to quantify phagocyto-
sis [4, 5]. They observed that, in contrast to starch particles,
polystyrene microspheres do not require a serum or phagocytosis-
promoting factor which the authors ascribed to the lipophilic
nature of these particles [5]. Since iodination (iodine-131) of poly-
styrene beads appeared to be not practicable, Wilkins reported in
1964 that these beads can be tagged with a green fluorescent dye
(“dimethyl-6:12-coeroxenol acetate”) [6]. His method made use
of the fact that during the preparation of polystyrene latex by emul-
sion polymerization, seed particles from an initial polymerization
are used to grow the particles to larger sizes in successive steps.
If an oil-soluble dye dissolved in fresh monomer is added to the
seed suspension during one of these steps, it will be incorporated
into the polymer particles [6]. This paved the way for the use of
fluorescent polystyrene beads in flow cytometric assays which
enabled the facile determination of the percent of phagocytic cells
and bead uptake [7–9]. In this regard, the mean fluorescence
intensity directly correlates with the binding and uptake of beads
because of uniformity in bead size and fluorescence distribution
throughout the outer radii of these beads [8]. Bead uptake was
also found to be dependent on bead size, concentration, and time.
As a general trend, cells incubated with either the 0.2 or 0.5 μm
microspheres resulted in a greater uptake of beads as compared to
Phagocytosis Assay for Microglia 123
the larger 2.0 μm beads [10]. Maximal values were approached

with increasing bead/cell ratios; rates were approximately linear up
to 60 min. Effective inhibitors (4 °C, colchicine, and energy source
deprivation) prevented phagocytosis and uptake was not present
when assayed at zero time [9]. Interestingly, this method was
applied to microglia as recently as 1997 by von Zahn et al., who
showed a slight reduction of the phagocytic activity by TGF-β1
and probably by IL-4 [11]. Since this method provides reliable,
fast, and objective measurements, flow cytometry-based microglial
phagocytosis assay has been used in numerous studies [12–17].
2 Materials
1. Dulbecco’s minimum essential medium supplemented with

10 % fetal bovine serum (FBS) and 1 % penicillin plus strepto-
mycin (DMEM+).
3. PBS containing 2 mM EDTA.
4. Polypropylene, 50-ml sterile, conical centrifuge tubes (Becton
Dickinson Labware).
5. Polystyrene, 5.5-ml tubes (Becton Dickinson Labware).
6. 6-well plate.
7. Fluorescent-labelled Fluoresbrite microspheres, 1 μm diameter
(Polysciences, Inc., Warrington, PA).
8. Trypan blue solution.
9. Propidium iodide solution 1 mg/ml.
10. Flow cytometer.
11. Incubator.
3 Methods
3.1 Plating 1. Microglia isolation is described elsewhere in this volume.

and Preparation The procedure, how microglia are isolated, is not relevant to
of Microglia the phagocytosis assay.
2. Centrifuge isolated microglia at 475 × g for 6 min and discard
the supernatant.
3. Resuspend the pellet in fresh DMEM+ and determine the
viable cell number by trypan blue staining.
4. Adjust the cell density to 2.5 × 105 cells per ml. Add 2 ml of cell
suspension (i.e., 5 × 105 cells) to each well of a 6-well plate and
incubate for 2 h at 37 °C in a 5 % CO2 humidified incubator.
This procedure allows the viable microglia to adhere to the cell
culture plastic while the dead cells remain floating in the

supernatant.
5. Aspirate the supernatants with floating cell debris and add 2 ml
fresh DMEM+ to each well. Return the plates to the incubator
for another 12–16 h.
3.2 Phagocytosis In order to distinguish the fluorescence resulting from active

of Latex Beads phagocytosis or from passive attachment of beads to the cells
surface, the assay is performed at 4 °C and 37 °C in parallel.
1. Considering the cell/beads ratio to be used in the assay,
prepare appropriate dilutions of latex beads in DMEM+
(see Note 1).
2. Remove the 6-well plates seeded with microglia from the incu-
bator and aspirate the medium (see Note 2).
3. Add 1 ml of appropriate bead dilution to the cells in 6-well
plates and incubate at 37 °C in a 5 % CO2 humidified incubator
for 1 h. As a control incubate an additional plate with beads at
4 °C for the same period of time (see Note 3).
4. After this incubation period, place the plates immediately
on ice. Further steps have to be strictly followed at 4 °C
(see Note 4).
5. To remove non-internalized beads, wash the wells by adding
2 ml of PBS and then aspirate it completely. Repeat this washing
step 5 times (see Note 5).
6. At the end of the last washing step, add 2 ml of ice-cold PBS
containing 2 mM ethylenediaminetetraacetic acid (EDTA) to
each well and incubate on ice for 10 min.
7. Of note, microglia strongly adhere to plastic (see Note 6).
To release the cells, vigorously pipette the PBS–EDTA includ-
ing each edge of the well. Collect the cell suspension into 5 ml
polystyrene tubes (see Note 7).
8. Centrifuge these tubes at 475 × g for 6 min at 4 °C.
9. Discard the supernatant and resuspend the cell pellet in 300 μl
of ice-cold PBS containing 2 mM EDTA. Store the samples at
4 °C until further use.
3.3 Flow Cytometric 1. Prior to sample acquisition, adjust the flow cytometer photo-
Analysis multiplier (PMT) voltage settings by using the microglia
sample that does not contain any beads.
2. Just before acquisition, add propidium iodide (PI) to each
sample and vortex thoroughly. Propidium iodide will later
enable discrimination of live and dead cells in data analysis.
3. Acquire and store ~20,000 live events for each sample
(see Note 8).
Fig. 2 Representative flow cytometry histograms display different mean fluorescence intensities (MFI) at 4 and
37 °C. Each peak represents the microglia population which has ingested 1, 2, 3, and 4 beads, respectively. Due
to overlapping, further discrimination of peaks is not possible. However, since the first peak is also detectable at
4 °C, it may just result from beads attached to the cell surface. Thus, the MFI at 37 °C minus the MFI at 4 °C is
the amount of incorporated fluorescent latex particles phagocytosed by a given number of microglia
4. For analysis, histogram profiles are generated for each sample

with log scale of corresponding fluorescence intensity taken on
the x-axis and the cell counts taken on the y-axis (Fig. 2).
5. The final data can be represented either as the percentage of
cells that have phagocytosed the fluorescent beads or as the
mean fluorescence intensity of the gated events. The fluores-
cence observed in the 4 °C control samples is subtracted from
the 37 °C samples to obtain the final result (Fig. 2).
4 Notes
1. Prior to the assay, it is important to determine the optimal

ratio of cell/beads for optimal results. Addition of excess beads
may lead to unwanted “background noise,” whereas insuffi-
cient number of beads may lower the significance of the findings.
We recommend for microglia a cell/beads ratio of 1:100.
2. Of course, other plates, i.e., 12- or 24-well plates, can be used
as well. However, then lower density seeding is required. Cell
numbers may then be insufficient to record an appropriate
number of life events.
3. Regardless of the uptake mechanism, it is considered plausible
that polystyrene beads spend some time on the exterior of the
cell membrane prior to translocation to the intracellular envi-
ronment. To distinguish between uptake and adsorption, other
methods can be used as well:
(a) “Quenching” of adsorbed fluorescent markers with agents
like ammonium acetate or trypan blue was proposed [18, 19].
Fig. 3 Representative z-stack series of confocal microscopic images of two microglia cells (red, labelled with
anti-OX42) definitely demonstrate that the particles were completely internalized and not merely attached to
the outer membrane
(b) Phagocytosis but not adsorption of the beads to the cells

can be completely abolished by cytochalasin D (5 μg/ml),
an inhibitor of actin polymerization [13]. The use of cyto-
chalasin D is a very elegant approach to correct for the
fluorescence/number of the adsorbed beads.
We recommend the use of confocal microscopy for a
qualitative distinction between beads taken up by cells or
merely attached to the plasma membrane ([13], Fig. 3). It
should be noted that fluorescent latex beads may be phago-
cytosed in a different way and at a different rate than other
particle types. Unfortunately, little work has been carried
out on exploiting the mechanism of the intracellular uptake
of polystyrene latex beads. To our knowledge, there is only
one report proposing that CD11c may be involved in the
uptake of non-opsonized polystyrene beads [20].
4. For better results it is important to maintain the cells at 4 °C
for further steps. All solutions used and the centrifuge have to
be precooled and maintained at 4 °C.
5. The number of washes is sufficient to remove unattached beads
since further washing did not reduce the adsorption-corrected
fluorescence. However, thorough washing is important to
eliminate all the non-phagocytosed beads that are either float-
ing in the medium or attached to the cell surface. Residual
beads can influence the final result as they can be detected on
the flow cytometer and lead to false signals.
6. Vigourous pipetting can result in death of few cells. The dead
cells can be excluded from the analysis by using propidium
iodide or any other viability tracking dyes (e.g., 7-AAD).
We strongly recommend to perform the phagocytosis assay
with adherent microglia because the bead uptake by cells in

suspension is very low.
7. There are also other methods to detach cells, for example,
trypsin–EDTA. Usage of trypsin–EDTA requires incubation
of cells at 37 °C, which indeed cannot be used for the above
described assay but may be applied to a cytochalasin D-based
phagocytosis assay (see Note 3). However, trypsin may digest
some surface antigens and thereby impair analysis of surface
receptors. We do not recommend the usage of cell scraper as
the cell viability of the cells is very low.
8. For a proper analysis, at least 10.000 events should be recorded.
References
1. Hamburger HJ (1912) Physikalisch-chemische by pro- and anti-inflammatory cytokines.

untersuchungen über phagozyten. Ihre bedeu- Neuroreport 8:3851–3856
tung von allgemein biologischem und pathol- 12. Pul R, Moharregh-Khiabani D, Skuljec J et al
ogischem gesichtspunkt. In: Bergmann JF (2011) Glatiramer acetate modulates TNF-
(ed). Wiesbaden alpha and IL-10 secretion in microglia and pro-
2. Fenn WO (1921) The phagocytosis of solid motes their phagocytic activity. J Neuroimmune
particles : I. Quartz. J Gen Physiol 3: Pharmacol 6:381–388
439–464 13. Stangel M, Joly E, Scolding NJ et al (2000)
3. Harvath L, Terle DA (1999) Assay for phago- Normal polyclonal immunoglobulins
cytosis. Methods Mol Biol 115:281–290 (‘IVIg’) inhibit microglial phagocytosis in
4. Crismon JM, Fuhrman FA (1947) Production vitro. J Neuroimmunol 106:137–144
of regional ischemia by intravascular injection 14. Tambuyzer BR, Casteleyn C, Van Cruchten S
of glass and plastic microspheres in graded et al (2012) Interferon-gamma modulates the
sizes. Fed Proc 6:93 functional profile of in-vitro-cultured porcine
5. Sbarra AJ, Karnovsky ML (1959) The bio- microglia. Neuroreport 23:519–524
chemical basis of phagocytosis. I. Metabolic 15. Dijkstra S, Geisert EE Jr, Dijkstra CD et al
changes during the ingestion of particles by (2001) CD81 and microglial activation in vitro:
polymorphonuclear leukocytes. J Biol Chem proliferation, phagocytosis and nitric oxide
234:1355–1362 production. J Neuroimmunol 114:151–159
6. Wilkins DJ (1964) Fluorescent Labelling of 16. Nagai A, Mishima S, Ishida Y et al (2005)
Polystyrene Latex for Tracing in Biological Immortalized human microglial cell line: phe-
Systems. Nature 202:798–799 notypic expression. J Neurosci Res 81:342–348
7. Dunn PA, Tyrer HW (1981) Quantitation of 17. Nagano T, Kimura SH, Takemura M (2010)
neutrophil phagocytosis, using fluorescent Prostaglandin E2 reduces amyloid beta-induced
latex beads. Correlation of microscopy and phagocytosis in cultured rat microglia. Brain
flow cytometry. J Lab Clin Med 98:374–381 Res 1323:11–17
8. Steinkamp JA, Wilson JS, Saunders GC et al 18. Lorenzi R, Brickell PM, Katz DR et al (2000)
(1982) Phagocytosis: flow cytometric quanti- Wiskott-Aldrich syndrome protein is necessary
tation with fluorescent microspheres. Science for efficient IgG-mediated phagocytosis.
215:64–66 Blood 95:2943–2946
9. Schroeder F, Kinden DA (1983) Measurement 19. Aslam R, Kim M, Speck ER et al (2007)
of phagocytosis using fluorescent latex beads. Platelet and red blood cell phagocytosis kinet-
J Biochem Biophys Methods 8:15–27 ics are differentially controlled by phosphatase
10. Sanchez-Martin RM, Muzerelle M, Chitkul N activity within mononuclear cells. Transfusion
et al (2005) Bead-based cellular analysis, sort- 47:2161–2168
ing and multiplexing. Chembiochem 6: 20. Keizer GD, Te Velde AA, Schwarting R et al
1341–1345 (1987) Role of p150,95 in adhesion, migra-
11. von Zahn J, Moller T, Kettenmann H et al tion, chemotaxis and phagocytosis of human
(1997) Microglial phagocytosis is modulated monocytes. Eur J Immunol 17:1317–1322
Chapter 15
Quantification of Microglial Proliferation and Apoptosis

by Flow Cytometry
Alicia A. Babcock, Martin Wirenfeldt, and Bente Finsen
Abstract
Microglia are innate immune cells that survey the central nervous system (CNS) and respond almost
immediately to any disturbance in CNS homeostasis. They are derived from primitive yolk sac myeloid
progenitors and in the mouse colonize the CNS during fetal development. As a population, microglia have
the potential to expand rapidly in response to inflammatory stimuli, injury, or any other pathological
changes, due to a high capacity for proliferation. In addition, apoptotic mechanisms can be evoked to
retract the microglial population, as reactivity declines. In the normal CNS, a low rate of proliferation and
apoptosis maintain a low rate of microglial turnover. Here, we describe quantitative analysis of prolifera-
tion and apoptosis of microglial cells isolated from individual adult mice by flow cytometry, which allows
distinction from perivascular or infiltrating macrophages, based on differential expression of CD45. These
methods can be applied to analyze microglial turnover in various models of neuroinflammation.
Key words Flow cytometry, Proliferation, Apoptosis, Microglia, Macrophages, Brain
1 Introduction
Microglia are the resident innate immune cells of the CNS. Despite
sometimes referred to as “resting” under normal conditions,
microglia are highly active cells that continuously survey defined
regions with their processes [1–4]. When damage is sensed,
microglial processes converge, and a reactive response program
that includes proliferative expansion is initiated [5–7]. As a popula-
tion, microglia have the potential to numerically expand very
rapidly after inflammatory stimuli or other pathological changes,
due to a high capacity for proliferation. For example, 30–40 % of
microglia proliferate 3 days after acute neural injury, a response
that begins 24 h after injury [8–12]. At the same time, apoptotic
mechanisms occur at low levels in normal CNS and can be induced
to retract an expanded microglial population, as reactivity to
damage subsides [8, 13]. Without pathological triggers, apoptosis
129
130 Alicia A. Babcock et al.
and proliferation are held in check. A low rate of microglial

turnover occurs in the normal CNS [14].
Cellular proliferation is often analyzed by measuring incorpo-
ration of bromodeoxyuridine (BrdU), a synthetic thymidine ana-
logue that becomes incorporated into the newly synthesized DNA
of proliferating cells. Incorporated BrdU is then detected with spe-
cific anti-BrdU antibodies, thus indicating cells (BrdU+) that were
replicating their DNA at the time the chemical was administered
[15, 16].
One of the earliest events in apoptosis is the externalization of
phosphatidylserine residues from the inner leaflet of the plasma
membrane to the outer leaflet [17]. This occurs prior to the loss
of membrane integrity that occurs in cell death during late apop-
tosis or necrosis. Externalized phosphatidylserine can be detected
on the cell surface with fluorochrome-conjugated Annexin V, a
calcium-dependent phospholipid-binding protein with a high
affinity for phosphatidylserine [18]. Annexin V is typically used in
combination with a live/dead discriminator dye, such as the DNA-
binding dye 7-Aminoactinomycin D (7-AAD), which permeates
the membranes of dead cells, but is excluded by viable cells with
intact membranes. This distinguishes early apoptotic events
(7-AAD− Annexin V+) from dead cells (7-AAD+).
Though microglial proliferation and apoptosis are frequently
assessed using immunohistochemical methods, both processes may
be monitored using flow cytometry [8, 10, 19]. Flow cytometry is
an invaluable technique for multiparametric analysis of single cells.
Using a panel of fluorescently conjugated reagents simultaneously
allows the detection and quantification of activation markers or
assessment of functional states by cells identified as microglia by
labeling with specific antibodies [8, 10, 19–22]. As myeloid-lineage
cells, microglia express CD11b. However, perivascular macro-
phages that express CD11b are present, even in perfused CNS.
CD11b+ microglia can be distinguished from CD11b+ macrophages
by differential expression of CD45 [23]. Studies in bone marrow
chimeric animals show that the vast majority of CD45dim cells are
resident microglial cells and that the vast majority of CD45high cells
are derived from the periphery [8, 19–22].
Here, we describe the analysis of proliferation and apoptosis
of microglial cells from individual adult mice by flow cytometry.
Unlike most protocols that use a Percoll gradient for cell isolation
and may require whole CNS or even pooling tissue from groups of
mice [19, 23–28], the method described in this chapter allows
investigation of isolated regions of the CNS in individual mice.
Since only a small amount of tissue is required, a single brain can
often be used for several different types of analyses. Our methods
can be applied to analyze microglial turnover in various models of
neuroinflammation.
Microglial Turnover by Flow Cytometry 131
2 Materials
2.1 BrdU Injections 1. ≥2 adult mice (see Note 1).

2. Scale.
3. BrdU (10 mg/ml solution): Dissolve BrdU in room tempera-
ture PBS in a laboratory chemical hood at a concentration of
10 mg/ml and filter sterilize (see Note 2).
4. 1 cc syringe and 26 gauge needles.
2.2 Perfusion/Tissue 1. Pentobarbital.

Collection 2. 70 % ethanol.
3. Perfusion block and absorbent material.
4. 23 gauge needles.
5. 20 cc syringe and 26 gauge needles.
6. Ice.
7. Phosphate buffered saline (PBS) for perfusion, preferably
ice-cold.
8. Surgical instruments for perfusion: blunt-ended large scissors,
forceps, hemostat, and sharp-pointed scissors.
9. Surgical instruments for removal of brain: small straight-edged
scissors, forceps (preferably with teeth), and scoop.
10. Surgical instruments for dissection of brain: blunt-ended forceps
and scalpel with #10 blade.
11. Petri dishes.
12. Hanks balanced salt solution (HBSS) or RPMI-1640 medium
(see Note 3).
13. Tubes for collecting samples: Eppendorf tubes or 24-well plate
for brain samples and 15 ml centrifuge tube for spleen. Samples
should be collected into a small amount of buffer (see Note 3).
2.3 Preparation 1. HBSS or RPMI (see Note 3), dispensed by squirt bottle or
of Cell Suspensions plastic bulb pipettes.
2. 35 mm Petri dishes.
3. 70 μm cell strainers.
4. Glass Pasteur pipettes and latex bulbs.
5. Plungers from 1 cc syringes.
6. 12 × 75 mm sample acquisition tubes for a flow cytometer.
7. Rack for 12 × 75 mm tubes, such as a test tube peg rack.
8. 15 ml centrifuge tubes.
9. Refrigerated centrifuge with racks for 12 × 75 mm tubes and
15 ml centrifuge tubes.
10. 0.83 % solution of ammonium chloride.
2.4 Staining and 1. Blocking buffer: HBSS containing 1 μg/ml anti-FcgIII/II

Flow Cytometry receptor antibody (clone 2.4G2), 50 μg/ml Syrian hamster Ig,
and 2 % FCS (see Note 4).
2. Staining buffer: HBSS containing 2 % FCS (see Note 4).
3. 10 % sodium azide solution: Dissolve 10 g of sodium azide in
100 ml of distilled H2O (see Note 4). The solution should be
prepared with care inside a chemical laboratory hood.
4. Refrigerated centrifuge with racks for 12 × 75 mm tubes and
15 ml centrifuge tubes.
5. Vortexer.
6. Aluminum foil.
7. PE-conjugated CD45 (clone 30-F11) and rat IgG2b isotype
control.
8. FITC-conjugated CD11b (clone M1/70) and rat IgG2b isotype
control.
9. PerCP-Cy™ 5.5-conjugated CD11b (clone M1/70) and rat
IgG2b isotype control.
10. 10× Annexin V binding buffer. Dilute to 1× with distilled
water prior to use.
11. 7-Aminoactinomycin D (7-AAD).
12. APC-conjugated Annexin V.
13. APC BrdU Flow Kit (BD Bioscience): contains BD Perm/
Wash™ Buffer (10×, diluted to 1× in deionized water prior
to use), BD Cytofix/Cytoperm™ Buffer, BD Cytoperm™
Permeabilization Buffer Plus, DNase I (1 mg/ml stock solu-
tion, see Note 5), and APC-conjugated anti-BrdU antibody.
14. Dulbecco’s PBS (D-PBS).
15. Ice.
16. FACS Calibur equipped with 488 nm and 635 nm lasers, capable
of detecting FITC, PE, 7-AAD, PerCP-Cy™ 5.5, and APC, or
a similar instrument.
17. CellQuest Pro or other software for analysis of flow cytometry
data.
3 Methods
3.1 Administration 1. Weigh the mice and calculate the volume of BrdU solution
of BrdU to administer (i.e., 90 mg/kg; 0.18 ml of a 10 mg/ml solution
to a 20 g mouse).
2. Use a 1 cc syringe and 26 gauge needle to administer BrdU
i.p. at 2 h, 10 h, and 22 h prior to perfusion (see Note 6).
3. Include one control mouse that has not been injected with
BrdU in the experiment.
3.2 Tissue Isolation 1. Anesthetize a mouse with an i.p. overdose of pentobarbital.

2. Pin the anesthetized mouse to a perfusion board on its back,
with limbs spread, using 23 gauge needles through the feet.
3. Wet the mouse with 70 % ethanol.
4. Carefully open the abdominal cavity using blunt-ended scissors
and forceps.
5. Cut the diaphragm, cut the ribs upwards, and clamp the sternum
with a hemostat, placing its handles near the mouse’s head to
expose the heart.
6. Place the tip of a 26 gauge needle connected to a 20 cc syringe
filled with ice-cold PBS into the left ventricle.
7. Cut the right atrium using sharp-pointed scissors so that blood
flows from the heart.
8. Slowly inject the PBS into the left ventricle, making sure not to
damage the heart (see Note 7). Blood flowing from the right
atrium will become clear.
9. Decapitate the mouse using a pair of large scissors.
10. Wet the head with 70 % ethanol.
11. Cut through the skin from the back of the neck to the scalp
and fold the skin down under the nose of the mouse for grip,
exposing the skull.
12. Insert the tip of a small, straight pair of scissors into the
cisterna magna and make small cuts through the bone, along
the midline, keeping the tip of the scissors pointed upwards to
not damage the brain.
13. Using a pair of forceps, carefully pull the skull bone away to
each side to reveal the brain.
14. Gently scoop out the brain into a small Petri dish containing
PBS (see Note 8).
15. Using forceps to grab the brain by the cerebellum, place the
rinsed brain onto the lid of the Petri dish.
16. Use a scalpel and a small pair of blunt-edged forceps to micro-
dissect the brain region of interest (see Notes 9–11).
17. Collect the brain region of interest by placing each individual
sample into labeled Eppendorf tubes or 24-well culture plates
containing a small volume of buffer (see Note 3).
18. Collect the spleen for CD45high/CD45dim controls (see Note 12).
Place the mouse on its right side and locate the spleen in the left
side of the abdominal cavity. Make a small cut to detach it, remove
any associated fat tissue, and place the spleen into HBSS.
3.3 Preparation 1. Place a 70 μm cell strainer into a 35 mm Petri dish, and place
of Cell Suspensions the plunger from a 1 cc syringe into the lid of the Petri
from Brain dish.
2. Using a short glass pipette fitted with a latex bulb, pick up

a brain sample, and place it on top of the cell strainer (see
Note 13).
3. Place the pipette temporarily into a 12 × 75 mm tube labeled to
match the tissue sample.
4. Fill a new, empty 12 × 75 mm tube nearly full with HBSS/
RPMI (see Note 3). Never place the pipette into this tube, or
you will have to change it every time.
5. Decant approximately half the HBSS onto the cell strainer in
the Petri dish.
6. Lift the cell strainer slightly, and using the 1 cc plunger, gently
spread the tissue over the cell strainer. Submerge in HBSS.
Repeat with the plunger.
7. Use the glass pipette and bulb to gently move the tissue down
through the cell strainer (see Note 14). All clumps should be
fully homogenized.
8. Collect the suspension into the previously labeled 12 × 75 mm
tube.
9. Wash the cell strainer with the remaining HBSS and transfer to
the previously labeled 12 × 75 mm tube.
10. Prepare other samples.
11. Spin all samples at 400 × g for 10 min at 4 °C (see Note 15).
3.4 Preparation 1. Prepare a single cell suspension using a 70 μm cell strainer and
of Cell Suspensions plunger, as for brain cells.
from Spleen 2. Collect cell suspension into a 15 ml centrifuge tube.
3. Spin at 4 °C for 7 min at 400 × g. Decant and vortex.
4. Resuspend well in 2 ml 0.83 % ammonium chloride. Keep at
room temperature for 10 min, rotating the tube every few
minutes.
5. Fill the centrifuge tube to 15 ml with HBSS.
6. Spin at 4 °C for 7 min at 400 × g. Decant and vortex (see Note 16).
7. Repeat HBSS washes three additional times.
3.5 Blocking 1. Resuspend spleen cells in 1 ml staining buffer. Vortex.

Non-specific Staining 2. Resuspend brain cells in 50 μl of blocking buffer per stain (see
and Setting Up Note 9). Vortex.
Samples/Controls
3. Incubate at room temperature for 20–30 min. During this time,
split brain cells into equivalent fractions (see Notes 9, 17) by
measuring the volume of the suspension with a pipette. Separate
tubes are required for analysis of proliferation and apoptosis.
4. Set up all controls, including compensation controls (Table 1)
and staining controls (Table 2) (see Note 18).
Table 1
Controls for adjusting photomultiplier tubes and compensation
Voltage Compensation
FL2- FL1- FL3- FL2- FL4- FL3-

Analysis Antibody/reagent FL1 FL2 FL3 FL4 %FL1 %FL2 %FL2 %FL3 %FL3 %FL4
Annexin V Unstained cells x x x x
CD11b FITC x x
CD45 PE x x x
7-AAD x x x
Annexin V APC x x
BrdU Unstained cells (x) x x x
(FITC or GFP) (x) (x)
CD45 PE x x x
CD11b PerCP-Cy™5.5 x x x
BrdU APC x x
5. Wash the cells with 1–2 ml of staining buffer.

6. Spin at 4 °C at 400 × g for 10 min (see Note 19).
7. Decant supernatant, then vortex well.
3.6 Annexin V 1. Stain for surface antigens (CD45 PE and CD11b FITC—or
Staining IgG2b controls) by adding 50 μl of a mixture of antibodies
diluted in staining buffer (see Note 20).
2. Vortex, and incubate for 20–30 min at room temperature, cov-
ered in foil.
3. Wash with 1–2 ml cold 1× binding buffer.
4. Spin at 4 °C at 400 × g for 10 min.
5. Decant supernatant and vortex.
6. Add 100 μl of 1× binding buffer and 10 μl of a 1:1 mix Annexin
V APC/7-AAD (see Note 21).
7. Incubate 15 min at room temperature, under foil.
8. Add 200 μl of 1× binding buffer, and store on ice under foil.
9. Acquire using a flow cytometer within 1 h.
3.7 BrdU Staining 1. Stain for surface antigens (CD45 PE and CD11b PerCP-Cy™5.5—
or IgG2b controls) by adding 50 μl of a mixture of antibodies
diluted in staining buffer (see Notes 20 and 22).
2. Vortex and incubate for 20–30 min at room temperature,
covered in foil.
3. Wash with 1–2 ml staining buffer.
4. Spin at 4 °C at 400 × g for 10 min.
Table 2
Staining controls for determining positive versus negative fluorescence levels
Fluorescence channels
Analysis Controls FL1 FL2 FL3 FL4
Annexin V Positive control CD11b FITC CD45 PE 7-AAD Annexin V APC

CD45high/CD45dim CD11b FITC CD45 PE 7-AAD Annexin V APC
controla
FL1 negative control Rat IgG2b FITC CD45 PE 7-AAD Annexin V APC
FL2 negative control CD11b FITC Rat IgG2b PE 7-AAD Annexin V APC
FL3 negative control CD11b FITC CD45 PE (omission)b Annexin V APC
FL4 negative control CD11b FITC CD45 PE 7-AAD (omission)b
Negative control Rat IgG2b FITC Rat IgG2b PE (omission)b (omission)b
BrdU Positive control (FITC or GFP)c CD45 PE CD11b BrdU APC
PerCP-Cy5.5
CD45high/CD45dim (FITC or GFP) CD45 PE CD11b BrdU APC
controla PerCP-Cy5.5
FL1 negative controlc (Ig or CD45 PE CD11b BrdU APC
GFP- cells) PerCP-Cy5.5
FL2 negative control (FITC or GFP) Rat IgG2b PE CD11b BrdU APC
PerCP-Cy5.5
FL3 negative control (FITC or GFP) CD45 PE Rat IgG2b BrdU APC
PerCP-Cy5.5
FL4 negative control (FITC or GFP) CD45 PE CD11b Mouse IgG1
PerCP-Cy5.5 APC
FL4 negative controld (FITC or GFP) CD45 PE CD11b BrdU APC
PerCP-Cy5.5
Negative controld (Ig Rat IgG2b PE Rat IgG2b BrdU APC
or GFP- cells) PerCP-Cy5.5
a
Use brain cells spiked with cells from spleen (or lymph nodes). Under some circumstances it may be ideal to run an
entire set of staining controls with a mixture of brain/spleen cells
b
The negative control should be omission of 7-AAD or Annexin V APC, rather than an Ig control
c
Inclusion of a FITC-labeled activation marker or GFP is optional. FITC-labeled Ig of the relevant isotype or GFP− cells
should be included as a negative control, respectively
d
Uses cells from a mouse that was not injected with BrdU (antigen omission). Cells are labeled as normal with anti-
BrdU APC antibody
5. Decant supernatant and vortex.

6. Fix and permeabilize the cells by adding 100 μl of Cytofix/
Cytoperm™ Buffer.
7. Vortex, and incubate cells for 30 min at room temperature,
under foil (see Note 23).
8. Wash with 1 ml 1× Perm/Wash™ Buffer. Spin at 4 °C at 400 × g
for 10 min, then decant supernatant (see Note 24), and vortex.
9. Resuspend cells in 100 μl of Cytoperm™ Permeabilization
Buffer Plus and incubate for 10 min at 4 °C, under foil.

for 10 min, then decant the supernatant and vortex.
11. Resuspend cells in 100 μl Cytofix/Cytoperm™ Buffer for
5 min at room temperature, under foil.
for 10 min, then decant supernatant and vortex.
13. Resuspend cells with 100 μl diluted DNAse I (diluted to
300 μg/ml in DPBS, giving 30 μg per tube) and incubate at
37 °C for 1 h.
15. Stain for incorporated BrdU by adding 50 μl of 1× Perm/
Wash™ Buffer containing anti-BrdU APC (1:50) (or control
mouse IgG1) and incubate for 20 min at room temperature.
16. Wash with 1 ml 1× Perm/Wash Buffer. Spin at 4 °C at 400 × g
17. Resuspend cells in staining buffer (~300 μl).
18. Acquire cells using a flow cytometer within 3 days.
3.8 Sample 1. Adjust the Forward Scatter (FSC) and Side Scatter (SSC)
Acquisition parameters on linear scales (see Note 25).
2. Set a FSC threshold, to exclude debris and reduce file size
(see Notes 17 and 25).
3. Adjust the fluorescence parameters so that the negative popu-
lations fall between 101 and 102 on logarithmic axes (Table 1).
4. Set up compensation using singly stained controls (Table 1).
5. Measure cell fluorescence by flow cytometry.
6. Acquire one to three million total events for samples and staining
controls (see Note 26, Table 2).
3.9 Analysis: 1. Identify cells of interest based on expression of surface markers

Identifying (see Note 27). Draw a series of gates (SSC vs. CD11b, SSC vs.
CD45+CD11b+ Cells CD45, FSC vs. CD11b, and FSC vs. CD45), based on the
fluorescence levels of isotype-stained controls (Table 2, Fig. 1).
2. Apply a combination of all gates and draw CD45/CD11b
plots, with CD45 on the y-axis and CD11b on the x-axis. Only
macrophages and microglia (all CD45+CD11b+ cells) should
be visible (Fig. 1).
3. Add quadrants to separate CD45high and CD45dim cells by
comparing brain cells to brain samples spiked with spleen cells
(Table 2, Fig. 1) (see Note 28).
3.10 Analysis: 1. After gating on CD11b+/CD45+ cells, exclude dead cells by gat-
Annexin V/Apoptosis ing only on viable (7-AAD−) cells, by comparing to the control
sample where 7-AAD is omitted (Table 2, Fig. 2) (see Note 29).
Fig. 1 Example of gating strategy and controls for analysis of CD45/CD11b flow cytometry profiles. (a) SSC/
CD45 gate depicted on SSC/CD45 or SSC/IgG2b flow cytometry profiles from brain homogenate spiked with
spleen cells. (b) FSC/CD45 gate shown on CD45/FSC or IgG2b/FSC flow cytometry profiles from brain homog-
enate spiked with spleen cells. (c) SSC/CD11b gate drawn on SSC/CD11b or SSC/IgG2b flow cytometry profiles
from brain homogenate spiked with spleen cells. (d) FSC/CD11b gate shown on CD11b/FSC or IgG2b/FSC flow
cytometry profiles from brain homogenate spiked with spleen cells. (e) Brain/spleen cells stained with isotype
controls or antibodies against CD45 and CD11b antibodies, shown as CD45/CD11b plots. Quadrants discrimi-
nate CD45high and CD45dim cells. Events shown in (a)–(e) are not gated and represent several million total
events collected. (f) CD45/CD11b profiles from brain/spleen mixture (left, gated on CD45+ cells as in (a) and
(b)) or neocortex from a 4-month-old mouse (right, gated on CD45+CD11b+ cells as in (a)–(d)). Quadrants
discriminate CD45high and CD45dim cells
2. Determine the percentage of Annexin V+ cells, by setting quad-

rants based on controls lacking Annexin V (Fig. 2) (see Note 30).
3.11 Analysis: 1. After gating on CD11b+/CD45+cells, set up quadrants to

BrdU/Proliferation determine the percentage of BrdU+ cells, versus controls
lacking BrdU and isotype-stained controls (Table 2, Fig. 3)
(see Note 30).
3.12 Application 1. Measure microglial proliferation and apoptosis in the experi-

mental system of interest (Fig. 4).
Fig. 2 Example of controls for analysis of apoptotic microglia/macrophages. (a) Live cells from the neocortex
of an adult mouse are gated by excluding 7-AAD+ cells on CD11b+CD45+ gated CD45/7-AAD flow cytometry
profiles (right ), based on fluorescence levels of the control lacking 7-AAD (left ). (b) Flow cytometry profiles gated
on viable (7-AAD−) CD11b+CD45+ cells show fluorescence with Annexin V labeling (right ) or binding buffer alone
(left ). Quadrants are drawn based on the fluorescence levels of the control lacking Annexin V, and discriminate
CD45high and CD45dim cells as described in Fig. 1
Fig. 3 Example of controls for analysis of BrdU+ microglia and macrophages. CD45/BrdU flow cytometry profiles,
gated on CD11b+CD45+ cells, from the neocortex of either a naïve mouse without BrdU injection (a) or a mouse
injected with BrdU (c), as well as a CD45/mIgG1 plot showing background staining in a BrdU-injected mouse (b).
Quadrants are drawn based on the fluorescence levels of the control mice for BrdU, and discriminate CD45high
and CD45dim cells using the strategy depicted in Fig. 1
Fig. 4 Example of analysis of Annexin V+ and BrdU+ microglia and macrophages in an experimental setting.
Microglial apoptosis (a) and proliferation (b) are increased in the ipsilateral hippocampus of mice 3 days after
perforant pathway axonal lesion (right ), compared to unlesioned contralateral hippocampus (left ). Axonal
lesions were carried out as previously described [8]
4 Notes
1. Each mouse is analyzed individually. At least one mouse should

serve as a “no BrdU” control that is not injected with BrdU.
2. Aliquots of diluted BrdU can be stored at −20 °C, though any
BrdU that precipitates upon freezing should be in solution
before use.
3. Tissue is usually collected into HBSS, although RPMI may be
preferred for analysis of apoptotic cells.
4. We usually add sodium azide to staining reagents at a final con-
centration of 0.1 %, except for analysis of Annexin V staining.
5. DNAse I stock solution may only be refrozen once before it
loses activity. Additional DNAse I can be purchased from
Sigma (D-4513), if necessary.
6. Many variations of the BrdU protocol are possible. We have
obtained similar results by administering BrdU i.p. over 3 × 8 h
intervals [8–10], single i.p. injections 1 h prior to perfusion
[11], or daily i.p. injections of a lower dose [19]. BrdU can also
be administered via drinking water [29]. In addition, analysis of
incorporated BrdU can be delayed by several days, to analyze
the fate of previously proliferating cells [29, 30].
7. The purpose of perfusion is to eliminate contamination of the
CNS by circulating blood cells. To ensure a good perfusion, do
not perfuse the mouse too quickly or cause major bleeding
elsewhere. Signs for good perfusion are that the forepaws turn
white and that the liver pales, turning yellow. When finished,
the brain should be white, not pink.
8. The meninges should be removed before analysis. Often the
meninges are detached when the skull is removed; however, it
may be necessary to do this with forceps.
9. Only a small amount of tissue is required for each stain for flow
cytometry. If appropriate to the experimental setup, brains can be
split into different hemispheres or regions, which may then be
collected for different purposes (i.e., immunohistochemistry,
RNA analyses). Furthermore, only a fraction of each brain region
is required per staining for flow cytometry. For example, only ½
of a hippocampus [8, 10, 31, 32], entorhinal cortex [24, 31], or
corpus callosum [19] should be stained. Use 1/6th to 1/8th of
the neocortex from a single hemisphere per stain [20, 21].
10. To isolate hippocampi, use a scalpel blade to make an incision
between the two cerebral hemispheres. At the front of the
brain, cut all the way down to the base. Cut down 1–2 mm at
the back of the cerebral hemispheres, just through the neocor-
tex. Using the blunted side of the tip of the scalpel blade, lift
the cortex away from the midbrain on each side. If necessary
make gentle cuts with the scalpel to fully push the neocortex
away (until flat), exposing the hippocampus. Make small inci-
sions at the fimbria, and gently roll the hippocampi backwards,
towards the entorhinal cortex, and remove. Though it is diffi-
cult to see after perfusion, the choroid plexus usually sticks to
the hippocampi; it should be removed before analysis. This can
be done by gently running the blunt side of the scalpel over the
isolated hippocampi.
11. To isolate neocortex, first remove the hippocampus (see
Note 10). Gently scoop out the striatum from the cortex, still
lying at each side of the brain, then dissociate the cortex from
the midbrain.
12. Alternatively, cells isolated from lymph nodes can be used to
distinguish CD45high cells from CD45dim cells. We often collect
inguinal lymph nodes for this purpose. They should be homog-
enized as indicated below; unlike spleen, no lysis of red blood
cells is required.
13. These are live cells that need to be processed shortly after tissues
are removed from mice. Analysis of apoptosis should be done
on the same day. Cells for BrdU analysis can be stored after the
fixation step. These procedures are lengthy, and it is recom-
mended to prepare as much as possible the day before and to
begin with only a few samples until a routine has been estab-
lished and techniques have been optimized.
14. This should be done as gently as possible, to prevent high levels
of cellular autofluorescence.
15. This protocol can be modified for the analysis of cytokine pro-
duction, by incubating cells with 1 μg/ml Brefeldin A for
4–6 h at 37 °C and 5 %CO2 [20, 21], before proceeding with
blocking and antibody staining.
16. Remove any clumps of dead cells that appear.
17. It is possible to estimate absolute cell numbers, as previously
described [8, 10, 19–21, 31], if one knows what fraction of the
brain region has been sampled. Thus it is important to split the
cells as accurately as possible. A second sampling step occurs
during sample acquisition at the flow cytometer. To account
for this, we usually measure the volume of the cell suspension,
before and after acquisition on the flow cytometer, to deter-
mine what fraction has been used. Alternatively, beads can be
used to measure this [33]—however, in this case, the FSC
threshold must be removed so that the beads remain visible,
making file sizes much larger.
18. For compensation controls, to set up the flow cytometer, a series
of cells stained with each antibody/dye individually is required,
in addition to unstained controls. Staining controls include
positive controls, Ig controls, and CD45high/CD45dim controls.
A low level of proliferation and apoptosis occurs in unmanipu-
lated mice [8]. Including a series of controls whereby only one
antibody/dye is replaced with Ig control/omitted allows
determination of background staining in specific cell popula-
tions. Including a mix of brain and spleen cells, for both sets of
stains, allows for consistent discrimination between CD45high/
CD45dim cells. Adding 5 μl of diluted spleen cells is sufficient
for generating spleen cell-spiked brain control samples.
19. Unless washing with Perm/Wash™ Buffer (see Note 24), spins
can be shortened to 6 min, if pressed for time.
20. Diluting to 0.5 μg/ml for anti-CD45 PE and to 1 μg/ml for
anti-CD11b (FITC or PerCP-Cy™5.5) is typically sufficient to
stain the number of cells described in this protocol (see Note 9).
However, the concentration of each antibody should be
titrated by the investigator to obtain optimal labeling.
21. Note that only 5 μl of Annexin V or 5 μl of 7-AAD should be
added for FL3 and FL4 negative controls (Table 2), respectively.
Stagger the addition of Annexin V/7-AAD if many samples
are to be run.
22. It is possible to include an additional antibody, for example, a

FITC-labeled activation marker, to determine whether its expres-
sion is related to cellular proliferation. We have successfully
detected cellular expression of GFP on FL1 in combination with
PE, PerCP-Cy5.5, and APC [8].
23. It is possible to pause the procedure here. Wash the cells with
staining buffer, spin at 4 °C at 400 × g for 10 min, decant, and
vortex. Cover with foil and store at 4 °C. Immediately prior to
resuming the procedure, vortex the cells and add 1–2 ml of
Perm/Wash™ Buffer for 1 h before continuing with step 9.
24. Use care when decanting after rinsing with Perm/Wash™ Buffer;
the pellets are looser than when washed with staining buffer.
25. We usually center the microglial population at approximately
250 on the FSC and SSC scales, and set the FSC threshold to
80. It may help to draw a SSC versus CD11b gate and then to
“back gate” to visualize CD11b+ cells on FSC/SSC plots, since
debris and dead cells have not been removed from the cell sus-
pension. Note that microglial FSC and SSC become increased
in the inflamed CNS [10, 27, 28].
26. A large number of total events should be collected, especially
for analysis of rare events [10], since some events are debris,
not actual cells.
27. Other cell types can be analyzed using this method by staining
with different antibodies, including T cells [10, 19, 21, 31,
32, 34], Gr1+ neutrophils [20, 21], CD11c+ cells [19], and
astrocytes [10].
28. It can help to visualize the brain/spleen plot after only gating
on SSC versus CD45 and FSC versus CD45, since the CD11b−
lymphocyte population in the spleen provides a very consistent
level of CD45 expression, which is useful for discriminating
CD45high/CD45dim cells on CD45/CD11b plots across mul-
tiple experiments (Fig. 1).
29. Alternatively, 7-AAD+ cells can be gated directly and excluded
from further analysis.
30. By setting quadrants based on the brain/spleen mix (Table 2),
it is possible to calculate separately for macrophages and
microglia (Fig. 2).
Acknowledgements
This work was supported by The Augustinus Foundation; Aase

og Ejnar Danielsens Fond; The Carlsberg Foundation, Grosserer
M. Brogaard og Hustrus Mindefond, Odense; Katrine og Vigo
Skovgaards Fond; Fonden til Lægevidenskabens Fremme;
The Lundbeck Foundation; The Novo Nordisk Foundation;

Overlægerådets Legatudvalg; the Danish Alzheimer’s Society; and
the Danish Medical Research Council.
References
1. Wirenfeldt M, Babcock AA, Vinters HV 11. Dissing-Olesen L, Ladeby R, Nielsen HH,

(2011) Microglia—insights into immune sys- Toft-Hansen H, Dalmau I, Finsen B (2007)
tem structure, function, and reactivity in the Axonal lesion-induced microglial prolifera-
central nervous system. Histol Histopathol tion and microglial cluster formation in the
26(4):519–530 mouse. Neuroscience 149(1):112–122.
2. Hanisch UK, Kettenmann H (2007) Microglia: doi:10.1016/j.neuroscience.2007.06.037
active sensor and versatile effector cells in the 12. Hailer NP, Grampp A, Nitsch R (1999)
normal and pathologic brain. Nat Neurosci Proliferation of microglia and astrocytes in the
10(11):1387–1394. doi:10.1038/nn1997 dentate gyrus following entorhinal cortex lesion:
3. Davalos D, Grutzendler J, Yang G, Kim JV, a quantitative bromodeoxyuridine-labelling
Zuo Y, Jung S, Littman DR, Dustin ML, Gan study. Eur J Neurosci 11(9):3359–3364
WB (2005) ATP mediates rapid microglial 13. Jones LL, Banati RB, Graeber MB, Bonfanti
response to local brain injury in vivo. Nat L, Raivich G, Kreutzberg GW (1997)
Neurosci 8(6):752–758. doi:10.1038/nn1472 Population control of microglia: does apopto-
4. Nimmerjahn A, Kirchhoff F, Helmchen F sis play a role? J Neurocytol 26(11):755–770
(2005) Resting microglial cells are highly 14. Lawson LJ, Perry VH, Gordon S (1992)
dynamic surveillants of brain parenchyma in Turnover of resident microglia in the normal
vivo. Science 308(5726):1314–1318. adult mouse brain. Neuroscience 48(2):
doi:10.1126/science.1110647 405–415
5. Ladeby R, Wirenfeldt M, Garcia-Ovejero D, 15. Gratzner HG, Leif RC (1981) An immuno-
Fenger C, Dissing-Olesen L, Dalmau I, Finsen fluorescence method for monitoring DNA
B (2005) Microglial cell population dynamics synthesis by flow cytometry. Cytometry
in the injured adult central nervous system. 1(6):385–393. doi:10.1002/cyto.990010606
Brain Res Brain Res Rev 48(2):196–206 16. Gratzner HG (1982) Monoclonal antibody to
6. Graeber MB, Tetzlaff W, Streit WJ, Kreutzberg 5-bromo- and 5-iododeoxyuridine: a new
GW (1988) Microglial cells but not astrocytes reagent for detection of DNA replication.
undergo mitosis following rat facial nerve Science 218(4571):474–475
axotomy. Neurosci Lett 85(3):317–321 17. Martin SJ, Reutelingsperger CP, McGahon AJ,
7. Streit WJ, Kreutzberg GW (1988) Response of Rader JA, van Schie RC, LaFace DM, Green
endogenous glial cells to motor neuron degen- DR (1995) Early redistribution of plasma
eration induced by toxic ricin. J Comp Neurol membrane phosphatidylserine is a general fea-
268(2):248–263. doi:10.1002/cne.902680209 ture of apoptosis regardless of the initiating
8. Wirenfeldt M, Dissing-Olesen L, Babcock AA, stimulus: inhibition by overexpression of Bcl-2
Nielsen M, Meldgaard M, Zimmer J, Azcoitia and Abl. J Exp Med 182(5):1545–1556
I, Leslie RG, Dagnaes-Hansen F, Finsen B 18. Koopman G, Reutelingsperger CP, Kuijten
(2007) Population control of resident and GA, Keehnen RM, Pals ST, van Oers MH
immigrant microglia by mitosis and apoptosis. (1994) Annexin V for flow cytometric detec-
Am J Pathol 171(2):617–631 tion of phosphatidylserine expression on B
9. Ladeby R, Wirenfeldt M, Dalmau I, Gregersen cells undergoing apoptosis. Blood 84(5):
R, Garcia-Ovejero D, Babcock A, Owens T, 1415–1420
Finsen B (2005) Proliferating resident microg- 19. Remington LT, Babcock AA, Zehntner SP,
lia express the stem cell antigen CD34 in Owens T (2007) Microglial recruitment, acti-
response to acute neural injury. Glia vation, and proliferation in response to pri-
50(2):121–131 mary demyelination. Am J Pathol 170(5):
10. Babcock AA, Wirenfeldt M, Holm T, Nielsen 1713–1724
HH, Dissing-Olesen L, Toft-Hansen H, 20. Lambertsen KL, Clausen BH, Babcock AA,
Millward JM, Landmann R, Rivest S, Finsen Gregersen R, Fenger C, Nielsen HH,
B, Owens T (2006) Toll-like receptor 2 signal- Haugaard LS, Wirenfeldt M, Nielsen M,
ing in response to brain injury: an innate Dagnaes-Hansen F, Bluethmann H,
bridge to neuroinflammation. J Neurosci Faergeman NJ, Meldgaard M, Deierborg T,
26(49):12826–12837. doi:10.1523/ Finsen B (2009) Microglia protect neurons
JNEUROSCI.4937-05.2006 against ischemia by synthesis of tumor necrosis
factor. J Neurosci 29(5):1319–1330. experimental allergic encephalomyelitis.

doi:10.1523/JNEUROSCI.5505-08.2009 Regulation by Th1 cytokines. J Immunol
21. Clausen BH, Lambertsen KL, Babcock AA, 154(2):944–953
Holm TH, Dagnaes-Hansen F, Finsen B 27. Sedgwick JD, Ford AL, Foulcher E, Airriess R
(2008) Interleukin-1beta and tumor necrosis (1998) Central nervous system microglial cell
factor-alpha are expressed by different subsets activation and proliferation follows direct
of microglia and macrophages after ischemic interaction with tissue-infiltrating T cell blasts.
stroke in mice. J Neuroinflammation 5(1):46 J Immunol 160(11):5320–5330
22. Wirenfeldt M, Babcock AA, Ladeby R, 28. Cao L, Palmer CD, Malon JT, De Leo JA
Lambertsen KL, Dagnaes-Hansen F, Leslie (2009) Critical role of microglial CD40 in the
RG, Owens T, Finsen B (2005) Reactive maintenance of mechanical hypersensitivity in
microgliosis engages distinct responses by a murine model of neuropathic pain. Eur J
microglial subpopulations after minor central Immunol 39(12):3562–3569. doi:10.1002/
nervous system injury. J Neurosci Res 82(4): eji.200939657
507–514 29. Tough DF, Sprent J (1996) Viruses and T cell
23. Sedgwick JD, Schwender S, Imrich H, Dorries turnover: evidence for bystander proliferation.
R, Butcher GW, ter Meulen V (1991) Isolation Immunol Rev 150:129–142
and direct characterization of resident microg- 30. Nielsen HH, Toft-Hansen H, Lambertsen KL,
lial cells from the normal and inflamed central Owens T, Finsen B (2011) Stimulation of
nervous system. Proc Natl Acad Sci USA adult oligodendrogenesis by myelin-specific T
88(16):7438–7442 cells. Am J Pathol 179(4):2028–2041.
24. Toft-Hansen H, Babcock AA, Millward JM, doi:10.1016/j.ajpath.2011.06.006
Owens T (2007) Downregulation of mem- 31. Babcock AA, Toft-Hansen H, Owens T (2008)
brane type-matrix metalloproteinases in the Signaling through MyD88 regulates leukocyte
inflamed or injured central nervous system. J recruitment after brain injury. J Immunol
Neuroinflammation 4:24. 181(9):6481–6490
doi:10.1186/1742-2094-4-24 32. Babcock AA, Kuziel WA, Rivest S, Owens T
25. Brambilla R, Persaud T, Hu X, Karmally S, (2003) Chemokine expression by glial cells
Shestopalov VI, Dvoriantchikova G, Ivanov D, directs leukocytes to sites of axonal injury in
Nathanson L, Barnum SR, Bethea JR (2009) the CNS. J Neurosci 23(21):7922–7930
Transgenic inhibition of astroglial NF-kappa B 33. Toft-Hansen H, Fuchtbauer L, Owens T
improves functional outcome in experimental (2011) Inhibition of reactive astrocytosis in
autoimmune encephalomyelitis by suppressing established experimental autoimmune encepha-
chronic central nervous system inflammation. J lomyelitis favors infiltration by myeloid cells
Immunol 182(5):2628–2640. doi:10.4049/ over T cells and enhances severity of disease.
jimmunol.0802954 Glia 59(1):166–176. doi:10.1002/glia.21088
26. Renno T, Krakowski M, Piccirillo C, Lin JY, 34. Khorooshi R, Babcock AA, Owens T (2008)
Owens T (1995) TNF-alpha expression by NF-kappaB-driven STAT2 and CCL2 expres-
resident microglia and infiltrating leukocytes sion in astrocytes in response to brain injury.
in the central nervous system of mice with J Immunol 181(10):7284–7291
Chapter 16
Fluorescence Imaging of Intracellular Ca2+, Na+,

and H+ in Cultured Microglia
Abstract
The behavior of microglial cells involves the activity of a variety of ion channels and ion transporters, which
are implicated in the regulation of ion concentrations, membrane potential, and cell volume of microglia.
Fluorescence imaging has been proven to be an elegant method to study ion concentration changes in
intact microglial cells under physiological and pathophysiological conditions. The development of highly
specific ion indicators has made it possible to detect changes in intracellular Ca2+, Na+, and H+ concentra-
tions of microglial cells as a result of ion channel or ion transporter activity. Fluorescence signals of isolated
dye-loaded microglial cells can be detected via a CCD camera equipped to a conventional microscope.
This chapter summarizes protocols of loading of microglial cells with small-molecule ion indicators as well
as protocols optimal for measurement and analysis of intracellular Ca2+, Na+, and H+ concentrations in
microglia in vitro.
Key words Fluorescence imaging, CCD camera, Ion channels, Ion transporters, Microglia,
Monochromator, Fura-2, SBFI, BCECF
1 Introduction
Fluorescence imaging provides a powerful method to investigate

changes in the intracellular milieu of intact microglial cells. Due to
the development of specific fluorescent ion indicators, it is now
possible to precisely measure concentrations of intracellular ion
concentrations and their changes in response to microglial stimula-
tion. Thus, in addition to the patch-clamp technique, fluorescence
imaging provides a useful tool to investigate ion channel activity in
microglial cells. Furthermore, in fluorescence imaging experiments,
the activity of electroneutral ion transporters can be assessed, which
is impossible using the patch-clamp technique. For example, fluo-
rescence imaging experiments on microglia can be used to deter-
mine the resting intracellular ion concentration of a microglial cell
and to provide detailed spatiotemporal information about ion
fluxes through microglial ion channels and ion transporters and
147
148 Tom Schilling and Claudia Eder
exchangers, about Ca2+ release from internal stores, and about H+/pH
changes in phagolysosomes and others.
For ion imaging, small-molecule ion indicators or genetically
encoded indicators can be used. Ratiometric indicators are of great
advantage as they allow accurate estimation of intracellular ion con-
centrations, since uneven dye loading, dye leakage, photobleaching,
and changes in cell volume can be corrected. In general, ion indica-
tors shift their absorption or emission spectrum or change their
fluorescence intensity after binding to a specific ion. The resulting
fluorescence can be detected by a camera, which is equipped to a
microscope and connected with a computer, while fluorescence
intensities of computer-generated images can be analyzed subse-
quently. Fluorescence imaging can be performed using conven-
tional microscopy or confocal microscopy. Conventional microscopy
can be used to detect ion changes in single cells or cell monolayers,
whereas confocal microscopy allows imaging of cells in tissue as it
provides an optical sectioning effect to reject out-of-focus fluores-
cence. In this chapter, we will introduce equipment and protocols
required for fluorescence imaging of Ca2+, Na+, and H+ signals in
cultured microglia using a CCD camera equipped to a conventional
microscope, which is suitable for ion measurements in isolated
microglial cells in vitro.
2 Materials
2.1 Equipment for Resolution of experimental data acquired with fluorescence imag-
Fluorescence Imaging ing techniques is highly dependent on the quality of the equip-
Experiments ment used. Essential parts are:
1. Microscope with an objective of at least 40× magnification.
2. Light source (laser, monochromator, mercury burner, or
xenon lamp).
3. Filter sets (see Subheading 2.3).
4. CCD camera.
5. Acquisition and analysis software.
6. Vibration isolation table and cage.
7. Recording chamber for microscope stage.
8. Superfusion system.
For analysis of single cells, inverted microscopes are more
valuable than upright microscopes, which are used for fluores-
cence imaging experiments on microglial cells in situ. The micro-
scope should be equipped with an objective lens of at least 40×
magnification (see Note 1). The microscope stage requires a
recording chamber connected to a superfusion system for rapid
and efficient solution exchange. For excitation purposes, a laser,
Imaging Microglial Ion Channel Activity 149
a monochromator (see Note 2), or a fluorescent light source, such

as a mercury or a xenon burner, can be used. The fluorescent
probe is usually excited by light, generated according to a protocol
of an acquisition software. Pictures of the emitted light signals are
taken by an imaging software-driven CCD camera. This camera
should be chosen according to emission spectra of the dyes; most
cameras for fluorescence imaging systems are able to detect all com-
mercially available dyes. For investigation of microglial ion channels
and transporter activity, spatial resolution of the camera is more
important than temporal resolution. A vibration isolation table, on
which the microscope is mounted, minimizes mechanically induced
artifacts during the recording. A cage increases signal to background
ratio and minimizes optical artifacts by other light sources.
2.2 Cells Cultured microglial (see Note 3) cells should be plated on sterile
glass cover slips. It is important to test in advance, whether these
cover slips are nonfluorescent for the excitation wavelength of the
fluorescent dye used. Cell density for glass cover slips, fitting into
a 24-well plate, should not exceed 100,000 cells per well to have a
sufficient amount of nonfluorescent area, i.e., an area free of cells, for
background subtraction.
2.3 Filter Sets To detect changes of intracellular Ca2+ concentrations of microglial

cells, Fura-2-acetoxymethyl ester (Fura-2AM) [1,2], Indo-1 [3],
2.3.1 Measurement
and Fluo-3 [4] have been used. In this chapter the application of
of Intracellular Calcium
Fura-2AM (see Note 4) will be described. With a Kd value of
Concentrations
224 nM, Fura-2AM is ideal to measure changes near the resting
calcium concentration of 100 nM (see Note 5).
If no monochromator is available, the microscope should be
equipped with band-pass filters of 340 nm and 380 nm wave-
lengths. To detect volume changes using Fura-2AM without a
monochromator, an additional band-pass filter of 360 nm is
required. Additionally, a 400 nm dichroic mirror and a 420 nm
longpass emission filter (see Note 6) should be installed.
2.3.2 Measurement Changes in intracellular sodium concentration of microglial cells

of Intracellular Sodium have been described using sodium-binding benzofuran-isophthalate
Concentrations acetoxymethyl ester (SBFI-AM) [5]. SBFI-AM (see Note 7) is a
ratiometric dye with a Kd value of 11.3 mM. No additional filters
are needed, if the microscope is equipped for calcium measure-
ments with Fura-2AM, i.e., either a monochromator or 340 nm
and 380 nm band-pass filters as well as a 400 nm dichroic mirror
and a 420 nm longpass emission filter should be installed.
2.3.3 Intracellular pH In microglial cells, intracellular pH (pHi) measurements have been

Measurements performed with 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluo-
rescein, acetoxymethyl ester (BCECF-AM) [6]. BCECF-AM has
a pKa of 6.98 (see Note 8) and is therefore a good choice to
detect changes in pHi. Like Fura-2AM and SBFI-AM, it has the

advantages of ratiometric dyes and is cell membrane permeable.
For dye excitation, the microscope should be equipped either with
a 440 nm and a 490 nm band-pass filter or a monochromator.
In addition, a 505 nm dichroic mirror and a 530 nm barrier filter
should be installed.
2.4 Solutions Experiments are performed using solution E1 (see Note 9) and
calcium-free solution E2.
2.4.1 Measurement
of Intracellular Calcium ●● Solution E1: 130 mM NaCl, 5 mM KCl, 10 mM
Concentrations 4-(2-
hydroxyethyl)piperazine-1-ethanesulfonic acid N-(2-
hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) (HEPES),
10 mM d-glucose, 2 mM CaCl2, and 1 mM MgCl2 (pH = 7.4;
adjusted with NaOH).
●● Calcium-free solution E2: 130 mM NaCl, 5 mM KCl, 10 mM
HEPES, 10 mM d-glucose, 1 mM ethylene glycol-bis
(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA), and
4 mM MgCl2 (pH = 7.4; adjusted with NaOH).
●● To investigate intracellular calcium stores, prepare a 10 mM
stock solution of the calcium ATPase inhibitor thapsigargin
(see Note 10) in DMSO.
●● To calibrate Fura-2AM fluorescence signals, calibration solu-
tions E3 and E4 should be prepared.
●● Calibration solution E3: 130 mM NaCl, 5 mM KCl, 10 mM
HEPES, 10 mM d-glucose, 2 mM EGTA, and 5 mM MgCl2
(pH = 7.4; adjusted with NaOH).
●● Calibration solution E4: 130 mM NaCl, 5 mM KCl, 10 mM
HEPES, 10 mM CaCl2, and 5 mM MgCl2 (pH = 7.4; adjusted
with NaOH).
Reconstitute ionomycin and thapsigargin as a 10 mM stock
solution in DMSO.
2.4.2 Measurement Experiments are performed using solution E1. Prepare additionally
of Intracellular Sodium sodium-free solution E5.
Concentrations ●● Solution E1: 130 mM NaCl, 5 mM KCl, 10 mM HEPES,
●● Sodium-free solution E5: 130 mM N-methyl-d-glucamine
(NMG)-Cl, 5 mM KCl, 10 mM HEPES, 10 mM d-glucose,
2 mM CaCl2, and 1 mM MgCl2 (pH = 7.4; adjusted with
NMGOH).
For calibration of SBFI-AM signals, stock solutions of gramicidin
(10 mM in ethanol), monensin (10 mM in ethanol), and ouabain
(20 mM in H2O) should be available (see Note 11).
2.4.3 Intracellular pH Solution E1 (see Note 12) is used for experiments. Experimental
Measurements data have to be calibrated after each experiment using solutions E6,
E7, and E8.
●● Solution E1: 130 mM NaCl, 5 mM KCl, 10 mM HEPES,
●● Solution E6: 130 mM KCl, 10 mM HEPES, 10 mM d-glucose,
2 mM CaCl2, and 1 mM MgCl2 (pH = 7.6; adjusted with KOH).
●● Solution E7: 130 mM KCl, 10 mM HEPES, 10 mM d-glucose,
2 mM CaCl2, and 1 mM MgCl2 (pH = 7.0; adjusted with KOH).
●● Solution E8: 130 mM KCl, 10 mM 2-(N-morpholino)ethane-
sulfonic acid hydrate, 4-morpholineethanesulfonic acid (MES),
adjusted with KOH).
Furthermore, dissolve nigericin in ethanol as a 10 mM stock
solution.
2.5 Data Analysis Data can be analyzed with most versions of commercially distrib-
uted fluorescence imaging software. A second software license is
advisable to be able to perform experiments while analyzing previ-
ous experiments on another computer at the same time. As an
open-source alternative, the NIH offers the platform-independent
program ImageJ (see Note 13), which is able to perform most
image processing tasks and can be accustomed with plug-ins to
one’s needs.
3 Methods
3.1 Ca2+ Imaging 1. Prepare a 3 mM stock solution (see Note 14) of Fura-2 AM in
DMSO (see Note 15).
3.1.1 Detection
of Intracellular Ca2+ 2. Dilute this stock solution in 2 ml solution E1 to a final concen-
Concentration tration of 3 μM Fura-2AM. Do not expose solution to light.
3. Place one glass cover slip with microglial cells in Fura-2AM
solution for 30 min at room temperature.
4. Wash microglial cells on glass cover slip in solution E1 for
30 min at room temperature.
5. Fill recording chamber with solution E1. Transfer glass cover
slip with dye-loaded microglial cells into recording chamber of
microscope.
6. Take a bright-field picture of a viewing field with a sufficient
number of microglial cells (see Note 16).
7. Take a sample picture of the same cells with excitation light at
380 nm. Increase exposure time until no further increase in
emitted fluorescence of microglial cells can be achieved. Adjust

the protocol of the imaging software for this exposure time.
8. Set the protocol of the imaging software to take one picture
with excitation at 340 nm and one at 380 nm every 20 s
(see Note 17). To monitor the experiment, calculate a ratio-
metric picture in the protocol editor by dividing the emission
signals at 340 and 380 nm.
9. Start the protocol. Wait for a baseline level of Ca2+ signals for
at least 15 data points.
10. Superfuse cells with solution E1 as a control experiment. No
change in Ca2+ signals should be observed (see Note 18).
11. Superfuse cells with activators dissolved in solution E1. Observe
changes in ratiometric calcium signals (see Note 19). Ca2+
signals of microglial cells should reach baseline levels in at least
ten data points before using another stimulus.
3.1.2 Detection of Ca2+ 1. Follow steps 1–10 of previous Subheading.

Influx from Extracellular 2. Superfuse cells with solution E1 containing the desired activator
Space and observe increases in intracellular calcium concentrations
(see Note 20). Record at least ten data points after calcium
signals reached baseline levels before proceeding with the
next steps.
3. Superfuse cells with calcium-free solution E2. A small decrease
of fluorescence intensities should be observed. Wait for a stable
baseline of Ca2+ signals under these conditions (see Note 21).
4. Stimulate microglial cells with E2 containing the desired activa-
tor. If the fluorescence ratio signal does not change under
these conditions, the activator induces calcium influx from the
extracellular space. If a Ca2+ signal is still detected (see Note 22),
the activator elicits intracellular calcium increases from intra-
cellular Ca2+ stores (Subheading 3.1.3).
5. Superfuse cells with solution E1. A moderate increase of the
ratiometric signal (see Note 23) should be observed.
6. Repeat step 2 to confirm that there is no inherent decline of
Ca2+ signals over time.
3.1.3 Ca2+ Release from 1. Follow steps 1–9 of Subheading 3.1.1. Change the protocol
Intracellular Stores to take a fluorescence picture every 5 s.
2. Superfuse microglial cells with solution E2. Observe a decrease
in the ratiometric signal.
3. Exchange solution with E2 containing 500 nM thapsigargin
(see Note 24). An increase in fluorescence intensity signals
should be detected, indicating depletion of intracellular Ca2+
stores. Wait for a stable baseline after this transient rise in intra-
cellular calcium concentrations.
4. Superfuse microglial cells with the desired activator dissolved

in E2. No change of fluorescence signals under these condi-
tions is a sign that the activator leads to Ca2+ release from intra-
cellular calcium stores.
3.1.4 Calibration 1. The system should always be recalibrated, when any part of the
of Intracellular Ca2+ Signals optical pathway (i.e., monochromator, filter sets, objective
lenses, camera, recording chamber, glass cover slips) has been
changed (see Note 25).
2. Follow steps 1–9 in Subheading 3.1.1.
3. Superfuse cells with solution E3 containing 10 μM ionomycin
and 1 μM thapsigargin. Wait for a stable ratiometric fluores-
cence signal.
4. Superfuse cells with solution E4 containing 10 μM ionomycin.
Wait until the intracellular fluorescence signals reach steady
state.
5. Repeat this calibration procedure at least two more times with
new cells on glass cover slips.
3.1.5 Data Analysis 1. In a first step, the calibration experiments should be analyzed.
2. Open the bright-field picture with your image analysis software.
3. Place several regions of interest (ROI) in background areas, in
which no cells are visible.
4. Determine the mean background fluorescence intensity in the
340 nm fluorescence picture. Subtract this value from the stack
of fluorescence pictures excited at 340 nm.
5. Repeat the procedure with fluorescence pictures excited at
380 nm.
6. Place ROIs in the bright-field picture in every visible cell.
7. Determine the fluorescence intensity F for each cell at 340 nm
and 380 nm excitation when exposed to E3 (F340min, F380min)
and E4 (F340max, F380max).
8. Calculate for each cell Rmin = F340min/F380min, Rmax = F340max/
F380max, and β = F380min/F380max.
9. Calculate the mean values for Rmin, Rmax, and β from all
experiments.
10. To determine changes in the intracellular calcium concentration,
repeat steps 2–5 for an experiment.
11. In the bright-field picture, put a ROI into each cell or into
identified subcellular structures.
12. Measure fluorescence intensities F340 and F380 for all ROIs
over time in background-corrected pictures (see Note 26).
Calculate for each ROI the ratiometric fluorescence signal
R = F340/F380 over time.

13. Calculate changes in intracellular calcium concentrations
[Ca2+]i for each ROI according to Eq. (1) [7]:
Ca 2+  = K d β (R − Rmin ) / (Rmax − R ) (1)
i
Use values for Rmin, Rmax, and β as derived in step 9 and a

Kd of 224 nM for Fura-2AM.
14. Measure the increase in intracellular calcium concentration for
each cell and analyze its dependence on extracellular calcium
and on release of calcium from intracellular stores. The follow-
ing parameters can be used to characterize calcium signals:
maximal increase in calcium concentration, time to increase of
calcium concentration, time to maximal calcium concentration,
and decay time of fluorescence signals. If the calcium concen-
tration declines to baseline levels, the area under the curve can
be measured.
3.2 Na+ Imaging 1. Prepare a 2 mM stock solution (see Note 27) of SBFI-AM in
equal volumes of DMSO and pluronic acid.
3.2.1 Detection of
Intracellular Na+ Changes 2. Dilute SBFI-AM stock solution in 2 ml solution E1 to a final
concentration of 10 μM. Prevent exposure to bright light.
3. Place microglial cells on glass cover slip in SBFI-AM solution
for 45–60 min at room temperature.
4. Wash dye-loaded microglial cells in solution E1 for 30 min at
room temperature.
5. Mount glass cover slip with dye-loaded microglial cells in
recording chamber filled with solution E1.
6. Check fluorescence intensity of SBFI-AM in microglial cells by
exciting the intracellular dye at 380 nm (see Note 28).
7. Choose a viewing field with several microglial cells. Take a
bright-field picture (see Note 29).
8. Take a sample picture of the same cells excited at 380 nm.
Increase the exposure time to maximize the emitted fluores-
cence signal. Use this exposure time in a protocol of the imaging
software.
9. Adjust the protocol of the imaging software to take every 20 s
a picture at 340 nm and at 380 nm. To observe the experi-
ment, create in the protocol editor a picture that divides the
340 nm signal by the 380 nm signal.
10. Start the protocol. Wait for a stable baseline level of the ratio-
metric signal.
11. Check the superfusion system by exchanging solution E1 in the
recording chamber. Ratiometric Na+ signals should not alter.
12. Superfuse cells with activators dissolved in solution E1. Monitor
changes in ratiometric SBFI-AM signals of microglial cells.
Ratiometric signals should reach baseline levels in at least ten

data points before using another stimulus.
13. Repeat experiments using solution E5 to ensure that changes
in SBFI fluorescence are dependent on extracellular sodium
(see Note 30).
3.2.2 Calibration of SBFI 1. Whenever an optical element like objectives or the type of glass
Fluorescence Signals cover slips is changed, the system should be recalibrated.
2. Follow steps 1–10 in Subheading 3.2.1.
3. Superfuse cells with solution E5 containing 5 μM gramicidin,
10 μM ouabain, and 10 μM monensin. Wait until a steady state
of the intracellular fluorescence signal has been achieved.
4. Exchange the solution in the recording chamber with solution
E1 containing 5 μM gramicidin, 10 μM ouabain, and 10 μM
monensin. Wait for a stable ratiometric fluorescence signal.
5. Repeat this calibration at least two more times with new cells.
3.2.3 Data Analysis Adapt the description from Subheading 3.1.5. To calculate intra-
cellular sodium concentrations, a Kd value of 11.3 mM for
SBFI-AM can be assumed (see Note 31).
3.3 Intracellular 1. Prepare a 2 mM stock solution (see Note 32) of BCECF-AM
pH Imaging with DMSO as the solvent. Do not expose BCECF-AM solution
to bright light.
3.3.1 Measurement
of pHi Changes 2. Dilute this stock solution in 2 ml E1 to achieve a 2 μM
BCECF-AM solution.
3. Transfer a glass cover slip with microglial cells into this
BCECF-AM solution.
4. After 10 min, wash dye-loaded microglial cells in solution E1
for another 10 min at room temperature.
5. Fill recording chamber of the microscope with solution E1
(see Note 33). Mount glass cover slip with BCECF-AM-loaded
microglial cells in recording chamber.
6. Take a bright-field picture of a viewing field. Take care that a
sufficient number of microglial cells are visible (see Note 34).
7. Take a sample picture of the same cells with excitation light at
490 nm. Change exposure time to a value, where the emitted
fluorescence intensity has a half-maximal value. Adjust the proto-
col of the imaging software for this exposure time.
8. Set the protocol of the imaging software to take one picture
with excitation at 490 nm and one at 440 nm every 20 s.
To monitor the experiment, calculate a ratiometric picture in
the protocol editor by dividing the emission signals at 490 and
440 nm excitation wavelengths.
9. Start the protocol. Wait for a baseline level of pHi signals for at
least 15 data points.
10. Superfuse cells with solution E1 as a control experiment. No
change in pHi signals should be observed (see Note 35).
11. Superfuse cells with activators dissolved in solution E1. Observe
pHi changes in microglial cells (see Note 36). Signals should
reach steady state in at least ten data points before using
another stimulus.
12. Calibrate BCECF-AM fluorescence signals (see Note 37) at
the end of each experiment. Superfuse cells with solution E1
containing 10 μM nigericin. Wait for a stable fluorescence signal
in at least ten data points.
13. Superfuse cells subsequently with extracellular solutions E6, E7,
and E8 containing 10 μM nigericin. After each solution, a
stable ratiometric signal should be observed.
3.3.2 Data Analysis 1. Open the bright-field picture with your image analysis software.
2. Place several regions of interest (ROI) in background areas, in
which no cells are visible.
3. Determine the mean background fluorescence intensity in the
490 nm fluorescence picture. Subtract this value from the stack
of fluorescence pictures excited at 490 nm.
4. Repeat the procedure with fluorescence pictures excited at
440 nm.
5. Create ROIs in the bright-field picture in every visible cell or
in identified subcellular compartments.
6. Determine the fluorescence intensity F for each cell at 490 nm
and 440 nm excitation from background subtracted pictures
(see Note 38).
7. Calculate for each cell the ratiometric fluorescence signal
R = F490/F440 over time.
8. Determine the ratio for each cell in the presence of extracellular
solution E6 (pH 7.6), E7 (pH 7.0), and E8 (pH 6.4).
9. Perform a linear regression fit of these three data points
between pHi and fluorescence ratio for each cell.
10. Convert fluorescence ratios for each cell into pHi values according
to this linear regression curve.
3.4 Determination 1. Follow steps 1–6 in Subheading 3.1.1.

of Cell Volume 2. Take a sample picture of the same viewing field with excitation
Changes (See Note 39) light at 360 nm. Change exposure time to optimize the fluo-
rescence signals emitted by microglial cells. Adjust exposure
time in protocol of the imaging software accordingly.
3. Start a protocol in imaging software for excitation at 360 nm
every 20 s.
4. Superfuse microglial cells with solution E1 containing desired

activators or a solution of increased/decreased osmolarity.
Observe changes in the emitted fluorescence signal (see Note 40).
5. For cell volume changes, analyze signals in background-
corrected fluorescence images excited at 360 nm. An increase
of the emitted fluorescence signal reflects an increase in Fura-
2AM concentration and therefore cell shrinkage. Accordingly,
a decrease of the Fura-2AM signal at 360 nm is detected during
cell swelling.
4 Notes
1. Objective lenses with 40× magnification are usually a good

compromise between cellular resolution and cell number per
viewing field. 20× magnification might be sufficient for detec-
tion of cytoplasmic fluorescence changes. If subcellular com-
partments are investigated, 63× magnification needs to be
considered. Inverted microscopes should preferentially be
equipped with water immersion objectives.
Fluorophores like Fura-2AM or SBFI-AM, which are
excited in the UV range, need special objectives for optimal
UV light transmission.
2. Monochromators have a variety of advantages. They can pro-
duce excitation light of a specific wavelength in a continuous
spectrum. Monochromators do not need excitation filters in
addition to dichroic and band-pass emission filters. Most
importantly, monochromators are able to change excitation
wavelengths fast and without mechanical impact on the
observed specimen. This is especially important if simultane-
ously patch-clamp recordings are performed.
3. Preparation of primary microglial cultures is described in ref.
[8]. Alternatively, a microglial cell line like BV-2 can be used.
4. Fura-2AM has several advantages. It is a ratiometric dye and
can therefore be calibrated in situ (see Subheading 3.1.4).
Additionally, photobleaching and dye leakage can easily be
detected and corrected for. Fura-2AM also emits a bright sig-
nal, which allows to load microglial cells at a concentration,
where the dye does not act as a significant calcium buffer.
The major disadvantage of Fura-2AM is its excitation wave-
length in the energy-rich ultraviolet range, which can lead to
chemical alterations of cellular proteins.
5. If intracellular calcium concentrations rise during the experi-
ment into the micromolar range, Fura-6F with a Kd value of
5.5 μM might be a better choice.
6. If simultaneously a second fluorophore with an emission spec-
trum similar to Fura-2AM is used, a 510 nm band-pass emission
filter might be necessary to separate both signals. Compared to

the suggested longpass filter, this band-pass filter reduces the
fluorescence intensity detected by the CCD camera.
7. Disadvantages of SBFI-AM are the excitation in the UV range
and the high Kd value. The former disadvantage can be over-
come by using the non-ratiometric dyes sodium green or
CoroNa green; both dyes have an even higher Kd value than
SBFI-AM.
8. This pKa is not optimal for measurements of pHi changes in acidic
compartments like lysosomes or mitochondria. A targeted expres-
sion of pHi-sensitive GFP variants might be indicated [9].
9. Alternatively, the following bicarbonate-buffered solution, fully
oxygenated at 37 °C with carbogen (95 % O2, 5 % CO2), can
be used: 129 mM NaCl, 3 mM KCl, 1.25 mM NaH2PO4,
10 mM d-glucose, 1.6 mM CaCl2, 1.8 MgSO4, and 21 mM
NaHCO3, 21. The calcium-free as well as the calibration solu-
tions need to be adjusted accordingly. This bicarbonate-
buffered solution should not be used to load fluorescent dyes
into microglial cells.
10. Thapsigargin binds irreversibly to calcium ATPases; a revers-
ible inhibitor is cyclopiazonic acid. A 10 mM stock solution of
cyclopiazonic acid should be prepared in DMSO.
11. Stock solutions of gramicidin and ouabain should be ultrasoni-
cated or vigorously vortexed before use.
12. To detect HCO3−-dependent processes (e.g., Cl−/HCO3−
exchanger or Na+/HCO3− co-transporter), use the following
bicarbonate-buffered solution fully oxygenated at 37 °C with
95 % O2 and 5 % CO2: 129 mM NaCl, 3 mM KCl, 1.25 mM
NaH2PO4, 10 mM d-glucose, 1.6 mM CaCl2, 1.8 MgSO4, and
21 mM NaHCO3. To load BCECF-AM into microglial cells,
HEPES-buffered solution E1 should be used.
13. A free download of ImageJ and its plug-ins is available under
http://rsbweb.nih.gov/ij/.
14. Stock solutions should be freshly prepared and kept in dark.
15. If an excessive amount of Fura-2AM crystals is visible during
the experiments, the dye is not sufficiently dissolved. In this
case, a mixture of 50 % DMSO and 50 % pluronic acid should
be used to prepare the stock solution.
16. Take care that several parts of the picture are not covered by
dye-loaded cells. These areas are essential for background
subtraction (see Subheading 3.1.5).
17. Some Ca2+ signals are faster and need a higher sample rate.
In most cases it is adequate to take a picture every 5 s.
18. If superfusion with extracellular solution alone evokes calcium
signals, microglial cells may detect mechanical stress.
Superfusion should be rearranged to prevent this. Another

common problem is a superfusion-induced movement of the
whole glass cover slip. Additional weights made of inert mate-
rial like platinum can be placed on the cover slip to prevent
movement of the observed cells.
19. An increase of the intracellular Ca2+ concentration is character-
ized by a decrease of the fluorescence intensity at 380 nm and
a simultaneous increase of the fluorescence intensity at 340 nm.
20. Some receptors show desensitization, which could influence
the second calcium response. It should be tested with solution
E1 whether repetitive stimulation evokes stable responses.
21. A prolonged (>10 min) exposure to calcium-free solutions
should be avoided because it can deplete intracellular calcium
stores.
22. Stock solution of activators should be prepared in a calcium-
free solution. Otherwise small calcium signals might be evoked
in calcium-free solution E2, which do not reflect calcium release
from intracellular stores.
23. This probably reflects in microglial cells the activity of calcium
release-activated calcium (CRAC) channels [10].
24. If a reversible inhibition of the calcium ATPase is necessary,
10 μM cyclopiazonic acid can be used.
25. This calibration method is not applicable for non-ratiometric
dyes like Fluo-3.
26. Fluorescence signals of all analyzed cells should be checked
thoroughly for dye leakage and photobleaching. This is indi-
cated by a rapid and simultaneous decline in all fluorescence
signals in one or more cells. As a consequence the values of
fluorescence intensities are very low, causing an excessive noise
signal in the ratiometric analysis.
28. In case of an insufficient fluorescence signal, ultrasonicate or
vortex SBFI-AM solutions or increase dye-loading time.
29. Take care that several parts of the picture are not covered by
dye-loaded cells. These areas are essential for background sub-
traction during data analysis (see Subheading 3.2.3).
30. Changes in SBFI-AM fluorescence signals do not necessarily
reflect increases in intracellular Na+ concentrations. Alterations
in intracellular pH, in the intracellular potassium concentra-
tion or direct chemical alterations of SBFI-AM can influence
the dye’s emission signal.
31. The Kd value of SBFI-AM can change depending on factors
like pH and potassium concentration. In reference [11] is
described, how the Kd value of SBFI-AM can be determined in
situ.

33. To detect HCO3−-dependent processes (e.g., Cl−/HCO3−
exchanger or Na+/HCO3− co-transporter), use the following
bicarbonate-buffered solution fully oxygenated at 37 °C with
95 % O2 and 5 % CO2: 129 mM NaCl, 3 mM KCl, 1.25 mM
NaH2PO4, 10 mM d-glucose, 1.6 mM CaCl2, 1.8 MgSO4, and
21 mM NaHCO3. To load BCECF-AM into microglial cells,
HEPES-buffered solution E1 should be used.
34. Several parts of the picture should not be covered by dye-loaded
cells. These areas are essential for background subtraction during
data analysis (see Subheading 3.3.2).
35. If superfusion with extracellular solution evokes changes in
ratiometric signals, microglial cells might be exposed to
mechanical stress and superfusion should be rearranged.
36. A change of the intracellular pH concentration is characterized
by an alteration of the fluorescence intensity at 490 nm with no
apparent change of the fluorescence intensity at the isosbestic
point at 440 nm.
37. A calibration method similar to the one described by [7] for
Fura-2AM signals can be found for BCECF-AM in ref. [12].
38. Fluorescence signals of all analyzed cells should be checked
thoroughly for dye leakage and photobleaching. This is indi-
cated by a decline in both fluorescence signals in one or more
cells. These cells should be excluded from further analysis.
39. Volume changes are detected as changes in Fura-2AM concen-
tration, measured at 360 nm excitation wavelength. This wave-
length is the isosbestic point for Fura-2 AM, at which the
fluorescence intensity is independent of the intracellular calcium
concentration.
40. This is not a ratiometric signal; the dye is only excited at
360 nm wavelength.
Acknowledgements
C.E. is supported by European Union FP7 collaborative grant

TargetBraIn (No. 279017).
References
1. Walz W, Ilschner S, Ohlemeyer C et al (1993) 2. Schilling T, Repp H, Richter H et al (2002)
Extracellular ATP activates a cation conduc- Lysophospholipids induce membrane hyperpo-
tance and a K+ conductance in cultured microg- larization in microglia by activation of IKCa1
lial cells from mouse brain. J Neurosci 13(10): Ca2+-dependent K+ channels. Neuroscience
4403–4411 109(4):827–835
3. Taupenot L, Ciesielskitreska J, Ulrich G et al improved fluorescence properties. J Biol Chem

(1996) Chromogranin a triggers a phenotypic 260(6):3440–3450
transformation and the generation of nitric 8. Eder C, Klee R, Heinemann U (1997)
oxide in brain microglial cells. Neuroscience Pharmacological properties of Ca2+-activated
72(2):377–389 K+ currents of ramified murine brain macro-
4. Nolte C, Möller T, Walter T et al (1996) phages. Naunyn Schmiedebergs Arch
Complement 5a controls motility of murine Pharmacol 356(2):233–239
microglial cells in vitro via activation of an 9. Pinton P, Rimessi A, Romagnoli A et al (2007)
inhibitory G-protein and the rearrangement of Biosensors for the detection of calcium and
the actin cytoskeleton. Neuroscience 73(4): pH. Methods Cell Biol 80(06):297–325
1091–1107 10. Nörenberg W, Cordes A, Blöhbaum G et al
5. Schilling T, Eder C (2004) A novel physio- (1997) Coexistence of purino- and pyrimidi-
logical mechanism of glycine-induced immu- noceptors on activated rat microglial cells. Br J
nomodulation: Na+-coupled amino acid Pharmacol 121(6):1087–1098
transporter currents in cultured brain macro- 11. Diarra A, Sheldon C, Church J (2001) In situ
phages. J Physiol 559(Pt 1):35–40 calibration and [H+] sensitivity of the
6. Faff L, Ohlemeyer C, Kettenmann H (1996) fluorescent Na+ indicator SBFI. Am J Physiol
Intracellular pH regulation in cultured microg- Cell Physiol 280(6):C1623–C1633
lial cells from mouse brain. J Neurosci Res 12. James-Kracke MR (1992) Quick and accurate
46(3):294–304 method to convert BCECF fluorescence to
7. Grynkiewicz G, Poenie M, Tsien R (1985) A pHi: calibration in three different types of cell
new generation of Ca2+ indicators with greatly preparations. J Cell Physiol 151(3):596–603
Chapter 17
Patch Clamp Protocols to Study Ion Channel

Activity in Microglia
Abstract
Microglia express a variety of ion channels, which can be distinguished based on their ion selectivity into K+,
H+, Na+, Ca2+, nonselective cation, and Cl− channels. With respect to their activation mode, voltage-, Ca2+-,
calcium release-, G protein-, swelling-, and stretch-activated ion channels have been described in microglia.
The best method to study the activity of microglial ion channels is the patch clamp technique. The activity
of microglial ion channels under physiological conditions is best explored using the perforated patch clamp
technique, which allows recordings of membrane potential or ion currents, while the intracellular milieu
of the cells remains intact. In whole-cell patch clamp recordings, application of specific voltage protocols
with defined intra- and extracellular solutions allows precise identification of a certain ion channel type in
microglia as well as the investigation of the channel’s biophysical and pharmacological properties.
This chapter summarizes patch clamp protocols optimal for recording and analysis of microglial ion channel
activity in vitro and in situ.
Key words Microglia, Ion channel, Patch clamp technique, Whole-cell recordings, Perforated patch
recordings, Voltage clamp, Current clamp, Voltage ramp, Voltage step
1 Introduction
Microglial cells express a variety of distinct ion channels, which play

pivotal roles in regulating microglial functions, such as prolifera-
tion, migration, cytokine production, and reactive oxygen species
generation (for reviews see refs. 1–4). Unlike many other cell types,
microglial cells do not constitutively express all ion channel types.
Expression of microglial ion channels strongly depends on the
functional state of the cells and on the cells’ microenvironment.
Furthermore, differences in the ion channel expression pattern
have been found between cultured microglial cells and microglia of
brain tissue, as well as between different species. Since microglial
ion channels provide potential therapeutic targets for neurological
and psychiatric diseases [3], it is of particular interest to identify
163
ion channels expressed in microglia under certain physiological

and pathophysiological conditions in the brain.
The patch clamp technique has been proven to be the best
method to study ion channel activity in excitable and non-excitable
cells, including microglia [5]. Here, we focus on protocols used for
whole-cell and perforated patch clamp recordings, since these
recording modes provide excellent approaches to study microglial
ion channel activity under physiological conditions. Whole-cell
recordings are used to assess ion channel activity of the entire cell
membrane. Due to the precise definition of intracellular and extra-
cellular solutions, it is possible to investigate certain ion channel
types in isolation. The perforated patch clamp method has been
developed, because intracellular second messenger systems can be
dialyzed out during whole-cell recordings leading to current run-
down. In perforated patch clamp recordings, pore-forming mole-
cules, such as nystatin, amphotericin, or gramicidin, allow electrical
access to the cell but prevent washout of larger intracellular con-
stituents. Therefore, whole-cell ion currents or membrane poten-
tial changes can be measured, while the cell interior remains intact
(see Note 1).
In this chapter, we list intracellular and extracellular solutions
and summarize voltage protocols optimal for measurements of
currents through known microglial K+, Na+, H+, nonselective
cation, Ca2+, and Cl− channels.
2 Materials
2.1 Equipment The patch clamp rig should at least consist of the following
for Patch Clamp components: vibration isolation table, microscope (inverted micro-
Experiments scope for in vitro recordings; upright microscope for in situ or
in vitro recordings) with an appropriate recording chamber, micro-
manipulators (for positioning of patch pipette and superfusion
pipette), and patch clamp amplifier and corresponding software for
data acquisition and analysis. For superfusion of cells with ion
channel activators or inhibitors, the use of a multibarrel microper-
fusion pipette is recommended in addition to a system allowing
complete exchange of bath solutions (see Note 2).
In addition, a puller is required to prepare patch pipettes prior
to patch clamp experiments. Patch pipettes are made from borosili-
cate glass capillaries (see Subheading 3.2). AgCl-coated silver wires
are used as patch and reference electrodes (see Subheading 3.3).
2.2 Cells Use cultured primary microglial cells (see Note 3) or a microglia
cell line (e.g., BV-2 cells) for in vitro studies. Prepare acute or cul-
tured organotypic brain slices (see Note 4) to study microglial cells
in situ.
Patch Clamp Recordings of Microglial Cells 165
2.3 Solutions Prepare and use extracellular solution E1 for storage of brain slices
and for in situ recordings of microglia in brain slices:
● Extracellular solution E1: 129 mM NaCl, 3 mM KCl, 1.8 mM
MgSO4, 21 mM NaHCO3, 1.25 mM NaH2PO4, 1.6 mM
CaCl2, 10 mM D-glucose (oxygenated with 95 % O2, 5 % CO2;
pH = 7.4) (see Note 5). For studies on the effects of polyvalent
cations, such as La3+, Gd3+, Ba2+, or Zn2+, replace MgSO4 by
MgCl2, and omit NaH2PO4 from this extracellular solution.
Prepare extracellular solutions E2–E6 for whole-cell record-
ings of microglial cells in vitro, and use intracellular solutions
I1–I7 (see Note 6) for both in vitro and in situ recordings of
microglia (see Note 7).
2.3.1 Solutions for ● Intracellular solution I1: 120 mM KCl, 5 mM NaCl, 2 mM

Measurements of MgCl2, 1 mM CaCl2, 10 mM N-[2-hydroxyethyl]piperazine-
Voltage-Activated K+ or N′-[2-ethanesulfonic acid] (HEPES), 11 mM
Na+ Currents ethylenebis(oxonitrilo)tetraacetate (EGTA) (pH = 7.3;
adjusted with KOH).
To investigate voltage-activated K+ or Na+ channels, use extra-
cellular solution E1 for in situ recordings or extracellular solution
E2 for in vitro measurements:
● Extracellular solution E2: 130 mM NaCl, 5 mM KCl, 2 mM
CaCl2, 1 mM MgCl2, 10 mM HEPES, 10 mM D-glucose
(pH = 7.4; adjusted with NaOH).
2.3.2 Solutions Ca2+-activated K+ currents can be measured using intracellular

for Measurements of solution I2 containing a low concentration of calcium buffer.
Ca2+-Activated K+ Currents Increases in the intracellular Ca2+ concentration should be elicited
either by stimulating Ca2+ influx from the extracellular space or by
inducing Ca2+ release from intracellular stores.
● Low Ca2+-buffered intracellular solution I2: 120 mM KCl,
2 mM MgCl2, 10 mM HEPES, 0.1 mM 1,2-Bis(2-
aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA)
(pH = 7.3; adjusted with KOH).
Alternatively, Ca2+-activated K+ currents can be measured using
intracellular solutions at which the free intracellular Ca2+ concen-
tration is clamped to values sufficient to directly activate Ca2+-
activated K+ channels. Using intracellular solution I3, an intracellular
free Ca2+ concentration of 1 μM is obtained (see Note 8).
● High Ca2+-containing intracellular solution I3: 120 mM KCl,
2 mM MgCl2, 5 mM BAPTA, 4.43 mM CaCl2, 10 mM HEPES
(pH = 7.3; adjusted with KOH).
Free Ca2+ concentrations of solutions can be calculated using

the program WINMAXC (see Note 9).
Use extracellular solution E1 for in situ measurements or extra-
cellular solution E2 for in vitro experiments.
2.3.3 Solutions for To measure G protein-activated K+ currents, use intracellular solution

Measurements of G I1 and add 50 μM GTPγS.
Protein-Activated K+ Use extracellular solution E1 for in situ measurements or extra-
Currents cellular solution E2 for in vitro experiments.
2.3.4 Solutions Intracellular and extracellular solutions are based on tetramethyl-

for Measurements ammonium (TMA) hydroxide and methanesulfonic (MeSO3) acid.
of H+ Currents ● Intracellular solution I4: 80 mM TMAMeSO3, 100 mM
2-(N-morpholino)ethanesulfonic acid (MES), 2 mM MgCl2,
1 mM BAPTA (pH 6.0; adjusted with TMAOH). When the
intracellular solution needs to be clamped to pH 7.0, use
100 mM HEPES instead of MES.
● Extracellular solution E3: 80 mM TMAMeSO3, 2 mM CaCl2,
100 mM HEPES, 1 mM EGTA (pH = 7.5 or 7.0). Use
100 mM MES instead of HEPES for studies at pH = 5.0, 6.0,
or 6.5 and use 100 mM tricine instead of HEPES for studies at
pH = 8.0.
2.3.5 Solutions for The extracellular and intracellular solution should be defined in a way
Measurements of that they differ in their reversal potentials for nonselective cation
Nonselective Cation (TRP) (TRP) and anion (Cl−) currents. Use extracellular solution E1 for
Channel Currents in situ measurements or extracellular solution E2 for in vitro exper-
iments. Prepare intracellular solution I5 (see Note 10).
● Intracellular solution I5: 20 mM NaCl, 100 mM Na-gluconate,
2 mM MgCl2, 1 mM CaCl2, 10 mM HEPES, 11 mM EGTA,
2 mM Na2ATP, 0.5 mM Na2GTP (pH = 7.3; adjusted with
NaOH).
Add 10 μM 4α-Phorbol 12,13-didecanoate (4α-PDD) to the
extracellular solution, to activate TRPV4 channels. To activate
TRPM2 channels, add 100 μM ADP-ribose to the intracellular
solution. Increase the free Ca2+ concentration of the intracellular
solution to 40 μM to measure TRPM4 channel activity. To activate
TRPM7 channels, omit Mg2+ and ATP from the intracellular
solution.
2.3.6 Solutions for Prepare extracellular solution E4 (see Note 11). Use E4 with intra-
Measurements of Calcium cellular solution I6 and add 10 μM inositol 1,4,5-trisphosphate
Release-Activated Ca2+ (InsP3) to intracellular solution I6.
(CRAC) Channels ● Extracellular solution E4: 130 mM NaCl, 10 mM CaCl2, 1 mM
MgCl2, 10 mM HEPES, 10 mM D-glucose (pH = 7.4; adjusted
with NaOH).
● Intracellular solution I6: 120 mM NaCl, 2 mM MgCl2, 1 mM

CaCl2, 10 mM HEPES, 11 mM EGTA, 2 mM Na2ATP,
0.5 mM Na2GTP (pH = 7.3; adjusted with NaOH).
2.3.7 Solutions for Solutions for Cl− current measurements should differ in their reversal
Measurements potentials for nonselective cation (TRP) and anion (Cl−) currents.
of Cl− Currents Use intracellular solution I5 with extracellular solution E1 for in situ
measurements or extracellular solution E2 for in vitro experiments
(see Note 10). Add 100 μM LaCl3 to the extracellular solution to
block voltage-gated H+ currents, if necessary.
To study swelling-activated Cl− currents, prepare intracellular
solution I6 and use this with isoosmolar extracellular solution E5
and hypoosmolar solution E6 (see Note 12):
● Isoosmolar (300 mosmol/kg) solution E5 containing 50 mM
NaCl, 170 mM D-mannitol, 2 mM CaCl2, 1 mM MgCl2, 10 mM
HEPES, 10 mM D-glucose (pH = 7.4; adjusted with NaOH).
● Hypoosmolar (200 mosmol/kg) solution E6 containing
50 mM NaCl, 70 mM D-mannitol, 2 mM CaCl2, 1 mM MgCl2,
10 mM HEPES, 10 mM D-glucose (pH = 7.4; adjusted with
NaOH).
2.3.8 Solutions Use extracellular solution E1 for in situ measurements or extracellular

for Perforated Patch solution E2 for in vitro experiments.
Clamp Recordings ● Prepare solution I7 containing 120 mM K2SO4, 16 mM KCl,
5 mM MgSO4, 10 mM HEPES (pH = 7.3; adjusted with KOH).
● Prepare a stock solution of nystatin in DMSO (25 mg/ml),
which should be stored in a freezer. Dilute this stock subse-
quently in solution I7 to a final concentration of 150–250 μg/ml
(see Note 13). Sonicate this solution for 10 min prior to its use
for backfilling the patch pipette. The final concentration of
nystatin used for perforated patch recordings needs to be adjusted
carefully (see Note 14). Amphotericin (200 μg/ml) or gramici-
din (5 μg/ml) can be used instead of nystatin (see Note 15).
3 Methods
3.1 Preparation To study ion channel activity of isolated microglial cells in vitro,
of Cells for In Vitro either primary cultured microglial cells (see Note 3) or a microglial
or In Situ Patch Clamp cell line can be used. Depending on the recording chambers used,
Experiments microglial cells should be plated either into plastic dishes or onto
glass coverslips at least 1 day prior to patch clamp recordings.
As microglial cells adhere well to glass or plastic surfaces, coating
of recording chambers is not required.
To study ion channel activity of microglial cells in situ, use
either acutely prepared or cultured organotypic brain slices
(see Note 4). For investigations of microglial cells in brain tissue,
use transgenic mice, in which microglial cells have been labeled

with a genetically encoded fluorescent dye, e.g., Iba1/GFP-labeled
microglia or CX3CR1/GFP-labeled microglia.
If animals with GFP-labeled microglia are not available, stain
microglial cells in brain slices prior to experiments with an isolectin-
B4-conjugated Alexa dye (e.g., Alexa 488-IB4). For microglial
staining, keep individual brain slices in a chamber with oxygenated
solution E1 containing 10 μg/ml Alexa-IB4 (see Note 16). Prepare
the final Alexa dye-containing solution from a stock solution of
1 mg/ml Alexa-IB4, which should also be prepared in solution E1.
Prior to each patch clamp experiment, incubate one brain slice in
the dark for 20 min in Alexa-IB4-containing solution E1 at room
temperature. After the incubation period, transfer the brain slice to
the recording chamber under the microscope and wash the slice
with oxygenated solution E1 for 20 min before starting patch
clamp recordings (see Note 17).
3.2 Preparation Patch electrodes consist of a patch pipette filled with an intracel-
of Patch Pipettes lular solution that has contact to the patch clamp amplifier via an
AgCl-coated silver wire. Patch pipettes are prepared from borosili-
cate glass capillaries, which are commercially available in varying
diameters and lengths (see Note 18). The appropriate capillary
diameter depends on the pipette holder of the patch clamp ampli-
fier in use.
Usually pullers form patch pipettes in two or more steps.
The first step defines the geometry of the tip (see Note 19). The
last step determines the pipette resistance. Increasing the tempera-
ture in this step leads to a smaller diameter of the tip orifice, i.e., a
higher pipette resistance.
Prepare patch electrodes of 3–5 MΩ (see Note 20) immedi-
ately before their use. Store patch electrodes in a box to protect
tips from dust exposure.
Polish pipette tips for a higher seal resistance, if necessary. Care
should be taken that this polishing procedure does not substan-
tially increase the pipette resistance above 5 MΩ. Coat pipette tips
with insulating polymers like Sylgard 184, if a decreased noise level
is desired (see Note 21).
3.3 Preparation A reference electrode (also called bath electrode) should connect
of Reference the ground of the patch clamp amplifier with the extracellular solu-
Electrodes tion in the recording chamber. These electrodes are usually silver
wires with a AgCl coating (see Note 22). They are commercially
distributed in different varieties.
A better reference electrode, especially when solutions contain
low chloride concentrations, is an agar bridge. They are commer-
cially available but can also be prepared by heating a 3 M KCl solu-
tion containing 2–5 % agar. Fill a glass capillary (see Note 23) with
this mixture and introduce a Ag/AgCl wire in one end of the glass
tube. Fixate this wire with heat-shrinkable tubing. After 1 day

remove excess agar outside the glass tube and discard tubes con-
taining air bubbles. Store agar bridges in 3 M KCl solution between
experiments.
3.4 Patch Clamp It is recommended to perform in situ patch clamp recordings at

Recordings: a temperature of 34–36 °C (see Note 24), while in vitro patch
Establishment clamp recordings can also be carried out at room temperature
of a Giga-Seal and (20–23 °C).
of a Perforated Patch 1. Maintain cells in extracellular solution E1 for in situ experiments
or Whole-Cell or in extracellular solution E2 for in vitro experiments.
Configuration
2. Bring the reference Ag/AgCl electrode in contact with the
extracellular solution in the recording chamber.
3. Fill a patch pipette with the appropriate intracellular solution.
Attach the pipette to the pipette holder (see Note 25). Make
sure that the electrode wire is in contact with the intracellular
solution and no air bubbles are trapped in the patch pipette.
4. Choose a microglial cell from which recordings should be per-
formed (see Note 26).
5. Apply a test pulse at an appropriate frequency (e.g., 10 mV for
10 ms every second) to visualize changes in pipette resistance,
when approaching the cell. The pipette resistance should be
3–5 MΩ (see Notes 20 and 27).
6. Bring the patch pipette in close contact with the cell surface,
while applying constantly a positive pressure on the pipette.
The recording pipette is in close proximity of the cell mem-
brane, when the pipette resistance increases by a few MΩ
(see Note 28). Apply gentle suction to the patch pipette to
increase the resistance between the patch pipette and the cell
membrane (see Note 29) until it reaches values of more than
1 GΩ, i.e., a giga-seal (see Note 30).
7. After formation of the giga-seal, break the membrane within
the pipette tip by fast application of additional suction to reach
the whole-cell configuration.
8. For perforated patch clamp recordings, the patch pipette tip
should be dipped briefly in solution I7 (see Note 31) and then
backfilled with solution I7 to which the pore-forming agent,
i.e., nystatin, amphotericin, or gramicidin, has been added.
After formation of the giga-seal, small pores are formed in
the membrane patch within the pipette. To visualize this
pore-forming process, apply a test pulse similar to the one used
during giga-seal formation. The pores lower the series resis-
tance allowing voltage or current clamp recordings of the
entire cell membrane. Optimally this pore-forming process
lasts for about 10–30 min, and the series resistance reaches a
stable value of less than 20 MΩ. Make sure that the input
resistance, i.e., the giga-seal, remains stable during the pore-
forming period.
9. After establishing electrical access to the inside of the cell,
series resistance and membrane capacitance (see Note 32)
should be compensated for. Some amplifier models are able to
automatically compensate for series resistance and membrane
capacitance. The series resistance in the whole-cell configura-
tion should not exceed two times the value of the pipette resis-
tance (see Note 33), while the input resistance should still be
in the GΩ range (see Note 34).
10. The superfusion pipette should be positioned very close (at a
distance of a few μm) to the recorded cell to permit a rapid and
complete exchange of solutions outside the cell (see Note 35).
3.5 Patch Clamp 1. Maintain cells in extracellular solution E1 for in situ experi-
Recordings: Detection ments or in extracellular solution E2 for in vitro experiments.
of Changes in 2. Switch the amplifier to current clamp mode, while no holding
Membrane Potential in current should be applied. This allows immediate determina-
the Current tion of the resting membrane potential of a cell.
Clamp Mode 3. To investigate changes in membrane potential during drug
application, record membrane potential continuously before,
during, and after application of certain drugs. All current clamp
recordings should ideally be performed in the perforated patch
configuration to allow recordings under physiological condi-
tions from intact cells and to prevent washout of intracellular
components.
3.6 Patch Clamp 1. Maintain cells in extracellular solution E1 for in situ experi-
of Ion Currents in the 2. Switch the amplifier to voltage clamp mode to measure ion
Voltage Clamp Mode currents in microglia.
3. Clamp cells to a holding potential, which is close to the resting
membrane potential of the cell, does not activate the investi-
gated ion channel and allows full activation of the ion channel
of interest (see Note 36).
4. To elicit and visualize ion channel activity, apply either a volt-
age ramp or a voltage step protocol (see Note 37). Voltage
ramp protocols allow determination of current activation
threshold and observation of voltage-dependent behavior.
Voltage-step protocols allow assessment of current kinetics,
i.e., time-dependent activation and inactivation behavior of the
currents at a certain voltage. When using voltage-step proto-
cols, it is recommended to record a leak protocol after each
measurement. For this, step from the holding potential to a
potential, where no channel activity is detected, e.g., from
−60 to −70 mV. The current measured in this protocol is the
leak current. To minimize noise, average ten traces of this leak

current recording.
5. Ion substitution experiments can be performed to identify
the activated channel. Omit the suspected charge carrier from
the solutions and observe, if the current can still be elicited.
Care should be taken that the substituted ion is not conducted
by the ion channel under investigation (see Note 38).
6. Use tail current protocols to determine current reversal poten-
tials (see Note 39). Tail current measurements are voltage-
step protocols, consisting of two consecutive voltage steps.
The first step is set to a fixed voltage value that fully activates the
channel under investigation (see Note 40). The second step
measures tail currents by varying the voltage around the
expected equilibrium potential. The tail currents, i.e., the
currents observed in this second voltage step, should switch
from outward to inward currents or vice versa. The voltage,
where no current flows, is the reversal (i.e., equilibrium)
potential for this ion channel.
7. Apply known ion channel activators and inhibitors and observe
their effect on the measured current. Usually a combination
of different inhibitors and measurement of their Kd value is
necessary to conclusively identify an ion channel type.
3.6.1 Patch Clamp 1. Maintain cells in extracellular solution E1 for in situ experiments
Recordings: Detection of or in extracellular solution E2 for in vitro experiments.
K+ Currents 2. Perform recordings in the perforated patch clamp or whole-
cell mode. For whole-cell recordings, fill the patch pipette with
intracellular solution I1 for recordings of inward-rectifying or
outward-rectifying voltage-activated K+ currents, with intracel-
lular solution I2 or I3 for recordings of Ca2+-activated K+ cur-
rents, or with intracellular solution I1 containing 50 μM GTPγS
for recordings of G protein-activated K+ currents.
3. Clamp cells to a holding potential of −60 mV (see Note 41).
4. Apply a voltage ramp from −150 mV to +30 mV for 400 ms.
5. Apply 200 ms lasting voltage steps to potentials between
−150 mV and +30 mV in 10 mV increments, while an inter-
val of 20 s between voltage pulses should be allowed (see
Note 42).
6. To block Kir2.1 inward-rectifying voltage-activated potassium
channels (see Note 43), use 100 μM Ba2+.
7. To identify Kv1.3 outward-rectifying voltage-activated K+
channels (see Note 44), test the effect of 100 nM margatoxin
or 50 nM agitoxin.
8. To identify KCa3.1 (see Note 45) Ca2+-activated K+ channels,
test the effect of 100 nM charybdotoxin (see Note 46) and
10 μM TRAM-34 (see Note 47).
9. To identify KCa1.1 (see Note 48) Ca2+-activated K+ channels,

use 10 μM paxillin or 100 nM iberiotoxin.
10. Apply 100 nM apamin or 5 nM tamapin, to identify KCa2.3
(see Note 49) Ca2+-activated K+ channels.
11. Add 2 mM GDP-βS to the intracellular solution to inhibit G
protein-activated K+ channels (see Note 50).
12. To identify ATP-sensitive K+ channels (KATP), test the effect
of 10 μM glibenclamide or 500 μM tolbutamide.
Recordings: Detection or in extracellular solution E2 for in vitro experiments.
of Na+ Currents 2. Perform recordings in the perforated patch clamp or whole-cell
mode. For whole-cell recordings, fill the patch pipette with
intracellular solution I1 (see Note 51).
3. Clamp cells to a holding potential of −80 mV.
−80 mV and +30 mV in 10 mV increments at an interval of 10 s
(see Note 52).
5. Test whether 1 μM tetrodotoxin (TTX) (see Note 53) induces
Na+ current inhibition.
3.6.3 Patch Clamp 1. Maintain cells in extracellular solution E1 for in situ experi-
of H+ Currents 2. Perform recordings in the perforated patch clamp or whole-
intracellular solution I4.
3. Clamp cells to a holding potential of −60 mV (see Note 54).
4. Apply voltage steps for a duration of 4 s in 20 mV increments
from −60 mV to +80 mV at 60 s intervals (see Notes 55
and 56).
5. Change pH of intracellular and/or extracellular solutions
and repeat the voltage-step protocol in order to test whether
the H+ current activation threshold changes as predicted
(see Note 57).
6. Test whether 100 μM Zn2+ or 100 μM La3+ induces H+ current
inhibition.
Recordings: Detection or in extracellular solution E2 for in vitro experiments.
of Nonselective Cation 2. Perform recordings in the perforated patch clamp or whole-
(TRP) Channel Currents cell mode. For whole-cell recordings, fill the patch pipette with
3. Clamp cells to a holding potential of 0 mV.
4. Apply voltage ramps from −90 mV to +90 mV for 400 ms
every 20 s.
5. Alternatively, apply 200 ms lasting voltage steps to potentials

between −90 mV and +90 mV in 10 mV increments at an
interval of 10 s (see Note 58).
6. Test first the effects of 50 μM Gd3+ or 10 μM ruthenium red,
which inhibit most TRP channels.
7. Test for TRPV1 channels by applying 10 μM capsazepine or
100 nM 5′-iodo-resiniferatoxin.
8. Test 50 μM 2-APB or 100 μM flufenamic acid, which inhibit
most TRPC channels (see Note 59).
9. Test 10 μM clotrimazole (see Note 60) to inhibit TRPM2
channels.
3.6.5 Patch Clamp 1. Maintain cells in extracellular solution E4.

Recordings: Detection of 2. Perform recordings in the whole-cell mode. Fill the patch
Calcium Release-Activated pipette with intracellular solution I6 to which 10 μM InsP3 has
Ca2+ (CRAC) Currents been added.
3. Clamp cells to a holding potential of 0 mV.
4. Apply voltage ramps from −90 mV to +90 mV for 400 ms
every 20 s (see Note 61).
−90 mV and +90 mV in 10 mV increments at an interval
of 10 s.
6. Test the inhibitory effect of 100 μM 2-APB or 10 μM Gd3+ on
CRAC currents.
3.6.6 Patch Clamp 1. Maintain cells in extracellular solution E1 for in situ experi-
of Cl− Currents 2. Perform recordings in the perforated patch clamp or whole-
3. Clamp cells to a holding potential of −40 mV.
4. Apply voltage ramps from –90 mV to +90 mV for 400 ms
every 10 s.
5. Alternatively, apply 200 ms lasting voltage steps to potentials
between −90 mV and +90 mV in 10 mV increments at an
interval of 10 s (see Note 62).
6. To activate swelling-induced Cl− currents, maintain cells in
isoosmolar extracellular solution E5 and use intracellular
solution I6.
7. Clamp cells to a holding potential of 0 mV and use voltage
ramp or voltage-step protocols as described before.
8. Superfuse cells with hypoosmolar solution E6.
9. Test the effect of 1 mM DIDS or 100 μM NPPB on
Cl− currents.
3.7 Data Analyses Determine the mean membrane potential of microglial cells, if
experiments have been performed in the current clamp mode.
3.7.1 Analysis of Current
These should be preferentially perforated patch clamp experiments.
Clamp Recordings
Compare this value with the mean membrane potential of cells in
experiments with certain ionic substitutions. Measure membrane
potentials of microglial cells before, during, and after application
of ion channel activators/inhibitors. Determine after drug applica-
tion the delay of changes in membrane potentials and the time to
reach a new stable membrane potential.
3.7.2 Analysis of Voltage Determine the activation threshold of voltage-activated channels

Clamp Recordings: (e.g., voltage-activated K+, Na+, or H+ channels) from voltage ramp
Voltage-Activated Currents or voltage-step protocols. If voltage-step protocols have been
used, subtract leak currents (see Note 63) from the recorded ion
currents.
Use voltage-step protocols to observe whether ion currents show
time-dependent (i.e., changes in current over time in one voltage
step) or time-independent (i.e., no current change over time) behav-
ior. Fit current traces with exponential functions (inward-rectifying
voltage-activated K+ channels, voltage-activated H+ channels) or
according to a n4j model of a Hodgkin-Huxley kinetics (outward-
rectifying voltage-activated K+ channels). Determine the activation
and inactivation time constants at different voltages.
Measure the reversal potential (see Note 64) of the investi-
gated ion current. The best way to determine the reversal potential
is to measure the voltage at which no tail currents flow (see Tail
current protocols in Subheadings 3.6). Using voltage ramps, this is
the membrane potential at which the current trace intersects with
the zero line. The reversal potential gives a hint to the selectivity
of the investigated channel (see Note 65).
Determine the current density (see Note 66), which is dependent
on the number of expressed channel proteins and their single channel
conductance (see Note 67). Measure the current amplitude at a
given potential and subtract the leak current (see Note 63) from
this value. Divide the leak-subtracted current amplitude by cell
capacitance (see Subheading 3.4) as a measure of cell surface area
to calculate the current density.
Quantify the effect of ion substitutions. Measure the change in
current density, reversal potential, activation threshold, and time-
dependent behavior before and after certain ion substitutions.
Quantify the effect of ion channel inhibitors. Measure (if pos-
sible, from leak-subtracted current amplitudes) the percentage
of current inhibition. Calculate IC50 values of inhibitors from a
concentration-response curve. For this calculation, measure-
ments with at least five different inhibitor concentrations are nec-
essary. Fit all data points according to a Hill equation to determine
the IC50 value. Determine whether current inhibition is voltage
dependent or voltage independent. Examine whether the
inhibitor changes activation threshold or reversal potential of the

current. Determine whether current inhibition is fully, partially,
or not reversible.
3.7.3 Analysis of Voltage Extract for ligand-activated (e.g., Ca2+-, G protein-, or calcium
Clamp Recordings: release-activated) channels the current induced by these agents by
Ligand-Activated Currents subtracting the current trace before drug application from the one
during drug application. The same principle applies to swelling-/
stretch-activated (e.g., Cl−) channels. Establish whether the cur-
rent exhibits a linear (i.e., a voltage independent) or nonlinear
(i.e., a voltage dependent) I-V relationship. Examine voltage-step
protocols for time-dependent or time-independent behavior of ion
currents.
Determine the reversal potential of ion currents either from
tail current protocols or from voltage ramp protocols.
Calculate current densities from leak-subtracted amplitudes
using voltage-step protocols. Alternatively, use voltage ramp
protocols to calculate the conductance and normalize this value to
cell capacitance.
Quantify changes in current density/normalized conductance,
reversal potential, and time- and voltage-dependent behavior
before and after certain ion substitutions.
Determine the effect of ion channel inhibitors. Calculate the
percentage of current or conductance inhibition and determine the
IC50 values of inhibitors using a Hill equation. Examine whether
inhibition of currents is voltage dependent or independent.
Compare reversal potentials before, during, and after application
of ion channel inhibitors. Determine whether current inhibition
is reversible.
4 Notes
1. A more detailed description on how to investigate biophysical

and pharmacological properties of ion channels using the patch
clamp technique can be found in [5].
2. Bath application of drugs is possible, but care should be taken
that the measured ion channel or receptor does not desensitize
when the extracellular solution is slowly exchanged.
Electrophoretical or pressure-induced drug application is
another possibility to consider.
3. A method for isolation of microglial cells and their subsequent
culture is described in [6].
4. Protocols for preparation of hippocampal brain slices can be
found in [7] for acutely isolated brain slices and in [8] for
organotypic cultures.
5. A 10× concentrated stock of extracellular solution E1 can be

prepared, if the NaHCO3 stock solution is stored separately.
Extracellular solution E1 should be prepared freshly every day
from these stock solutions.
6. Add 2 mM Na2ATP and 0.5 mM Na2GTP to the intracellular
solution if current rundown is observed during whole-cell
recordings.
7. Check osmolarity of all solutions to prevent swelling or shrink-
age of cells. It is recommended that the extracellular solution
has a slightly higher osmolarity than the intracellular solution.
8. To measure KCa1.1 (BK) channels, it is advisable to increase the
free calcium concentration of intracellular solution I3 to 10 μM.
9. A Windows version of this program can be downloaded from
http://www.stanford.edu/~cpatton/downloads.htm.
10. The reversal potential for extracellular solution E1 or E2 versus
intracellular solution I5 is +5 mV for nonselective cation channels
and −40 mV for Cl− channels.
11. Alternatively, omit KCl from extracellular solution E1 and raise
the CaCl2 concentration to 10 mM.
12. The reversal potential for extracellular solution E5 or E6 versus
intracellular solution I6 is −20 mV for nonselective cation
channels and +20 mV for Cl− channels.
13. Do not store the final nystatin-containing pipette solution
for more than 1 h at room temperature. It is recommended to
prepare for each new cell under investigation a fresh nystatin-
containing solution from stock.
14. The final concentration of the pore-forming substances nystatin,
amphotericin, or gramicidin used for perforated patch record-
ings may vary between cell preparations depending on their
membrane properties. A concentration should be chosen, at
which the series resistance decreases within 10–30 min continu-
ously, reaching stable values below 20 MΩ. The use of concen-
trations above the optimum results in either losing the giga-seal
or in rupture of the cell membrane under the pipette leading
to a whole-cell configuration. No or insufficient changes in
series resistance occur over a period of more than 30 min
recording time, if the concentration of the agent is too low.
15. Amphotericin or gramicidin does not have any substantial
advantage compared to nystatin. They have similar limitations
regarding their chemical stability.
16. The stock and final solution of fluorescent dyes should not be
exposed to bright light and kept in the dark prior and during
the experiment.
17. For patch clamp recordings of microglial cells in situ, we rec-
ommend to add a fluorescent dye with a different excitation
wavelength (e.g., Alexa647) to the intracellular solution.

This approach provides a good control whether an extracellu-
larly labeled microglial cell is identical to the cell stained intra-
cellularly during the patch clamp recording. Add the Alexa dye
at a final concentration of 2 μM to the intracellular solution
used. Prepare the final solution from a frozen stock solution of
1 mM Alexa dye, which can be prepared in intracellular solution
or water.
18. The silver chloride coating of the electrode wire is easily scraped
off by glass capillaries. Therefore, it is recommended to fire pol-
ish both ends of the capillaries before pulling the patch pipette.
19. A short conical tip facilitates diffusion of the intracellular solu-
tion into the cell. This is of advantage for perforated patch
recordings as well as for measurements of Ca2+-activated, G
protein-activated, and CRAC channels.
20. A pipette resistance of up to 10 MΩ is tolerable, if the intracel-
lular solution mainly contains large molecules as charge carriers
(e.g., to measure proton channel activity, see Subheading 2.3.4).
21. Neither tip polishing nor tip coating are usually required for
patch clamp recordings of microglial cells with an exemption
being single channel recordings. Try first to renew the filament
of the microelectrode puller, if the seal resistance is constantly
insufficient.
22. All Ag/AgCl electrodes, i.e., pipette and reference electrodes,
should be stored in bleach overnight.
23. Alternatively, disposable pipette tips with a trimmed opening
can be used instead of a small glass tube.
24. Due to the bicarbonate buffer, the pH value of extracellular
solution E1 is temperature dependent.
25. During this procedure the grounded table should be touched
to avoid electrostatic discharges, which can destroy the patch
clamp amplifier.
26. IB4-conjugated Alexa dyes label microglial cells as well as endo-
thelial cells. Due to their different morphology, microglia can
be easily distinguished from endothelial cells, which delineate
blood vessels.
27. A much higher resistance can be caused by air bubbles in the
patch pipette, lack of contact between electrode wire and intra-
cellular solution, no contact between reference electrode and
extracellular solution, or a very small pipette tip opening.
28. Alternatively, it is possible to observe cell membrane dimpling,
when the recording pipette is close enough to the cell’s surface.
29. This membrane resistance is also called input resistance. It is
mainly determined by the quality of the seal and the conduc-
tivity of all ion channels.
30. Check that patch pipettes do not drift during the recording
if you experience problems with this step. Make sure that
vibrations from the surroundings are not transmitted to the
patch pipette. Use only freshly pulled pipettes; do not store
patch pipettes for more than 1 day. Polish pipette tips using a
microforge, if necessary.
31. Do not omit this step even if you use glass capillaries with fila-
ments. Tip filling with solution I7 not containing pore-forming
molecules is required to achieve giga-seal formation.
32. The membrane capacitance correlates with the cell surface area
and can therefore be used to normalize currents.
33. A high series resistance, especially in combination with a low
input resistance, causes a voltage drop, i.e., the command voltage
from the amplifier is not fully detected by ion channels in the
cell membrane. Additionally, diffusion of intracellular solution
from the patch pipette into the cell is limited.
34. A decreased input resistance can be caused by losing the tight
seal between pipette and cell membrane. An input resistance in
the MΩ range is also detectable, when channels are already
open, e.g., a high calcium concentration in the pipette gates
calcium-activated potassium channels.
35. The flow of the superfusion pipette should be tested with con-
trol solution before the actual patch clamp recording is per-
formed. Sometimes the flow from the superfusion pipette does
not exchange the solution around the patched cell; the super-
fusion pipette needs to be repositioned in this case. Another
common problem is blockage of the superfusion system by air
bubbles. Test the superfusion pipette flow at a safe distance
from the patched cell. It should also be tested with drug-free
control solution that ion channel activity is not altered during
solution exchange.
36. If possible, activation of ion channels other than the investigated
voltage-activated ion channel should be avoided. Otherwise a
permanent ion influx or efflux is induced with subsequent cell
volume changes.
37. Adjust the sample frequency to a value that the relevant voltage
step or voltage ramp records at least 2,000 data points. Typical
sample frequencies are 5–10 kHz for K+ channel measure-
ments, 50–100 kHz for Na+ channel measurements, or
0.5–1 kHz for H+ channel measurements.
38. Typical substitutions are N-methyl-D-glucamine (NMG) or
TMA for monovalent cations and gluconate for Cl− ions.
Omission of Ca2+ or Mg2+ should be accompanied by inclusion
of chelators like BAPTA or EGTA.
39. A reversal potential of a channel is also called equilibrium

potential and allows to determine which ions are conducted by
the investigated channel.
40. Care should be taken that the current, elicited in this first step,
does not change substantially during the whole measurement.
If changes occur, change the voltage value in this first step or
allow more time between steps.
41. For voltage-activated potassium channel measurements, the
membrane potential should be −60 mV. Outward-rectifying
Kv1.3 K+ channels are partially inactivated at more positive
holding potentials; inward-rectifying Kir2.1 K+ channels activate
at more negative holding potentials.
42. Some ion channels in microglia, e.g., voltage-activated Kv1.3
K+ channels, exhibit strong use dependence. Therefore, during
voltage-step protocols, sufficient time for recovery should be
allowed between individual voltage steps.
43. An example for an inward-rectifying K+ current in microglia
can be found in [1], Fig. 1.
44. A typical outward-rectifying voltage-activated K+ current is
shown in [1], Fig. 2.
45. This channel is also known as KCNN4, IKCa1, IK1, and SK4.
A representative KCa3.1 current is depicted in [1], Fig. 3.
46. Charybdotoxin also inhibits Kv1.3 and KCa1.1 channels.
47. TRAM-34 does not only inhibit Ca2+-activated K+ channels
but also nonselective cation channels in microglia [9].
48. This KCa1.1 channel has also the following names: KCNMA1,
BK, Slo, and maxiK. A representative current recording for this
channel is shown in [7], Fig. 4.
49. KCa2.3 channels are also called KCNN3 and SK3. A represen-
tative current recording is shown in Fig. 1 of [10].
50. Figure 6 in [11] shows an example of a G protein-activated K+
current.
51. Substitute K+ ions in all solutions with Cs+, if large K+ currents
interfere with the measurement of Na+ currents.
52. A typical Na+ current recording in microglia is shown in Fig. 2
of [12].
53. Microglial cells predominantly express TTX-sensitive Nav1.6
and Nav1.1 sodium channels but also TTX-resistant Nav1.5
subunits [13].
54. The holding potential should be set below the activation
threshold Vthreshold of voltage-activated H+ channels, which can
be calculated from Vthreshold = 20 mV − 40 mV(pHo − pHi) [14].
55. For investigation of voltage-activated H+ channels, time

between individual voltage steps should be at least 30 s to
minimize intracellular proton depletion and accumulation of
protons outside the cell membrane [15].
56. Figure 5 in [1] shows a typical recording of a voltage-activated
H+ current.
57. The activation threshold of voltage-activated H+ channels
can be calculated using the equation Vthreshold = 20 mV − 40 mV
(pHo − pHi) [14].
58. Nonselective cation currents are shown in Fig. 1 of [ 16 ].
A typical example of a TRPV1 current is shown in [17] Fig. 1,
a TRPV4 current in [18] Fig. 2, a TRPM2 current in [19]
Fig. 3, a TRPM4 current in [20] Fig. 4, and a TRPM7 current
in [21] Fig. 1.
59. In contrast to other TRPC channels, TRPC6-mediated currents
are potentiated by flufenamic acid.
60. Clotrimazole also inhibits Ca2+-activated KCa3.1 K+ channels
61. A representative example of a CRAC current is shown in [22],
Fig. 7.
62. Figure 6 in [1] shows a typical recording of a microglial stretch-
activated Cl− current.
63. Leak currents have a linear I-V relationship. Therefore, the
leak current for a voltage step of 20 mV is two times the leak
current for a 10 mV voltage step. This leak subtraction method
cannot be applied to ion channels, which are not voltage acti-
vated (e.g., the calcium-activated channel KCa3.1).
64. The reversal potential (or equilibrium potential) is often called
Nernst potential in the literature. They are distinct, when more
than one ion channel with a large permeability is open or the
ion channel is permeable for more than one ionic species. In
other words, if only one channel type is open and this channel
is selective for one ion, reversal and Nernst potential are identical.
The Nernst potential for one ion can be determined using the
following online calculator: http://www.physiologyweb.com/
calculators/nernst_potential_calculator.html
65. For example, the activated channel is most likely a potassium
channel, if the reversal potential is near the Nernst potential for
potassium. The reversal potential is difficult to interpret, when
Nernst potentials for several ions do not differ significantly.
Often the Nernst potential for Cl− is 0 mV, which is also the
reversal potential for nonselective cation/TRP channels. This
is not the case for the solutions described in Subheadings 2.3.5
and 2.3.7—the reversal potentials for Cl− and cation currents
differ significantly.
66. Microglial activation often results in changes of cell size and

morphology. Current density normalizes the current for cell
surface area and is therefore a better marker for ion channel
expression than current amplitude.
67. Both channel number and single channel conductance can
change depending on intra- or extracellular activators/
modulators.
Acknowledgements
C.E. is supported by European Union FP7 collaborative grant

TargetBraIn (No. 279017).
References
1. Eder C (1998) Ion channels in microglia 11. Schilling T, Repp H, Richter H et al (2002)
(brain macrophages). Am J Physiol 275(2 Pt Lysophospholipids induce membrane hyper-
1):C327–C342 polarization in microglia by activation of
2. Eder C (2005) Regulation of microglial behav- IKCa1 Ca(2+)-dependent K(+) channels.
ior by ion channel activity. J Neurosci Res Neuroscience 109(4):827–835
81(3):314–321 12. Nörenberg W, Illes P, Gebicke-Haerter PJ
3. Eder C (2010) Ion channels in monocytes and (1994) Sodium channel in isolated human
microglia/brain macrophages: promising ther- brain macrophages (microglia). Glia 10(3):
apeutic targets for neurological diseases. J 165–172
Neuroimmunol 224(1–2):51–55 13. Black JA, Waxman SG (2011) Sodium chan-
4. Kettenmann H, Hanisch U-K, Noda M et al nels and microglial function. Exp Neurol
(2011) Physiology of microglia. Physiol Rev 234(2):302–315
91(2):461–553 14. Cherny VV, Markin VS, DeCoursey T (1995)
5. Hille B (2001) Ion channels of excitable mem- The voltage-activated hydrogen ion conduc-
branes, 3rd edn. Sinauer Associates, tance in rat alveolar epithelial cells is deter-
Sunderland, MA mined by the pH gradient. J Gen Physiol
6. Eder C, Klee R, Heinemann U (1997) 105(6):861–896
Pharmacological properties of Ca2+-activated 15. DeCoursey TE (2003) Voltage-gated proton
K+ currents of ramified murine brain macro- channels and other proton transfer pathways.
phages. Naunyn Schmiedebergs Arch Physiol Rev 83(2):475–579
Pharmacol 356(2):233–239 16. Schilling T, Eder C (2009) Non-selective cat-
7. Schilling T, Eder C (2007) Ion channel expres- ion channel activity is required for
sion in resting and activated microglia of hip- lysophosphatidylcholine-induced monocyte
pocampal slices from juvenile mice. Brain Res migration. J Cell Physiol 221(2):325–334
1186:21–28 17. Schilling T, Eder C (2009) Importance of the
8. Stoppini L, Buchs P-A, Muller D (1991) A sim- non-selective cation channel TRPV1 for
ple method for organotypic cultures of nervous microglial reactive oxygen species generation.
tissue. J Neurosci Methods 37(2):173–182 J Neuroimmunol 216(1–2):118–121
9. Schilling T, Eder C (2007) TRAM-34 inhibits 18. Konno M, Shirakawa H, Iida S et al (2012)
nonselective cation channels. Pflugers Arch Stimulation of transient receptor potential
454(4):559–563 vanilloid 4 channel suppresses abnormal acti-
10. Liu B-S, Ferreira R, Lively S et al (2013) vation of microglia induced by lipopolysaccha-
Microglial SK3 and SK4 currents and activation ride. Glia 60(5):761–770
state are modulated by the neuroprotective 19. Kraft R, Grimm C, Grosse K et al (2004)
drug, riluzole. J Neuroimmune Pharmacol Hydrogen peroxide and ADP-ribose induce
8(1):227–237 TRPM2-mediated calcium influx and cation
currents in microglia. Am J Physiol Cell 21. Jiang X, Newell EW, Schlichter LC (2003)
Physiol 286(1):C129–C137 Regulation of a TRPM7-like current in rat
20. Beck A, Penner R, Fleig A (2008) brain microglia. J Biol Chem 278(44):
Lipopolysaccharide-induced down-regulation 42867–42876
of Ca2+ release-activated Ca2+ currents (ICRAC) 22. Nörenberg W, Cordes A, Blöhbaum G et al
but not Ca2+-activated TRPM4-like currents (1997) Coexistence of purino- and pyrimidi-
(ICAN) in cultured mouse microglial cells. noceptors on activated rat microglial cells. Br J
J Physiol 586(2):427–439 Pharmacol 121(6):1087–1098
Part VI
Analysis of Microglia Polarization

Chapter 18
Studying M1 and M2 States in Adult Microglia

Sadanand M. Gaikwad and Michael T. Heneka
Abstract
Microglial cell function receives increasing interest. To date, the majority of experiments are performed by
using immortalized microglia-like cells or primary microglia prepared from pre- or postnatal rodent brain.
As those may not adequately reflect the microglial biology in the adult brain, this protocol advocates a
procedure which allows for the isolation, purification, and subsequent analysis of microglial cells. Once
isolated, the principal state of activation, M1 or M2, can be determined in adult microglia using
fluorescence-activated cell sorting, quantitative PCR, and/or Western blotting. Likewise, adult microglia
generated by this protocol can be used for functional analysis through cell cultivation for a limited time.
Key words Adult microglia, M1/M2 phenotype, Neurodegeneration, Cytokine, Inflammatory gene,
Alzheimer, Neuroinflammation
1 Introduction
The notion that the innate immune system is widely implicated in

a variety of physiological and pathological processes in the brain,
including synaptic pruning, debris clearance by phagocytosis, and
trophic factor supply to surrounding bystander cells, has instigated
research on microglial cells, the major cell type representing the
brain’s innate immune defense in recent years. Until now, however,
the majority of studies on microglia have been performed using
immortalized microglia-like cells and primary microglia derived
from pre- or postnatal rat or mouse brain preparations. These cells,
however, may only in part reflect the biology of microglial cells as
they may lack substantial information which may only be acquired
over time, once brain development has been finalized and the mor-
phology and function of microglia has been shaped by neurotrans-
mitters, endocrine factors, and their specific neuronal and astroglial
environment. Based on this, one can assume that the experimental
results obtained from in vitro studies using immortalized microglia-
like cells or primary microglia should be further backed up by the
analysis of adult microglia. Since not all of these studies can be
185
186 Sadanand M. Gaikwad and Michael T. Heneka
done in the living animal, e.g., by the use of two-photon in vivo

laser scanning microscopy of microglia-expressing fluorophores of
interest, protocols are required which allow for the analysis of adult
microglia in the context of a specific physiological state or disease.
Microglial activation in response to physiological or pathophys-
iological stimuli will determine their morphology and function, and
thus microglia can be polarized into various directions of which
two, the M1 and M2 phenotypes, just have begun to be investi-
gated and described. It is therefore likely that with an increasing
knowledge about microglia functions in the brain, further activa-
tion states will be discovered. The M1 phenotype usually refers to
microglial cells stimulated by interferon gamma and/or tumor
necrosis factor alpha which expresses proinflammatory cytokines,
while M2 cells produce anti-inflammatory cytokines, including
interleukin 10 and interleukin 4, and are thought to dampen the
immune response and instead promote tissue repair, similar to the
concept of TH1/TH2 T cell polarization [1].
Neurodegenerative processes inevitably lead to the activation
of microglial cells, which may acutely serve a beneficial purpose,
but when chronically activated further promote and exacerbate
neurodegeneration [2]. Using the example of Alzheimer’s disease,
we below describe a procedure how adult microglia can be isolated
from a related transgenic mouse model and then studied to delineate
their precise role in this devastating neurological disorder. A brief
description of the mouse strains analyzed is followed by a listing of
all materials and tools needed for the subsequent steps. In brief,
the described methodology includes the murine anesthesia, surgery,
and intracardial perfusion which is performed to clear the brain’s
vasculature from circulating immune cells and immune mediators.
Thereafter the gentle dissociation of the brain is described, followed
by centrifugation and separation procedures to clear the microglia
from myelin and debris. Then the protocol pictures fluorescence-
activated cell sorting (FACS), which through the use of adequate
markers allows for the identification and isolation of adult microglia
from the brain. Given the technical possibilities modern FACS
provides, subsequent steps of analysis may include the purification
of RNA and analysis of the transcriptome by qPCR or similar meth-
ods. Alternatively and depending on the amount of cells, these
methods may also be used for the analysis of microglia by Western
blot, immunocytochemistry, and others. A further option is to use
the isolated cells and to cultivate those further. This procedure
allows to perform a functional analysis, e.g., using assays which
allow the assessment of migration or phagocytosis (not described).
Importantly, the described methods of microglial isolation and
analysis can be adopted for other models of neurodegenerative
disease including but not restricted to the analysis of microglia in
the substantia nigra of MPTP-treated rodents or for models of
experimental allergic encephalitis.
Studying M1 and M2 States in Adult Microglia 187
2 Materials
2.1 Animals 1. APP/PS1 transgenic animals (# 005864, The Jackson

and Ages Laboratory, [3]) and wild-type (WT) animals (The Jackson
Laboratory) both were from C57/Bl6 genetic background.
2. Animal ages during experiments were 6–10 months.
3. Mice were housed under standard conditions at 22 °C, a 12h
light–dark cycle with free access to food and water.
4. Animal care and handling was performed according to the
declaration of Helsinki and approved by the local ethical
committee.
5. The following animal groups were analyzed: WT and APP/PS1.
2.2 Intracardial Tools needed:

Perfusion of Mice
1. Blunt-ended scissors, with approximately 2.5- to 3-in. blade
length.
2. Blunt-ended forceps, approximately 6 in., iridectomy scissors,
hemostats [2].
3. Metal spatula, scalpel, a small, single-use needle (6) (see Note 1).
4. Pointed scissors, with approximately 1-in. blade length.
5. Container for the mouse with a corked surface (see Note 1).
Solutions:
1. Ice-cold 1× PBS including approximately 10 U/ml heparin
(a 25 kU vial will make 2.5 L of PBS+Hep) (see Note 2).
2. 70 % Ethanol and dH2O.
Perfusion setup:
1. Perfusion pump (which maintains a flow no higher than
0.5 ml/min for up to 3–5 min) (see Note 3).
2. Extension tubing 100 cm.
3. Butterfly needles, 23G, with 12-in. tubing.
Anesthesia:
1. Isoflurane (2-chloro-2-(difluoromethoxy)-1,1,1-trifluoroethane)
(see Note 4).
2.3 Adult Mouse 1. Medium B: 45 ml of 1× Hank’s buffered salt solution (HBSS)

Brain Microglia and 5 ml of fetal calf serum (FCS) 10 % mixed by vortex and
Isolation Collagenase Type IV (0.2 mg/ml, working activity of 770 U/mg)
(see Note 5).
2. To inactivate the enzymatic reaction, 20 ml of 20 % fetal calf
serum (FCS) in 1× HBSS.
3. 100 % Percoll, mixture of 100 % Percoll and 10× HBSS was

prepared in ratio of 9:1 (45 ml of 100 % Percoll and 5 ml of
ice-cold 10× HBSS).
4. 75 % Percoll, for one brain 7.5 ml of 100 % was mixed with
2.5 ml of ice-cold PBS (1×).
5. 25 % Percoll, for one brain 2.5 ml of 100 % Percoll was mixed
with 7.5 ml of ice-cold PBS (1×) (see Note 6).
6. 18G, 23G needles with 20 ml syringes, cell strainer with 70 μm
nylon mesh (see Note 7).
1. CD16/32 antibody (Fc receptor blocking antibody).

2.4 Cell Staining 2. Anti-mouse CD11b- and CD45-conjugated antibody (see Note 8).
and Sorting
3. Eppendorf tubes (1.5 ml) and microcentrifuge tubes—
centrifuge.
4. FACS tubes.
5. Cytometry and cell sorting facility (see Note 9).
2.5 Real-Time PCR 1. mRNA Isolation Kit.

Components 2. cDNA Preparation Kit.
3. PCR primers or fluorescent probe and real-time PCR machine
(see Note 10).
2.6 Immunoblotting Prepare all solutions using ultrapure water (prepared by purifying
deionized water to attain a sensitivity of 18 MΩ cm at 25 °C) and
analytical grade reagents. Prepare and store all reagents at room
temperature (unless otherwise indicated). Diligently follow all
waste disposal regulations when disposing waste materials. We do
not add sodium azide to the reagents.
2.7 Solutions 1. To prepare 1× phosphate-buffered saline (PBS), add 8 g

and Reagents (137 mM) of NaCl, 0.2 g (2.7 mM) of KCl, 1.44 g (10 mM)
of Na2HPO4, and 0.24 g (10 mM) of KH2PO4 to 800 ml of
dH2O. Then adjust the pH to 7.4 with HCl (see Note 16) and
adjust the volume to 1liter with the addition of dH2O and
autoclave to sterilize.
2. To prepare 1 ml volume of 2× SDS lysis buffer: take 200 μl
(100 mM) of Tris–HCl prepared in pH 6.8 then 200 μl (4 %) of
SDS, 100 μl (10 %) of glycerol, and 20 μl (2 %) of 2-mercap-
toEtOH mixed and store at 4 °C until use (see Note 11).
3. Just before use add protease inhibitors to the lysis buffer 3 μl
Pic and 3 μl (200 mM of) PMS (see Note 12).
4. To prepare 10 % APS, 1 g ammonium persulfate is mixed with
10 ml dH2O. Mix well.
5. 1× transfer buffer was prepared by adding 3.03 g of Tris–HCl,

14.4 g of glycine, 200 ml of methanol, and 1 ml of SDS (10 %)
(see Note 13).
6. To prepare 10× PAGE buffer, 144 g of glycine and 30 g of Tris
base are filled to 1 L with dH2O and mixed.
7. Running buffer: 100 ml of 10× SDS-PAGE buffers is mixed
with 5 ml of 20 % SDS and filled up to 1 L with dH2O and
mixed (see Note 14).
8. Blotting buffer is prepared by mixing 100 ml of 10× PAGE
buffer and 200 ml of methanol and filling up to 1 L with dH2O
and mixed.
9. Tank buffer 10× is prepared by mixing 15 g of Tris–HCl, 72 g
glycine, and 50 ml of SDS (10 %) and filling up to by adding
500 ml of dH2O.
10. Wash buffer: Use 1× TBST is prepared by adding 1 L of TBS
1× and 1 ml of Tween20.
11. 100 ml of 5 % nonfat dry milk is prepared in 1× TBST by add-
ing 5 g of milk powder (5 %), 10 ml of 10× TBS (20 mM Tris–
HCl pH8, 100 mM NaCl), and 50 μl Tween20 (0.05 %).
12. Blocking buffer: Use 1× TBST with 5 % nonfat dry milk; for
150 ml, add 7.5 g nonfat dry milk to 150 ml 1× TBST and mix
well (see Note 15), then store at 4 °C.
13. Bovine serum albumin (BSA).
14. Primary antibody dilution buffer: Use 1× TBST with 5 % BSA
or 5 % nonfat dry milk as indicated on the primary antibody
datasheet; for 20 ml, add 1.0 g BSA or nonfat dry milk to
20 ml 1× TBST and mix well (see Note 15) store at 4 °C.
15. HRP Western Blot Detection System: Secondary antibody
conjugated to horseradish peroxidase (HRP), Anti-Biotin HRP-
linked antibody, chemiluminescent reagent, and peroxide.
16. Prestained Protein Marker, Broad Range (premixed format).
17. Blotting membrane: This protocol has been optimized for
nitrocellulose membranes (recommended). PVDF membranes
may also be used. A pore size of 0.2 μm is generally
recommended.
3 Methods
3.1 Intracardial 1. Place tubing into ice-cold 1× PBS buffer. Connect the exten-
Perfusion and Surgery sion tube through the perfusion pump and the other end with
butterfly needles, 23G, with 12-in. tubing.
2. Turn on the pump to flow PBS through the extension tube
to wash the extension tube and butterfly needles along with
tubing. Make sure that there are no bubbles in any of the tubes
(see Note 3).
3. Place the mouse in a container that has a small amount of
isoflurane (or equivalent) (see Note 4).
4. Once the mouse no longer shows a response to the painful
stimulus, wet the abdomen with 70 % ethanol. Place the mouse
on the corked surface with the abdomen facing up, fixated
using small needles (see Note 1).
5. Grab the skin with a blunt-ended forceps at the level of the
diaphragm, and cut to expose the liver. Cut laterally and then
up, cutting through the ribs. Lift flap and continue cutting
until the heart is easy to access. Pin the loose flap of the skin
out of the way and free the heart by tearing any connective
tissue with the forceps (see Note 16).
6. Secure the chest flap to the corked surface with a 20G needle
(see Note 17).
7. Place the butterfly needle into the left ventricle, and turn
on the perfusion pump to flow no faster than 0.5 ml/min of
ice-cold 1× PBS BUFFER. Then immediately cut the right
atrium (see Note 18).
8. Continue the buffer flow until the liver has become a light coffee
color. This should be noticeable within 4–5 min.
9. Turn off pump. If perfusing more mice, lift the tube out of buffer
for a couple of seconds and switch the pump-to-pump buffer.
Allow the pump to flow until any air bubbles have been expelled.
This ensures that only buffer is in the line. Turn off pump.
10. Decapitate mouse at the level of the shoulders. Cut through
the scalp with a razor blade to expose the skull. Using a small
sharp-ended scissors and a scalpel, scrape off the skin and
muscle at the back of the brain.
11. Carefully expose the brain by scraping through the skull bone
(see Note 19). After most of the skull bone is removed, with
the help of metal spatula gently take out the brain and place it
into ice-cold 1×HBSS medium.
3.2 Adult Microglia 1. Remove the mouse brain and determine the weight of tissue in
Isolation 1 ml of ice-cold 1× HBSS-contained 35 mm dish to make sure
the 0.4 g limit per brain is not exceeded.
2. With the help of a new scalpel, separate the two hemispheres,
most of the visible meninges must be carefully removed under
a microscope.
3. Transfer the brain of one animal into one small Petri dish con-
taining 5 ml of Medium B (enzymatic solution) and finely chop
the brain with a scalpel and incubate it for 90 min at 37 °C, 5 %
CO2, and continuously stir (Fig. 1) (see Note 20).
Mechanical dissociation
Medium B
Adult Mouse Brain

chopped into small
paces
Myelin /debris separation
1x PBS
Myelin debris
Microglia
25% Percoll at the interface
75% Percoll
Cell Pellet
Microglia Purification
CD11b Positive
Microglia
Sorting
Cell Culture
Genomics
Functional
Proteomics
assays
Fig. 1 Protocol steps to isolate microglia from an adult mouse brain are outlined
and illustrated in this schematic procedure workflow
4. After this incubation transfer the tissue to a 50 ml falcon tube

and inactivate enzymatic reaction using 20 ml of 20 % fetal calf
serum (FCS) in 1× HBSS, and centrifuge at 200 × g for 10 min
at 4 °C.
5. Gently mix the tissue suspension with a 20 ml syringe and 18G
needle 10–12 times and then repeat it with a 23G needle
(see Note 21). Remove all clumps and make single-cell suspen-
sions, filtering them through the cell strainer with 70 μm nylon
mesh (see Note 7) into a new 50 ml conical tube. Centrifuge
at 155 × g for 10 min at 4 °C.
6. A discontinuous gradient density centrifugation step using
Percoll is then performed: Resuspend the pellet with 1 ml of
75 % Percoll and then add the remaining 9 ml of 75 % Percoll.
7. Very carefully add 10 ml of 25 % Percoll by using a serological
5 ml pipette, avoid mixing the phases! On top of this very
slowly add 10 ml pure 1× PBS and centrifuge at 600 × g for
25 min at room temperature with slow acceleration and no

brake (see Note 22).
8. Gently discard the material at the interphase between 25 %
Percoll and 1× PBS, which contains myelin and cell debris
(Fig. 1).
9. Next, gently collect the lightly visible interphase ring between
75 % and 25 %, which contains a mixed glial cell population, into
a new 50 ml flacon tube (around 10 ml from the ring) (Fig. 1).
10. Add threefold ice-cold 1× PBS (around 30 ml) to the same
50 ml flacon tube.
11. Perform a final centrifugation at 200 × g for 10 min at 4 °C
with slow brake.
12. Gently discard the supernatant, keeping in mind that approxi-
mately 100 μl may remain in the falcon tube. Add additional
ice-cold 900 μl 1×HBSS and count the cell number for further
surface marker staining.
3.3 CD11b Positive FACS staining protocol: Dual-color method used APC and PE
Microglia Staining and conjugates.
Sorting
1. After the cell count, distribute equal numbers of the cells into
the appropriate numbers of 1.5 ml Eppendorf tubes (e.g.,
CD11b, CD45, CD11b+CD45, unstained) (see Note 23).
2. Centrifuge at 9,391 × g for 1 min and discard the HBSS-
contained supernatant.
3. Gently resuspend the cell pellet in 500 μl of 1× HBSS buffer.
4. Centrifuge at 9,391 × g for 1 min and discard the 1× HBSS
buffer.
5. Gently resuspend the cell pellet in 200 μl of 1× HBSS buffer
and add 2 μl of CD16/32 Fc receptor blocking antibody
(~0.5–1 μg/100 μl) and incubate for 20 min on ice.
6. Add 1 μl of fluorochrome-conjugated primary antibodies
respectively to each tube except to the unstained control
(~0.5–1 μg/100 μl) and incubate for 30 min on ice (see Note 8).
7. Add 400 μl of 1× HBSS buffer (just to avoid stress during cen-
trifugation), centrifuge at 9,391 × g for 1 min, and discard the
1× HBSS buffer.
8. Wash the cells one more time by gently resuspending the cell
pellet in 500 μl of 1× HBSS buffer and centrifuging at 9,391 × g
for 1 min to completely remove unbound fluorochrome-con-
jugated antibodies.
9. Finally, gently resuspend the cell pellet in 500 μl of 1× HBSS
buffer and filter through 70 μm nylon mesh into the FACS
tubes. CD11b-high and CD45-low positive cells are sorted
and collected by a flow cytometric cell sorter like the BD
FACSAria III (Fig. 1).
Table 1
List of the M1 and M2 gene markers
Microglia phenotype Gene target

Neutral Iba1
C1qa
Slc13a3
M1 iNOS
IL-1, IL-6
TNFα, Mcp-1
Cxcl-2
Tspo
M2 IL-10
Ym1
Arg-I
Fizz
TGFβ
Lgals3
3.4 Real-Time PCR After sorting CD11b-high and CD45-low positive microglia cell,
populations are further used for transcriptional quantification of
inflammatory gene markers for M1 and M2. (Selected M1 and M2
markers are listed in Table 1.)
mRNA isolation and cDNA preparation are done according
to respective datasheet (see Note 24). The primers’ list is given
in Table 2 (we use Taqman probe for further gene quantification)
(see Note 7).
3.5 Immunoblotting After sorting CD11b-high and CD45-low positive microglia cell,
populations are further used for quantification of inflammatory
protein markers for M1 and M2 translational regulation. (Selected
M1 and M2 protein markers are listed in Table 3.)
Preparation of Cell Lysates and Protein Blotting
1. Sorted microglial cells are washed once with 1× PBS. Cell lysis
is conducted by adding 2× SDS lysis buffer (at approximately
2 × 106 to 1 × 107 cells per ml).
2. Transfer extract to a microcentrifuge tube (1.5 ml Epi) and
heat to 100 °C, heating at this temperature for 5 min, then
sonicate with three to four bursts of 5–10 s each. Next cool it
on ice.
3. Microcentrifuge the sample for 5 min at full speed (12,000 rpm).
4. Measure the protein concentration from each sample by a
standard available method (see Note 25).
5. Load the same concentration of each sample (around 20 μl)
onto SDS-PAGE gel (10 cm × 10 cm) and either 10 μl of
prestained molecular weight markers for electrotransfer
Table 2
List of Taqman probes used for detection gene transcripts
Gene target Taqman probe

Iba1 Mm00479862_g1
C1qa Mm00432142_m1
Slc13a3 Mm00475280_m1
iNOS Mm00440485_m1
IL-1, IL-6 Mm00434228_m1, Mm00446190_1
TNFα, Mcp-1 Mm00443258_m1, Mm00441242_m1
Cxcl-2 Mm00436450_m1
Tspo Mm00437828_m1
IL-10 Mm00439616_m1
Ym1 Mm00657889_mH
Arg-I Mm00475988_m1
Fizz Mm00445109_m1
TGFβ Mm00498255_m1
Lgals3 Mm00802901_m1
Table 3
List of the M1 and M2 protein markers
Phenotype Protein
Neutral Mac-1
F4/80
Isolectin B4
M1 iNOS
IL-6, IL-1
TNFα
MARCO
MHC II
TSPO
M2 CD163
CD206
Ym1
Arginase I
FIZZ
verification or 10 μl of biotinylated protein ladder to determine

molecular weights.
6. Finally, transfer the electrophoresed protein to nitrocellulose
or PVDF membrane by electrotransfer.
Blocking Membrane and Antibody Incubation
Please make sure the sizes of the membranes listed below have a
volume of 10 cm × 10 cm (100 cm2).
1. Blocking membrane: After the electrotransfer of membrane,
wash with 20 ml of 1× TBS for 5 min at room temperature.
2. Add 20 ml of blocking buffer and incubate the membrane for
1 h at room temperature.
3. Wash the membrane once for 5 min with 20 ml of 1× TBST
buffer.
4. Incubate with primary antibody at the appropriate recommended
antibody dilutions. A total 10 ml of primary antibody dilution
buffer should be used at 4 °C overnight with gentle agitation.
5. The next day wash the membrane three times for 5 min with
20 ml of TBST buffer.
6. Prepare the HRP-linked antibody in a dilution of 1:1,000 in
20 ml of blocking buffer (or if you have it, Anti-biotin) and
incubate for 1 h at room temperature with gentle agitation.
7. Wash the membrane three times for 5 min with 20 ml of TBST
buffer and proceed to developing the image.
Developing Images
1. Take a new membrane incubation plastic tray and mix the
Chemiluminescent Detection Substrates in a 1:1 ratio. The amount
may vary depending on the size of the membrane, in each case
1 ml (for small membranes) or 2 ml (for large membranes).
2. Carefully transfer the membrane onto the plastic tray prepared
in step 1 above (carefully agitate the tray to and fro).
3. Now cautiously use a blunt-ended forceps to remove the
membrane from the tray and place on the white board in the
documentation system device set.
4. Put paraffin paper over the membrane; use a small roller to
remove air bubbles.
5. Switch on the documentation unit on the Western blot.
6. Turn on the Epi white light to visualize the membrane and
adjust the membrane to the proper orientation.
7. Turn off the Epi white light and adjust the illumination intensity
and the duration of membrane exposure.
8. Expose to film, develop the image, and save it for further use.
4 Notes
1. The needle number is not specified. We use a 23G needle to

fix the mouse on corked surface and secure the four paws to
surface spreading them as wide as possible.
2. The use of heparin is not a must, but you need to be quick to
perfuse if you do not want to use it.
3. Not all laboratories have a perfusion pump facility. Instead, at
the one end of the extension tube, attach a 50 ml syringe filled
with 1×PBS and at the other end attach the butterfly needle
(23G) with a 12-in. tubing. Make sure that there are no bubbles
in any of the tubes.
4. This is only to relax the mouse enough such that the mouse
can easily die. If kept too long in the container, pinch the toes
to judge its level of response to painful stimulus.
5. Other enzymes have been used for this protocol but we have
found that these enzymes yield a good end product. Prepare
freshly prior to the start of the experiment.
6. Prepared 75 % and 25 % Percoll can be stored on ice until use.
7. We used Taqman probe and based a gene expression assay,
but one could order real-time PCR primers and study the
expression pattern.
8. This preparation generates huge amounts of debris which
cannot be removed by gradient centrifugation. Cell suspen-
sions can be improved using 70 μm nylon mesh, or mesh of an
even finer grade, as necessary.
9. Conjugated antibodies need to be selected carefully since the
selected fluorochrome peak excitation and emission wave-
lengths should not overlap each other in double-stain FACS
analysis.
10. As the cell sorting facility may be inaccessible, one could alter-
natively use the CD11b magnetic cell sorting.
11. SDS precipitates at 4 °C; therefore, the lysis buffer needs to be
gently heated prior to use.
12. Any serine protease inhibitor may be used.
13. Dilute 100 ml of 10× native buffer (0.25 M Tris–HCl and
1.92 M glycine) to 800 ml with dH2O and mix 200 ml of
methanol. Avoid mixing methanol directly into the 10× buffer
since it precipitates its ingredients. In this case the precipitate
can be redissolved by the addition of 800 ml of dH2O.
14. Another simple method to prepare running buffer is to prepare
10× native buffer, measure 30.3 g Tris–HCl and 144 g glycine,
mix well, add makeup to 1 L with dH2O, then add 10 ml of
10 % SDS. Always add SDS solution at last, since it makes
bubbles.
15. Add 100 ml of 10× TBS to a 1 L volume graduated cylinder

and filling it to about 800 ml with dH2O. Transfer 50 g of skim
milk powder into the cylinder and stir until dissolved. Fill to 1 L
with dH2O. Separate 500 ml for the blocking solution. To the
remaining 500 ml solution, add 250 μl of Tween20, dissolve,
and use it as diluents for the antibody.
16. Be sure to draw the scissors away from the organs when cutting
to avoid damaging circulation more than necessary.
17. Do not use scissors to free the heart—this causes undesired bleeding.
18. This process needs to be done quickly before onset of blood
clotting.
19. This needs to be done carefully as it can easily damage the
hemisphere.
20. Some groups have shown that even a 2-h incubation time may
be used.
21. To have single-cell suspensions, better to use first 18G needle
and then 23G needle may improve the preparation and avoid
tissue clumps.
22. This step must be done very slowly. When properly done, one
should clearly see different layers.
23. From one mouse, approximately 700,000 cells can be obtained.
24. Any high-quality mRNA Isolation Kit may be used which
provides high-quality mRNA from sorted cell populations.
Any cDNA Isolation Kit may be used to prepare cDNA from
isolated high-quality mRNA.
25. Any available standard method may be used to quantification
of protein concentration.
Acknowledgement
This work was supported by a grant of the Deutsche

Forschungsgemeinschaft to MTH (KFO177, TP8) and of the
EU-FP7 program on neuroinflammation (INMIND).
References
1. Gordon S (2003) Alternative activation of mac- 3. Jankowsky JL, Slunt HH, Ratovitski T et al
rophages. Nat Rev Immunol 3:23–35 (2001) Co-expression of multiple transgenes in
2. Heneka MT, O’Banion MK, Terwel D et al mouse CNS: a comparison of strategies. Biomol
(2010) Neuroinflammatory processes in Eng 17:157–165
Alzheimer’s disease. J Neural Transm 117:
919–947
Chapter 19
Isolating, Culturing, and Polarizing Primary Human

Adult and Fetal Microglia
Bryce A. Durafourt, Craig S. Moore, Manon Blain, and Jack P. Antel
Abstract
Microglia are an important component of the innate immune system within the central nervous system
(CNS). Isolation and in vitro culturing of microglia can provide insight towards the basic biology of these
cells as well as their interactions with neurons, astrocytes, and oligodendrocytes. While studies of rodent
microglia and microglial cell lines have provided a basis for our understanding of these cells, human adult
microglia exhibit distinct properties when compared to rodent cells. Furthermore, the study of human
fetal microglia provides a window into the developing CNS. This chapter describes the protocols used to
isolate, purify, and culture both human adult and fetal microglia. Under basal culture conditions, human
microglia survive for extended periods in the absence of growth factors, thus allowing their properties to
be investigated under resting conditions. In addition, both human adult and fetal microglia can be used
to study how they respond to different polarization conditions. As is the case with macrophages, it is
also possible to polarize microglia towards the pro-inflammatory “M1” and the anti-inflammatory “M2”
phenotypes, as described in this chapter.
Key words Microglia, Myeloid cells, Human, Isolation, Culture, Polarization, M1, M2
1 Introduction
In the scientific literature, much of our understanding as it pertains

to the biology of microglia has relied on the use of primary rodent-
derived cells and transformed human cell lines. While the use of
these cells/cell lines has led to many discoveries that we predict will
translate and apply to humans, these continue to require validation
in primary human culture systems. Human microglia possess distinct
properties, including their capacity to survive for prolonged periods
under basal culture conditions. Access to both human fetal and adult
central nervous system (CNS) tissue provides the unique and invalu-
able opportunity to isolate primary human neural cells and gain
insight into their basic properties during development, and in the
mature brain under physiologic and pathological conditions.
Microglia represent an important member of the innate immune
199
200 Bryce A. Durafourt et al.
system within the CNS and play critical roles in influencing the
adaptive immune response and biological activities of other CNS-
resident cells (e.g., neurons, astrocytes, oligodendrocytes). In this
chapter, we will discuss in detail the protocols that can be used to
successfully isolate, purify, and culture human microglia from either
fetal or adult human CNS tissue. In addition, we will also provide
protocols that can be used to polarize microglia into either a classi-
cal, pro-inflammatory “M1” or alternative, anti-inflammatory “M2”
phenotype. Polarization of microglia is performed using different
disease-relevant molecules, reflecting the plasticity of these cells in
vivo. In culture, cell polarization allows us to compare and contrast
the molecular mechanisms that may help distinguish microglia from
other myeloid cell types and provide rationale and insight for further
in situ investigation in diseased tissue.
Based on whether CNS tissue is fetal- or adult-derived, in this
chapter we will provide different cell isolation, purification, and cul-
turing protocols. Human fetal samples are obtained at a pre-myelinating
gestational age between 14 and 20 weeks; adult tissue is surgically
derived and most often collected from tissue resections from either
the frontal or temporal lobe of patients undergoing surgery for
intractable epilepsy. The same isolation and culture techniques
have also been applied to autopsy-derived material. The use of
different cell isolation techniques is primarily due to the presence
of myelin in the adult tissue, which must be successfully removed
to prevent cell toxicity and unwanted cellular biology that could
incur as a consequence of the presence of myelin during culture. It
should be noted that compared to surgically resected material,
CNS tissue derived at autopsy introduces both premorbid (such as
preterminal conditions including ischemia and hypoxia) and post-
mortem (delay to obtaining tissue) variables. We acknowledge that
variables in working with human tissue exceed those encountered
using tissue obtained from rodents or other animals raised under
strict laboratory conditions. Use of surgical tissue for research pur-
poses must be coordinated with examination of tissue for clinical
diagnostic purposes.
2 Materials
2.1 General Material 1. Gloves, lab coat.

2. Incubator (37 °C, 5 % CO2).
3. Shaking water bath (37 °C).
4. Tissue culture hood.
5. Vacuum aspirator apparatus.
6. Swinging bucket centrifuge.
7. Disposal beaker containing ~50 ml bleach.
Isolating, Culturing, and Polarizing Human Microglia 201
8. Parafilm.
9. Pipette aid and 5, 10, and 25 ml serological pipettes.
10. 10 μl and 200 μl micropipettes and tips.
11. Autoclaved 100 ml and 500 ml bottles.
12. Mashing filter (see Note 6 on preparing the filter).
13. Plunger of 20 ml syringe.
15. DNase.
16. Trypsin.
17. 50 ml conical tubes.
18. Fetal calf serum (FCS).
19. Poly-L-lysine (optional).
20. Tissue-coated culture flasks (T12.5, T25, T75, T175) and plates
(24-well plates, 12-well plates, or 6-well plates as required).
21. Hemocytometer.
2.2 Adult Prep 1. Percoll®.

2. Minimum Essential Medium (MEM) containing 5 % FCS, 0.1 %
glucose, 1 % penicillin–streptomycin, and 1 % glutamine.
3. Plastic 3 ml transfer pipette.
4. Pathology cassette and formalin.
5. Glass Pasteur pipettes and bulb.
6. 40 ml Nalgene centrifuge tubes.
7. Fixed rotor centrifuge.
8. Ethylenediaminetetraacetic acid (EDTA).
2.3 Fetal Prep 1. Petri dish (150 × 15 mm).

2. Scalpels.
3. Sterile forceps.
4. Dulbecco’s Modified Eagle Medium (DMEM) containing
5 % FCS, 0.1 % glucose, 1 % penicillin-streptomycin, and 1 %
glutamine.
5. 70 μm cell strainer.
2.4 Polarization 1. Macrophage colony-stimulating factor (M-CSF).

2. Granulocyte-macrophage colony-stimulating factor (GM-CSF).
3. Interleukin-4 (IL-4).
4. Interleukin-13 (IL-13).
5. Interferon gamma (IFN-γ).
6. Lipopolysaccharide (LPS; serotype 0127:B8).
3 Methods
These protocols must be carried out in a sterile tissue culture hood,

using autoclaved solutions and autoclaved reusable or sterile dis-
posable equipment. Universal precautions must be followed when
processing human tissue; personnel must wear lab coats and gloves.
Special care must also be taken especially when handling sharp
instruments. Any reusable equipment must be left in a 10 % bleach
solution for at least 24 h prior to being washed. Unless otherwise
stated, all centrifugation steps should be carried out at room tem-
perature, in a swinging bucket rotor, with brakes set to “high.”
3.1 Isolation of 1. Human brain tissue (see Note 1) is obtained in a Cavitron

Human Adult Microglia Ultrasonic Surgical Aspirator (CUSA®) bag, in the form of
pieces less than 1 mm3. Following resection from the CNS,
process tissue as quickly as possible (ideally within 1 h).
2. Before starting the procedure, turn on a shaking water bath
to 37 °C.
3. Carefully open the CUSA ® bag in the tissue culture hood
and pour contents into 50 ml conical tubes. A maximum of 14
tubes should be used; if there is still liquid in the CUSA® bag,
keep the bag for use in the next step.
4. Allow the tissue to settle at the bottom of the tubes, and then
pour out ¾ of the liquid. Pour any remaining CUSA® bag con-
tents into the conical tubes. Repeat until the CUSA® bag is
empty, at which point the CUSA® bag should be rinsed with
50–100 ml PBS, taking care not to leave any tissue sticking to
the inside of the CUSA® bag.
5. Allow the tissue to settle and pour off ¾ of the liquid as in Step 4
above, then add PBS in order to wash and remove as much
blood as possible. Repeat this step until the liquid becomes
almost clear, usually after two to three washes.
6. Pool the tissue to one tube by first allowing tissue to settle and
pouring off ¾ of the liquid as above. Once it has settled, note
the total amount of tissue (see Note 2).
7. Use a transfer pipette to remove 1 ml of tissue and place it into
a pathology tissue cassette. Place the cassette into a specimen
container containing formalin and return this specimen to the
pathology department (see Note 3).
8. Remove as many blood clots as possible using a Pasteur pipette
(see Note 4).
9. Pour the cleaned tissue into a 100 ml bottle and rinse the conical
tube with ~5 ml PBS.
10. Add 5 ml 0.5 % trypsin and 5 ml DNase (25 μg/ml). If total
tissue volume is >15 ml, add an additional 5 ml each of trypsin
and DNase.
11. Add PBS to a final volume of 68 ml.

12. Tightly cap the bottle, and wrap the mouth of the bottle and
cap with parafilm (see Note 5) and shake at 180 rpm, 37 °C for
30 min.
13. During the shaking period, set up the required materials for
the following steps. Set up mashing filter (see Note 6) in the
neck of a 500 ml bottle. Remove the plunger of a 20 ml syringe
and place on top of the mashing filter.
14. Following the 30 min shaking incubation, add 10 ml of FCS to
inactivate the trypsin.
15. Determine the maximum volume of PBS to be used for the
mashing steps (see Note 7).
16. Pipette 5–10 ml of tissue into the filter, add PBS, and mash
using the syringe plunger. When adding PBS, do not exceed a
maximum volume of 50 ml per tube calculated above. If the
prep is very large and this volume exceeds 500 ml, move the
filter setup to a new 500 ml bottle when the first bottle is full.
17. Continue adding 5–10 ml of tissue into the filter and mashing
with PBS (see Note 8).
18. Once all tissue has been mashed, wash the bottle that con-
tained the tissue with 20 ml of PBS, ensuring all pieces of
tissue have been collected and added to the filter. Continue
mashing with PBS until tissue no longer passes through the
filter. It is normal that some clumps of myelin debris may not
pass through the filter.
19. Divide the contents of the bottle equally into the number of
tubes determined in Step 15 above. Fill to 50 ml with PBS and
spin 300 × g, 10 min.
20. During this spin, add 9 ml of Percoll® to each sterile Nalgene
centrifuge tube. Prepare a 50 ml conical tube with PBS for use
in the next step.
21. Pour out the entire supernatant of each tube and pour in
~20 ml PBS. Do not add >21 ml PBS as this is the volume used
in the next step. Adding <21 ml at this step ensures that no
tissue is wasted.
22. Resuspend the tissue by pipetting up and down. Process one
conical tube at a time. Transfer first 10 ml from the conical
tube into a Percoll® tube. Next, transfer 11 ml, using the tube
prepared with PBS to supplement to the required volume.
23. Weigh the centrifuge tubes in order to match for balance. The
maximum difference in weights between paired tubes should
not exceed 0.05 g. Add PBS if required to bring weights within
acceptable range (adding 10 μl for each 0.01 g difference).
24. Spin 31,000 × g, 4 °C, 30 min, with no brakes, in a fixed rotor
centrifuge.
25. During this spin, set up an aspirator with 10 ml of bleach, and

prepare 50 ml conical tubes to match the number of Nalgene
tubes used above.
26. Carefully remove the Nalgene tubes from the centrifuge. The top
layer contains myelin, the middle layer contains glial cells, and
the bottom layer contains red blood cells.
27. Using a vacuum aspirator, remove the top layer of myelin
(see Note 9).
28. Set a pipette aid to “slow” and transfer the cell layer (~15–18 ml)
to a 50 ml conical tube (see Note 10).
29. Fill each conical tube with PBS to a final volume of 50 ml.
Invert tubes to mix.
30. Spin 900 × g, 10 min.
31. During this spin, warm MEM containing 5 % FCS, 0.1 % glu-
cose, 1 % penicillin-streptomycin, and 1 % glutamine to 37 °C.
32. Carefully pour out the supernatant as much as possible.
33. Add 7 ml of medium to one tube, resuspend the pellet, and
transfer to the next tube. Continue this process until all pellets
within each tube have been resuspended.
34. Use 10 ml of medium to wash the empty tubes, ensuring all
cells are collected in the final tube.
35. Spin 300 × g, 10 min, using low brakes.
36. Resuspend with 2 ml MEM 5 % FCS and remove a small
aliquot (10 μl) to count.
37. Plate at 2 × 106 cells per ml in a T12.5 plastic flask (max 3 ml)
or 6 ml per T25 flask if the prep is larger (i.e., >6 × 106 cells).
38. Allow the microglia 24 h to adhere to the flask. Floating cells
(including oligodendrocytes and progenitors) should be
removed at this point (see Note 11). Wash flask 2× with 3 ml
medium (MEM 5 % FCS). Do not touch the pipette to the
surface of the flask to avoid disturbing adherent microglia.
39. Add fresh MEM 5 % FCS (3 ml for T12.5 flask, 6 ml for T25
flask).
40. Allow microglia to recover from the isolation process for
3–5 days prior to trypsinization and plating.
41. Set up a vacuum aspirator, and prepare a 50 ml conical tube
containing 5 ml of warm medium (MEM 5 % FCS).
42. Prepare 10 ml trypsin solution (0.05 % trypsin, 2 mM EDTA).
43. Aspirate medium and rinse with 3–5 ml (3 ml for T12.5, 5 ml
for T25) of PBS.
44. Aspirate the PBS and replace with 3–5 ml (3 ml for T12.5,
5 ml for T25) of trypsin solution.
Fig. 1 Human adult microglia (25×). Cells were isolated according to the described
protocol and grown in MEM containing 5 % FCS, 0.1 % glucose, 1 % penicillin–
streptomycin, and 1 % glutamine. In culture, adult microglia cells adopt a bipolar
morphology and exhibit numerous vesicles visible under light microscopy
45. After 10 s, aspirate the trypsin, leaving only a thin film of liquid
covering the surface of the flask.
46. Place flask in the incubator for 5 min or until cells have lifted
(see Note 12).
47. Hit the flask three times on a counter surface (quickly and
forcefully) to dislodge the cells. Verify under a microscope that
the cells have been dislodged.
48. Quickly return to the hood, pipette 3–5 ml medium into the
flask (expel liquid to further detach cells), and transfer cells to
the tube prepared with medium. Rinse the flask 2× with
medium, transferring each time to the tube.
49. Spin 300 × g, 10 min.
50. Resuspend in 1–2 ml of medium and remove a small aliquot
(10 μl) to count.
51. Plate the microglia at desired density, usually 100,000 cells/
ml, in either a 24-well plate (1 ml/well) or a 6-well plate
(3–4 ml/well) (see Note 13). See Fig. 1 for the appearance of
typical human adult microglia after 4–5 days in culture.
3.2 Isolation of 1. Before starting the procedure, warm the shaking water bath to
Human Fetal Microglia 37 °C.
2. Upon receipt of fetal brain tissue (see Note 14), transfer the
tissue to a 50 ml conical tube.
3. Allow the tissue to settle at the bottom of tube, and then pour
off excess liquid. If the liquid contains a significant amount of
blood, rinse two to three times with PBS (each time allowing
the tissue to settle and carefully pouring off ¾ of the liquid).
4. Pour off ¾ of the liquid and transfer the tissue to a 150 × 15 mm
Petri dish.
5. Add 10 ml PBS onto the tissue and remove the meninges
(if present) using sterile forceps.
6. Use the forceps to remove blood clots, taking care to avoid
removing brain tissue.
7. Use two scalpels to crosscut the brain into small pieces of
1–2 mm3.
8. Add 5–10 ml PBS to the minced tissue and transfer with a
pipette into a 100 ml bottle. Add another 10 ml PBS to wash
the Petri dish and transfer to the bottle. Fill with PBS to a final
volume of 40 ml.
9. Add 5 ml 0.5 % trypsin and 10 ml DNase (25 μg/ml). If volume
of tissue exceeds 20 ml, add an additional 5 ml 0.5 % trypsin
and 5 ml DNase (25 μg/ml).
10. Tightly cap the bottle, wrap the mouth of the bottle and cap
with parafilm (see Note 5), and shake at 180 rpm, 37 °C for
15 min.
11. During this incubation step, prepare reagents for the following
steps. Set up mashing filter (see Note 6) in the neck of a 500 ml
bottle. Remove the plunger of a 20 ml syringe and place on
top of the mashing filter.
12. At the end of the incubation, add 5 ml of FCS to inactivate the
trypsin.
13. Transfer the dissociated brain pieces onto the filter by adding 5 ml
of tissue at a time, adding PBS and mashing with the plunger.
14. Once all the tissue has been mashed, wash the bottle with PBS
and transfer onto the filter. Continue to mash and wash with
PBS until the mesh is as clear as possible. It is normal that some
debris (such as blood clots) may be left on the mesh. The final
volume in the bottle should be 150–400 ml, depending on the
size of prep.
15. Distribute the liquid from the bottle into 1 conical tube per
50 ml volume (i.e., three to eight tubes total), and spin at
450 × g, 10 min.
16. During this spin, warm the medium (DMEM containing 5 %
FCS, 0.1 % glucose, 1 % penicillin–streptomycin, 1 % gluta-
mine) to 37 °C.
17. Carefully pour off the supernatant, being careful not to disturb
the pellet; 10–15 ml of liquid should remain.
18. Resuspend the first pellet by pipetting 5 ml of medium.
19. Transfer this first volume to the second pellet, and resuspend.
Continue until all the pellets have been resuspended and the
tissue has been pooled to a single tube. Wash the empty tubes
using another 5 ml of medium and transfer to the tube with
the pooled tissue.
20. Take an aliquot of the tissue (50 μl) to count, noting that this
sample may need to be diluted more than 1:10 if the prep is
large.
21. Spin 300 × g, 10 min.
22. During this spin, count the cells, taking care to exclude the red
blood cells from the count.
23. Pour off the supernatant and resuspend in 20 ml of medium.
If cell suspension contains clumps (i.e., dead cells), the suspen-
sion should be passed through a 70 μM cell strainer.
24. Bring cell suspension to a concentration of 100 × 106 cells/ml
in DMEM 5 % FCS (see Note 15).
25. Plate cells at 6 × 106 cells/ml in 40–50 ml per T175 flask or
15–20 ml per T75 flask (see Note 16).
26. Allow astrocytes and microglia to grow for 5–7 days in the flask.
At this time, if the astrocyte bed is thick and many floating
microglia are visible under the microscope, the microglia can
be harvested (go to Step 31). If the astrocyte layer is not yet
thick and microglia are still adherent, follow the steps below
(27–30) to change the medium in the flask.
27. Pour the medium out of the flask into a 50 ml conical tube,
carefully by inverting the flask, without disturbing the astrocyte
layer.
28. Add 25 ml (T175 flask) or 10 ml (T75 flask) fresh medium to
the flask, by inverting the flask and expelling liquid onto the
surface opposite the cell layer (this avoids disturbing the astro-
cyte bed). Return the flask to the incubator for the duration of
the following spin.
29. Spin the collected medium 300 × g, 10 min.
30. Pour off the supernatant and resuspend in 25 ml (T175 flask)
or 10 ml (T75 flask) fresh medium. Add this medium back to
the flask, again expelling onto the surface of the flask opposite
the cell layer. Return the flask to the incubator for another
5–7 days.
31. When the astrocyte layer is thick and microglia are floating,
collect the microglia by carefully pouring off the liquid from
the flask into a 50 ml conical tube (see Note 17).
32. Spin the collected medium 300 × g, 10 min.
33. Resuspend in 1–2 ml of fresh medium and remove a small aliquot
(10 μl) to count.
Fig. 2 Human fetal microglia (25×). Cells were isolated according to the described
protocol and grown in DMEM containing 5 % FCS, 0.1 % glucose, 1 % penicillin–
streptomycin, and 1 % glutamine. In culture, fetal microglia are mitotic with long,
slender, bipolar processes
34. Plate the microglia at desired density, usually 100,000 cells/ml,

in either a 24-well plate (1 ml/well) or a 6-well plate (3–4 ml/
well) (see Note 13). See Fig. 2 for the appearance of typical
human fetal microglia after 4–5 days in culture.
3.3 Polarization 1. Plated microglia can be left untreated for multiple weeks in
of Human Microglia culture; medium should be changed every 5–7 days. To gener-
ate classically activated (M1) microglia, follow Steps 2 and 3
below; to generate alternatively activated (M2) microglia, fol-
low Steps 4 and 5 below.
2. To generate M1 microglia, treat cells with GM-CSF (5 ng/ml)
for 48 h, then remove half the medium and replace with fresh
medium containing a full dose of GM-CSF (such that the final
concentration of GM-CSF is again 5 ng/ml) for an additional
48–72 h.
3. Activate M1 cells with IFN-γ (20 ng/ml) for 1 h followed by
24–48 h treatment with LPS (serotype 0127:B8, 100 ng/ml).
4. To generate M2 microglia (see Note 18), treat the cells as per
Step 2 above, using M-CSF (25 ng/ml) instead of GM-CSF.
5. Activate M2 cells by treatment with IL-4 (20 ng/ml) and
IL-13 (20 ng/ml) for 24 h. M2 cells must be primed with an
additional dose of IL-4 (20 ng/ml) and IL-13 (20 ng/ml)
(without changing the medium) for 24 h in order to achieve
full M2 polarization.
6. Polarization can be confirmed by flow cytometry assessing the

expression of surface markers associated with the M1 (CD80
and CCR7) and M2 (CD206 and CD209) phenotypes [1].
4 Notes
1. Tissue is obtained as a mixture of white and grey matter of

temporal lobe brain tissue from patients undergoing surgery
for non-tumor-related intractable epilepsy. Tissue is derived
from outside of the suspected focal site. As the material in the
CUSA® bag is in small pieces in PBS, tissue hypoxia likely
occurs later than if tissue is obtained in block format. If tissue
is received in block format, tissue should be manually chopped
into small sections as quickly as possible prior to processing.
2. Recording the volume of tissue is required to determine the
number of centrifuge tubes that will be used in subsequent
steps. Tissue volume is also helpful in approximating cell yield
(see Fig. 3).
3. This specimen is required in case the diagnosis is later ques-
tioned or if a tumor is suspected. In addition, sections can later
be made from this stored tissue and be useful in the event that
cells obtained from this specimen exhibit abnormal properties.
4. This process can be made easier by dividing the tissue into mul-
tiple tubes. Alternatively, the tissue may be spread in a large Petri
dish. The objective is to make the tissue as clean as possible, but
should not exceed 20 min to limit loss of viable cells.
5. Wrapping the bottle cap in parafilm prevents contamination
from splashing of water in the shaking water bath.
Fig. 3 Total cell yield versus starting tissue volume. Average total cell yield
(including microglia and oligodendrocytes) is approximately 1.4 × 106 cells per ml
of starting tissue
6. Prepare the mashing filter using a two-piece Buchner funnel

and a 132 μm nylon mesh. Cut a square piece of nylon mesh
1–2 cm larger than the diameter of the funnel and place tightly
between the two funnel pieces. Tape tightly with autoclave
tape to ensure the mesh does not become detached during
mashing. Autoclave the prepared mashing filter prior to use.
7. At this point, it is appropriate to determine the number of
tubes that will be used for the Percoll® spin. As our fixed rotor
centrifuge can accommodate a maximum of 14 tubes, we first
determine if one or two spins will be necessary. Our guideline
is to use 1–2 ml of tissue per Nalgene tube. For ≤14 ml of
tissue, use 1 tube per ml, rounding down to an even number
of tubes (to facilitate balancing the centrifuge). For 15–20 ml
of tissue, use 14 tubes. For >20 ml of tissue, more than 14
tubes are required for best yield, and thus two spins are neces-
sary. In this case, use 1 tube per ml, rounding down to an even
number of tubes and using a maximum of 28 tubes.
8. Adding too much tissue without mashing or adding PBS can
clog the mesh and make the process slower.
9. Remove as much myelin as possible, especially if also collecting
other cell types, as free myelin is toxic to oligodendrocytes.
10. Care should be taken not to touch the red blood cell layer
below; this process can be facilitated by slightly tilting the tube.
11. Methods of culturing oligodendrocytes from adult human
brain tissue will not be discussed here; see [2].
12. After 5 min, verify under a microscope that some cells are float-
ing. If cells are not yet floating, return flask to incubator for an
additional 1–2 min. Leaving cells in trypsin for too long can be
detrimental to the cells.
13. If using glass chamber slides, surfaces should be treated with
poly-L-lysine prior to plating. Tissue culture-coated plates and
flasks do not require additional coating prior to use.
14. Tissue should be processed within 24 h of dissection; a delay of
36 h or longer may result in reduced cell viability.
15. If other glial cells (progenitors, astrocytes, or neurons) are to
be cultured, an appropriate volume of cell suspension should
be removed at this point; for specific methods, see [3–5].
16. To improve adhesion of cells to the flasks, poly-L-lysine can be
used to coat the flasks prior to plating.
17. The astrocyte layer is extremely fragile at this point, and care
should be taken when handling the flask and when pouring off
the medium. Astrocytes must not be dislodged during this step
as this may result in astrocyte contamination of the microglial
cultures.
18. Alternatively activated microglia generated using IL-4 and

IL-13 are referred to as M2a cells. Other classes of M2 cells
also exist, namely, M2b (generated by activation with immune
complexes and Toll-like receptor ligands) and the “deacti-
vated” M2c (generated using IL-10). The appropriate activa-
tion paradigm should be selected accordingly.
References
1. Durafourt BA, Moore CS, Zammit DA et al tors to growth factors and axon signals.
(2012) Comparison of polarization proper- J Neuropathol Exp Neurol 69(9):930–944
ties of human adult microglia and blood- 4. Darlington PJ, Goldman JS, Cui QL et al (2008)
derived macrophages. Glia 60(5): Widespread immunoreactivity for neuronal
717–727 nuclei in cultured human and rodent astrocytes.
2. Ruffini F, Arbour N, Blain M et al (2004) J Neurochem 104(5):1201–1209
Distinctive properties of human adult brain- 5. D’Souza S, Alinauskas K, McCrea E et al (1995)
derived myelin progenitor cells. Am J Pathol Differential susceptibility of human CNS-derived
165(6):2167–2175 cell populations to TNF-dependent and indepen-
3. Cui QL, Fragoso G, Miron VE et al (2010) dent immune-mediated injury. J Neurosci
Response of human oligodendrocyte progeni- 15(11):7293–7300
Part VII
Co-culture Systems to Analysis Microglia Interactions

with Other Cell Types
Chapter 20
Understanding Microglia–Neuron Cross Talk: Relevance

of the Microglia–Neuron Cocultures
Fernando G. Correa, Miriam Hernangómez, and Carmen Guaza
Abstract
Microglia–neuron interaction is a complex process involving a plethora of ligands and receptors. The outcome
of this intricate process will depend on the prevailing signals (i.e., whether the microglial cells will produce
pro-inflammatory cytokines and/or phagocyte a dying neuron or whether it will produce neurotrophic
factors and support neuronal growth, among other possible scenarios).
In order to study this complex process, several tools have been developed, ranging from in vivo models
(knockout and knock-in mice, conditional transgenic mice, imaging techniques) to in vitro models (microglia–
neuron cocultures, transwell cell cultures). Here we describe a protocol for primary microglia–neuron
coculture. This coculture allows to combine neurons and microglial cells coming from wild-type and
KO mice, making this coculture a useful method to study in vitro the interaction of different sets of
ligand–receptor.
Key words Microglia, Neurons, Cocultures, Innate immunity, Neuroimmunoregulatory molecules,

CD200, CD200R
1 Introduction
Microglia constitute the highly versatile resident macrophages in

the central nervous system (CNS), comprising 5–20 % of the total
glial cell population (Lawson et al.). As reviewed by Yang et al. [1],
in 1919 a disciple of Ramón y Cajal, Pío del Río-Hortega recog-
nized microglial cells as a different cell type from the other glial cells.
He also characterized microglial transformation from a ramified
phenotype into ameboid phagocytic macrophage-like cells in the
stab wounds made in animal brains [2]. From these observations,
del Río-Hortega concluded that microglia originated from periph-
eral mononuclear cells [2].
Interestingly, and after a long debate in which even the existence
of microglial cells was questioned (reviewed in [3]), microglia has
finally been established as a distinct glial cell population from a
myelomonocity origin [4]. Recently, it has been demonstrated that
215
216 Fernando G. Correa et al.
microglial cells are originated from a distinct myeloid precursor

that migrates from the hematopoietic islands in the yolk sac to the
developing brain parenchyma at embryonic day 8.5 [5].
Microglia are widely distributed in the adult CNS, but differ-
ences in their cellular density among different brain areas have been
reported in mice [6, 7] and humans [8]. Thus, microglial cells are
more abundant in the telencephalon than in the diencephalon or
the mesencephalon, with the rhombencephalon containing the
lowest amount of these cells [7] (reviewed in [9]). Not only there is
an uneven distribution of microglia across the cerebral anatomical
regions but between the gray and white matter (the latter contain-
ing higher microglial cell density than the former). Moreover,
cumulative data show that microglial expression of tissue macro-
phage markers varies within the different brain regions as well as
their response to different stimuli [10–12].
In the adult healthy brain, microglia has been described to be
in “resting” or “quiescent” state with a ramified morphology.
However, this term has limitations and can lead to the misleading
idea that microglia are dormant cells awaiting a signal that will
wake them up. On the contrary, recent elegant in vivo imaging
experiments show “resting” microglia as highly dynamic cells, con-
tinuously branching thin processes that survey and sample their
microenvironment [13, 14] in search for potential threats and dan-
ger signals. Thanks to this routine immunosurveillance of the
CNS, microglia will remove apoptotic bodies and other potentially
toxic cellular debris (myelin debris, amyloid deposits, protein
aggregates, etc.) [15–17]. In order to accomplish this task, it is
essential that microglial cells are capable of distinguishing between
“self” and “nonself” signals (Fig. 1). Clearance of pathogens and
toxic cell debris during infection or tissue damage is based on the
recognition of “nonself” and “altered-self” patterns by microglia,
but also astrocytes, oligodendrocytes, and neurons have been
shown to be able to recognize those patterns [18, 19]. There is a
plethora of the so-called “eat-me” signals expressed by pathogens
and apoptotic or necrotic cells. Some of these signals are a hetero-
geneous group of molecules known as pathogen-associated molec-
ular patterns (PAMPs) and are characterized by being highly
conserved through evolution with little antigenic variability [15].
These PAMPs are constituents of the microbial structure which
induce in the host a strong innate immune reaction directed towards
the removal of the pathogen by phagocytosis [20]. The classical
example of a PAMP is the lipopolysaccharide (LPS), component of
the Gram-negative external membrane. Analogous to PAMPs, it
has been proposed that cells undergoing programmed cell death
express de novo apoptotic cell-associated molecular patterns
(ACAMPs) [15, 21, 22]. These ACAMPs would play a key role in
the embryonic process in which whole cell populations need to be
cleared out without mounting an inflammatory response and
Microglia–Neuron Cross Talk 217
Fig. 1 Scheme showing how microglial cells differentiate “self” from “nonself” cues. Adapted from Medzhitov
and Janeway [18]
minimizing the presence of cellular debris [23]. Some ACAMPs

include oxidized low-density proteins, alteration of the membrane
electrical charges, nucleic acids, and phosphatidylserine [19, 24].
Similarly, damaged or stressed tissues release/express the so-called
danger-associated molecular patterns (DAMPs). Some DAMPs
including heat shock proteins (HSP), adenosine, ATP, high-
mobility group box chromosomal protein 1 (HMGB-1), galectins,
and thioredoxin present adjuvant and pro-inflammatory activity
[25]. Phagocytic cells recognize these PAMPs, ACAMPs, and/or
DAMPs, which can be either membrane bound or soluble, through
their pattern recognition receptors (PRRs) [15, 22, 26]. Some of
the PRRs include toll-like receptors (TLR), scavenger and man-
nose receptors, CD14, CD36, complement receptors, phosphati-
dylserine receptor (PSR), and milk fat globulin (MFG-EGF8)
(reviewed in [27]). Therefore, the activation of microglia, rather
than an unspecific process, is highly dependent on the stimulus
that originated it.
In addition to this plethora of signals that might elicit an innate

immune activation directed to the elimination of the pathogen,
the apoptotic cell, or the tissue debris, there is a complex set of
interactions that silence and reshape microglial response [18]. For
instance, electrically active neurons inhibit the interferon-gamma
(IFNγ)-induced increase in MHC class II expression in microglia
[28]. Some neurotransmitters have modulatory effects on microg-
lial response, whereas others, like substance P, enhance the activa-
tion of microglia [29]. Neurotrophins secreted by healthy neurons
such as nerve growth factor (NGF), brain-derived growth factor
(BDNF), and to a lower extent neurotrophin-3 (NT-3) were able
to reverse the induction of MHC class II molecules in microglia
[30]. Interestingly, neurons express membrane-bound molecules
and/or secrete soluble mediators that function as “don’t-eat-me”
signals, reshaping microglial response and inhibiting their phago-
cytic activity on the site of injury [19, 31]. These signals were orig-
inally grouped in a family of heterogeneous molecules called
self-associated molecular patterns (SAMPs). It was suggested that
these SAMPs could interact with novel inhibitory PRRs, negatively
modulating the innate immune response and promoting tissue repair
[19]. These SAMPs signals are present in almost all host “normal-
self” cells but not in pathogens, and they are downregulated in apop-
totic/necrotic cells (altered self). Thus, this “normal-self” identity
can be lost. An example of a molecular marker of “normal self” is the
presence of sialic acid in the terminal of glycoproteins and glycolipids,
which can interact with different microglial receptors [32]. One of
these inhibitory receptors belongs to a family known as siglecs [33].
Siglecs contain immunoreceptor tyrosine-based inhibitory motifs
(ITIMs) in their cytoplasmic tail which downregulate phagocytosis
in macrophages and microglia [33, 34]. Most microorganisms lack
sialic acid in their glycolipids and/or glycoproteins, and in some
cases, virally infected or transformed cells can show a deficient
pattern or sialylation, which is a strong signal of “altered self” to
promote phagocytosis. Also in apoptotic cells, a reduced expression
of sialic acid has been shown [27].
Of particular interest is the physical interaction (by means of
their membrane-bound proteins) between neurons and microglia.
Thus, a group of SAMPs was renamed neuroimmune regulatory
proteins (NIRegs) to highlight their role in modulating and
reshaping an adverse immune response and skewing microglia
towards a protective phenotype. Interestingly, many of these
NIRegs contain an ITIM domain in their cytoplasmic tails which
negatively regulated microglial response to insults.
As mentioned before, siglecs are members of a subgroup of the
Ig superfamily that recognize the sialic acid residues on the periph-
ery of cell surface glycolipids and glycoprotein. Siglecs contain an
ITIM domain in their cytoplasmic tails which signal via the recruit-
ment of tyrosine phosphatases such as SHP1 (reviewed in [35]).
a Resting
Neuron Microglia
CD200 CD200R1
b Chronic Inflammatory situation

Neuron Microglia
CD200 CD200R1
Neuronal damage
CD200 CD200R1
Fig. 2 (a) In a basal situation the cross talk between neuron and microglia via
the CD200–CD200R interaction is correct and the inhibitory signal is working
adequately. (b) In an inflammatory situation the cross talk between neuron and
microglia through the CD200–CD200R interaction is altered. The reduced inhibitory
input from CD200 causes a disturbed equilibrium which results in activation of
microglia and neuronal damage
In this sense, when microglial siglecs interact with the neuronal

glycocalyx, neurotoxicity is alleviated [36].
Another interesting NIReg protein is the pair CD200–CD200
receptor (CD200R). CD200 is a 41–47 KDa protein belonging to
the immunoglobulin Ig supergene family characterized by two
IgSF domains [37, 38]. This surface protein is highly conserved
and within the CNS is mainly expressed by neurons and vascular
endothelium [39]. In contrast, CD200R, which also contains two
IgSF domains and a longer cytoplasmic tail with an ITIM domain
[37], is chiefly expressed by myeloid cells and microglia [39, 40].
CD200–CD200R interaction plays a critical role in neuronal
protection in the setting of inflammatory-mediated neurodegen-
eration (Fig. 2); in particular, this immunoregulatory system, when
deficient, may contribute to chronic inflammation in multiple scle-
rosis, Alzheimer disease, or in the aging brain. Deletion of CD200
results in myeloid cells dysregulation and enhanced susceptibility
to autoimmune response [40–42].
Another neuron–microglia ligand–receptor interaction that
can modulate microglial responsiveness is the neuronal CD47 (also
expressed by endothelium, astrocytes, and dendritic cells) [43] and
the microglial signal-regulatory protein SIRPα ((CD172a), which
contains three extracellular Ig domains) [44]. CD47–CD172a
interaction recruits tyrosine phosphatases SHP-1 and SHP-2 with
inhibition of phagocytosis and synthesis of the anti-inflammatory

cytokine TGF-β (reviewed in [27]). Another NIRegs with a potential
critical role in neuroprotection include the pair neuronal CX3CL1
and microglial CX3CR1 [45].
Interestingly, some membrane receptors contain immunore-
ceptor tyrosine-based activation motifs (ITAM). ITAM-containing
signalling adaptor proteins are associated with receptor subunits.
After ligand–receptor interaction, tyrosine residues of the ITAMs
become phosphorylated by members of the Src kinase family and
subsequently serve as docking sites for Src homology 2 (SH2)
domains of Syk protein kinases which then mediate cellular activation
via a number of downstream cascades (reviewed in [35]). Triggering
receptor expressed on myeloid cells 2 (TREM-2) is an innate
immune glycoprotein heterogeneously expressed in brain microglia
[46]. It belongs to the immunoglobulin- and lectin-like superfamily
with a short cytoplasmic tail [47]. Interaction of TREM-2 with
its uncharacterized ligand induces its association with the ITAM-
containing adaptor molecule DAP12 followed by the recruitment
of ZAP70, SYK, PI3K, and phospholipase Cγ [48]. Activation of
TREM-2/DAP12 leads to augmented phagocytosis with reduced
expression of TNFα, IL-6, and iNOS (reviewed in [49]). Mutation
of either TREM2 or DAP12 proteins leads to the rare Nasu–
Hakola disease (NHD), also known as polycystic lipomembranous
osteodysplasia with sclerosing leukoencephalopathy (PLOSL),
an autosomal recessive inherited disease [50]. While NHD patients
carrying TREM-2 mutation present an early-onset presenile
dementia followed by delayed bone symptoms [51], patients with
mutations in DAP12 display an early-onset combination of presenile
dementia and systemic bone cysts [52]. Interestingly, Piccio et al.
[53, 54] showed that a soluble form of TREM-2 was present in the
cerebrospinal fluid (CSF) of patients with multiple sclerosis and that
blockade of TREM-2 with a blocking antibody exacerbated experi-
mental autoimmune encephalitis in rodents. These observations
suggest a crucial role of TREM-2 in preventing neurodegenerative
processes. So far the nature of the TREM-2 ligand remains unknown.
It has been suggested that TREM-2 binds anionic ligands of the
surface of Gram-positive and Gram-negative bacteria, including
Neisseria gonorrhoeae, as well as an unidentified ligand in the astro-
cytoma cell line HTB12 (reviewed in [49]). Importantly, Hsieh
et al. [55], using a TREM2/Fc chimera, showed that neurons
undergoing apoptosis increased the expression of a TREM-2 ligand
(TREM-2L), inducing their phagocytosis by BV2 cells. This effect
could be reversed with the use of a blocking antibody against
microglial TREM-2 [55].
To add to the complexities of neuron–microglia interactions, it
has been recently described the existence of receptors containing
inhibitory ITAM domains (ITAMi), and conversely some activating
ITIMs have been described [56].
Fig. 3 Scheme showing the different arrays of neuron–microglia cocultures (wt microglia/wt neurons, KO
microglia/wt neurons, wt microglia/KO neurons, KO microglia/KO neurons) that allow exploring different
situations
Many different tools have been developed for the study of

neuron–microglia interaction, ranging from transgenic mice to in
vitro coculture models. One advantage of the cocultures is that it
allows combining neuronal cells from wild-type or KO mice for the
molecule of interest with microglial cells from wild-type or KO mice
in a much simpler and less time and money-consuming way than
developing double KO mice. In this sense, by combining the differ-
ent possibilities (wild-type neurons/wild-type microglia, wild-type
neurons/KO microglia, KO neurons/wild-type microglia, KO neu-
rons/KO microglia), distinct scenarios can be explored. An example
is summarized in Fig. 3. Moreover, since neurons are cultured sepa-
rately from microglia prior the coculture is performed, these cells
can be subjected to different treatments before being put together.
This setting allows the dissection of drug effects on the neuron–
microglia interaction by targeting one kind of cell only.
Here we describe a coculture model for studying the cell-to-cell
contact between neurons and microglial cells [40].
2 Materials
The protocols described here are based on the ones by McCarthy

and de Vellis [57] with minor modifications [58] and by Rose et al.
[59] with modifications [40, 60]. All experiments involving live
animals must conform to national and institutional regulations.
2.1 Reagents 1. Mice or rats in postnatal days 0–2 (P0–P2).

2. Dulbecco’s modified Eagle’s medium (DMEM).
3. Neurobasal medium.
5. Fetal bovine serum (FBS).
6. Horse serum (HS).
7. Penicillin/streptomycin.
8. Poly-D-lysine.
9. Trypsin/EDTA.
10. MilliQ H2O.
2.2 Reagent Setup 1. DMEM 10:10:1. (DMEM, 10 % FBS, 10 % HS, 1 % penicillin/

streptomycin).
2. Trypsin/EDTA solution. (0.05 % Trypsin + 0.02 % EDTA in
HBSS w/o Ca2+/Mg2+).
3. Poly-D-lysine coating solution. Dilute 5 mg of the poly-D-lysine
in 50 ml of sterile MilliQ H2O, to make a 20× stock solution.
Aliquots of 10 ml each can be stored at −20 °C and diluted in
190 ml of MilliQ H2O to make the 1× working concentration
solution (5 μg/ml).
2.3 Equipments 1. Laminar flow hood.

2. Dissecting magnifying glass.
3. Water bath at 37 °C.
4. Humidified tissue culture incubator (37 °C, 5 % CO2).
5. Sterilized microdissecting instruments: large dissecting scissors,
mouse-teeth forceps, curved forceps, and fine Dumont forceps.
6. Orbital shaker.
7. Tabletop centrifuge.
8. 50 ml plastic conical tubes.
9. T75 cm2 tissue culture flasks with plug seal.
10. Tissue culture plates.
11. Medium-size Petri dishes.
12. 50 μm sterile nylon mesh.
2.4 Equipment Setup 1. Poly-D-lysine-coated flasks. Coat the culture flasks with 5 ml of
1× poly-D-lysine (for 1–2 h or overnight) in a 37 °C incubator.
Remove the coating solution and wash once with 5 ml of sterile
MilliQ H2O. Keep in the laminar flow hood until use.
2. Poly-D-lysine-coated plates. Coat the culture plates with 1×
poly-D-lysine and keep overnight in a 37 °C incubator. Remove
the coating solution, wash once with 5 ml of sterile MilliQ
H2O, and air dry in the laminar flow hood until use for neuro-
nal cultures. Microglia cell cultures generally do not require
coating; however, glass cover slips must be poly-D-lysine coated
to ensure a proper adherence of neuronal and microglial cells
to the glass surface. In all cases, plates can be stored at 4 °C
under sterile conditions until use.
3 Methods
3.1 Preparation 1. Dissect the brains: decapitate P0–P2 neonatal rat or mouse
of Mixed Glial Cell pups using sterile large scissors and gently place the head into
Culture (Timing: a Petri dish containing 70 % ethanol. A whole litter (8–12 pups)
15 Days) should be used to ensure high yield rates.
2. Transfer the head to a Petri dish containing cold DMEM.
3. Use mouse-teeth forceps to hold the rostral portion of the
head, and follow the midline to cut the skin and the skull from
the nose to the foramen magnum using curved forceps (see
Note 1).
4. Expose the whole brain and remove it from the skull base and
place it in a new Petri dish containing cold DMEM.
5. Under a dissecting magnifying glass and using thin Dumont
forceps, remove the meningeal layer from the brain. Briefly,
remove the meningeal layer from the inner midbrain, cut the
meninges between the hemispheres, and eliminate them. Then,
clamp the olfactory bulbs and remove the meningeal layer
from both hemispheres. Carefully open the hemispheres,
remove the choroid plexus covering of the inside, and remove
brainstem and cerebellum (see Note 2).
6. Place all the forebrains in new Petri dishes containing cold
DMEM.
7. Using the Dumont forceps, transfer the forebrains to a 50 ml
Falcon tube and add ~0.5 ml of DMEM medium to the tube.
8. Gently triturate and dissociate the nervous tissue with a serum-
coated Pasteur pipette, adding small amounts of DMEM as
long as the content is being aspirated and discarded, until a
homogenate can be found in the medium (see Note 3).
9. Centrifuge the tubes for 10 min to 168 × g.
10. Discard the supernatant by aspiration and suspend the pellet

in 50 ml of warm DMEM 10:10:1.
11. Seed poly-D-lysine-coated 75 cm2 flasks with 10 ml of cell
suspension/each.
12. Incubate at 37 °C in water saturated with 5 % CO2 and 95 %
air atmosphere for 15 days, without changing the culture
medium along this time.
3.2 Isolation 1. After the incubation time, properly close the flasks and place
of Microglial Cells them in an orbital shaker.
(Timing ~4–24 h) 2. Shake at 230 rpm for 3 h and collect the supernatants in 50 ml
Falcon tube (see Note 4).
3. Centrifuge the cell suspension for 10 min to 168 × g.
4. Discard the supernatant by aspiration and suspend the pellet in
1 ml of warm DMEM 10:10:1.
5. Determine the number of viable cells by gently mixing 10 μl of
homogenate with 80 μl of PBS and 10 μl of trypan blue.
6. Dilute the cell suspension to the desired cell concentration
with warm DMEM 10:10:1. Incubate for 2 h in a tissue cul-
ture incubator with 5 % CO2 at 37 °C. To eliminate nonadher-
ent cells, replace the medium with warm DMEM 10:10:1, and
incubate them for 24 h in a tissue culture incubator with 5 %
CO2 at 37 °C.
7. Assess the purity of microglial cultures by examining the char-
acteristic cell morphologies under phase-contrast microscopy,
and confirm it by immunostaining with Mac-1 anti-CD11b
antibody.
8. At this point, microglial cells can be used to perform in vitro
experiments.
3.3 Preparation 1. Remove the E17-E18 embryos from pregnant rat or mouse:
of Cortical Neuronal euthanize a pregnant rat or mouse with CO2 or by cervical
Cell Culture (Timing: dislocation. Spray the abdomen region with 70 % EtOH and
7 Days) open the abdominal cavity with clean scissors. Remove the
uterus containing the embryos and place it in a Petri dish
containing DMEM.
2. Dissect the brains: remove individual embryos from the uterine
horns and decapitate them using sterile large scissors and gently
place the head into a Petri dish containing DMEM.
3. Using forceps gently remove the brain from the skull and place
them in a new Petri dish containing cold DMEM.
4. Under a dissecting magnifying glass and using thin Dumont
forceps, remove the meningeal layer from the brain. Briefly,
remove the meningeal layer from the inner midbrain, cut the
meninges between the hemispheres, and eliminate them.

Carefully open the hemispheres, remove the choroid plexus
covering of the inside, and remove brainstem and cerebellum.
5. Dissect the cortices from the rest of the brain and transfer them
into a 15 ml Falcon with 1 ml of PBS containing 0.25 % trypsin
and 1 mg/ml DNAse I at 37 °C for 15 min.
6. Gently triturate and dissociate the nervous tissue with a serum-
coated Pasteur pipette, adding small amounts of DMEM as
long as the content is being aspirated and discarded, until a
single-cell suspension is achieved.
7. Centrifuge at 500 rpm for 10 min, discard the supernatant by
aspiration and resuspend the pellet in 1 ml of DMEM supple-
mented with 10 % heat-inactivated horse serum.
8. Determine the number of viable cells by gently mixing 10 μl of
homogenate with 80 μl of PBS and 10 μl of trypan blue.
9. Dilute the cell suspension to the desired cell concentration
with warm DMEM 10:10:1. Incubate for 3 h in a tissue culture
incubator with 5 % CO2 at 37 °C. To eliminate nonadherent
cells, replace the medium with warm neurobasal medium
supplemented containing 1 % penicillin/streptomycin, 2 mM
glutamax, and 1× of the antioxidant B-27 supplement.
10. After 3 days in vitro (DIV), halt nonneuronal cell division by
exposure to 10 μM cytosine-D-arabinofuranoside (Ara C), and
continue growing neuronal cells in a humidified environment
containing 5 % CO2 and held at a constant temperature of
37 °C for about 7–8 days.
11. Assess the purity of neuronal cell cultures by examining the
characteristic cell morphologies under phase-contrast micros-
copy, and confirm it by immunostaining with using anti-MAP-
2. The amount of cells other than neurons can be quantified
using GFAP for astrocytes, Mac-1, anti-CD11b antibody for
microglia, and PDFGRα for OPCs. Discard any neuronal cell
culture containing more than 2 % of nonneuronal cells.
12. At this point, neuronal cells can be used to perform in vitro
experiments.
3.4 Preparation 1. In order to prepare neuron–microglia cocultures, it is extremely

of Neuron–Microglia important to coordinate the timing of the preparation of neu-
Coculture ronal cultures and of mixed glial cell cultures followed by the
isolation of microglia.
2. Start by preparing a mixed glial cell culture as described above,
and after 7 DIV, proceed to perform a cortical neuronal cell
culture as previously described. This will ensure that the isolation
of microglia (15 DIV) will correspond to the 7–8 DIV of the
neuronal culture.
Neuron-microglia interaction:
120 Co-cultures
100
80
Cell death
60
40
20
0
Microglia (cells/mL) 1.5x104 3x104 6x104
Neurons (cells/mL) 5x105
Fig. 4 The magnitude of neuronal cell death is dependent on the number of

“resting” (white bars) or “activated” (black bars) microglial cells added to neuronal
cultures (color figure online)
3. In order to establish the adequate coculture conditions,

preliminary experiments varying the density of microglial cells
and the neuron/microglia ratio should be performed.
4. In Hernangómez et al. [40], we described that the optimized
coculture conditions for our study of CD200–CD200R1
interaction were achieved by adding 1.5 × 104 primary microg-
lial onto 5 × 105 neuronal cultures that had been cultured for
7–8 days. Densities of 1.5 × 104, 3 × 104, or 6 × 104 cells per well
in 12-well plates containing 5 × 105 neurons per well were
tested (Fig. 4), obtaining the best results with the proportion
of 1.5 × 104 microglial cells per 5 × 105 neurons (1:33 ratio).
5. However, the optimal ratio may vary depending on the experi-
mental design and should be established accordingly.
4 Notes
Some tricks can be applied to improve the development of the

culture:
1. In the dissection procedure, leave the olfactory bulbs intact so
they can help you to remove quickly the meninges from the
cerebral hemispheres.
2. If the meninges have not been completely removed, forebrain-
cell suspension can be passed through a 150 μm nylon mesh
after the mechanical dissociation, to prevent the contamination
with fibroblasts.
3. If the mechanical dissociation of the nervous tissue is not

enough to obtain a homogenate, a serum-coated fire-polished
Pasteur pipette (with decreased diameter of the tip) can be
used to ensure the triturating of the forebrains.
4. Avoid the use of plastic/glass pipettes by passing the cells
directly from the flask to the Falcon or the Petri dish; if needed,
they can be coated with horse serum or fetal serum simply tak-
ing some milliliters and releasing them, as this coating provides
a smooth hydrophobic barrier so the adherence of cells to the
glass/plastic surface is greatly reduced and minimizes physical
damage to the cells.
Acknowledgments
The authors thank to REEM (Red Española de Esclerosis Múltiple)

RD07/0060/0010, Instituto de Salud Carlos III, and Plan
Nacional, SAF 2010/17501, Ministerio de Economía y
Competitividad for financial support.
References
1. Yang I, Han SJ, Kaur G, Crane C, Parsa AT 8. Mittelbronn M, Dietz K, Schluesener HJ,
(2010) The role of microglia in central ner- Meyermann R (2001) Local distribution of
vous system immunity and glioma immunol- microglia in the normal adult human central
ogy. J Clin Neurosci 17:6–10 nervous system differs by up to one order of
2. del Río-Hortega P (1919) El tercer elemento magnitude. Acta Neuropathol 101:249–255
de los centros nerviosos. I La microglia en 9. Olah M, Biber K, Vinet J, Boddeke HW
estado normal. II Intervención de la microglia (2011) Microglia phenotype diversity. CNS
en los procesos patológicos. III Naturaleza Neurol Disord Drug Targets 10:108–118
probable de la microglia. Bol de la Soc esp de 10. Hanisch UK, Kettenmann H (2007) Microglia:
biol 9:69–120 active sensor and versatile effector cells in the
3. Graeber MB (2010) Changing face of microg- normal and pathologic brain. Nat Neurosci
lia. Science 330:783–788 10:1387–1394
4. Ling EA, Wong WC (1993) The origin and 11. Kraft AD, McPherson CA, Harry GJ (2009)
nature of ramified and amoeboid microglia: a his- Heterogeneity of microglia and TNF signaling
torical review and current concepts. Glia 7:9–18 as determinants for neuronal death or survival.
5. Saijo K, Glass CK (2011) Microglial cell origin Neurotoxicology 30:785–793
and phenotypes in health and disease. Nat Rev 12. Ransohoff RM, Perry VH (2009) Microglial
Immunol 11:775–787 physiology: unique stimuli, specialized
6. Barnham KJ, Cappai R, Beyreuther K, Masters responses. Annu Rev Immunol 27:119–145
CL, Hill AF (2006) Delineating common 13. Davalos D, Grutzendler J, Yang G, Kim JV,
molecular mechanisms in Alzheimer’s and Zuo Y, Jung S, Littman DR, Dustin ML, Gan
prion diseases. Trends Biochem Sci 31: WB (2005) ATP mediates rapid microglial
465–472 response to local brain injury in vivo. Nat
7. Lawson LJ, Perry VH, Dri P, Gordon S Neurosci 8:752–758
(1990) Heterogeneity in the distribution 14. Nimmerjahn A, Kirchhoff F, Helmchen F
and morphology of microglia in the normal (2005) Resting microglial cells are highly
adult mouse brain. Neuroscience dynamic surveillants of brain parenchyma in
39:151–170 vivo. Science 308:1314–1318
15. Gasque P, Dean YD, McGreal EP, VanBeek J, 30. Neumann H, Wekerle H (1998) Neuronal
Morgan BP (2000) Complement components control of the immune response in the central
of the innate immune system in health and nervous system: linking brain immunity to neu-
disease in the CNS. Immunopharmacology rodegeneration. J Neuropathol Exp Neurol
49:171–186 57:1–9
16. Barnum SR (2002) Complement in central 31. Grimsley C, Ravichandran KS (2003) Cues for
nervous system inflammation. Immunol Res apoptotic cell engulfment: eat-me, don’t eat-
26:7–13 me and come-get-me signals. Trends Cell Biol
17. Wyss-Coray T, Mucke L (2002) Inflammation 13:648–656
in neurodegenerative disease–a double-edged 32. Meri S, Pangburn MK (1990) Discrimination
sword. Neuron 35:419–432 between activators and nonactivators of the
18. Medzhitov R, Janeway CA Jr (2002) Decoding alternative pathway of complement: regula-
the patterns of self and nonself by the innate tion via a sialic acid/polyanion binding site on
immune system. Science 296:298–300 factor H. Proc Natl Acad Sci USA 87:
19. Elward K, Gasque P (2003) “Eat me” and 3982–3986
“don’t eat me” signals govern the innate 33. Crocker PR (2005) Siglecs in innate immunity.
immune response and tissue repair in the CNS: Curr Opin Pharmacol 5:431–437
emphasis on the critical role of the comple- 34. Angata T, Kerr SC, Greaves DR, Varki NM,
ment system. Mol Immunol 40:85–94 Crocker PR, Varki A (2002) Cloning and char-
20. Matzinger P (2007) Friendly and dangerous acterization of human Siglec-11. A recently
signals: is the tissue in control? Nat Immunol evolved signaling molecule that can interact
8:11–13 with SHP-1 and SHP-2 and is expressed by
21. Franc NC, White K, Ezekowitz RA (1999) tissue macrophages, including brain microglia.
Phagocytosis and development: back to the J Biol Chem 277:24466–24474
future. Curr Opin Immunol 11:47–52 35. Linnartz B, Wang Y, Neumann H (2010)
22. Gregory CD (2000) CD14-dependent clear- Microglial immunoreceptor tyrosine-based acti-
ance of apoptotic cells: relevance to the immune vation and inhibition motif signaling in neuro-
system. Curr Opin Immunol 12:27–34 inflammation. Int J Alzheimers Dis 2010
23. Imler JL, Hoffmann JA (2002) Toll receptors 36. Wang Y, Neumann H (2010) Alleviation of
in Drosophila: a family of molecules regulating neurotoxicity by microglial human Siglec-11.
development and immunity. Curr Top J Neurosci 30:3482–3488
Microbiol Immunol 270:63–79 37. Barclay AN, Wright GJ, Brooke G, Brown
24. Fadok VA, Bratton DL, Rose DM, Pearson A, MH (2002) CD200 and membrane protein
Ezekewitz RA, Henson PM (2000) A receptor interactions in the control of myeloid cells.
for phosphatidylserine-specific clearance of Trends Immunol 23:285–290
apoptotic cells. Nature 405:85–90 38. Wright GJ, Jones M, Puklavec MJ, Brown MH,
25. Amor S, Puentes F, Baker D, van der Valk P Barclay AN (2001) The unusual distribution of
(2010) Inflammation in neurodegenerative the neuronal/lymphoid cell surface CD200
diseases. Immunology 129(2):154–169 (OX2) glycoprotein is conserved in humans.
Immunology 102:173–179
26. Parnaik R, Raff MC, Scholes J (2000)
Differences between the clearance of apoptotic 39. Koning N, Swaab DF, Hoek RM, Huitinga I
cells by professional and non-professional (2009) Distribution of the immune inhibitory
phagocytes. Curr Biol 10:857–860 molecules CD200 and CD200R in the normal
central nervous system and multiple sclerosis
27. Griffiths M, Neal JW, Gasque P (2007) Innate lesions suggests neuron-glia and glia-glia
immunity and protective neuroinflammation: interactions. J Neuropathol Exp Neurol
new emphasis on the role of neuroimmune 68:159–167
regulatory proteins. Int Rev Neurobiol
82:29–55 40. Hernangómez M, Mestre L, Correa FG, Loría
F, Mecha M, Íñigo PM, Docagne F, Williams
28. Neumann H, Boucraut J, Hahnel C, Misgeld RO, Borrell J, Guaza C (2012) CD200-
T, Wekerle H (1996) Neuronal control of CD200R1 interaction contributes to neuro-
MHC class II inducibility in rat astrocytes and protective effects of anandamide on
microglia. Eur J Neurosci 8(12):2582–2590 experimentally induced inflammation. Glia
29. McCluskey LP, Lampson LA (2000) Local 60:1437–1450
neurochemicals and site-specific immune regu- 41. Hoek RM, Ruuls SR, Murphy CA, Wright GJ,
lation in the CNS. J Neuropathol Exp Neurol Goddard R, Zurawski SM, Blom B, Homola
59:177–187 ME, Streit WJ, Brown MH, Barclay AN,
Sedgwick JD (2000) Down-regulation of the 52. Paloneva J, Kestila M, Wu J, Salminen A,

macrophage lineage through interaction with Bohling T, Ruotsalainen V, Hakola P, Bakker
OX2 (CD200). Science 290:1768–1771 AB, Phillips JH, Pekkarinen P, Lanier LL,
42. Gorczynski RM, Chen Z, Yu K, Hu J (2001) Timonen T, Peltonen L (2000) Loss-of-
CD200 immunoadhesin suppresses collagen- function mutations in TYROBP (DAP12)
induced arthritis in mice. Clin Immunol result in a presenile dementia with bone cysts.
101:328–334 Nat Genet 25:357–361
43. de Vries HE, Hendriks JJ, Honing H, De 53. Piccio L, Buonsanti C, Mariani M, Cella M,
Lavalette CR, van der Pol SM, Hooijberg E, Gilfillan S, Cross AH, Colonna M, Panina-
Dijkstra CD, van den Berg TK (2002) Signal- Bordignon P (2007) Blockade of TREM-2
regulatory protein alpha-CD47 interactions exacerbates experimental autoimmune enceph-
are required for the transmigration of mono- alomyelitis. Eur J Immunol 37:1290–1301
cytes across cerebral endothelium. J Immunol 54. Piccio L, Buonsanti C, Cella M, Tassi I,
168:5832–5839 Schmidt RE, Fenoglio C, Rinker J, Naismith
44. Brown EJ, Frazier WA (2001) Integrin- RT, Panina-Bordignon P, Passini N, Galimberti
associated protein (CD47) and its ligands. D, Scarpini E, Colonna M, Cross AH (2008)
Trends Cell Biol 11:130–135 Identification of soluble TREM-2 in the cere-
45. Lyons A, Lynch AM, Downer EJ, Hanley R, brospinal fluid and its association with multiple
O’Sullivan JB, Smith A, Lynch MA (2009) sclerosis and CNS inflammation. Brain 131:
Fractalkine-induced activation of the phosphati- 3081–3091
dylinositol-3 kinase pathway attenuates microg- 55. Hsieh CL, Koike M, Spusta SC, Niemi EC,
lial activation in vivo and in vitro. J Neurochem Yenari M, Nakamura MC, Seaman WE (2009)
110(5):1547–1556 A role for TREM2 ligands in the phagocytosis
46. Schmid CD, Sautkulis LN, Danielson PE, of apoptotic neuronal cells by microglia.
Cooper J, Hasel KW, Hilbush BS, Sutcliffe JG, J Neurochem 109:1144–1156
Carson MJ. Heterogeneous expression of the 56. Barrow AD, Trowsdale J (2006) You say
triggering receptor expressed on myeloid ITAM and I say ITIM, let’s call the whole
cells-2 on adult murine microglia. J Neurochem thing off: the ambiguity of immunoreceptor
83(6):1309–1320. signalling. Eur J Immunol 36:1646–1653
47. Bouchon A, Hernandez-Munain C, Cella M, 57. McCarthy KD, de Vellis J (1980) Preparation
Colonna M (2001) A DAP12-mediated path- of separate astroglial and microglial cell cul-
way regulates expression of CC chemokine tures from rat cerebral tissue. J Cell Biol
receptor 7 and maturation of human dendritic 85:890–902
cells. J Exp Med 194:1111–1122 58. Mecha M, Íñigo PM, Mestre L, Hernangómez
48. Lanier LL, Corliss BC, Wu J, Leong C, Phillips M, Borrell J, Guaza C (2011) An easy and fast
JH (1998) Immunoreceptor DAP12 bearing a way to obtain a high number of glial cells from
tyrosine-based activation motif is involved in rat cerebral tissue: a beginners approach. Protocol
activating NK cells. Nature 391:703–707 Exchange. doi:10.1038/protex.2011.218
49. Ford JW, McVicar DW (2009) TREM and 59. Rose K, Goldberg MP, Choi DW (1993)
TREM-like receptors in inflammation and Cytotoxicity in murine neocortical cell culture.
disease. Curr Opin Immunol 21:38–46 In: Tyson CA, Frazier JM (eds) Methods in
50. Colonna M (2003) DAP12 signaling: from neurotoxicology. Academic, San Diego, pp
immune cells to bone modeling and brain 46–60
myelination. J Clin Invest 111:313–314 60. Loría F, Petrosino S, Hernangómez M, Mestre
51. Chouery E, Delague V, Bergougnoux A, L, Spagnolo A, Correa F, Di Marzo V,
Koussa S, Serre JL, Megarbane A (2008) Docagne F, Guaza C (2010) An endocannabi-
Mutations in TREM2 lead to pure early-onset noid tone limits excitotoxicity in vitro and in a
dementia without bone cysts. Hum Mutat model of multiple sclerosis. Neurobiol Dis
29:E194–E204 37(1):166–176
Chapter 21
Preparation of Rodent Primary Cultures for Neuron–Glia,

Mixed Glia, Enriched Microglia, and Reconstituted Cultures
with Microglia
Shih-Heng Chen, Esteban A. Oyarzabal, and Jau-Shyong Hong
Abstract
Microglia, neurons, and macroglia (astrocytes and oligodendrocytes) are the major cell types in the central
nervous system. In the past decades, primary microglia-enriched cultures have been widely used to study
the biological functions of microglia in vitro. In order to study the interactions between microglia and
other brain cells, neuron–glia, neuron–microglia, and mixed glia cultures were developed. The aim of this
chapter is to provide basic and adaptable protocols for the preparation of these microglia-containing
primary cultures from rodent. Meanwhile, we also want to provide a collection of tips from our collective
experiences doing primary brain cell cultures.
Key words Microglia, Neuron–glia culture, Mixed glia culture, Microglia-enriched culture,
Reconstituted neuron–microglia culture
1 Introduction
Although the in vitro study of microglia goes as far back as 1930

[1, 2], it was not until 1986, with the development of a new pro-
tocol to selectively isolate and culture microglia from mammalian
brain [3], that the use of microglia in cell culture systems of the
brain became popular. Since then, and with the advent of several
different microglia-containing cell culture protocols and microglia-
specific cell markers, our understanding of the physiology of
microglia continues to vastly improve.
Several cell culture systems exist to study microglia derived
from rodent brains, including mesencephalic neuron–glia, mixed
glia, enriched microglia, and reconstituted neuron–microglia
cultures. Each culture system can serve different purposes to
study the molecular mechanisms and the interactions between
microglia and neurons, microglia and astrocytes, or microglia
themselves under normal physiological or pathological conditions.
231
232 Shih-Heng Chen et al.
Mesencephalic neuron–glia cultures serve as an excellent in vitro

model of the midbrain—both in the diversity and ratio of differ-
ent cell types when compared to the normal physiology of the
brain. This culture is predominantly composed of astrocytes
(~50 % GFAP immunopositive cells), neurons (~40 % NeuN
immunopositive cells), and microglia (~10 % OX-42 immu-
nopositive cells) [4] and is commonly used as a platform to study
neuron–glia interaction, especially in culture models that focus
on degeneration or neuroprotection of dopaminergic neurons
(which account for 1–2 % of the neurons in this culture)—such as
in toxin-induced models of Parkinson’s disease [4–7].
Primary mixed glia cultures are composed of astrocytes
(~80 % GFAP immunopositive cells) and microglia (~20 % OX-42
immunopositive cells) [8] and are commonly used to study the
interaction between microglia and astrocytes. In mixed glia cul-
tures, microglia grow loosely attached to the confluent mono-
layer of astrocytes; therefore, microglia can be shaken off to
produce a microglia-enriched culture that is composed of more
than 95 % microglia, as characterized by immunocytochemical
staining with OX-42 [9]. Primary rodent microglia-enriched cul-
tures are widely used to study microglial biology in vitro. Lastly,
enriched microglia can be seeded on top of neuron-enriched cul-
tures to generate reconstituted neuron–microglia cocultures
which are a useful tool to investigate direct neuron and microglia
interaction in vitro.
Over the past decade these protocols have successfully allowed
our laboratory to investigate the physiology of microglia. Our
hope is to share our experience to help other researchers establish
their own microglia-containing primary cultures.
2 Materials
2.1 Equipment 1. A guillotine or an acrylic box connected to CO2 tank for euth-
Component anizing animals (see Note 1).
2. One pair operating scissors (5½″).
3. One pair dissecting scissors (4¼″)
4. One pair microdissecting scissors (4½″, 17 mm blades).
5. One pair curved dissecting forceps (5″).
6. One pair microdissecting tweezers (no. 5).
7. One pair curved microdissecting tweezers (no. 5).
8. Dissection microscope.
9. Sterile petri dishes.
10. Sterile 50 mL conical vials.
11. Sterile 10 and 25 mL serological pipettes.
12. Sterile 200 μL and 1 mL pipette tips.
Preparation of Primary Cultures Containing Microglia 233
13. Sterile 70 μm cell strainer (BD Bioscience, San Jose, CA, USA).
14. Sterile tissue culture-grade plates (Corning Inc., Corning, NY,
USA) or flasks (BD Bioscience, San Jose, CA, USA).
15. Hemocytometer, 0.4 % Trypan Blue Stain solution (Invitrogen,
Carlsbad, CA, USA) and bright-field microscope.
16. Humidified cell culture incubator (37 °C, 5 % CO2).
2.2 Reagents and 1. Neuron–Glia Dissection Buffer: Minimum Essential Medium

Solutions Component (MEM) (cat. No. 11090, Invitrogen, Carlsbad, CA, USA).
2. Mixed Glia Dissection Buffer: Dulbecco’s modified Eagle’s
medium/Nutrient Mixture F-12 (DMEM/F12) (cat. No.
11330, Invitrogen, Carlsbad, CA, USA).
3. Poly-D-lysine working solution: dilute 100 μg/mL poly-D-
lysine (Sigma-Aldrich, St. Louis, MO, USA) in ddH2O to a
final concentration of 20 μg/mL. Store at 4 °C.
4. Cytosine β-D-arabinofuranoside (ara-C) (Sigma-Aldrich, St.
Louis, MO, USA) solution: dilute 10 mM of stock solution in
Neuron–Glia Maintenance Medium to the desired final
concentration.
5. Neuron–Glia Maintenance Medium (for 500 mL): combine
380 mL of MEM, 50 mL of heat-inactivated fetal bovine
serum (FBS), 50 mL of heat-inactivated horse serum (HS),
5 mL of 100× nonessential amino acids (Invitrogen, Carlsbad,
CA, USA), 5 mL of 100 mM sodium pyruvate (Sigma-Aldrich,
St. Louis, MO, USA), 5 mL of 200 mM L-glutamine, 5 mL of
5,000 U/mL penicillin and 5,000 μg/mL streptomycin, and
0.5 g of D-glucose. Stir gently to dissolve, filter sterilize, and
store in the dark at 4 °C.
6. Mixed Glia Maintenance Medium (for 500 mL): combine
430 mL of DMEM/F12 (Invitrogen, Carlsbad, CA, USA),
5 mL of 100× nonessential amino acids, 5 mL of 100 mM
sodium pyruvate, 5 mL of 200 mM L-glutamine, and 5 mL of
5,000 U/mL penicillin and 5,000 μg/mL streptomycin. Stir
gently to dissolve, adjust pH to 7.2, add 50 mL of heat-
inactivated FBS, filter sterilize, and store in the dark at 4 °C.
3 Methods
3.1 Preparing 1. All cell culture plates and flasks are pre-coated with poly-D-
for Primary Culture lysine before seeding with single cell suspension. Coat 24- and
6-well plates with 0.5 and 1 mL of the poly-D-lysine working
solution, respectively, and 25 mL to 175 cm2 culture flasks.
Place in 37 °C incubator for at least 1 h.
2. After coating, rinse the culture plates or flasks three times:
twice with sterile ddH2O and once with sterile PBS. The volume
Fig. 1 Illustration of the steps to isolate the mesencephalic region. (a) Separate the rostral forebrain (use the
traverse sinus as hallmark) and the caudal hindbrain from the mesencephalic region as indicated the dotted
lines. (b) Butterfly the tissue by inserting the microdissection scissors through the inside of the mesencephalic
region of the neural tubule and slicing open the tissue along the dorsal midline. (c) Remove the meninges from
the butterflied mesencephalic tissue
of ddH2O and PBS in different culture plates and flask is 1 mL

in 24-well plate, 2 mL in 6-well plate, and 25 mL in 175 cm2
flask. Store culture plates and flasks in PBS at 37 °C until
immediately before cell seeding.
3. Preheat aliquoted culture medium in water bath at 37 °C, and
sterilize dissection instruments and work surfaces with 70 %
ethanol beforehand.
3.2 Mesencephalic 1. Euthanize two to four rat dams at gestation day 14–15 and
Neuron–Glia Culture mouse dams at gestation day 13–14 (see Notes 1 and 2).
Protocol 2. Lay the rodents on their back and clean the surface of abdominal
area with 70 % ethanol. Lift the abdominal skin with the curved
dissecting forceps, and using operating scissors make a vertical
incision to expose the abdominal muscle layer. Using the dis-
section scissors make a vertical incision through the muscle
layer to expose the abdominal cavity. Carefully remove the
uterine horns and place them in a petri dish filled with ice-cold
Neuron–Glia Dissection Buffer.
3. Carefully remove the embryos from the amniotic membrane
and transfer them into a petri dish with ice-cold Neuron–Glia
Dissection Buffer and place on ice. Rinse the embryos twice
with ice-cold Neuron–Glia Dissection Buffer and transfer them
into a new petri dish with ice-cold Neuron–Glia Dissection
Buffer. Place this dish on top of a petri dish filled with ice and
set on the floor of a dissection microscope.
4. Under the microscope, hold down the body of the embryo with
a pair of microdissecting tweezers, and make two perpendicular
incisions with the microdissecting scissor to cut out the mesence-
phalic region of the brain (see Fig. 1a) [10]. Insert one blade of
the scissors into the tubular structure, and butterfly the tissue by
making an incision down the dorsal midline (see Fig. 1b, c).
Carefully separate the tissue from the meninges, and transfer the
ventral midbrain tissue into a 50 mL conical tube filled with
10 mL of ice-cold dissection Neuron–Glia Dissection Buffer.
5. In a biological safety cabinet, triturate the midbrain tissues into
a single cell suspension by slowly passing the tissue through a
10 mL serological pipette until the tissues are smaller than
the opening of the 1 mL pipette tip. Fit a sterile 1 mL pipette
tip to the end of the 10 mL serological pipette, and continue
to triturate the tissue until it is smaller than the opening of a
200 μL pipette tip. Remove the 1 mL pipette tip and fit a 200
μL pipette tip to the end of the 10 mL serological pipette and
triturate the tissues three to five times and strain the cell
suspension through a 70 μm cell strainer into a fresh 50 mL
conical tube (see Note 3).
6. Centrifuge the cell suspension at 430 × g for 6 min at 4 °C.
7. Decant the supernatant and resuspend the pellet in 10 mL of
warm Neuron–Glia Maintenance Medium.
8. Using Trypan Blue Stain solution and a hemocytometer, deter-
mine the cell count and viability of the cell suspension. Adjust the
density of the viable cells with warm Neuron–Glia Maintenance
Medium to 1 × 106 cells/mL. Cap and invert the conical tube to
ensure a thorough mixture of the cell suspension. Add 0.5 mL
per well of the cell suspension to a poly-D-lysine-coated 24-well
plates (see Note 4). Gently agitate the plate in several direc-
tions to ensure even seeding, and place it in a humidified
37 °C, 5 % CO2 incubator. Avoid stacking plates because it
may affect the cellular air exchange of the plates.
9. Gently supplement each well with 0.5 mL of warm Neuron–
Glia Maintenance Medium on the third day after seeding.
Be sure to add medium along the side of the wells to avoid
disturbing the cellular monolayer.
10. Cells are ready for treatment 7 days after their initial seeding.
The cellular composition of the neuron–glia culture can be
determined by immunocytochemistry staining (see Note 5).
3.3 Mixed Glia 1. Wipe the head of 0–1-day-old rodent pups with 70 % ethanol
Culture Protocol and euthanize them by decapitation.
2. Remove the whole brain by securing the rostral end of the
head (dorsal side upward) with a pair of tweezers and making
an incision through the foramen magnum towards the ear and
curving up (near the eye sockets) towards the coronal suture.
Peel back the skin and skull to expose the brain with a pair of
tweezers and gently scoop up the entire brain, and immerse it
in a petri dish filled with Mixed Glia Dissection Buffer. Place
this dish on top of a petri dish filled with ice and set on the
floor of a dissection microscope.
Table 1
Seeding concentration for primary murine mixed glia cultures
Mouse Rat
6-well plate 1.5 × 106 cells/well 1 × 106 cells/well
24-well plate 1.5 × 105 cells/well 1 × 105 cells/well
175 cm2 flask 5 brains/flask 2.5 brains/flask
3. Remove the olfactory bulbs, cerebellum, and brain stem using

microdissecting tweezers and curved microdissecting tweezers.
Dissociate the remaining brain tissue into cerebral hemispheres
and midbrain. Carefully remove all the meninges and blood ves-
sels from each tissue. Pool the brain tissue into a 50 mL conical
tube with 15 mL of ice-cold Mixed Glia Dissection Buffer.
4. In a biological safety cabinet, triturate the pooled cerebral
hemisphere and midbrain tissues into a single cell suspension by
slowly passing the tissue through a 10 mL serological pipette
until the tissues are smaller than the opening of the 1 mL pipette
tip. Fit a sterile 1 mL pipette tip to the end of the 10 mL sero-
logical pipette, and continue to triturate the tissue until it is
smaller than the opening of a 200 μL pipette tip. Remove the
1 mL pipette tip and fit a 200 μL pipette tip to the end of the
10 mL serological pipette and triturate the tissues three to five
times and strain the cell suspension through a 70 μm cell strainer
into a fresh 50 mL conical tube (see Note 3).
6. Decant the supernatant and resuspend the pellet in 10 mL of
warm mixed glia culture maintenance medium.
7. Using Trypan Blue Stain and a hemocytometer, determine the
cell count and viability of the cell suspension. Adjust the density
of the viable cells with warm Mixed Glia Maintenance Medium
to acquire the proper seeding concentration (see Table 1).
8. Refresh the medium every three days by completely removing
and replacing the spent medium. Add 25 mL for flasks, 1 mL/
well for 24-well plates, and 3 mL/well for 6-well plates
(see Note 6).
9. Cells are ready for treatment 7–14 days after their initial seeding,
depending on size of the culture container. Cultures seeded on
6- or 24-well plates are ready for treatment 7 days after seeding
(see Note 7).
3.4 Enriched 1. Grow mixed glial cultures (see Subheading 3.3) in 175 cm2
Microglia Culture culture flasks for about 14–16 days after seeding, tightly cap
Protocol the flasks, and seal them with parafilm (see Note 8).
2. Stack the flasks on a flat platform shaker and shake the flasks at
180 rpm for 30 min to 1 h at 37 °C (see Note 9). After shaking,
collect the medium into 50 mL conical tubes, and centrifuge
the cell suspension for 6 min at 430 × g at 4 °C.
3. Resuspend the cell pellet in an appropriate volume of Mixed
Glia Maintenance Medium.
4. Using Trypan Blue Stain and a hemocytometer, determine
the cell count and viability of the cell suspension. Adjust the
density of the viable cells with warm Mixed Glia Maintenance
Medium to 1 × 106 cells/mL. Seed the cell suspension into
either a 96-, 24-, or 6-well plate with 0.1, 0.5, and 2 mL of the
cell suspension per well, respectively. Store cells overnight in a
humidified incubator (37 °C, 5 % CO2).
5. Cells are ready for treatment the next day.
3.5 Reconstituted 1. Grow neuron–glia cultures (see Subheading 3.2) in 24-well

Neuron–Microglia plates for 48 h, and gently supplement each well with 0.5 mL
Coculture Protocol of warm Neuron–Glia Maintenance Medium containing
ara-C with a final concentration of 5–15 μM (see Notes 10
and 11).
2. Refresh the medium at day 6 after seeding by completely
removing and replacing the spent ara-C-containing medium
with warm neuron–glia medium containing a cell suspension
of enriched microglia, seeding at 5 × 104 cells/well. Store cells
overnight in a humidified incubator (37 °C, 5 % CO2).
3. Cells are ready for treatment the next day.
4 Notes
1. Rat and mouse dams are euthanized without anesthesia by

decapitation and cervical dislocation, respectively, because
anesthetic agents have been shown to affect brain chemistry
[11]. Rapid asphyxiation with CO2 gas is a suitable alternative
method that appears to have a minimal impact on embryonic
brain cells.
2. An accurate estimate of embryonic age is extremely important
when culturing dopaminergic neurons—this is because cells
must be transplanted during a critical 2-day window during
which dopaminergic neurons are undergoing differentiation.
Transplants outside this critical window result in poor survival.
The embryonic age can be confirmed by comparing the crown-
to-rump length (CRL) (see Fig. 2) of your embryos to the
average lengths (see Table 2) [12, 13]. Differences in CRLs
may also be attributed to differences among strains or animals
with partial or complete genetic ablations. Thus it is important
Fig. 2 The crown-to-rump length (CRL) of the embryo is the greatest distance
from the top of the skull to the buttock
Table 2
Average CRL measurements from E13 to E15 rodents as derived from data
compiled by Butler and Juurlink [12]
Gestation days Mouse Rat

E13 7–8 mm 7–9 mm
E14 9–10 mm 10–11 mm
E15 10.5–11.5 mm 12.5–13 mm
to conduct pilot studies to determine the most suitable critical

time window for your cultures.
3. The frequency of tissue trituration varies among cell culture
users. The number of passages may depend on the pipette-aid
and the size and quantity of brain tissue. A gradual decrease
in the size of the tip in respect to the decrease in the size of
the tissue reduces the sheering pressure exerted on tissue
resulting in less cellular rupturing (optimally less than 10 % of
the total cells).
4. Based on our laboratory’s experience, optimal growth of
neuron–glia cultures has been observed in 24-well plate from
Corning but not BD Falcon.
5. Immunocytochemical staining using cell type-specific markers
(see Table 3) is routinely used to determine the cellular hetero-
geneity of the neuron–glia culture—which should be composed
of ~50 % astrocytes, ~40 % neurons, and ~10 % microglia.
A variation in the ratio of microglia, either above or below 10 % of
Table 3
Markers used to identify brain cells
Cell type Marker

Astrocytes GFAP
Microglia OX-42
Iba-1
Neurons NeuN
MAP-2
Dopaminergic neurons TH
Oligodendrocytes MBP
the total cells, may result in highly variable experimental results.

Thus, it is important to monitor the cellular heterogeneity of
the neuron–glia culture to ensure cell culture stability from
batch to batch.
6. The ability of the cells to adhere to the culture plates may be
damaged if cultures are perturbed during the first three days.
Gently agitate the plates or flasks before refreshing the medium
to dislodge any cellular debris—reducing potential endogenous
ligands that could activate microglia via damage-associated
molecular pattern receptors.
7. The growth of microglia may be delayed in different strains or
transgenic mice—pilot studies may be required to optimize the
microglia composition in the culture.
8. Covering the filter cap of flasks with parafilm prevents the
fluctuation of pH in medium and reduces the risk for
contamination.
9. The speed and duration of shaking is relatively slower and
shorter than other protocols, resulting in fewer but more pure
cell suspensions of microglia. The time of shaking should be
optimized by individual to reduce Type II astrocyte and oligo-
dendrocyte contamination.
10. The consistency of the neuron–glia culture is important dur-
ing the proper preparation of neuron-enriched cultures. Once
the culture is consistent, optimize the treatment concentra-
tion of ara-C to ensure the highest glial toxicity and lowest
neurotoxicity.
11. Ara-C and leucine methyl ester (LME) are routinely used to
ablate glial cell and microglia, respectively, from culture sys-
tems. It is important to note that as little as 1 % microglial
contamination can result in a detectable proinflammatory
response.
Acknowledgements
This research was supported [in part] by the Intramural Research

Program of the NIH, National Institute of Environmental Health
Sciences. We would like to acknowledge Dr. Bin Liu for his contri-
bution in developing these protocols.
References
1. Costero J (1930) Estudie del compotamento polysaccharide: relevance to the etiology of
de la microlgia cultivade on vetro. Datos con- Parkinson's disease. J Neurosci 23(4):
cernientes a su histogenesis. Mem R Soc cep 1228–1236
Hist nat 14:125–182 8. Chang RC, Chen W, Hudson P et al (2001)
2. Kettenmann H, Hanisch UK, Noda M et al Neurons reduce glial responses to lipopolysac-
(2011) Physiology of microglia. Physiol Rev charide (LPS) and prevent injury of microglial
91(2):461–553 cells from over-activation by LPS. J Neurochem
3. Giulian D, Baker TJ (1986) Characterization of 76(4):1042–1049
ameboid microglia isolated from developing 9. Liu B, Du L, Hong JS (2000) Naloxone protects
mammalian brain. J Neurosci 6(8):2163–2178 rat dopaminergic neurons against inflammatory
4. Gao HM, Hong JS, Zhang W et al (2002) damage through inhibition of microglia activa-
Distinct role for microglia in rotenone- tion and superoxide generation. J Pharmacol
induced degeneration of dopaminergic neu- Exp Ther 293(2):607–617
rons. J Neurosci 22(3):782–790 10. Paxinos G, Tork I, Tecott LH et al (1991)
5. Gao HM, Jiang J, Wilson B et al (2002) Atlas of the developing rat brain. Academic,
Microglial activation-mediated delayed and San Diego
progressive degeneration of rat nigral dopami- 11. Hong JS, Wood PL, Gillin JC et al (1980)
nergic neurons: relevance to Parkinson's dis- Changes of hippocampal Met-enkephalin con-
ease. J Neurochem 81(6):1285–1297 tent after recurrent motor seizures. Nature
6. Qin L, Liu Y, Cooper C et al (2002) Microglia 285(5762):231–232
enhance beta-amyloid peptide-induced toxicity 12. Butler H, Juurlink BHJ (1987) An atlas for
in cortical and mesencephalic neurons by pro- staging mammalian and chick embryos, 1st
ducing reactive oxygen species. J Neurochem edn. CRC, Boca Raton
83(4):973–983 13. Torres EM, Weyrauch UM, Sutcliffe R et al
7. Gao HM, Hong JS, Zhang W et al (2003) (2008) A rat embryo staging scale for the gen-
Synergistic dopaminergic neurotoxicity of the eration of donor tissue for neural transplanta-
pesticide rotenone and inflammogen lipo- tion. Cell Transplant 17(5):535–542
Part VIII
Analysis of Microglia Functions In Vivo

Chapter 22
Microglia Detection by Enzymatic Histochemistry

Beatriz Almolda, Berta González, and Bernardo Castellano
Abstract
Visualization of microglia by means of histochemistry has been for years a reliable method to demonstrate
this population of cells in the central nervous system (CNS). Wide range of data on microglia has been
published using lectin and enzymatic histochemistry. While at present, in most laboratories, the use of
specific antibodies is the first choice, histochemical detection of microglia remains a powerful method as it
has certain advantages upon immunohistochemical methods because it is faster, cheaper, and can be used
in different species including human. In this chapter we want to present the detailed methodology for
microglial staining using the histoenzymatic demonstration of the enzyme nucleoside-diphosphatase
(NDPase), a phosphatase found in the plasma membrane of microglia that is absent in the plasma mem-
brane of other glial cells and neurons. With this technique it is possible to visualize amoeboid microglia
during development, ramified microglia in the adult brain, and also reactive microglia. As the technique
also stains the blood vessels, it allows the analysis of the relationship between microglia and vasculature.
This method can be performed in either histological sections or cell cultures for light microscopy analysis.
Furthermore, we described how to combine this histochemical method with conventional immunohisto-
chemistry for double labelling using other markers, and finally we give details to perform the procedure
not only for optical microscopic studies but also for transmission electron microscopy (TEM).
Key words Microglia, Histochemistry, NDPase, Immunohistochemistry, Transmission electron

microscopy, Cell culture, Double labelling
1 Introduction
It was during the first part of the twentieth century that Pio del Rio
Hortega through the development of a specific technique such as
the silver carbonate method [1] was able to describe the microglial
cells and distinguish them from other cells in nervous tissue [2].
For many years silver carbonate has been the only tool that neuro-
scientist has had to visualize this population of cells. However,
between 1981 and 1982, Murabe and Sano reported that using the
method of Novikoff and Goldfisher [3], intense thiamine pyro-
phosphatase (TPPase) and nucleoside-diphosphatase (NDPase)
enzymatic activities were selectively found at the plasma membrane
of glial cells [4, 5]. Authors identified these glial cells as microglia
243
244 Beatriz Almolda et al.
because they were morphologically similar to microglial cells in the

sections stained with Rio Hortega’s silver impregnations. Although
it is unclear whether a single enzyme performs both enzymatic
activities or if there are different isoenzymes [6, 7], when performed
in the same conditions, the reaction product was always found in
the same location, without any difference, and therefore histo-
chemical demonstration of either TPPase or NDPase has been used
indiscriminately as a selective marker of microglia. In our hands,
although we have used both thiamine pyrophosphate and nucleo-
side diphosphate as substrate, and all worked well, in most studies
we have used nucleoside diphosphates because stains showed up
cleaner and stronger, also requiring less incubation time.
For years, NDPase histochemistry has been used for the selec-
tive labelling of microglial cells in the central nervous system of
different species including fishes and even humans [8–14]. NDPase
histochemistry stains not only amoeboid and ramified microglia in
the normal intact nervous system [15, 16] but also activated-
reactive microglia in the experimentally lesioned brain [11, 17]
and in neurological diseases such as Alzheimer’s disease [18] and
other disorders [12, 19]. In front to immunohistochemical meth-
ods, NDPase histochemistry is a faster and cheaper technique that
can be developed not only on histological sections but also in cell
cultures [20] allowing the fast identification of microglial cells,
which are usual contaminants of “pure” neuronal or astrocyte
cultures.
The method is based on the detection of the enzyme that
hydrolyzes the nucleoside diphosphates such as 5′-inosine-
diphosphate (IDP) to 5′-inosine-monophosphate (IMP). To
accomplish that, histological sections obtained with a vibratome
are incubated in a buffered medium containing the substrate
(IDP), Mn2+ ions as cofactor of the enzymatic reaction, and a lead
salt that act as donor of Pb2+ ions. When the enzyme NDPase acts
on the substrate, it yields inorganic phosphate that reacts with the
Pb2+ ions and precipitate “in status nascendi” as lead phosphate
which is insoluble in aqueous solution. As the lead phosphate is
white and cannot be distinguished at the light microscope, it is
necessary to develop the histochemical reaction by using a sulfide
salt which will yield lead sulfide that is brown and therefore visible
at the light microscope. An ulterior treatment with a silver salt will
yield silver sulfide, a more stable reaction product (Fig. 1). Another
advantage of the NDPase histochemistry is that as the first reac-
tion product of the enzymatic activity is electron dense, the sections
can be analyzed at the ultrastructural level using an electron
microscope.
Finally, although not obvious, this technique can be combined
with other markers allowing the double labelling of microglia and
other cells [21]. For instance, double labelling of microglia and
astrocytes can be obtained by combining NDPase with
NDPase Histochemistry for Microglia Detection 245
Fig. 1 Histochemical demonstration of NDPase. In the CNS, NDPase enzyme is located on the plasma membrane
of both microglial cells (M) and endothelial cells of the blood vessels (BV). The enzyme NDPase, which requires
manganese ions (Mn+2) as cofactor, catalyzes the transformation of the substrate inosine 5′-diphosphate into
inosine 5′-monophosphate and inorganic phosphate (PO4−3). If lead ions (Pb+2) are present in the incubation
medium, then the inorganic phosphate precipitates “in status nascendi” as lead phosphate (Pb3(PO4)2).
This salt is white and therefore not observable at the light microscope. Then to visualize it, it is necessary to
treat the samples with ammonium sulfide that reacts with the lead phosphate producing lead sulfide (PbS), an
insoluble salt with brown color. Finally, treatment with silver nitrate yields silver sulfide (Ag2S), a more stable
salt that allows indefinite storage of samples
immunohistochemical detection of GFAP (a recognized astrocyte

marker). To accomplish that, after the histoenzymatic incubation
and before the treatment with sulfide, the process is interrupted
and followed with immunohistochemical staining. Only after the
immunohistochemical staining is finished, the development of the
histoenzymatic reaction is done. As the final product of the histo-
chemical technique is brownish black, it is necessary to develop the
immunohistochemical reaction in a completely different color
(e.g., light blue) to distinguish both labels.
Fig. 2 Microglia labelling by NDPase histochemistry. Photographs showing the NDPase-labelled microglial cells
[arrows in (a–b)] observed in the CNS of different species: frog (a), quail (b), lizard (c), rat (d, g–i), mouse (e), and
human (f). In photographs (c, e, and f), toluidine blue was used as counterstain. (g) Double labelling combining
NDPase histochemistry and GFAP immunohistochemistry. NDPase-labelled microglia (arrows) are seen in brown,
whereas GFAP-stained astrocytes (arrowheads) are visualized in blue. In addition to microglia, NDPase histo-
chemistry also stains blood vessels (BV in b, d–i) allowing the study of microglial interaction with the vasculature.
Activated microglia (arrowheads) are found in close relationship with blood vessels during aging in RLA rats (h).
Microglial precursors and amoeboid microglia (arrowheads) in the developing rat brain (i)
In this chapter we describe how to perform the NDPase enzy-

matic histochemistry in vibratome sections and cultures for quick
identification of microglia (Figs. 2 and 3); how to combine this
histochemical technique with other immunohistochemical label-
ling, in particular GFAP immunohistochemistry that allows to the
visualization of astrocytes (Figs. 2g and 3b); and finally how to
process the NDPase histochemically reacted sections for electron
microscopy analysis (Fig. 4).
Fig. 3 NDPase labelling of microglia in cell cultures. Primary astroglial cell cultures from brains of newborn
mice showing the presence of contaminating microglia (arrows) visualized by NDPase histochemistry. In (a)
the cell body and microglial processes are seen in brown and the nucleus in blue because the culture has been
counterstained with toluidine blue. In (b) NDPase histochemistry has been combined with GFAP immunohisto-
chemistry: microglia are visualized in brown, whereas astrocytes are stained in blue
Fig. 4 NDPase histochemistry for TEM studies. (a) Semithin section showing NDPase-labelled microglia in the
white matter of mice spinal cord. The section has been counterstained with toluidine blue. Note the product of
the NDPase enzymatic activity in both the prolongations (arrows) and cell body (arrowheads) of microglial
cells. (b) Semithin section showing NDPase-labelled microglia (arrow) in a sample from human cerebral cortex.
Note that other cells like neurons (N) are not stained with this histochemical method. The section has been
counterstained with toluidine blue. (c) Ultrathin section from mice spinal cord showing the reaction product of
NDPase activity, which precipitates at the plasma membrane (arrows) of microglia (M). Note that the histo-
chemical reaction allows the detection of microglial processes (arrowheads) in the neuropilum. NDPase label-
ling was also observed in the wall of blood vessels (BV). Note that other glial cells (A) do not show NDPase stain
2 Materials
2.1 Solutions for 1. Cacodylate stock solution (CSS): 0.5 M cacodylate solution at
Fixation and Washing pH 7.4. Dissolve 107 g of sodium cacodylate × 3 H2O (molec-
ular weight 214) in 800 mL of distilled water. Make up to 1 L
with distilled water and adjust the pH by adding some drops of
0.2 M NaOH. Store at 4 °C.
2. Fixative solution (FS): 4 % paraformaldehyde in 0.1 M caco-
dylate buffer at pH 7.4, with 5 % sucrose. For 1 L of FS, dis-
solve 40 g of paraformaldehyde in 700 mL of hot distilled
water (70 °C) by stirring. The solution will appear milky. Add
0.2 M NaOH dropwise until the solution becomes transpar-
ent. Add 200 mL of CSS and filter the solution. Add 50 g of
sucrose to the solution and make up to 1 L with distilled water.
Adjust the pH to 7.4 and store at 4 °C.
3. Fixative solution with glutaraldehyde (FSG): 2 % paraformalde-
hyde in 0.1 M cacodylate buffer and 0.5 % glutaraldehyde at pH
7.4, with 5 % sucrose. For 1 L of FSG, dissolve 20 g of parafor-
maldehyde in 700 mL of hot distilled water (70 °C) by stirring.
The solution will appear milky. Add 0.2 M NaOH dropwise until
the solution becomes transparent. Add 200 mL of CSS and filter
the solution. Add 50 mL of 25 % glutaraldehyde. Finally, add
50 g of sucrose to the solution and make up to 1 L with distilled
water. Adjust the pH to 7.4 and store at 4 °C until use.
4. Cacodylate washing buffer with sucrose (CWBS): 0.1 M caco-
dylate buffer at pH 7.4, with 7.5 % sucrose. For 1 L, dissolve
75 g of sucrose in 700 mL of distilled water. Add 200 mL of
CSS, make up to 1 L with distilled water, and adjust the pH by
adding some drops of 0.2 M NaOH. Store at 4 °C.
2.2 Solutions 1. Substrate: 5′-inosine-diphosphate sodium salt (Sigma-Aldrich,

for the Nucleoside- I4375; molecular weight 428.19) stored at −20 °C.
Diphosphatase 2. Trizma maleate buffer (TMB) stock solution: 0.2 M Trizma
Histochemical maleate, pH 7.4. Dissolve 3.083 g of Trizma maleate (Sigma-
Reaction Aldrich, T3128: molecular weight 237.21) in 65 mL of dis-
tilled water and adjust the pH to 7.4. Store at room temperature
(RT) but do not use after 4 weeks.
3. Manganese (II) chloride solution: Dissolve 0.25 g of MnCl2 in
50 mL of distilled water to obtain a 0.5 % solution. Store at RT.
4. Lead (II) nitrate solution: Dissolve 0.4 g Pb(NO3)2 in 40 mL
of distilled water to obtain a 1 % solution. Store at RT.
5. Ammonium sulfide solution: Add 5 mL of 20 % (NH4)2S to
50 mL of distilled water to obtain a 2 % ammonium sulfide
solution. Prepare fresh just before use.
6. Silver nitrate solution: Dissolve 0.5 g Ag(NO3) in 50 mL of dis-
tilled water to obtain a 1 % solution. Prepare fresh just before use.
2.3 Solutions 1. Acetic acid solution (AAS): Add 2.75 mL of glacial acetic acid
for Toluidine Blue (≥99.7 %) to 240 mL of distilled water to obtain a 0.2 M
Counterstaining solution.
2. Sodium acetate solution (SAS): Dissolve 2.63 g of sodium ace-
tate (molecular weight 82.03) in 160 mL of distilled water to
obtain a 0.2 M solution.
3. Walpole buffer (WB): 0.2 M Walpole buffer solution, pH 4.5.
Mix three parts of AAS with two parts of SAS and adjust the
pH by adding some drops of the 0.2 M AAS.
4. Toluidine blue solution: Dissolve 0.1 g of toluidine blue in
100 mL of WB in order to obtain a 0.1 % solution. Filter and
store at RT.
2.4 Solutions 1. Cacodylate buffer (CB): 0.2 M cacodylate buffer, pH 7.4. Mix
for Resin-Embedding 40 mL CSS with 60 mL of distilled water. Adjust the pH and
Procedure for store at 4 °C.
Electron Microscopy 2. Osmium tetroxide solution (OTS): 1 % osmium tetroxide
solution in 0.1 M cacodylate buffer. Mix 5 mL of 2 % OsO4
aqueous solution (Electron Microscopy Sciences, 19172) with
5 mL CB. Store at 4 °C in the dark.
3. Uranyl acetate solution (UAS): 2 % uranyl acetate solution in
70 % ethanol. Mix 2 g uranyl acetate powder with 100 mL
70 % ethanol. Filter and store at 4 °C.
4. Araldite I: Mix 10 g of component AM epoxy resin (Sigma-
Aldrich ACM Fluka, 44611) with 10 g of component B
(Sigma-Aldrich ACM Fluka, 44612) and 0.3 g of component
D (Sigma-Aldrich ACM Fluka, 44614) in a glass container.
Mix well using a glass rod and maintain in the oven at 55 °C.
Before use, be sure that there are no air bubbles in the mixture.
Araldite I should be freshly prepared 1–2 h before starting the
embedding process (see Note 1).
5. Araldite II: Mix the same components used to prepare Araldite
I and add 0.3 g of component C (Sigma-Aldrich ACM Fluka,
44613). Mix well with a glass rod and stand in the oven at
55 °C. Prepare fresh just before use. Do not use until all air
bubbles are gone (see Note 1).
2.5 Solutions for 1. Tris-buffered solution (TB): 0.05 M tris-buffered solution, pH

Double Labelling: 7.4. Add 12.11 g of Trizma base (Sigma-Aldrich, T1503;
NDPase Histochemistry molecular weight 121.14) and 7.2 mL of 37 % chlorhydric acid
with GFAP solution (HCl) to 2 L of distilled water. Adjust the pH to 7.4
Immuno- and store at RT.
histochemistry 2. Tris-buffered saline solution (TBS): 0.05 M tris-buffered
saline solution, pH 7.4. Add 12.11 g of Trizma base (Sigma-
Aldrich, T1503; molecular weight 121.14), 17.53 g of NaCl
(Sigma-Aldrich, 71379; molecular weight: 58.44), and 7.2 mL

of 37 % HCl to 2 L of distilled water. Mix well, adjust the pH
to 7.4 and store at RT.
3. TBS with Triton X-100 (TBST): 0.05 M tris-buffered saline
solution with 1 % Triton X-100, pH 7.4. Add 10 mL of Triton
X-100 to 900 mL of TBS and mix well until the Triton X-100
is completely dissolved. Make up to 1 L with TBS. Store at RT.
Discard when the solution is not transparent.
4. Buffer blocking solution (BBS): 0.05 M TBST, pH 7.4 with
10 % fetal bovine serum (FBS), and 3 % bovine serum albumin
(BSA). Add 10 mL of FBS (Sigma-Aldrich, F7524) to 90 mL
of TBST pH 7.4 and dissolve 300 mg of BSA (Sigma-Aldrich,
A9647) in the previous solution. Mix well until the BSA is
completely dissolved. Prepare fresh just before use.
5. Primary antibody: Polyclonal rabbit anti-GFAP antibody
(DAKOPatts, Z0334).
6. Secondary antibody: Biotinylated anti-rabbit Ig (affinity puri-
fied) antibody from goat (Vector Laboratories, BA-1000).
7. Horseradish peroxidase streptavidin (Vector Laboratories,
SA-5004).
8. 1-Naphthol working solution: 1-Naphthol solution with
0.033 % hydrogen peroxide. Dissolve 50 mg of 1-naphthol
(Sigma-Aldrich, N1000; molecular weight 144.17) in 0.5 mL
ethanol. Mix this solution with 10 mL of 1 % ammonium car-
bonate and 89.5 mL of TBS. After filtering the solution, add
0.01 mL 30 % hydrogen peroxide. Prepare fresh just before use.
9. Azur A working solution: 0.05 % Azur A in 0.05 M TBS pH 8.
Dissolve 0.05 g of Azur A chloride (Sigma-Aldrich, 861049)
in 100 mL of TBS. Adjust the pH to 8.00 with some drops of
0.2 M NaOH.
3 Methods
3.1 Intracardiac 1. Anesthetized animals (see Note 2) are intracardially perfused

Perfusion, Postfixation, with the FS (FSG for TEM analysis) at 4 °C for 10 min (see
and Vibratome Note 3). In case of human samples from biopsies or autopsies,
Sections the samples should be immediately submerged in FS (or FSG)
at 4 °C. The sample size should be not thicker than 1 cm for
best results.
2. The areas of interest (telencephalon, cerebellum, spinal cord,
and others) should be dissected out as quickly as possible and
immersed in glass vials containing the same fixative solution
(4 °C) to complete a total time of 4 h of fixation for light
microscopy (1 h for TEM analysis). Take into account that
Fig. 5 Diagrams showing the sample handling previous to vibratome sectioning. (a) Sample dissection during
postfixation time facilitates the penetration of the fixative solution and prepares the samples for ulterior vibratome
sectioning. (b) Tissue samples are attached to squares of filter paper using super glue gel. (c)The assembly is
attached to the vibratome platform and submerged in the vibratome container previously filled with the buffer
volume of FS (or FSG) should be at least ten times the volume

of the tissue pieces. Remove the meninges when possible.
Special care should be taken in order to prevent any desicca-
tion during all the process including the extraction of the brain
(see Note 4), and the subsequent dissection because otherwise
the NDPase enzymatic activity in the plasma membrane of
microglia may be completely destroyed.
3. Dissect out the samples to be cut with the vibratome (Fig. 5a).
This process can be done during the time of fixation (see Note 5).
4. Once completed the fixation, wash the samples in CWBS
(3 × 10 min, 4 °C) and store them in this buffer (4 °C) until
cutting (see Note 6).
5. Attach the piece to be cut to a small square of filter paper using
a super glue gel (see Note 7). To achieve that, place some drops
of glue depending on the size of the tissue sample, on the sur-
face of a square piece of filter paper to produce a little dome.
Immediately, with the aid of tweezers, take the piece of tissue
to be attached to the filter paper, and before placing on it,

quickly absorb the liquid from its basis by gently touching it
during 2 s with a strip of filter paper. This is important to assure
an appropriated attachment. Once the tissue piece is placed on
the center of the glue dome, press gently down with a flat
object (as the handle of the tweezers) to be sure that the basis
is attached as flat as possible to the filter paper. Wait for two
additional seconds and quickly take the square and the attached
tissue sample and immediately submerge the assembly in cold
(4 °C) CWBS until to be cut (see Notes 8 and 9) (Fig. 5b).
6. With the aid of some drops of super glue gel, attach the assem-
bly (square of the filter paper plus the tissue sample) with glue to
the vibratome platform and quickly submerge the platform with
the assembly in the specific container of the vibratome that
should be previously filled with cold (4 °C) CWBS (see Note
10) (Fig. 5c). Obtain sections 30–50 μm thick for optical micro-
scope analysis and 60–100 μm thick for resin-embedding and
ulterior transmission electron microscopy (TEM) studies. Free-
floating sections obtained should be progressively taken from
the vibratome container with the aid of a soft brush and intro-
duced in 5 mL glass vials filled with cold CWBS (4 °C) until the
histochemical reaction will be performed.
3.2 Histochemical 1. Mix 50 mL of 0.2 M TMB stock solution with 50 mL of dis-

Reaction for Single tilled water in order to obtain 0.1 M TMB solution.
NDPase Demonstration 2. Remove the CWBS from the glass vials with the aid of a Pasteur
and Ulterior pipette and immediately add enough 0.1 M TMB solution,
Counterstaining with usually 2 mL per vial. The TMB solution should always cover
Toluidine Blue for the free-floating sections within the glass vial.
Light Microscopy 3. Bring the glass vials into an oven at 37 °C for 15 min (mean-
while prepare the fresh incubation medium).
4. Prepare the incubation medium taking into account that you
will need around 2 mL of solution per vial containing around
20 vibratome sections. To prepare the medium (for around ten
vials) mix always the components in this order:
(a) Dissolve 25 mg of substrate in 7 mL of distilled water.
(b) Add 10 mL of 0.2 M TMB stock solution.
(c) Add 5 mL manganese (II) chloride solution.
(d) Add 3 mL lead (II) nitrate solution (see Note 11).
5. Filter the incubation medium. This should become mainly
transparent then.
6. Take into account that as negative control of the histochemical
reaction, some sections should be incubated in a medium con-
taining the same components but lacking the substrate.
7. Remove the TMB solution from the glass vials with a Pasteur
pipette and add the incubation medium.
8. Incubate at 37 °C in the oven for 20–25 min. Shake the vials
every 5 min to ensure a homogeneous histochemical reaction
in all the sections.
9. Stop the histochemical reaction by removing the incubation
medium and adding cold (4 °C) 0.1 M TMB for 1 min.
10. Wash 3 × 10 min with 0.1 M CWBS at RT (see Note 12).
11. Visualize the histochemical reaction by removing the CWBS
and adding ammonium sulfide solution at RT for 1 min.
Immediately the sections become brown, which is an indicative
that the histochemical reaction has worked properly.
12. Wash 3 × 3 min in distilled water.
13. Remove the distilled water and stabilize the histochemical
reaction product by adding to the vials the silver nitrate solu-
tion for 1 min at RT.
14. Wash 3 × 3 min in distilled water.
15. Mount the sections on gelatin-coated slides and allow to dry
for at least 24 h before applying the toluidine blue counter-
staining (see Note 13).
16. Submerge the slides in distilled water for 3 min.
17. Counterstain with toluidine blue for 1 min (see Note 14).
18. Wash with distilled water for 1 min to remove the excess of
colorant.
19. Dehydrate by submerging the slides in graded ethanol: 50, 70,
90, and 3× 100 %. No more than 30 s in each alcohol.
20. Differentiate with 1-butanol for 1 min.
21. Submerge in xylol (2× 2 min).
22. Finally, coverslip the slides using DPX as permanent mounting
medium.
23. Wait at least 24 h in order to allow complete polymerization of
the DPX before proceeding with the analysis at the light
microscope.
3.3 NDPase This technique can also be used for visualization of microglia at the
Histochemistry for ultrastructural level. To accomplish that, go ahead with the steps
Transmission Electron detailed in the previous part until step 10. As the reaction product
Microscopy of the histochemical reaction, that is to say, the lead phosphate, is
electrodense, this product will be already seen at the electron
microscope; however, it is not possible to see it in semithin sec-
tions. If microglia need to be analyzed in semithin sections, then it
is necessary to visualize the reaction product by treating the sec-
tions with ammonium sulfide solution (step 11) in order to obtain
lead sulfide that can be visualized with the light microscope.
Subsequently in both cases, follow the steps detailed below:

1. Mix 50 mL of 0.2 M of CB with 50 mL distilled water in order
to obtain 0.1 M CB.
2. Wash the free-floating sections by adding to the vials contain-
ing the free-floating sections, the 0.1 M CB (3 × 10 min).
3. Postfix the sections in OTS for 2 h at RT in the dark.
4. Wash sections 3 × 10 min in 0.1 M CB at 4 °C.
5. Remove the 0.1 M CB from the glass vials and start the dehy-
dration process with graded ethanol as indicated:
(a) 30 % ethanol for 15 min at 4 °C.
(b) 50% ethanol for 15 min at 4 °C.
(c) 70 % ethanol for 15 min at 4 °C.
6. En bloc stain with UAS for 1.5 h at 4 °C.
7. Wash in 70 % ethanol for 15 min at 4 °C.
8. Continue the dehydration as follows:
(a) 96 % ethanol for 15 min at 4 °C.
(b) 100 % ethanol 3 × 10 min at 4 °C.
9. Remove the last 100 % ethanol and add propylene oxide
3 × 10 min at 4 °C.
10. Start the embedding process by removing the propylene oxide
and adding to the glass vials a mixture of one part of Araldite I
and three parts of propylene oxide for 2 h at RT (see Note 15).
Prepare the mixture just before use.
11. Remove the previous mixture and add a new mixture of one
part of Araldite I and one part of propylene oxide for 2 h at
RT. Prepare the mixture just before use.
12. Again remove the previous mixture and add a new mixture of
three parts of Araldite I and one part of propylene oxide for
2 h at RT. Prepare the mixture just before use.
13. After removing the last mixture, add Araldite I to the glass vials
and maintain them in the oven at 55 °C overnight.
14. Change the Araldite I and maintain at 55 °C for 3 h. Meanwhile
prepare the Araldite II (see Subheading 2).
15. Remove the Araldite I and add Araldite II for 2 h at 55 °C.
16. Take the samples from the glass vials with the aid of a wood
spatula and carefully transfer the sections to silicone molds pre-
viously filled with Araldite II. Make sure that all sections are
completely flat and correctly oriented.
17. Maintain the silicone molds with the sections in the oven at
65 °C for 48 h to ensure the correct polymerization of the resin.
3.4 Double Labelling Simultaneous demonstration of microglial and astroglial cells can
Technique for Light be achieved easily by combining NDPase histochemistry with
Microscopy Combining GFAP immunohistochemistry through three sequential steps:
NDPase (1) incubation for histochemical demonstration of NDPase activity,
Histochemistry (2) incubation for immunohistochemical detection of GFAP, and
and GFAP (3) visualization of the histochemical and the immunohistochemi-
Immunohistochemistry cal reaction products for light microscopy.
1. Incubation for histochemical demonstration of NDPase activ-
ity: Follow the steps 1–10 previously specified in
Subheading 3.2 (see Note 16).
2. Incubation for the immunohistochemical detection of GFAP:
(a) Remove the CWBS from the glass vials and wash the sec-
tions 3 × 10 min with TBS at RT.
(b) Wash 3 × 10 min with TBST at RT.
(c) Block the unspecific binding with BBS for 1 h at RT.
(d) Incubate with the primary GFAP antibody diluted in BBS
(1:1,800) overnight at 4 °C (see Note 17).
(e) Wash the sections 3 × 10 min in TBST at RT.
(f) Incubate sections in the secondary biotinylated anti-rabbit
Ig antibody diluted in BBS (1:500) for 1 h at RT.
(g) Wash the sections 3 × 10 min in TBST at RT.
(h) Incubate in the horseradish peroxidase streptavidin diluted
in BBS (1:500) for 1 h at RT.
(i) Wash the sections 3 × 10 min in TBS at RT.
3. Visualization of the histochemical and the immunohistochemi-
cal reaction products for light microscopy. To achieve a suc-
cessful visualization of the double staining, the final steps
should be executed in this specific sequence:
(a) Remove TBS from glass vials and incubate 15 min in
1-naphthol working solution at RT (see Note 18).
(b) Wash the sections 3 × 10 min in TBS at RT.
(c) Visualize the NDPase histochemistry by incubating the sec-
tions with 2 % ammonium sulfide solution for 1 min at RT.
(d) Wash the sections 3 × 10 min in TBS at RT.
(e) Remove the TBS and add the Azur A working solution for
30 min at RT.
(f) Wash the sections 4 × 10 min with TBS at RT with con-
tinuous stirring.
(g) Wash 2 × 5 min in distilled water.
(h) Stabilize the histochemical reaction product with 1 % silver
nitrate solution for 1 min at RT.
(i) Wash 2 × 5 min in distilled water.

(j) Mount the sections on gelatin-coated slides and allow dry
for 24 h at RT.
(k) Dehydrate by submerging the slides in graded ethanol: 50,
70, 90, and 3× 100 %. No more than 30 s in each
alcohol.
(l) Submerge in xylol (2× 2 min).
(m) Finally, coverslip the slides using DPX as permanent
mounting medium.
(n) Wait for at least 24 h in order to allow complete polymer-
ization of the DPX before proceeding with the analysis at
the light microscope.
3.5 Cytochemical Visualization of microglia in cell cultures can also be performed by

Demonstration of demonstration of NDPase enzymatic activity as follows:
NDPase in Cell
1. Fixation should be done by removing the cell culture medium
Cultures for Microglia from culture plates or flasks and adding enough FS at 4 °C to
Visualization cover the cells in the culture.
2. Fix the cultures for 2 h at 4 °C and then wash in CWBS (3×
10 min, 4 °C).
3. Follow the steps indicated in Subheading 3.2 from 1 to 14
taking into account that you are not using sections mounted
on slides but cell plates; therefore, all changes should be done
accordingly (see Note 19).
4. If you want to counterstain with toluidine blue, add toluidine
blue and stain for 1 min, then wash with distilled water for
1 min.
5. Follow the steps indicated in Subheading 3.2 from 19 to 23
(see Note 19).
6. In the case you want to demonstrate the presence of microglia
and astroglia combining the NDPase visualization with the
GFAP immunostaining, follow the steps indicated in
Subheading 3.4. Again, take into account that you are not
using sections mounted on slides but cell plates (see Note 19).
4 Notes
1. Take into account how much volume you will need in function
of the number of changes to do, samples, volume of glass vials,
and volume of the silicon molds to be used.
2. Animals can be anesthetized using any anesthetic as isofluor-
ane, pentobarbital, ketamine/xylazine, and others.
3. Ideally the perfusion in rats and mice should be done by insert-

ing the canule through the left ventricle until reaching the
aorta and tie up the blood vessel with a forceps. Then make a
cut in the right auricula and open the flux of the fixative. To
assure that the cold (4 °C) fixative does not cause any undesir-
able vasoconstriction, it is necessary to use an appropriated
device that provides a constant pressure and enough flux of the
FS (or FSG) to push out the blood during the first 10 s of
perfusion.
4. After perfusion, during bone removal and brain extraction, it is
important to maintain the exposed brain surfaces always wet by
adding some drops of FS (or FSG) with a Pasteur pipette.
5. To facilitate the dissection of the samples, it is recommendable
to do it after some time (30 min) of fixation. Dissections can
be done in a Petri dish using a razor blade and, if necessary,
under the binocular lens. Tissue pieces should be covered all
the time by the FS (or FSG) at 4 °C, in order to prevent desic-
cation. The size of the pieces should be enough to be appro-
priately cut with the vibratome. Coronal sections 5–6 mm
thick of a whole rat brain may be obtained and cut without
problems. Preferably the basis of the pieces should be as flat as
possible to assure a good and stable attachment to the vibra-
tome. Once the samples are dissected out, they should be
transferred back from the Petri dish to the glass vials to com-
plete the total time of fixation (4 h for light microscopic study
and 1 h for TEM) established in Subheading 3.
6. Best results are obtained when samples are cut on the same day
and immediately stained. However, enzymatic activity can still
remain during at least the next 24 h. If it is not possible to do
the enzymatic reaction on the same day of perfusion, it is rec-
ommendable to keep the dissected samples in CWBS (4 °C)
and to obtain the vibratome sections in the following day.
7. Although any super glue can be used, the gel type is the most
recommendable because it does not spread when placed in the
filter paper and it holds better the tissue piece, especially if the
piece is not completely flat in the base.
8. If there are many pieces, they can be attached to different filter
paper squares and maintain them submerged in cold CWBS
until they are cut.
9. All the process should be as fast as possible to minimize the
possible desiccation of the samples. Do your best to ensure
that the time spent in attaching the samples to the filter paper
is only some few seconds.
10. If the vibratome device you use does not have a container
where temperature can be regulated, then change frequently
the CWBS, or if the vibratome device allows it, add some ice
around the container to maintain cold the CWBS.
11. When lead (II) nitrate is added to the incubation medium, it
becomes whitish.
12. If necessary, sections can be stored in 0.1 M CWBS at 4 °C for
some hours before treatment with ammonium sulfide.
13. To check if the histochemical NDPase reaction has been suc-
cessful, add a drop of glycerine to one of the slides, coverslip,
and have a look in the microscope. Profiles of microglia and
blood vessels should be distinguished.
14. Toluidine blue solution for counterstaining can be reused, but
it is important to filter always the solution before use. Note
that the time of counterstaining should be adjusted in every
reuse; therefore, try first with one slice to determine the appro-
priated time.
15. From now, remove the lids of the glass vials in order to facili-
tate the evaporation of propylene oxide.
16. Visualization and stabilization of the reaction with ammonium
sulfide and silver nitrate are omitted now, as these reagents may
interfere with antigenicity. However, to check the histochemi-
cal reaction, some sections can be immediately treated with
ammonium sulfide to visualize the reaction product.
17. As negative control of the immunohistochemical reaction,
some sections should be processed in parallel without the pri-
mary antibody.
18. 1-Naphthol basic dye method of Mauro et al. [22] has been
chosen for demonstration of the horseradish peroxidase in the
immunohistochemical staining for GFAP, because this gives a
blue HRP reaction product that was easily distinguished from
the brownish-black NDPase staining.
19. The addition and removal of the different reagents to the cell
plates should be done carefully using a Pasteur pipette. It has to
pay special care not to touch the cell culture with the pipette tip.
References
1. Rio HP (1918) Noticia de un nuevo y fácil método 4. MurabeY,SanoY(1981)Thiaminepyrophosphatase
para la coloración de la neuroglia y del tejido con- activity in the plasma membrane of microglia.
juntivo. Trab Lab Invest Biol 15:367–378 Histochemistry 71:45–52
2. Rio HP (1920) El tercer elemento de los cen- 5. Murabe Y, Sano Y (1982) Morphological
tros nerviosos. I. La microglía en estado nor- studies on neuroglia V. Microglial cells in the
mal. Bol Soc Esp Biol 8:68–82 cerebral cortex of the rat, with special refer-
3. Novikoff AB, Goldfisher S (1961) ence to their possible involvement in synaptic
Nucleosidediphosphatase activity in the Golgi function. Cell Tissue Res 223:493–506
apparatus and its usefulness for cytological 6. Yamazaki M, Hayaishi O (1968) Allosteric
studies. Proc Natl Acad Sci 47:802–810 properties of nucleoside diphosphatase and its
identity with thiamine pyrophosphatase. J Biol and postnatal rabbit retina. J Comp Neurol
Chem 243–2934 282:249–263
7. Sano S, Matsuda Y, Miyamoto S et al (1984) 15. Dalmau I, Finsen B, Zimmer J et al (1998)
Thiamine pyrophosphatase and nucleoside Development of microglia in the postnatal rat
diphosphatase in rat brain. Biochem Biophys hippocampus. Hippocampus 8:458–474
Res Commun 118:292–298 16. Dalmau I, Vela JM, González B et al (2003)
8. Salvador-Silva M, Vidal-Sanz M, Villegas- Dynamics of microglia in the developing rat
Pérez MP (2000) Microglial cells in the retina brain. J Comp Neurol 458:144–157
of Carassius auratus: effects of optic nerve 17. Jørgensen MB, Finsen BR, Jensen MB et al
crush. J Comp Neurol 417:431–447 (1993) Microglial and astroglial reactions in
9. Castellano B, González B, Dalmau I et al ischemic and kainic acid induced lesions in the
(1991) Identification and distribution of adult rat hippocampus. Exp Neurol 120:70–88
microglial cells in the cerebral cortex of the liz- 18. Vostrikov VM (1985) Electron-cytochemical
ard: a histochemical study. J Comp Neurol study of microglia in Alzheimer’s disease and
311:434–444 senile dementia. ZhNevropatol PsikhiatrIm S
10. Lopez-Garcia C, Nacher J, Castellano B et al SKorsakova 85:974–976
(1994) Transitory disappearance of microglia 19. Vela JM, Dalmau I, González B et al (1998)
during the regeneration of the lizard medial Glial abnormalities in genetically determined
cortex. Glia 12:52–61 disorders of myelin. In: Castellano B, González
11. Jensen MB, González B, Castellano B et al B, Nieto M (eds) Understanding glial cells,
(1994) Microglial and astroglial reactions to Chapter 18. Kluwer Academics Publisher,
anterograde axonal degeneration: A histo- Boston, pp 363–384
chemical and inmunocytochemical study of 20. Dalmau I, Castellano B, Pedersen EB et al
the adult rat fascia dentata after entorhinal per- (1996) Reduction of the microglial cell number
forant path lesions. Exp Brain Res in rat primary glial cell cultures by exogenous
98:245–260 addition of dibutyryl cyclic adenosine mono-
12. Almolda B, Costa M, Montoya M et al (2009) phosphate. J Neuroimmunol 70:123–129
CD4 microglial expression correlates with 21. Castellano B, González B, Jensen MB et al
spontaneous clinical improvement in the acute (1991) A double staining technique for
Lewis rat EAE model. J Neuroimmunol simultaneous demonstration of astrocytes and
209:65–80 microglia in brain sections and astroglial cell
13. Vela JM, Dalmau I, González B et al (1995) cultures. J Histochem Cytochem
Morphology and distribution of microglial 39:561–568
cells in the young and adult mouse cerebellum. 22. Mauro A, Germano I, Giaccone G et al (1985)
J Comp Neurol 361:602–616 1-Naphthol basic dye (1-NBD). An alternative
14. Schnitzer J (1989) Enzyme-histochemical to diaminobenzidine (DAB) in immuno peroxi-
demonstration of microglial cells in the adult dase techniques. Histochem 83:97
Chapter 23
Tomato Lectin Histochemistry for Microglial Visualization

Nàdia Villacampa, Beatriz Almolda, Berta González,
and Bernardo Castellano
Abstract
The use of different lectins for the study of microglial cells in the central nervous system (CNS) is a valuable
tool that has been extensively used in the last years for the selective staining of this glial cell population,
not only in normal physiological conditions, but also in a wide range of pathological situations where the
normal homeostasis of the parenchyma is disturbed. In this chapter we accurately describe the methodol-
ogy for the selective labelling of microglial cells by using the tomato lectin (TL), a protein lectin obtained
from Lycopersicum esculentum with specific affinity for poly-N-acetyl lactosamine sugar residues which are
found on the plasma membrane and in the cytoplasm of microglia. Here we describe how to perform this
technique on vibratome, frozen, and paraffin sections for optical microscopy, as well as for transmission elec-
tron microscopy (TEM) studies. Using this methodology it is possible to visualize amoeboid microglia in
the developing brain, ramified microglia in the adult, and activated/reactive microglia in the experimen-
tally damaged brain. In addition, as TL also recognized sugar residues in endothelial cells, this technique
is very useful for the study of the relationship established between microglia and the CNS vasculature.
Key words Microglia, Histochemistry, Tomato lectin, Transmission electron microscopy, Double
labelling, Fluorescence
Abbreviations
AAS Acetic acid solution
BBS Buffer blocking solution
CNS Central nervous system
DS Developing solution
DSN Developing solution with nickel
DSNC Developing solution with nickel-cobalt-enhanced DAB
EPBS Endogenous peroxidase blocking solution
FS Fixative solution
FSG Fixative solution with glutaraldehyde
GFAP Glial fibrillary acidic protein
In memoriam of Laia Acarin Pérez-Simó (1970–2011)
261
262 Nàdia Villacampa et al.
GSA-IB4 Griffonia simplicifolia B4 isolectin

MHC Major histocompatibility complex
ML-1 Mistletoe lectin
OS Olmos solution
OTS Osmium tetroxide solution
PBS Phosphate buffer solution
PWB Phosphate washing buffer
RCA-1 Ricinus communis agglutinin-1
SAS Sodium acetate solution
SS Sucrose solution
TB Tris-buffered solution
TBS Tris-buffered saline solution
TBST TBS with Triton X-100
TEM Transmission electron microscopy
TL Tomato lectin
UAS Uranyl acetate solution
WB Walpole buffer
WGA Wheat germ agglutinin
1 Introduction
Lectins are sugar-binding proteins that have been of great interest

for their use in the characterization of the membranes composition
and the identification of different cell types [1]. In the central ner-
vous system (CNS), several lectins have been used as selective
microglial markers. Some of them are wheat germ agglutinin
(WGA), mistletoe lectin (ML-1), Ricinus communis agglutinin-1
(RCA-1), Griffonia simplicifolia B4 isolectin (GSA-IB4), tomato
lectin (TL) [2], and more recently galectins [3]. The TL specifi-
cally recognizes the poly-N-acetyl lactosamine residues, a sugar
found in the plasma membrane and cytoplasm of microglia and
endothelial cells [2] (Fig. 1).
Since its first description by Acarin et al. in 1994, the use of TL
as microglial marker has been a useful tool for the characterization of
this glial cell population in the rodent adult normal brain during
development [4, 5] and aging [6] but also after different experimen-
tal injuries, such as excitotoxicity [7–10], aspiration lesions [11],
experimental autoimmune encephalomyelitis [12], as well as for
microglial studies in genetically modified animals such as myelin
mutants [13, 14] and transgenic mice [15]. In addition to microglia,
TL histochemistry also visualized the blood vessels allowing the study
of the relationship between microglial cells and vasculature. As acti-
vated microglia in addition to changes in morphology also increase
the expression of N-acetyl lactosamine residues, TL histochemistry
easily demonstrates the reactive changes in this glial cell population.
Methodologically, the use of TL as microglial marker has some
advantages in front of other techniques such as immunodetection
of specific antigens and those based on the detection of specific
Tomato Lectin in Microglia 263
Fig. 1 Histochemical demonstration of tomato lectin. TL is a carbohydrate-binding protein that recognizes

specifically poly-N-acetyl lactosamine residues located on the plasma membrane and cytoplasm of both
microglial cells (M) and endothelial cells of the blood vessels (BV). To visualize these poly-N-acetyl lactosamine
residues, biotinylated TL is used. Afterwards, biotin is recognized by a streptavidin-horseradish complex added
to the samples, which in turn carries conjugated the peroxidase enzyme. Finally, visualization of the reaction
product is performed by incubating the samples with 3,3′-diaminobenzidine (DAB). In presence of peroxidase
enzyme and hydrogen peroxide (H2O2), DAB will be oxidated and quickly precipitated in the place where peroxi-
dase enzyme is located, showing the characteristic brown staining observed in the picture
phosphatases such as PNPase [16] and NDPase enzymatic histo-

chemistry (see Chapter 22), because sugar residues are very stable
and therefore can be easily detected after a wide range of fixation
and histological processing. Therefore TL technique can be per-
formed on both frozen sections and in sections obtained from
paraffin-embedded tissue. Moreover, TL histochemistry not only
allows the specific staining of microglia in histological sections but
also has been used for the characterization of microglia in organo-
typic cultures [17, 18]. Finally, we want to point out that TL
histochemistry can also be used for the analysis of microglia at the

electron microscope level [2].
In this chapter we described the detailed methodology to
perform the TL histochemistry method in vibratome, paraffin, and
cryostat sections for the identification of microglial cells for both
light (Fig. 2a–d) and fluorescence microscopy (Fig. 3a). Moreover,
we give the details to combine this technique with any immunohis-
tochemical labelling, and as example we have detailed how to com-
bine the TL histochemistry with the immunodetection of glial
fibrillary acidic protein (GFAP) for specific visualization of astro-
cytes (Fig. 2e, f) and immunodetection of major histocompatibility
complex class I (MHC-I), a molecule involved in antigen presenta-
tion (Fig. 3b). Finally we describe how to process vibratome
TL-stained samples for electron microscopy studies (Fig. 4).
2 Materials
2.1 Solutions for 1. Phosphate buffer stock solution (PBS): 0.2 M phosphate buf-
Fixation and Washing fer at pH 7.4. Mix 6.6 g of NaH2PO4⋅2H2O (molecular weight:
156.01) and 43.58 g of Na2HPO4⋅7H2O (molecular weight:
268.07) in 700 mL of distilled water. Make up to 1 L with
distilled water and adjust the pH by adding some drops of
0.2 M NaOH. Store at 4 °C.
2. Fixative solution (FS): 4 % paraformaldehyde in 0.1 M phos-
phate buffer at pH 7.4. For 1 L of FS, dissolve 40 g of parafor-
maldehyde in 500 mL of hot distilled water (70 °C) by stirring.
The solution will appear milky. Add 0.2 M NaOH dropwise
until the solution becomes transparent. Add 500 mL of PBS
and filter the solution. Adjust the pH to 7.4 and store at 4 °C
until use during the following hours.
3. Phosphate washing buffer (PWB): phosphate buffer 0.1 M at
pH 7.4. Mix 500 mL of PBS with 500 mL distilled water.
Store at 4 °C.
4. Sucrose solution (SS): sucrose 30 % in phosphate buffer 0.1 M
at pH 7.4. Dissolve 90 g of sucrose in 200 mL of PWB and
wait until sucrose is completely dissolved. Make up to 300 mL
with PWB. Store at 4 °C.
5. Fixative solution with glutaraldehyde (FSG): 2 % paraformal-
dehyde in 0.1 M phosphate buffer and 0.5 % glutaraldehyde at
pH 7.4. For 1 L of FSG, dissolve 20 g of paraformaldehyde in
700 mL of hot distilled water (70 °C) by stirring. The solution
will appear milky. Add 0.2 M NaOH dropwise until the solu-
tion becomes transparent. Add 200 mL of PBS and filter the
solution. Add 50 mL of 25 % glutaraldehyde. Finally, make up
to 1 L with distilled water and adjust the pH to 7.4. Store at
4 °C until use during the following hours.
Fig. 2 Microglial labelling using tomato lectin histochemistry for light microscopy. Photographs showing
TL-labelled microglial cells (arrows) and blood vessels (BV) in vibratome sections (a, b), cryostat sections
(c), and paraffin sections (d). Note that vibratome sections (a, b) have been counterstained with toluidine blue.
(e) Double labelling combining TL histochemistry and GFAP immunohistochemistry allows visualization of
microglial cells in brown (arrows) and astrocytes in bluish black (arrowheads). (f) Magnification of the same
area shown in (e) where microglial cells in brown (arrow) and astrocytes in bluish black (arrowhead) can be
seen with more detail
Fig. 3 Microglial labelling using tomato lectin for fluorescence microscopy. (a) Photograph showing TL-labelled
microglial cells (arrow ) and blood vessels (BV) in the spinal cord of adult rats. (b) Double fluorescence labelling
combining TL (green ) and MHC-I (red ) showing co-localization (arrows ) of both markers in some activated/
reactive microglial cells found in the spinal cord of EAE-induced rats. Note that blood vessels (BV) do not
expressed MHC-I
2.2 Solutions 1. Olmos solution (OS): 20 % sucrose, 1 % polyvinylpyrrolidone

for Cryopreserving (PVP), and 30 % ethylene glycol in distilled water. Dissolve
200 g of sucrose in 400 mL of distilled water. Add 300 mL
ethylene glycol to the solution. Dissolve 10 g of PVP (Sigma-
Aldrich, PVP40, molecular weight: 40) and wait until com-
pletely dissolved (see Note 1). Finally, make up to 1 L with
distilled water and store at −20 °C. The solution should not
freeze.
2.3 Solutions 1. Acetic acid solution (AAS): Add 2.75 mL of glacial acetic acid
for Toluidine Blue (≥99.7 %) to 240 mL of distilled water to obtain a 0.2 M
Counterstaining solution.
2. Sodium acetate solution (SAS): Dissolve 2.63 g of sodium ace-
tate (molecular weight 82.03) in 160 mL of distilled water to
obtain a 0.2 M solution.
3. Walpole buffer (WB): 0.2 M Walpole buffer solution, pH 4.5.
Mix three parts of AAS with two parts of SAS and adjust the
pH by adding some drops of the 0.2 M AAS.
4. Toluidine blue solution: Dissolve 0.1 g of toluidine blue in
100 mL of WB in order to obtain a 0.1 % solution. Filter and
store at RT.
Fig. 4 Tomato lectin histochemistry for TEM studies. Semithin sections reacted with TL and counterstained
with toluidine blue of (a) adult rat brain showing resting microglial cells (arrows) and (b1, b2) 5-day-old rat
brain showing strongly stained amoeboid microglial cells (arrows) and blood vessels (BV). (c) Electron micro-
photograph showing amoeboid microglial cells from 5-day-old rat. TL binding shows an intracellular localiza-
tion, whereas intracytoplasmic vacuoles (V) remain unstained (Reproduced from ref. 2). (d) Electron micrograph
showing a blood vessel with intense TL staining (arrowhead), whereas pericytes (P) remain TL-negative
(Reproduced from ref. 2)
2.4 Solutions for 1. Tris-buffered solution (TB): 0.05 M tris-buffered solution,

Single Labelling with pH 7.4. Add 12.11 g of Trizma base (Sigma-Aldrich, T1503;
TL Histochemistry and molecular weight 121.14) and 7.2 mL of 37 % chlorhydric acid
Double Labelling solution (HCl) to 1.8 L of distilled water. Mix well and make
Combining TL up to 2 L. Adjust the pH at 7.4 and store at RT.
Histochemistry with 2. Tris-buffered saline solution (TBS): 0.05 M tris-buffered saline
Either GFAP or MHC-I solution, pH 7.4. Add 12.11 g of Trizma base (Sigma-Aldrich,
Immunohisto- T1503; molecular weigh 121.14), 17.53 g of NaCl (molecular
chemistry weight 58.44), and 7.2 mL of 37 % HCl to 1.8 L of distilled
water. Mix well, make up to 2 L, adjust the pH at 7.4, and
store at RT.
3. TBS with Triton X-100 (TBST): 0.05 M tris-buffered saline
solution with 1 % Triton X-100, pH 7.4. Add 10 mL of Triton
X-100 to 900 mL of TBS and mix well until the Triton X-100
is completely dissolved. Make up to 1 L with TBS. Store at RT.

Discard when the solution is not transparent.
4. Endogenous peroxidase blocking solution (EPBS): 70 % meth-
anol solution with 0.66 % hydrogen peroxide in distilled water.
Mix 70 mL methanol with 28 mL distilled water. Finally add
2 mL 30 % hydrogen peroxide. Prepare fresh just before use.
5. Buffer blocking solution (BBS): 0.05 M TBST, pH 7.4 with
10 % fetal bovine serum (FBS) and 3 % bovine serum albumin
(BSA). Add 10 mL of FBS (Sigma-Aldrich) to 90 mL of TBST
pH 7.4. Add to this solution 300 mg of BSA (Sigma-Aldrich).
Mix well until the BSA is completely dissolved. Prepare fresh
just before use.
6. Biotinylated TL from Lycopersicum esculentum (Sigma-
Aldrich, L-9389).
7. Primary polyclonal rabbit anti-GFAP antibody (DAKOPatts,
Z0334).
8. Primary monoclonal mouse anti-MHC-I antibody (AbD
Serotec, MCA51G).
9. Secondary biotinylated anti-rabbit Ig (affinity purified) anti-
body from goat (Vector Laboratories, BA-1000).
10. Secondary Alexa Fluor® 555 donkey anti-mouse IgG
(Molecular Probes, A-31570).
11. Horseradish peroxidase streptavidin (Vector Laboratories,
SA-5004).
12. Streptavidin-Alexa Fluor® 488 conjugate (Molecular Probes,
S-32354).
13. Developing solution (DS): 0.5 mg/mL 3,3′-diaminobenzi-
dine solution with 0.011 % hydrogen peroxide in TBS. Dissolve
50 mg 3,3′-diaminobenzidine (Sigma-Aldrich, ≥96 %, molec-
ular weight 360.11) in 100 mL TBS. Mix well and add 33 μL
30 % hydrogen peroxide. Prepare fresh just before use.
14. Developing solution with nickel (DSN): 0.5 mg/mL
3,3′-diaminobenzidine solution with 0.011 % hydrogen perox-
ide and 0.026 g/mL nickel ammonium sulfate in TBS. Add
2.6 g nickel ammonium sulfate (Sigma-Aldrich, 03750, molec-
ular weight 62.07) to 100 mL of DS.
15. Developing solution with nickel-cobalt-enhanced DAB
(DSNC): 0.5 mg/mL 3,3′-diaminobenzidine solution with
0.011 % hydrogen peroxide, 0.02 % nickel ammonium sul-
fate, and 0.025 % cobalt chloride. Add 4 mL 1 % of nickel
ammonium sulfate (Sigma-Aldrich, molecular weight 62.07)
and 5 mL 1 % of cobalt chloride (Sigma-Aldrich) to 200 mL
of DS.
2.5 Solutions for 1. Osmium tetroxide solution (OTS): 1 % osmium tetroxide

Resin-Embedding solution in distilled water. Mix 5 mL of 2 % OsO4 aqueous
Procedure for Electron solution (Electron Microscopy Sciences, 19172) with 5 mL
Microscopy distilled water. Store at 4 °C in the dark.
2. Uranyl acetate solution (UAS): 2 % uranyl acetate solution in
70 % ethanol. Mix 2 g uranyl acetate powder with 100 mL
70 % ethanol. Filter and store at 4 °C.
3. Araldite I: Mix 10 g of component AM epoxy resin (Sigma-
Aldrich ACM Fluka, 44611) with 10 g of component B
(Sigma-Aldrich ACM Fluka, 44612) and 0.3 g of component
D (Sigma-Aldrich ACM Fluka, 44614) in a glass container.
Mix well using a glass rod and maintain in the oven at 55 °C.
Before use, be sure that there are not air bubbles in the mix-
ture. Araldite I should be freshly prepared 1–2 h before start-
ing the embedding process (see Note 2).
4. Araldite II: Mix the same components used to prepare Araldite
I and add 0.3 g of component C (Sigma-Aldrich ACM Fluka,
44613). Mix well with a glass rod and stand in the oven at
55 °C. Prepare fresh just before use. Do not use until all air
bubbles are gone (see Note 2).
3 Methods
3.1 Intracardiac 1. Anesthetized animals (see Note 3) are intracardially perfused with
Perfusion and the FS (FSG for TEM analysis) at 4 °C for 10 min (see Note 4).
Postfixation 2. The areas of interest (telencephalon, cerebellum, spinal cord,
and others) should be dissected out as quickly as possible and
immersed in glass vials containing the same fixative solution
(4 °C) to complete a total time of 4 h of fixation for light and
fluorescence microscopy (1 h for TEM analysis). Take into
account that volume of FS (or FSG) should be at least ten
times the volume of the tissue pieces. Remove the meninges
when possible. Special care should be taken in order to prevent
any desiccation during all the process including the extraction
of the brain (see Note 5).
3. Dissect out the samples to be cut. This process can be done
during the time of fixation (see Note 6).
3.2 Obtaining 1. Once completed the fixation, wash the samples in PWB
Vibratome Sections (3 × 10 min, 4 °C) and store them in this buffer (4 °C) until
cutting (see Note 7).
2. Attach the piece to be cut to a small square of filter paper
using a super glue gel (see Note 8). To achieve that, place
some drops of glue depending on the size of the tissue sample
on the surface of a square piece of filter paper to produce a
little dome. Immediately, with the aid of tweezers, take the

piece of tissue to be attached to the filter paper and, before
placing on it, quickly absorb the liquid from its basis by gen-
tly touching it during 2 s with a strip of filter paper. This is
important to assure an appropriated attachment. Once the
tissue piece is placed on the center of the glue dome, press
gently down with a flat object (as the handle of the tweezers)
to be sure that the basis is attached as flat as possible to the
filter paper. Wait for two additional seconds and quickly take
the square and the attached tissue sample and immediately
submerge the assembly in cold (4 °C) PWB until to be cut
3. With the aid of some drops of super glue gel, attach the assem-
bly (square of the filter paper plus the tissue sample) with glue to
the vibratome platform and quickly submerge the platform with
the assembly in the specific container of the vibratome, which
should be previously filled with cold (4 °C) PWB (see Note 11).
Obtain sections 30–50 μm thick for optical microscope analysis
and 50–80 μm thick for resin-embedding and ulterior transmis-
sion electron microscopy (TEM) studies. Free-floating sections
obtained should be progressively taken from the vibratome con-
tainer with the aid of a soft brush and introduced in 5 mL vials
filled either with cold PWB (4 °C) if the histochemical reaction
will be performed immediately after cutting or with OS to store
the sections at −20 °C in case of posterior use.
3.3 Obtaining 1. Once completed the fixation, wash the samples in cold PWB
Cryostat Sections (4 °C) (2 × 5 min).
2. Remove the PWB and submerge samples in SS for 48 h at 4 °C
(see Note 12).
3. Place the tissue in the correct orientation and glue on a filter
paper strip with Tissue-Tek® OCT™ Compound (Sakura,
4583) (see Note 13).
4. Carefully drop the assembly of the sample and the filter paper
into cold 2-methylbutane solution (Sigma-Aldrich, 320404,
molecular weight 72.15) (see Note 14). Freeze for 30–45 s
before inserting the sample into precooled storage tubes and
store at −80 °C.
5. With the aid of a cryostat, obtain sections 25–30 μm thick.
Free-floating sections obtained should be progressively taken
from the cryostat device with the aid of a soft brush and intro-
duced in 2 mL plastic vials filled with cold OS (−20 °C) (see
Note 15). Store at −20 °C until their posterior use.
3.4 Obtaining 1. After the 4 h of fixation, wash the samples in cold PWB (4 °C)
Paraffin Sections (2 × 5 min).
2. Submerge the samples in graded ethanols as follows:

(a) 50 % ethanol for 1 h at RT.
(b) 70 % ethanol for 1 h at RT.
(c) 95 % ethanol for 1 h at RT.
(d) 100 % ethanol 3 × 1 h at RT.
3. After the last ethanol, submerge samples in xylene, 2 × 1 h at RT.
4. Remove the xylene and with the aid of the tweezers put each
samples in a metal mold previously filled with hot paraffin
(58 °C). Maintain samples in paraffin for 2 h, changing the
paraffin after the first hour.
5. Embed samples in paraffin blocks (Fig. 5):
(a) Choose a new mold with an appropriate size for the tissue
(see Note 16).
(b) Dispense a small amount of molten paraffin in the mold.
(c) Transfer tissue into the mold with the aid of warm for-
ceps, taking into account that the side to be cut must face
the bottom of the mold (Fig. 5a).
(d) To hold the tissue in the correct position, transfer the mold
with the tissue to a cold plate, and gently press tissue flat.
Paraffin will solidify in a thin layer holding the tissue.
(e) Once the tissue is in the right position, place a labelled
tissue cassette on the top of the mold. Press firmly
(Fig. 5b).
(f) Gently add hot paraffin to the mold using the paraffin
dispenser. The face of the plastic cassette must be com-
pletely covered with paraffin (Fig. 5c).
(g) After 30 min, paraffin should be solidified. When the wax
is cold and hard, the paraffin block can be easily removed
from the mold and should not stick (see Note 17)
(Fig. 5d). In case of cracked wax or tissue not well aligned,
melt them again and start over. Paraffin blocks can be
stored at room temperature for years.
6. Obtain sections 10 μm thick with the aid of a microtome and
collect them on gelatine-coated slides. Keep the slides stored at
RT until their use for the staining procedure.
7. Dewax sections just before starting the TL staining procedure
following the next steps:
(a) Heat the slides in dry oven at 55–60 °C for 20 min.
(b) Immerse slides in xylene (3 × 10 min).
(c) Immerse slides in 100 % ethanol for 2 min.
(d) Immerse slides in 95 % ethanol for 1 min.
(e) Immerse slides in 70 % ethanol for 1 min.
(f) Immerse slides in 50 % ethanol for 1 min.
Fig. 5 Schematic representation of the different steps for the embedding of samples in paraffin blocks.
(a) Place tissue in a metal mold with a small amount of molten paraffin. Be sure the tissue is in the correct
orientation. (b) Transfer mold to a cold plate to allow hardening of paraffin and place the cassette on top of the
mold. (c) Move the mold to the cold plate and refill completely with molten paraffin using the paraffin dis-
penser. Be sure paraffin covers all the cassette holes. (d) After 30 min of cooling, remove the cassette with the
paraffin block from the metal mold
(g) Immerse slides in 30 % ethanol for 1 min.

(h) Immerse slides in 1× PBS (Gibco, 14190) for 2 min.
(i) Proceed with the staining procedure.
3.5 Histochemical 1. Wash the sections or the gelatine-coated slides with TBS at RT
Reaction Using Single (3 × 10 min).
Tomato Lectin 2. Incubate the sections or slides in EPBS at RT for 10 min (see
Labelling for Light Notes 18 and 19).
Microscopy and 3. Wash 3 × 10 min with TBST at RT.
Fluorescence
4. Incubate sections with the TL diluted in TBST (1:150) over-
night at 4 °C and afterwards 1 h at RT (see Note 20).
5. Wash the sections 3 × 10 min in TBST at RT.

6. Incubate sections with the horseradish peroxidase streptavidin
diluted in TBST (1:500) for 1 h at RT. In case of fluorescence
microscopy, incubate with the streptavidin-Alexa Fluor® 488 con-
jugate diluted in TBST (1:1,000) for 1 h at RT (see Note 21).
7. Wash the sections 3 × 10 min in TBS at RT.
8. Wash the sections 3 × 10 min in TB at RT.
9. For light microscopy, visualize the peroxidase reaction using the
DS. Incubate sections in the DS solution for exactly 3 min. Be
accurate over time in this step. For fluorescence microscopy, skip
this step. If desired in these fluorescence-processed sections, a
nuclei counterstaining can be performed by incubating the sec-
tions in 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI;
Sigma-Aldrich, D9542) in TB (1:10,000) for 3 min.
10. In case of vibratome and cryostat free-floating samples, mount
the sections on gelatine-coated slides. Allow dry for at least
24 h. In case of light microscopy, proceed with the toluidine
blue counterstaining following steps 11–18. In case of fluo-
rescence, skip steps from 11–13 and proceed with the dehy-
dration (step 14) and coverslip the slides as specified below
(steps 17 and 18).
11. Submerge the slides in distilled water for 3 min.
12. Counterstain with toluidine blue for 1 min (see Note 22).
13. Wash with distilled water for 1 min to remove the excess of
colorant.
90, and 3 × 100 %. No more than 30 s in each alcohol.
15. Differentiate with 1-butanol for 1 min.
16. Submerge in xylene (2 × 2 min).
medium (see Note 23).
the DPX before proceeding with the analysis at the light
microscope.
3.6 Double Labelling Double labelling can be visualized by using either light or fluores-
Technique Combining cence microscopy. In case of light microscopy, follow the steps 1,
Tomato Lectin 3, and 4; in case of fluorescence microscopy, follow the steps 2–4:
Histochemistry and
1. Incubation for the immunohistochemical detection of GFAP
Either GFAP or MHC-I for light microscopy:
Immunohisto-
(a) Remove the sections stored at −20 °C on Olmos solution
chemistry
or the previously dewax slides and wash 3 × 10 min with
TBS at RT.
(b) After washing, block the endogenous peroxidase incubat-

ing for exactly 10 min with the EPBS at RT (see Note 18).
(c) Remove EPBS and wash 3 × 10 min with TBST at RT.
(d) Block the unspecific binding with BBS for 1 h at RT.
(e) Incubate with the primary GFAP antibody diluted in BBS
(1:1,800) overnight at 4 °C and 1 h at RT (see Note 24).
(f) Wash the samples 3 × 10 min in TBST at RT.
(g) Incubate samples in the secondary biotinylated anti-rab-
bit Ig antibody diluted in BBS (1:500) for 1 h at RT.
(h) Wash the samples 3 × 10 min in TBST at RT.
(i) Incubate in the horseradish peroxidase streptavidin
diluted in BBS (1:500) for 1 h at RT.
(j) Wash the sections 3 × 10 min in TBS at RT.
(k) Visualize the peroxidase reaction incubating the samples
for 3 min in DSN at RT.
(l) Remove the DSN and wash sections 3 × 10 min in TB at RT.
(m) After washing, block the endogenous peroxidase incubat-
ing for exactly 10 min with the EPBS at RT.
2. Incubation for the immunohistochemical detection of MHC-I
for fluorescence microscopy:
(a) Remove the sections stored at −20 °C on Olmos solu-
tions or the previously dewax slides and wash 3 × 10 min
with TBS at RT.
(b) Wash 3 × 10 min with TBST at RT.
(c) Block the unspecific binding with BBS for 1 h at RT.
(d) Incubate with the primary MHC-I antibody diluted in
BBS (1:1,000) overnight at 4 °C (see Note 24).
(e) Wash the sections 3 × 10 min in TBST at RT.
(f) Incubate sections in the secondary Alexa Fluor® 555 don-
key anti-mouse IgG antibody diluted in BBS (1:1,000)
for 1 h at RT.
(g) Wash the sections 3 × 10 min in TBST at RT.
3. For both light and fluorescence microscopy, after the GFAP or
MHC-I immunohistochemistry demonstration, incubate for
the histochemical detection of TL. For this purpose, follow the
steps 3–9, specified previously in Subheading 3.5 taking into
account which type of microscopy will be used to visualize the
staining.
4. Finally, in case of cryostat or vibratome samples, mount the

sections on gelatine-coated slides and allow dry for at least
24 h. Skip this step in case of paraffin-processed samples.
90, and 3 × 100 %. No more than 30 s in each alcohol.
6. Submerge in xylene (2 × 2 min). Skip this step in case of fluo-
rescence microscopy, as the treatment with xylene can elimi-
nate the fluorescence signal.
medium.
the DPX before proceeding with the analysis at the microscope.
3.7 Tomato Lectin This technique can also be used for visualization of microglia at the
Histochemistry for ultrastructural level as follows:
Transmission Electron
1. Fixation must be done following the steps specified in
Microscopy Subheading 3.1 but using FSG instead of FS. To obtain the
sections to be processed, follow the steps detailed above in
Subheading 3.2.
2. Once 50–80 μm thick sections will be obtained, process the
samples for the histochemical demonstration of TL following
the steps indicated previously in Subheading 3.5 from 1 to 8
using the solutions needed for light microscopy.
3. For better visualization of the peroxidase reaction product
with the electron microscope, perform the step 9 specified in
Subheading 3.5 but using DSNC instead of DS.
4. Transfer the sections to glass vials and wash by adding 0.1 M
PWB (3 × 10 min) to the vials.
5. Postfix the sections in OTS for 2 h at RT in the dark.
6. Wash sections 3 × 10 min in 0.1 M PWB at 4 °C.
7. Remove the 0.1 M PWB from the glass vials and start the
dehydration process with graded ethanol as indicated:
(a) 30% ethanol for 15 min at 4 °C.
(b) 50% ethanol for 15 min at 4 °C.
(c) 70% ethanol for 15 min at 4 °C.
8. En bloc stain with UAS for 1.5 h at 4 °C.
9. Wash in 70 % ethanol for 15 min at 4 °C.
10. Continue the dehydration as follows:
(a) 96% ethanol for 15 min at 4 °C.
(b) 100% ethanol 3 × 10 min at 4 °C.
11. Remove the last 100 % ethanol and add propylene oxide
3 × 10 min at 4 °C.
12. Start the embedding process by removing the propylene oxide
and adding to the glass vials a mixture of one part of Araldite I
and three parts of propylene oxide for 2 h at RT (see Note 25).
Prepare the mixture just before use.
13. Remove the previous mixture and add a new mixture of one
part of Araldite I and one part of propylene oxide for 2 h at
RT. Prepare the mixture just before use.
14. Again remove the previous mixture and add a new mixture of
three parts of Araldite I and one part of propylene oxide for
2 h at RT. Prepare the mixture just before use.
15. After removing the last mixture, add Araldite I to the glass vials
and maintain them in the oven at 55 °C overnight.
16. Change the Araldite I and maintain at 55 °C for 3 h. Meanwhile
prepare the Araldite II (see Subheading 2).
17. Remove the Araldite I and add Araldite II for 2 h at 55 °C.
18. Take the samples from the glass vials with the aid of a wood
spatula and carefully transfer the sections to silicone molds pre-
viously filled with Araldite II. Make sure that all sections are
completely flat and correctly oriented.
19. Maintain the silicone molds with the sections in the oven at
65 °C for 48 h to ensure the correct polymerization of the resin.
20. Using an ultramicrotome, obtain either semithin (1–2 μm) or
ultrathin (300–500 Å) sections. Semithin sections can be visu-
alized with a conventional light microscope whereas for the
study of ultrathin sections a transmission electron microscope
should be used.
4 Notes
1. Take into account that polyvinylpyrrolidone is a polymer and is

difficult to dissolve. To make it easy, add gradually in 1 g por-
tions with constant stirring and wait until the first gram is com-
pletely dissolved to continue adding. Make sure there are no
lumps in the solution.
2. Take into account how much volume you will need in function
of the number of changes to do, samples, volume of glass vials,
and volume of the silicon molds to be used.
3. Animals can be anesthetized using any anesthetic as isoflurane,
pentobarbital, ketamine/xylazine, and others.
4. Ideally the perfusion in rats and mice should be done inserting
the cannula through the left ventricle until reaching the aorta
and tie up the blood vessel with a forceps. Then make a cut in
the right auricle and open the flux of the fixative. To assure
that the cold (4 °C) fixative does not cause any undesirable
vasoconstriction, it is necessary to use an appropriated device
that provides a constant pressure and enough flux of the FS (or
FSG) to push out the blood during the first 10 s of perfusion.
5. After perfusion, during bone removal and brain extraction, it is
important to maintain the exposed brain surfaces always wet by
adding some drops of FS (or FSG) with a Pasteur pipette.
6. To facilitate the dissection of the samples, it is recommendable
to do it after some time (30 min) of postfixation. Dissections can
be done in a Petri dish using a razor blade and, if necessary,
under the binocular lens. Tissue pieces should be covered all the
time by the FS (or FSG) at 4 °C, in order to prevent desiccation.
Preferably the basis of the pieces should be as flat as possible to
assure a good and stable attachment to the cutting device. Once
the samples are dissected out, they should be transferred back
from the Petri dish to the glass vials to complete the total time
of fixation (4 h for light microscopic study and 1 h for TEM)
established in the method section. In the case of vibratome sec-
tions, the size of the pieces should be enough to be appropri-
ately cut. In this sense, coronal sections 5–6 mm thick of a whole
rat brain may be obtained and cut without problems.
7. Best results are obtained when samples are cut on the same
day. If not possible to do the histochemistry the same day of
perfusion, it is recommendable to keep the dissected samples
in PWB (4 °C) and to obtain the vibratome sections in the fol-
lowing day.
8. Although any super glue can be used, the gel type is the most
recommendable because it does not spread when placed in the
filter paper and it holds better the tissue piece, especially if the
piece is not completely flat in the base.
9. If there are many pieces, they can be attached to different filter
paper squares and maintained submerged in cold PWB until
they are cut.
10. All the process should be as fast as possible to minimize the
possible desiccation of the samples. Do your best to ensure
that the time spend in attaching the samples to the filter paper
is only some few seconds.
11. If the vibratome device you use does not have a container
where temperature can be regulated, then change frequently
the PWB, or if the vibratome device allows it, add some ice
around the container to maintain the PWB cold.
12. SS is a cryopreservative solution and avoids the formation of
water crystals that can break the tissue structure. It is important
to maintain the samples at list 48 h in this solution to ensure a

good preservation. Samples sunk at the bottom of the glass vial
indicate completely penetration of sucrose in the tissue.
13. Tissue-Tek® OCT™ Compound is special medium that freezes
and provides a correct support of the sample during the cut-
ting process.
14. The freezing process should be as quick as possible to ensure
that all the tissue freezes at the same time. For that purpose,
allow 2-methylbutane to reach cold enough temperature by
putting the solution surrounded by dry ice at least 15 min before
starting the freezing procedure. In order to be sure that
2-methylbutane is cold enough, drop a piece of tissue into it.
Tissue must become white in 1–2 s indicating a rapid freezing.
15. Keep the plastic vials filled with OS inside the cryostat to main-
tain the temperature of OS at −20 °C during the cutting
process.
16. An appropriate mold for a tissue is one that ensures that all the
sides of the tissue will be surrounded by paraffin. This is impor-
tant because it gives a best cutting support.
17. To increase the speed of paraffin block cooling, the mold can
be submerged into cold water.
18. The incubation time should be the same for all the samples,
especially if you want to perform a quantitative analysis of the
staining. Skip this step in case of fluorescence microscopy.
19. In case of paraffin sections mounted on gelatine-coated slides,
it may be useful to draw a hydrophobic mark around the tissue
using a specific pencil (IHCworld, Super PAP Pen Liquid
Blocker, SPM0928) in order to prevent the solutions from
spreading. In addition, the different incubations should be
performed in a humid chamber which will ensure that the solu-
tions will not evaporate during the incubation time.
20. As a negative control of the histochemistry, some sections
should be processed in parallel without the lectin.
21. From now on, keep the sections away from direct light, as fluo-
rochrome has been used and fluorescence may fade easily.
22. Toluidine blue solution for counterstaining can be reused, but
it is important to filter always the solution before use. Note
that the time of counterstaining should be adjusted in every
reused; therefore, try first with one slice to determine the
appropriated time.
23. DPX could be used as well for fluorescence microscopy, but
fluorescent mounting medium (Dako, S3023) is strongly rec-
ommended. In this case avoid steps 15 and 18 and coverslip
the slides just after mounting (step 11).
24. As negative control of the immunohistochemical reaction,

some sections should be processed in parallel without the pri-
mary antibody.
25. From now, remove the lids of the glass vials in order to facili-
tate the evaporation of propylene oxide.
References
1. Roth J (2011) Lectins for histochemical dem- 10. Acarin L, Gonzalez B, Castellano B (2000)
onstration of glycans. Histochem Cell Biol Neuronal, astroglial and microglial cytokine
136(2):117–130 expression after an excitotoxic lesion in the
2. Acarin L, Vela JM, Gonzalez B et al (1994) immature rat brain. Eur J Neurosci
Demonstration of poly-N-acetyl lactosamine 12(10):3505–3520
residues in ameboid and ramified microglial 11. Sanz O, Acarin L, Gonzalez B et al (2001)
cells in rat brain by tomato lectin binding. Expression of 27 kDa heat shock protein
J Histochem Cytochem 42(8):1033–1041 (Hsp27) in immature rat brain after a cortical
3. Pesheva P, Urschel S, Frei K et al (1998) aspiration lesion. Glia 36(3):259–270
Murine microglial cells express functionally 12. Almolda B, Gonzalez B, Castellano B (2010)
active galectin-3 in vitro. J Neurosci Res 51(1): Activated microglial cells acquire an immature
49–57 dendritic cell phenotype and may terminate
4. Dalmau I, Vela JM, Gonzalez B et al (1997) the immune response in an acute model of
Expression of LFA-1alpha and ICAM-1 in the EAE. J Neuroimmunol 223(1–2):39–54
developing rat brain: a potential mechanism 13. Vela JM, Dalmau I, Gonzalez B et al (1996)
for the recruitment of microglial cell precur- The microglial reaction in spinal cords of jimpy
sors. Brain Res Dev Brain Res 103(2): mice is related to apoptotic oligodendrocytes.
163–170 Brain Res 712(1):134–142
5. Dalmau I, Vela JM, Gonzalez B et al (2003) 14. Vela JM, Hidalgo J, Gonzalez B et al (1997)
Dynamics of microglia in the developing rat Induction of metallothionein in astrocytes and
brain. J Comp Neurol 458(2):144–157 microglia in the spinal cord from the myelin-
6. Campuzano O, Castillo-Ruiz MM, Acarin L deficient jimpy mouse. Brain Res 767(2):
et al (2009) Increased levels of proinflamma- 345–355
tory cytokines in the aged rat brain attenuate 15. Minten C, Terry R, Deffrasnes C et al (2012)
injury-induced cytokine response after excito- IFN regulatory factor 8 is a key constitutive
toxic damage. J Neurosci Res 87(11): determinant of the morphological and molec-
2484–2497 ular properties of microglia in the CNS. PLoS
7. Acarin L, Gonzalez B, Castellano B et al One 7(11):e49851
(1996) Microglial response to N-methyl-D- 16. Castellano B, Gonzalez B, Finsen BR et al
aspartate-mediated excitotoxicity in the imma- (1990) Histochemical demonstration of
ture rat brain. J Comp Neurol 367(3): purine nucleoside phosphorylase (PNPase) in
361–374 microglial and astroglial cells of adult rat brain.
8. Acarin L, Gonzalez B, Castellano B et al J Histochem Cytochem 38(11):1535–1539
(1997) Quantitative analysis of microglial 17. Montero Dominguez M, Gonzalez B, Zimmer
reaction to a cortical excitotoxic lesion in the J (2009) Neuroprotective effects of the anti-
early postnatal brain. Exp Neurol 147(2): inflammatory compound triflusal on ischemia-
410–417 like neurodegeneration in mouse hippocampal
9. Acarin L, Gonzalez B, Castro AJ et al (1999) slice cultures occur independent of microglia.
Primary cortical glial reaction versus secondary Exp Neurol 218(1):11–23
thalamic glial response in the excitotoxically 18. Montero M, Gonzalez B, Zimmer J (2009)
injured young brain: microglial/macrophage Immunotoxic depletion of microglia in mouse
response and major histocompatibility complex hippocampal slice cultures enhances ischemia-
class I and II expression. Neuroscience like neurodegeneration. Brain Res 1291:
89(2):549–565 140–152
Chapter 24
Immunohistochemical Detection of Microglia

Rocío M. de Pablos, Ana M. Espinosa-Oliva, and Antonio J. Herrera
Abstract
Immunohistochemistry (IHC) is a technique that allows the localization of antigens or proteins in tissue
sections using the high specificity and affinity of antibodies to recognize molecules and join them. The
commercial offer and the standardization of protocols make this technique a simple, fast, and powerful
method. Microglia, the resident macrophage cells of the central nervous system, can exist in three different
forms that can be identified using different antibodies. The aim of this chapter is to describe the methods
to perform IHC using these different antibodies.
Key words Iba1, Immunohistochemistry, Microglia, OX-6, OX-42
1 Introduction
Immunohistochemistry (IHC) is a technique that allows the local-

ization of antigens or proteins in tissue sections; it uses labelled anti-
bodies as specific reagents through antigen–antibody interactions
visualized by a variety of markers, such as enzymes, fluorescent dyes,
or colloidal gold. Since the first experiments back in the 1930–
1940s, methods of tissue fixation, protein conjugation, labels, and
microscopy have improved greatly, making IHC a simple, powerful,
and essential tool in diagnostic and research laboratories.
The technique is based on the high specificity and affinity of anti-
bodies to recognize molecules and join them. It enables the visualiza-
tion (using light or confocal microscopy) of the tissue distribution of
specific antigens (epitopes). Specific monoclonal and polyclonal anti-
bodies target proteins of interest in a process called antibody incubation
(Fig. 1). The use of avidin–biotin immunohistochemistry on recent
years takes advantage of the high affinity that avidin (or streptavidin)
has for biotin, making possible the use of multiple binding sites on the
target molecules, thus resulting in the amplification of tissue signal,
which is very helpful to detect low amounts of antigens. Conjugation
or combination of the antibodies with fluorescing substances enables
the detection of minute amounts in the tissue sections.
281
282 Rocío M. de Pablos et al.
Fig. 1 Schematic representation of the immunohistochemical staining using

antibodies. Primary antibody specifically binds the antigen through its variable
region, whereas the biotinylated secondary antibody recognizes the constant
region of the primary one. The avidin–peroxidase conjugate binds biotin attached
to the secondary antibody. The peroxidase was visualized with a standard diami-
nobenzidine/hydrogen peroxide reaction
Microglia is the resident macrophage cell of the central ner-

vous system, showing a highly specialized morphology and unusual
antigenic phenotype. Microglia can exist in three different forms
serving different functional roles: amoeboid, ramified, and reac-
tive/activated microglia. It has generally been accepted that rest-
ing microglia—with small cell bodies and thin processes—increase
in their soma size as well as thicken and shorten their processes in
early stage of activation to become activated microglia, which are
finally transformed into amoeboid microglia with phagocytic
activity [1–4]. Microglia share many properties with peripheral
monocytes and macrophages but retain unique morphological
and molecular characteristics [5, 6]. Microglia can bind several
lectins, including Griffonia simplicifolia agglutinin I B4, Ricinus
communis agglutinin, and mistletoe agglutinin, as well as express
complement receptor 3 (CR3), even in the resting state. In addi-
tion, activated microglia expresses high levels of major histocom-
patibility complex (MHC) class I and II [1, 2, 7, 8]. These
different forms of microglial cells can be identified using different
antibodies. Iba1 is a 17-kDa EF hand protein that is specifically
expressed in macrophages/microglia and is upregulated during
the activation of these cells. Iba1 is markedly expressed in microg-
lia (Fig. 2), but is never expressed in neurons or astrocytes. These
observations indicate that Iba1 is a macrophage-specific protein
Microglia Immunohistochemistry 283
Fig. 2 Immunostaining of microglial cells using the Iba1 antibody
and that anti-Iba1 antibody is valuable for identification of

microglia and macrophages. In addition, anti-Iba1 antibody has a
great advantage over other marker antibodies since it is so far the
only polyclonal antibody applicable to dual labelling studies of the
macrophage/microglia lineage even in paraffin sections [9].
OX-42 is an antibody to CR3, a marker expressed by microglia
(Fig. 3), whereas OX-6 is an antibody recognizing the MHC II
antigen expressed by activated microglia (Fig. 4).
The aim of this chapter is to describe the methods to perform
an IHC using the three antibodies mentioned above.
2 Materials
2.1 General 1. Section-staining tray (humid chamber for sections staining).

2. Coplin jars.
3. Hydrophobic pen (Dako, Glostrup, Denmark).
4. Coverslips.
5. Tissue-Tek® OCT™ compound (Sakura, Alphen aan den Rijn,
the Netherlands).
6. Ethanol solutions (50–100 %).
7. Methanol.
8. Hydrogen peroxide (30 %).
9. Triton® X-100.
Fig. 3 Immunostaining of microglial cells using the OX-42 antibody. The picture
shows resting microglial cells with small cell bodies and thin processes
Fig. 4 Immunohistochemistry of activated microglial cells using the OX-6 anti-

body. The picture shows microglial cells in different states of activation
10. Histolemon (Carlo Erba, Milano, Italy).

11. DPX mounting medium (VWR International Ltd, Poole,
England).
2.2 Reagents 1. ABC solution (Vectastain®; Vector Laboratories Inc.).

2. Proteinase K (Sigma-Aldrich).
3. 4 % paraformaldehyde in PBS, pH 7.4. Weigh 20 g paraformal-

dehyde and 7.1 g Na2HPO4 in 100–150 mL of distilled water.
Heat to 55 °C with continuous stirring. Add a few pellets of
NaOH. Add 25 g sucrose and 2 mL 25 % glutaraldehyde.
Adjust pH, top up to 500 mL with distilled water, and filter.
Store at 4 ºC.
4. Tris-buffered saline (TBS), pH 7.4, 10×: 121 g Tris base, and
1,470 g NaCl in 7 L of distilled water. Adjust pH, top up to
10 L with distilled water, and filter. Store at room temperature.
5. Tris-EDTA (TE) buffer for proteinase K, pH adjusted to 8.0:
6.1 g Tris base, 0.37 g EDTA, and 5 mL Triton X-100 in 1 L
of distilled water. Store at room temperature.
6. Phosphate-buffered saline (PBS), pH 7.4, 10×: 7.933 g
NaH2PO4⋅H2O, 8.9 g Na2HPO4⋅2H2O, and 75.86 NaCl in
500 mL of distilled water. Adjust pH, top up to 1 L with dis-
tilled water, and filter. Store at room temperature.
7. Diaminobenzidine (DAB; Sigma-Aldrich): Dissolve 1 tablet
(10 mg) in 100 mL TBS 1×. Filter and make aliquots of 2 mL,
approximately. Store at −20 ºC protected from the light. DAB–
hydrogen peroxidase reaction solution should be prepared
freshly, adding 2 μL of H2O2 to 2 mL of DAB solution just
before use.
8. Proteinase K: 2 mg of proteinase K, 2.5 mL TE buffer, and
2.5 mL glycerol. Gently mix proteinase K and TE buffer and
then add glycerol. Mix and make aliquots (75 μL) stored at
−20 ºC. In this condition, aliquots are viable for 2–3 years.
2.3 Antibodies 1. Primary antibodies: Mouse-derived OX-6 (Serotec, Oxford,

and Serums UK; 1:200), mouse-derived OX-42 (Serotec; 1:100), and
rabbit-derived anti-Iba1 (Wako, Osaka, Japan; 1:500).
2. Secondary antibodies: Biotinylated horse anti-mouse IgG
(1:200) and biotinylated goat anti-rabbit IgG (1:200) from
Vector Laboratories (Burlingame, CA).
3. Horse and goat serum (Vector Laboratories, 1:100).
3 Methods
3.1 Tissue To obtain good results, fixation and processing of the tissue must
Preparation be taken into consideration.
1. Perfuse through the heart (left ventricle) under deep anesthe-
sia with approximately 100 mL of saline solution to remove
brain blood.
2. Perfuse with 150–200 mL of 4 % paraformaldehyde in 0.1 M
PBS, pH 7.4.
3. Remove the brain from the skull and then cryoprotect it serially
in sucrose in PBS, pH 7.4 (see Note 1).
4. Freeze the brains in isopentane at −80 ºC (see Note 2).
If brains are not used immediately, store them at −40 ºC
(see Note 3).
5. Mount specimens with OCT; cut 20-μm sections in a cryostat
(−20 °C) and mount them in gelatine-coated slides. Sections
can be stored at −20 ºC for years (see Note 4).
3.2 Antigen Retrieval Most proteins and peptides retain their antigenicity with 4 % para-
formaldehyde, although low molecular weight antigens (e.g., bio-
genic amines) may lose it. Reaction with many antigens can be
significantly improved by the pretreatment with an antigen retrieval
reagent that breaks the protein cross-links formed by formalin fixa-
tion, uncovering hidden antigenic sites. Different antigens and
antibodies may require different antigen retrieval methods, as the
heat-induced epitope retrieval (HIER) or the proteolytic-induced
epitope retrieval (PIER), among others.
Proteinase K: Formalin and other aldehyde fixatives form protein
cross-links that mask antigenic sites in tissue specimens, thereby
producing weak or false-negative staining in the immunohisto-
chemical detection process of some proteins. Proteinase K solu-
tions (PIER) break these protein cross-links, unmasking antigens
and epitopes in formalin-fixed and paraffin-embedded tissue sec-
tions, which enhances staining intensity of antibodies. This method
is highly recommended for Iba1 immunostaining (to be done
before immunostaining procedure), proceeding as follows:
1. Place a humid chamber in an incubator at 37 ºC for 10 min.
2. Prepare the proteinase K solution by adding 1,425 μL of TE
buffer to an aliquot of 75 μL proteinase K prepared as described
below (dilution 1:20).
3. Incubate sections with this solution in the humid chamber for
10 min at 37 ºC and then for 10 min at room temperature
(see Note 5).
4. Rinse twice in TBS 1×, 10 min each (see Note 6).
3.3 Immuno- 1. Completely thaw the sections for at least 15–30 min (see Note 7).
histochemistry 2. Surround sections with the hydrophobic pen (see Note 8).
3. Rinse in TBS 1× for 10 min in a jar.
4. Block endogenous peroxidase with 0.3 % H2O2 in methanol in
a jar.
5. Rinse twice in TBS 1× for 10 min each in a jar.
6. Incubate in a solution containing TBS 1× and 1 % normal
serum (from the same animal species as the biotinylated anti-
body) for 60 min in a humid chamber. Add 50 μL approxi-
mately to each section (see Note 9).
7. Drain slides without washing.

8. Incubate overnight at 4 ºC in the humid chamber with 50 μL
per section of the primary antibody in TBS 1× containing 1 %
serum and 0.25 % Triton X-100 (see Note 10).
9. Rinse three times in TBS 1× for 10 min each in a jar.
10. Incubate with the secondary biotinylated antibody (1:200
dilution) in TBS 1× containing 0.25 % Triton X-100 (50 μL
per section) for 2 h at room temperature in the humid
chamber.
12. Allow to react with ABC solution (50 μL per section) for 1 h
at room temperature (see Note 11).
14. Allow reacting with DAB–hydrogen peroxide solution (50 μL
per section) for 3–6 min.
15. Rinse twice in TBS 1× for 10 min each, changing the jar after
the first wash (see Note 12).
16. Dehydrate sections with 50, 70, and 90 % ethanol (5 min each)
and with 100 % ethanol for 10 min, then twice in Histolemon,
10 min each (see Note 13).
17. Mount sections in a permanent medium (DPX) with cover
glasses and observe under microscope (see Note 14).
4 Notes
1. Sucrose displaces part of the water in the tissue, reducing freez-

ing artifacts during sectioning. Put the brain in 10 % sucrose
for 24 h and then in 20 and 30 % sucrose until it sunk (2–5
days).
2. To prevent deformation of the brain during the process, put it
and remove it slowly several times before dropping it in the
frozen isopentane.
3. Slow freezing can form ice crystal, altering the normal archi-
tecture of the tissue and producing holes, in a sort of “Swiss
cheese” effect. Thus, freezing should be made quickly to avoid
ice crystals keeping water in a vitreous form that does not
expand when solidifies. Make sure freezing medium is cold
enough. Brains can be frozen in isopentane at −80 ºC to
obtain a fast freezing and then kept at −40 ºC. It might be
thought that liquid nitrogen (nearly −200 ºC, one of the cold-
est liquids routinely available; it does not mix with tissue)
would be very appropriate for fast freezing. However, it boils
in contact with objects over their own temperature (virtually
all the objects in the lab, including specimens), creating a vapor
barrier that causes freezing in a slower, unpredictable pattern.
Because of this, tissue in OCT often cracks due to the

unpredictable freezing pattern. We do not recommend liquid
nitrogen to freeze brain specimens.
4. Sometimes, sections can detach from slides during the long
incubation process. This can be due to the handling of sections
while cutting; if sections are not well stretched over the glass,
they can be moved smoothly by using a fine brush and saline
solution. Sections should not be frozen until the saline solu-
tion has dried; keep sections at room temperature for
15–30 min before storing them frozen.
5. This method can damage under-fixed tissues. An appropriate
incubation time (5–20 min) and temperature (20–60 ºC) are
required for specific application; avoid over-digesting tissues,
especially under-fixed ones.
6. Other types of antigen retrieval system (HIER) greatly enhance
immunohistochemistry results after microwave irradiation of
formalin-fixed, paraffin-embedded specimens in buffer. The
energy provided by irradiation breaks some of the bonds
formed during fixation, increasing the intensity of reactions
and thus the number of positive cells available, although the
exact mechanism involved is unclear.
Introduce the slides in a coplin jar with citrate solution, pH 6
(0.65 g citric acid and 0.48 g sodium citrate in 2 L of distilled
water. Store at room temperature), and put this jar within a big-
ger jar with water. Heat in the microwave for 4 min at approxi-
mately 500 W. Repeat replacing water and citrate solution. Place
sections in a coplin jar with TBS for 10 min to cool.
7. As mentioned in Note 4, sections can detach from slides dur-
ing the incubation process. To avoid this, sections must be
absolutely dried before starting the protocol. Putting them in
a hot plate during 15–30 min could be helpful.
8. Alternatively, you can drain and dry the slides around the sec-
tions with filter paper after every wash. Once you start the pro-
tocol, sections should not dry in any moment. Be especially
careful in long incubations.
9. The humid chamber can be prepared by placing some filter
paper at the bottom of a staining tray, wetting it with water or
buffer.
10. Detergents (e.g., Triton X-100) should be used for antiserum
raised against an intracellular antigen or domain. As a general
rule, antibodies may be stored frozen but repeated thaw/
frozen cycles should be avoid. Prepare small aliquots adjusted
to the amounts normally used for the staining and store those
frozen.
OX-6 can recognize a monomorphic determinant of the
rat I-A antigen present on B lymphocytes, dendritic cells,
macrophages, and certain epithelial cells. The anti-Iba1 anti-

body also recognized a number of tissue-resident macro-
phages, including alveolar macrophages, Kupffer cells, and
splenocytes. OX-42 recognizes most macrophages (includ-
ing resident peritoneal and activated macrophages), Kupffer
cells, and about 35 % alveolar macrophages [10]. This anti-
body also labels dendritic cells extensively, granulocytes, and
cells with the morphology of microglia in the brain. Note
that when using these antibodies, you can be staining not
only microglial cells. To distinguish microglial cells from
macrophages, other techniques, such as flow cytometry, are
required.
11. ABC solution should be made 30 min before use by mixing A
and B solutions (10 μL each in 1 mL of TBS 1×).
12. Remove DAB solution carefully with a pipette, discarding it in
an appropriate vial for waste; stop the reaction by transferring
to TBS 1×. Check the reaction (brown color) by observing
under a light microscope.
DAB is a carcinogenic compound that must be manipu-
lated carefully. We recommend to develop the DAB reaction
by placing the staining tray on aluminum foil and discard it
when finished. The coplin jar used to the first wash after the
DAB reaction must be cleaned thoroughly.
13. For proper conservation of the slides, ethanol and Histolemon
should not be reused many times (they accumulate water) to
ensure the dehydrating process.
14. When mounting the slides with mounting medium, remove
bubbles by pressing the coverslip with Dumont forceps.
References
1. Streit WJ, Kreutzberg GW (1988) Response of resident microglial cells from the normal and
endogenous glial cells to motor neuron degen- inflamed central nervous system. Proc Natl
eration induced by toxic ricin. J Comp Neurol Acad Sci USA 88:7438–7442
268:248–263 7. Kaur C, Ling EA (1992) Activation and re-
2. Mor G, Nilsen J, Horvath T et al (1999) expression of surface antigen in microglia fol-
Estrogen and microglia: a regulatory system lowing an epidural application of kainic acid in
that affects the brain. J Neurobiol 40: the rat brain. J Anat 180:333–342
484–496 8. Kreutzberg GW (1996) Microglia: a sensor for
3. Stence N, Waite M, Dailey ME (2001) Dynamics pathological events in the CNS. Trends
of microglial activation: a confocal time-lapse Neurosci 19:312–318
analysis in hippocampal slices. Glia 33:256–266 9. Imai Y, Kohsaka S (2002) Intracellular signal-
4. Cho BP, Song DY, Sugama S et al (2006) ing in M-CSF-induced microglia activation:
Pathological dynamics of activated microglia role of Iba1. Glia 40:164–174
following medial forebrain bundle transection. 10. Robinson AP, White TM, Mason DW (1986)
Glia 53:92–102 Macrophages heterogeneity in the rat as delin-
5. Perry VH, Gordon S (1991) Macrophages and eated by two monoclonal antibodies MRC
the nervous system. Int Rev Cytol 125:203–244 OX-41 and MRC OX-42, the latter recogniz-
6. Sedgwick JD, Schwender S, Imrich H et al ing complement receptor type 3. Immunology
(1991) Isolation and direct characterization of 57:239–247
Chapter 25
Intrathecal Infusion of Microglia Cells

and Kazuhide Inoue
Abstract
Spinal microglia have been implicated in the pathogenesis of neuropathic pain after peripheral nerve injury
concomitant with diseases such as diabetes and cancer. To reveal the etiological roles of microglia in behav-
ioral pain hypersensitivity or neuronal excitability, technical approaches have been used. Here, we describe
a technique for intrathecal transfer of cultured microglial cells through a catheter surgically implanted into
the spinal subarachnoid space.
Key words Microglia, Pain, Spinal cord
1 Introduction
Following nerve injury arising from bone compression in cancer,

diabetes, infection, autoimmune disease, or physical injury, spinal
microglia transform into a reactive state through a progressive
series of cellular and molecular changes and contribute to the for-
mation of debilitating chronic pain conditions (referred to as neu-
ropathic pain) [1, 2]. Various approaches have been used to
determine the microglial genes crucial for neuropathic pain, such
as those encoding cell surface receptors and diffusible mediators
(e.g., proinflammatory cytokines), and a number of microglial
genes have been characterized as “pain-related molecules” [1, 2].
Here, we demonstrate the intrathecal infusion of primary cultured
microglia as a tool for verifying the direct link between pain hyper-
sensitivity and microglial genes, in combination with behavioral or
electrophysiological approaches.
2 Materials
Prepare and store the following at room temperature (unless oth-

erwise specified). We do not add sodium azide to the reagents.
291
1. 10–12-week-old male mice (25–30 g).

2. Isoflurane.
3. Intrathecal catheter: connect 2.9 cm of 32-gauge catheter
(ReCathCo, Allison Park, PA) (indwelling end) and 5 cm of
polyethylene catheter (PE-10) (BD) (external end) using sur-
gical adhesive.
4. Primary microglial cells: prepare according to the method
described previously [3, 4].
5. Culture medium for microglial cells: 10 % fetal bovine serum
(FBS), penicillin (100 U/mL), streptomycin (100 μg/mL),
and L-glutamine (2 mM) in Dulbecco’s modified Eagle
medium (DMEM). Store at 4 °C.
6. Lentiviral particles: prepare as previously described (ref. 5, or
see the Chapter 8). Store at −80 °C.
7. 35-mm cell culture dish.
8. 24-well plate.
9. Cell lifter.
10. Cell counter.
11. Microsyringe.
3 Methods
3.1 Intrathecal 1. Anesthetize mice with 2 % isoflurane.

Catheterization 2. Clean the skin over the nape of the neck using 75 % ethanol
in Mice and incise for about 1 cm.
3. Expose the atlanto-occipital membrane by detaching the mus-
cle on either side of the external occipital crest.
4. Incise the membrane using the tip of a needle (see Note 1).
5. Insert an indwelling 32-gauge catheter through the membrane
into the intrathecal space in the midline, dorsal to the spinal
cord (see Note 2).
6. Externalize the exit end of the catheter (PE-10) through a
separate opening in the skin.
7. After more than 5 days for recovery, administer 1.5 μl of 2 %
lidocaine intrathecally to confirm that the tip of catheter was
placed appropriately for the intended target (see Note 3).
3.2 Intrathecal 1. Seed primary microglial cells in a cell culture dish (see Note 4).
Infusion of Gene- 2. Transduce the microglial cells by adding an appropriate vol-
Transduced Microglial ume of viral particles onto the cells (see Note 5).
Cells
3. After 72 h for transduction, aspirate culture medium.
Intrathecal Delivery of Microglia Cells 293
4. Gently wash the dish twice with PBS to remove culture medium
(see Note 6).
5. Harvest microglial cells using a cell lifter and transfer to a 1.5-
mL tube.
6. Spin down the cells by centrifuging at 400 × g for 1 min.
7. Discard the supernatant.
8. Resuspend the cells by gently pipetting with 100 μl of PBS.
9. Measure the density of microglia using a cell counter, and
adjust the cell density to 2.0 × 104 cells per 5 μl by adding an
appropriate volume of PBS (see Note 7).
10. Infuse 5 μl of cell suspension using a microsyringe through an
intrathecal catheter implanted into the intrathecal space, fol-
lowed by injection of PBS (5–10 μl).
3.3 Intrathecal 1. Seed primary microglial cells in a culture dish (see Note 4).
Infusion of Microglial 2. After they have adhered to the bottom, aspirate culture
Cells After Drug medium.
Treatment
3. Gently wash the dish twice with PBS to remove culture medium
(see Note 6).
4. Harvest microglial cells by using a cell lifter and transfer to a
1.5-mL tube.
5. Spin down the cells by centrifuging at 400 × g for 1 min.
6. Discard the supernatant.
7. Resuspend the cells by gently pipetting with 100 μl of PBS.
8. Measure the density of microglia using a cell counter, and
adjust the cell density according to the experimental condition
(see Note 7).
9. Treat the cells with drug, such as a neutralizing antibody.
10. Infuse the cells plus supernatant as described above (see
4 Notes
1. The escape of cerebrospinal fluid can be observed when the atlanto-

occipital membrane is incised (for Subheading 3.1, step 4).
2. To place the tip of the catheter in proximity to the L3–L4 dor-
sal surface of the spinal cord, insert the catheter intrathecally
about 2.9 cm (for Subheading 3.1, step 5).
3. Mice showing severe motor weakness under the normal condi-
tion should be excluded from the study (for Subheading 3.1,
step 7).
4. We usually seed primary microglial cells at a density of 5 × 105

cells per 35-mm culture dish or 1.2 × 105 cells per well (24-well
plate) (for Subheading 3.2, step 1, and Subheading 3.3, step 1).
5. Transduce the cells with viral particles according to our previous
paper [5] (or see the Chapter 8) (for Subheading 3.2, step 2).
6. PBS can be used at room temperature. To avoid adding culture
medium to the cell suspension, remove culture medium as
carefully as possible (for Subheading 3.2, step 4, and
7. Adjust the density of microglial cells according to experimental
condition (for Subheading 3.2, step 9, and Subheading 3.3,
step 8).
References
1. McMahon SB, Malcangio M (2009) Current chal- 4. Tsuda M, Masuda T, Kitano J et al (2009) IFN-
lenges in glia-pain biology. Neuron 64:46–54 gamma receptor signaling mediates spinal
2. Inoue K, Tsuda M (2009) Microglia and neu- microglia activation driving neuropathic pain.
ropathic pain. Glia 57:1469–1479 Proc Natl Acad Sci USA 106:8032–8037
3. Nakajima K, Shimojo M, Hamanoue M et al 5. Masuda T, Tsuda M, Yoshinaga R et al (2012)
(1992) Identification of elastase as a secretory IRF8 is a critical transcription factor for trans-
protease from cultured rat microglia. J forming microglia into a reactive phenotype.
Neurochem 58:1401–1408 Cell Rep 1:334–340
Chapter 26
Intracranial Injection of LPS in Rat as Animal Model

of Neuroinflammation
Ana M. Espinosa-Oliva, Rocío M. de Pablos, and Antonio J. Herrera
Abstract
Animal models of neuroinflammatory processes are needed to study the involvement of inflammation in
neurodegenerative disorders such as Parkinson’s and Alzheimer’s diseases. One of the models used is based
on lipopolysaccharide (LPS) as brain inflammation-inducing agent. This toxin is a potent inducer of
inflammation and has different effects on cells of the immune system, as microglial cells. This chapter
describes a protocol for the model of brain inflammation in rats based on the unilateral stereotaxic injec-
tion of LPS, which mimics the inflammatory milieu produced in some brain diseases.
Key words Animal models, Lipopolysaccharide, Neurodegeneration, Neuroinflammation,

Stereotaxic surgery
1 Introduction
The possible role of neuroinflammatory processes in the develop-

ment of neurodegenerative diseases (i.e., Parkinson’s disease, PD)
is receiving growing support. Early on 1988, McGeer’s group
showed the presence of activated microglia in the substantia nigra
(SN) of PD patients, suggesting the implication of neuroinflamma-
tion as a component of the neurodegenerative progress of the dis-
ease [1]; for review, see refs. 2, 3. In fact, they showed a huge
proliferation of reactive amoeboid macrophages and microglia in
the SN of postmortem parkinsonian brains; this finding, along
with the oxidative stress marks observed, was compatible with the
existence of inflammation, as it had been previously described for
Alzheimer’s disease (AD) [4] and multiple sclerosis [5].
Animal models of neuroinflammation have become essential to
study the possible implication of neuroinflammatory processes in
the development of these disorders. Stereotaxic injection of lipo-
polysaccharide (LPS) is now commonly used to produce neuroin-
flammation within the brain. LPS, a component of the Gram-negative
bacteria cell wall [6], is a potent inducer of inflammation and has
295
296 Ana M. Espinosa-Oliva et al.
diverse effects on cells of the immune system [7]. Several works

have shown that LPS activated glial cells in vivo after intracranial
injection [8–12]; microglial cells release several neurotoxins [13–
18], as inflammatory cytokines, N-methyl-D-aspartate (NMDA)
receptor agonists, oxygen free radicals, and nitric oxide (NO),
which could contribute to neurotoxicity. All these data suggest
that the injection of LPS in different areas of the brain could pro-
vide a good experimental model to ascertain the importance of
inflammation in this organ. Stereotaxic surgery allows the experi-
menter to reach any part of the brain with great accuracy and thus
is widely used for delivering drugs and other substances within the
brain. Two fundamental things are required to carry out stereo-
taxic injection of LPS: a stereotaxic apparatus and a stereotaxic
brain atlas.
The stereotaxic apparatus allows the researcher to place an
object (cannula, electrode, etc.) or substance at a specific location
within the brain. It includes a support to keep the animal’s skull in
the appropriate position, as well as a holder for the syringe or the
electrode. The holder is attached to a calibrated mechanism that
moves the holder through the three spatial axes, the so-called ste-
reotaxic coordinates: anterior-posterior (AP), lateral-medial (LM),
and dorsal-ventral (DV).
The stereotaxic brain atlas is a set of photographs or drawings
of brain sections, ordered along one of the three axes mentioned
above (Fig. 1a). As it will be explained later on, injections are per-
formed through a small drill hole made on the skull, trying to
minimize the damage inflicted to the animal. Thus, brain coordi-
nates have to be referred to an external reference point. Two of
these points are used: bregma (the anatomical point on the skull at
which the coronal suture is intersected perpendicularly by the sag-
ittal suture) and lambda (a dense, fibrous connective tissue joint on
the posterior part of the skull that connects the parietal and tem-
poral bones with the occipital bone; Fig. 1b). For a discussion on
bregma and lambda points, see Paxinos and Watson [19].
Pages of a stereotaxic atlas typically correspond to coronal sec-
tions taken at various distances anterior and posterior to bregma
and lambda. Each plate includes the LM and the DV distances to
these reference points. Sagittal and horizontal sections are also
provided. Brain structures can be easily found in an index (list of
structures) included in the stereotaxic atlas. Thus, brain structures
(that we cannot see in the animal alive) are easily located following
the three coordinates (with respect to bregma or lambda points on
the skull) provided by the atlas.
Stereotaxic surgery can be made either unilateral or bilateral.
In the first case, the intact side serves as absolute control. Bilateral
surgery is performed when sham and true surgery are required in
the same animal.
Injection of LPS in Brain 297
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7
a
9 1
8 2
7 3
6 4
5 5
4 6
3 7
2 8
1 9
0 10
Interaural 3.40 mm Bregma –5.60 mm
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7
Bregma
Lambda
Interaural Line
9.0 mm
Bregma Lambda
10.0 mm
3.3 mm
Incisor Bar
Interaural Line
Fig. 1 (a) Diagram showing the three stereotaxic coordinates and the structures at 5.6 mm posterior to
bregma. (b) Skull diagram showing bregma and lambda points as well as the position of the tooth bar to get
a flat position of the skull (Reproduced with permission from The Rat Brain in Stereotaxic Coordinates,
Paxinos G. and Watson C.; 1986. Academic Press)
2 Materials
2.1 Surgical 1. Stereotaxic instrument for small animals supplemented with

Apparatus and Tools ear bars.
2. Ten-microliter Hamilton syringe (Hamilton Bonaduz AG,
Bonaduz, Switzerland).
3. Electric shaver.
4. Scissors.
5. Scalpel.
6. Metal clips.
7. Sharp forceps.
8. Needles.
9. Suture thread.
10. Forceps.
11. Disinfectant.
12. Syringe.
13. Sterile cotton pads.
14. Cloth.
15. Small drill.
2.2 Reactives 1. Anesthetics: Several anesthetic agents can be used as described

in the literature:
– Chloral hydrate: Dissolve 6 g in 100 ml of distilled water;
inject 400 mg per kg of body weight intraperitoneal (i.p.).
– Fentanyl citrate salt, 50 μg/ml + medetomidine hydro-
chloride (DormitorVet®, Pfizer Animal Health) 50 μg/ml,
20:1 mixture; inject 6.3 ml per kg of body weight (i.p.).
– Ketamine 80 mg/kg + xylazine 12 mg/kg (ketamine
hydrochloride/xylazine hydrochloride solution, Sigma-
Aldrich); inject 1 ml of solution per kg of body weight
intramuscular.
2. Saline solution (0.9 % NaCl, w/v in distilled water).
3. 70 % ethanol.
4. Vehicle: 1 % Monastral Blue inert tracer in saline solution.
5. LPS: Dissolve 1 mg in 1 ml of vehicle. Aliquot small amounts
(ca 15–20 μl) into Eppendorf® tubes. Keep the tubes on ice if
LPS is going to be used fresh. For long storage, use screw cap
vials; glass vials are preferred. Store the aliquots at −20 °C and
avoid repeated thawing/freezing cycles.
3 Methods
The procedure described below is carried out in accordance with

the guidelines of the European Union Council (86/609/EU) for the
use of laboratory animals. Before starting with the surgery, check
that all reagents and materials are ready.
1. Prepare a couple of Eppendorf® tubes containing distilled
water and 70 % ethanol, respectively, for washing the syringe.
2. Thaw an aliquot of LPS; keep it on ice.
3. Prepare all the surgical material listed above, disinfecting them
with 70 % ethanol.
4. Clean the Hamilton syringe barrel inner space with distilled
water and 70 % ethanol (see Note 1).
5. Place the Hamilton syringe in the syringe holder (see Note 2).
6. Place a clean cloth over the stereotaxic apparatus platform
where the animal will be placed (see Note 3).
7. Weigh and anesthetize the animal. Different injectable anes-
thetics can be used as already mentioned above (see Note 4).
8. Place the animal in a cage until it is deeply asleep (see Note 5).
9. Take an aliquot of LPS solution and fill the Hamilton syringe
avoiding bubbles.
10. To decrease the chance of infection at the incision, shave the
fur on the top of the head with an electric shaver.
11. Place the animal in the stereotaxic apparatus on horizontal
position. Put the ears and the tooth bar, immobilizing the skull
to avoid movements of the head (see Note 6) (see Figs. 2 and 3).
Cover the body with the cloth (see Note 7).
12. Wipe the shaved zone with a sterile cotton pad soaked in 70 %
ethanol. Let it dry.
13. Make a midline incision on the skin with the scalpel, starting
between the eyes; about 2 cm in length should be enough.
14. Keep the skin open with two metal clips. Scratch and remove
the membrane above the bone; clean the bleeding with a sterile
cotton pad and keep the bone dry; the bregma is easily located
(see Fig. 4a).
15. Place the needle on the bregma (see Note 8).
16. Read the three coordinates (AP, LM, and DV) on the manipu-
lator (in mm) and write them down; this is our starting point.
Move the needle up a few millimeters.
17. Set the AP and LM coordinates of the injection point by add-
ing the previously calculated coordinates to the manipulator.
Mark the point on the skull (see Note 9).
Fig. 2 (a) Placing ear bars. (b) Adequate position of the bar with respect to the
bar holder
18. Retract the arm with the needle and drill a small hole in the
skull bone above the mark (see Note 10; see Fig. 4b).
19. Put the arm in position and check that the needle enters (just
the tip) through the drill hole centered. If the needle tip is
close to the borders, make the hole a bit bigger but do not
change needle coordinates (see Note 11).
20. Slowly, set the DV coordinate of the injection point; this inserts
the needle tip into the brain till the injection point. Wait 1 min
with needle in place.
21. Start releasing the LPS solution (2 μl) at an approximate flow
rate of 0.5 μl/min (see Note 12).
22. Leave the needle in place for 5 min to avoid reflux along the
injection track (see Note 13).
Fig. 3 (a) Tooth bar. The lower part hooks the teeth and the upper part pushes
the rat nose down to the required flat position of the skull. Rat before (b) and
after (c) fixing the head with the tooth bar
Fig. 4 (a) Rat skull after cutting the skin and removing bone membrane. The
arrow shows the bregma point. (b) Drill hole (arrow) at the projection point of
the SN on the skull
23. Take the needle out very slowly. Retract the arm, empty the
syringe, and clean it as described above.
24. Close the wound area by suturing, apply disinfectant, and mark
the rat for further identification.
25. Place the animal in a clean cage until it regains consciousness
and recovers (Notes 7 and 14).
26. Disinfect the surgical tools and the table.
4 Notes
1. Fill and empty the syringe several times with distilled water,
then with 70 % ethanol, and again with distilled water.
Cleanliness of the needle is very important to prevent infection
and to check if the syringe is occluded.
2. Make sure that the needle is not bended at all; needle orienta-
tion must be absolutely vertical with respect to the apparatus
platform to avoid injection out of the desired structure.
3. Animals can experiment some hypothermia under anesthesia.
4. Inhaled anesthetics can also be used. In such cases, the animal
is usually introduced in a chamber where the gas is released
until it is completely asleep. Gases commonly used are either
isoflurane or halothane. A stereotaxic adapter is needed to
influx anesthesic gas during the entire surgical procedure. In
such a case, surgery should be performed in a ventilated bench
or fume hood to prevent gas inhalation.
5. Look for reflexes by pinching the tail or paws to ensure that
the animal is anesthetized.
6. The procedure is delicate and requires time until the necessary
ease is got; you may experience some difficulty the first few
times you do this if you have no previous experience. With
practice, placing the rat in the apparatus takes less than 1 min.
It is essential that the head is positioned correctly so that the
plane containing bregma and lambda points is flat.
Ear bars should not be pushed with strength to avoid
unnecessary damage to the skull and brain; pressing with one
finger is enough. Ear bars include a graduated scale; when the
rat is placed in the right position, the number 1 on the bars
should coincide approximately with the 0.5 that is in the bar
holder (Fig. 2b). When the bars enter the earhole in the skull,
a soft click can be heard; this is normally accompanied by eye
blinking of the rat. At this moment, the up-down movement
of the head is still free around the interaural line.
Next, fix the teeth of the animal on the tooth bar to prevent
the head moving up and down. This operation is much easier
than placing the ear bars. The tooth bar can be moved freely
backward and forward; softly, push the bar inside the rat mouth,
push the rat nose down, and pull the tooth bar out of the rat’s
mouth. If there is resistance to traction, the bar is in the correct
position and has hooked the incisors. Be sure to tighten the
screw that secures the position of the bar to the U frame. Clip
the nose of the rat by lowering the metal piece softly (Fig. 3c).
7. Cover the rat with a cloth to keep it warm; body temperature
decreases because of anesthetics. Alternatively, you may use a
thermal blanket lying below the animal.
8. It is very important not to damage the needle tip, just touch the
skull with the very tip, and no pressure is required at all. When
using a beveled needle, be much careful not to blunt the tip.
9. This allows finding the vertical projection of the desired struc-
ture on the skull surface. Move the needle down with much
care, looking for the touching point on the skull. Move the
needle up and mark the point with a pencil (a waterproof ink

pen is also fine); check that the needle touches the skull just in
the marked point.
The stereotaxic atlas let us to determine the coordinates of
the desired brain structure. Coordinates are expressed in mil-
limeters relative to bregma. These are the coordinates of some
structures we use frequently.
SN pars compacta: AP = −5.5, LM = ±1.8, DV = −8.3;
Striatum: AP = +0.5, LM = ±3.0, DV = −6.4;
Hippocampus: AP = −3.6, LM = ±2.0, DV = −3.4.
Negative values have different meanings depending on the
coordinate; for the AP, negative means posterior; for the LM,
it means right hemisphere; and for DV, down (the only way to
reach the brain, actually).
10. Adjust drill speed to your convenience; low to medium speeds
can provide more confidence to beginners. No pressure is
required, just hold the drill and push down very softly with
your fingertips. Be very careful in this step; drill tip should
never touch the brain when drilling. In fact, meninges under
the bone should not be broken by the drill. Once the hole is
open in the bone, punch the meninges with a sterile needle.
There might be some bleeding now, especially when drilling
near the midline. Clean the blood with a sterile cotton pad. Do
not continue with the injection procedure if there is bleeding
through the hole to avoid putting blood into the brain with
the needle.
11. Depending on the shape of the drill tip, hole walls may not be
straight vertical, with the lower part of the hole narrower than
the upper. Due to the needle flexibility, if the needle touches
the lower hole border, its tip will end far from the desired
injection place although the needle is apparently straight over
the hole.
12. Do this very slowly; it helps turning the plunger softly (as if
screwed) instead of pushing it down.
13. Interestingly, other substances may need longer times.
14. Alternatively, use a heat lamp.
References
1. McGeer PL, Itagaki S, Boyes BE et al (1988) 3. Klegeris A, McGeer EG, McGeer PL (2007)
Reactive microglia are positive for HLA-DR in Therapeutic approaches to inflammation in
the substantia nigra of Parkinson’s and neurodegenerative disease. Curr Opin Neurol
Alzheimer’s disease brains. Neurology 20:351–357
38:1285–1291 4. Rogers J, Civin WH, Styren SD et al (1992)
2. McGeer PL, McGeer EG (2008) Glial reac- Immune-related mechanisms of Alzheimer’s
tions in Parkinson’s disease. Mov Disord disease pathogenesis. In: Khachatunan ZS, Blass
23:474–483 JP (eds) Alzheimer’s disease, new treatment
strategies. Marcel Dekker, New York, NY, pp intracerebral injection: relative contribution of
147–163 peripheral monocytes and activated microglia.
5. Compston A (1993) Inflammation and the Brain Res 724:55–66
brain. Mol Chem Neuropathol 19:47–64 12. Szczepanik AM, Fishkin RJ, Rush DK et al
6. Herrera AJ, Castaño A, Venero JL et al (2000) (1996) Effects of chronic intrahippocampal
The single intranigral injection of LPS as a new infusion of lipopolysaccharide in the rat.
model for studying the selective effects of Neuroscience 70:57–65
inflammatory reactions on dopaminergic sys- 13. Piani D, Frei K, Do KQ et al (1991) Murine
tem. Neurobiol Dis 7:429–447 brain macrophages induce NMDA receptor
7. Burrell R (1990) Immunomodulation by bac- mediated neurotoxicity in vitro by secreting
terial endotoxin. Crit Rev Microbiol 17: glutamate. Neurosci Lett 133:159–162
189–208 14. Chao CC, Hu S, Molitor TW et al (1992)
8. Bourdiol F, Toulmond S, Serrano A et al Activated microglia mediate neuronal cell
(1991) Increase in ω3 (peripheral type benzo- injury via a nitric oxide mechanism. J Immunol
diazepine) binding sites in the rat cortex and 149:2736–2741
striatum after local injection of interleukin-1, 15. Banati RB, Gehrmann J, Schubert P et al
tumour necrosis factor-α and lipopolysaccha- (1993) Cytotoxicity of microglia. Glia 7:
ride. Brain Res 543:194–200 111–118
9. Andersson PB, Perry VH, Gordon S (1992) The 16. Giulian D (1993) Reactive glia as rivals in reg-
acute inflammatory response to lipopolysaccha- ulating neuronal survival. Glia 7:102–110
ride in CNS parenchyma differs from that in 17. Giulian D, Vaca K, Corpuz M (1993) Brain
other body tissues. Neuroscience 48:169–186 glia release factors with opposing actions upon
10. Montero-Menei CN, Sindji L, Pouplard- neuronal survival. J Neurosci 13:29–37
Barthelaix A et al (1994) Lipopolysaccharide 18. Lees GJ (1993) The possible contribution of
intracerebral administration induces minimal microglia and macrophages to delayed neuro-
inflammatory reaction in rat brain. Brain Res nal death after ischemia. J Neurol Sci 114:
653:101–111 119–122
11. Montero-Menei CN, Sindji L, Garcion E et al 19. Paxinos G, Watson C (1986) The rat brain in
(1996) Early events of the inflammatory reac- stereotaxic coordinates, 2nd edn. Academic,
tion induced in rat brain by lipopolysaccharide London
Chapter 27
Analyses of Microglia Effector Function

Using CX3CR1-GFP Knock-In Mice
Jenny A. Garcia, Sandra M. Cardona, and Astrid E. Cardona
Abstract
The generation of bone marrow radiation chimeric mice is a beneficial tool to utilize when studying
inflammation of the central nervous system (CNS). It is widely accepted that blood-derived progenitors
are capable of populating the CNS during chronic diseases and severe injuries; however, they are neither
consistent nor efficient in doing so. The lack of the appropriate recruitment could explain delays in recov-
ery and repair after an increase of toxic proteins in chronic neurodegenerative diseases. With the inge-
nious development of bone marrow chimeric mice, some of these concerns can be addressed and allow us
to hypothesize about further implications and possible mechanisms that may lead to medicinal applica-
tions. Bone marrow chimeric mice are often used to distinguish the intrinsic versus extrinsic effects of
specific mutations. In our case, chimeras help us to better understand the role of CX3CR1 in microglia
and peripheral myeloid cells. To detect cell autonomous effects on myeloid cell differentiation, CX3CR1-
deficient mice are used as donors and wild-type mice are used as recipients. In order to detect effects on
the “immune cell environment,” wild-type donors are used for the transfer into Cx3cr1−/− recipients. The
resulting chimeric mice can then be used for the analysis of microglial motility, regulation of neuroinflam-
mation, and persistence. This technique can be applied to a broad spectrum of research ranging from
neurodegenerative diseases to viral and parasitic pathogenicity and everything in between. This protocol
describes the approach to generate chimeric mice and analyze the role of CX3CR1 in CNS inflammation
in bone marrow radiation chimeras.
Key words CX3CR1, Bone marrow, Chimeras, Microglia, Trafficking, Chemokines, Radiation
1 Introduction
Microglia are specialized, surveillant phagocytic cells [1–3] that col-

onize the CNS during embryonic development and are best known
for their ability to scan the environment with their extensive, motile
processes. The knowledge of responses and pathogenic functions of
microglia can be useful in developing inhibitors and therapeutic tar-
gets for disease modulation [3]. In the CNS, microglia are the only
cells that express CX3CR1, also known as the fractalkine receptor
[4]. It has been shown via in vivo models of autoimmune diseases
307
308 Jenny A. Garcia et al.
that in the absence of CX3CR1, dysregulation of microglial responses

occurs leading to neurotoxicity. CX3CR1 deficiency has been associ-
ated with neuronal cell death subsequent to lipopolysaccharide
(LPS) challenge [4]. To monitor the role of CX3CR1 in the CNS,
we utilized mice in which the Cx3cr1 gene was disrupted by the
insertion of DNA encoding green fluorescent protein (GFP), allow-
ing for the direct visualization of microglial cells [4–8]. The genera-
tion of bone marrow chimeric mice allows us to investigate the
effects of infiltration and adaptive immunity triggered by extraneous
bone marrow cells on their endogenous counterparts. Considerable
speculation of the chimeric model is focused on the ability of radia-
tion to break down the blood–brain barrier (BBB). This issue can be
addressed by shielding the brain from irradiation while knocking out
the rest of the immune system. The only downfall is that the efficiency
of chimerism is sacrificed due to the fact that skull bone marrow can
contribute to hematopoiesis [9]. To further investigate roles of adap-
tive immunity when exploring the pathogenicity of multiple sclerosis
(MS), we utilize the animal model experimental autoimmune
encephalomyelitis in conjunction with the chimeric model [2, 10].
In such an experiment, C57BL6 recipient mice are challenged with
MOG35–55 peptide 6 weeks post-chimerism via active or passive
immunization. A variety of combinations can be investigated when
using this approach. For instance, CD45.1 or Ly5.1 donor cells defi-
cient of CX3CR1 can be tracked in recipient CD45.2 or Ly5.2 WT
mice via flow cytometry analysis by the detection of the CD45.1
marker in addition to the microglia expressing GFP. Alternatively,
CX3CR1-deficient mice can serve as recipients for CD45.2 Cx3cr1
WT donor cells. Overall, this protocol can be applied to any genetic
models of interest with potential implications in translational and
medicinal research in the development of systems to moderate
microglial function.
2 Materials
Prepare solutions and handle all tools under sterile conditions to

prevent any source of contamination.
2.1 Radiation 1. RS 2000 Rad Source Biological Irradiator with radiation output
of 160 kV, 25 mA, 0.3 mm Cu filter (Rad Source Technologies,
Alpharetta, GA, USA) or any X-ray machine or cesium source
of gamma irradiation available to your institution.
2. 1/16″ lead helmets (New Shield Inc., Pomona, NY, USA)
(see Note 1).
3. Anesthesia cocktail: Mix Ketaset and Xylazine in 1× HBSS for a
final concentration of 10 mg/ml Ketaset and 1 mg/ml Xylazine,
and inject 100 mg/kg Ketaset and 10 mg/kg Xylazine.
Microglia and CNS Inflammation 309
2.2 Bone Marrow 1. Mice: B6.SJL-Ptprca Pep3b/BoyJ (CD45.1 WT) #002014,

Isolation B6.129P (Cg)-Ptprca Cx3cr1tm1Litt/LittJ (CD45.1 KO), and
C57BL/6J (CD45.2 WT) #000664 (The Jackson Laboratory,
Bar Harbor, MA, USA).
2. 6-well plate (BD Falcon).
3. KimWipes or autoclaved paper towels.
4. 70 % ethanol.
5. 70 μM cell strainer (Fisher Brand #22363548 Fisher Scientific).
6. 2–200 ml beakers for holding tools in 1× PBS and 1 sterile
with 70 % ethanol for submerging mouse to disinfect skin.
7. Sterile 1× PBS.
8. Absorbent pads.
9. Paper towels.
10. Plastic bags to collect carcass.
11. Scissors.
12. Forceps.
13. 10, 50 ml polypropylene conical tubes.
14. 1.5 ml graduated microcentrifuge tubes.
15. 3 ml Luer-Lock tip syringes (BD #309657).
16. 0.4 % trypan blue diluted to 0.05 % in HBSS (Gibco).
17. Iscove’s/10 % FBS/0.5 % gentamicin reagent solution (fetal
bovine serum – heat inactivated) (Sigma-Aldrich).
18. Needles: PrecisionGlide 25G 5/8 and 23G ¾ (BD #305122 and
#305143, respectively).
19. Hemocytometer.
20. Bright-field microscope.
21. Carbon dioxide asphyxiation chamber.
2.3 Injection of Bone 1. Iscove’s Modified Dulbecco’s Medium supplemented with

Marrow Cells L-glutamine, 25 mM HEPES (Gibco #12440), and 50 μg/ml
gentamicin reagent solution (Gibco).
2. 1 ml tuberculin slip tip syringe with PrecisionGlide 27G ½
needle (BD #309623).
3. 1.5 ml graduated microcentrifuge tubes.
4. Isoflurane and vaporizer with oxygen chamber.
2.4 Flow Cytometry 1. Double-distilled sterile water.

Analysis of Peripheral 2. 1×HBSS free of calcium, magnesium chloride, and magne-
Blood sium sulfate (Gibco).
3. 10 mM HEPES.
4. 10× HBSS.
5. 15 ml conical tubes.
6. 1.5 ml microcentrifuge tubes.
7. Sterile 5 mm Goldenrod Animal Lancet.
8. FC block: rat anti-mouse CD16/CD32 clone 2.4G2 in cell
staining buffer (optimal concentration should be determined
by the investigator) (BD Pharmingen).
9. Cell staining buffer (BioLegend).
10. Fluorescent antibodies: CD45.1 anti-mouse PE-Cy7 clone
A20 in cell staining buffer (eBioscience) and CD45.2 anti-
mouse APC clone 104 in cell staining buffer (eBioscience)
(Note: optimal concentration should be determined by the
investigator).
11. Prepare Sorenson’s buffer by adding 3.59 g sodium phosphate
monobasic and 24.7 g sodium phosphate dibasic heptahydrate
to 400 ml ddH2O. Ensure that pH is between 7.0 and 7.3.
Adjust final volume to 500 ml with ddH2O and store at room
temperature.
12. 4 % paraformaldehyde in phosphate buffer (Sigma-Aldrich):
First, prepare 8 % PFA by dissolving 40 g paraformaldehyde to
400 ml ddH2O and stir for 15 min at 55–65 °C. Add 100 µl
5 M NaOH for 100 ml of 8 % PFA solution. Stir at 30 °C and
add ddH2O for a final volume of 500 ml. Add 200 ml of
Sorenson’s buffer to 500 ml of 8 % PFA and 300 ml ddH2O.
Filter, sterilize, and store at −20 °C.
13. Heparin sodium salt 5,000 U/ml (Sigma-Aldrich).
2.5 Flow Cytometry 1. Isotonic Percoll (GE Healthcare) buffered with 10× HBSS.
Analysis of Microglia Prepare 100 % ISP by adding 45 ml of isotonic Percoll to 5 ml
10× HBSS. Prepare 70 % Percoll by adding 21 ml of 100 %
ISP to 9 ml 1× HBSS. This is enough for 9 samples. Scale up
volumes as needed.
2. RPMI 1640 supplemented with L-glutamine without phenol
red (Gibco).
3. Dounce homogenizers for 7 and 15 ml volume with A (loose)
and B (tight) pestles (Pyrex, Wheaton).
4. S1 Pipet Filler (Thermo Scientific #9531) or Pipet-Aid.
5. 1× HBSS without calcium, magnesium chloride, and magne-
sium sulfate (Gibco) with 10 mM HEPES (Gibco).
6. FC block, antibodies, and reagents as in Subheading 2.4 in
addition to any other antibodies of interest.
3 Methods
3.1 Radiation 1. Subject the recipient mice (see Note 1) to radiation to deplete
Conditioning of the immune system. This will also help prevent any rejection
Recipient Mice of injected cells. Determine if your experiment requires total
body or brain-protected radiation.
2. For total body radiation, subject the recipient mice to 900
rads, a sublethal dose of radiation. Place all mice in an animal
cage that does not contain any metal and place cage into radia-
tion chamber with lid secured. Deliver the required amount of
radiation and remove cage from radiation chamber.
3. If you decide to protect the head from radiation treatment,
anesthetize the mice to ensure that they will not move around
in the cage potentially causing their protective helmet to fall
off (see Note 2). Once the mice have been anesthetized, roll
them on to their back and place the helmet onto the face of
the mouse. The idea is to protect the brain while keeping the
thymus exposed to radiation treatment. Once the helmets are
in place, replace the cage lid and place the cage into the radia-
tion chamber. For protected head recipients, it is important to
note that two consecutive doses of 600 rads are required.
4. To ensure a good percentage of chimerism, it is recommended
to treat recipient mice with radiation the same day just prior to
injecting bone marrow-isolated cells as seen in the next step.
5. If anesthesia was used, it is imperative to monitor the mice
until they regain consciousness.
3.2 Isolation of Bone 1. Euthanize donor mice by CO2 asphyxiation and submerge
Marrow Cells them in 70 % ethanol to disinfect the fur and skin. Place disin-
fected mouse onto absorbent pads or paper towels to collect
any bodily fluids or tissues that may be extracted.
2. Remove the skin from the lower limbs and cut away the mus-
cles revealing the femur and tibia bones.
3. After removing as much muscle as possible, cut the entire leg
off with scissors by cutting above the femoral head making
sure to keep the femur intact (see Note 3).
4. Once the hind limb is removed from the body, you can care-
fully cut off the foot making sure to keep the tibia bones intact.
Clean the bones with KimWipes or sterile paper towels of any
excess muscle and collect bones in a 6-well plate with Iscove’s
medium containing 10 % FBS and 50 µg/ml gentamicin and
keep on ice.
5. After collecting all hind limbs, separate the femur and tibia
bones by cutting at the patella, area between the femur and
tibia, making sure not to fracture the bones (see Note 4).
6. Using KimWipes or autoclaved paper towels, carefully remove

any excess tissue from the bones until you can clearly see the
epiphyses and entire shaft of the bone. This is important so
that you can see when the red bone marrow flushes out.
7. Fill a 3 ml syringe with Iscove’s/10 %FBS/50 µg/ml genta-
micin. Cut off the epiphyses at both ends of the femur and use
filled syringe with 25G 5/8 needle to flush out the bone mar-
row collecting the flow through in a 50 ml conical tube. While
holding the bone tightly with forceps, insert the needle inside
the bone in an up and down motion while pressing the plunger.
The bone will turn white as the cells are flushed out. Flip the
bone over and repeat flushing from the opposite end (see Note 5).
Follow the same procedure for the tibia.
8. After you have collected all marrow cells, pour them through
a 70 μM mesh filter and collect in a clean 50 ml conical tube.
If there are clumps on the filter, dissociate them with a syringe
plunger and wash with Iscove’s/10 % FBS/50 μg/ml genta-
micin to ensure that you recover as many cells as possible.
Aspirate supernatant and suspend pellet in 1 ml Iscove’s with
0.5 % gentamicin.
10. Count cells and adjust the final volume with Iscove’s with
50 μg/ml gentamicin to prepare cells at 130–140 × 106 cells/ml.
3.3 Injection of Bone 1. Anesthetize recipient mice with an isoflurane vaporizer by

Marrow Cells placing the mice in the gas-filled chamber.
2. Remove mouse from the chamber and inject 150–200 µl of
your cell suspension to deliver 10–20 × 106 cells per mouse via
retro-orbital injection using a 1 ml syringe with 27 G needle
(see Note 6).
3. Monitor recipient mice until they are fully awake.
4. Treat recipient mice with 50 μg/ml gentamicin in drinking water
for 2 weeks to avoid infection of immunocompromised mice.
5. Four weeks later, evaluate percentage of chimerism by flow
cytometry analysis of peripheral blood.
3.4 Flow Cytometry 1. Using a 5 mm lancet, draw 100–150 μl of blood from cheek
Analysis of Peripheral pouch and collect in 60 μl heparin sodium salt solution (see
Blood Note 7). Be sure to mix well to prevent coagulation. Transfer
blood to 15 ml conical tube and measure with a pipette to
determine the volume of water (ddH2O) needed to lyse red
blood cells. Determine the appropriate amount of ddH2O and
10× HBSS buffer needed to lyse your sample on a 20× final
volume. For example, to lyse 0.1 ml of blood, you will need
1.7 ml of water and 0.2 ml of 10× HBSS, so the final volume
of the lysed suspension is 2 ml (20× excess of the initial 0.1 ml
of blood) and in 1× buffer.
2. Have the appropriate volumes of water, 10× HBSS and 1×

HBSS/10 mM HEPES, measured out and ready as overtreat-
ment with water can lead to lysis of white blood cells which are
needed for staining. Add the correct amount of water, close cap
tightly, and mix by inversion for 20 s. Immediately add 10× HBSS
and invert tube twice to mix. Then add 5 ml 1× HBSS/10 mM
HEPES to ensure complete buffering of your cell suspension.
3. Centrifuge for 7 min at 700 × g at 4 °C. Aspirate supernatant
and suspend pellet in 1 ml cell staining buffer and transfer cells
to 1.5 ml microcentrifuge tube.
4. Centrifuge for 1 min at 8,000 × g at room temperature.
Aspirate supernatant and suspend in 50 μL of FC block
(CD16/32) to prevent nonspecific binding of antibodies (see
Note 8). Incubate on ice for 10 min.
5. Prepare antibody mix at appropriate dilutions for all samples. Be
sure to choose antibodies that will allow you to stain for donor
and recipient cells (i.e., if you are using CD45.1 donor mice and
CD45.2 recipients, be sure that you are using antibodies that
can identify those markers). Add 50 μl antibody mix to all sam-
ples and incubate for 30 min on ice protected from light.
6. Add 1 ml of cell staining buffer to wash any excess antibodies
from the cell suspension and centrifuge at 8,000 × g for 1 min
at room temperature.
7. Aspirate supernatant and suspend pellet in 150 μL cell staining
buffer and then transfer to equal volume of 4 % paraformalde-
hyde (PFA) making a final concentration of 2 % PFA. The final
volume of the cells depends on the volume needed to acquire
by your flow cytometer. Cells can be kept at 4 °C protected
from light until ready to acquire, but no longer than 36 h.
8. Analyze results by arranging markers for recipient versus donor
mice on the X and Y axis respectively. The resulting popula-
tions will show the percentage of chimerism (Fig. 1).
3.5 Flow Cytometric 1. Anesthetize mouse with Ketaset/Xylazine cocktail at

Analysis of Microglia 100 mg/kg Ketaset and 10 mg/kg Xylazine and perfuse with
1× HBSS by cutting the thoracic cavity open and cut the right
atrium of the heart allowing blood to flow out. Inject 20 ml of
ice cold 1× HBSS into the left ventricle to completely remove
blood from tissues.
2. Remove the brain from the skull and obtain the spinal cord by
flushing with 1× HBSS (see Note 9). Transfer tissues to 6-well
plate containing RPMI and keep on ice until all tissues are
collected.
3. Fill Dounce homogenizers with 3.5 ml RPMI without phenol
red, label them accordingly, and keep on ice. It is possible to
homogenize up to two brains in one 7 ml homogenizer. When
combining more than two brains, use a 15 ml homogenizer.
Fig. 1 Peripheral blood analysis from a CD45.2 WT mouse reconstituted with

CD45.1 KO bone marrow cells. CD45.1 cells are stained with anti-mouse CD45.1
phycoerythrin-Cy7 (PE-Cy7). CD45.2 (recipient) cells are stained with anti-mouse
CD45.2 allophycocyanin (APC). Data shows 96.6 % CD45.1 reconstitution of the
CD45.2 recipient indicating a successful chimerism
4. Transfer tissues to their respective homogenizer and grind the

tissue using the “A” or “loose” plunger until the suspension
has reached confluency and then use the “B” or “tight”
plunger. Once the suspension is homogeneous, transfer the
suspension to a 15 ml conical tube and top off with RPMI to
7 ml. Continue this step with all samples and keep on ice
before moving to the next step.
5. Add 3 ml of 100 % Percoll solution to the cell suspension and
mix well to gain confluency. Slowly layer the homogenized
suspension onto a new 15 ml conical tube containing 2 ml of
70 % Percoll. The idea is to set up a density gradient in which
the brain mononuclear cells including microglia will layer atop
the 70 % Percoll (see Note 10).
6. Centrifuge at 550 × g for 30 min at 18 °C.
7. Locate the interphase which contains the microglia just above
the 2 ml graduation on the 15 ml tube. It is easily seen as a
cloudy halo when held up to the light. Using a transfer pipette,
carefully remove the interphase, approximately 2 ml, and
transfer to a 15 ml tube containing 8 ml of 1× HBSS/10 mM
HEPES to wash cells of Percoll.
8. Centrifuge for 7 min at 700 × g at 4 °C. Aspirate supernatant
and suspend pellet in 1 ml cell staining buffer and transfer to
1.5 ml tubes.
9. Centrifuge cells for 1 min at 8,000 × g, aspirate supernatant, and
suspend pellet in volume necessary for blocking step (see Note
8). Incubate cells in FC block solution for 10 min on ice.
Fig. 2 (a) Cell homogenates were prepared from the brain of EAE-affected Cx3cr1gfp/gfp → wild-type chimeric
mice and separated over density Percoll gradients. CD45 (APC) antibody was used to locate the CD45hi popula-
tion of hematogenous cells and CD45lo or resident microglia. (b) From the same population of cells seen in (a),
we can determine that 18.7 % of the population is GFP positive (FITC) indicating donor microglia. The other
population (18.4 %) is FITC negative representative of resident microglia. Additional markers can be incorpo-
rated to determine differences in effector function between the different experimental groups
10. Without washing, add equal volume of antibody mix to the

cells and incubate for 30 min on ice protected from light.
11. Wash cells of any excess antibody by adding 1 ml cell staining
buffer to each tube and centrifuge for 1 min at 8,000 × g.
12. Aspirate supernatant and resuspend pellet in cell staining buf-
fer and then transfer to equal volume of 4 % paraformaldehyde
(PFA) for a final concentration of 2 % PFA. The final volume
of the cells depends on the volume needed to acquire by your
flow cytometer. Cells can be kept at 4 °C protected from light
until ready to acquire.
13. To analyze microglia, you must first distinguish your donor
cells from your recipients based on the expression of CD45
and the presence or absence of GFP (see Note 11). With
CD45 on the Y axis and SSC on the X axis, locate the microg-
lial population expressed as CD45lo. This population denotes
both donor and recipient microglia. By changing the X axis to
FITC to observe the presence of GFP, it is now possible to
differentiate donor from recipient microglia. (Note: other
antibodies can be included to compare effector function or
proliferation) (Fig. 2).
4 Notes
1. Select donor and recipient mice properly, ensuring the differ-

entiation of peripheral cells from injected cells by using differ-
ent strains of mice that you can identify by flow cytometry
staining. For example, in the figures shown, we used CD45.1

KO donor mice and CD45.2 WT recipients. Anti-CD45.1-
and anti-CD45.2-specific antibodies are then used to deter-
mine the percentage of chimerism of the recipient mice. Our
Cx3cr1−/− mice contain a knock-in of GFP within CX3CR1
making it possible to determine KO from WT cells.
2. To protect the mouse brain from radiation, you must first
anesthetize it using 100 mg/kg Ketaset/10 mg/kg Xylazine.
For this procedure, inject 100 μl of the solution for every 10 g
of body weight via intraperitoneal (i.p.) route. Secure the hel-
met as discussed in the procedure and treat with correct dos-
age of radiation. Be sure to monitor the mice until they regain
righting reflex and normal activity which takes approximately
1–2 h after injection of cocktail.
3. It can be helpful to cut into the spine just above the femoral head
when detaching the limb. This will ensure that you will not break
the femoral bone resulting in the loss of cells. After removing the
skin and muscle, locate the femoral epiphysis and cut into the
spine just above it. Once you have removed the limb, you can cut
off any excess tissue and spinal bones with greater ease.
4. You can separate the bones by bending the knee backwards
with a twisting motion until the bones detach. This aids in
keeping the femur and tibia bones intact.
5. As you flush out cells from either end of the bone shaft, you
might find a population of cells located in the middle portion of
the shaft. Once you see that the epiphyseal end of the bone is
white, you can trim the bone to better reach the middle of the
shaft. Take care when cutting the bone as splitting can occur.
6. Retro-orbital injections take some practice. It is recommended
to practice by injecting 150 μl of saline or PBS into other mice
prior to your experiment. After you have anesthetized the
mice as listed in the procedure, roll the mouse on its side.
Secure the head with your index finger behind the skull and
thumb under the chin. Insert the needle at a 45° angle into
the conjunctiva, inner corner of the eye, until you feel a soft
spot about a quarter inch deep. Begin to inject saline/cells. If
you see the fluid coming out of the eye, you are not deep
enough. Insert the needle deeper and continue injecting fluid.
After injecting the correct amount of saline/cells, remove the
needle and blot any excess fluid from the eye. The mice should
wake within 2 min. Be sure to monitor the mice until they
regain consciousness.
7. First, locate the submandibular vein. This is usually found
behind the cheek below the ear and above the jaw bone. Make
a deep puncture with an animal lancet and collect up to 200 μl
of blood in a tube with 60 μl heparin. Mix blood with heparin
instantly to prevent coagulation.
8. It is recommended to suspend sample pellets in 50 μl of FC

block solution for incubation. After blocking the cells, add
50 μl of the antibody mix to the cells in FC block for a final
volume of 100 μl.
9. To flush out the spinal cord, you will need to remove the
entire spinal cord from the carcass. With a 10 ml syringe and
blunt 16 gauge needle, find the spinal cord within the column;
hold the column tight while you press the plunger with great
force. It is important to hold tight to the spinal column as the
suction is required to expel the spinal cord. The spinal cord
should shoot out in one piece.
10. If cells are pipetted too fast or into the 70 % Percoll instead of
on top, the cells will be lost indefinitely. It is helpful to use the
gravity or lowest setting on the pipette if possible.
11. As previously stated, we use CD45.1 donor mice and use the
CD45.1 anti-mouse PE-Cy7 to identify those cells. We use
CD45.2 anti-mouse APC to label our recipient endogenous
cells. The Cx3cr1−/− mice contain GFP which allows us to
visualize donor microglia. The CD45lo GFP- population rep-
resents the resident microglia. The CD45lo GFP+ population
of cells is representative of CD45.1 donor microglia. CD45hi
populations are indicative of infiltrating monocytes and can
also be segregated by the presence or absence of GFP.
Acknowledgments
This work was supported by NIH SCIGM095426 and National

Multiple Sclerosis Society TA-3021-A-1 grant to AEC.
References
1. Soulet D, Rivest S (2008) Bone-marrow- 6. Geissmann F, Gordon S, Hume DA et al

derived microglia: myth or reality? Curr Opin (2010) Unravelling mononuclear phagocyte
Pharmacol 8:508–518 heterogeneity. Nat Rev Immunol 10:453–460
2. Mizutani M, Pino PA, Saederup N et al (2012) 7. Auffray C, Fogg DK, Narni-Mancinelli E et al
The fractalkine receptor but not CCR2 is pres- (2009) CX3CR1+ CD115+ CD135+ com-
ent on microglia from embryonic development mon macrophage/DC precursors and the role
throughout adulthood. J Immunol 188:29–36 of CX3CR1 in their response to inflammation.
3. Prinz M, Priller J, Sisodia SS et al (2011) J Exp Med 206:595–606
Heterogeneity of CNS myeloid cells and their 8. Cardona AE, Sasse ME, Mizutani M et al
roles in neurodegeneration. Nat Neurosci (2008) Scavenging roles of chemokine recep-
14:1227–1235 tors: chemokine receptor deficiency is associ-
4. Cardona A, Pioro EP, Sasse ME et al (2006) ated with increased levels of ligand in circulation
Control of microglial neurotoxicity by the and tissues. Blood 112(2):256–263
fractalkine receptor. Nat Neurosci 9:917–924 9. Prinz M, Mildner A (2011) Microglia in the
5. Jung S, Aliberti J, Graemmel P et al (2000) CNS: immigrants from another world. Glia
Analysis of fractalkine receptor CX(3)CR1 59:177–187
function by targeted deletion and green fluo- 10. Ransohoff RM, Cardona AE (2010) Microglia:
rescent protein reporter gene insertion. Mol myeloid cells of the central nervous system
Cell Biol 20:4106–4114 parenchyma. Nature 468:253–262
Chapter 28
In Vivo Two-Photon Microscopy of Microglia

Satoru Kondo and Shigeo Okabe
Abstract
In vivo imaging with two-photon microscopy is becoming an indispensable technique to investigate
cellular and subcellular phenomenon in living tissues including the central nervous system. This micros-
copy enables to image dynamics of molecules, morphology, and excitability with minimal invasion to tis-
sues. Microglia are residual immune-responsive cells in the central nervous system and show highly
dynamic response to the environmental alterations. Diverse roles of microglial functions in the intact and
pathological brain are still largely unknown. In this chapter we describe the detailed method to image the
dynamics of microglia in the mouse brain in vivo.
Key words Microglia, Two-photon microscopy, In vivo imaging, Cell dynamics, Mouse
1 Introduction
Among in vivo imaging techniques, two-photon microscopy is the

only method so far to image biological phenomenon with both
spatially and temporally high resolution. This method was first
applied for brain and now has been used for various tissue observa-
tions. The central nervous system (CNS) is mainly composed of
neurons and glia whose cell diameter is only around 10 μm with
fine subcellular structures. In the brain, two-photon microscopy
has revealed the neuronal dynamics of structure by observing the
morphology of fluorescently labeled neurons and excitability by
measuring the calcium dynamics with calcium indicator. Historically,
glia have been long regarded as static cells that only support the
structure of CNS. However, recent findings suggest that glia are
rather highly dynamic cells and indispensable for the neuronal
functions. Microglia are one of glial subtypes in the CNS. They are
quiescent cells with fine ramified processes in the non-pathological
state but once insult occurs, they rapidly become activated (pro-
cesses are retracted and thicken, become amoeboid) and migrate
to the site of injury [1]. Microglia are electrically non-excitable but
319
320 Satoru Kondo and Shigeo Okabe
structurally highly dynamic. Several transgenic mouse lines that

express fluorescent proteins specifically in microglia are available
[2, 3]. By utilizing such mice, microglial dynamics in the intact or
pathological state can be investigated [4–8]. These studies revealed
that even in a non-pathological state, their fine processes are highly
motile and continuously survey brain parenchyma. This protocol
describes the imaging of microglia in the living mouse brain using
two-photon microscopy.
2 Materials
2.1 Reagents 1. Mice: Iba1-EGFP transgenic mice or Cx3cr1-EGFP knock-in

mice are useful to image structural dynamics of microglia in
vivo. Both mice express EGFP in microglia, and fine processes
can be imaged. Iba1-EGFP transgenic mice were generated by
Hirasawa et al. [2]. Iba1 (ionized calcium-binding adaptor
molecule 1) is specifically expressed in microglia and macro-
phages, and EGFP is expressed under the regulation of this
gene promoter. The availability of the mice should be addressed
to the authors. CX3CR1 (CX3C chemokine receptor 1) is a
receptor protein for chemokine and expressed in microglia,
monocyte, and other immune-related cells. Cx3cr1-EGFP
mice were generated by Jung et al. [3] and are available from
The Jackson Laboratory (stock # 005582).
2. Anesthetics: Ketamine hydrochloride salt (50 mg/ml).
Xylazine hydrochloride (20 mg/ml). Ketamine/xylazine mix-
ture is prepared at the concentration of 10 mg/ml of ketamine
and 1 mg/ml of xylazine diluted in saline. Lidocaine hydro-
chloride (2 % xylocaine jelly, 2 % xylocaine solution).
3. 0.9 % Saline: Saline is consisted of 0.9 % NaCl. Sterilized solu-
tion is available from pharmacy or can be prepared in labora-
tory by autoclaving.
4. Disinfectant (0.1 % Gentacin ointment).
5. Iodine tincture (povidone-iodine solution).
6. Ophthalmic ointment (Mycochlorin).
7. Absorber: High absorbable cotton paper.
8. Absorbable gelatin sponge.
9. Cotton buds.
10. Adhesive: Jelly-type instant adhesive.
11. Hair removal cream (Nair, USA).
2.2 Equipments 1. Two-photon excitation microscopy: Commercially built micro-

and Tools scopes are available from some companies (e.g., FV1000MP,
In Vivo Two-Photon Microscopy of Microglia 321
Olympus, Japan; A1RMP, Nikon, Japan; LSM7MP, Zeiss,

Germany). These microscopes are fully equipped, ready to use,
and almost maintenance-free. Half custom-made microscopy
can be constructed by purchasing system or kit from some com-
panies (e.g., Thorlabs, USA; Prairie Technologies, USA; Sutter
Instruments, USA; Femtonics, Hungary; and Scientifica, UK).
For the construction of full custom-made microscopes, refer
some literatures [9].
2. Low-magnified, high numerical aperture, water-immersion
objective lens (e.g., XlPlan N 25× 1.05 NA, Olympus, Japan;
CFI Apo LWD 25× 1.1 NA, Nikon, Japan).
3. Drill and bur: High-speed (ca 30,000 rpm) drill. Micro drill
bur (φ0.7 mm, Fine Science Tools, USA). It is better to choose
micro drill bur made from sterilizable stainless steel.
4. Scissors (BH-12S, Napox, Poland).
5. Forceps (Dumont #5 and #7, Fine Science Tools, USA).
6. Needle holder (Olsen-Hegar with scissors, Fine Science Tools,
USA).
7. Microsurgical blade.
8. Scalpel.
9. Surgical needle with suture.
10. Head plate: Custom-made. 60 mm (L) × 10 mm (W) × 0.5 mm
(H) stainless steel (Fig. 1).
11. Head-holding stage: Custom-assembled. All parts are avail-
able from some companies (e.g., Breadbord MB4, Clamping
Fork CF175, Pedestal Post Holder PH1E, Stainless Steel Post
TR1.5, Thorlabs, USA) (Fig. 1).
12. Balance: This is for weighing mouse. Readability of 0.1 g is
preferable.
13. Cold light source: High luminance LED is preferable to use.
14. CCD or Digital camera: Either black/white or color CCD
camera can be used. Blood vessels can be distinguished easily
with color CCD camera. Compact digital camera for photog-
raphy can be alternatively used. Attachment parts to mount on
various microscopes is available from some companies (e.g.,
Microscope Network Co., Japan).
15. Stereomicroscope: Microscope with large field of view and
high magnification (ca 40×) is recommended (S6, Leica,
Germany).
16. Heat pad: Temperature-controlled heat pad with rectal probe
(DC Temperature Control System Cat# 40-90-8, FHC,
USA). Pet heat pad (Flexiguard, Petnap Ltd., UK).
17. Shaver for small animal.
Fig. 1 Experimental equipments to hold the head of mouse. (a) Head-holding

stage is shown. Height of the pole is adjustable by the screw on the side, depend-
ing on the size of mouse. Head plate is fixed on top of the poles with screw.
(b) Head plate is shown with its sizes. This is made of stainless steel and can be
reused repetitively. Square hole in the middle is for the observation window
3 Methods
To perform in vivo imaging of fluorescently labeled cells in the

brain, skull that hinders the penetration of excitation and emission
light should be treated to increase the transmission of light. To this
end, there are two methods to prepare the observation window in
the skull [10, 11]. One is called open-skull and other is thinned-
skull window. Open-skull window is formed by a craniotomy, in
which a part of skull is removed and replaced with cover glass.
Thinned-skull window is prepared by scraping a part of skull with
microsurgical blade to the thickness of 10–20 μm but leaves the
skull intact. Distortion of light is smaller in open-skull window and
better images can be obtained from deeper depth of tissue than
thinned-skull window. However, it is reported that open-skull win-
dow activates microglia and astrocytes by inflammatory processes
[9], yet this is still in debate [11]. Thinned-skull window can avoid
activation of microglia and astrocytes, but optical accessibility is
limited. Due to the larger light scattering and aberration by the
remained skull, maximal depth of imaging of fine structure is

limited to 100–200 μm from the cortical surface. Despite of some
limitations for the observation, transcranial imaging might be pref-
erable for the studies of microglial dynamics that is sensitive to
minimal invasion to the brain tissue.
Microglia is present not only in the brain but in the spinal cord as
well. Different technique is necessary for the spinal cord imaging [12].
3.1 Attachment of 1. Sterilize all the surgical tools and materials before surgery
Head Plate to the (see Note 1).
Mouse Skull 2. Place the mouse in a small box and weigh it. Administer
ketamine/xylazine anesthesia based on the body weight
3. Transfer mouse on the temperature-controlled heat pad
(see Notes 4 and 5). Protect mouse’s eyes with ophthalmic
ointment.
4. Shave fur with shaver from the base of ears to eyes. Alternatively,
hair remover can be used.
5. Disinfect the skull skin with 70 % ethanol and iodine tincture.
6. Apply local anesthesia (2 % lidocaine) into the skull skin.
7. Make incision at the midline from ears to eyes and retract skin
left and right. Remove the periosteum of the skull and clean
the surface (see Notes 6 and 7).
8. Thoroughly dry up the skull surface (see Note 8). Mark the
center of the area where images are going to be taken.
9. Put glue on both skull and head plate and then attach a head
plate to the skull tightly being the marked spot in the center
10. Move the mouse to the head-holding stage and attach head
plate to the head-holding stage tightly (see Note 11).
3.2 Thinned-Skull Thinning process consists of two steps. First step is carried out by
Surgery high-speed drill for wider area, and second step is done by microsur-
gical blade and narrows the thinning area. To avoid the mechanical
damage to the brain parenchyma (to avoid the activation of microg-
lia) during the thinning, the area for the optical access (thinned-skull
window size) should be limited to 0.5 mm square (ca. 0.3 mm2).
1. Determine the thinning area for the first step (see Note 12).
The area of thinning is ca. 1.5 mm square (Fig. 2a).
2. Start thinning with high-speed drill (see Note 13). The speed of
drill is 20,000 rpm keeping the drill bur angle at 30° (see Note 14).
3. The skull should be cooled frequently with cold saline to pre-
vent the heating due to the frictional heat between bone and
drill bur.
Fig. 2 Example photographs of the thinned-skull window. (a) A low-magnified photograph after the thinned-
skull preparation. The thinning areas for the high-speed drill (blue) and for the microsurgical blade (green) are
shown. The sizes of area are 1.5 mm square and 0.5 mm square, respectively. Black arrow shows blood ves-
sels in the skull. (b) After the experiment, the imaged area is marked (dotted square). (c) High-magnified
photographs of imaged area. Imaged area is marked with dotted square. Vasculature pattern helps to find the
same area for the next imaging session. Scale bars, 1 mm
4. The thinning is performed until the spongy bone is removed (see

Note 15). When the thinning reached to the third layer (com-
pact bone), stop the thinning and investigate the thickness (see
Note 16). At this stage, the skull thickness is still around 100 μm.
5. Determine the final thinning area (0.5 mm square) (see Note
17 and Fig. 2a).
6. The thinning is further continued with drill. If the bone
becomes too soft to thin with drill, thinning is switched to pur-
sue with microsurgical blade by hand. The thinning should be
done with greater caution. It is necessary to avoid pushing
downward to the skull. Keep the angle of the microsurgical
blade at 45° (see Note 18).
7. Stop thinning when thickness reached 10–20 μm (see Note 19).
8. Take image of cranial window to record the thinning area by
CCD camera attached to the stereo microscope (see Note 20
and Fig. 4a).
3.3 Imaging Choice of objective lens is quite important to obtain the bright and
Microglia with high-resolution images with low excitation laser power. Higher
Two-Photon laser power exposure may cause the microglial activation. Recently
Microscopy released objective lenses with low magnification and high numeri-
cal aperture are recommended to use. Olympus 25× NA1.05 or
Nikon 25× NA1.10 can achieve better excitation of fluorophores
and better collection of fluorescence in vivo.
Some methods to improve the excitation and detection effi-
ciencies are recently reported, but most of them are still not
installed in the commercially built microscopes (e.g., adaptive
optics for wave-front correction or passive pulse splitters).
1. Transfer the mouse mounted on the head-holding stage under
the microscope. Set and fix the position of mouse properly.
2. By using microscope’s ocular and low-magnified objective lens
(4×), adjust the focus on the skull surface by epi-illumination
with cold light. Center the cranial window and take photograph
by CCD camera attached on two-photon microscopy to record
the vasculature pattern for the future relocation (Fig. 2a).
3. Change objective lens to higher magnification and immerse
saline between objective lens and skull. Shield with blackout
curtain or plate to prevent the light leakage.
4. By using microscope’s ocular and epi-illumination with cold
light, adjust the image area by comparing the previously taken
low-magnified photograph (Fig. 2a). Vasculature pattern can
be used as a landmark.
5. Switch to the LSM mode and start imaging. Set the HV value
of PMT and laser intensity as appropriate (see Note 21).
6. If the focus is on top of the skull, initial image is the autofluo-
rescence from skull. As the objective lens goes downward,
subsequent image is the autofluorescence from thin layer of
dura mater. Finally fluorescently labeled microglia will appear
(see Note 22).
7. Acquire imaging data as required (see Notes 23 and 24)
(Fig. 3).
8. At the end of each imaging session, move up the objectives to
the surface and take photograph to record the vasculature pat-
tern (Fig. 2c) and mark the position of imaged area on the
low-magnified photograph of blood vessel pattern (see Note
25 and Fig. 2b).
9. During the imaging, take care of the depth of the anesthesia
and administer the additional anesthetics as necessary.
10. After all the imaging, detach the mouse from head holder and
remove the head plate from the skull carefully. Clean the cranial
b1 b2
c d
10 : Process 1
Motility of processes
Motility of processes
: Process 2 4
(µm/3 min)
(µm/3 min)
: Process 3
0
2
−10 0
5 15 25
Time (min)
Fig. 3 An example of in vivo time-lapse imaging of microglial processes. (a) Morphology of microglia imaged
by in vivo two-photon microscopy of Iba1-EGFP mice. Time-lapse imaging was performed, and images with
15 min intervals are shown. Scale bar, 10 μm. (b1) This image shows the binary image shown in A at time point
of zero min, with an overlay of green lines (numbered from 1 to 10) on individual processes. The tips of these
green lines were recorded, and their distances between adjacent time frames (3 min intervals) were calculated
as an index of process motility shown in c and d. Scale bar, 10 μm. (b2) Binary images of a microglial process
(green line) extracted from time-lapse sequences in the image area marked by a dotted rectangle in b1, illus-
trating the extent (yellow line) of rapid growth and shrinkage (red line). Numbers in the upper left corner of
images indicate elapsed time in minute. Scale bar, 2 μm. (c) The motility of microglial processes. Extension
and retraction of processes were plotted for three representative processes with time intervals of 3 min.
Positive values indicate process extension, and negative values indicate retraction. (d) Summation of absolute
distances of extension and retraction during the observation period of 30 min was calculated, and the average
speeds per frame (3 min) were estimated from 10 processes
window with sterilized water and suture the skin. Put the dis-
infectant on the sutured skin and return the mouse into the
cage. Keep the cage on the heat pad until awakening.
11. Clean the objective lens with appropriate cleaning solution and
disinfect the area under the microscope with 70 % ethanol.
3.4 Repetitive Bone has ability to grow again, and skull can gradually recover the
Imaging thickness where thinned-skull operation was carried out. Therefore,
skull rethinning is necessary, depending on the situation, before
the repetitive imaging. The degree of bone regrowth depends on
the time interval of imaging. Usually within 2–3 days after the last
imaging, it is not necessary to do the rethinning. However, after
around 4 days, connective tissue appears on the cranial window
and regrowth of bones becomes evident after a week. Basically
rethinning can be performed by the same procedure as the thin-
ning process at the first time. Regrowth bone is something differ-
ent from the original bone. It has inhomogeneous hardness and
irregular structure. Because the difficulty of rethinning increases as
the number of times of this process, maximum number of times
that imaging can be performed is limited to 3–4.
Recently transparent cyanoacrylate cement can stabilize the
thinned-skull window and prevent bone regrowth, achieving the
repetitive observation without rethinning [13].
1. Place the mouse in a small box and weigh. Administer ket-
amine/xylazine anesthesia based on the body weight (see
Notes 2 and 3).
2. Transfer mouse on the temperature-controlled heat pad (see Notes
4 and 5). Protect mouse’s eyes with ophthalmic ointment.
3. Remove fur if necessary. Disinfect the skull skin with 70 %
ethanol and iodine tincture.
4. Apply local anesthesia (2 % lidocaine) into the skull skin.
5. Make incision at the midline from ears to eyes and retract skin
left and right.
6. Confirm the location of the observation window comparing
the photograph that was taken at the previous imaging
(Fig. 28.4a). Clean up the skull with cotton buds or soft paper
not to damage the observation window (see Note 26). At this
stage do not clean the observation window yet.
7. Put glue on both skull and head plate and then attach a head
plate to the skull tightly being the observation window in the
center (see Notes 9 and 10).
8. Move the mouse with heat pad to the head-holding stage and
attach head plate to the head-holding stage tightly (see Note 11).
9. Start cleaning the observation window. Under the stereomicro-
scope investigate the condition of cranial window. If it is covered
Fig. 4 Bone regrowth occurs after long-time intervals. (a) Photograph of thinned-skull window at day 0. High
transparency of thinned-skull area is observed. (b, c, d) Photographs of thinned-skull window at day 21. (b)
Thinned-skull area is covered with connective tissues. (c) Connective tissues are thoroughly removed.
Transparency of thinned-skull window is lost, and bone regrowth evidently occurred. (d) Rethinning with high-
speed drill and microsurgical blade recovered the high transparency of the observation window. Scale bar, 1 mm
only with a small piece of tissue, high transparency of thinned-

skull window should be maintained beneath. Usually after the
removal of the tissue observation can be done without the
necessity of rethinning. If the cranial window is covered with
dense connective tissue and blood vessels, bone regrowth could
occur (Fig. 4b). Firstly, connective tissue should be thoroughly
removed with scalpel or microsurgical blade. Untransparent
regrowth bone will appear (Fig. 4c). Secondly, skull rethinning
is curried out with high-speed drill and microsurgical blade as
described in the Subheading 3.2. As mentioned earlier, the
quality of regrowth bone is something different from the origi-
nal bone. More attention should be paid to perform the skull
rethinning than the first-time thinning.
10. Transfer mouse under the microscope and set the position of
the mouse on the head-holding stage.
11. By using low-magnified objective lens (4×) and microscope’s
ocular, observe the skull surface by epi-illumination with cold
Fig. 5 An example of time-lapse imaging over different days. (a1, a2) Low-magnified images of Iba1-EGFP
mouse. Same areas taken at two different days are shown. Several microglia can be seen, and cell bodies of
microglia observed on both days are marked with arrow and arrow heads. Most of the microglia remained allo-
cated at the same position during this time period (2 days). (b1, b2) Higher-magnified images of microglia from
a1 and a2 (indicated by arrow) are shown. Although the position of cell bodies remained unchanged, extension
patterns of their processes are largely altered as can be predicted from the short-term imaging (Fig. 3)
light. Compare observation window with the reference image

that was taken previously and locate the area where imaging
was performed previously (see Note 27).
12. Acquire imaging data as required (see Note 28) (Fig. 5).
13. After all the imaging, detach the mouse from head holder and
remove the head plate from the skull carefully. Clean the cra-
nial window with sterilized water and suture the skin. Put the
disinfectant on the sutured skin and return the mouse into the
cage. Keep the cage on the heat pad until awakening.
14. Clean the objective lens with appropriate cleaning solution and
disinfect the area under the microscope with 70 % ethanol.
3.5 Simultaneous Besides the immune function of microglia, increasing number of

Imaging of Microglia studies has demonstrated its interactions with neurons, especially
and Neuron at the synapses during physiological and pathological states.
Simultaneous two-photon in vivo imaging of microglia and neu-
rons could elucidate the roles of microglia in the synapses. To char-
acterize the microglia-synapse interactions, double transgenic mice
(Iba1-EGFP/Thy1-YFP or Cx3cr1-EGFP/Thy1-YFP), in which
both microglia and layer V neurons are fluorescently labeled, are
used [6, 7, 12]. Although these mice are quite useful for simulta-
neous imaging, emission wavelength of EGFP and YFP are very
close to each other, and it is difficult to separate these signals with
optical filter [14]. One of the alternative ways is to perform in
utero electroporation of gene that codes red fluorescent protein to
Iba1-EGFP or Cx3cr1-EGFP mouse embryo and label neurons.
Gene expression in neurons persists for a long time, and imaging at
later postnatal stage is possible. Red fluorescent protein is a better
choice than the neuronal labeling. Basically there is no large differ-
ence for the imaging method as mentioned in the earlier sections.
Detailed description of in utero electroporation technique is out
of the scope of this chapter. Please refer appropriate literatures [15]:
1. In utero electroporation of gene is performed to Iba1-EGFP
or Cx3cr1-EGFP mouse embryos and raise mouse until appro-
priate age (see Note 29).
2. Surgical preparations and setup for observation are carried out
by the same methods as described in the Subheadings 3.1–3.3.
3. Acquire imaging data as required (see Note 30) (Fig. 6).
4. Finish the imaging procedure as described in the
Subheading 3.3.
4 Notes
1. All the surgical tools made of stainless steel should be prepared

to sterilize in the dry heat oven. Sterilization for other surgical
instruments and materials can be achieved by autoclave (high-
pressure steam). Autoclaved materials can be dried up in the
low-temperature oven kept at 50 centigrade.
2. The dose is 10 μl/g body weight (0.1 mg of ketamine and
0.01 mg of xylazine/g body weight). This mixture of anesthe-
sia can be administered either intraperitoneally or
intramuscularly.
3. Alternatively, isoflurane or other anesthesia can be chosen. In
terms of the controllable and stable anesthesia, isoflurane is pref-
erable, but from our experiences, motion movement of images is
less in mouse anesthetized by ketamine/xylazine mixture.
Fig. 6 Example images of simultaneous dual color imaging of microglia and neurons. Simultaneous dual color
time-lapse imaging was performed with Iba1-GFP mouse, in which red fluorescent protein dsRed2 is expressed
in neurons by in utero gene electroporation method. Cell body of microglia (green) is marked as M, and neu-
ronal processes (red) are visible. Time-lapse images over 15 min show dendritic structures contacted by
microglia (yellow, white, and red arrowheads). Numbers in the upper left corner of images indicate elapsed
time in minute. Scale bar, 10 μm
4. In the case of isoflurane, mouse is placed on the heating pad

attaching nose cone for maintaining anesthesia.
5. Confirm that mouse is deeply anesthetized by checking the
absence of paw withdrawal from pinching the toe. If with-
drawal reflex remains, adjust anesthetic level by additional
administration of ketamine/xylazine or increasing the isoflu-
rane concentration.
6. Sometimes bleeding from skull occurs during the cleaning,
but it stops spontaneously within a short while. However if it
prolongs, resolve the bleeding by using gel form.
7. To attach head plate, it is not necessary to remove cervical
muscle for somatosensory or motor cortex imaging, but for
other areas of the cortex, retract or remove cervical muscle as
necessary.
8. This process is quite important. If the surface of skull is not
exposed enough and cleaned, metal head plate will not be
adhered firmly. Immediate adhesive glue is convenient to use.
The head plate should be fixed tightly during surgery and
imaging period yet be detached from skull after imaging. And
also head plate is ideally adhered to the skull as immediate as
possible.
9. It is important to check that the surface of the head plate and
the skull is in parallel. In some area of cortex, the skull outline
and cortical surface are not in parallel.
10. Usually wait for about 10 min until the head plate attaches to
skull completely.
11. By narrowing the free space for the head of the mouse (the dis-
tance between the top of the head plate and the base of the
head holder), spontaneous movements of mouse due to the
breath or heartbeat is partially suppressed.
12. It is quite important to avoid large blood vessel within the
observation window. Before starting the thinning process,
confirm that there is no large blood vessel under the area that
was marked in the previous section. Vasculatures on the corti-
cal surface become visible by immersing the skull surface with
saline. If any large vessel is visible, it is better to avoid it by
shifting the thinning area.
13. The thinning should be performed uniformly and keep the
skull surface flat.
14. Great caution should be applied not to push the skull during
thinning.
15. The skull is consisted of three layers, two compact bones sand-
wich a spongy bone, and the total thickness is ca. 0.5 mm. As the
thinning is going on, the transparency of the skull becomes
higher, and blood vessels on the brain surface can be seen clearer.
It is possible to know which layer of the skull is currently thinned;

within compact bone layer, drilling causes small bone powder
whereas rough powder in spongy bone layer. Spongy bone layer
often contains small canals for blood vessels and bone marrow
(Fig. 2a), and bleeding may occur during the thinning but usu-
ally stops within a few minutes.
16. As the thinning proceeds, the bone becomes very soft not like
bone but something like paper sheet. Drill bur should be
moved as if it traces on a sheet of thin paper. The thickness of
the third layer (compact bone) is ca. 100 μm. At this thickness,
the blood vessel on the cortical surface can be seen without
immersing the bone with saline.
17. Usually it is the center of the initial thinning area. If large
blood vessel that may hinder the imaging appears during thin-
ning, set the final thinning area properly.
18. It is easier to scalp when the bone is dried, but saline immer-
sion should be done frequently to remove the heat, bone shav-
ing, and assess thinning progress. The thinning should be
done as flat as possible to decrease the light scattering and
increase the resolution of image.
19. The thickness of the bone can be assessed by two-photon
microscopy observation by measuring the autofluorescence of
bone. If the thickness does not reach 10–20 μm, thinning pro-
cess should be repeated.
20. This image is indispensable for the repeating imaging. If the
imaging interval is more than a week, rethinning of observa-
tion window is necessary to do. This image helps to locate the
thinning area precisely.
21. Initially laser intensity should begin from minimal; otherwise,
microglia can be activated unexpectedly. As the laser power
increases, autofluorescence of skull will be seen. The thickness
of the bone can be measured by subtracting the depth at the
start and the end point of autofluorescence. Dura mater has
autofluorescence as well, but the appearance of autofluores-
cence of bone and dura mater is different. The autofluores-
cence of bone is relatively homogeneous but of dura mater
looks fiber-like structure.
22. Due to the remained skull, light scattering and aberration are
significant in thinned-skull preparation. The maximal depth of
imaging is around 200 μm from the surface, and typically
imaging depth is restricted to 100 μm for fine structure obser-
vation of microglial processes without any correction of aber-
ration. It is better to avoid the recording site close to the large
blood vessels, because aberration and also motion artifact is
larger surround.
23. One of problems during imaging is motion artifact. This arti-

fact problem is substantial to the three-dimensional stack
image and time-lapse imaging. Motion artifact primarily
results from respiration and heartbeat. Lateral image move-
ment can be aligned offline by registration algorithm to a cer-
tain degree (e.g., Image J plugins), but axial movement
remains difficult to align. To reduce the motion artifact,
changing the posture of mouse or depth of anesthesia level is
sometimes effective. Artificial respiration may help to diminish
the motion. Respiration can be controlled with tightly attached
mask or by tracheal intubation. These methods can be also
applied for mouse, but it is difficult to maintain during a long
time period imaging. It is recommended to check periodically
if the lens is immersed enough. Deterioration of images dur-
ing the imaging is often related with the inadequate or lack of
immersion of the lens.
24. The resolution of images varied within the same window. It
often happens despite of the careful flat thinning of skull. It is
presumably related with the natural inhomogeneity in bone
and unflatness of the opposite surface of skull.
25. This photograph is indispensable to relocate the same area at
later time point.
26. The skull may be covered with connective tissues and some-
times contains newly invaded blood vessels. The thickness of
connective tissue depends on the interval from the previous
imaging. If intensive removal is necessary, use scalpel and
forceps.
27. Usually branching and running pattern of blood vessels are
almost unchanged. Therefore, vasculature pattern is a good
landmark to find the previously imaged area.
28. Dimmer or more distorted images are sometimes observed at
repetitive imaging. First point to suspect is the setting of
microscopy. It is necessary to check the output laser power
and excitation and collection light path. During the long-term
repetitive imaging expression, the level of fluorescent protein
may decrease. Bone regrowth on the opposite side may occur
and increased inhomogeneities in bone resulted in the reduc-
tion of signal intensity. Further rethinning may improve the
image quality. Increase of excitation power should be per-
formed carefully not to give photodamage and activate
microglia.
29. Among red fluorescent proteins, dsRed2 or tdTomato is use-
ful for the simultaneous dual color imaging, because they can
be excited at the same wavelength as GFP and the emission
spectra of GFP and these red fluorescent proteins have very
little overlap. Therefore, this combination would be excellent
for separating each signal.
30. For the simultaneous dual color imaging, microscope is neces-

sary to be equipped with two PMTs. The excitation wave-
length for dual color imaging should be determined by the
two-photon cross-section spectra. For the case for GFP and
dsRed2, wavelength at 920 nm can be used for the simultane-
ous excitation with high efficiency but low water absorption
[16]. Emission maxima of GFP are around 510 nm; hence,
dsRed2 is around 580 nm upon simultaneous excitation. The
appropriate filter set to separate these two signals could be
500–550 nm for GFP, 570–600 for dsRed2, and 560 nm for
the dichroic mirror.
Acknowledgments
S.O. is supported by Grants-in-Aid for Scientific Research

(18200025, 20019013, 21220008, and 22650070), Global COE
Program (Integrative Life Science Based on the Study of Biosignaling
Mechanisms), and Strategic Research Program for Brain Sciences
from MEXT Japan and by Takeda Science Foundation.
References
1. Tremblay ME, Stevens B, Sierra A et al (2011) response on cortical spine dynamics revealed by
Mini-Symposium: The role of microglia in the two-photon microscopy in vivo. Mol Brain 4:27
healthy brain. J Neurosci 31:16064–16069 9. Majewska A, Yiu G, Yuste R (2000) A custom-
2. Hirasawa T, Ohsawa K, Imai Y et al (2005) made two-photon microscope and deconvolu-
Visualization of microglia in living tissues tion system. Eur J Physiol 441:398–408
using Iba1-EGFP transgenic mice. J Neurosci 10. Xu HT, Pan F, Yang G, Gan WB (2007)
Res 81:357–362 Choice of cranial window type for in vivo
3. Jung S, Aliberti J, Graemmel P et al (2000) imaging affects dendritic spine turnover in the
Analysis of fractalkine receptor CX3CR1 func- cortex. Nat Neurosci 10:549–551
tion by targeted deletion and green fluorescent 11. Holtmaat A, Bonhoeffer T, Chow DK et al
protein reporter gene insertion. Mol Cell Biol (2009) Long-term, high-resolution imaging in
20:4106–4114 the mouse neocortex through a chronic cranial
4. Nimmerjahn A, Kirchhoff F, Helmchen F window. Nat Protocols 4:1128–1144
(2005) Resting microglial cells are highly 12. Farrar MJ, Bernstein IM, Schlafer DH et al
dynamic surveillants of brain parenchyma in (2012) Chronic in vivo imaging in the mouse
vivo. Science 308:1314–1318 spinal cord using an implanted chamber. Nat
5. Davalos D, Grutzendler J, Yang G et al (2005) Methods 9:297–302
ATP mediates rapid microglial response to 13. Drew PJ, Shih AY, Driscoll JD et al (2010)
local brain injury in vivo. Nat Neurosci Chronic optical access through a polished and
8:752–758 reinforced thinned skull. Nat Methods
6. Wake H, Moorhouse AJ, Jinno S et al (2009) 7:981–984
Resting microglia directly monitor the func- 14. Feng G, Mellor RH, Bernstein M et al (2000)
tional state of synapses in vivo and determine Imaging neuronal subsets in transgenic mice
the fate of ischemic terminals. J Neurosci expressing multiple spectral variants of GFP.
29:3974–3980 Neuron 28:41–51
7. Tremblay ME, Lowery RL, Majewska AK 15. Saito T (2006) In vivo electroporation in the
(2010) Microglial interactions with synapses embryonic mouse central nervous system. Nat
are modulated by visual experience. PLoS Biol Protocols 1:1552–1558
8:e1000527 16. Drovizhev M, Makarov NS, Tillo SE et al
8. Kondo S, Kohsaka S, Okabe S (2011) Long- (2011) Two-photon absorption properties of
lasting effect of a transient peripheral immune fluorescent proteins. Nat Methods 8:393–399
Chapter 29
Use of Confocal Microscopy in the Study of Microglia

in a Brain Metastasis Model
Manuel Sarmiento
Abstract
Confocal imaging of brain slices is a worthwhile analysis method to study the structure and function of
resting and activated microglia with submicrometer resolution. This chapter will focus on acquisition of
high-resolution confocal image stacks where we will discuss the technical aspects of confocal imaging in
brain sections as well as some of the currently suitable fluorescent markers for this type of work.
Key words Confocal, Microglia, Brain, Cancer, Metastasis, Immunofluorescence
1 Introduction
The initial steps of brain metastasis are difficult to detect in vivo,

and this is one of the primary reasons for poor prognosis in patients
with this disease; the median survival of untreated patients is 1–2
months [1]. Confocal microscopy has become a widely used method
in the study of brain diseases. This widespread adoption of confocal
microscopy is the result of the advantages of the method, namely,
eliminating out-of-focus light which produces sharp, high-contrast
images of cells and subcellular structures even within thick sections,
combined with the availability of increasingly powerful, sensitive,
and user-friendly instruments from several manufacturers.
Metastasis is the movement or spreading of cancer cells from
one organ or tissue to another. In the case of brain metastasis, the
lymph nodes are not considered due to the absence of a lymphatic
system within the central nervous system (CNS). The process of
brain metastasis depends on the success of several steps, including
cancer cell escape from the primary tumor, distribution and sur-
vival within the circulation, adhesion and penetration of the blood–
brain barrier (BBB), and proliferation within the brain
microenvironment. Importantly, the brain is the only site of tumor
337
338 Manuel Sarmiento
relapse in ~60 % of lung cancer patients, ~25 % of breast cancer

patients, and ~55 % of melanoma patients [2] and is a frequent site
of therapeutic failure.
The purpose of this chapter is to provide guidance in the use of
confocal microscopy in the study of the behavior of microglia in
one metastasis model. Microglia are the resident macrophages of
the CNS [3]. Their roles in neurodegenerative and neuroinflam-
matory response have been intensely investigated, whereas their
potential function in metastasis received almost no attention.
Microglia have been described as enhancing invasion and coloniza-
tion of brain tissue by breast cancer cells, serving both as active
transporters and guiding rails to further invasion [4]. Recent
insights into the mechanisms of cancer progression demonstrate
the critical role of the tumor microenvironment in brain metasta-
sis, tempting speculation with the fact that the resident
macrophages/microglia may play an even more important role for
colonization than the peripheral blood-derived macrophages.
Other clinical reports state that once the tumor is in the CNS (i.e.,
glioma grade IV), macrophages/microglia support tumor growth
by changing its microenvironment via neovascularization [5].
Several features of confocal scanning make it advantageous
compared to conventional wide-field microscopy observations and
measurements in brain sections. Probably the most important fea-
ture is that it allows the visualization of much smaller objects (e.g.,
microglia engulfing tumor cells). Moreover, it also makes possible
the ability to correctly interpret the 3-dimensional organization of
fine brain structure, for example, cell-to-cell contacts.
2 Materials
2.1 Tissue 1. Heparinized sodium chloride: Make up 50 mL of 0.9 % sodium

Preparation chloride in distilled water and add 500 U of heparin (sodium
heparin 125,000 U/5 mL).
2. PLPlight: Periodate-lysine-paraformaldehyde (PLP) containing
only 0.025 % glutaraldehyde (see Note 1).
3. 0.1 M phosphate buffer: Mix 3.1 g of sodium dihydrogen ortho-
phosphate in 500 mL distilled water. Mix 13 g of Sorenson’s
salt ((Na2HPO4)2H2O) in 400 mL distilled water. Add together
and make up to 1 L. If necessary adjust pH to 7.4.
4. 30 % sucrose (in phosphate buffer): The brains will be cryopro-
tected in a high-concentration sucrose solution composed by
30 g of sucrose in 100 mL of phosphate buffer (see Note 2).
5. Isopentane: To speed the freezing process, the brains will be
frozen in isopentane at −20 ºC for at least 30 min.
Microglia and Metastasis Under Confocal Sight 339
2.2 Tissue Other chapters within this book describe stereotaxical procedures
Processing involved with in vivo models. Thus, this chapter aims to begin
explanation of the methods used to process our samples once the
tumor cells are already implanted in the brain parenchyma.
1. PBS: Dissolve 50 tablets of phosphate-buffered saline
(Dulbecco A, Oxoid, UK) in 5 L of distilled water.
2. Quench solution: Add 2.5 mL of hydrogen peroxide 30 %
(Sigma Aldrich, UK) to 250 mL of methanol (see Note 3).
3. Blocking solution: Block endogenous streptavidin and biotin
using the streptavidin/biotin kit (Vector Laboratories, USA).
After this, we need also to block sections with Tris–NaCl
blocking buffer (TNB, Perkin-Elmer, USA).
4. Image-IT® FX signal enhancer (Invitrogen, USA) for 30 min.
This reagent will improve the signal of the fluorophore.
5. Primary antibody against microglia (1:200 in TNB). Iba-1
(DAKO, USA. Ionized calcium binding adaptor molecule 1).
6. Secondary antibody: Biotinylated antibody against the species
where the primary antibody was raised in (1:200 in TNB;
Vector Laboratories Inc, USA).
7. Streptavidin-HRP (1:100 in TNB; Perkin-Elmer, USA).
8. TSA-biotin (Tyramide Signal Amplification-biotin) (1:100 in
amplification buffer; Perkin-Elmer, USA) (see Note 4).
9. Streptavidin-Cy3 fluorophore (1:200 in TNB; Invitrogen, USA).
10. Vectashield mounting medium (Vector Labs, USA) with DAPI.
11. Cover slips.
2.3 Imaging 1. Confocal microscope (LSM-710, Carl Zeiss Microimaging,

Jena, Germany) (see Fig. 1).
2. Oil immersion (either oil or water immersion solution depend-
ing on the lens).
3. Lens tissues (see Note 5) (see Fig. 1a, b).
3 Methods
Fluorescence occurs when a molecule absorbs one or more photons

to reach the excited state from which it can relax to the ground state;
this process is accompanied by emission of a photon. Therefore,
confocal imaging in brain sections requires that the cells or their ele-
ments must be tagged with some kind of fluorophore. The labelling
can be achieved not only by immunofluorescence but also by other
techniques such as viral gene expression systems or DNA transfec-
tion. In this case, our mouse mammary carcinoma cells (4T1) are
tagged with a previously transfected GFP fluorophore [6].
Fig. 1 Comparison of a tumor colony within the striatum of a mouse brain taken
under two different microscopes. (a) Photomicrograph captured from an epi-
fluorescence microscope. (b) Picture taken from a Zeiss confocal microscope.
Both pictures were acquired from the same brain section. Green: tumor cells
tagged with GFP, Blue: DAPI to stain nuclei, Red: Iba-1 antibody against microglia.
Scale bars, 50 μm
Conventional confocal imaging of cells within brain sections is

usually limited to the upper 60 μm of a section. The average pen-
etration of commercial antibodies is around 20 μm; therefore, the
thickness of the sections should be restricted to these measures.
3.1 Tissue 1. Inject anesthetic to the animal and start transcardiac perfu-
Preparation sion. Inject into the left ventricle 50 mL of heparinized saline
followed by 200 mL of PLPlight at ~40 mL/min. Ensure that
the right atrium of the heart is cut to relieve pressure.
2. Harvest the brain and postfix it for 4 h at 4 °C in PLPlight. After
this, transfer the brain into sucrose 30 % until sink, at 4 °C
(after 24–36 h the brain will sink, indicating that cryoprotec-
tion is complete).
3. Place the brain in isopentane at –20 °C for 30 min. Once frozen,
wrap it in tin foil and transfer to −20 °C for long-term storage.
4. The brains will be placed in a Leica cryostat at −20 °C tempera-
ture, and the thickness of the sections will be of 10 μm. Once
cut, the sections will be placed on the slides (see Note 6).
3.2 Tissue 1. Hydrate 10 μm sections for 5 min in PBS (see Subheading 3.1).
Processing 2. Quench the sections with 1 % hydrogen peroxide (see
Subheading 2.2, item 2) for 15 min.
3. Streptavidin- and biotin-blocked (Vector Laboratories, USA;
15 min each; see Subheading 2.2, item 3).
4. 30 min incubation in Image-IT® enhancer (Subheading 2.2,
item 4) plus 30 min in Tris–NaCl blocking buffer (TNB; see
Subheading 2.2, item 3).
5. Incubate overnight with anti-Iba-1 antibody raised in rabbit
(Wako, USA; see Subheading 2.2, item 5).
6. Rinse with PBS (three rinses, 10 min each).
7. Incubate with a biotinylated secondary antibody anti-rabbit
(Subheading 2.2, item 6) for 1 h.
8. Rinse sections (three rinses, 10 min each).
9. Incubate with streptavidin-HRP (see Subheading 2.2, item 7)
in TNB for 30 min.
10. Rinse with PBS (three times, 10 min each).
11. Incubate sections for 8 min in the dark with TSA-biotin
(see Subheading 2.2, item 8) diluted in amplification buffer
(see Note 4).
12. Rinse slides in PBS (three times, 10 min)
13. Incubate sections with a streptavidin-Cy3 fluorophore (1:200 in
TNB; Invitrogen, USA) (see Subheading 2.2, item 9) for 30 min.
14. Cover-slip the slides using Vectashield mounting medium (see
Subheading 2.2, item 10) with DAPI to stain nuclei. The
slides are now ready to move to the microscope.
3.3 Examination Images can be acquired using an inverted confocal microscope

Under Confocal (LSM-710, Carl Zeiss Microimaging, Jena, Germany) and analyzed
Microscopy using Image J (rsbweb.nih.gov/ij) and Zen (Carl Zeiss) software.
1. Place the slide in the sample holder. Start from a low magnifi-
cation lens in order to find the tumor colony.
2. Apply immersion medium (oil or glycerol depending on which
class of lens used) to the objective of the ×63. Focus on the
region of interest and choose the channels to be used depend-
ing on the fluorophores used in your immunofluorescence. In
this case we will use lasers at 405 nm (DAPI), 488 nm (GFP),
and 568 nm (Cy3).
3. Adjust laser intensity. In order to avoid saturation of the
images, each channel’s laser intensity must be checked indi-
vidually and adjust appropriately (see Note 7).
4. Set all the pinholes at ≤1 airy unit (AU). With a pinhole diam-
eter <1 AU, resolution improves (better point separation),
which is penalized by a drastic loss in energy (see Note 8).
5. Detection ranges were set to eliminate crosstalk between fluo-
rophores: 409–485 nm for DAPI, 494–553 nm for Alexa
Fluor, and 564–712 nm for Cy3. Detection windows can be
tuned individually in order to avoid the overlap of the emis-
sion spectra of the fluorophores (see Note 9).
6. Z-stack: use the micrometer to focus at both levels of the sam-
ple (upper and lower). Save both coordinates in order to set
the thickness of our range of detection (see Note 10).
7. Adjust the confocal parameters to obtain the better resolution
image (see Note 11) (see Fig. 2).
4 Notes
1. One of the reasons to use PLPlight (a “soft” fixative) is the use

of frozen sections during the processing of the samples. This
kind of fixative helps an easier penetration of the antibody
within the sample, allowing a more efficient binding between
the Iba-1 antibody and its epitope.
Recipe: Dissolve 3.4 g lysine HCl in 94 mL distilled water.
Dissolve 0.45 g disodium hydrogen orthophosphate dehy-
drate (Sorenson’s salt) in 25 mL distilled water. Add this to
lysine solution in order to adjust the pH to 7.4. Discard any
remaining phosphate solution. Dissolve 5 g of paraformalde-
hyde in 50 mL of distilled water at 60 °C. Add paraformalde-
hyde and slowly add a few drops of 1 M NaOH as a catalyst
while stirring until the solution clears. Leave to cool down.
Prepare 0.53 g of sodium metaperiodate. Mix the lysine solu-
tion, paraformaldehyde solution, and the sodium periodate
together. Make up to 250 mL with 0.1 M phosphate buffer
Fig. 2 Photomicrograph of a tumor within the striatum of a mouse 5 days after

implantation. You can see the tumor cells growing along the vessels and microglial
activation surrounding the boundaries of the colony. Green: tumor cells tagged with
GFP, Blue: DAPI to stain nuclei, Red: Iba-1 against microglia. Scale bar, 100 μm
and adjust pH to 7.2 with NaOH. Finally, add 0.25 mL of

glutaraldehyde 0.01 % for fixing tissue.
2. Depending on the accuracy of the transcardial perfusion,
sometimes the brains take longer to sink when left in sucrose
solution; if so, it is recommended to use sucrose gradient
embedding. After the first 4 h in PLPlight, transfer the brain to
10 % until it sinks. Repeat this procedure with 20 % and finally
30 % solutions of sucrose. In this way we can ensure that the
sucrose will perfuse inside the brain.
3. Some fluorophores are affected by the methanol quenching
solution. In these cases, instead of methanol, use PBS.
4. The Perkin–Elmer kit provides the TNB blocking buffer and
the amplification buffer into which the reagents will be diluted.
It is possible to substitute these with PBS if necessary.
5. It is important to thoroughly clean the lens after using the
immersion oil in order to avoid staining other lenses. The air
lenses could be seriously affected if immersion agents are
allowed to come into contact with them.
6. Once cut the brain, to avoid the detachment of the sections
from the slide, it is important to leave them dry properly; in this
case we left them overnight at room temperature in darkness to
avoid photobleaching of the GFP present in the tumor cells.
7. We need to check that the images are not overexposed to the

fluorophore; to do that, we must check that the image has no
saturated voxels. The exact method for doing so varies between
different manufacturer’s software, but in principle there will
be an option to highlight every saturated pixel. If necessary,
we can reduce the gain/power of the laser to remove them.
8. A small pinhole diameter will increase the depth of focus, but
reduce the light intensity received so that the projected image
will be dimmer. For a proper confocal microscopy, the pinhole
shouldn’t be wider than 1 AU. However, it should be consid-
ered that it will depend on the signal level which noise source
dominates. With high-amplitude signals (number of detected
photons >10,000), laser noise is the dominating effect, whereas
the quality of low signals (number of detected photons <1,000)
is limited by the shot noise of the light. 1 Airy Unit is a good
value to enable a confocal fluorescence XY-image to be obtained.
9. Even without a fluorescent dye, brain sections contain fluores-
cent molecules. Some of these molecules, such as tyrosine,
tryptophan, and NADH, will only be fluorescent when excited
with UV light (<400 nm). In confocal imaging, UV light is
not normally used as it does not penetrate well into the tissue.
The 488 nm light from an argon laser (the most commonly
used wavelength in a single-photon system) excites other
endogenous molecules, in particular flavins and porphyrins. It
is important to note that most of these molecules emit in the
green–yellow part of the spectrum (500–600 nm). Thus, it is
important to take into account the autofluorescence problems
of the fluorophores variants of green fluorescent protein
(GFP) that are imaged using 458 or 488 nm excitation and so
emit in the green part of the spectrum, where both absorption
and emission of endogenous molecules are the strongest [7].
Molecules that are mainly responsible for autofluorescence
are mostly unexcitable by >600 nm light (e.g., 633 nm line of
red diode lasers). Thus, the red part of the spectrum seems to
be ideal for single-photon imaging of difficult and deep objects
in brain sections where autofluorescence is intense (i.e., Cy3
or Cye5 dye) [7].
10. Confocal imaging gives us the option of either choosing a sin-
gle plane of the section to image or to obtain the average of all
the slices exposed (known as maximum intensity projection).
Depending on the quality of our immunofluorescence, and
our objectives, either thing is possible (see Fig. 3).
11. The system determines the settings for the Gain (Master) and
digital offset to match the scan speed, pinhole size, and laser
power. The higher digital offset, the lower background, but
the intensity of the signal received will be also reduced (see
Fig. 3a, b).
Fig. 3 These two photomicrographs are taken from a single tumor colony in a
mouse 5 days after tumor implantation. (a) A single X–Y plane section of the
slide. (b) A combined image using the average of intensity of 20 planes of the
same slide (maximum intensity projection). Scales bars, 50 μm
References
1. Eichler AF, Loeffler JS (2007) Multidisciplinary 5. Komohara Y, Ohnishi K, Kuratsu J et al (2008)
management of brain metastases. Oncologist Possible involvement of the M2 anti-
12:884–898 inflammatory macrophage phenotype in growth
2. Sampson JH, Carter JH Jr, Friedman AH et al of human gliomas. J Pathol 216:15–24
(1998) Demographics, prognosis, and therapy 6. Aslakson CJ, Miller FR (1992) Selective events
in 702 patients with brain metastases from in the metastatic process defined by analysis of
malignant melanoma. J Neurosurg 88:11–20 the sequential dissemination of subpopulations
3. Hanisch UK, Kettenmann H (2007) Microglia: of a mouse mammary tumor. Cancer Res
active sensor and versatile effector cells in the 52:1399–1405
normal and pathologic brain. Nat Neurosci 7. Kasparov S, Teschemacher AG, Paton JF
10:1387–1394 (2002) Dynamic confocal imaging in acute
4. Pukrop T, Dehghani F, Chuang HN et al brain slices and organotypic slice cultures
(2010) Microglia promote colonization of using a spectral confocal microscope with sin-
brain tissue by breast cancer cells in a Wnt- gle photon excitation. Exp Physiol
dependent way. Glia 58:1477–1489 87:715–724
INDEX
A Caspase-3 ............................................................. 6, 113–120

Caspase-7 ......................................................... 113, 114, 120
7-AAD. See 7-Aminoactinomycin D (7-AAD) Caspase-8 ............................................................. 6, 113–120
Adult microglia.................. 42, 44, 47, 49, 185–197, 202–205 Caspase assay ............................................................114–117
Alzheimer ......................................5, 144, 186, 219, 244, 295 Caspases ............................................................... 6, 113–120
7-Aminoactinomycin D (7-AAD) .......................... 126, 130, CBA. See Cytokine bead array (CBA)
132, 135–137, 139, 142, 143 CCD camera ............................. 148, 149, 158, 321, 324, 325
Amplex Red assay ..................................................... 105, 107 CD45 ................................... 30, 34, 37, 38, 48, 50, 130, 132,
Animal models ................................................. 295–304, 308 133, 135–139, 141–143, 188, 192, 193, 315, 317
Annexin V .........................................130, 132, 135–140, 142 CD200...................................................................... 219, 226
Antibody.........................................22, 34–37, 50, 57, 71–74, CD11b ............................11, 25, 30, 34–38, 57, 59, 130, 131,
94–100, 106, 109, 111, 115, 118, 120, 130, 132, 135, 135–139, 142, 143, 188, 191–193, 196, 224, 225
136, 138, 142, 143, 188, 189, 192, 195–197, 220, 224, CD200R ...........................................................................219
225, 250, 255, 258, 268, 274, 279, 281–289, 293, 310, Cell biology .............................................................. 185, 200
313, 315–317, 339–342 Cell culture ............................... 12–15, 34, 46, 56, 57, 64, 65,
Antigen retrieval ....................................................... 286, 288 72–73, 98, 100, 108, 110, 119, 200, 223–226, 231,
Apoptosis.................................................. 113, 118, 129–144 233, 238, 239, 244, 247, 256, 258, 292
B Cell dynamics ........................................................... 216, 318
Cell lines ...............11–15, 67, 80, 99, 157, 164, 167, 199, 220
BBB. See Blood brain barrier (BBB) Central nervous system (CNS) .............................. 3–5, 7, 17,
BCECF. See 2′,7′-Bis-(2-carboxyethyl)-5-(and-6)- 25, 55, 63, 71, 72, 83, 84, 90, 93, 129, 130, 141, 143,
carboxyfluorescein (BCECF) 199, 200, 215, 216, 219, 244–246, 262, 282, 307, 308,
Biochemical methods .........................................................43 319, 337, 338
Biological irradiator ..........................................................308 Chemokines.........................................3, 5–7, 17, 33, 71, 320
Biological methods ...........................................................319 Chimeras ..........................................................................220
2′,7′-Bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein CHME5 cells .....................................................................13
(BCECF)...............................149, 155, 156, 158, 160 Cl− currents ............................................... 166, 167, 173, 180
Blood brain barrier (BBB) ..................................... 14, 17, 55, CNS. See Central nervous system (CNS)
84, 308, 337 Co-cultures ............................................................... 213, 226
Bone marrow ............... 27, 130, 308, 309, 311–312, 314, 333 Confocal ......................................50, 126, 148, 281, 337–345
Bone marrow chimeric mice .............................................308 Conjugated-Saporin .....................................................57–59
Brain ........................................... 3, 11, 17, 25, 33, 41, 55, 72, Cortical neuronal cell culture....................................224–226
85, 130, 163, 199, 215, 231, 244, 262, 285, 295, 308, Cryopreserving .................................................................266
319, 337 Cryostat sections ........................................ 87, 264, 265, 270
BrdU. See Bromodeoxyuridine (BrdU) Current clamp .................................................. 169, 170, 174
Bromodeoxyuridine (BrdU)...................... 130–132, 135–142 CX3CR1 ...................................168, 220, 307–317, 320, 330
BV2 cells................................................................... 164, 220 Cx3cr1-EGFP knock-in mice .................................. 320, 330
Cytokine ....................................... 3–7, 12, 17, 22, 25–27, 30,
C 33, 69, 71–81, 83–90, 93–100, 104, 142, 163, 186,
Calcium release......................................... 159, 166, 173, 175 220, 291, 296
Caspase-1 .....................................................................6, 113 Cytokine bead array (CBA) ..........................................71–80
DOI 10.1007/978-1-62703-520-0, © Springer Science+Business Media New York 2013
347
MICROGLIA: METHODS AND PROTOCOLS
348 Index
D Histology/histochemistry ........................................ 243–258,

261–279, 286–287
DEVDase activity.............................................................114 Human ................................. 11–13, 41–50, 64, 96, 105, 115,
Developmental biology.....................................................185 118, 119, 199–211, 216, 244, 246, 247, 250
Diagnostics ............................................................... 200, 281 Human brain tissue ............................................ 42, 202, 210
Diaphorase staining .......................................... 104–106, 109
Double labelling .......................................244, 246, 249–250, I
255–256, 265, 267–268, 273–275
Iba-1 ................................................................. 239, 339–343
E Iba1-EGFP transgenic mice ............................................320
IETDase activity .............................................. 114, 119, 120
Electro-chemiluminescence................................................95 Immunoblotting ............................................... 188, 193–195
ELISA. See Enzyme-linked immunosorbent Immunohistochemistry ................................83, 88, 141, 246,
assay (ELISA) 247, 249–250, 255–256, 265, 267–268, 273–245,
Enzymatic histochemistry ................................ 243–258, 263 281, 284, 286–288
Enzyme-conjugated oligodeoxynucleotide probes ..............83 Immunology .......................................................................17
Enzyme-linked immunosorbent assay Inducible nitric oxide synthase (iNOS) ....................... 34, 38,
(ELISA) ..................................................... 71, 94, 95 104–105, 108–110, 193, 194, 220
Experimental autoimmune Inflammation ..........................................4, 25, 26, 72–75, 78,
encephalomyelitis.......................................... 262, 308 104, 219, 295, 296
Inflammatory gene ...........................................................193
F Inflammatory gene expression ............................................35
FACS. See Fluorescence-activated cell sorting (FACS) Innate immunity ...........................................71, 72, 129, 185,
Flow cytometry...................................22, 34–36, 71–81, 114, 199, 216, 218, 220
115, 117–119, 123, 125, 129–143, 192, 209, 289, iNOS. See Inducible nitric oxide synthase (iNOS)
308–310, 312–313, 315 In situ hybridization .....................................................83–90
Fluorescence-activated cell sorting (FACS)........... 36, 72–76, Interleukin (IL) .................................. 5, 6, 12, 26, 30, 34, 38,
79, 119, 132, 186, 188, 192, 196 74, 80, 84, 94, 96, 98, 123, 186, 193, 194, 201, 208,
Fluorescence imaging ...............................................147–160 211, 220
Fluorogenic substrate .......................................................113 Interleukin-1 (IL-1) ................................................. 193, 194
Fluorometric 2,3-diaminonaphtalene Intrathecal catheter................................................... 292, 293
assay .............................................. 104, 105, 108–109 In vivo imaging ......................................... 216, 319, 322, 330
Fluorometry.............................................. 104, 105, 108–109 Ion channels ............................................. 147, 149, 163–181
Fractalkine receptor ..........................................................307 Ion transporters ................................................................147
Fura-2 ........................................149–151, 154, 157, 158, 160 Isolation ..........................................11, 17–22, 26, 28, 33–38,
41–50, 72, 123, 130, 133, 148, 149, 164, 175,
G 186–188, 190–193, 197, 200, 202, 204–208, 224,
225, 309, 311–312
Genetics.................................................56, 63, 187, 237, 308
Gene transduction ....................................................292–293 K
GFAP. See Glial fibrillary acidic protein (GFAP)
Glial fibrillary acidic protein (GFAP) ....................... 22, 225, K+ currents ........................................ 165–166, 171–172, 179
232, 239, 245, 249, 250, 255, 256, 258, 264, 265, 267,
268, 273–274 L
Glioma ..................................................................... 5, 6, 338 Lectins ...................................................................... 262, 282
GM-CSF. See Granulocyte/macrophage colony-stimulating Lentiviral vector ...........................................................65–67
factor (GM-CSF) Leucine methyl ester (LME)................................ 56–59, 239
G protein-activated K+ currents ........................ 166, 171, 179 Lipopolysaccharide (LPS) ............................5, 14, 15, 22, 30,
Granulocyte/macrophage colony-stimulating factor 34, 35, 37, 38, 71, 104, 110, 119, 201, 208, 216,
(GM-CSF) ..................................27, 30, 50, 201, 208 295–304, 308
Griess reaction .................................................. 104, 105, 108 Lipoteichoic acid (LTA) ...................................................103
LME. See Leucine methyl ester (LME)
H
LPS. See Lipopolysaccharide (LPS)
H+ currents ............................................... 166, 167, 172, 180 Luminogenic substrate .....................................................114
High content analysis ...................................................41–50 Luminometry ...................................................................116
Index
349
M Neurotoxicity .........................................6, 219, 239, 296, 308

NF-κB ..............................................................................6, 7
Macrophage .................................... 4, 5, 7, 11, 17, 26, 27, 33, Nitric oxide (NO) ....................4, 17, 103–105, 108–110, 296
50, 59, 71, 84, 94, 130, 137, 139, 140, 143, 201, 215, 3-Nitrotyrosine ......................................... 104, 106, 109, 111
216, 218, 282, 283, 289, 295, 320, 338 3-Nitrotyrosine immunocytochemistry .................... 106, 109
Macrophage colony-stimulating factor nNOS. See Neuronal nitric oxide synthase (nNOS)
(M-CSF) .....................................27, 30, 50, 201, 208 Non-selective cation (TRP) channel
Magnetic cell separation ...............................................33–38 currents .................. 166, 167, 172–173, 176, 179, 180
Major histocompatibility complex class I Nucleoside-diphosphatase (NDPase) ...................... 243–256,
(MHC-I) .............................. 264, 266–268, 273–275 258, 263
M-CSF. See Macrophage colony-stimulating factor
(M-CSF) O
Medium .................................... 13, 14, 18–20, 22, 28–30, 46,
OX-6 ................................................................ 283–285, 288
56–59, 64–67, 72, 80, 96, 97, 104, 106–108, 110, 116,
OX-42 .......................................111, 232, 239, 283–285, 289
117, 119, 123, 124, 126, 131, 187, 190, 204–208, 210,
222–225, 233–237, 239, 244, 245, 252, 253, 256, P
258, 273, 275, 278, 284, 287, 289, 292–294, 304,
309, 311, 339, 341, 342 Pain ..............................................................................6, 291
Mesencephalic neuron-glia Culture.......... 231, 232, 234–235 Paraffin sections.................................265, 270–272, 278, 283
Metastasis .................................................................337–345 Parkinson .............................................................. 5, 232, 295
MHC-I. See Major histocompatibility complex Patch clamp technique ..................................... 147, 164, 175
class I (MHC-I) Perforated patch recordings .............................. 167, 176, 177
Microglia Peripheral nerve injury .........................................................6
activation ............................................................ 5, 14, 33 Peroxynitrite (OONO−)............................ 104, 106, 109, 110
cell line .......................... 11–15, 67, 80, 99, 157, 164, 167 Phagocytic .................................... 4, 27, 30, 48, 50, 122, 123,
depletion .................................................................55–59 215, 217, 218, 282, 307
enriched culture .......................................... 232, 236–237 Phagocytic NADPH oxidase............................................104
Microglial marker ........................................... 30, 36, 37, 262 Phagocytosis ................................. 4, 6, 12, 26, 42, 44, 47–50,
Microglial proliferation ............................................129–143 121–127, 185, 186, 216, 218, 220
Microglia-neuron cross-talk .....................................215–227 Pharmacology ............................................................. 56, 175
Mixed glia cultures ........................................... 232, 235–236 Physiology ................................................................ 231, 232
M1/M2 phenotype ................................................... 186, 200 Polarization........................................186, 200, 201, 208–209
MOG35-55 peptide .............................................................308 Polystyrene latex beads ..................................... 121, 122, 126
Molecular genetics .................................56, 63, 187, 237, 308 Postnatal ................................ 4, 25–30, 33–38, 185, 222, 330
Monochromator ....................................... 148–150, 153, 157 Primary astrocyte culture ..............................................56, 57
Mouse....................................... 11, 12, 17–22, 25–30, 35, 72, Primary cells ..................................................... 12, 34, 55–59
73, 78–80, 84, 85, 88, 90, 96, 115, 118, 119, 132, 133, Primary human culture ............................................... 50, 199
136–141, 185–188, 190, 191, 196, 222–224, 234, Primary neuronal culture ....................................................58
236–238, 246, 268, 274, 285, 309–314, 316, 317, Proliferation................................ 17, 26, 27, 49, 93, 129–143,
320–323, 325, 327–332, 334, 339, 340, 343, 345 163, 295, 315, 337
mRNA .............................................38, 83–90, 188, 193, 197 Proteinase K .............................................................284–286
Myeloid cells .............................................. 55, 200, 219, 220 Protein kinase C ...................................................................6
N R
Na+ currents ...................................................... 165, 172, 179 Radiation .......................................................... 308, 311, 316
NDPase. See Nucleoside-diphosphatase (NDPase) Rat .................................... 12, 25–30, 96, 115, 132, 136, 185,
Neurodegeneration ..................................25, 42, 93, 186, 219 222–224, 234, 237, 246, 257, 266, 267, 276, 277, 288,
Neuroimmunoregulatory molecules..................................219 295–304, 310
Neuroinflammation .............................4, 5, 17, 130, 295–304 Reactive nitrogen species (RNS) ................................ 17, 103
Neuronal nitric oxide synthase (nNOS) ................... 109, 110 Reactive oxygen species (ROS)..........4, 17, 26, 103, 104, 163
Neuron-glia culture .....................34, 232, 234–235, 237–239 Real-time PCR........................................... 38, 188, 193, 196
Neurons ............................................ 3–5, 17, 55, 56, 59, 104, Reconstituted neuron-microglia culture ...........................231
109, 200, 210, 216, 218–221, 225, 226, 231, 232, Resin-embedding procedure ..................................... 249, 269
237–239, 247, 282, 319, 330, 331 RNS. See Reactive nitrogen species (RNS)
Neuropathic pain ..........................................................6, 291 ROS. See Reactive oxygen species (ROS)
350 Index
S Tomato lectin ...........................................................261–279

Trafficking ........................................................................307
SBFI. See Sodium-binding benzofuran-isophthalate (SBFI) Transmission electron microscopy (TEM) .............. 247, 250,
Serum ..................................... 14, 18–20, 30, 64, 65, 97, 109, 252–254, 257, 267, 269, 270, 275–277
122, 191, 223, 225, 227, 285–287 Tumor necrosis factor (TNF) .......................................6, 186
Sodium-binding benzofuran-isophthalate Two-photon microscopy ...........................................319–335
(SBFI)...................................................................155
Spinal cord...................................... 4, 6, 11, 13, 59, 247, 250, V
266, 269, 292, 293, 313, 317, 323
Stereotaxic surgery............................................................296 Vibratome sections ...................................246, 250–252, 257,
SULFO-TAG ...............................................................95, 96 265, 269–270, 277
Superoxide (O2−) ........................................... 4, 103–107, 110 Voltage clamp ........................................... 170–171, 174–175
Voltage ramp .................................................... 170–175, 178
T Voltage step .............................................. 170–175, 178–180
TEM. See Transmission electron microscopy (TEM) W

TNF. See Tumor necrosis factor (TNF)
Toluidine blue counterstaining ................. 249, 253, 266, 273 Whole-cell recordings .......................164, 165, 171–173, 176

Micro Glia

Uploaded by

Copyright:

Available Formats

Micro Glia

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Micro Glia

Uploaded by

Copyright:

Available Formats

Methods in

Molecular Biology 1041

For further volumes:

Methods and Protocols

José Luis Venero

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Library of Congress Control Number: 2013940920

© Springer Science+Business Media New York 2013

Printed on acid-free paper

Humana Press is a brand of Springer

Stockholm, Sweden Bertrand Joseph

PART I OVERVIEW OF MICROGLIA BIOLOGY

1 A Brief Overview of Multitalented Microglia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

PART II ISOLATION AND CULTURE OF MICROGLIA

2 Cell Culturing of Human and Murine Microglia Cell Lines . . . . . . . . . . . . . . . . . 11

PART III DEPLETION AND TRANSDUCTION OF MICROGLIA

7 Depletion of Microglia from Primary Cellular Cultures . . . . . . . . . . . . . . . . . . . . 55

PART IV ANALYSIS OF MICROGLIAL CYTOKINE PRODUCTION

9 Microglial Activation: Measurement of Cytokines by Flow Cytometry. . . . . . . . . 71

11 Use of Meso-Scale Discovery™ to Examine Cytokine Content

PART V ANALYSIS OF MICROGLIA ACTIVATION

12 Analysis of Microglial Production of Reactive Oxygen and Nitrogen Species . . . . 103

PART VI ANALYSIS OF MICROGLIA POLARIZATION

18 Studying M1 and M2 States in Adult Microglia . . . . . . . . . . . . . . . . . . . . . . . . . 185

PART VII CO-CULTURE SYSTEMS TO ANALYSIS MICROGLIA INTERACTIONS

20 Understanding Microglia–Neuron Cross Talk: Relevance

PART VIII ANALYSIS OF MICROGLIA FUNCTIONS IN VIVO

22 Microglia Detection by Enzymatic Histochemistry . . . . . . . . . . . . . . . . . . . . . . 243

24 Immunohistochemical Detection of Microglia . . . . . . . . . . . . . . . . . . . . . . . . . . 281

BEATRIZ ALMOLDA • Universitat Autònoma de Barcelona, Barcelona, Spain

MICHAEL T. HENEKA • Clinical Neuroscience Unit, Department of Neurology,

Overview of Microglia Biology

A Brief Overview of Multitalented Microglia

1 Overview of the Multitalented Microglia

stroke, toxins, and other stimuli are capable of producing an acute

including extracellular matrix proteases and cytokines, which

and IkB kinase (IKK), thereby leading to the nuclear translocation

Isolation and Culture of Microglia

Cell Culturing of Human and Murine Microglia Cell Lines

The concept of microglia cells was first described by Rio-Hortega

Cell line Species Immortalization method Year Authors

the important characteristics of microglial cells, like the ability of

The access to human samples for isolating microglia from the

1. Dulbecco’s Modified Eagle Medium (DMEM: high glucose

4. As soon as the cells have detached, add 5–10 ml of complete

1. RPMI 1640 medium can also be used for culturing of microglia

Another source of diversity here can also be the choice of LPS.

Microglia Isolation from Adult Mouse Brain

Microglia, the myeloid-derived resident macrophages of the brain,

stepwise method (modified from one previously described [7]) to

3 Media and Solutions

1. Dissection medium (50 mL).

5. Culture medium (50 mL).

1. Place the freshly perfused adult brain in 2 mL serum-free

6. Aspirate the medium slowly (be extremely cautious not to

21. Resuspend pellet in 500 μL of HBSS and transfer to small 0.6

1. Percoll should be kept at room temperature.

Preparation of Primary Microglia Cultures

Key words Microglia, Inflammation, Cytokine, Postnatal, Cell culture